Goldman-Cecil Medicine [26 ed.] 0323532667, 9780323532662

To be the best doctor you can be, you need the best information. For more than 90 years, what is now called Goldman-Ceci

6,133 1,167 143MB

English Pages 2664 [4629] Year 2019

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Goldman-Cecil Medicine [26 ed.]
 0323532667, 9780323532662

Table of contents :
Front Matter
Copyright Page
Associate Editors
Preface
Contributors
Video Contents
SECTION 1: SOCIAL AND ETHICAL ISSUES IN MEDICINE
1 Approach to Medicine, the Patient, and the Medical Profession
2 Bioethics in the Practice of Medicine
3 Palliative Care
4 Disparities in Health and Health Care
5 Global Health
SECTION 2: PRINCIPLES OF EVALUATION AND MANAGEMENT
6 Approach to the Patient
7 Approach to the Patient with Abnormal Vital Signs
8 Statistical Interpretation of Data and Using Data for Clinical Decisions
9 Measuring Health and Health Care
10 Quality, Safety, and Value
11 Comprehensive Chronic Disease Management
SECTION 3: PREVENTIVE AND ENVIRONMENTAL ISSUES
12 The Periodic Health Examination
13 Physical Activity
14 Adolescent Medicine
15 Immunization
16 Principles of Occupational and Environmental Medicine
17 Radiation Injury
18 Bioterrorism
19 Chronic Poisoning
SECTION 4: AGING AND GERIATRIC MEDICINE
20 Epidemiology of Aging
21 Geriatric Assessment
22 Common Clinical Sequelae of Aging
23 Urinary Incontinence
24 Neuropsychiatric Aspects of Aging
25 Delirium in the Older Patient
SECTION 5: CLINICAL PHARMACOLOGY
26 Principles of Drug Therapy
27 Pain
28 Biology of Addiction
29 Nicotine and Tobacco
30 Alcohol Use Disorders
31 Drugs of Abuse
32 Immunomodulatory Drugs
33 Biologic Agents and Signaling Inhibitors
34 Complementary, Alternative, and Integrative Medicine
SECTION 6: GENETICS
35 Principles of Genetics
36 Clinical Genomics—Genome Structure and Variation
37 Applications of Molecular Technologies to Clinical Medicine
38 Regenerative Medicine, Cell, and Gene Therapies
SECTION 7: PRINCIPLES OF IMMUNOLOGY AND INFLAMMATION
39 The Innate Immune System
40 The Adaptive Immune System
41 Mechanisms of Immune-Mediated Tissue Injury
42 Mechanisms of Inflammation and Tissue Repair
43 Transplantation Immunology
44 Complement System in Disease
SECTION 8: CARDIOVASCULAR DISEASE
45 Approach to the Patient with Possible Cardiovascular Disease
46 Epidemiology of Cardiovascular Disease
47 Cardiac and Circulatory Function
48 Electrocardiography
49 Echocardiography
50 Noninvasive Cardiac Imaging
51 Catheterization and Angiography
52 Heart Failure
53 Heart Failure
54 Diseases of the Myocardium and Endocardium
55 Principles of Electrophysiology
56 Approach to the Patient with Suspected Arrhythmia
57 Approach to Cardiac Arrest and Life-Threatening Arrhythmias
58 Supraventricular Cardiac Arrhythmias
59 Ventricular Arrhythmias
60 Electrophysiologic Interventional Procedures and Surgery
61 Congenital Heart Disease in Adults
62 Angina Pectoris and Stable Ischemic Heart Disease
63 Acute Coronary Syndrome
64 ST Elevation Acute Myocardial Infarction and Complications of Myocardial Infarction
65 Interventional and Surgical Treatment of Coronary Artery Disease
66 Valvular Heart Disease
67 Infective Endocarditis
68 Pericardial Diseases
69 Diseases of the Aorta
SECTION 9: VASCULAR MEDICINE
70 Arterial Hypertension
71 Atherosclerotic Peripheral Arterial Disease
72 Other Peripheral Arterial Diseases
73 Thrombotic Disorders
74 Venous Thrombosis and Embolism
75 Pulmonary Hypertension
76 Antithrombotic and Antiplatelet Therapy
SECTION 10: RESPIRATORY DISEASES
77 Approach to the Patient with Respiratory Disease
78 Imaging in Pulmonary Disease
79 Respiratory Testing and Function
80 Disorders of Ventilatory Control
81 Asthma
82 Chronic Obstructive Pulmonary Disease
83 Cystic Fibrosis
84 Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders
85 Alveolar Filling Disorders
86 Interstitial Lung Disease
87 Occupational Lung Disease
88 Physical and Chemical Injuries of the Lung
89 Sarcoidosis
90 Acute Bronchitis and Tracheitis
91 Overview of Pneumonia
92 Diseases of the Diaphragm, Chest Wall, Pleura, and Mediastinum
93 Interventional and Surgical Approaches to Lung Disease
SECTION 11: CRITICAL CARE MEDICINE
94 Approach to the Patient in a Critical Care Setting
95 Respiratory Monitoring in Critical Care
96 Acute Respiratory Failure
97 Mechanical Ventilation
98 Approach to the Patient with Shock
99 Cardiogenic Shock
100 Shock Syndromes Related to Sepsis
101 Disorders Due to Heat and Cold
102 Acute Poisoning
103 Medical Aspects of Trauma and Burns
104 Envenomation, Bites, and Stings
105 Rhabdomyolysis
SECTION 12: RENAL AND GENITOURINARY DISEASES
106 Approach to the Patient with Renal Disease
107 Structure and Function of the Kidneys
108 Disorders of Sodium and Water
109 Potassium Disorders
110 Acid-Base Disorders
111 Disorders of Magnesium and Phosphorus
112 Acute Kidney Injury
113 Glomerular Disorders and Nephrotic Syndromes
114 Tubulointerstitial Diseases
115 Diabetes and the Kidney
116 Vascular Disorders of the Kidney
117 Nephrolithiasis
118 Cystic Kidney Diseases
119 Hereditary Nephropathies and Developmental Abnormalities of the Urinary Tract
120 Benign Prostatic Hyperplasia and Prostatitis
121 Chronic Kidney Disease
122 Treatment of Irreversible Renal Failure
SECTION 13: GASTROINTESTINAL DISEASES
123 Approach to the Patient with Gastrointestinal Disease
124 Diagnostic Imaging Procedures in Gastroenterology
125 Gastrointestinal Endoscopy
126 Gastrointestinal Hemorrhage
127 Disorders of Gastrointestinal Motility
128 Functional Gastrointestinal Disorders
129 Diseases of the Esophagus
130 Acid Peptic Disease
131 Approach to the Patient with Diarrhea and Malabsorption
132 Inflammatory Bowel Disease
133 Inflammatory and Anatomic Diseases of the Intestine, Peritoneum, Mesentery, and Omentum
134 Vascular Diseases of the Gastrointestinal Tract
135 Pancreatitis
136 Diseases of the Rectum and Anus
SECTION 14: DISEASES OF THE LIVER, GALLBLADDER, AND BILE DUCTS
137 Approach to the Patient with Liver Disease
138 Approach to the Patient with Jaundice or Abnormal Liver Tests
139 Acute Viral Hepatitis
140 Chronic Viral and Autoimmune Hepatitis
141 Toxin- and Drug-Induced Liver Disease
142 Bacterial, Parasitic, Fungal, and Granulomatous Liver Diseases
143 Alcoholic and Nonalcoholic Steatohepatitis
144 Cirrhosis and Its Sequelae
145 Hepatic Failure and Liver Transplantation
146 Diseases of the Gallbladder and Bile Ducts
SECTION 15: HEMATOLOGIC DISEASES
147 Hematopoiesis and Hematopoietic Growth Factors
148 The Peripheral Blood Smear
149 Approach to the Anemias
150 Microcytic and Hypochromic Anemias
151 Autoimmune and Intravascular Hemolytic Anemias
152 Hemolytic Anemias
153 The Thalassemias
154 Sickle Cell Disease and Other Hemoglobinopathies
155 Megaloblastic Anemias
156 Aplastic Anemia and Related Bone Marrow Failure States
157 Polycythemia Vera, Essential Thrombocythemia, and Primary Myelofibrosis
158 Leukocytosis and Leukopenia
159 Approach to the Patient with Lymphadenopathy and Splenomegaly
160 Histiocytoses
161 Eosinophilic Syndromes
162 Approach to the Patient with Bleeding and Thrombosis
163 Thrombocytopenia
164 Von Willebrand Disease and Hemorrhagic Abnormalities of Platelet and Vascular Function
165 Hemorrhagic Disorders
166 Hemorrhagic Disorders
167 Transfusion Medicine
168 Hematopoietic Stem Cell Transplantation
SECTION 16: ONCOLOGY
169 Approach to the Patient with Cancer
170 Epidemiology of Cancer
171 Cancer Biology and Genetics
172 Myelodysplastic Syndromes
173 The Acute Leukemias
174 Chronic Lymphocytic Leukemia
175 Chronic Myeloid Leukemia
176 Non-Hodgkin Lymphomas
177 Hodgkin Lymphoma
178 Plasma Cell Disorders
179 Amyloidosis
180 Tumors of the Central Nervous System
181 Head and Neck Cancer
182 Lung Cancer and Other Pulmonary Neoplasms
183 Neoplasms of the Esophagus and Stomach
184 Neoplasms of the Small and Large Intestine
185 Pancreatic Cancer
186 Liver and Biliary Tract Cancers
187 Tumors of the Kidney, Bladder, Ureters, and Renal Pelvis
188 Breast Cancer and Benign Breast Disorders
189 Gynecologic Cancers
190 Testicular Cancer
191 Prostate Cancer
192 Malignant Tumors of Bone, Sarcomas, and Other Soft Tissue Neoplasms
193 Melanoma and Nonmelanoma Skin Cancers
SECTION 17: METABOLIC DISEASES
194 Approach to Inborn Errors of Metabolism
195 Disorders of Lipid Metabolism
196 Glycogen Storage Diseases
197 Lysosomal Storage Diseases
198 Homocystinuria and Hyperhomocysteinemia
199 The Porphyrias
200 Wilson Disease
201 Iron Overload (Hemochromatosis)
SECTION 18: NUTRITIONAL DISEASES
202 Nutrition’s Interface with Health and Disease
203 Protein-Energy Malnutrition
204 Malnutrition
205 Vitamins, Trace Minerals, and Other Micronutrients
206 Eating Disorders
207 Obesity
SECTION 19: ENDOCRINE DISEASES
208 Approach to the Patient with Endocrine Disease
209 Principles of Endocrinology
210 Neuroendocrinology and the Neuroendocrine System
211 Anterior Pituitary
212 Posterior Pituitary
213 Thyroid
214 Adrenal Cortex
215 Adrenal Medulla, Catecholamines, and Pheochromocytoma
216 Diabetes Mellitus
217 Hypoglycemia and Pancreatic Islet Cell Disorders
218 Polyglandular Disorders
219 Neuroendocrine Tumors
220 Sexual Development and Identity
221 The Testis and Male Hypogonadism, Infertility, and Sexual Dysfunction
222 Ovaries and Pubertal Development
223 Reproductive Endocrinology and Infertility
SECTION 20: WOMEN'S HEALTH
224 Approach to Women’s Health
225 Contraception
226 Common Medical Problems in Pregnancy
227 Menopause
228 Intimate Partner Violence
SECTION 21: DISEASES OF BONE AND MINERAL METABOLISM
229 Approach to the Patient with Metabolic Bone Disease
230 Osteoporosis
231 Osteomalacia and Rickets
232 The Parathyroid Glands, Hypercalcemia, and Hypocalcemia
233 Paget Disease of Bone
234 Osteonecrosis, Osteosclerosis/Hyperostosis, and Other Disorders of Bone
SECTION 22: DISEASES OF ALLERGY AND CLINICAL IMMUNOLOGY
235 Approach to the Patient with Allergic or Immunologic Disease
236 Primary Immunodeficiency Diseases
237 Urticaria and Angioedema
238 Systemic Anaphylaxis, Food Allergy, and Insect Sting Allergy
239 Drug Allergy
240 Mastocytosis
SECTION 23: RHEUMATIC DISEASES
241 Approach to the Patient with Rheumatic Disease
242 Laboratory Testing in the Rheumatic Diseases
243 Imaging Studies in the Rheumatic Diseases
244 Inherited Diseases of Connective Tissue
245 The Systemic Autoinflammatory Diseases
246 Osteoarthritis
247 Bursitis, Tendinitis, and Other Periarticular Disorders and Sports Medicine
248 Rheumatoid Arthritis
249 The Spondyloarthropathies
250 Systemic Lupus Erythematosus
251 Systemic Sclerosis (Scleroderma)
252 Sjögren Syndrome
253 Inflammatory Myopathies
254 The Systemic Vasculitides
255 Giant Cell Arteritis and Polymyalgia Rheumatica
256 Infections of Bursae, Joints, and Bones
257 Crystal Deposition Diseases
258 Fibromyalgia, Chronic Fatigue Syndrome, and Myofascial Pain
259 Systemic Diseases in Which Arthritis Is a Feature
260 Surgical Treatment of Joint Diseases
SECTION 24: INFECTIOUS DISEASES
261 Introduction to Microbial Disease
262 The Human Microbiome
263 Principles of Anti-Infective Therapy
264 Approach to Fever or Suspected Infection in the Normal Host
265 Approach to Fever and Suspected Infection in the Immunocompromised Host
266 Prevention and Control of Health Care–Associated Infections
267 Approach to the Patient with Suspected Enteric Infection
268 Approach to the Patient with Urinary Tract Infection
269 Approach to the Patient with a Sexually Transmitted Infection
270 Approach to the Patient Before and After Travel
BACTERIA
271 Antibacterial Chemotherapy
272 Staphylococcal Infections
273 Streptococcus Pneumoniae Infections
274 Nonpneumococcal Streptococcal Infections and Rheumatic Fever
275 Enterococcal Infections
276 Diphtheria and Other Corynebacterium Infections
277 Listeriosis
278 Anthrax
279 Erysipelothrix Infections
280 Clostridial Infections
281 Diseases Caused by Non–Spore-Forming Anaerobic Bacteria
282 Neisseria Meningitidis Infections
283 Neisseria Gonorrhoeae Infections
284 Haemophilus and Moraxella Infections
285 Chancroid
286 Cholera and Other Vibrio Infections
287 Campylobacter Infections
288 Escherichia Coli Enteric Infections
289 Infections Due to Other Members of the Enterobacteriaceae, Including Management of Multidrug-Resistant Strains
290 Pseudomonas and Related Gram-Negative Bacillary Infections
291 Diseases Caused by Acinetobacter and Stenotrophomonas Species
292 Salmonella Infections (Including Enteric Fever)
293 Shigellosis
294 Brucellosis
295 Tularemia and Other Francisella Infections
296 Plague and Other Yersinia Infections
297 Whooping Cough and Other Bordetella Infections
298 Legionella Infections
299 Bartonella Infections
300 Granuloma Inguinale (Donovanosis)
301 Mycoplasma Infections
302 Diseases Caused by Chlamydiae
303 Syphilis
304 Nonsyphilitic Treponematoses
305 Lyme Disease
306 Relapsing Fever and Other Borrelia Infections
307 Leptospirosis
308 Tuberculosis
309 The Nontuberculous Mycobacteria
310 Leprosy (Hansen Disease)
311 Rickettsial Infections
312 Zoonoses
313 Actinomycosis
314 Nocardiosis
FUNGUS
315 Systemic Antifungal Agents
316 Endemic Mycoses
317 Cryptococcosis
318 Candidiasis
319 Aspergillosis
320 Mucormycosis
321 Pneumocystis Pneumonia
322 Mycetoma and Dematiaceous Fungal Infections
PARASITES
323 Antiparasitic Therapy
324 Malaria
325 African Sleeping Sickness
326 Chagas Disease
327 Leishmaniasis
328 Toxoplasmosis
329 Cryptosporidiosis
330 Giardiasis
331 Amebiasis
332 Babesiosis and Other Protozoan Diseases
333 Cestodes
334 Trematode Infections
335 Nematode Infections
VIRUS
336 Antiviral Therapy (Non-HIV)
337 The Common Cold
338 Respiratory Syncytial Virus
339 Parainfluenza Viral Disease
340 Influenza
341 Adenovirus Diseases
342 Coronaviruses
343 Measles
344 Rubella (German Measles)
345 Mumps
346 Cytomegalovirus, Epstein-Barr Virus, and Slow Virus Infections of the Central Nervous System
347 Parvovirus
348 Smallpox, Monkeypox, and Other Poxvirus Infections
349 Papillomavirus
350 Herpes Simplex Virus Infections
351 Varicella-Zoster Virus (Chickenpox, Shingles)
352 Cytomegalovirus
353 Epstein-Barr Virus Infection
354 Retroviruses Other Than Human Immunodeficiency Virus
355 Enteroviruses
356 Rotaviruses, Noroviruses, and Other Gastrointestinal Viruses
357 Viral Hemorrhagic Fevers
358 Arboviruses Causing Fever and Rash Syndromes
359 Arboviruses Affecting the Central Nervous System
SECTION 25: HIV AND THE ACQUIRED IMMUNODEFICIENCY SYNDROME
360 Epidemiology and Diagnosis of Human Immunodeficiency Virus Infection and Acquired Immunodeficiency Syndrome
361 Immunopathogenesis of Human Immunodeficiency Virus Infection
362 Biology of Human Immunodeficiency Viruses
363 Prevention of Human Immunodeficiency Virus Infection
364 Antiretroviral Therapy for Human Immunodeficiency Virus and Acquired Immunodeficiency Syndrome
365 Microbial Complications in Patients Infected With Human Immunodeficiency Virus
366 Systemic Manifestations of HIV/AIDS
367 Immune Reconstitution Inflammatory Syndrome in HIV/AIDS
SECTION 26: NEUROLOGY
368 Approach to the Patient with Neurologic Disease
369 Psychiatric Disorders in Medical Practice
370 Headaches and Other Head Pain
371 Traumatic Brain Injury and Spinal Cord Injury
372 Mechanical and Other Lesions of the Spine, Nerve Roots, and Spinal Cord
373 Regional Cerebral Dysfunction
374 Cognitive Impairment and Dementia
375 The Epilepsies
376 Coma, Vegetative State, and Brain Death
377 Sleep Disorders
378 Approach to Cerebrovascular Diseases
379 Ischemic Cerebrovascular Disease
380 Hemorrhagic Cerebrovascular Disease
381 Parkinsonism
382 Other Movement Disorders
383 Multiple Sclerosis and Demyelinating Conditions of the Central Nervous System
384 Meningitis
385 Brain Abscess and Parameningeal Infections
386 Acute Viral Encephalitis
387 Prion Diseases
388 Nutritional and Alcohol-Related Neurologic Disorders
389 Congenital, Developmental, and Neurocutaneous Disorders
390 Autonomic Disorders and Their Management
391 Amyotrophic Lateral Sclerosis and Other Motor Neuron Diseases
392 Peripheral Neuropathies
393 Muscle Diseases
394 Disorders of Neuromuscular Transmission
SECTION 27: EYE, EAR, NOSE, AND THROAT DISEASES
395 Diseases of the Visual System
396 Neuro-Ophthalmology
397 Diseases of the Mouth and Salivary Glands
398 Approach to the Patient with Nose, Sinus, and Ear Disorders
399 Smell and Taste
400 Hearing and Equilibrium
401 Throat Disorders
SECTION 28: MEDICAL CONSULTATION
402 Principles of Medical Consultation
403 Preoperative Evaluation
404 Overview of Anesthesia
405 Postoperative Care and Complications
406 Medical Consultation in Psychiatry
SECTION 29: SKIN DISEASES
407 Approach to Skin Diseases
408 Principles of Therapy of Skin Diseases
409 Eczemas, Photodermatoses, Papulosquamous (Including Fungal) Diseases, and Figurate Erythemas
410 Macular, Papular, Purpuric, Vesicobullous, and Pustular Diseases
411 Urticaria, Drug Hypersensitivity Rashes, Nodules and Tumors, and Atrophic Diseases
412 Infections, Hyperpigmentation and Hypopigmentation, Regional Dermatology, and Distinctive Lesions in Black Skin
413 Diseases of Hair and Nails
Appendix Reference Intervals and Laboratory Values
Endsheet 5

Citation preview

GOLDMAN-CECIL MEDICINE

Organized and Edited by: Min Thant Thaw

GOLDMAN-CECIL MEDICINE 26TH EDITION Volume 1

EDITED BY

LEE GOLDMAN, MD

Harold and Margaret Hatch Professor Chief Executive, Columbia University Irving Medical Center Dean of the Faculties of Health Sciences and Medicine Columbia University New York, New York

ANDREW I. SCHAFER, MD Professor of Medicine Director, Richard T. Silver Center for Myeloproliferative Neoplasms Weill Cornell Medical College New York, New York

Elsevier 1600 John F. Kennedy Blvd. Ste 1600 Philadelphia, PA 19103-2899

GOLDMAN-CECIL MEDICINE, TWENTY-SIXTH EDITION

INTERNATIONAL EDITION Copyright © 2020 by Elsevier, Inc. All rights reserved.

ISBN: 978-0-323-53266-2   Volume 1 ISBN: 978-0-323-76018-8   Volume 2 ISBN: 978-0-323-76019-5 ISBN: 978-0-323-64033-6   IE Volume 1 ISBN: 978-0-323-75998-4   IE Volume 2 ISBN: 978-0-323-75999-1

No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Chapter 104: Julian White retains copyright to his original figures/images appearing in the chapter. The following contributors are US government employees and their contributions are in public domain: David Atkins – Chapter 12 John O’Shea – Chapter 33 Leslie Biesecker – Chapter 36 Amy Klion – Chapter 161 Donna Krasnewich & Ellen Sidransky – Chapter 197 Lynnette Nieman – Chapter 208, 214, 218 Richard Siegel & Daniel Kastner – Chapter 245 Roland Sutter – Chapter 276 Paul Mead – Chapter 296 Joseph Kovacs – Chapter 321 Louis Kirchhoff – Chapter 326 Theodore Nash – Chapter 330 Neal Young – Chapter 347 Jeffrey Cohen – Chapter 351

Notice Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds or experiments described herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. To the fullest extent of the law, no responsibility is assumed by Elsevier, authors, editors or contributors for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Previous editions copyrighted 2016, 2012, 2008, 2004, 2000, 1996, 1991, 1988, 1982, 1979, 1975, 1971, 1963, 1959, 1955, 1951, 1947, 1943, 1940, 1937, 1933, 1930, 1927 by Saunders, an imprint of Elsevier Inc. Copyright renewed 1991 by Paul Beeson. Copyright renewed 1979 by Russell L. Cecil and Robert F. Loeb. Copyright renewed 1987, 1975, 1971, 1965, 1961, 1958, 1955 by Elsevier Inc. Library of Congress Control Number: 2019909209

Publishing Director, Medical Reference: Dolores Meloni Content Development Manager: Laura Schmidt Publishing Services Manager: Catherine Jackson Senior Project Manager: Daniel Fitzgerald Designer: Maggie Reid Printed in the United States of America Last digit is the print number: 9 8 7 6 5 4 3 2 1

ASSOCIATE EDITORS Mary K. Crow, MD

Joseph P. Routh Professor of Rheumatic Diseases in Medicine Weill Cornell Medical College Physician-in-Chief and Benjamin M. Rosen Chair in Immunology and Inflammation Research Hospital for Special Surgery New York, New York

Nancy E. Davidson, MD

Professor of Medicine and Raisbeck Endowed Chair President Seattle Cancer Care Alliance Senior Vice President and Director, Clinical Research Division Fred Hutchinson Cancer Research Center Chief Division of Medical Oncology University of Washington School of Medicine Seattle, Washington

Jeffrey M. Drazen, MD

Distinguished Parker B. Francis Professor of Medicine Harvard Medical School Senior Physician Department of Medicine Brigham and Women’s Hospital Boston, Massachusetts

Robert C. Griggs, MD

Professor of Neurology, Medicine, Pediatrics, Pathology & Laboratory Medicine University of Rochester School of Medicine & Dentistry Rochester, New York

Donald W. Landry, MD, PhD

Samuel Bard Professor and Chair Department of Medicine Columbia University Vagelos College of Physicians and Surgeons Physician-in-Chief Columbia University Irving Medical Center New York, New York

Wendy Levinson, MD Professor of Medicine Chair Emeritus Department of Medicine University of Toronto Toronto, Ontario, Canada

Anil K. Rustgi, MD

Irving Professor of Medicine Director Herbert Irving Comprehensive Cancer Center Chief NewYork-Presbyterian Hospital/Columbia University Irving Medical Center Cancer Service Columbia University Vagelos College of Physicians and Surgeons New York, New York

W. Michael Scheld, MD

Bayer-Gerald L. Mandell Professor of Infectious Diseases Professor of Medicine Clinical Professor of Neurosurgery David A. Harrison Distinguished Educator University of Virginia Health System Charlottesville, Virginia

Allen M. Spiegel, MD

Dean Emeritus Professor of Medicine Albert Einstein College of Medicine Bronx, New York

PREFACE In the more than 90 years since the first edition of the Cecil Textbook of Medicine was published, almost everything we know about internal medicine has changed. Progress in medical science is now occurring at an ever-accelerating pace, and it is doing so within the framework of transformational changes in clinical practice and the delivery of health care at individual, social, and global levels. This textbook and its associated electronic products incorporate the latest medical knowledge in multiple formats that should appeal to students and seasoned practitioners regardless of how they prefer to access this rapidly changing information. Even as Cecil’s specific information has changed, however, we have remained true to the tradition of a comprehensive textbook of medicine that carefully explains the why (the underlying genetics, genomics, and pathobiology of disease) and the how (now expected to be evidence-based from randomized controlled trials and meta-analyses). Descriptions of physiology and pathophysiology include the latest genetic advances in a practical format that strives to be useful to the nonexpert so that care can truly be as precise and personalized as possible. Medicine has entered an era when the acuity of illness and the limited time available to evaluate a patient have diminished the ability of physicians to satisfy their intellectual curiosity. As a result, the acquisition of information, quite easily achieved in this era, is often confused with knowledge. We have attempted to address this dilemma with a textbook that not only informs but also stimulates new questions and gives a glimpse of the future path to new knowledge. Grade A evidence is specifically highlighted in the text and referenced at the end of each chapter. In addition to the information provided in the textbook, the Cecil website supplies expanded content and functionality. In many cases, the full articles referenced in each chapter can be accessed from the Cecil website. The website is also continuously updated to incorporate subsequent Grade A information, other evidence, and new discoveries. The sections for each organ system begin with a chapter that summarizes an approach to patients with key symptoms, signs, or laboratory abnormalities associated with dysfunction of that organ system. As summarized in E-Table 1-1, the text specifically provides clear, concise information regarding how a physician should approach more than 100 common symptoms, signs, and laboratory abnormalities, usually with a flow diagram, a table, or both for easy reference. In this way, Cecil remains a comprehensive text to guide diagnosis and therapy, not only for patients with suspected or known diseases but also for patients who may have undiagnosed symptoms or signs that require an initial evaluation. Just as each edition brings new authors, it also reminds us of our gratitude to past editors and authors. Previous editors of Cecil include a short but remarkably distinguished group of leaders of American medicine: Russell Cecil, Paul Beeson, Walsh McDermott, James Wyngaarden, Lloyd H. Smith, Jr., Fred Plum, J. Claude Bennett, and Dennis Ausiello. As we welcome a new associate editor—Nancy Davidson—we also express our appreciation to James Doroshow and other associate editors from the previous editions on whose foundation we have built. Our returning associate editors—Mary K. Crow, Jeffrey M. Drazen, Robert C. Griggs, Donald W. Landry, Wendy Levinson, Anil Rustgi, W. Michael Scheld, and Allen M. Spiegel—continue to make critical contributions to the selection of authors and the review and approval of all manuscripts. The editors, however, are fully responsible for the book as well as the integration among chapters. The tradition of Cecil is that all chapters are written by distinguished experts in each field. Two of those authors, Frank A. Lederle, author of the chapter on “Diseases of the Aorta,” and Ronald Victor, author of the chapter on “Arterial Hypertension,” passed away after submitting their chapters, and we mourn their passing. We are also most grateful for the editorial assistance in New York of Timothy Gahr, Maribel Lim, Eva Allen, and Magdalena Fuentes. These individuals and

others in our offices have shown extraordinary dedication and equanimity in working with authors and editors to manage the unending flow of manuscripts, figures, and permissions. This edition of Goldman-Cecil Medicine includes many new authors. We would also like to thank outgoing authors, who often provided figures that are included in this edition as well as tables that have been included or modified for this edition. Furthermore, because of the templated format and extensive editing that are characteristic of Goldman-Cecil Medicine, some new chapters incorporate principles, concepts, and organizational aspects from those prior chapters, often revised extensively prior to publication. Among prior authors who deserve our appreciation, in the numerical order of their chapters, are Victoria M. Taylor, Steven A. Schroeder, Thomas B. Newman, Charles E. McCulloch, Thomas H. Lee, F. Daniel Duffy, Lawrence S. Neinstein, Steven E. Hyman, Grant W. Cannon, Cem Gabay, Carlo Patrono, Jack Hirsh, Adam Perlman, Sandesh C.S. Nagamani, Paweł Stankiewicz, James R. Lupski, Sekar Kathiresan, David Altshuler, Göran K. Hansson, Anders Hamsten, L. David Hillis, Bruce W. Lytle, William C. Little, Donna Mancini, Yoshifumi Naka, Dennis E. Niewoehner, Frank J. Accurso, Emanuel P. Rivers, Marsha D. Ford, Geoffrey K. Isbister, Itzchak Slotki, Mark L. Zeidel, David H. Kim, Perry J. Pickhardt, Martin J. Blaser, Stephen Crane Hauser, H. Franklin Bunn, Gordon D. Ginder, Martin H. Steinberg, Aśok C. Antony, Ayalew Tefferi, Michael Glogauer, Marc E. Rothenberg, William L. Nichols, Lawrence T. Goodnough, Adrian R. Black, Kenneth H. Cowan, Susan O’Brien, Elias Jabbour, Marshall R. Posner, Charles D. Blanke, Douglas O. Faigel, David Spriggs, John D. Hainsworth, F. Anthony Greco, Clay F. Semenkovich, Stephen G. Kaler, Bruce R. Bacon, Bruce R. Bistrian, Stephen A. McClave, Mark E. Molitch, Matthew Kim, Paul W. Ladenson, Kenneth R. Hande, Robert W. Rebar, Deborah Grady, Elizabeth Barrett-Connor, Samuel A. Wells, Jr., Stephen I. Wasserman, Larry Borish, Suneel S. Apte, Joel A. Block, Carla Scanzello, Robert M. Bennett, Ilseung Cho, S. Ragnar Norrby, Lionel A. Mandell, Donald E. Low, Kenneth L. Gage, Atis Muehlenbachs, Stuart Levin, Kamaljit Singh, Richard L. Guerrant, Dirk M. Elston, Larry J. Anderson, Martin Weisse, Mark Papania, Letha M. Healey, Tamsin A. Knox, Christine Wanke, Kristina Crothers, Alison Morris, Toby A. Maurer, Thomas S. Uldrick, Robert Yarchoan, Robert Colebunders, Ralph F. Józefowicz, Michael Aminoff, Eelco F.M. Wijdicks, Myron Yanoff, Douglas Cameron, David H. Chu, James C. Shaw, Neil J. Korman, and Ronald J. Elin. We also thank Michael G. House, who contributed to the chapter on “Diseases of the Gallbladder and Bile Ducts,” and Anna Louise Beavis, who contributed to the chapter on “Gynecologic Cancers.” Chapters written by public employees reflect recommendations and conclusions of the authors and do not necessarily reflect the official position of the entity for which they work. At Elsevier, we are most indebted to Dolores Meloni and Laura Schmidt, and also thank Lucia Gunzel, Dan Fitzgerald, and Maggie Reid, who have been critical to the planning and production. We have been exposed to remarkable physicians in our lifetimes and would like to acknowledge the mentorship and support of several of those who exemplify this paradigm—Eugene Braunwald, the late Lloyd H. Smith, Jr., Frank Gardner, and William Castle. Finally, we would like to thank the Goldman family—Jill, Jeff, Abigail, Mira, Samuel, Daniel, Morgan, Robyn, Tobin, Dashel, and Alden—and the Schafer family—Pauline, Eric, Melissa, Nathaniel, Caroline, Pam, John, Evan, Samantha, Kate, Sean, Patrick, and Meghan—for their understanding of the time and focus required to edit a book that attempts to sustain the tradition of our predecessors and to meet the needs of today’s physician. LEE GOLDMAN, MD ANDREW I. SCHAFER, MD

CONTRIBUTORS Charles S. Abrams, MD Francis C. Wood Professor of Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania Thrombocytopenia

Deborah K. Armstrong, MD Professor of Gynecology and Obstetrics, Johns Hopkins University School of Medicine, Baltimore, Maryland Gynecologic Cancers

Ronald S. Adler, MD, PhD Professor of Radiology, New York University School of Medicine; NYU Langone Health, New York, New York Imaging Studies in the Rheumatic Diseases

M. Amin Arnaout, MD Professor of Medicine, Chief Emeritus, Division of Nephrology, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts Cystic Kidney Diseases

Cem Akin, MD, PhD Professor of Medicine, Internal Medicine, University of Michigan Medical School, Ann Arbor, Michigan Mastocytosis Allen J. Aksamit, Jr., MD Professor of Neurology, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Acute Viral Encephalitis Qais Al-Awqati, MB ChB Robert F. Loeb Professor, Medicine, and Physiology & Cellular Biophysics, Columbia University Vagelos College of Physicians & Surgeons, New York, New York Structure and Function of the Kidneys; Disorders of Sodium and Water Ban Mishu Allos, MD Associate Professor of Medicine, Division of Infectious Diseases, Vanderbilt University School of Medicine, Nashville, Tennessee Campylobacter Infections Jeffrey L. Anderson, MD Professor of Medicine, Division of Cardiovascular Medicine, University of Utah School of Medicine; Distinguished Clinical and Research Physician, Intermountain Medical Center Heart Institute, Salt Lake City, Utah ST Elevation Acute Myocardial Infarction and Complications of Myocardial Infarction Derek C. Angus, MD, MPH Professor and Mitchell P. Fink Endowed Chair, Department of Critical Care Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania Approach to the Patient with Shock Gerald B. Appel, MD Professor of Medicine and Director, Glomerular Center, Columbia University Irving Medical Center, New York, New York Glomerular Disorders and Nephrotic Syndromes Frederick R. Appelbaum, MD Professor of Medicine, University of Washington School of Medicine; Executive Senior VP and Deputy Director, Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, Washington The Acute Leukemias James O. Armitage, MD Professor of Internal Medicine, University of Nebraska Medical Center College of Medicine, Omaha, Nebraska Non-Hodgkin Lymphomas

Robert M. Arnold, MD Distinguished Service Professor, Chief, Section of Palliative Care and Medical Ethics, University of Pittsburgh School of Medicine; Chief Medical Officer, UPMC Palliative and Supportive Institute, UPMC Health Plan, Pittsburgh, Pennsylvania Palliative Care David Atkins, MD, MPH Director, Health Services Research and Development, Office of Research and Development, Dept. of Veterans Affairs (10P9H), Washington, D.C. The Periodic Health Examination John P. Atkinson, MD Professor of Medicine, Division of Rheumatology, Washington University School of Medicine in St. Louis, St. Louis, Missouri Complement System in Disease John Z. Ayanian, MD, MPP Alice Hamilton Professor of Medicine; Director, Institute for Healthcare Policy and Innovation, University of Michigan Medical School, Ann Arbor, Michigan Disparities in Health and Health Care Larry M. Baddour, MD Professor of Medicine, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Infective Endocarditis Grover C. Bagby, MD Professor of Medicine, Molecular and Medical Genetics, Oregon Health & Science University, Portland, Oregon Aplastic Anemia and Related Bone Marrow Failure States Barbara J. Bain, MBBS Professor in Diagnostic Haematology, Haematology, St Mary’s Hospital Campus of Imperial College London, London, United Kingdom The Peripheral Blood Smear Dean F. Bajorin, MD Attending Physician and Member, Memorial Sloan Kettering Cancer Center; Professor of Medicine, Weill Cornell Medical College, New York, New York Tumors of the Kidney, Bladder, Ureters, and Renal Pelvis Robert W. Baloh, MD Professor of Neurology, David Geffen School of Medicine at UCLA, Los Angeles, California Neuro-Ophthalmology; Smell and Taste; Hearing and Equilibrium

viii

Contributors

Charles R.M. Bangham, BM BCh Professor of Medicine, Faculty of Medicine, Imperial College London School of Medicine, London, United Kingdom Retroviruses Other Than Human Immunodeficiency Virus

Hasan Bazari, MD Associate Professor of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts Approach to the Patient with Renal Disease

Jonathan Barasch, MD, PhD Samuel W Lambert Professor of Medicine, Professor of Pathology and Cell Biology, Columbia University Vagelos College of Physicians & Surgeons, New York, New York Structure and Function of the Kidneys

Jeffrey J. Bazarian, MD, MPH Professor of Emergency Medicine, University of Rochester School of Medicine & Dentistry, Rochester, New York Traumatic Brain Injury and Spinal Cord Injury

Richard L. Barbano, MD, PhD Professor of Neurology and Chief of the Movement Disorders Division, University of Rochester School of Medicine & Dentistry, Rochester, New York Mechanical and Other Lesions of the Spine, Nerve Roots, and Spinal Cord Bruce Barrett, MD, PhD Professor, Department of Family Medicine and Community Health, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin The Common Cold John R. Bartholomew, MD Professor of Medicine and Section Head Vascular Medicine, Cleveland Clinic Lerner College of Medicine, Cleveland, Ohio Other Peripheral Arterial Diseases

John H. Beigel, MD Associate Director for Clinical Research, Division of Microbiology and Infectious Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland Antiviral Therapy (Non-HIV) Elisabeth H. Bel, MD, PhD Professor and Head of the Department of Respiratory Medicine, Amsterdam University Medical Center, University of Amsterdam, The Netherlands Asthma George A. Beller, MD Emeritus Professor of Cardiology, Department of Medicine, University of Virginia Health System, Charlottesville, Virginia Noninvasive Cardiac Imaging

J.D. Bartleson, MD Professor of Neurology, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Mechanical and Other Lesions of the Spine, Nerve Roots, and Spinal Cord

Joseph R. Berger, MD Professor of Neurology, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania Cytomegalovirus, Epstein-Barr Virus, and Slow Virus Infections of the Central Nervous System; Brain Abscess and Parameningeal Infections

Mary Barton, MD, MPP Vice President, Performance Measurement, National Committee for Quality Assurance, Washington, D.C. The Periodic Health Examination

Paul D. Berk, MD Professor of Medicine, Columbia University Vagelos College of Physicians & Surgeons, New York, New York Approach to the Patient with Jaundice or Abnormal Liver Tests

Robert C. Basner, MD Professor of Medicine, Columbia University Vagelos College of Physicians and Surgeons, New York, New York Sleep Disorders

Nancy Berliner, MD H. Franklin Bunn Professor of Medicine; Chief, Division of Hematology, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts Leukocytosis and Leukopenia; Histiocytoses

Anne R. Bass, MD Professor of Clinical Medicine, Weill Cornell Medical College; Attending Physician, Hospital for Special Surgery, New York, New York Immunomodulatory Drugs Stephen G. Baum, MD Professor of Medicine and of Microbiology and Immunology, Albert Einstein College of Medicine, Bronx, New York Mycoplasma Infections Julie E. Bauman, MD, MPH Professor of Medicine, University of Arizona Cancer Center, Tucson, Arizona Head and Neck Cancer Daniel G. Bausch, MD, MPH&TM Director, United Kingdom Public Health Rapid Support Team, Public Health England/London School of Hygiene and Tropical Medicine, London, United Kingdom Viral Hemorrhagic Fevers Arnold S. Bayer, MD Distinguished Professor of Medicine, David Geffen School of Medicine at UCLA; Senior Investigator-LA Biomedical Research Institute At Harbor-UCLA, Los Angeles, California Infective Endocarditis

James L. Bernat, MD Professor of Neurology and Medicine, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire and Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire Coma, Vegetative State, and Brain Death Philip J. Bierman, MD Professor of Internal Medicine, University of Nebraska Medical Center College of Medicine, Omaha, Nebraska Non-Hodgkin Lymphomas Leslie G. Biesecker, MD Chief, Medical Genomics and Metabolic Genetics Branch, National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland Clinical Genomics—Genome Structure and Variation Michael R. Bishop, MD Professor of Medicine and Director of the Cellular Therapy Program, Section of Hematology and Oncology, University of Chicago Pritzker School of Medicine, Chicago, Illinois Hematopoietic Stem Cell Transplantation Joseph J. Biundo, MD Clinical Professor of Medicine, Tulane Medical Center, New Orleans, Louisiana Bursitis, Tendinitis, and Other Periarticular Disorders and Sports Medicine

Contributors Joel N. Blankson, MD, PhD Professor of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland Immunopathogenesis of Human Immunodeficiency Virus Infection

Lucy Breakwell, PhD, MSc Epidemiologist, Global Immunization Division, Centers for Disease Control and Prevention, Atlanta, Georgia Diphtheria and Other Corynebacterium Infections

William A. Blattner, MD Chief Executive Officer, Salt Run Global Health and Research, Saint Augustine, Florida Retroviruses Other Than Human Immunodeficiency Virus

David J. Brenner, PhD, DSc Higgins Professor of Radiation Biophysics, Center for Radiological Research, Columbia University Irving Medical Center, New York, New York Radiation Injury

Thomas P. Bleck, MD Professor of Neurology, Northwestern University Feinberg School of Medicine; Professor Emeritus of Neurological Sciences, Neurosurgery, Medicine, and Anesthesiology, Rush Medical College, Chicago, Illinois Arboviruses Affecting the Central Nervous System Karen C. Bloch, MD, MPH Associate Professor of Medicine (Infectious Diseases) and Health Policy, Vanderbilt University School of Medicine, Nashville, Tennessee Tularemia and Other Francisella Infections Henk J. Blom, PhD Professor of Biochemistry of Inherited Metabolic Disease, Department of Clinical Genetics, Center for Lysosomal and Metabolic Diseases, Erasmus MC, Rotterdam, The Netherlands Homocystinuria and Hyperhomocysteinemia Olaf A. Bodamer, MD, PhD Park Gerald Chair of Genetics and Genomics, Department of Medicine, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts Approach to Inborn Errors of Metabolism William E. Boden, MD Professor of Medicine, Boston University School of Medicine; Lecturer in Medicine, Harvard Medical School; Scientific Director, Clinical Trials Network, Department of Medicine, VA Boston Healthcare System, Boston, Massachusetts Angina Pectoris and Stable Ischemic Heart Disease Guy Boivin, MD Professor of Microbiology, Immunology and Infectiology, CHU de Québec-Laval University, Quebec City, Quebec, Canada Cytomegalovirus Jean Bolognia, MD Professor of Dermatology, Yale University School of Medicine, New Haven, Connecticut Infections, Hyperpigmentation and Hypopigmentation, Regional Dermatology, and Distinctive Lesions in Black Skin William Bonnez, MD Professor Emeritus of Medicine, University of Rochester School of Medicine & Dentistry, Rochester, New York Papillomavirus Robert A. Bonomo, MD Professor of Medicine, Case Western Reserve University School of Medicine; Chief of Medicine, Cleveland VA Hospital, Cleveland, Ohio Diseases Caused by Acinetobacter and Stenotrophomonas Species Sarah L. Booth, PhD Professor of Nutrition, Tufts University; Director, USDA Human Nutrition Research Center on Aging; Director, Vitamin K Laboratory, USDA Human Nutrition Research Center on Aging, Boston, Massachusetts Vitamins, Trace Minerals, and Other Micronutrients Patrick J. Bosque, MD Associate Professor of Neurology, University of Colorado School of Medicine; Chief, Neurology Division, Department of Medicine, Denver Health Medical Center, Denver, Colorado Prion Diseases

ix

Laurent Brochard, MD Keenan Chair in Critical Care and Respiratory Medicine and Professor of Medicine and Interdepartmental Division Director for Critical Care, University of Toronto Faculty of Medicine; Division of Critical Care, Saint Michael’s Hospital, Toronto, Ontario, Canada Mechanical Ventilation Itzhak Brook, MD Professor of Pediatrics, Georgetown University School of Medicine, Washington, D.C. Diseases Caused by Non–Spore-Forming Anaerobic Bacteria; Actinomycosis Enrico Brunetti, MD Associate Professor, Department of Clinical, Surgical, Diagnostic and Pediatric Sciences and Staff Physician, Department of Infectious and Tropical Diseases, San Matteo Hospital Foundation, University of Pavia, Pavia, Italy Cestodes Amy E. Bryant, PhD Associate Professor of Medicine, University of Washington School of Medicine, Seattle, Washington and Research Career Scientist, Infectious Diseases Section, VA Medical Center, Boise, Idaho Nonpneumococcal Streptococcal Infections and Rheumatic Fever David M. Buchner, MD, MPH Professor Emeritus, Department of Kinesiology & Community Health, University of Illinois Urbana Champaign, Champaign, Illinois Physical Activity Pierre A. Buffet, MD, PhD Professor of Cell Biology, Faculty of Medicine, Paris University and Consultant Physician, Institut Pasteur Medical Center, Paris, France Leishmaniasis David A. Bushinsky, MD John J. Kuiper Distinguished Professor of Medicine and of Pharmacology and Physiology, University of Rochester School of Medicine & Dentistry, Rochester, New York Nephrolithiasis Vivian P. Bykerk, MD Associate Professor of Medicine, Weill Cornell Medical College; Associate Attending Physician, Hospital for Special Surgery, New York, New York Approach to the Patient with Rheumatic Disease John C. Byrd, MD Distinguished University Professor, Ohio State University, Columbus, Ohio Chronic Lymphocytic Leukemia Peter A. Calabresi, MD Professor of Neurology and Neuroscience, Director of the Richard T Johnson Division of Neuroimmunology and Neuroinfectious Diseases; Director of the Multiple Sclerosis Center, Johns Hopkins University School of Medicine, Baltimore, Maryland Multiple Sclerosis and Demyelinating Conditions of the Central Nervous System

x

Contributors

David P. Calfee, MD, MS Professor of Medicine and of Health Policy & Research, Weill Cornell Medical College; Chief Hospital Epidemiologist, NewYork-Presbyterian Hospital/Weill Cornell, New York, New York Prevention and Control of Health Care–Associated Infections Clara Camaschella, MD Professor of Medicine, Division of Genetics and Cell Biology, San Raffaele Scientific Institute, Milano, Italy Microcytic and Hypochromic Anemias Michael Camilleri, MD Atherton and Winifred W. Bean Professor of Medicine, Pharmacology, and Physiology, Mayo Clinic College of Medicine and Science; Consultant, Division of Gastroenterology and Hepatology, Department of Medicine, Mayo Clinic, Rochester, Minnesota Disorders of Gastrointestinal Motility Maria Domenica Cappellini, MD Professor of Internal Medicine, Department of Clinical Sciences and Community Health, University of Milan; and Ca’ Granda Foundation-Policlinico Hospital, Milan, Italy The Thalassemias

Lin H. Chen, MD Associate Professor of Medicine, Harvard Medical School, Boston, Massachusetts and Director of the Travel Medicine Center, Division of Infectious Diseases and Travel Medicine, Mount Auburn Hospital, Cambridge, Massachusetts Approach to the Patient before and after Travel Sharon C-A Chen, MB, PhD Professor of Medicine, University of Sydney and Centre for Infectious Diseases and Microbiology, ICPMR and Westmead Hospital, New South Wales, Australia Cryptococcosis William P. Cheshire, Jr., MD Professor of Neurology, Mayo Clinic College of Medicine and Science, Jacksonville, Florida Autonomic Disorders and Their Management Arun Chockalingam, PhD Professor of Epidemiology, Medicine and Global Health, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada Global Health

Blase A. Carabello, MD Professor of Cardiovascular Sciences and Chief, Division of Cardiology, East Carolina University Brody School of Medicine, Greenville, North Carolina Valvular Heart Disease

David C. Christiani, MD Professor of Medicine, Harvard Medical School; Physician, Pulmonary and Critical Care, Massachusetts General Hospital; Elkan Blout Professor of Environmental Genetics, Environmental Health, Harvard School of Public Health, Boston, Massachusetts Physical and Chemical Injuries of the Lung

Edgar M. Carvalho, MD, PhD Professor of Medicine, Federal University of Bahia, Oswaldo Cruz Foundation (Fiocruz), Instituto de Pesquisa Gonçalo Moniz (IGM), Salvador-Bahia, Brazil Trematode Infections

Edward Chu, MD, MMS Professor and Chief, Division of Hematology-Oncology, Deputy Director, UPMC Hillman Cancer Center, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania Neoplasms of the Small and Large Intestine

William H. Catherino, MD, PhD Professor and Chair-Research Division, Department of Obstetrics and Gynecology, Uniformed Services University of the Health Sciences, Bethesda, Maryland Ovaries and Pubertal Development; Reproductive Endocrinology and Infertility

Theodore J. Cieslak, MD, MPH Associate Professor of Epidemiology, Co-Medical Director, Nebraska Biocontainment Unit, College of Public Health, University of Nebraska, Omaha, Nebraska Bioterrorism

Jane A. Cauley, DrPH Distinguished Professor of Epidemiology, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania Epidemiology of Aging: Implications of an Aging Society

George A. Cioffi, MD Edward S. Harkness Professor and Chair, Jean and Richard Deems Professor of Ophthalmology, Columbia University Vagelos College of Physicians and Surgeons, New York, New York Diseases of the Visual System

Naga P. Chalasani, MD David W. Crabb Professor and Director, Division of Gastroenterology and Hepatology, Indiana University School of Medicine, Indianapolis, Indiana Alcoholic and Nonalcoholic Steatohepatitis Henry F. Chambers, MD Professor of Medicine and Director, Clinical Research Services, Clinical Translational Science Institute, University of California, San Francisco, School of Medicine, San Francisco, California Staphylococcal Infections Larry W. Chang, MD, MPH Associate Professor of Medicine, Epidemiology, and International Health, Johns Hopkins University School of Medicine and Bloomberg School of Public Health, Baltimore, Maryland Epidemiology and Diagnosis of Human Immunodeficiency Virus Infection and Acquired Immunodeficiency Syndrome

Carolyn M. Clancy, MD Clinical Associate Professor of Internal Medicine, George Washington University School of Medicine; Assistant Deputy Undersecretary for Health, Quality, Safety and Value, Veterans Administration, Washington, D.C. Measuring Health and Health Care Heather E. Clauss, MD Associate Professor of Medicine, Section of Infectious Diseases, Lewis Katz School of Medicine at Temple University, Philadelphia, Pennsylvania Listeriosis Daniel J. Clauw, MD Professor of Anesthesiology, Medicine (Rheumatology) and Psychiatry, Director, Chronic Pain and Fatigue Research Center, University of Michigan Medical School, Ann Arbor, Michigan Fibromyalgia, Chronic Fatigue Syndrome, and Myofascial Pain David R. Clemmons, MD Kenan Professor of Medicine, University of North Carolina School of Medicine; Attending Physician, Medicine, UNC Hospitals, Chapel Hill, North Carolina Approach to the Patient with Endocrine Disease

Contributors David Cohen, MD Professor of Medicine, Columbia University Vagelos College of Physicians and Surgeons, New York, New York Treatment of Irreversible Renal Failure Jeffrey Cohen, MD Chief, Laboratory of Infectious Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland Varicella-Zoster Virus (Chickenpox, Shingles) Myron S. Cohen, MD Yeargan-Bates Eminent Professor of Medicine, Microbiology and Epidemiology, Associate Vice Chancellor for Global Health; Director, Institute of Global Health and Infectious Diseases, University of North Carolina School of Medicine, Chapel Hill, North Carolina Approach to the Patient with a Sexually Transmitted Infection; Prevention of Human Immunodeficiency Virus Infection Steven P. Cohen, MD Professor of Anesthesiology & Critical Care Medicine, Neurology and Physical Medicine & Rehabilitation and Chief, Pain Medicine Division, Johns Hopkins School of Medicine; Director of Pain Research and Professor of Anesthesiology and Physical Medicine & Rehabilitation, Walter Reed National Military Medical Center, Uniformed Services University of the Health Sciences, Baltimore, Maryland Pain Steven L. Cohn, MD Professor Emeritus, Department of Medicine, University of Miami Miller School of Medicine, Miami, Florida; Clinical Professor of Medicine Emeritus, SUNY Downstate, Brooklyn, New York Preoperative Evaluation Joseph M. Connors, MD Emeritus Professor, BC Cancer Centre for Lymphoid Cancer and the University of British Columbia, Vancouver, British Columbia, Canada Hodgkin Lymphoma Deborah J. Cook, MD, MSc Professor of Medicine, Clinical Epidemiology & Biostatistics, McMaster University Michael G. DeGroote School of Medicine, Hamilton, Ontario, Canada Approach to the Patient in a Critical Care Setting David S. Cooper, MD Professor of Medicine, Division of Endocrinology and Metabolism, The Johns Hopkins University School of Medicine, Baltimore, Maryland Thyroid Joseph Craft, MD Paul B. Beeson Professor of Medicine and Professor of Immunobiology, Departments of Internal Medicine and Immunobiology, Yale University, New Haven, Connecticut The Adaptive Immune System Jill P. Crandall, MD Professor of Medicine and Chief, Division of Endocrinology, Albert Einstein College of Medicine, Bronx, New York Diabetes Mellitus Simon L. Croft, PhD Professor of Parasitology, Faculty of Infectious and Tropical Diseases, London School of Hygiene & Tropical Medicine, London, United Kingdom Leishmaniasis

xi

Mary K. Crow, MD Joseph P. Routh Professor of Rheumatic Diseases in Medicine, Weill Cornell Medical College; Physician-in-Chief and Benjamin M. Rosen Chair in Immunology and Inflammation Research, Hospital for Special Surgery, New York, New York The Innate Immune System; Approach to the Patient with Rheumatic Disease; Systemic Lupus Erythematosus John A. Crump, MB ChB, MD, DTM&H McKinlay Professor of Global Health, Centre for International Health, University of Otago, Dunedin, Otago; Adjunct Professor of Medicine, Pathology, and Global Health, Division of Infectious Diseases and International Health, Duke University Medical Center, Durham, North Carolina Salmonella Infections (Including Enteric Fever) Merit E. Cudkowicz, MD Professor of Neurology, Harvard Medical School and Chair of Neurology, Massachusetts General Hospital, Boston, Massachusetts Amyotrophic Lateral Sclerosis and Other Motor Neuron Diseases Mark R. Cullen, MD Professor of Medicine, Stanford University School of Medicine, Stanford, California Principles of Occupational and Environmental Medicine Charlotte Cunningham-Rundles, MD, PhD David S Gottesman Professor of Medicine, Icahn School of Medicine at Mount Sinai, New York, New York Primary Immunodeficiency Diseases Inger K. Damon, MD, PhD Director, Division of High Consequence Pathogens and Pathology, Centers for Disease Control and Prevention, Atlanta, Georgia Smallpox, Monkeypox, and Other Poxvirus Infections Troy E. Daniels, DDS, MS Professor Emeritus of Oral Pathology & Pathology, University of California, San Francisco, School of Medicine, San Francisco, California Diseases of the Mouth and Salivary Glands Richard Dart, MD, PhD Professor of Emergency Medicine, University of Colorado School of Medicine and Director, Rocky Mountain Poison and Drug Center, Denver Health and Hospital Authority, Denver, Colorado Envenomation, Bites, and Stings Nancy E. Davidson, MD Professor of Medicine and Raisbeck Endowed Chair; President, Seattle Cancer Care Alliance; Senior Vice President and Director, Clinical Research Division, Fred Hutchinson Cancer Research Center; Chief, Division of Medical Oncology, University of Washington School of Medicine, Seattle, Washington Breast Cancer and Benign Breast Disorders Lisa M. DeAngelis, MD Lillian Rojtman Chair in Honor of Jerome B Posner, Acting Physician-inChief, Memorial Hospital, Chair, Department of Neurology, Memorial Sloan-Kettering Cancer Center, New York, New York Tumors of the Central Nervous System Malcolm M. DeCamp, MD Professor of Surgery and Chair, Division of Cardiothoracic Surgery, K. Craig Kent Chair in Strategic Leadership, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin Interventional and Surgical Approaches to Lung Disease

xii

Contributors

Carlos Del Rio, MD Hubert Professor and Chair, Hubert Department of Global Health, Rollins School of Public Health of Emory University; Professor, Department of Medicine, Emory University School of Medicine, Atlanta, Georgia Prevention of Human Immunodeficiency Virus Infection Gabriele C. DeLuca, MD, DPhil Associate Professor, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, Oxfordshire, United Kingdom Approach to the Patient with Neurologic Disease David W. Denning, MBBS Professor of Infectious Diseases in Global Health and Director of the National Aspergillosis Centre, University of Manchester and Wythenshawe Hospital, Manchester, United Kingdom Systemic Antifungal Agents Patricia A. Deuster, PhD, MPH Professor and Director, Department of Military and Emergency Medicine, Director, Consortium for Health and Military Performance, Uniformed Services University, Bethesda, Maryland Rhabdomyolysis Robert B. Diasio, MD William J and Charles H Mayo Professor of Molecular Pharmacology and Experimental Therapeutics, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Principles of Drug Therapy David J. Diemert, MD Associate Professor, Departments of Medicine and Microbiology, Immunology and Tropical Medicine, George Washington University School of Medicine and Health Sciences, Washington, D.C. Nematode Infections Kathleen B. Digre, MD Professor of Neurology and Ophthalmology, University of Utah School of Medicine, Salt Lake City, Utah Headaches and Other Head Pain

W. Lawrence Drew, MD, PhD Professor Emeritus of Laboratory Medicine and Medicine, University of California, San Francisco, School of Medicine, San Francisco, California Cytomegalovirus George L. Drusano, MD Professor of Medicine and Director, Institute for Therapeutic Innovation, University of Florida College of Medicine, Orlando, Florida Antibacterial Chemotherapy Thomas D. DuBose, Jr., MD Professor Emeritus of Medicine, Wake Forest School of Medicine, Winston-Salem, North Carolina; Visiting Professor of Medicine, University of Virginia School of Medicine, Charlottesville, Virginia Vascular Disorders of the Kidney J. Stephen Dumler, MD Professor and Chairperson, Joint Departments of Pathology, Uniformed Services University, Walter Reed National Military Medical Center, and Joint Pathology Center, Bethesda, Maryland Zoonoses Herbert L. DuPont, MD Professor of Infectious Diseases, University of Texas School of Public Health, Mary W. Kelsey Chair, University of Texas McGovern Medical School, Houston, Texas Approach to the Patient with Suspected Enteric Infection Madeleine Duvic, MD Professor and Deputy Chairman, Department of Dermatology, University Texas MD Anderson Cancer Center, Houston, Texas Urticaria, Drug Hypersensitivity Rashes, Nodules and Tumors, and Atrophic Diseases Kathryn M. Edwards, MD Sarah H. Sell and Cornelius Vanderbilt Chair in Pediatrics, Vanderbilt University School of Medicine, Nashville, Tennessee Parainfluenza Viral Disease

James. H. Doroshow, MD Deputy Director for Clinical and Translational Research, Director, Division of Cancer Treatment and Diagnosis, National Cancer Institute, National Institutes of Health, Bethesda, Maryland Approach to the Patient with Cancer

N. Lawrence Edwards, MD Professor and Vice Chairman, Department of Medicine, University of Florida College of Medicine; Chief, Section of Rheumatology Medicine, Malcolm Randall Veterans Administration Medical Center, Gainesville, Florida Crystal Deposition Diseases

John M. Douglas, Jr., MD Executive Director, Tri-County Health Department, Greenwood Village, Colorado Papillomavirus

Lawrence H. Einhorn, MD Distinguished Professor of Medicine, Indiana University School of Medicine, Indianapolis, Indiana Testicular Cancer

Jeffrey M. Drazen, MD Distinguished Parker B. Francis Professor of Medicine, Harvard Medical School and Senior Physician, Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts Asthma

George M. Eliopoulos, MD Professor of Medicine, Harvard Medical School; Physician, Beth Israel Deaconess Medical Center, Boston, Massachusetts Principles of Anti-Infective Therapy

Dimitri Drekonja, MD, MS Associate Professor of Medicine, University of Minnesota and Chief, Infectious Diseases Section, Minneapolis VA Health Care System, Minneapolis, Minnesota Approach to the Patient with Urinary Tract Infection Stephen C. Dreskin, MD, PhD Professor of Medicine and Immunology, University of Colorado School of Medicine, Aurora, Colorado Urticaria and Angioedema

Perry M. Elliott, MBBS, MD Professor of Cardiovascular Medicine, Institute of Cardiovascular Science, University College London & St. Bartholomew’s Hospital, London, United Kingdom Diseases of the Myocardium and Endocardium Jerrold J. Ellner, MD Professor of Medicine, Rutgers-New Jersey Medical School; Director of Research Innovations, Center for Emerging Pathogens, Newark, New Jersey Tuberculosis

Contributors

xiii

Ezekiel J. Emanuel, MD, PhD Vice Provost for Global Initiatives, Office of the Provost; Chair, Department of Medical Ethics and Health Policy, University of Pennsylvania, Philadelphia, Pennsylvania Bioethics in the Practice of Medicine

Lee A. Fleisher, MD Robert D. Dripps Professor and Chair, Anesthesiology and Critical Care; Professor of Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania Overview of Anesthesia

Joel D. Ernst, MD Professor and Chief, Division of Experimental Medicine, University of California, San Francisco, School of Medicine, San Francisco, California Leprosy (Hansen Disease)

Paul W. Flint, MD Professor and Chair of Otolaryngology-Head & Neck Surgery, Oregon Health & Science University, Portland, Oregon Throat Disorders

Gregory T. Everson, MD Professor of Medicine, University of Colorado Denver; Director of Hepatology, Hepatology and Transplant Center, University of Colorado Hospital, Aurora, Colorado Hepatic Failure and Liver Transplantation

Evan L. Fogel, MD, MSc Professor of Medicine, Indiana University School of Medicine, Indianapolis, Indiana Diseases of the Gallbladder and Bile Ducts

Amelia Evoli, MD Associate Professor of Neurology, Institute of Neurology, Catholic University, Roma, Italy Disorders of Neuromuscular Transmission Matthew E. Falagas, MD, MSc, DSc Director, Alfa Institute of Biomedical Sciences and Chief, Department of Medicine, Henry Dunant Hospital Center, Athens, Greece; Adjunct Associate Professor of Medicine, Tufts University School of Medicine, Boston, Massachusetts Pseudomonas and Related Gram-Negative Bacillary Infections Gary W. Falk, MD, MS Professor of Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania Diseases of the Esophagus James C. Fang, MD Professor of Medicine, University of Utah School of Medicine; Executive Director, Cardiovascular Service Line, University of Utah Health Sciences, Salt Lake City, Utah ST Elevation Acute Myocardial Infarction and Complications of Myocardial Infarction Gene Feder, MBBS, MD Professor, Centre for Academic Primary Care, Population Health Sciences, Bristol Medical School, University of Bristol; General Practitioner, Helios Medical Centre, Bristol, United Kingdom Intimate Partner Violence David J. Feller-Kopman, MD Professor of Medicine, Anesthesiology, Otolaryngology-Head & Neck Surgery and Director, Bronchoscopy & Interventional Pulmonology, Johns Hopkins University School of Medicine, Baltimore, Maryland Interventional and Surgical Approaches to Lung Disease Thomas McDonald File, Jr., MD, MSc Professor and Chair, Infectious Disease Section, Northeast Ohio Medical University, Rootstown, Ohio; Chair, Infectious Disease Division, Summa Health, Akron, Ohio Streptococcus Pneumoniae Infections Gary S. Firestein, MD Professor of Medicine, Dean, and Associate Vice Chancellor of Clinical and Translational Research, University of California, San Diego, School of Medicine, La Jolla, California Mechanisms of Inflammation and Tissue Repair Glenn I. Fishman, MD William Goldring Professor of Medicine and Director, Leon H. Charney Division of Cardiology, New York University School of Medicine, New York, New York Principles of Electrophysiology

Chris E. Forsmark, MD Professor of Medicine, University of Florida College of Medicine, Gainesville, Florida Pancreatitis Pierre-Edouard Fournier, MD, PhD Professor of Medical Bacteriology-Virology and Hygiene, Faculté de Médecine, Aix-Marseille Université and Institut Hospitalo-Universitaire Méditerranée-Infection, Marseille, France Rickettsial Infections Vance G. Fowler, Jr., MD, MHS Professor of Medicine and of Molecular Genetics and Microbiology, Duke University School of Medicine, Durham, North Carolina Infective Endocarditis Manuel A. Franco, MD, PhD Professor, Instituto de Genética Humana, Facultad de Medicina, Pontificia Universidad Javeriana, Bogotá, Colombia Rotaviruses, Noroviruses, and Other Gastrointestinal Viruses David O. Freedman, MD Professor Emeritus of Infectious Diseases, University of Alabama at Birmingham School of Medicine; Medical Director, Shoreland Travax, Birmingham, Alabama Approach to the Patient before and after Travel Martyn A. French, MB ChB, MD Emeritus Professor in Clinical Immunology, University of Western Australia Medical School and School of Biomedical Sciences, Faculty of Health and Medical Sciences, Perth, Australia Immune Reconstitution Inflammatory Syndrome in HIV/AIDS Karen M. Freund, MD, MPH Professor of Medicine and Vice Chair for Faculty Affairs and Quality Improvement, Tufts University School of Medicine, Boston, Massachusetts Approach to Women’s Health John N. Galgiani, MD Professor of Medicine and Director, Valley Fever Center for Excellence, University of Arizona College of Medicine; Chief Medical Officer, Valley Fever Solutions, Tucson, Arizona Endemic Mycoses Patrick G. Gallagher, MD Professor of Pediatrics, Pathology and Genetics, Yale University School of Medicine, New Haven, Connecticut Hemolytic Anemias: Red Blood Cell Membrane and Metabolic Defects Leonard Ganz, MD Director of Cardiac Electrophysiology, Heart and Vascular Center, Heritage Valley Health System, Beaver, Pennsylvania Electrocardiography

xiv

Contributors

Hasan Garan, MD, MS Dickinson W. Richards, Jr. Professor of Medicine, Director, Cardiac Electrophysiology, Columbia University Vagelos College of Physicians and Surgeons, New York, New York Ventricular Arrhythmias Guadalupe Garcia-Tsao, MD Professor of Medicine, Yale University School of Medicine, New Haven, Connecticut; Chief of Digestive Diseases, School of Medicine, VA-CT Healthcare System, West Haven, Connecticut Cirrhosis and Its Sequelae William M. Geisler, MD, MPH Professor of Medicine, University of Alabama at Birmingham School of Medicine, Birmingham, Alabama Diseases Caused by Chlamydiae Tony P. George, MD Professor of Psychiatry and Director, Division of Brain and Therapeutics, University of Toronto; Chief, Addictions Division, Centre for Addiction and Mental Health, Toronto, Ontario, Canada Nicotine and Tobacco Lior Gepstein, MD, PhD Sohnis Family Professor in Medicine, Technion - Israel Institute of Technology; Director, Cardiology Department, Rambam Health Care Campus, Haifa, Israel Regenerative Medicine, Cell, and Gene Therapies Susan I. Gerber, MD Chief, Respiratory Viruses Branch, Division of Viral Diseases, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia Coronaviruses Dale N. Gerding, MD Professor (retired) of Medicine, Loyola University Chicago Stritch School of Medicine, Maywood, Illinois; Research Physician, Medicine, Edward Hines Jr. VA Hospital, Hines, Illinois Clostridial Infections Morie A. Gertz, MD Roland Seidler Jr. Professor of the Art of Medicine and Chair Emeritus, Internal Medicine, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Amyloidosis Khalil G. Ghanem, MD, PhD Associate Professor of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland Granuloma Inguinale (Donovanosis); Syphilis; Nonsyphilitic Treponematoses Christopher J. Gill, MD, MS Associate Professor of Global Health, Boston University School of Public Health, Boston, Massachusetts Whooping Cough and Other Bordetella Infections Jeffrey S. Ginsberg, MD Professor of Medicine, McMaster University Michael G. DeGroote School of Medicine, Hamilton, Ontario, Canada Venous Thrombosis and Embolism Geoffrey S. Ginsburg, MD, PhD Professor of Medicine and Pathology and Director, Duke Center for Applied Genomics & Precision Medicine, Duke University, Durham, North Carolina Applications of Molecular Technologies to Clinical Medicine Marshall J. Glesby, MD, PhD Professor of Medicine, Weill Cornell Medical College, New York, New York Systemic Manifestations of HIV/AIDS

John W. Gnann, Jr., MD Professor of Medicine, Medical University of South Carolina, Charleston, South Carolina Mumps; Herpes Simplex Virus Infections Matthew R. Golden, MD, MPH Professor of Medicine, University of Washington School of Medicine; Director, HIV/STD Program, Public Health - Seattle & King County, Seattle, Washington Neisseria Gonorrhoeae Infections David L. Goldman, MD Associate Professor of Pediatrics, Microbiology and Immunology, Children’s Hospital at Montefiore/Albert Einstein College of Medicine, Bronx, New York Mycoplasma Infections Lee Goldman, MD Harold and Margaret Hatch Professor, Chief Executive, Columbia University Irving Medical Center, Dean of the Faculties of Health Sciences and Medicine, Columbia University, New York, New York Approach to Medicine, the Patient, and the Medical Profession: Medicine as a Learned and Humane Profession; Approach to the Patient with Possible Cardiovascular Disease Larry B. Goldstein, MD Ruth L Works Professor and Chairman, Department of Neurology, University of Kentucky College of Medicine; Co-Director, Kentucky Neuroscience Institute, Lexington, Kentucky Approach to Cerebrovascular Diseases; Ischemic Cerebrovascular Disease Richard M. Gore, MD Professor of Radiology, University of Chicago Pritzker School of Medicine; Chief, Section of Gastrointestinal Radiology, NorthShore University HealthSystem, Evanston, Illinois Diagnostic Imaging Procedures in Gastroenterology Jason Gotlib, MD, MS Professor of Medicine, Stanford University School of Medicine, Stanford Cancer Institute, Stanford, California Polycythemia Vera, Essential Thrombocythemia, and Primary Myelofibrosis Eduardo Gotuzzo, MD Professor Emeritus, Alexander von Humboldt Tropical Medicine Institute, Universidad Peruana Cayetano Heredia; Principal Professor of Medicine and Tropical Diseases, National Hospital Cayetano Heredia, Lima, Peru Cholera and Other Vibrio Infections; Trematode Infections Leslie C. Grammer, MD Professor of Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois Drug Allergy Hartmut Grasemann, MD, PhD Professor of Pediatrics, The Hospital for Sick Children and University of Toronto, Toronto, Ontario, Canada Cystic Fibrosis M. Lindsay Grayson, MBBS, MD, MS Professor of Medicine, University of Melbourne, Director, Infectious Diseases & Microbiology, Austin Health, Melbourne, Victoria, Australia Principles of Anti-Infective Therapy Harry B. Greenberg, MD Professor of Medicine and of Microbiology and Immunology, Stanford University School of Medicine, Stanford, California Rotaviruses, Noroviruses, and Other Gastrointestinal Viruses

Contributors Steven A. Greenberg, MD Professor of Neurology, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts Inflammatory Myopathies David M. Greer, MD, MA Professor and Chair of Neurology, Boston University School of Medicine, Boston, Massachusetts Coma, Vegetative State, and Brain Death Robert C. Griggs, MD Professor of Neurology, Medicine, Pediatrics, Pathology & Laboratory Medicine, University of Rochester School of Medicine & Dentistry, Rochester, New York Approach to the Patient with Neurologic Disease Lev M. Grinberg, MD, PhD Professor and Chair, Department of Pathology, Ural State Medical University, Ekaterinburg, Russia Anthrax Daniel Grossman, MD Professor of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, School of Medicine, San Francisco, California Contraception Lisa M. Guay-Woodford, MD Richard L. Hudson Professor of Pediatrics, George Washington University School of Medicine and Health Science and Director, Center for Translational Research, Children’s National Medical Center, Washington D.C. Hereditary Nephropathies and Developmental Abnormalities of the Urinary Tract Roy M. Gulick, MD, MPH Professor of Medicine, Weill Cornell Medical School; Attending Physician, NewYork-Presbyterian Hospital, New York, New York Antiretroviral Therapy for Human Immunodeficiency Virus and Acquired Immunodeficiency Syndrome Rajesh Gupta, MD, MEd Associate Professor of Medicine, University of Toronto; General Internist, Medicine, St. Michael’s Hospital, Toronto, Ontario, Canada Medical Consultation in Psychiatry Colleen Hadigan, MD, MPH Staff Clinician, National Institutes of Health, Laboratory of Immunoregulation, NIAID, Bethesda, Maryland Microbial Complications in Patients Infected with Human Immunodeficiency Virus Melissa M. Hagman, MD Associate Professor of Medicine, Program Director, Internal Medicine Residency-Boise, University of Washington, Boise, Idaho Nonpneumococcal Streptococcal Infections and Rheumatic Fever Klaus D. Hagspiel, MD Professor of Radiology, Medicine (Cardiology) and Pediatrics; Chief, Division of Noninvasive Cardiovascular Imaging, Department of Radiology and Medical Imaging, University of Virginia School of Medicine, Charlottesville, Virginia Noninvasive Cardiac Imaging H. Hunter Handsfield, MD Professor of Medicine Emeritus, University of Washington School of Medicine, Seattle, Washington Neisseria Gonorrhoeae Infections

xv

Raymond C. Harris, MD Anne and Roscoe R. Robinson Chair and Professor of Medicine and Associate Chair, Division of Nephrology, Medicine, Vanderbilt University School of Medicine, Nashville, Tennessee Diabetes and the Kidney Frederick G. Hayden, MD Stuart S. Richardson Professor Emeritus of Clinical Virology and Professor Emeritus of Medicine, Department of Medicine, University of Virginia School of Medicine, Charlottesville, Virginia Influenza Frederick M. Hecht, MD Professor of Medicine, University of California, San Francisco, School of Medicine, San Francisco, California Complementary, Alternative, and Integrative Medicine Douglas C. Heimburger, MD, MS Professor of Medicine, Vanderbilt University School of Medicine; Associate Director for Education & Training, Vanderbilt Institute for Global Health, Vanderbilt University, Nashville, Tennessee Nutrition’s Interface with Health and Disease Donald D. Hensrud, MD, MPH Associate Professor of Preventive Medicine and Nutrition, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Nutrition’s Interface with Health and Disease Erik L. Hewlett, MD Professor of Medicine, Microbiology, Immunology and Cancer Biology, University of Virginia School of Medicine, Charlottesville, Virginia Whooping Cough and Other Bordetella Infections Richard J. Hift, MMed(Med), PhD Professor of Medicine, University of KwaZulu-Natal, Durban, KwaZulu-Natal, South Africa The Porphyrias David R. Hill, MD, DTM&H Professor of Medical Sciences, Director, Global Public Health, Quinnipiac University Frank H Netter MD School of Medicine, Hamden, Connecticut Giardiasis Nicholas S. Hill, MD Professor of Medicine, Tufts University School of Medicine; Chief, Division of Pulmonary, Critical Care and Sleep Medicine, Tufts Medical Center, Boston, Massachusetts Respiratory Monitoring in Critical Care Christopher D. Hillyer, MD President and Chief Executive Officer, New York Blood Center; Professor of Medicine, Weill Cornell Medical College, New York, New York Transfusion Medicine Brian D. Hoit, MD Professor of Medicine, Physiology and Biophysics, Case Western Reserve University School of Medicine; Director of Echocardiography, Harrington Heart & Vascular Center, University Hospital Cleveland Medical Center, Cleveland, Ohio Pericardial Diseases Steven M. Holland, MD Director, Division of Intramural Research, Chief, Immunopathogenesis Section, National Institute of Allergy and Infectious Diseases, NIH, Bethesda, Maryland The Nontuberculous Mycobacteria

xvi

Contributors

Steven M. Hollenberg, MD Professor of Medicine, Cooper Medical School of Rowan University; Director, Coronary Care Unit, Cooper University Hospital, Camden, New Jersey Cardiogenic Shock Edward W. Hook, III, MD Professor of Medicine and Director, Division of Infectious Diseases, University of Alabama at Birmingham School of Medicine, Birmingham, Alabama Granuloma Inguinale (Donovanosis); Syphilis; Nonsyphilitic Treponematoses

Dennis M. Jensen, MD Professor of Medicine, David Geffen School of Medicine at UCLA; Staff Physician, Medicine-GI, VA Greater Los Angeles Healthcare System; Director, Human Studies Core & GI Hemostasis Research Unit, CURE Digestive Diseases Research Center, Los Angeles, California Gastrointestinal Hemorrhage Michael D. Jensen, MD Professor of Medicine, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Obesity

Jo Howard, MB BChir Consultant Haematologist and Lead Clinician, Haematology, Guy’s and St Thomas’ National Health Service Foundation Trust; Honorary Reader, King’s College London, London, United Kingdom Sickle Cell Disease and Other Hemoglobinopathies

Robert T. Jensen, MD Chief, Cell Biology Section, Digestive Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland Neuroendocrine Tumors

David J. Hunter, MBBS, MPH, ScD Richard Doll Professor of Epidemiology and Medicine, Nuffield Department of Population Health, University of Oxford, Oxford, United Kingdom Epidemiology of Cancer

Alain Joffe, MD, MPH Retired. Most recently, Associate Professor of Pediatrics, Johns Hopkins University School of Medicine and Director, Student Health and Wellness Center, Johns Hopkins University, Baltimore, Maryland Adolescent Medicine

Khalid Hussain, MB ChB, MD, MSc Professor of Pediatrics, Weill Cornell Medicine-Qatar; Division ChiefEndocrinology, Vice Chair for Research, Program Director-Research, Sidra Medicine, OPC, Doha, Qatar Hypoglycemia and Pancreatic Islet Cell Disorders

Stuart Johnson, MD Professor of Medicine/Infectious Disease, Loyola University Chicago Stritch School of Medicine, Maywood, Illinois; Physician Researcher, Research Service, Hines VA Hospital, Hines, Illinois Clostridial Infections

Michael C. Iannuzzi, MD, MBA Professor and Chairman, Department of Internal Medicine, Northwell-Staten Island University Hospital and Donald and Barbara Zucker School of Medicine at Hofstra/Northwell, New York Sarcoidosis

Robin L. Jones, MD, BSc, MB Consultant Medical Oncologist, Royal Marsden Hospital and Institute of Cancer Research, London, United Kingdom Malignant Tumors of Bone, Sarcomas, and Other Soft Tissue Neoplasms

Robert D. Inman, MD Professor of Medicine and Immunology, University of Toronto and Kremil Research Institute, University Health Network, Toronto, Ontario, Canada The Spondyloarthropathies Sharon K. Inouye, MD, MPH Professor of Medicine, Harvard Medical School; Director, Aging Brain Center, Marcus Institute for Aging Research-Hebrew SeniorLife, Boston, Massachusetts Neuropsychiatric Aspects of Aging; Delirium in the Older Patient Michael G. Ison, MD, MS Professor of Medicine (Infectious Diseases) and Surgery (Organ Transplantation), Northwestern University Feinberg School of Medicine, Chicago, Illinois Influenza; Adenovirus Diseases Karen R. Jacobson, MD, MPH Assistant Professor of Medicine, Medical Director, Boston Tuberculosis Clinic, Boston University School of Medicine, Boston, Massachusetts Tuberculosis Michael R. Jaff, DO Professor of Medicine, Harvard Medical School, Boston, Massachusetts; President, Newton-Wellesley Hospital, Newton, Massachusetts Other Peripheral Arterial Diseases Joanna C. Jen, MD, PhD Professor of Neurology, David Geffen School of Medicine at UCLA, Los Angeles, California Neuro-Ophthalmology; Smell and Taste; Hearing and Equilibrium

Sian Jones, MD Associate Professor of Clinical Medicine, Weill Cornell Medical College, New York, New York Systemic Manifestations of HIV/AIDS Jacqueline Jonklaas, MD, PhD Professor of Medicine, Georgetown University School of Medicine, Washington, D.C. Thyroid Richard C. Jordan, DDS, PhD Professor of Pathology, Oral Pathology & Radiation Oncology, University of California, San Francisco, School of Medicine, San Francisco, California Diseases of the Mouth and Salivary Glands Charles J. Kahi, MD, MS Professor of Clinical Medicine, Indiana University School of Medicine; GI Section Chief, Richard L. Roudebush VAMC, Indianapolis, Indiana Vascular Diseases of the Gastrointestinal Tract Moses R. Kamya, MB ChB, MMed, MPH, PhD Professor of Medicine, Makerere University School of Medicine, Kampala, Uganda Malaria Louise W. Kao, MD Associate Professor of Clinical Emergency Medicine and Director, Medical Toxicology Fellowship Program, Indiana University School of Medicine, Indianapolis, Indiana Chronic Poisoning: Trace Metals and Others Steven A. Kaplan, MD Professor of Urology, Icahn School of Medicine at Mount Sinai; Director, Men’s Health Program, Mount Sinai Health System, New York, New York Benign Prostatic Hyperplasia and Prostatitis

Contributors Daniel L. Kastner, MD, PhD Scientific Director, Division of Intramural Research, National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland The Systemic Autoinflammatory Diseases David A. Katzka, MD Professor of Medicine, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Diseases of the Esophagus Debra K. Katzman, MD Professor of Pediatrics, The Hospital for Sick Children and University of Toronto; Senior Associate Scientist, Research Institute; Director, Health Science Research, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada Adolescent Medicine Carol A. Kauffman, MD Professor of Internal Medicine and Chief, Infectious Diseases, Veterans Affairs Ann Arbor Healthcare System, University of Michigan Medical School, Ann Arbor, Michigan Endemic Mycoses; Cryptococcosis; Candidiasis Kenneth Kaushansky, MD Professor of Medicine, Senior Vice President for Health Sciences, and Dean, Stony Brook University School of Medicine, Stony Brook, New York Hematopoiesis and Hematopoietic Growth Factors Keith S. Kaye, MD, MPH Professor of Medicine, University of Michigan Medical School, Ann Arbor, Michigan Diseases Caused by Acinetobacter and Stenotrophomonas Species Armand Keating, MD Professor of Medicine and Professor, Institute of Biomaterials and Biomedical Engineering, University of Toronto, Toronto, Canada Hematopoietic Stem Cell Transplantation Robin K. Kelley, MD Associate Professor of Clinical Medicine, University of California, San Francisco, School of Medicine, San Francisco, California Liver and Biliary Tract Cancers Morton J. Kern, MD Professor of Medicine and Associate Chief of Cardiology, University of California, Irvine, Orange, California; Chief of Medicine, Long Beach Veterans Health Care System, Long Beach, California Catheterization and Angiography Gerald T. Keusch, MD Professor of Medicine, Boston University School of Medicine, Boston, Massachusetts Shigellosis Fadlo R. Khuri, MD President and Professor of Hematology and Medical Oncology, American University of Beirut; Adjunct Professor of Medicine, Pharmacology and Otolaryngology, Emory University School of Medicine, Atlanta, Georgia Lung Cancer and Other Pulmonary Neoplasms Louis V. Kirchhoff, MD, MPH Professor of Internal Medicine (Infectious Diseases), Psychiatry, and Epidemiology, University of Iowa Carver College of Medicine and College of Public Health, Iowa City, Iowa Chagas Disease

xvii

Ajay J. Kirtane, MD Associate Professor of Medicine, Columbia University Vagelos College of Physicians and Surgeons; Chief Academic Officer, Center for Interventional Vascular Therapy; Director, Columbia University Irving Medical Center Cardiac Catheterization Laboratories, New York New York Catheterization and Angiography Amy D. Klion, MD Senior Clinical Investigator, Laboratory of Parasitic Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland Eosinophilic Syndromes David S. Knopman, MD Professor of Neurology, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Regional Cerebral Dysfunction: Higher Mental Functions; Cognitive Impairment and Dementia Christine J. Ko, MD Professor of Dermatology and Pathology, Yale University School of Medicine, New Haven, Connecticut Approach to Skin Diseases Dimitrios P. Kontoyiannis, MD, ScD Texas 4000 Distinguished Endowed Professor For Cancer Research, Deputy Head, Division of Internal Medicine, University of Texas MD Anderson Cancer Center, Houston, Texas Mucormycosis; Mycetoma and Dematiaceous Fungal Infections Barbara S. Koppel, MD Chief of Neurology, Metropolitan Hospital, New York, New York and Professor of Clinical Neurology, New York Medical College, Valhalla, New York Nutritional and Alcohol-Related Neurologic Disorders Kevin M. Korenblat, MD Professor of Medicine, Washington University School of Medicine in St. Louis, St. Louis, Missouri Approach to the Patient with Jaundice or Abnormal Liver Tests Bruce R. Korf, MD, PhD Professor of Genetics, University of Alabama at Birmingham and Chief Genomics Officer, UAB Medicine, Birmingham, Alabama Principles of Genetics Mark G. Kortepeter, MD, MPH Professor of Epidemiology, College of Public Health, University of Nebraska, Omaha, Nebraska; Adjunct Professor of Preventive Medicine and Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland Bioterrorism Shyamasundaran Kottilil, MD, PhD Professor of Medicine and Associate Chief of Infectious Diseases at the Institute of Human Virology, University of Maryland School of Medicine, Baltimore, Maryland Antiviral Therapy (Non-HIV) Joseph A. Kovacs, MD Senior Investigator, Critical Care Medicine Department, National Institutes of Health Clinical Center, Bethesda, Maryland Pneumocystis Pneumonia Thomas O. Kovacs, MD Professor of Medicine, David Geffen School of Medicine at UCLA, Los Angeles, California Gastrointestinal Hemorrhage

xviii

Contributors

Kris V. Kowdley, MD Director, Liver Care Network and Organ Care Research, Swedish Medical Center; Clinical Professor of Medicine, Washington State University, Elson S. Floyd College of Medicine, Seattle, Washington Iron Overload (Hemochromatosis) Monica Kraft, MD Robert and Irene Flinn Professor and Chair, Department of Medicine, Deputy Director, Asthma and Airway Disease Research Center, University of Arizona Health Sciences, Tucson, Arizona Approach to the Patient with Respiratory Disease Christopher M. Kramer, MD Ruth C. Heede Professor of Cardiology and Professor of Radiology, University of Virginia School of Medicine, Charlottesville, Virginia Noninvasive Cardiac Imaging Donna M. Krasnewich, MD, PhD Program Director, NIGMS, National Institutes of Health, Bethesda, Maryland Lysosomal Storage Diseases Alexander Kratz, MD, PhD, MPH Professor of Clinical Pathology and Cell Biology, Columbia University Vagelos College of Physicians and Surgeons; Director, Automated Core Laboratory and Point of Care Testing Service, Columbia University Irving Medical Center and NewYork-Presbyterian Hospital, New York, New York Reference Intervals and Laboratory Values Virginia Byers Kraus, MD, PhD Professor of Medicine, Adjunct Professor of Pathology and Orthopaedic Surgery, Duke University School of Medicine, Duke Molecular Physiology Institute, Durham, North Carolina Osteoarthritis William E. Kraus, MD Richard and Pat Johnson Distinguished University Professor, Duke University School of Medicine, Durham, North Carolina Physical Activity Peter J. Krause, MD Senior Research Scientist, Yale University School of Public Health, Yale University School of Medicine, New Haven, Connecticut Babesiosis and Other Protozoan Diseases Daniela Kroshinsky, MD, MPH Associate Professor of Dermatology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts Macular, Papular, Purpuric, Vesicobullous, and Pustular Diseases John F. Kuemmerle, MD Charles M. Caravati Professor of Medicine, Chair, Division of Gastroenterology, Hepatology and Nutrition, Medical College of Virginia, Virginia Commonwealth University, Richmond, Virginia Inflammatory and Anatomic Diseases of the Intestine, Peritoneum, Mesentery, and Omentum Ernst J. Kuipers, MD, PhD Professor of Medicine, Erasmus MC University Medical Center, Rotterdam, Netherlands Acid Peptic Disease Daniel Laheru, MD Ian T. MacMillan Professorship in Clinical Pancreatic Research, Johns Hopkins University School of Medicine, Baltimore, Maryland Pancreatic Cancer

Donald W. Landry, MD, PhD Samuel Bard Professor and Chair, Department of Medicine, Columbia University Vagelos College of Physicians and Surgeons and Physician-in-Chief, Columbia University Irving Medical Center, New York, New York Approach to the Patient with Renal Disease Anthony E. Lang, MD Jack Clark Chair in Parkinson’s Disease Research and Director, Division of Neurology, University of Toronto; Director, Morton and Gloria Shulman Movement Disorders Clinic and Edmond J Safra Program in Parkinson’s Disease, University Health Network, Toronto Western Hospital, Toronto, Ontario, Canada Parkinsonism; Other Movement Disorders Richard A. Lange, MD, MBA Rick and Ginger Francis Endowed Professor and President, Texas Tech University Health Sciences Center, El Paso; Dean, Paul L. Foster School of Medicine, El Paso, Texas Acute Coronary Syndrome: Unstable Angina and Non–ST Elevation Myocardial Infarction Frank A. Lederle, MD† Formerly Professor of Medicine, University of Minnesota School of Medicine; Director of the Minneapolis Veterans Administration Center for Epidemiological and Clinical Research, Minneapolis, Minnesota Diseases of the Aorta William M. Lee, MD Meredith Mosle Chair in Liver Disease and Professor of Internal Medicine, University of Texas Southwestern Medical Center at Dallas, Dallas, Texas Toxin- and Drug-Induced Liver Disease James E. Leggett, MD Department of Medical Education, Providence Portland Medical Center; Associate Professor of Medicine Emeritus, Division of Infectious Diseases, Oregon Health & Science University, Portland, Oregon Approach to Fever or Suspected Infection in the Normal Host Glenn N. Levine, MD Professor of Medicine, Baylor College of Medicine; Director, Cardiac Care Unit, Michael E. DeBakey VA Medical Center, Houston, Texas Antithrombotic and Antiplatelet Therapy Marc S. Levine, MD Emeritus Professor of Radiology, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania Diagnostic Imaging Procedures in Gastroenterology Stephanie M. Levine, MD Professor of Medicine, University of Texas Health San Antonio, San Antonio, Texas Alveolar Filling Disorders Gary R. Lichtenstein, MD Professor of Medicine, University of Pennsylvania Perelman School of Medicine; Director, Center for Inflammatory Bowel Disease, Department of Medicine, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania Inflammatory Bowel Disease Jeffrey M. Liebmann, MD Shirlee and Bernard Brown Professor and Vice Chair, Department of Ophthalmology, Columbia University Vagelos College of Physicians and Surgeons, New York, New York Diseases of the Visual System



Deceased.

Contributors Henry W. Lim, MD Chairman and C.S. Livingood Chair Emeritus of Dermatology, Henry Ford Hospital; Senior Vice President for Academic Affairs, Henry Ford Health System, Detroit, Michigan Eczemas, Photodermatoses, Papulosquamous (Including Fungal) Diseases, and Figurate Erythemas Aldo A.M. Lima, MD, PhD Professor, Institute of Biomedicine, Federal University of Ceara, Fortaleza, Ceará, Brazil Cryptosporidiosis; Trematode Infections Geoffrey S.F. Lin, MD, PhD Professor of Neurology, Johns Hopkins University School of Medicine, Baltimore, Maryland Traumatic Brain Injury and Spinal Cord Injury Mark S. Link, MD Professor of Medicine, University of Texas Southwestern Medical Center, Dallas, Texas Electrocardiography Donald M. Lloyd-Jones, MD, ScM Chair and Eileen M. Foell Professor of Preventive Medicine, Senior Associate Dean for Clinical & Translational Research, Northwestern University Feinberg School of Medicine, Chicago, Illinois Epidemiology of Cardiovascular Disease Bennett Lorber, MD, DSc Thomas M. Durant Professor of Medicine and Professor of Microbiology and Immunology, Lewis Katz School of Medicine at Temple University, Philadelphia, Pennsylvania Listeriosis Arnold Louie, MD Professor of Medicine, Molecular Genetics and Microbiology and Associate Director, Institute for Therapeutic Innovation, University of Florida College of Medicine, Orlando, Florida Antibacterial Chemotherapy Daniel R. Lucey, MD, MPH Adjunct Professor, Department of Medicine/Infectious Diseases, Georgetown University Medical Center, Washington, D.C. Anthrax Jeffrey M. Lyness, MD Professor of Psychiatry & Neurology and Senior Associate Dean for Academic Affairs, University of Rochester School of Medicine & Dentistry, Rochester, New York Psychiatric Disorders in Medical Practice C. Ronald MacKenzie, MD C. Ronald MacKenzie Chair in Ethics and Medicine, Hospital for Special Surgery; Professor of Clinical Medicine and Medical Ethics, Weill Cornell Medical College, New York, New York Surgical Treatment of Joint Diseases Harriet L. MacMillan, CM, MD, MSc Chedoke Health Chair in Child Psychiatry and Professor of Psychiatry & Behavioural Neurosciences and of Pediatrics, Offord Centre for Child Studies, McMaster University Michael G. DeGroote School of Medicine, Hamilton, Ontario, Canada Intimate Partner Violence Robert D. Madoff, MD Professor of Surgery, University of Minnesota, Minneapolis, Minnesota Diseases of the Rectum and Anus Frank Maldarelli, MD, PhD Head, Clinical Retrovirology Section, HIV Dynamics and Replication Program, NCI-Frederick, Frederick, Maryland Biology of Human Immunodeficiency Viruses

xix

Atul Malhotra, MD Kenneth M. Moser Professor of Medicine, Chief of Pulmonary and Critical Care Medicine, Director of Sleep Medicine, University of California, San Diego, School of Medicine, La Jolla, California Disorders of Ventilatory Control Mark J. Manary, MD Helene B. Roberson Professor of Pediatrics, Washington University School of Medicine in St. Louis, St. Louis, Missouri; Senior Lecturer, Department of Community Health, University of Malawi College of Medicine, Blantyre, Malawi Protein-Energy Malnutrition Peter Manu, MD Professor of Medicine and Psychiatry, Donald and Barbara Zucker School of Medicine at Hofstra/Northwell, Hempstead, New York; Director of Medical Services, South Oaks Hospital, Amityville, New York Medical Consultation in Psychiatry Luis A. Marcos, MD, MPH Associate Professor of Clinical Medicine, School of Medicine, Stony Brook University, Stony Brook, New York Trematode Infections Ariane J. Marelli, MD, MPH Professor of Medicine and Director, McGill Adult Unit for Congenital Heart Disease, McGill University Health Centre, Montreal, Quebec, Canada Congenital Heart Disease in Adults Xavier Mariette, MD, PhD Professor of Rheumatology, Université Paris-Sud, AP-HP, Le Kremlin Bicêtre, France Sjögren Syndrome Andrew R. Marks, MD Wu Professor and Chair, Department of Physiology and Cellular Biophysics, Director, Helen and Clyde Wu Center for Molecular Cardiology, Columbia University Vagelos College of Physicians & Surgeons, New York, New York Cardiac and Circulatory Function Kieren A. Marr, MD Professor of Medicine and Oncology and Director, Transplant and Oncology Infectious Diseases, John Hopkins University School of Medicine, Baltimore, Maryland Approach to Fever and Suspected Infection in the Immunocompromised Host Thomas J. Marrie, MD Professor of Medicine and Dean Emeritus, Faculty of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada Legionella Infections Paul Martin, MD Professor of Medicine and Chief, Division of Gastroenterology and Hepatology, University of Miami Miller School of Medicine, Miami, Florida Approach to the Patient with Liver Disease Fernando J. Martinez, MD, MS Bruce Webster Professor of Internal Medicine and Chief, Division of Pulmonary and Critical Care Medicine, Weill Cornell Medical College, New York, New York Interstitial Lung Disease Joel B. Mason, MD Professor of Medicine and Nutrition, Tufts University School of Medicine; Director, Vitamins & Carcinogenesis Laboratory, U.S.D.A. Human Nutrition Research Center at Tufts University, Boston, Massachusetts Vitamins, Trace Minerals, and Other Micronutrients

xx

Contributors

Henry Masur, MD Chief, Critical Care Medicine Department, Clinical Center, National Institutes of Health, Bethesda, Maryland Microbial Complications in Patients Infected with Human Immunodeficiency Virus Amy J. Mathers, MD Associate Professor of Medicine and Pathology, Associate Director of Clinical Microbiology, Medical Director Antimicrobial Stewardship, University of Virginia School of Medicine, Charlottesville, Virginia Infections Due to Other Members of the Enterobacteriaceae, Including Management of Multidrug-Resistant Strains Eric L. Matteson, MD, MPH Professor of Medicine, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Infections of Bursae, Joints, and Bones Michael A. Matthay, MD Professor of Medicine and Anesthesia, University of California, San Francisco, San Francisco, California Acute Respiratory Failure Emeran A. Mayer, MD Professor of Medicine and Psychiatry, Executive Director G. Oppenheimer Center for Neurobiology of Stress and Resilience, David Geffen School of Medicine at UCLA, Los Angeles, California Functional Gastrointestinal Disorders: Irritable Bowel Syndrome, Dyspepsia, Esophageal Chest Pain, and Heartburn Stephan A. Mayer, MD William T. Gossett Endowed Chair of Neurology, Henry Ford Health System, Professor of Neurology, Wayne State University School of Medicine, Detroit, Michigan Hemorrhagic Cerebrovascular Disease F. Dennis McCool, MD Professor of Medicine, Warren Alpert Medical School of Brown University, Providence, Rhode Island; Memorial Hospital of Rhode Island, Pawtucket, Rhode Island Diseases of the Diaphragm, Chest Wall, Pleura, and Mediastinum Iain McInnes, PhD Professor of Experimental Medicine and Director, Institute of Infection, Immunity and Inflammation, University of Glasgow, Glasgow, United Kingdom Rheumatoid Arthritis William J. McKenna, MD Emeritus Professor of Cardiology, Institute of Cardiovascular Science, University College London, London, United Kingdom Diseases of the Myocardium and Endocardium Vallerie McLaughlin, MD Professor of Medicine, University of Michigan Medical School; Director, Pulmonary Hypertension Program, Ann Arbor, Michigan Pulmonary Hypertension John J.V. McMurray, BSc, MB ChB, MD Professor of Medical Cardiology, British Heart Foundation Cardiovascular Research Centre, University of Glasgow; Honorary Consultant Cardiologist, Queen Elizabeth University Hospital Glasgow, Glasgow, Scotland, United Kingdom Heart Failure: Management and Prognosis Kenneth R. McQuaid, MD Professor of Clinical Medicine and Vice-Chair, Department of Medicine, University of California, San Francisco, School of Medicine; Chief of Gastroenterology and of the Medical Service, San Francisco Veterans, Affairs Medical Center, San Francisco, California Approach to the Patient with Gastrointestinal Disease

Paul S. Mead, MD, MPH Chief, Bacterial Diseases Branch, Division of Vector-Borne Diseases, Centers for Disease Control and Prevention, Fort Collins, Colorado Plague and Other Yersinia Infections Robert T. Means, Jr., MD Professor of Internal Medicine, East Tennessee State University James H. Quillen College of Medicine, Johnson City, Tennessee Approach to the Anemias Graeme Meintjes, MB ChB, MPH, PhD Professor of Medicine, University of Cape Town, Cape Town, South Africa Immune Reconstitution Inflammatory Syndrome in HIV/AIDS Genevieve B. Melton-Meaux, MD, PhD Professor of Surgery, University of Minnesota Medical School, Minneapolis, Minnesota Diseases of the Rectum and Anus Samuel T. Merrick, MD Professor of Clinical Medicine, Weill Cornell Medical College, New York, New York Systemic Manifestations of HIV/AIDS Marc Michel, MD Professor and Head of the Unit of Internal Medicine, Henri Mondor University Hospital, Assistance Publique Hopitaux de Paris, Université Paris-Est Créteil, Creteil, France Autoimmune and Intravascular Hemolytic Anemias Jonathan W. Mink, MD, PhD Professor of Neurology, University of Rochester School of Medicine & Dentistry, Rochester, New York Congenital, Developmental, and Neurocutaneous Disorders William E. Mitch, MD Professor of Medicine, Baylor College of Medicine, Houston, Texas Chronic Kidney Disease Bruce A. Molitoris, MD Distinguished Professor of Medicine, Indiana University School of Medicine, Indianapolis, Indiana Acute Kidney Injury José G. Montoya, MD Professor of Medicine, Division of Infectious Diseases and Geographic Medicine, Stanford University School of Medicine, Stanford, California; Director, Palo Alto Medical Foundation Toxoplasma Serology Laboratory, National Reference Center for the Study and Diagnosis of Toxoplasmosis, Palo Alto, California Toxoplasmosis Ernest Moy, MD, MPH Executive Director, Office of Health Equity, Veterans Health Administration, Washington, D.C. Measuring Health and Health Care Debabrata Mukherjee, MD, MS Professor and Chairman, Department of Internal Medicine, Chief, Cardiovascular Medicine, Texas Tech University Health Sciences Center, El Paso, Texas Acute Coronary Syndrome: Unstable Angina and Non–ST Elevation Myocardial Infarction Andrew H. Murr, MD Professor and Chairman, Department of Otolaryngology-Head and Neck Surgery, University of California, San Francisco, School of Medicine, San Francisco, California Approach to the Patient with Nose, Sinus, and Ear Disorders

Contributors Daniel M. Musher, MD Distinguished Service Professor of Medicine and Professor of Molecular Virology and Microbiology, Baylor College of Medicine; Staff Physician, Infectious Disease Section, Michael E. DeBakey VA Medical Center, Houston, Texas Overview of Pneumonia Robert J. Myerburg, MD Professor of Medicine and Physiology, Department of Medicine, University of Miami Miller School of Medicine, Miami, Florida Approach to Cardiac Arrest and Life-Threatening Arrhythmias Kari C. Nadeau, MD, PhD Naddisy Family Foundation Professor of Allergy and Director, Sean N. Parker Center for Allergy and Asthma Research at Stanford University, Stanford, California Approach to the Patient with Allergic or Immunologic Disease Stanley J. Naides, MD President, Stanley J. Naides, M.D., P.C., Dana Point, California Arboviruses Causing Fever and Rash Syndromes Theodore E. Nash, MD Principal Investigator, Clinical Parasitology Section, National Institutes of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland Giardiasis Avindra Nath, MD Chief, Section of Infections of the Nervous System, National Institutes of Neurological Diseases and Stroke, National Institutes of Health, Bethesda, Maryland Cytomegalovirus, Epstein-Barr Virus, and Slow Virus Infections of the Central Nervous System; Meningitis: Bacterial, Viral, and Other; Brain Abscess and Parameningeal Infections Genevieve Neal-Perry, MD, PhD Professor of Obstetrics and Gynecology and Director of the Reproductive Endocrinology and Infertility Center, University of Washington School of Medicine, Seattle, Washington Menopause Anne T. Neff, MD Professor of Medicine, Hematology/Medical Oncology, Cleveland Clinic Lerner College of Medicine; Staff Physician, Cleveland Clinic Foundation, Cleveland, Ohio Von Willebrand Disease and Hemorrhagic Abnormalities of Platelet and Vascular Function

xxi

Eric J. Nestler, MD, PhD Nash Family Professor of Neuroscience, Director, Friedman Brain Institute, Icahn School of Medicine at Mount Sinai, New York, New York Biology of Addiction Anne B. Newman, MD, MPH Distinguished Professor and Chair, Department of Epidemiology, Katherine M. Detre Endowed Chair of Population Health Sciences; Director, Center for Aging and Population Health, Professor of Medicine, and Clinical and Translational Science Graduate School of Public Health, University of Pittsburgh; Clinical Director, Aging Institute of UPMC and Pitt, Pittsburgh, Pennsylvania Epidemiology of Aging: Implications of an Aging Society Lindsay E. Nicolle, MD Professor Emeritus, Department of Internal Medicine, University of Manitoba, Winnipeg, Manitoba, Canada Approach to the Patient with Urinary Tract Infection Lynnette K. Nieman, MD Senior Investigator, Diabetes, Endocrinology and Obesity Branch, NIDDK/NIH, Bethesda, Maryland Approach to the Patient with Endocrine Disease; Adrenal Cortex; Polyglandular Disorders Gaetane Nocturne, MD, PhD Associate Professor of Rheumatology, Université Paris-Sud, AP-HP, Le Kremlin Bicêtre, France Sjögren Syndrome Christopher M. O’Connor, MD Adjunct Professor of Medicine, Duke University School of Medicine, Durham, North Carolina; CEO, Inova Heart and Vascular Institute, Fairfax, Virginia Heart Failure: Pathophysiology and Diagnosis Francis G. O’Connor, MD, MPH Professor and Medical Director, Consortium for Health and Military Performance, Uniformed Services University of the Health Sciences, Bethesda, Maryland Disorders Due to Heat and Cold; Rhabdomyolysis Patrick G. O’Connor, MD, MPH Dan Adams and Amanda Adams Professor and Chief, General Internal Medicine, Yale University School of Medicine, New Haven, Connecticut Alcohol Use Disorders

Eric G. Neilson, MD Vice President for Medical Affairs and Lewis Landsberg Dean and Professor of Medicine and of Cell and Developmental Biology, Northwestern University Feinberg School of Medicine, Chicago, Illinois Tubulointerstitial Diseases

James R. O’Dell, MD Bruce Professor and Vice Chair of Internal Medicine, University of Nebraska Medical Center College of Medicine; Chief of Rheumatology, Medicine, Omaha VA, Omaha, Nebraska Rheumatoid Arthritis

Christina A. Nelson, MD, MPH Medical Officer, Bacterial Diseases Branch, Division of Vector-Borne Diseases, Centers for Disease Control and Prevention, Fort Collins, Colorado Plague and Other Yersinia Infections

Anne E. O’Donnell, MD The Nehemiah and Naomi Cohen Chair in Pulmonary Disease Research, Chief, Division of Pulmonary, Critical Care and Sleep Medicine, Georgetown University Medical Center, Washington, D.C. Bronchiectasis, Atelectasis, Cysts, and Localized Lung Disorders

Lewis S. Nelson, MD Professor and Chair, Department of Emergency Medicine; Director, Division of Medical Toxicology, Rutgers New Jersey Medical School, Newark, New Jersey Acute Poisoning

Jae K. Oh, MD Professor of Medicine, Mayo Clinic College of Medicine and Science, Rochester, Minnesota; Director, Heart Vascular Stroke Institute, Samsung Medical Center, Seoul, Gangnam, South Korea Pericardial Diseases

xxii

Contributors

Pablo C. Okhuysen, MD Professor of Infectious Diseases, Infection Control and Employee Health, University of Texas MD Anderson Cancer Center; Adjunct Professor of Infectious Diseases, Baylor College of Medicine; Adjunct Professor of Epidemiology, Human Genetics and Environmental Health, University of Texas School of Public Health; Adjunct Professor of Infectious Diseases, McGovern Medical School at the University of Texas Health Science Center at Houston, Houston, Texas Approach to the Patient with Suspected Enteric Infection Michael S. Okun, MD Professor and Chair of Neurology, Fixel Institute for Neurological Diseases, University of Florida College of Medicine, Gainesville, Florida Parkinsonism; Other Movement Disorders Jeffrey E. Olgin, MD Gallo-Chatterjee Distinguished Professor and Chief of Cardiology, University of California, San Francisco, School of Medicine, San Francisco, California Approach to the Patient with Suspected Arrhythmia Nancy J. Olsen, MD Professor of Medicine, Penn State Milton S. Hershey Medical Center, Hershey, Pennsylvania Biologic Agents and Signaling Inhibitors Walter A. Orenstein, MD, DSc Professor of Medicine, Pediatrics, Epidemiology & Global Health, Emory University School of Medicine; Associate Director, Emory Vaccine Center, Atlanta, Georgia Immunization John J. O’Shea, MD Scientific Director, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, Bethesda, Maryland Biologic Agents and Signaling Inhibitors Douglas R. Osmon, MD Professor of Medicine, Mayo Clinic College of Medicine and Science; Consultant, Division Infectious Disease, Mayo Clinic, Rochester, Minnesota Infections of Bursae, Joints, and Bones Catherine M. Otto, MD J. Ward Kennedy-Hamilton Endowed Chair in Cardiology and Professor of Medicine, University of Washington School of Medicine; Director, Heart Valve Clinic, Associate Director, Echocardiography, University of Washington Medical Center, Seattle, Washington Echocardiography Martin G. Ottolini, MD Professor of Pediatrics and Director, Capstone Student Research Program, Uniformed Services University of the Health Sciences; Consultant, Pediatric Infectious Diseases, Pediatrics, Walter Reed National Military Medical Center, Bethesda, Maryland Measles Peter G. Pappas, MD Professor of Medicine, University of Alabama at Birmingham School of Medicine, Birmingham, Alabama Candidiasis; Mycetoma and Dematiaceous Fungal Infections Ben Ho Park, MD, PhD The Donna S. Hall Professor of Medicine, Vanderbilt University School of Medicine; Co-Leader Breast Cancer Research; Director of Precision Oncology; Associate Director for Translational Research, Vanderbilt-Ingram Cancer Center, Nashville, Tennessee Cancer Biology and Genetics

Pankaj Jay Pasricha, MD Professor of Medicine and Neuroscience, Johns Hopkins University School of Medicine, Baltimore, Maryland Gastrointestinal Endoscopy Manisha Patel, MD, MS Measles, Mumps, Rubella, Herpesvirus, and Domestic Polio Epidemiology Team Lead, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia Mumps Robin Patel, MD Elizabeth P. and Robert E. Allen Professor of Individualized Medicine and Professor of Medicine and of Microbiology; Chair, Division of Clinical Microbiology; Consultant, Divisions of Clinical Microbiology and Infectious Diseases; Director, Infectious Diseases Research Laboratory, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Introduction to Microbial Disease: Pathophysiology and Diagnostics David L. Paterson, MBBS, PhD Professor of Medicine and Director, Centre for Clinical Research, University of Queensland, Herston, Queensland; Consultant Infectious Diseases Physician, Department of Infectious Diseases, Royal Brisbane and Women’s Hospital, Brisbane, Australia Infections Due to Other Members of the Enterobacteriaceae, Including Management of Multidrug-Resistant Strains Jean-Michel Pawlotsky, MD, PhD Professor, Department of Virology, Henri Mondor University Hospital, Creteil, France Acute Viral Hepatitis; Chronic Viral and Autoimmune Hepatitis Thomas H. Payne, MD Professor of Medicine, University of Washington School of Medicine; Medical Director, Information Technology Services, UW Medicine, Seattle, Washington Statistical Interpretation of Data and Using Data for Clinical Decisions Richard D. Pearson, MD Professor Emeritus of Medicine, University of Virginia School of Medicine, Charlottesville, Virginia Antiparasitic Therapy Trish M. Perl, MD, MSc Jay Sanford Professor of Medicine and Chief of Infectious Diseases and Geographic Medicine, University of Texas Southwestern Medical Center Dallas, Texas Enterococcal Infections Michael A. Pesce, PhD Professor Emeritus of Pathology and Cell Biology, Columbia University Vagelos College of Physicians and Surgeons, New York, New York Reference Intervals and Laboratory Values Brett W. Petersen, MD, MPH Epidemiology Team Lead, Poxvirus and Rabies Branch, Centers for Disease Control and Prevention, Atlanta, Georgia Smallpox, Monkeypox, and Other Poxvirus Infections William A. Petri, Jr., MD, PhD Wade Hampton Frost Professor of Epidemiology and Vice Chair for Research, Department of Medicine, University of Virginia School of Medicine, Charlottesville, Virginia Relapsing Fever and Other Borrelia Infections; African Sleeping Sickness; Amebiasis

Contributors

xxiii

Marc A. Pfeffer, MD, PhD Dzau Professor of Medicine, Harvard Medical School; Senior Physician, Brigham and Women’s Hospital, Boston, Massachusetts Heart Failure: Management and Prognosis

James D. Ralston, MD, MPH Senior Investigator, Kaiser Permanente Washington Health Research Institute, Seattle, Washington Comprehensive Chronic Disease Management

David S. Pisetsky, MD, PhD Professor of Medicine and Immunology, Duke University School of Medicine, Chief, Rheumatology, VA Medical Center, Durham, North Carolina Laboratory Testing in the Rheumatic Diseases

Stuart H. Ralston, MB ChB Professor of Rheumatology, University of Edinburgh, Edinburgh, United Kingdom Paget Disease of Bone

Frank Powell, PhD Professor of Medicine, University of California, San Diego, School of Medicine, La Jolla, California Disorders of Ventilatory Control Reed E. Pyeritz, MD, PhD Professor of Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania Inherited Diseases of Connective Tissue Thomas C. Quinn, MD, MSc Professor of Medicine and Pathology, Director, Center for Global Health, Johns Hopkins University School of Medicine; Associate Director, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Baltimore, Maryland Epidemiology and Diagnosis of Human Immunodeficiency Virus Infection and Acquired Immunodeficiency Syndrome Jai Radhakrishnan, MD, MS Professor of Medicine, Columbia University Vagelos College of Physicians and Surgeons; Clinical Chief, Division of Nephrology, Columbia University Irving Medical Center, New York, New York Glomerular Disorders and Nephrotic Syndromes Jerald Radich, MD Associate Professor of Medical Oncology, Clinical Research Division, Fred Hutchinson Cancer Research Center and University of Washington School of Medicine, Seattle, Washington Chronic Myeloid Leukemia Petros I. Rafailidis, MD, PhD, MSc Assistant Professor Internal Medicine-Infectious Diseases, Democritus University of Thrace; Beta University Department of Internal Medicine, University General Hospital of Greece, Alexandroupolis, Greece; Senior Researcher, Alfa Institute of Biomedical Sciences, Athens, Greece Pseudomonas and Related Gram-Negative Bacillary Infections Ganesh Raghu, MD Professor of Medicine and Laboratory Medicine (adjunct), University of Washington School of Medicine; Director, Center for Interstitial Lung Diseases, UW Medicine; Co-Director, Scleroderma Clinic, University of Washington Medical Center, Seattle, Washington Interstitial Lung Disease

Didier Raoult, MD, PhD Professor, Aix-Marseille Université, Faculté de Médecine, Chief, Institut Hospitalo-Universitaire Méditerranée-Infection, Marseille, France Bartonella Infections; Rickettsial Infections Adam J. Ratner, MD, MPH Associate Professor of Pediatrics and Microbiology and Chief, Division of Pediatric Infectious Diseases, New York University School of Medicine, New York, New York Haemophilus and Moraxella Infections Annette C. Reboli, MD Dean and Professor of Medicine, Cooper Medical School of Rowan University and Cooper University Hospital, Camden, New Jersey Erysipelothrix Infections K. Rajender Reddy, MD Ruimy Family President’s Distinguished Professor of Internal Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania Bacterial, Parasitic, Fungal, and Granulomatous Liver Diseases Donald A. Redelmeier, MD Professor of Medicine, University of Toronto; Canada Research Chair, Medical Decision Science; Senior Scientist, Evaluative Clinical Sciences, Sunnybrook Research Institute; Staff Physician, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada Postoperative Care and Complications Susan E. Reef, MD Medical Epidemiologist, Global Immunization Division, Centers for Disease Control and Prevention, Atlanta, Georgia Rubella (German Measles) John Reilly, MD Richard D. Krugman Endowed Chair and Dean, School of Medicine, and Vice Chancellor for Health Affairs, University of Colorado School of Medicine, Aurora, Colorado Chronic Obstructive Pulmonary Disease Megan E. Reller, MD, PhD Associate Professor of Medicine, Duke University School of Medicine, Durham, North Carolina Zoonoses

Margaret V. Ragni, MD, MPH Professor of Medicine, and Clinical Translational Science, University of Pittsburgh School of Medicine; Director, Hemophilia Center of Western Pennsylvania, Pittsburgh, Pennsylvania Hemorrhagic Disorders: Coagulation Factor Deficiencies

Neil M. Resnick, MD Thomas Detre Professor of Medicine and Chief, Division of Geriatric Medicine and Gerontology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania Urinary Incontinence

Srinivasa N. Raja, MD Professor of Anesthesiology, Critical Care Medicine, and Neurology; Director of Pain Research, Division of Pain Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland Pain

David B. Reuben, MD Archstone Professor and Chief, Division of Geriatrics, David Geffen School of Medicine at UCLA, Los Angeles, California Geriatric Assessment

S. Vincent Rajkumar, MD Edward W. and Betty Knight Scripps Professor of Medicine, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Plasma Cell Disorders

Jennifer G. Robinson, MD, MPH Professor of Epidemiology and Medicine, Director, Prevention Intervention Center, Department of Epidemiology, University of Iowa Carver College of Medicine, Iowa City, Iowa Disorders of Lipid Metabolism

xxiv

Contributors

Inez Rogatsky, PhD Professor of Microbiology and Immunology, Weill Cornell Medical College; Senior Scientist, Arthritis and Tissue Degeneration Program, Hospital for Special Surgery, New York, New York Immunomodulatory Drugs Joseph G. Rogers, MD Professor of Medicine, Duke University School of Medicine, Durham, North Carolina Heart Failure: Pathophysiology and Diagnosis Jean-Marc Rolain, PharmD, PhD Professor, Aix-Marseille Université and Institut Hospitalo-Universitaire Méditerranée Infection, Marseille, France Bartonella Infections Barrett J. Rollins, MD, PhD Linde Family Professor of Medicine, Dana-Farber Cancer Institute, Brigham & Women’s Hospital and Harvard Medical School, Boston, Massachusetts Histiocytoses José R. Romero, MD Horace C. Cabe Professor of Infectious Diseases, Department of Pediatrics, University of Arkansas for Medical Sciences; Director, Pediatric Infectious Diseases Section, Arkansas Children’s Hospital; Director, Clinical Trials Research, Arkansas Children’s Research Institute, Little Rock, Arkansas Enteroviruses Karen Rosene-Montella, MD President, Karen Rosene, LLC; Senior Consultant the Levinson Institute; Professor Emerita of Medicine, Warren Alpert Medical School at Brown University, Providence, Rhode Island Common Medical Problems in Pregnancy Philip J. Rosenthal, MD Professor of Medicine, University of California, San Francisco, School of Medicine, San Francisco, California Malaria James A. Russell, MD Professor of Medicine, University of British Columbia, Vancouver, British Columbia Shock Syndromes Related to Sepsis Anil K. Rustgi, MD Irving Professor of Medicine, Director, Herbert Irving Comprehensive Cancer Center; Chief, NewYork-Presbyterian Hospital/Columbia University Irving Medical Center Cancer Service, Columbia University Vagelos College of Physicians and Surgeons, New York, New York Neoplasms of the Esophagus and Stomach Daniel E. Rusyniak, MD Professor of Emergency Medicine, Indiana University School of Medicine, Indianapolis, Indiana Chronic Poisoning: Trace Metals and Others George Sakoulas, MD Associate Adjunct Professor, Division of Host-Microbe Systems & Therapeutics, University of California, San Diego, School of Medicine, La Jolla, California; Infectious Disease Consultant, Sharp Healthcare, San Diego, California Staphylococcal Infections Robert A. Salata, MD STERIS Chair of Excellence in Medicine, Professor and Chairman, Department of Medicine, Case Western Reserve University School of Medicine; Physician-in-Chief, Master Clinician in Infectious Diseases, University Hospitals Cleveland Medical Center, Cleveland, Ohio Brucellosis

Jane E. Salmon, MD Collette Kean Research Chair, Hospital for Special Surgery; Professor of Medicine, Weill Cornell Medical College, New York, New York Mechanisms of Immune-Mediated Tissue Injury Edsel Maurice T. Salvana, MD, DTM&H Associate Professor of Medicine and Director, Institute of Molecular Biology and Biotechnology, National Institutes of Health, University of the Philippines College of Medicine, Manila, Philippines Brucellosis Nanette Santoro, MD Professor and E. Stewart Taylor Chair, Department of Obstetrics and Gynecology, University of Colorado School of Medicine, Aurora, Colorado Menopause Renato M. Santos, MD Assistant Professor of Medicine, Emory University School of Medicine, Emory Heart and Vascular Center, John’s Creek, Georgia Vascular Disorders of the Kidney Peter A. Santucci, MD Professor of Medicine, Loyola University Medical Center, Maywood, Illinois Electrophysiologic Interventional Procedures and Surgery Patrice Savard, MD, MSc Assistant Professor of Microbiology and Immunology, Université de Montréal; Director, Unité de Prévention, Centre Hospitalier de l’Université de Montréal, Québec, Canada Enterococcal Infections Michael N. Sawka, PhD Professor, School of Biological Sciences, Georgia Institute of Technology, Atlanta, Georgia Disorders Due to Heat and Cold Paul D. Scanlon, MD Professor of Medicine, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Respiratory Testing and Function Andrew I. Schafer, MD Professor of Medicine, Director, Richard T. Silver Center for Myeloproliferative Neoplasms, Weill Cornell Medical College, New York, New York Approach to Medicine, the Patient, and the Medical Profession: Medicine as a Learned and Humane Profession; Thrombotic Disorders: Hypercoagulable States; Approach to the Patient with Bleeding and Thrombosis; Hemorrhagic Disorders: Disseminated Intravascular Coagulation, Liver Failure, and Vitamin K Deficiency William Schaffner, MD Professor of Preventive Medicine, Vanderbilt University School of Medicine, Nashville, Tennessee Tularemia and Other Francisella Infections W. Michael Scheld, MD Bayer-Gerald L. Mandell Professor of Infectious Diseases; Professor of Medicine; Clinical Professor of Neurosurgery; David A. Harrison Distinguished Educator, University of Virginia Health System, Charlottesville, Virginia Introduction to Microbial Disease: Pathophysiology and Diagnostics Manuel Schiff, MD, PhD Associate Professor of Pediatrics and Head of Metabolic Unit, Reference Center for Inborn Errors of Metabolism, Robert Debré University Hospital, Paris, France Homocystinuria and Hyperhomocysteinemia

Contributors Michael L. Schilsky, MD Professor of Medicine and Surgery, Yale University School of Medicine, New Haven, Connecticut Wilson Disease Robert T. Schooley, MD Professor of Medicine, University of California, San Diego, School of Medicine, San Diego, California Epstein-Barr Virus Infection David L. Schriger, MD, MPH Professor of Emergency Medicine, David Geffen School of Medicine at UCLA, Los Angeles, California Approach to the Patient with Abnormal Vital Signs Lynn M. Schuchter, MD Professor of Medicine, C. Willard Robinson Professor and Chair of the Division of Hematology-Oncology, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania Melanoma and Nonmelanoma Skin Cancers Sam Schulman, MD, PhD Professor of Medicine, McMaster University Michael G. DeGroote School of Medicine, Hamilton, Ontario, Canada Antithrombotic and Antiplatelet Therapy Lawrence B. Schwartz, MD, PhD Charles and Evelyn Thomas Professor of Medicine, Medical College of Virginia, Virginia Commonwealth University, Richmond, Virginia Systemic Anaphylaxis, Food Allergy, and Insect Sting Allergy Carlos Seas, MD, MSc Associate Professor of Medicine, Universidad Peruana Cayetano Heredia; Vice Director, Alexander von Humboldt Tropical Medicine Institute, Attending Physician, Infectious and Tropical Medicine, Hospital Nacional Cayetano Heredia, Lima, Peru Cholera and Other Vibrio Infections Steven A. Seifert, MD Professor of Emergency Medicine, University of New Mexico School of Medicine; Medical Director, New Mexico Poison and Drug Information Center, Albuquerque, New Mexico Envenomation, Bites, and Stings Julian Lawrence Seifter, MD James G. Haidas Distinguished Chair in Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts Potassium Disorders; Acid-Base Disorders Duygu Selcen, MD Professor of Neurology and Pediatrics, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Muscle Diseases Carol E. Semrad, MD Professor of Medicine, University of Chicago Pritzker School of Medicine, Chicago, Illinois Approach to the Patient with Diarrhea and Malabsorption Harry Shamoon, MD Professor of Medicine and Senior Associate Dean for Clinical & Translational Research, Albert Einstein College of Medicine; Director, Harold and Muriel Block Institute for Clinical and Translational Research at Einstein and Montefiore, Bronx, New York Diabetes Mellitus Pamela J. Shaw, DBE, MBBS, MD Professor of Neurology, Sheffield Institute for Translational Neuroscience, University of Sheffield, Sheffield, United Kingdom Amyotrophic Lateral Sclerosis and Other Motor Neuron Diseases

xxv

Beth H. Shaz, MD Chief Medical and Scientific Officer, New York Blood Center; Adjunct Assistant Professor, Department of Pathology and Cell Biology, Columbia University Vagelos College of Physicians and Surgeons, New York, New York Transfusion Medicine Robert L. Sheridan, MD Professor of Surgery, Harvard Medical School and Massachusetts General Hospital, COL (ret), U.S. Army, Boston, Massachusetts Medical Aspects of Trauma and Burns Stuart Sherman, MD Glen A. Lehman Professor of Gastroenterology and Professor of Medicine and Radiology; Clinical Director of Gastroenterology and Hepatology, Indiana University School of Medicine, Indianapolis, Indiana Diseases of the Gallbladder and Bile Ducts Wun-Ju Shieh, MD, MPH, PhD Deputy Chief/Medical Officer, Infectious Diseases Pathology Branch, Centers for Disease Control and Prevention, Atlanta, Georgia Leptospirosis Michael E. Shy, MD Professor of Neurology and Pediatrics, University of Iowa Carver College of Medicine, Iowa City, Iowa Peripheral Neuropathies Ellen Sidransky, MD Chief, Section of Molecular Neurogenetics, Medical Genetics Branch, NHGRI, National Institutes of Health, Bethesda, Maryland Lysosomal Storage Diseases Richard M. Siegel, MD, PhD Clinical Director and Chief, Autoimmunity Branch, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, Bethesda, Maryland The Systemic Autoinflammatory Diseases Costi D. Sifri, MD Professor of Medicine and Medical Director, Immunocompromised Infectious Diseases Program, University of Virginia Health System, Charlottesville, Virginia Approach to Fever and Suspected Infection in the Immunocompromised Host Robert F. Siliciano, MD, PhD Professor of Medicine, Johns Hopkins University School of Medicine; Investigator, Howard Hughes Medical Institute, Baltimore, Maryland Immunopathogenesis of Human Immunodeficiency Virus Infection Michael S. Simberkoff, MD Professor of Medicine, New York University School of Medicine and Chief of Staff, VA New York Harbor Healthcare System, New York, New York Haemophilus and Moraxella Infections David L. Simel, MD, MHS Professor of Medicine, Duke University School of Medicine; Chief of Medical Service, Durham Veterans Affairs Medical Center, Durham, North Carolina Approach to the Patient: History and Physical Examination Karl Skorecki, MD Professor and Dean, Azriel Faculty of Medicine, Bar-Ilan University, Ramat Gan, Israel Regenerative Medicine, Cell, and Gene Therapies

xxvi

Contributors

Arthur S. Slutsky, CM, MD Professor of Medicine, Director, Interdepartmental Division of Critical Care Medicine, University of Toronto; Vice President (Research), St Michael’s Hospital; Keenan Research Centre, Li Ka Shing Knowledge Institute, Toronto, Ontario, Canada Mechanical Ventilation Eric J. Small, MD Professor of Medicine, Deputy Director and Chief Scientific Officer, UCSF Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, School of Medicine, San Francisco, California Prostate Cancer Gerald W. Smetana, MD Professor of Medicine, Harvard Medical School and Physician, Division of General Medicine and Primary Care, Beth Israel Deaconess Medical Center, Boston, Massachusetts Principles of Medical Consultation Gordon Smith, MD Professor and Chair of Neurology, Medical College of Virginia, Virginia Commonwealth University, Richmond, Virginia Peripheral Neuropathies Frederick S. Southwick, MD Professor of Medicine, University of Florida College of Medicine, Gainesville, Florida Nocardiosis Allen M. Spiegel, MD Dean Emeritus and Professor of Medicine, Albert Einstein College of Medicine, Bronx, New York Principles of Endocrinology; Polyglandular Disorders Robert Spiera, MD Professor of Clinical Medicine, Weill Cornell Medical College; Director, Scleroderma, Vasculitis, & Myositis Center, Hospital for Special Surgery, New York, New York Giant Cell Arteritis and Polymyalgia Rheumatica Stanley M. Spinola, MD Professor of Medicine, Microbiology and Immunology, Pathology and Laboratory Medicine and Chair, Microbiology and Immunology, Indiana University School of Medicine, Indianapolis, Indiana Chancroid Sally P. Stabler, MD Professor of Medicine and Cleo Scott & Mitchell Vincent Allen Chair in Hematology Research, University of Colorado School of Medicine, Aurora, Colorado Megaloblastic Anemias Stephanie M. Stanford, PhD Assistant Professor of Medicine, University of California, San Diego, School of Medicine, La Jolla, California Mechanisms of Inflammation and Tissue Repair

Theodore S. Steiner, MD Professor and Associate Head, Division of Infectious Diseases, University of British Columbia, Vancouver, British Columbia, Canada Escherichia Coli Enteric Infections David S. Stephens, MD Stephen W. Schwarzmann Distinguished Professor of Medicine and Chair, Department of Medicine, Emory University School of Medicine, Atlanta, Georgia Neisseria Meningitidis Infections David A. Stevens, MD Professor of Medicine, Stanford University School of Medicine, Stanford, California; President and Principal Investigator, Infectious Diseases Research Laboratory, California Institute for Medical Research, San Jose, California Systemic Antifungal Agents Dennis L. Stevens, PhD, MD Professor of Medicine, University of Washington School of Medicine, Seattle, Washington; Research & Development Service, Veterans Affairs Medical Center, Boise, Idaho Nonpneumococcal Streptococcal Infections and Rheumatic Fever James K. Stoller, MD, MS Professor and Chairman, Education Institute, Jean Wall Bennett Professor of Medicine, Samson Global Leadership Endowed Chair, Cleveland Clinic Lerner College of Medicine, Cleveland Clinic, Cleveland, Ohio Respiratory Monitoring in Critical Care John H. Stone, MD, MPH Professor of Medicine, Harvard Medical School, Director, Clinical Rheumatology, Massachusetts General Hospital, Boston, Massachusetts The Systemic Vasculitides Richard M. Stone, MD Professor of Medicine, Harvard Medical School; Chief of the Medical Staff, Dana-Farber Cancer Institute, Boston, Massachusetts Myelodysplastic Syndromes Raymond A. Strikas, MD, MPH Medical Officer, Immunization Services Division, Centers for Disease Control and Prevention, Atlanta, Georgia Immunization Edwin P. Su, MD Associate Professor of Clinical Orthopaedics, Weill Cornell Medical College; Associate Attending Orthopaedic Surgeon, Hospital for Special Surgery, New York, New York Surgical Treatment of Joint Diseases Roland W. Sutter, MD, MPH&TM Special Adviser to Director, Polio Eradication Department, World Health Organization, Geneva, Switzerland Diphtheria and Other Corynebacterium Infections

Paul Stark, MD Professor Emeritus of Radiology, University of California, San Diego, School of Medicine; Chief of Cardiothoracic Radiology, VA San Diego Healthcare System, La Jolla, California Imaging in Pulmonary Disease

Ronald S. Swerdloff, MD Professor of Medicine, David Geffen School of Medicine at UCLA, Los Angeles, California; Chief, Division of Endocrinology, Metabolism and Nutrition, Harbor-UCLA Medical Center, Senior Investigator, Los Angeles Biomedical Research Institute, Torrance, California The Testis and Male Hypogonadism, Infertility, and Sexual Dysfunction

David P. Steensma, MD Associate Professor of Medicine, Harvard Medical School and Physician, Dana-Farber Cancer Institute, Boston, Massachusetts Myelodysplastic Syndromes

Heidi Swygard, MD, MPH Professor of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina Approach to the Patient with a Sexually Transmitted Infection

Contributors Megan Sykes, MD Michael J. Friedlander Professor of Medicine, Director, Columbia Center for Translational Immunology, Columbia University Vagelos College of Physicians and Surgeons, New York, New York Transplantation Immunology H. Keipp Talbot, MD, MPH Associate Professor of Medicine, Vanderbilt University School of Medicine, Nashville, Tennessee Respiratory Syncytial Virus Marian Tanofsky-Kraff, PhD Professor of Medical and Clinical Psychology and of Medicine, Uniformed Services University of the Health Sciences, Bethesda, Maryland Eating Disorders Susan M. Tarlo, MBBS Professor of Medicine, University of Toronto; Respiratory Physician, University Health Network, Toronto, Ontario, Canada Occupational Lung Disease Paul S. Teirstein, MD Chief of Cardiology; Director, Interventional Cardiology, Scripps Clinic, La Jolla, California Interventional and Surgical Treatment of Coronary Artery Disease Sam R. Telford, III, ScD Professor of Infectious Disease and Global Health, Tufts University School of Veterinary Medicine, North Grafton, Massachusetts Babesiosis and Other Protozoan Diseases Rajesh V. Thakker, MD May Professor of Medicine, Radcliffe Department of Medicine, University of Oxford, Oxford, United Kingdom The Parathyroid Glands, Hypercalcemia, and Hypocalcemia Judith Therrien, MD Professor of Medicine, Jewish General Hospital, Montreal, Quebec, Canada Congenital Heart Disease in Adults George R. Thompson, III, MD Associate Professor of Clinical Medicine, University of California, Davis School of Medicine, Davis, California Endemic Mycoses Antonella Tosti, MD Fredric Brandt Endowed Professor of Dermatology, Dr. Phillip Frost Department of Dermatology and Cutaneous Surgery, University of Miami Miller School of Medicine, Miami, Florida Diseases of Hair and Nails Indi Trehan, MD, MPH, DTM&H Associate Professor of Pediatrics, Washington University School of Medicine in St. Louis, St. Louis, Missouri; Executive Director and Medical Director, Lao Friends Hospital for Children, Luang Prabang, Lao People’s Democratic Republic Protein-Energy Malnutrition Ronald B. Turner, MD Professor of Pediatrics, University of Virginia School of Medicine, Charlottesville, Virginia The Common Cold Anthony Michael Valeri, MD Associate Professor of Medicine, Vagelos College of Physicians and Surgeons; Medical Director, Hemodialysis, Columbia University Irving Medical Center, New York, New York Treatment of Irreversible Renal Failure

xxvii

John Varga, MD John and Nancy Hughes Distinguished Professor of Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois Systemic Sclerosis (Scleroderma) Bradley V. Vaughn, MD Professor of Neurology, University of North Carolina, Chapel Hill, North Carolina Sleep Disorders Alan P. Venook, MD Professor of Clinical Medicine, University of California, San Francisco, School of Medicine, San Francisco, California Liver and Biliary Tract Cancers Joseph G. Verbalis, MD Professor of Medicine, Georgetown University; Chief, Endocrinology and Metabolism, Georgetown University Hospital, Washington, D.C. Posterior Pituitary Ronald G. Victor, MD† Formerly Burns & Allen Professor of Medicine, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles, California Arterial Hypertension Angela Vincent, MBBS, MSc Emeritus Professor, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom Disorders of Neuromuscular Transmission Tonia L. Vincent, PhD Professor of Musculoskeletal Biology, Arthritis Research UK Senior Fellow and Consultant Rheumatologist; Director, Arthritis Research UK Centre for Osteoarthritis Pathogenesis, University of Oxford, Oxford, England Osteoarthritis Robert M. Wachter, MD Holly Smith Professor and Chairman, Department of Medicine, University of California, San Francisco, School of Medicine, San Francisco, California Quality, Safety, and Value Edward H. Wagner, MD, MPH Director Emeritus, MacColl Center for Health Care Innovation, Group Health Research Institute, Seattle, Washington Comprehensive Chronic Disease Management Edward E. Walsh, MD Professor of Medicine, University of Rochester School of Medicine & Dentistry; Unit Chief, Infectious Diseases, Rochester General Hospital, Rochester, New York Respiratory Syncytial Virus Thomas J. Walsh, MD Professor of Medicine, Pediatrics, Microbiology & Immunology and Chief, Infectious Diseases Translational Research Laboratory, Weill Cornell Medical College, New York, New York; Adjunct Professor of Pathology, Johns Hopkins University School of Medicine; Adjunct Professor of Medicine, University of Maryland School of Medicine, Baltimore, Maryland Aspergillosis Jeremy D. Walston, MD Raymond and Anna Lublin Professor of Geriatric Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland Common Clinical Sequelae of Aging



Deceased.

xxviii

Contributors

Roland B. Walter, MD, PhD, MS Associate Professor of Medicine, University of Washington School of Medicine and Associate Member, Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, Washington The Acute Leukemias Christina Wang, MD Professor of Medicine, David Geffen School of Medicine at UCLA, Los Angeles, California; Clinical and Translational Science Institute, Los Angeles Biomedical Research Institute and Division of Endocrinology, Department of Medicine, Harbor-UCLA Medical Center, Torrance, California The Testis and Male Hypogonadism, Infertility, and Sexual Dysfunction Lorraine B. Ware, MD Professor of Medicine, Pathology, Microbiology, and Immunology, Vanderbilt University School of Medicine, Nashville, Tennessee Acute Respiratory Failure Cirle A. Warren, MD Associate Professor of Medicine, University of Virginia School of Medicine, Charlottesville, Virginia Cryptosporidiosis John T. Watson, MD, MSc Respiratory Viruses Branch, Division of Viral Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia Coronaviruses Thomas J. Weber, MD Associate Professor of Medicine, Duke University School of Medicine, Durham, North Carolina Approach to the Patient with Metabolic Bone Disease; Osteoporosis Geoffrey A. Weinberg, MD Professor of Pediatrics, University of Rochester School of Medicine & Dentistry; Director, Clinical Pediatric Infectious Diseases & Pediatric HIV Program, Golisano Children’s Hospital, University of Rochester Medical Center, Rochester, New York Parainfluenza Viral Disease David A. Weinstein, MD, MMSc Professor of Pediatrics, University of Connecticut School of Medicine, Farmington, Connecticut; Director, Glycogen Storage Disease Program, Connecticut Children’s Medical Center, Hartford, Connecticut Glycogen Storage Diseases Robert S. Weinstein, MD Professor of Medicine, University of Arkansas for Medical Sciences; Staff Endocrinologist, Central Arkansas Veterans Health Care System, Little Rock, Arkansas Osteomalacia and Rickets Roger D. Weiss, MD Professor of Psychiatry, Harvard Medical School, Boston, Massachusetts; Chief, Division of Alcohol and Drug Abuse, McLean Hospital, Belmont, Massachusetts Drugs of Abuse Roy E. Weiss, MD, PhD Kathleen & Stanley Glaser Distinguished Chair and Chairman, Department of Medicine, University of Miami Miller School of Medicine, Miami, Florida; Esformes Professor Emeritus, Department of Medicine, University of Chicago Pritzker School of Medicine, Chicago, Illinois Neuroendocrinology and the Neuroendocrine System; Anterior Pituitary Jeffrey I. Weitz, MD Professor of Medicine & Biochemistry, McMaster University Michael G. DeGroote School of Medicine; Executive Director, Thrombosis & Atherosclerosis Research Institute, Hamilton, Ontario, Canada Venous Thrombosis and Embolism

Richard P. Wenzel, MD, MSc Professor and Former Chairman, Internal Medicine, Medical College of Virginia, Virginia Commonwealth University, Richmond, Virginia Acute Bronchitis and Tracheitis Victoria P. Werth, MD Professor of Dermatology, University of Pennsylvania Perelman School of Medicine; Chief of Dermatology, Corporal Michael J. Crescenz VAMC, Philadelphia, Pennsylvania Principles of Therapy of Skin Diseases Sterling G. West, MD Professor of Medicine, University of Colorado School of Medicine, Aurora, Colorado Systemic Diseases in Which Arthritis Is a Feature A. Clinton White, Jr., MD Professor of Internal Medicine, University of Texas Medical Branch, Galveston, Texas Cestodes Christopher J. White, MD Chairman and Professor of Medicine, Ochsner Clinical School of the University of Queensland, Ochsner Medical Institutions, New Orleans, Louisiana Atherosclerotic Peripheral Arterial Disease Julian White, MBBS, MD Professor and Head, Toxinology Department, Women’s & Children’s Hospital, North Adelaide, South Australia, Australia Envenomation, Bites, and Stings Perrin C. White, MD Professor of Pediatrics, University of Texas Southwestern Medical Center; Chief of Endocrinology, Children’s Medical Center, Dallas, Texas Sexual Development and Identity Richard J. Whitley, MD Distinguished Professor of Pediatrics, Loeb Eminent Scholar Chair in Pediatrics, Professor of Microbiology, Medicine, and Neurosurgery, Pediatrics, University of Alabama at Birmingham School of Medicine, Birmingham, Alabama Herpes Simplex Virus Infections Michael P. Whyte, MD Professor of Medicine, Pediatrics, and Genetics, Washington University School of Medicine in St. Louis; Medical-Scientific Director, Center for Metabolic Bone Disease and Molecular Research, Shriners Hospital for Children, St. Louis, Missouri Osteonecrosis, Osteosclerosis/Hyperostosis, and Other Disorders of Bone Samuel Wiebe, MD, MSc Professor of Clinical Neurosciences, Community Health Sciences and Pediatrics, University of Calgary Cumming School of Medicine, Calgary, Alberta, Canada The Epilepsies Jeanine P. Wiener-Kronish, MD Henry Isaiah Dorr Professor of Research and Teaching in Anaesthesia, Department of Anesthesia, Critical Care and Pain Medicine, Harvard Medical School; Anesthestist-in-Chief, Massachusetts General Hospital, Boston, Massachusetts Overview of Anesthesia David J. Wilber, MD George M Eisenberg Professor of Medicine, Loyola University Chicago Stritch School of Medicine; Director, Division of Cardiology, Loyola University Medical Center, Maywood, Illinois Electrophysiologic Interventional Procedures and Surgery

Contributors Beverly Winikoff, MD, MPH President, Gynuity Health Projects; Professor of Clinical Population and Family Health, Population and Family Health, Columbia University Mailman School of Public Health, New York, New York Contraception Jane N. Winter, MD Professor of Medicine, Robert H Lurie Comprehensive Cancer Center and the Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois Approach to the Patient with Lymphadenopathy and Splenomegaly Edward M. Wolin, MD Professor of Medicine, Albert Einstein College of Medicine; Director, Neuroendocrine Tumor Program, Department of Medical Oncology, Montefiore Einstein Center for Cancer Care, Bronx, New York Neuroendocrine Tumors Gary P. Wormser, MD Professor of Medicine and of Microbiology and Immunology and Pharmacology, New York Medical College; Chief, Division of Infectious Diseases, Valhalla, New York Lyme Disease Neal S. Young, MD Chief, Hematology Branch, NHLBI, National Heart, Lung, and Blood Institute, Bethesda, Maryland Parvovirus Vincent B. Young, MD, PhD William Henry Fitzbutler Professor of Internal Medicine/Infectious Diseases, Professor of Microbiology & Immunology, University of Michigan Medical School, Ann Arbor, Michigan The Human Microbiome

xxix

William F. Young, Jr., MD, MSc Professor of Medicine, Tyson Family Endocrinology Clinical Professor, Mayo Clinic College of Medicine and Science, Rochester, Minnesota Adrenal Medulla, Catecholamines, and Pheochromocytoma Alan S.L. Yu, MB BChir Harry Statland and Solon Summerfield Professor, University of Kansas Medical Center; Director, The Kidney Institute, University of Kansas Medical Center, Kansas City, Kansas Disorders of Magnesium and Phosphorus Anita K. M. Zaidi, MBBS, SM Director, Enteric and Diarrheal Diseases; and Vaccine Development and Surveillance, Bill and Melinda Gates Foundation, Seattle, Washington Shigellosis Sherif Zaki, MD, PhD Chief, Infectious Diseases Pathology Branch, Centers for Disease Control and Prevention, Atlanta, Georgia Leptospirosis Thomas R. Ziegler, MD Department of Medicine, Division of Endocrinology, Metabolism and Lipids, Emory University School of Medicine, Atlanta, Georgia Malnutrition: Assessment and Support Peter Zimetbaum, MD Richard and Susan Smith Professor of Cardiovascular Medicine, Harvard Medical School; Associate Chief and Director of Clinical Cardiology, Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts Supraventricular Cardiac Arrhythmias

VIDEO CONTENTS This icon appears throughout the book to indicate chapters with accompanying video available on Expertconsult.com. AGING AND GERIATRIC MEDICINE Confusion Assessment Method Chapter 25, Video 1 – SHARON K. INOUYE

CARDIOVASCULAR DISEASE Standard Echocardiographic Views Chapter 49, Video 1 – CATHERINE M. OTTO Dilated Cardiomyopathy Chapter 49, Video 2 – CATHERINE M. OTTO Three-Dimensional Echocardiography Chapter 49, Video 3 – CATHERINE M. OTTO Stress Echocardiography Chapter 49, Video 4 – CATHERINE M. OTTO Pericardial Effusion Chapter 49, Video 5 – CATHERINE M. OTTO Secundum Atrial Septal Defect Chapter 61, Video 1 – ARIANE J. MARELLI Perimembranous Ventricular Septal Defect Chapter 61, Video 2 – ARIANE J. MARELLI Coronary Stent Placement Chapter 65, Video 1 – PAUL S. TEIRSTEIN Guidewire Passage Chapter 65, Video 2 – PAUL S. TEIRSTEIN Delivering the Stent Chapter 65, Video 3 – PAUL S. TEIRSTEIN Inflating the Stent Chapter 65, Video 4 – PAUL S. TEIRSTEIN Final Result Chapter 65, Video 5 – PAUL S. TEIRSTEIN

RESPIRATORY DISEASES Wheezing Chapter 81, Video 1 – JEFFREY M. DRAZEN Inhaler Use Chapter 81, Video 2 – JEFFREY M. DRAZEN VATS Wedge Resection Chapter 93, Video 1 – MALCOLM M. DeCAMP

CRITICAL CARE MEDICINE Ventilation of an Ex Vivo Rat Lung Chapter 97, Video 1 – ARTHUR S. SLUTSKY, GEORGE VOLGYESI, AND TOM WHITEHEAD

GASTROINTESTINAL DISEASES Ulcerative Colitis, Moderately Severe Chapter 132, Video 1 – GARY R. LICHTENSTEIN

DISEASES OF THE LIVER, GALLBLADDER, AND BILE DUCTS Endoscopic Ultrasound of Biliary Ascariasis, Endoscopic Retrograde Cholangiopancreatography of Biliary Ascariasis, and Extraction of the Helminthic Worm Chapter 142, Video 1 – K. RAJENDER REDDY

ONCOLOGY Laparascopic Assisted Double Balloon Enteroscopy with Polypectomy of a Jejunal Adenoma Followed by Surgical Oversew of the Polypectomy Site Chapter 184, Video 1 – SHABANA PASHA

Snare Polypectomy of a Colon Adenoma Chapter 184, Video 2 – JONATHAN LEIGHTON Endoscopic Mucosal Resection Using Saline Lift Polypectomy of a Colon Adenoma, Followed by Closure of the Mucosal Defect with Clips Chapter 184, Video 3 – WAJEEH SALAH

ENDOCRINE DISEASES Pituitary Surgery Chapter 211, Video 1 – IVAN CIRIC

NEUROLOGY Focal Seizure Disorder—Temporal Lobe Epilepsy Chapter 368, Video 1 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Generalized Seizure Disorder—Myoclonic Epilepsy Chapter 368, Video 2 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Pronator Drift Chapter 368, Video 3 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Basal Ganglia: Parkinsonism Chapter 368, Video 4 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Brainstem (Medial Longitudinal Fasciculus): Internuclear Ophthalmoplegia (INO) Chapter 368, Video 5 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Cerebellum and Spinal Cord: Spastic-Ataxic Gait Chapter 368, Video 6 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Sustained Clonus Chapter 368, Video 7 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Anterior Horn Cell and Axon: Fasciculations (Tongue and Lower Limb) Chapter 368, Video 8 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Brachial Plexus: Brachial Plexopathy Chapter 368, Video 9 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Peripheral Nerve: Length-Dependent Peripheral Neuropathy Chapter 368, Video 10 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Neuromuscular Junction: Fatigable Ptosis, Dysarthria, and Dysphonia Chapter 368, Video 11 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Muscle: Myotonic Dystrophy Chapter 368, Video 12 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Facioscapulohumeral Muscular Dystrophy Chapter 368, Video 13 – GABRIELE C. DeLUCA AND ROBERT C. GRIGGS Cervical Provocation Chapter 372, Video 1 – J.D. BARTLESON AND RICHARD L. BARBANO Spurling Maneuver Chapter 372, Video 2 – J.D. BARTLESON AND RICHARD L. BARBANO Cervical Distraction Test Chapter 372, Video 3 – J.D. BARTLESON AND RICHARD L. BARBANO Straight Leg Raise Chapter 372, Video 4 – J.D. BARTLESON AND RICHARD L. BARBANO Contralateral Straight Leg Raise Chapter 372, Video 5 – J.D. BARTLESON AND RICHARD L. BARBANO Seated Straight Leg Raise Chapter 372, Video 6 – J.D. BARTLESON AND RICHARD L. BARBANO Discectomy Chapter 372, Video 7 – J.D. BARTLESON AND RICHARD L. BARBANO

xl

Video Contents

Typical Absence Seizure in a 19-Year-Old Woman (Generalized Absence Seizure) Chapter 375, Video 1 – SAMUEL WIEBE Left Motor Rolandic Seizure Evolving to Bilateral Tonic-Clonic Seizure in a Woman with Post-Traumatic Epilepsy (Left Focal Motor to Bilateral Tonic Clonic Seizure) Chapter 375, Video 2 – SAMUEL WIEBE Left Temporal Focal Impaired Awareness Seizure Chapter 375, Video 3 – SAMUEL WIEBE Left Temporal Focal Impaired Awareness Seizure - Postictal Confusion Chapter 375, Video 4 – SAMUEL WIEBE Left Temporal Focal Impaired Awareness Seizure, to Bilateral Tonic Clonic Chapter 375, Video 5 – SAMUEL WIEBE Focal Right Supplementary Sensory-Motor Seizure in a Patient with a Normal MRI (Focal Motor Aware) Chapter 375, Video 6 – SAMUEL WIEBE Right Posterior Temporal Seizure - Dramatic Hyperkinetic Semiology Chapter 375, Video 7 – SAMUEL WIEBE Right Mesial Frontal Hyperkinetic Seizure Chapter 375, Video 8 – SAMUEL WIEBE Nonconvulsive Generalized Status Epilepticus Chapter 375, Video 9 – SAMUEL WIEBE Generalized Tonic-Clonic Seizure, Tonic Phase Chapter 375, Video 10 – SAMUEL WIEBE Generalized Tonic-Clonic Seizure, Clonic Phase Chapter 375, Video 11 – SAMUEL WIEBE Generalized Myoclonic Seizure Involving the Face in a Patient with Juvenile Myoclonic Epilepsy Chapter 375, Video 12 – SAMUEL WIEBE Tonic Seizure in a Patient with Lennox-Gastaut Syndrome (Generalized Tonic Seizure, Could Also Be Classified as Epileptic Spasm) Chapter 375, Video 13 – SAMUEL WIEBE Atonic Seizure in a Patient with Lennox-Gastaut Syndrome (Generalized Atonic Seizure) Chapter 375, Video 14 – SAMUEL WIEBE Reflex Auditory Seizure in a Patient with Normal MRI (Focal Reflex Impaired Awareness Seizure) Chapter 375, Video 15 – SAMUEL WIEBE Early Parkinson Disease Chapter 381, Video 1 – MICHAEL S. OKUN AND ANTHONY E. LANG Freezing of Gait in Parkinson Disease Chapter 381, Video 2 – MICHAEL S. OKUN AND ANTHONY E. LANG Gunslinger Gait in Progressive Supranuclear Palsy Chapter 381, Video 3 – MICHAEL S. OKUN AND ANTHONY E. LANG Supranuclear Gaze Palsy in Progressive Supranuclear Palsy Chapter 381, Video 4 – MICHAEL S. OKUN AND ANTHONY E. LANG Applause Sign in Progressive Supranuclear Palsy Chapter 381, Video 5 – MICHAEL S. OKUN AND ANTHONY E. LANG Apraxia of Eyelid Opening in Progressive Supranuclear Palsy Chapter 381, Video 6 – MICHAEL S. OKUN AND ANTHONY E. LANG Cranial Dystonia in Multiple-System Atrophy Chapter 381, Video 7 – MICHAEL S. OKUN AND ANTHONY E. LANG Anterocollis in Multiple-System Atrophy Chapter 381, Video 8 – MICHAEL S. OKUN AND ANTHONY E. LANG Stridor in Multiple-System Atrophy Chapter 381, Video 9 – MICHAEL S. OKUN AND ANTHONY E. LANG

Alien Limb Phenomenon in Corticobasal Syndrome Chapter 381, Video 10 – MICHAEL S. OKUN AND ANTHONY E. LANG Myoclonus in Corticobasal Syndrome Chapter 381, Video 11 – MICHAEL S. OKUN AND ANTHONY E. LANG Levodopa-Induced Dyskinesia in Parkinson Disease Chapter 381, Video 12 – MICHAEL S. OKUN AND ANTHONY E. LANG Essential Tremor Chapter 382, Video 1 – MICHAEL S. OKUN AND ANTHONY E. LANG Huntington’s Disease Chapter 382, Video 2 – MICHAEL S. OKUN AND ANTHONY E. LANG Hemiballism Chapter 382, Video 3 – MICHAEL S. OKUN AND ANTHONY E. LANG Blepharospasm Chapter 382, Video 4 – MICHAEL S. OKUN AND ANTHONY E. LANG Oromandibular Dystonia Chapter 382, Video 5 – MICHAEL S. OKUN AND ANTHONY E. LANG Cervical Dystonia Chapter 382, Video 6 – MICHAEL S. OKUN AND ANTHONY E. LANG Writer’s Cramp Chapter 382, Video 7 – MICHAEL S. OKUN AND ANTHONY E. LANG Embouchure Dystonia Chapter 382, Video 8 – MICHAEL S. OKUN AND ANTHONY E. LANG Sensory Trick in Cervical Dystonia Chapter 382, Video 9 – MICHAEL S. OKUN AND ANTHONY E. LANG Generalized Dystonia Chapter 382, Video 10 – MICHAEL S. OKUN AND ANTHONY E. LANG Tics Chapter 382, Video 11 – MICHAEL S. OKUN AND ANTHONY E. LANG Tardive Dyskinesia Chapter 382, Video 12 – MICHAEL S. OKUN AND ANTHONY E. LANG Hemifacial Spasm Chapter 382, Video 13 – MICHAEL S. OKUN AND ANTHONY E. LANG Wernicke Encephalopathy Eye Movements: Before Thiamine Chapter 388, Video 1 – BARBARA S. KOPPEL Wernicke Encephalopathy Eye Movements: After Thiamine Chapter 388, Video 2 – BARBARA S. KOPPEL Central Pontine Myelinolysis: Man with Slow, Dysconjugate Horizontal Eye Movements Chapter 388, Video 3 – BARBARA S. KOPPEL Limb Symptoms and Signs in ALS Chapter 391, Video 1 – PAMELA J. SHAW Bulbar Symptoms and Signs in ALS Chapter 391, Video 2 – PAMELA J. SHAW Video Fluoroscopy of Normal Swallowing and of Swallowing in an ALS Patient with Bulbar Dysfunction Chapter 391, Video 3 – PAMELA J. SHAW Charcot-Marie-Tooth Disease Examination and Walk Chapter 392, Video 1 – GORDON SMITH AND MICHAEL E. SHY

EYE, EAR, NOSE, AND THROAT DISEASES Skin Testing Chapter 398, Video 1 – LARRY BORISH Nasal Endoscopy Chapter 398, Video 2 – LARRY BORISH

2

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION: MEDICINE AS A LEARNED AND HUMANE PROFESSION LEE GOLDMAN AND ANDREW I. SCHAFER

  APPROACH TO MEDICINE

Medicine is a profession that incorporates science and the scientific method with the art of being a physician. The art of tending to the sick is as old as humanity itself. Even in modern times, the art of caring and comforting, guided by millennia of common sense as well as a more recent, systematic approach to medical ethics (Chapter 2), remains the cornerstone of medicine. Without these humanistic qualities, the application of the modern science of medicine is suboptimal, ineffective, or even detrimental. The caregivers of ancient times and premodern cultures tried a variety of interventions to help the afflicted. Some of their potions contained what are now known to be active ingredients that form the basis for proven medications (Chapter 26). Others (Chapter 34) have persisted into the present era despite a lack of convincing evidence. Modern medicine should not dismiss the possibility that these unproven approaches may be helpful; instead, it should adopt a guiding principle that all interventions, whether traditional or newly developed, can be tested vigorously, with the expectation that any beneficial effects can be explored further to determine their scientific basis. When compared with its long and generally distinguished history of caring and comforting, the scientific basis of medicine is remarkably recent. Other than an understanding of human anatomy and the later description, albeit widely contested at the time, of the normal physiology of the circulatory system, almost all of modern medicine is based on discoveries made within the past 150 years, during which human life expectancy has more than doubled.1 Until the late 19th century, the paucity of medical knowledge was also exemplified best by hospitals and hospital care. Although hospitals provided caring that all but well-to-do people might not be able to obtain elsewhere, there is little if any evidence that hospitals improved health outcomes. The term hospitalism referred not to expertise in hospital care but rather to the aggregate of iatrogenic and nosocomial afflictions that were induced by the hospital stay itself. The essential humanistic qualities of caring and comforting can achieve full benefit only if they are coupled with an understanding of how medical science can and should be applied to patients with known or suspected diseases. Without this knowledge, comforting may be inappropriate or misleading, and caring may be ineffective or counterproductive if it inhibits a sick person from obtaining appropriate, scientific medical care. Goldman-Cecil Medicine focuses on the discipline of internal medicine, from which neurology and dermatology, which are also covered in substantial detail in this text, are relatively recent evolutionary branches. The term internal medicine, which is often misunderstood by the lay public, was developed in 19th-century Germany. Inneren medizin was to be distinguished from clinical medicine because it emphasized the physiology and chemistry of disease, not just the patterns or progression of clinical manifestations. Goldman-Cecil Medicine follows this tradition by showing how pathophysiologic abnormalities cause symptoms and signs and by emphasizing how therapies can modify the underlying pathophysiology and improve the patient’s well-being. Modern medicine has moved rapidly past organ physiology to an increasingly detailed understanding of cellular, subcellular, and genetic mechanisms. For example, the understanding of microbial pathogenesis and many inflammatory diseases (Chapter 241) is now guided by a detailed understanding of the human immune system and its response to foreign antigens (Chapters 39 to 44). Advances in our understanding of the human microbiome raise the possibility that our complex interactions with microbes, which outnumber our cells by a factor of 10, will help explain conditions ranging from inflammatory bowel disease (Chapter 132) to obesity (Chapter 207). Health, disease, and an individual’s interaction with the environment are also substantially determined by genetics. In addition to many conditions that may be determined by a single gene, medical science increasingly understands

the complex interactions that underlie multigenic traits (Chapter 36). The decoding of the human genome holds the promise that personalized health care increasingly can be targeted according to an individual’s genetic profile, in terms of screening and presymptomatic disease management, as well as in terms of specific medications, their complex interactions, and their adjusted dosing schedules.2 Knowledge of the structure and physical forms of proteins helps explain abnormalities as diverse as sickle cell anemia (Chapter 154) and prion-related diseases (Chapter 387). Proteomics, which is the study of normal and abnormal protein expression of genes, also holds extraordinary promise for developing drug targets for more specific and effective therapies. Gene therapy is currently approved by the U.S. Food and Drug Administration (FDA) for only a few diseases—Leber congenital amaurosis (Chapter 395), retinal dystrophy, and hemophilia (Chapter 165)—but many more are in development and clinical testing. Cell therapy is now beginning to provide vehicles for the delivery of cells engineered to address a patient’s particular chimeric antigen receptor (CAR),3 and CAR-T cell therapy is now FDAapproved for non-Hodgkin lymphoma (Chapter 176) and acute lymphoblastic leukemia (Chapter 173). Regenerative medicine to help heal injured or diseased organs and tissues is in its infancy, but cultured chondrocytes are now FDAapproved to repair cartilaginous defects of the femoral condyle and the knee. Immune checkpoint inhibitors have revolutionized the approach to cancer, especially melanoma (Chapter 193).4 In the future, immunotherapy will likely find applications not only for malignancies but also for the treatment of refractory infectious diseases, autoimmunity, and allergy.5 Concurrent with these advances in fundamental human biology has been a dramatic shift in methods for evaluating the application of scientific advances to the individual patient and to populations. The randomized controlled trial, sometimes with thousands of patients at multiple institutions, has replaced anecdote as the preferred method for measuring the benefits and optimal uses of diagnostic and therapeutic interventions (Chapter 8). And now, even the well-established randomized controlled trial model is being challenged. To reduce costs as well as overcome inefficiencies, redundancies, and the late failure of many clinical trials (at the phase 3 stage) inherent in classical randomized controlled trials, technologic advances are enabling new methods, tools, and approaches to bring clinical trials into the 21st century. These methods include: disease modeling and simulation; alternative trial methods such as response-adaptive randomized designs (Chapter 8); novel objective outcome measures, and engagement of clinical trial “participants” (rather than “human subjects”) to expand the pool of patients willing to be involved in clinical research. As studies progress from those that show biologic effect, to those that elucidate dosing schedules and toxicity, and finally to those that assess true clinical benefit, the metrics of measuring outcome has also improved from subjective impressions of physicians or patients to reliable and valid measures of morbidity, quality of life, functional status, and other patient-oriented outcomes (Chapter 9). These marked improvements in the scientific methodology of clinical investigation have expedited extraordinary changes in clinical practice, such as recanalization therapy for acute myocardial infarction (Chapter 64), and have shown that reliance on intermediate outcomes, such as a reduction in asymptomatic ventricular arrhythmias with certain drugs, may unexpectedly increase rather than decrease mortality. Just as physicians in the 21st century must understand advances in fundamental biology, similar understanding of the fundamentals of clinical study design as it applies to diagnostic and therapeutic interventions is important. Studies can be designed to show benefit or to show noninferiority, and newer pragmatic designs (Chapter 8) help with the study of topics that would be challenging using traditional approaches. An understanding of human genetics can also help stratify and refine the approach to clinical trials by helping researchers select fewer patients with a more homogeneous disease pattern to study the efficacy of an intervention. Such an approach has been especially relevant in cancer, where tumors with certain genetic mutations can respond to a drug specifically designed for that target, whereas other tumors with similar microscopic but different genomic characteristics will not.6 Genomic, transcriptomic, epigenomic, proteomic, metabolomic, and other “omic” technologies provide a more holistic view of the molecular makeup of a normal or abnormal organism, tissue, or cell. Systems biology, which is the integration of all these techniques, can enable the development of new predictive, preventive, and personalized approaches to disease. Sophisticated computerized analyses of radiographs and retinal images7 are also poised to revolutionize the interpretation of these images much as computerized electrocardiographic interpretation (Chapter 48) changed clinical cardiology. Electronic medical records also can detect patterns of drug side

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

ABSTRACT

The medical profession incorporates both the science of medicine as well as the art of being a physician. Physicians cannot help patients unless they are well-grounded in the latest information about medical diagnosis and therapy, which increasingly is based on randomized clinical trials as well as specific information about the genetics and genomics of individual patients. However, this scientific expertise must also be applied in the context of understanding the patient as an individual person. In applying both scientific knowledge and medical professionalism, the physician must also recognize the importance of social justice as well as how to advocate for and help each individual patient in the context of broader societal issues.

KEYWORDS

medical professionalism scientific medicine evidence-based medicine approach to the patient signs and symptoms

2.e1

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

effects or interactions that can then guide molecular analyses that confirm new risks or even genetic diseases.8 Although it is too soon to know whether patients would routinely benefit from sequencing and analysis of their exome or full genome, such information is increasingly becoming affordable and more accurate, with potential usefulness for identifying mendelian disease patterns9 and informing reproductive planning to avoid autosomal recessive diseases. Despite much hope, however, genetic profiling has had very limited positive impact on drug selection and dosing. This explosion in medical knowledge has led to increasing specialization and subspecialization, defined initially by organ system and more recently by locus of principal activity (inpatient vs. outpatient), reliance on manual skills (proceduralist vs. nonproceduralist), or participation in research. Nevertheless, it is becoming increasingly clear that the same fundamental molecular and genetic mechanisms are broadly applicable across all organ systems and that the scientific methodologies of randomized trials and careful clinical observation span all aspects of medicine. The advent of modern approaches to managing data now provides the rationale for the use of health information technology. Computerized health records, oftentimes shared with patients in a portable format, can avoid duplication of tests, assure that care is coordinated among the patient’s various health care providers, and increase the value of health care.10 Real-time electronic records can also be used to alert physicians about patients whose vital signs (Chapter 7) might warrant urgent evaluation to avoid more serious clinical decompensation. However, a current downside is that for every hour physicians provide direct clinical face time to patients in the office, nearly two additional hours may be spent on electronic health records and desk work within the work day.11 Extraordinary advances in the science and practice of medicine, which have continued to accelerate with each recent edition of this textbook, have transformed the global burden of disease. Life expectancies for men and women have been increasing, a greater proportion of deaths are occurring among people older than age 70 years, and far fewer children are dying before the age of 5 years. In the United States, however, overall life expectancy has surprisingly declined in the last several years. Explanations include obesity-related diseases12 as well as so-called deaths of despair owing to alcohol, drugs, and suicide.13 Nevertheless, huge regional disparities remain, and disability from conditions such as substance abuse, mental health disorders, injuries, diabetes, musculoskeletal disease, and chronic respiratory disease have become increasingly important issues for all health systems.

  APPROACH TO THE PATIENT

Patients commonly have complaints (symptoms), but at least one third of these symptoms will not be readily explainable by any detectable abnormalities on examination (signs) or on laboratory testing. Even in our modern era of advanced diagnostic testing, the history and physical examination are estimated to contribute at least 75% of the information that informs the evaluation of symptoms, and symptoms that are not explained on initial comprehensive evaluation rarely are manifestations of a serious underlying disease. Conversely, asymptomatic patients may have signs or laboratory abnormalities, and laboratory abnormalities can occur in the absence of symptoms or signs. Symptoms and signs commonly define syndromes, which may be the common final pathway of a wide range of pathophysiologic alterations. The fundamental basis of internal medicine is that diagnosis should elucidate the pathophysiologic explanation for symptoms and signs so that therapy may improve the underlying abnormality, not just attempt to suppress the abnormal symptoms or signs. When patients seek care from physicians, they may have manifestations or exacerbations of known conditions, or they may have symptoms and signs that suggest malfunction of a particular organ system. Sometimes the pattern of symptoms and signs is highly suggestive or even pathognomonic for a particular disease process. In these situations, in which the physician is focusing on a particular disease, Goldman-Cecil Medicine provides scholarly yet practical approaches to the epidemiology, pathobiology, clinical manifestations, diagnosis, treatment, prevention, and prognosis of entities such as acute myocardial infarction (Chapter 64), chronic obstructive lung disease (Chapter 82), inflammatory bowel disease (Chapter 132), gallstones (Chapter 146), rheumatoid arthritis (Chapter 248), hypothyroidism (Chapter 213), and tuberculosis (Chapter 308), as well as newly described disorders such as emerging zoonoses, small fiber neuropathies, nephrogenic systemic fibrosis, mitochondrial diseases, autoinflammatory diseases, and clonal disorders of indeterminate potential.

3

Many patients, however, have undiagnosed symptoms, signs, or laboratory abnormalities that cannot be immediately ascribed to a particular disease or cause. Whether the initial manifestation is chest pain (Chapter 45), diarrhea (Chapter 131), neck or back pain (Chapter 372), or a variety of more than 100 common symptoms, signs, or laboratory abnormalities, Goldman-Cecil Medicine provides tables, figures, and entire chapters to guide the approach to diagnosis and therapy (see E-Table 1-1 or table on inside back cover). By virtue of this dual approach to known disease as well as to undiagnosed abnormalities, this textbook, similar to the modern practice of medicine, applies directly to patients regardless of their mode of manifestation or degree of previous evaluation. The patient-physician interaction proceeds through many phases of clinical reasoning and decision making. The interaction begins with an elucidation of complaints or concerns, followed by inquiries or evaluations to address these concerns in increasingly precise ways. The process commonly requires a careful history or physical examination, ordering of diagnostic tests, integration of clinical findings with test results, understanding of the risks and benefits of the possible courses of action, and careful consultation with the patient and family to develop future plans. Physicians can increasingly call on a growing literature of evidence-based medicine to guide the process so that benefit is maximized while respecting individual variations in different patients. Throughout Goldman-Cecil Medicine, the best current evidence is highlighted with specific grade A references that can be accessed directly in the electronic version. The increasing availability of evidence from randomized trials to guide the approach to diagnosis and therapy should not be equated with “cookbook” medicine.14 Evidence and the guidelines that are derived from it emphasize proven approaches for patients with specific characteristics. Substantial clinical judgment is required to determine whether the evidence and guidelines apply to individual patients and to recognize the occasional exceptions. Even more judgment is required in the many situations in which evidence is absent or inconclusive. Evidence must also be tempered by patients’ preferences, although it is a physician’s responsibility to emphasize evidence when presenting alternative options to the patient. The adherence of a patient to a specific regimen is likely to be enhanced if the patient also understands the rationale and evidence behind the recommended option. To care for a patient as an individual, the physician must understand the patient as a person. This fundamental precept of doctoring includes an understanding of the patient’s social situation, family issues, financial concerns, and preferences for different types of care and outcomes, ranging from maximum prolongation of life to the relief of pain and suffering (Chapters 2 and 3). If the physician does not appreciate and address these issues, the science of medicine cannot be applied appropriately, and even the most knowledgeable physician will fail to achieve the desired outcomes. Even as physicians become increasingly aware of new discoveries, patients can obtain their own information from a variety of sources, some of which are of questionable reliability. The increasing use of alternative and complementary therapies (Chapter 34) is an example of patients’ frequent dissatisfaction with prescribed medical therapy. Physicians should keep an open mind regarding unproven options but must advise their patients carefully if such options may carry any degree of potential risk, including the risk that they may be relied on to substitute for proven approaches. It is crucial for the physician to have an open dialogue with the patient and family regarding the full range of options that either may consider. Another manifestation of problematic interactions and care is medical malpractice litigation, which commonly is a result of both suboptimal medical care and suboptimal communication (Chapter 10). Of note is that about 1% of all physicians account for 32% of paid malpractice claims nationally,15 thereby suggesting that individual physician characteristics are important and addressable contributors. The physician does not exist in a vacuum, but rather as part of a complicated and extensive system of medical care and public health. In premodern times and even today in some developing countries, basic hygiene, clean water, and adequate nutrition have been the most important ways to promote health and reduce disease. In developed countries, adoption of healthy lifestyles, including better diet (Chapter 202) and appropriate exercise (Chapter 13), is the cornerstone to reducing the epidemics of obesity (Chapter 207), coronary disease (Chapter 46), and diabetes (Chapter 216). Public health interventions to provide immunizations (Chapter 15) and to reduce injuries and the use of tobacco (Chapter 29), illicit drugs (Chapter 31), and excess alcohol (Chapter 30) can collectively produce more health benefits than nearly any other imaginable health intervention.

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

E-TABLE 1-1 GUIDE TO THE APPROACH TO COMMON SYMPTOMS, SIGNS, AND LABORATORY ABNORMALITIES CHAPTER

SPECIFIC TABLES OR FIGURES

SYMPTOMS Constitutional Fever

264, 265

Figures 265-1, 265-2; Tables 264-1 to 264-8

Fatigue

258

E-Table 258-1

Poor appetite

123

Table 123-1

Weight loss

123, 206

Figure 123-4; Tables 123-4, 206-1, 206-2

Obesity

207

Figure 207-1

Snoring, sleep disturbances

377

Table 377-6

Headache

370

Tables 370-1, 370-2

Visual loss, transient

395, 396

Tables 395-2, 396-1

Ear pain

398

Table 398-3

Hearing loss

400

Figure 400-1

Ringing in ears (tinnitus)

400

Figure 400-2

Vertigo

400

Figure 400-3

Nasal congestion, rhinitis, or sneezing

398

Loss of smell or taste

399

Table 399-1

Dry mouth

397

Table 397-7

Sore throat

401

Figure 401-2; Table 401-1

Hoarseness

401

Head, Eyes, Ears, Nose, Throat

Cardiopulmonary Chest pain

45, 128

Tables 45-2, 128-5, 128-6

Bronchitis

90

Shortness of breath

45, 77

Figure 77-3

Palpitations

45, 56

Figure 56-1; Tables 45-4, 56-5

Dizziness

45, 56, 400

Figure 56-1; Table 400-1

Syncope

56

Figure 56-1; Tables 56-1, 56-2, 56-4

Cardiac arrest

57

Figures 57-2, 57-3

Cough

77

Figure 77-1; Tables 77-2, 77-3

Hemoptysis

77

Tables 77-6, 77-7

Nausea and vomiting

123

Figure 123-5; Table 123-5

Dysphagia, odynophagia

123, 129

Table 123-1

Hematemesis

126, 144

Figure 126-3; Table 126-1

Heartburn/dyspepsia

123, 128 to 130

Figures 123-6, 129-2; Tables 128-3, 128-4, 130-1

Abdominal pain  Acute  Chronic

123, 133 123, 128

Figures 123-1, 123-2; Tables 123-2, 123-3, 133-1 Figure 123-3; Tables 123-2, 129-1

Diarrhea

128, 131

Figures 128-1, 131-1 to 131-4

Melena, blood in stool

126

Figures 126-3, 126-4, 126-6; Table 126-4

Constipation

127, 128

Figures 127-3, 128-1; Table 127-2

Fecal incontinence

136

Figure 136-5

Anal pain

136

Gastrointestinal

Genitourinary Dysuria

268, 269

Tables 268-3, 268-5, 269-2

Frequency

268

Table 268-3

Incontinence

23

Tables 23-1 to 23-3

Renal colic

117

Figure 117-1

Vaginal discharge

269

Menstrual irregularities

223

Figure 223-3; Tables 223-3, 223-4

Female infertility

223, 227

Table 223-5

Hot flushes

227

Table 227-1

Erectile dysfunction

221

Figure 221-10

Male infertility

221

Figures 221-8, 221-9; Table 221-7

Scrotal mass

190

Figure 190-1

Genital ulcers or warts

269

Table 269-1

3.e1

3.e2

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

E-TABLE 1-1 GUIDE TO THE APPROACH TO COMMON SYMPTOMS, SIGNS, AND LABORATORY ABNORMALITIES—cont’d CHAPTER

SPECIFIC TABLES OR FIGURES

Musculoskeletal Neck or back pain

372

Figures 372-4, 372-5, 372-6; Tables 372-3 to 372-5

Painful joints

241

Figure 241-1; Tables 241-1, 241-3

Swollen feet, ankles, or legs  Bilateral  Unilateral

45 74

Figure 45-8 Figure 74-2; Table 74-2

Claudication

71

Table 71-3

Acute limb ischemia

71

Figure 71-4; Table 71-2

Weakness

368, 392 to 394

Tables 368-1, 392-2, 393-2, 393-4

Sensory loss

368, 392

Figure 392-1; Tables 392-1, 392-3 to 392-5

Memory loss

374

Figures 374-1, 374-2; Tables 374-1 to 374-6

Abnormal gait

368

Table 368-2

Seizures

375

Tables 375-1 to 375-6

Abnormal bleeding

162

Table 162-1

Rash

407, 412

Figure 407-1; Tables 407-1 to 407-6, 412-5

Hives

237, 411

Figure 237-2; Tables 237-1, 411-1, 411-2

Abnormal pigmentation

412

Table 412-2

Alopecia and hirsutism

413

Tables 413-1, 413-3

Nail disorders

413

Table 413-4

Fever

264, 265

Figure 265-1; Tables 264-1 to 264-8, 265-2

Heat illness/hyperthermia

101

Tables 101-1 to 101-3

Hypothermia

7, 101

Tables 101-4 to 101-6

Tachycardia/bradycardia

7, 56, 58, 59

Figures 56-2, 56-3; Tables 58-4, 59-2

Hypertension

70

Tables 70-3, 70-7 to 70-11

Hypotension/shock

7, 98

Figures 98-3, 100-1; Tables 98-1, 99-1, 99-2

Altered respiration

7, 80, 96

Tables 80-1, 80-2, 96-2

Eye pain

395

Table 395-3

Red eye

395

Tables 395-4, 395-6

Dilated pupil

396

Figure 396-4

Nystagmus

396

Table 396-4

Papilledema

396

Table 396-2

Strabismus

396

Figure 396-6

Jaundice

138

Figure 138-2; Tables 138-1 to 138-3

Rhinitis

398

Table 398-3

Sinusitis

398

Tables 398-1, 398-2, 398-4, 398-5

Oral ulcers and discolorations

397

Tables 397-1 to 397-4

Salivary gland enlargement

397

Table 397-6

Neck mass

181

Figure 181-3

Lymphadenopathy

159

Tables 159-1 to 159-4

Thyroid nodule

213

Figure 213-5

Thyromegaly/goiter

213

Figures 213-2, 213-3

Extremities

Neurologic

Integumentary

SIGNS Vital Signs

Head, Eyes, Ears, Nose, Throat

Neck

Breast Breast mass

188

Lungs Wheezes

77

Table 77-4

Heart murmur or extra sounds

45

Figure 45-5; Tables 45-7, 45-8

Jugular venous distention

45

Table 45-6

Carotid pulse abnormalities

45

Figure 45-4

Cardiac

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

3.e3

E-TABLE 1-1 GUIDE TO THE APPROACH TO COMMON SYMPTOMS, SIGNS, AND LABORATORY ABNORMALITIES—cont’d CHAPTER

SPECIFIC TABLES OR FIGURES

Abdomen Hepatomegaly

137

Figure 137-5

Splenomegaly

159

Table 159-5

Acute abdomen

133, 134

Figure 134-1; Table 133-1

Abdominal swelling/ascites

133, 144

Table 144-3

Rectal bleeding/positive stool

126, 184

Figures 126-3, 126-4, 126-6; Table 126-4

Hemorrhoids

136

Table 136-1

Arthritis

241

Figure 241-1

Edema

45

Figure 45-7

Cyanosis

45

Clubbing

45

Musculoskeletal/Extremities

Neurologic Delirium

25

Figure 25-1; Tables 25-1, 25-2

Psychiatric disturbances

369

Tables 369-1 to 369-4, 369-6 to 369-8, 369-10, 369-11, 369-13, 369-14

Coma

376

Tables 376-1 to 376-4

Stroke

379, 380

Figure 379-1; Tables 379-2, 379-3, 379-5, 379-6, 380-5, 380-6

Movement disorders

381, 382

Tables 381-4, 382-1 to 382-8

Neuropathy

392

Tables 392-1 to 392-4, 392-6

Suspicious mole

193

Table 193-1

Nail diseases

413

Table 413-4

Anemia

149

Tables 149-2 to 149-6

Polycythemia

157

Table 157-4

Leukocytosis

158

Figure 158-4; Table 158-1

Lymphocytosis

158

Table 158-3

Monocytosis

158

Table 158-2

Eosinophilia

161

Figure 161-1; Table 161-1

Neutropenia   With fever

158 265

Figure 158-7; Tables 158-4 and 158-5 Figure 265-1

Thrombocytosis

157

Table 157-5

Thrombocytopenia

163

Figure 163-1; Tables 163-1, 163-3

Prolonged PT or PTT

162

Figure 162-4

Urinalysis

106, 112

Tables 106-2, 112-6

Abnormal liver enzymes

138

Figures 138-1 to 138-3

Elevated BUN/creatinine  Acute  Chronic

112 121

Figure 112-1; Tables 112-1 to 112-5 Table 121-1

Hyperglycemia

216

Tables 216-1, 216-2

Hypoglycemia

217

Tables 217-1, 217-2

Electrolyte abnormalities

108, 109

Figures 108-3, 108-4; Tables 108-7, 109-3

Acid-base disturbances

110

Figures 110-1 to 110-3; Tables 110-1 to 110-7

Hypercalcemia

232

Figure 232-3; Tables 232-2 to 232-4

Hypocalcemia

232

Figure 232-4; Table 232-6

Hypo- and hyperphosphatemia

111

Tables 111-2, 111-3

Magnesium deficiency

111

Table 111-1

Elevated Pco2

80

Figure 80-2

Solitary pulmonary nodule

182

Figure 182-2

Pleural effusion

92

Tables 92-3 to 92-5

ECG abnormalities

48

Tables 48-2 to 48-5

Skin and Nails

COMMON LABORATORY ABNORMALITIES Hematology/Urinalysis

Chemistries

Chest Radiograph/ECG

BUN = blood urea nitrogen; ECG = electrocardiogram; PT = prothrombin time; PTT = partial thromboplastin time.

TABLE 1-1 PROFESSIONAL RESPONSIBILITIES Commitment to: Professional competence Honesty with patients Patient confidentiality Maintaining appropriate relations with patients Improving the quality of care Improving access to care Just distribution of finite resources Scientific knowledge Maintaining trust by managing conflicts of interest Professional responsibilities From Brennan T, Blank L, Cohen J, et al. Medical professionalism in the new millennium: a physician charter. Ann Intern Med. 2002;1136:243-246.

  APPROACH TO THE MEDICAL PROFESSION

Medical professionalism should emphasize three fundamental principles: the primacy of patient welfare, patient autonomy, and social justice.16 As modern medicine brings a plethora of diagnostic and therapeutic options, the interactions of the physician with the patient and society become more complex and potentially fraught with ethical dilemmas (Chapter 2). To help provide a moral compass that is not only grounded in tradition but also adaptable to modern times, the primacy of patient welfare emphasizes the fundamental principle of a profession. The physician’s altruism, which begets the patient’s trust, must be impervious to the economic, bureaucratic, and political challenges that are faced by the physician and the patient (Chapter 4). The principle of patient autonomy asserts that physicians make recommendations but patients make the final decisions. The physician is an expert advisor who must inform and empower the patient to base decisions on scientific data and how these data can and should be integrated with a patient’s preferences. The importance of social justice symbolizes that the patient-physician interaction does not exist in a vacuum. The physician has a responsibility to the individual patient and to broader society to promote access, to eliminate disparities in health and health care, and to bring science to even the most contentious political issues. For example, research into the relationship of firearms to rates of murder and suicide17 can be useful for preventive medicine and public policy regardless of an individual’s position on background checks and licensing for gun owners. To promote these fundamental principles, a series of professional responsibilities (Table 1-1) represents practical, daily traits that benefit the physician’s own patients and society as a whole. Physicians who use these and other attributes to improve their patients’ satisfaction with care are not only promoting professionalism but also reducing their own risk for liability and malpractice. By comparison, the recent emphasis on maintenance of certification requirements is of uncertain benefit for improving patient outcomes. An interesting new aspect of professionalism is the increasing reliance on team approaches to medical care, as exemplified by physicians whose roles are defined by the location of their practice—historically in the intensive care unit or emergency department and more recently on the inpatient general hospital floor. Quality care requires coordination and effective communication across inpatient and outpatient sites among physicians who themselves now typically work defined hours. This transition from reliance on a single, always available physician to a team, ideally with a designated coordinator, places new challenges on physicians, the medical care system, and the medical profession. An ongoing challenge for a profession that values dedication, attention to detail, and selflessness is the risk of burnout, which is characterized by emotional exhaustion and depersonalization. Both individual-focused and structural or organizational modifications in the work environment can result in clinically meaningful reductions in physician burnout.18 The changing medical care environment is placing increasing emphasis on standards, outcomes, and accountability. As purchasers of insurance become more cognizant of value rather than just cost (Chapter 10), outcomes ranging from rates of screening mammography (Chapter 188) to mortality rates with coronary artery bypass graft surgery (Chapter 65) become metrics by which rational choices can be made. Clinical guidelines and critical pathways derived from randomized controlled trials and evidence-based medicine can potentially lead to more cost-effective care and better outcomes. These major changes in many Western health care systems bring with them many major risks and concerns. If the concept of limited choice among

physicians and health care providers is based on objective measures of quality and outcome, channeling of patients to better providers is one reasonable definition of better selection and enlightened competition. If the limiting of options is based overwhelmingly on cost rather than measures of quality, outcomes, and patient satisfaction, physicians and their patients can be seriously disadvantaged. Another risk is that the same genetic information that could lead to more effective, personalized medicine will be used against the very people whom it is supposed to benefit—by creating a stigma, raising health insurance costs, or even making someone uninsurable. The ethical approach to medicine (Chapter 2), genetics (Chapter 35), and genetic counseling provides means to protect against this adverse effect of scientific progress. In this new environment, the physician often has a dual responsibility: to the health care system as an expert who helps create standards, measures of outcome, clinical guidelines, and mechanisms to ensure high-quality, costeffective care; and to individual patients who entrust their well-being to that physician to promote their best interests within the reasonable limits of the system. A health insurance system that emphasizes cost-effective care, that gives physicians and health care providers responsibility for the health of a population and the resources required to achieve these goals, that must exist in a competitive environment in which patients can choose alternatives if they are not satisfied with their care, and that places increasing emphasis on health education and prevention can have many positive effects. In this environment, however, physicians must beware of overt and subtle pressures that could entice them to underserve patients and abrogate their professional responsibilities by putting personal financial reward ahead of their patients’ welfare. The physician’s responsibility to represent the patient’s best interests and avoid financial conflicts by doing too little in the newer systems of capitated care provides different specific challenges but an analogous moral dilemma to the historical American system in which the physician could be rewarded financially for doing too much. In the current health care environment, all physicians and trainees must redouble their commitment to professionalism. At the same time, the challenge to the individual physician to retain and expand the scientific knowledge base and process the vast array of new information is daunting. In this spirit of a profession based on science and caring, Goldman-Cecil Medicine seeks to be a comprehensive approach to modern internal medicine. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 1  APPROACH TO MEDICINE, THE PATIENT, AND THE MEDICAL PROFESSION  

GENERAL REFERENCES 1. Goldman L. Three stages of health encounters over 8000 human generations and how they inform future public health. Am J Public Health. 2018;108:60-62. 2. Vassy JL, Christensen KD, Schonman EF, et al. The impact of whole-genome sequencing on the primary care and outcomes of healthy adult patients: a pilot randomized trial. Ann Intern Med. 2017;167:159-169. 3. June CH, O’Connor RS, Kawalekar OU, et al. CAR T cell immunotherapy for human cancer. Science. 2018;359:1361-1365. 4. Ribas A, Wolchok JD. Cancer immunotherapy using checkpoint blockade. Science. 2018;359: 1350-1355. 5. Wraith DC. The future of immunotherapy: a 20-year perspective. Front Immunol. 2017;8:1-6. 6. Kalemkerian GP, Narula N, Kennedy EB, et al. Molecular testing guideline for the selection of patients with lung cancer for treatment with targeted tyrosine kinase inhibitors: American society of clinical oncology endorsement of the college of American Pathologists/international association for the study of lung Cancer/association for molecular pathology clinical practice guideline update. J Clin Oncol. 2018;36:911-919. 7. Nam JG, Park S, Hwang EJ, et al. Development and validation of deep learning–based automatic detection algorithm for malignant pulmonary nodules on chest radiographs. Radiology. 2019;290: 218-228. 8. Wolford BN, Willer CJ, Surakka I. Electronic health records: the next wave of complex disease genetics. Hum Mol Genet. 2018;27:R14-R21.

4.e1

9. Bastarache L, Hughey JJ, Hebbring S, et al. Phenotype risk scores identify patients with unrecognized mendelian disease patterns. Science. 2018;359:1233-1239. 10. Smith CD, Levinson WS. A commitment to high-value care education from the internal medicine community. Ann Intern Med. 2015;162:639-640. 11. Sinsky C, Colligan L, Li L, et al. Allocation of physician time in ambulatory practice: a time and motion study in 4 specialties. Ann Intern Med. 2016;165:753-760. 12. Hales CM, Fryar CD, Carroll MD, et al. Trends in obesity and severe obesity prevalence in US youth and adults by sex and age, 2007-2008 to 2015-2016. JAMA. 2018;319:1723-1725. 13. Case A, Deaton A. Mortality and morbidity in the 21st century. Brookings Pap Econ Act. 2017;2017:397-476. 14. Yeh RW, Kramer DB. Decision tools to improve personalized care in cardiovascular disease: moving the art of medicine toward science. Circulation. 2017;135:1097-1100. 15. Studdert DM, Bismark MM, Mello MM, et al. Prevalence and characteristics of physicians prone to malpractice claims. N Engl J Med. 2016;374:354-362. 16. Egener BE, Mason DJ, McDonald WJ, et al. The charter on professionalism for health care organizations. Acad Med. 2017;92:1091-1099. 17. Kaufman EJ, Morrison CN, Branas CC, et al. State firearm laws and interstate firearm deaths from homicide and suicide in the United States: a cross-sectional analysis of data by county. JAMA Intern Med. 2018;178:692-700. 18. West CP, Dyrbye LN, Shanafelt TD. Physician burnout: contributors, consequences and solutions. J Intern Med. 2018;283:516-529.

4

CHAPTER 2  Bioethics in the Practice of Medicine  

2  BIOETHICS IN THE PRACTICE OF MEDICINE EZEKIEL J. EMANUEL It commonly is argued that the bioethical dilemmas physicians face today result primarily from modern advances in medical technology. The rise of antibiotics, transplantation, intensive care units, genetics, implantable devices, and other technologies have created novel bioethical concerns. In reality, however, concerns about ethical issues are as old as the practice of medicine itself. The Hippocratic Oath, composed sometime around 400 bc, attests to the need even of ancient Greek physicians for advice on how to address the many bioethical dilemmas that they confronted. The Oath addresses issues of confidentiality, abortion, euthanasia, sexual relations between physicians and patients, divided loyalties, and, at least implicitly, charity care and executions. Whether we agree with the advice it dispensed or not, the mere existence of the Oath serves as a reminder that bioethical conundrums are inherent to medical practice. Technology may make these issues more common and change the context in which they arise, but many, if not most, bioethical issues that physicians regularly confront are timeless. During their training, many physicians are taught that four main principles can be invoked to address bioethical dilemmas: autonomy, nonmaleficence,

4.e2

ABSTRACT

CHAPTER 2  Bioethics in the Practice of Medicine  

It commonly is argued that the bioethical dilemmas physicians face today result primarily from modern advances in medical technology. In reality, however, concerns about ethical issues are as old as the practice of medicine itself. The Hippocratic Oath, composed sometime around 400 bc, attests to the need even of ancient Greek physicians for advice on how to address the many bioethical dilemmas that they confronted. Technology may make these issues more common and change the context in which they arise, but many, if not most, bioethical issues that physicians regularly confront are timeless. A multitude of bioethical dilemmas arise in medical practice each year, including issues of genetics, conscientious objection by providers, and termination of care. In clinical practice, the most common issues revolve around informed consent, termination of life-sustaining treatments, euthanasia and physicianassisted suicide, and conflicts of interest.

KEYWORDS

bioethics informed consent life-sustaining treatment euthanasia physician-assisted suicide conflict of interest

CHAPTER 2  Bioethics in the Practice of Medicine  

beneficence, and justice. Autonomy is the idea that people should have the right and freedom to choose, pursue, and revise their own life plans. Nonmaleficence is the idea that people should not be knowingly harmed or injured. This principle is encapsulated in the oft-repeated phrase that a physician must “first do no harm”—primum non nocere. Interestingly, this phrase is not found in the Hippocratic Oath; the only related, but still not identical, Hippocratic phrase is “at least, do not harm.” Beneficence refers to the positive actions that a physician should undertake to promote the well-being of his or her patients. In clinical practice, this obligation usually arises from the implicit and explicit commitments and promises central to the physician-patient relationship. Finally, the principle of justice is defined by the fair distribution of benefits and burdens that result from a clinical interaction. Although helpful in providing an initial framework, these principles are too broad to have more than limited value. The principles are also frequently underdeveloped and likely to conflict with each other, thereby resulting in bioethical dilemmas. The principles themselves do not offer guidance on how they should be balanced or specified to resolve dilemmas. Given that they are focused on physician-patient encounters, the principles are also unhelpful when considering bioethical issues at the system or institutional level, such as the allocation of scarce vaccines or transplant organs. Finally, these four principles are not comprehensive. Other fundamental ethical principles and values—such as priority to the worst off, duties to future generations, and professional integrity—are important in bioethics but not fully encapsulated by these four principles. There is no formula that can magically determine how to solve bioethical dilemmas. Instead, medical professionals should follow an orderly analytic process. First, practitioners need to obtain the facts relevant to the situation. Second, they must delineate the fundamental bioethical issue. Third, they must identify all the crucial principles and values that relate to, and potentially conflict with, the case. Fourth, because many ethical dilemmas have been previously analyzed and subjected to empirical study, practitioners should examine the relevant literature so they may potentially identify new values, understand existing principles, reformulate the issue at hand, and see if there is an accepted resolution. Fifth, with this information, the practitioner must distinguish clearly unethical practices from a range of ethically permissible actions. Finally, it is important not only to come to a resolution but also to state clearly the justification for such decisions. Although unanimous decisions are ideal, the reality remains that such consensus may be elusive. Reasonable physicians must therefore take care to explain what principles and interpretations they relied upon to resolve ethical dilemmas. A multitude of bioethical dilemmas arise in medical practice each year, including issues of genetics, conscientious objection by providers, and termination of care. In clinical practice, the most common issues revolve around informed consent, termination of life-sustaining treatments, euthanasia and physician-assisted suicide, and conflicts of interest.

  PHYSICIAN-PATIENT RELATIONSHIP:

INFORMED CONSENT



History

The requirement of informed consent dates as far back as Plato. The first recorded legal case on informed consent took place in England in 1767 when a patient complained that he had not given his consent for two surgeons to refracture his leg after it had healed improperly. An 18th-century English court ultimately ruled that obtaining a patient’s consent prior to a procedure was the “rule of the profession” and thus a legal obligation of surgeons. Failure to obtain consent, the court declared, was inexcusable. In more contemporary times, a landmark 1957 U.S. court ruling stated that physicians have a positive legal obligation to disclose information about risks, benefits, and alternative treatments to patients; this decision popularized the term informed consent.  

Definition and Justification

Informed consent is a person’s autonomous authorization to permit a physician—or other health care professional—to undertake diagnostic or therapeutic interventions for himself or herself. The patient understands that he or she is taking responsibility for the decision while empowering someone else, the physician, to implement it. However, agreement to a course of medical treatment does not necessarily qualify as informed consent. The four fundamental requirements for valid informed consent are: mental capacity, disclosure, understanding, and voluntariness. First, informed consent assumes that people have the mental capacity to make decisions; disease, development, or medications can compromise patients’ mental capacity to provide informed consent. Adults are presumed to have the legal competence

5

to make medical decisions, and whether an adult is incompetent to make medical decisions is a legal determination. Practically, physicians usually decide whether patients are competent based on whether they can understand the information disclosed, appreciate its significance for their own situation, and use logical and consistent thought processes in decision making. Incompetence in medical decision making does not mean a person is incompetent in all types of decision making and vice versa. Second, crucial information relevant to the decision must be disclosed, usually by the physician, to the patient. Third, the patient should understand the information and its implications for his or her interests and life goals. Finally, the patient’s decision must be made voluntarily, i.e., without coercion or manipulation by the physician. It is a mistake to view informed consent as a one-time event, such as the signing of a form. Informed consent is viewed more accurately as a process that evolves throughout the course of a patient’s diagnosis and subsequent treatment. Typically, a patient’s autonomy is the value invoked to justify informed consent. Other values, such as bodily integrity and beneficence, have also been cited, especially in early legal rulings.  

Empirical Data

Extensive research on informed consent shows that physicians frequently do not communicate all the relevant information needed for patients to make an informed decision in clinical settings. The more complex the medical decisions, the more likely it is that physicians will obtain all the elements of informed consent. Interestingly, data suggest that disclosure, both in informed consent documents and in discussion, is better in research than clinical settings. Greater disclosure in the research setting may be the result of the research-specific requirement of having a written informed consent document reviewed by an independent committee, such as an institutional review board or a research ethics committee. Patients frequently fail to recall crucial information disclosed during the process of obtaining informed consent, although they usually think they have sufficient information to make a decision. Whether patients fail to recall key information because they are overwhelmed by the information or because they find much of it not salient to their decision is unclear. The issue therefore lies more in determining what patients understand at the point of decision making, rather than what they recall later. For common medical interventions, such as elective surgery, the ideal informed consent would include the risks and benefits as quantified in randomized controlled trials, relevant data on the surgeon, the institution’s clinical outcomes for the procedure, and a list of acceptable alternatives.1 Studies aimed at improving informed consent in clinical settings suggest that interactive media, such as videos and interactive computer software, can improve patients’ understanding of such competing alternatives. A1  A review of 115 studies on shared decision making found that, compared with those receiving usual care, patients who used a decision aid had greater knowledge of the evidence, felt more clear about what mattered to them, had more accurate understanding of risks and benefits, and participated more in the decisionmaking process. These decision aids can be especially important in preferencesensitive conditions—situations with several treatment options and important tradeoffs that are dependent on a patient’s values, such as decisions about prostatectomy for early stage prostate cancer. Computer and web-based decision aides are now available for over 200 common conditions and procedures. A more modern challenge in obtaining informed consent is the introduction of electronic methods into the process, including smartphone applications in acute situations such as ischemic stroke. A2  Digital models for informed consent often are not user-friendly for older patients, and evidence suggests that most people do not read click-through agreements on computers and mobile devices. Concern also exists that it may be challenging to obtain true voluntary choice without being able to assess the body language, tone, and emotion that can be observed during a person-to-person interaction.2 One of the most important results of empirical research on informed consent is that there is a gap between a desire for information and a desire for actual decision making. Many studies show that although most patients want information, far fewer actually want to make decisions about their own care. In one study, for example, only one third of patients desired decision-making authority, and patients’ decision-making preferences were not correlated with their information-seeking preferences (which were high). Patients’ preferences for decision-making authority generally increase with higher educational levels and decline with age. Most importantly, the more serious the illness, the more likely patients are to prefer that physicians make the decisions. Several studies

6

CHAPTER 2  Bioethics in the Practice of Medicine  

TABLE 2-1 FUNDAMENTAL ELEMENTS FOR DISCLOSURE TO PATIENTS Diagnosis and prognosis Nature of proposed intervention Reasonable alternative interventions Risks associated with each alternative intervention Benefits associated with each alternative intervention Probable outcomes of each alternative intervention

suggest that patients who have less of a desire to make their own decisions generally are more satisfied with how the decisions are ultimately made.  

Practical Considerations

Implementing informed consent raises questions about the extent to which information should be disclosed and how to disclose it. Physicians should disclose at least six fundamental elements of information to patients: (1) diagnosis and prognosis; (2) nature of the proposed intervention; (3) alternative interventions, including no treatment; (4) risks associated with each alternative; (5) benefits of each alternative; and (6) likely outcomes of these alternatives (Table 2-1). Because risk is usually a physician’s principal concern, physicians also should disclose (1) the nature of the risks, (2) their magnitude, (3) the probability that each risk will occur, and (4) when the consequence might occur. Increasingly, these disclosures should include data both from clinical trials and from the institution and physician performing the test and treatments. In general, all serious risks, such as death, paralysis, stroke, infections, or chronic pain, even if rare, should be disclosed, as should common risks. The key challenge in providing this information is doing so within reasonable time constraints and without overwhelming the patient with unnecessarily complex or technical details. Fortunately, time constraints can be somewhat ameliorated by using interactive electronic media that allow patients to view information on their own schedules while facilitating the transfer of basic information. The question of how much physicians should disclose has been approached differently state-by-state. Generally, states have adopted one of two divergent legal standards. The physician or customary standard, adapted from malpractice law, states that the physician should disclose all information “which a reasonable medical practitioner would make under the same or similar circumstances.” Conversely, the reasonable person or lay-oriented standard states that physicians should disclose all information that a “reasonable person in the patient’s circumstances would find material” to the medical decision. The physician standard is factual and can be determined empirically, but the patient-oriented standard, which is meant to engage physicians with patients, is hypothetical. Currently, each standard is used by about half of the United States. In 2015, the United Kingdom’s Supreme Court ruled that the standard for what information a physician should disclose should not be determined by what a reasonable physician would do, but rather by what a reasonable patient would want. The requirement of informed consent has no exceptions. In emergency situations, consent can be assumed under the belief that patients’ interests are in survival and retaining maximal mental and physical functioning; as a result, reasonable persons would want treatment. In some circumstances, physicians may believe the process of informed consent could pose a serious psychological threat. In rare cases, the “therapeutic privilege” of promoting a patient’s well-being trumps autonomy, but physicians should be wary of invoking this exception too readily. If patients are deemed mentally incompetent to make medical decisions, family members—beginning with spouse, children, parents, siblings, then more distant relatives—usually are selected as surrogates or proxies, although there may be concerns about conflicting interests or knowledge of the patient’s wishes. In the relatively rare circumstance in which a patient has formally designated a proxy, that person has decision-making authority. The substituted judgment standard states that the proxy should choose what the patient would choose if he or she were competent. The best interests standard states that the proxy should choose what is best for the patient. However, it is often not clear what the patient would have decided, because the situation was not discussed with the patient and he or she left no living will. Similarly, what is considered “best” for a patient can be controversial because of tradeoffs between quality of life and pure survival. These problems are complicated by the poor ability of many proxies to predict a patient’s quality of life; proxies

also tend to underestimate patients’ future functional status and satisfaction. Similarly, a proxy’s predictions on a mentally incapacitated patient’s lifesustaining preferences are often inaccurate. In cases in which the patient is diagnosed with dementia, families tend to agree with patients on decisions regarding life-sustaining treatment two thirds of the time, better but not much better than the 50% agreement based on chance alone. Such confusion on how to decide for incapacitated patients can create conflicts among family members or between the family and medical providers. In such circumstances, an ethics consultation may be helpful.



  TERMINATION OF MEDICAL INTERVENTIONS History

Since the origins of medicine, withholding medical treatment from terminally ill patients while still providing palliative care, thereby allowing “nature to take its course,” has been deemed ethical.3 Hippocrates argued that physicians should “refuse to treat those [patients] who are overmastered by their disease.” In the 19th century, prominent American physicians advocated withholding cathartic and emetic “treatments” from the terminally ill. In 1900, the editors of The Lancet argued that physicians should intervene to ease the pain of death and that they did not have an obligation to prolong a clearly terminal life. The contemporary debate on terminating care began in 1976 with the Quinlan case, in which the New Jersey Supreme Court ruled that patients had a right to refuse life-sustaining interventions on the basis of a right to privacy, and that the family could exercise that right for a patient in a persistent vegetative state.  

Definition and Justification

It generally is agreed that all patients have a right to refuse medical interventions. Ethically, this right is based on the patient’s right to autonomy and is implied by the doctrine of informed consent. Legally, state courts have cited the right to privacy, the right to bodily integrity, and common law to justify the right to refuse medical treatment. In the 1990 Cruzan case, and in the subsequent physician-assisted suicide cases, the U.S. Supreme Court affirmed that there is a “constitutionally protected right to refuse lifesaving hydration and nutrition.” The Court stated that “[A] liberty interest [based on the 14th Amendment] in refusing unwanted medical treatment may be inferred from our prior decisions.” All patients have both a constitutional and an ethical right to refuse medical interventions. These rulings were the basis of consistent state and federal court rulings in the Schiavo case to permit the husband to terminate artificial nutrition and hydration for his terminally ill wife in a vegetative state (Chapter 376).  

Empirical Data

Data show that termination of medical treatments is now the norm, and the trend has shifted toward stopping medical interventions based on the preferences of patients and their surrogate decision makers. Over 85% of Americans and 90% of decedents in intensive care units do not receive cardiopulmonary resuscitation. Of decedents in intensive care units, more than 85% die after the withholding or withdrawal of medical treatments, with an average of 2.6 interventions being withheld or withdrawn per decedent. Despite extensive public support for use of advance care directives and the passage of the Patient Self-Determination Act mandating that health care institutions inform patients of their right to complete such documents, less than 40% of Americans appear to have completed any written form of endof-life decisions. Among Americans ages 75 and older, 1 in 5 have neither written nor talked with someone about their wishes for medical treatment at the end of their lives. Data suggest that although over 40% of patients required active decision making about terminating medical treatments in their final days, more than 70% lacked actual decision-making capacity, thereby emphasizing the importance of completing advance directives. Efforts to improve completion of advance care directives have generated mixed results. In La Crosse County, Wisconsin, for example, after health care organizations in the county added an “Advance Directive” section to their electronic medical records, 90% of decedents had some type of advance directive. Unfortunately, even successful pilot efforts like La Crosse County’s have not been adopted or easily scaled. A persistent problem has been that even when patients complete advance care directives, the documents frequently are not readily available, physicians do not know they exist, or they tend to be too general or vague to guide decisions. The increasing use of electronic health records should make it possible for advance directives to be available whenever and wherever the patient presents to a health care provider. Although electronic health records will help in making existing advance directives

CHAPTER 2  Bioethics in the Practice of Medicine  

available, they will not solve the problem of actually having a conversation between the physician and the patient about advance care planning. Starting that conversation still seems to be a persistent barrier. Just as proxies are poor at predicting patients’ wishes, data show that physicians are even worse at determining patients’ preferences for life-sustaining treatments. In one study, for example, 30% of family conferences between clinicians and surrogates did not discuss preferences for end-of-life decision making for patients who were at high risk of death.4 In many cases, life-sustaining treatments are continued even when patients or their proxies desire them to be stopped. Conversely, many physicians discontinue or never begin interventions unilaterally without the knowledge or consent of patients or their surrogate decision makers. These discrepancies emphasize the importance of engaging patients early on in their care about treatment preferences.  

Practical Considerations

Many practical considerations are applicable to enacting the right to terminate medical treatment (Table 2-2). First, patients have a right to refuse any and all medical interventions. The question of what medical interventions can be terminated—or not started—is a recurrent topic of debate among physicians and other health care providers. Initiation of cardiopulmonary resuscitation was the focus of early court cases. Courts have made clear that any treatment prescribed by a physician and administered by a health care provider can be stopped if it is more harmful than beneficial.5 The issue is not whether the treatment is ordinary, extraordinary, or heroic, or whether it is high-technology or lowtechnology. Treatments that can be stopped include not only ventilators, artificial nutrition, and hydration, but also dialysis, pacemakers, ventricular assist devices, antibiotics, and any medication. Second, there is no ethical or legal difference between withholding an intervention and withdrawing it. If a respirator or other treatment is started because physicians are uncertain whether a patient would have wanted it, they always can stop it later when information clarifies the patient’s wishes. Although

TABLE 2-2 PRACTICAL CONSIDERATIONS IN TERMINATION OF MEDICAL TREATMENTS PRACTICAL QUESTION

ANSWER

Is there a legal right to refuse medical interventions?

Yes. The U.S. Supreme Court declared that competent people have a constitutionally protected right to refuse unwanted medical treatments based on the 14th Amendment.

What interventions can be legally and ethically terminated?

Any and all interventions (including respirators, antibiotics, pacemakers, ventricular assist devices, intravenous or enteral nutrition and hydration) can be legally and ethically terminated.

Is there a difference between withholding life-sustaining interventions and withdrawing them?

No. The consensus is that there is no important legal or ethical difference between withholding and withdrawing medical interventions. Stopping a treatment once begun is just as ethical as never having started it.

Whose view about terminating The views of a competent adult patient prevail. life-sustaining interventions It is the patient’s body and life. prevails if there is a conflict between the patient and family? Who decides about terminating life-sustaining interventions if the patient is incompetent?

If the patient appointed a proxy or surrogate decision maker when competent, that person is legally empowered to make decisions about terminating care. If no proxy was appointed, there is a legally designated hierarchy, usually (1) spouse, (2) adult children, (3) parents, (4) siblings, and (5) available relatives.

Are advance care directives legally enforceable?

Yes. As a clear expression of the patient’s wishes, they are a constitutionally protected method for patients to exercise their right to refuse medical treatments. In almost all states, clear and explicit oral statements are legally and ethically sufficient for decisions about withholding or withdrawing medical interventions.

7

physicians and nurses might find stopping a treatment to be more difficult psychologically, withdrawal is ethically and legally permitted—and required— when it is consonant with the patient’s wishes. Third, competent patients have the exclusive right to make decisions about terminating their own care. If there is a conflict between a competent patient and his or her family, the patient’s wishes are to be followed. It is the patient’s right to refuse treatment, not the family’s right. For mentally incompetent patients, the situation is more complex. If the patients left clear indications of their wishes, whether as explicit oral statements or as written advance care directives, these wishes should be followed. Physicians should not be overly concerned about the precise form patients use to express their wishes; because patients have a constitutional right to refuse treatment, the real concern is whether the wishes are clear and relevant to the situation. If an incompetent patient did not leave explicit indications of his or her wishes or designate a proxy decision maker, the physician should identify a surrogate decision maker and rely on the decision maker’s wishes. Some state courts have restricted what treatments a proxy decision-maker can terminate, thereby requiring the incompetent patient to have given very specific instructions about the particular treatments he or she does not want to receive and the conditions under which care should be withheld or withdrawn. This requirement severely limits the authority and power of proxy decision makers. Fourth, the right to refuse medical treatment does not translate into a right to demand any treatment, especially treatments that have no pathophysiologic rationale, have already failed, or are known to be harmful. Futility has become a justification to permit physicians unilaterally to withhold or withdraw treatments despite the family’s requests for treatment. Some states, such as Texas, have enacted futility laws, which prescribe procedures by which physicians can invoke futility either to transfer a patient or to terminate interventions. However, the principle of futility is not easy to implement in medical practice. Initially, some commentators advocated that an intervention was futile when the probability of success was 1% or lower. Although this threshold seems to be based on empirical data, it is a covert value judgment. Because the declaration of futility is meant to justify unilateral determinations by physicians, it generally has been viewed as an inappropriate assertion that undermines physician-patient communication and violates the principle of shared decision making. Similar to the distinction between ordinary and extraordinary care, futility is increasingly viewed as more obfuscating than clarifying and is therefore being invoked much less often. For example, a recent California case involved a 13-year-old girl who suffered a cardiac arrest during a tonsillectomy and adenoidectomy and who was subsequently declared brain dead. Her family refused to accept the determination of death and sued. After several legal appeals, courts agreed that she was dead. Her body was nevertheless given on a respirator to the county coroner, who then transferred the body to the parents. The parents kept the body and authorized that a tracheostomy and a feeding tube be inserted. The court ruled that neither physicians nor a medical facility had any obligation to provide treatments to a dead body, even if the parents asserted, contrary to medical experts, that the patient was not dead.



  ASSISTED SUICIDE AND EUTHANASIA History

As far back as the time of Hippocrates, euthanasia and physician-assisted suicide were controversial issues. In 1905, a bill was introduced into the Ohio legislature to legalize euthanasia; it was defeated. In the mid-1930s, similar bills were introduced and defeated in the British Parliament and the Nebraska legislature. As of January 2017, physician-assisted suicide—but not euthanasia—has been made legal in Oregon, Washington, California, Colorado, Vermont, and Washington, D.C. In Montana, the Supreme Court did not recognize a constitutional right to physician-assisted suicide, but it ruled that the law permitting the termination of life-sustaining treatment protected physicians from prosecution if they helped hasten the death of a consenting, rational, terminally ill patient. Of note, however, is that the American College of Physicians does not currently support the legalization of physician-assisted suicide.6 Both euthanasia and physician-assisted suicide are legal in the Netherlands, Belgium, and Luxembourg, and physician-assisted suicide is legal in Switzerland.  

Definition and Justification

The terms euthanasia and physician-assisted suicide require careful definition (Table 2-3). So-called passive and indirect euthanasia are misnomers and not actual instances of euthanasia; rather, they are ethical and legal ways to terminate care.

8

CHAPTER 2  Bioethics in the Practice of Medicine  

TABLE 2-3 DEFINITIONS OF ASSISTED SUICIDE AND EUTHANASIA TERM

DEFINITION

Voluntary active euthanasia

Intentional administration of medications or other interventions to cause the patient’s death with the patient’s informed consent

Involuntary active euthanasia

Intentional administration of medications or other interventions to cause the patient’s death when the patient was competent to consent but did not consent (e.g., the patient may not have been asked)

Nonvoluntary active Intentional administration of medications or other euthanasia interventions to cause the patient’s death when the patient was incompetent and was mentally incapable of consenting (e.g., the patient might have been in a coma) Passive euthanasia

Withholding or withdrawal of life-sustaining medical treatments from a patient to let him or her die (termination of life-sustaining treatments)—a poor term that should not be used

Indirect euthanasia

Administration of narcotics or other medications to relieve pain with the incidental consequence of causing sufficient respiratory depression to result in the patient’s death

Physician-assisted suicide

A physician provides prescription medications or other interventions to a patient with the understanding that the patient can use them to commit suicide

There are four arguments against permitting euthanasia and physician-assisted suicide. First, Kant and Mill, the philosophical champions of individual autonomy, believed that autonomy itself did not allow a person voluntarily to end conditions that made them autonomous. As a result, both philosophers were against voluntary enslavement and suicide. They therefore argued that the exercise of autonomy cannot include the ending of life, which would mean ending the possibility of exercising autonomy. Second, many dying patients may experience pain and suffering as the result of not receiving appropriate care. It is therefore possible that adequate care and pain management (Chapter 27) might relieve suffering without the need for euthanasia or physician-assisted suicide (Chapter 3). Although some patients may experience uncontrolled pain despite optimal end-of-life care, relatively few give pain as the justification for seeking euthanasia or physician-assisted suicide. Third, there is a clear ethical distinction between intentionally ending a life and terminating lifesustaining treatments. Both the motivations and physical acts are different. Injecting a life-ending medication, or providing a prescription for one, is not the same as removing or refraining from introducing an invasive medical intervention. Finally, permitting euthanasia and physician-assisted suicide may introduce adverse consequences. There are disturbing reports of involuntary euthanasia in the Netherlands and Belgium, and many worry about coercion of expensive or burdensome patients to accept euthanasia or physician-assisted suicide. Permitting euthanasia and physician-assisted suicide is likely to lead to further intrusions of lawyers, courts, and legislatures into the physicianpatient relationship. There are four parallel arguments for permitting euthanasia and physicianassisted suicide. First, it is argued that autonomy justifies euthanasia and physician-assisted suicide. To respect autonomy requires permitting individuals to decide when and how it is better to end their lives. Second, beneficence— furthering the well-being of individuals—supports permitting euthanasia and physician-assisted suicide. In some cases, living can create more pain and suffering than death; ending a painful life relieves more suffering and produces a net good for the patient. Just the reassurance of having the option of euthanasia or physician-assisted suicide, even if not used, can provide “psychological insurance” and be beneficial to people. Third, euthanasia and physician-assisted suicide are no different from termination of life-sustaining treatments that are recognized as ethically justified. In both cases, the patient consents to die; in both cases, the physician intends to end the patient’s life and takes some action to end the patient’s life; and in both cases, the final result is the same: the patient’s death. With no difference in the patient’s consent, the physician’s intention, or the final result, there can be no difference in the ethical justification. Fourth, the supposed slippery slope that would result from permitting euthanasia and physician-assisted suicide is unlikely. The idea that permitting

euthanasia and physician-assisted suicide would undermine the physicianpatient relationship or lead to forced euthanasia is completely speculative and not borne out by the available data. In its 1997 decisions, the U.S. Supreme Court stated that there is no constitutional right to euthanasia and physician-assisted suicide, but that there also is no constitutional prohibition against states legalizing these interventions. Consequently, five states and the District of Columbia (see above) have constitutionally legalized physician-assisted suicide, and others may do so either by legislation or ballot measure.  

Empirical Data

Attitudes and practices related to euthanasia and physician-assisted suicide have been studied extensively.7 Two thirds of Americans say there are some situations in which a patient should be allowed to die, but 30% say that medical professionals should always do everything possible to save a patient’s life. About 60% of adults think that these interventions are moral for a person who has an incurable disease and is suffering great pain with no hope of improvement. However, public support dramatically decreases to below 40% for patients who are ready to die because living is a burden or for patients who are burdensome to their families. Overall, public support for physicianassisted suicide in the United States remains just below 50%. Physicians tend to be much less supportive of euthanasia and physicianassisted suicide than the public, with oncologists, palliative care physicians, and geriatricians among the least supportive. Among American and British physicians, the majority opposes legalizing either practice. Approximately 25% of American physicians have received requests for euthanasia or physician-assisted suicide, including about 50% of oncologists. Studies also indicate that less than 5% of American physicians have performed euthanasia or physician-assisted suicide. Surveys of oncologists indicate that about 4% have performed euthanasia and about 11% have performed physicianassisted suicide during their careers. Safeguards for euthanasia and physician-assisted suicide are frequently violated. For example, one study found 54% of euthanasia requests came from the family. In about 40% of euthanasia and 20% of physician-assisted suicide cases, the patient was depressed; in only half of the cases was the request repeated, irrespective of treatment. Oregon has legally permitted physician-assisted suicide for the longest of any U.S. jurisdiction. Data show that over 70% of patients receiving physicianassisted suicide had cancer. Other characteristics strongly linked to requesting physician-assisted suicide included age over 65 years, white race, more formal education, and having medical insurance. Importantly, use of physician-assisted suicide is rare. Over 20 years, less than 0.4% of all dying patients died by physician-assisted suicide.8 In the Netherlands and Belgium, where both euthanasia and physician-assisted suicide are legal, less than 2% of all deaths are by these measures, with 0.4 to 1.8% of all deaths as the result of euthanasia without the patient’s consent.9 Counterintuitively, in all jurisdictions where it has been studied, pain is not the primary motivation for requesting euthanasia or physician-assisted suicide. In Oregon, loss of autonomy, dignity, and fear of being a burden are cited by patients as the predominant motives. In addition, psychological distress, especially depression and hopelessness, seem to be more important than pain. Interviews with physicians and with patients with amyotrophic lateral sclerosis, cancer, or infection with human immunodeficiency virus show that pain is not associated with interest in euthanasia or physician-assisted suicide; instead, depression and hopelessness are the strongest predictors of interest. These findings raise important issues about the involvement of mental health experts in attempts to determine whether psychiatric treatment would change a patient’s views.10 Finally, data from the Netherlands and the United States suggest that there are significant problems in performing euthanasia and physician-assisted suicide. Dutch researchers reported that physician-assisted suicide causes complications in 7% of cases. Furthermore, the patients did not die, awoke from coma, or vomited up the medication in 15% of cases. Ultimately, in nearly 20% of physician-assisted suicide cases, the physician ended up injecting the patient with life-ending medication, converting physician-assisted suicide to euthanasia. These data raise serious questions about how to address complications of physician-assisted suicide when euthanasia is illegal or unacceptable.  

Practical Considerations

There is widespread agreement that if euthanasia and physician-assisted suicide are used, they should be considered only after all reasonable attempts at physical

CHAPTER 2  Bioethics in the Practice of Medicine  

and psychological palliation have failed. A consensus—with slight differences— among American states and European countries has emerged on safeguards. These safeguards include: (1) the patient must be competent and must request euthanasia or physician-assisted suicide repeatedly and voluntarily; (2) in the Netherlands and other European countries, the patient must have unbearable pain or other suffering that cannot be relieved by optimal palliative interventions; by comparison, there is no requirement for suffering in the United States, but the patient must be terminally ill; (3) there should be a waiting period to ensure that the patient’s desire for euthanasia or physician-assisted suicide is stable and sincere; and (4) the physician should obtain a second opinion from an independent physician. Although there have been some prosecutions in the United States, there have been no convictions—except for Dr. Kevorkian—when physicians and others have participated in euthanasia and physician-assisted suicide.



  FINANCIAL CONFLICTS OF INTEREST History

Worrying about how payment structures and fees compromise the integrity of medical decision making is not new. In 1899, a physician reported that more than 60% of surgeons in Chicago were willing to provide a 50% commission to physicians for referring cases. He subsequently argued that in some cases, this fee splitting led to unnecessary surgical procedures. A 1912 study by the American Medical Association confirmed that fee splitting was a common practice and it added to the list of physicians’ financial conflicts of interest acts, which included selling patented medicines and patenting surgical instruments. In the 1990s, the ethics of pharmaceutical and biotech companies paying clinical researchers and physicians again raised the issue of financial conflicts of interest.  

Definition and Justification

A conflict of interest occurs when a physician’s secondary interests, such as making money, risks compromising or undermining a physician’s primary interest, especially promoting a patient’s well-being. Physicians also have other primary interests: (1) to advance biomedical research, (2) to educate future physicians, and, more controversially, (3) to promote public health (Table 2-4). Physicians also have other, secondary interests, such as earning income, raising a family, contributing to the profession, and pursuing avocational interests, such as hobbies. These secondary interests are not evil; typically, they are legitimate, even admirable. A conflict of interest occurs when one of these secondary interests could compromise pursuit of a primary interest, especially the patient’s well-being. Conflicts of interest are problematic because they can, or at least appear to, compromise the integrity of physicians’ judgment, the patient’s well-being, or research. Conflicts of interest can induce a physician to do something— perform a procedure, fail to order a test, or distort data—that may not be in a patient’s best interest. These conflicts can undermine the trust of both the patient and the public, not only in an individual physician but also in the medical profession at large. Sometimes a distinction may be claimed between actual and potential conflicts of interest, suggesting that a conflict exists only when a physician’s judgment is actually distorted or undermined. This concept is wrong. An actual conflict of interest occurs when a reasonable person could suspect that the physician’s judgment could have been altered by the secondary interest. Appearances can be damaging, because it is difficult for patients and the public to determine what motives influence a physician’s decision and it often is impossible to know whether judgment actually has been distorted. Financial conflicts of interest are of particular concern, not because they are worse than other types of conflicts, but rather because they are more pervasive, identifiable, and regulated compared with other conflicts. Since ancient times, the ethical norm on conflicts has been clear: the physician’s primary obligation is to the patient’s well-being, and a physician’s personal financial well-being comes second and should never compromise this duty.

TABLE 2-4 PRIMARY INTERESTS OF PHYSICIANS Promotion of the health and well-being of their patients Advancement of biomedical knowledge through research Education of future physicians and health care providers Promotion of public health



9

Empirical Data

Financial conflicts are not rare, but they are frequently under-reported. The more imaging facilities and specialty referrals a practice has, the greater the utilization of medical services and the higher the health care spending—often without any clear benefit to the patients. In Florida, nearly 40% of physicians are owners of freestanding facilities to which they refer patients. In one study, 4 to 4.5 times more imaging examinations were ordered by self-referring physicians than by physicians who referred patients to radiologists. Similarly, patients referred to joint-venture physical therapy facilities have an average of 16 visits compared with 11 at non–joint-venture facilities. A recent study of urologists found that those who had integrated radiation facilities into their practices increased their radiation use by 2.5 times compared with urologists who did not have financial relationships with radiation facilities. Similarly, multiple studies have shown that interaction with pharmaceutical representatives can lead to prescribing of new drugs, nonrational prescribing, and decreased use of generic drugs by physicians. Industry funding for continuing medical education payment for travel to educational symposia increases prescribing of the sponsor’s drug. A study of 1,400 FDA advisory committee members found that 13% had some financial interest in a drug company whose product was being reviewed by that committee; these members had a 63% chance of voting for its approval, and an 84% chance of doing so if they sat on advisory boards for that company. A separate study found that 80% of U.S.-based hematologist-oncologists who use Twitter, often to tweet about pharmaceutical products, have at least one financial conflict of interest, with median payments of over $1000.11 Regarding researcher conflicts of interest, the available data suggest that corporate funding does not appear to compromise the design and methodology of clinical research; in fact, commercially funded research may be methodologically more rigorous than government- or foundation-supported research. Conversely, data suggest that financial interests do distort researchers’ interpretation of data. The most important impact of financial interests, however, appears to be on dissemination of research studies. Growing evidence suggests the suppression or selective publication of data unfavorable to corporate sponsors but the repeated publication of favorable results.  

Practical Considerations

First, financial conflicts of interest are inherent in any profession in which the professional earns income from rendering a service. Second, conflicts come in many different forms, from legitimate payment for services rendered and ownership of medical laboratories and facilities, to drug company dinners, payment for attendance at pharmaceutical meetings, and consultation with companies. Third, in considering how to manage conflicts, it is important to note that people are poor judges of their own potential conflicts. Individuals often cannot distinguish the various influences that guide their judgments, do not think of themselves as inherently bad, and do not imagine that payment shapes their judgments. Physicians tend to be defensive about charges of conflicts of interest. In addition, conflicts tend to act insidiously, subtly changing practice patterns so that they then become what appear to be justifiable norms. Fourth, rules—whether laws, regulations, or professional standards—to regulate conflicts of interest are based on two considerations: (1) the likelihood that payment or other secondary interests would create a conflict, with greater financial interest tending to increase the risk of compromised judgment, and (2) the magnitude of the potential harm if judgment is compromised. Rules tend to be of three types: (1) disclosure of conflicts, (2) management of conflicts, and (3) outright prohibition. Federal law bans certain types of self-referral of physicians in the Medicare program. The American Medical Association and the Pharmaceutical Research and Manufacturers of America have established joint rules that permit physicians to accept gifts of minimal value but “refuse substantial gifts from drug companies, such as the costs of travel, lodging, or other personal expenses…for attending conferences or meetings.” Additionally, the Physician Payment Sunshine Act, which was passed in 2010 as part of the Affordable Care Act and went into effect in August 2013, requires that drug and device manufacturers report all payments and transfers of value given to physicians to the Centers for Medicare and Medicaid Services so that such information can be published on a searchable public website. Fifth, there is much emphasis on disclosure of conflicts, with the implicit idea being that sunshine is the best disinfectant. Disclosure may be useful in publications for peers, but it is unclear whether this is a suitable safeguard in

the clinical setting. Disclosure may instead make patients worry more. Patients may have no context in which to place the disclosure or to evaluate the physician’s clinical recommendation, and they may have few other options in selecting a physician or getting care, especially in an acute situation. Furthermore, selfdisclosure often is incomplete, even when required. Finally, some conflicts can be avoided by a physician’s own action. Physicians can refuse to engage in personal investments in medical facilities or to accept gifts from pharmaceutical companies at relatively little personal cost. In other circumstances, the conflicts may be institutionalized, and minimizing them can occur only by changing the way organizations structure reimbursement incentives. Capitation encourages physicians to limit medical services, and its potentially adverse effects are likely to be managed by institutional rules rather than by personal decisions.

  FUTURE DIRECTIONS

In the near future, as genetics moves from the research to the clinical setting, practicing physicians are increasingly likely to encounter ethical issues surrounding genetic testing, counseling, and treatment. The use of genetic tests without the extensive counseling so common in research studies would alter the nature of the bioethical issues. Because these tests have serious implications for the patient and others, scrupulous attention to informed consent must occur. The bioethical issues raised by genetic tests for somatic cell changes, such as tests that occur commonly in cancer diagnosis and risk stratification, are no different from the issues raised with the use of any laboratory or radiographic test. In some cases, ethics consultation services may be of assistance in resolving bioethical dilemmas, although current data suggest that consultation services are used mainly for problems that arise in individual cases and are not used for more institutional or policy problems.

  Grade A References A1. Stacey D, Légaré F, Lewis K, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev. 2017;4:CD001431. A2. Haussen DC, Doppelheuer S, Schindler K, et al. Utilization of a smartphone platform for electronic informed consent in acute stroke trials. Stroke. 2017;48:3156-3160.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 2  Bioethics in the Practice of Medicine  

GENERAL REFERENCES 1. Fernandez Lynch H, Joffe S, Feldman EA. Informed consent and the role of the treating physician. N Engl J Med. 2018;378:2433-2438. 2. Grady C, Cummings SR, Rowbotham MC, et al. Informed consent. N Engl J Med. 2017;376: 856-867. 3. CaringInfo, a program of the National Hospice and Palliative Care Organization. http://www.caringinfo.org/. Accessed May 10, 2019. 4. Scheunemann LP, Cunningham TV, Arnold RM, et al. How clinicians discuss critically ill patients’ preferences and values with surrogates: an empirical analysis. Crit Care Med. 2015;43:757-764. 5. Robinson EM, Cadge W, Zollfrank AA, et al. After the DNR: surrogates who persist in requesting cardiopulmonary resuscitation. Hastings Cent Rep. 2017;47:10-19.

10.e1

6. Snyder Sulmasy L, Mueller PS. Ethics and the legalization of physician-assisted suicide: an American college of physicians position paper. Ann Intern Med. 2017;167:576-578. 7. Emanuel EJ, Onwuteaka-Philipsen BD, Urwin JW, et al. Attitudes and practices of euthanasia and physician-assisted suicide in the United States, Canada, and Europe. JAMA. 2016;316:79-90. 8. Hedberg K. New C. Oregon’s death with dignity act: 20 years of experience to inform the debate. Ann Intern Med. 2017;167:579-583. 9. Preston R. Death on demand? An analysis of physician-administered euthanasia in The Netherlands. Br Med Bull. 2018;125:145-155. 10. Sheehan K, Gaind KS, Downar J. Medical assistance in dying: special issues for patients with mental illness. Curr Opin Psychiatry. 2017;30:26-30. 11. Tao DL, Boothby A, McLouth J, et al. Financial conflicts of interest among hematologist-oncologists on twitter. JAMA Intern Med. 2017;177:425-427.

10.e2

CHAPTER 2  Bioethics in the Practice of Medicine  

REVIEW QUESTIONS Medical content: End-of-life care 1. A 26-year-old woman collapses in her apartment because of a cardiac arrest. Her husband calls 9-1-1 after discovering her unconscious. Paramedics take her to the hospital, where she is intubated, ventilated, and ultimately diagnosed as being in a persistent vegetative state. The husband, appointed by the court as his wife’s legal guardian, moves to petition to remove the feeding tube. The woman’s parents oppose the movement. The woman has no living will. Who has the legal right to make end-of-life decisions in this case? A . The husband, as he is the legal guardian. B. The parents, as they are next of kin. C. The state, as the woman lacked a living will. D. The physicians, as they are the ones who would actually terminate care. E. The hospital’s ethics committee. Answer: A  This patient was legally incompetent and therefore unable to make end-of-life care decisions. Health care surrogates are selected according to the following priority: guardian, spouse, adult son or daughter, parents, adult siblings, adult relative, close friend, guardian of the estate. Her husband was therefore lawfully appointed to be her guardian and proxy decision maker, as he has higher priority than the woman’s parents. Despite the parents’ objections, the husband has full legal authority to make the final decisions.

Medical content: Conflict of interest 2. A patient with a headache goes to see his primary care physician. The patient describes the pain as a dull throbbing, not particularly painful, and as having appeared 12 hours earlier after a poor night’s sleep. The physician sits on an advisory board of a pharmaceutical company that makes pain relievers. Which of the following represents the MOST ethical course of action by the physician? A . Offer to enroll the patient in the study on a new pain relief medication for migraines. B. Order magnetic resonance imaging. C. Prescribe the patient pain relievers from the company on whose board the physician sits. D. Refer the patient to another provider, as the physician has too many conflicts of interest to be involved in this case. E. Recommend the patient take an over-the-counter pain reliever, go home, and call back if pain persists or worsens. Answer: E  The patient’s symptoms are neither severe nor life-threatening. Especially given the patient’s poor sleep prior to the headache’s onset, it is reasonable for the physician to only recommend over-the-counter pain relievers and follow-up should the pain persist. Enrolling the patient in the migraine study presents a conflict of interest because the physician would be compensated for such action, and it does not appear the patient actually suffers from migraines. Ordering a magnetic resonance imaging scan would be an example of unnecessary care simply to increase the physician’s compensation. Prescribing medications from the company on which the physician sits on the board is also a conflict of interest, because such prescription pills are not needed. If the physician is cognizant of his potential conflicts of interest, he can still provide high-quality care without having to refer the patient to another provider.

10

CHAPTER 3  Palliative Care  

3  PALLIATIVE CARE ROBERT M. ARNOLD

By 2030, 20% of the U.S. population will be older than 65 years, and people older than 85 years constitute the fastest growing segment of the population. Owing to successes in public health and medicine, many of these people will live the last years of their lives with chronic medical conditions such as cirrhosis, end-stage kidney disease, heart failure, and dementia. Even human immunodeficiency virus (HIV) and many cancers, once considered terminal, have turned into chronic diseases. The burden associated with these illnesses and their treatments is high. Chronically ill patients report multiple physical and psychological symptoms that lower their quality of life. The economic pressures associated with medical care adversely affect patients’ socioeconomic status and cause family stress, especially among caregivers, who spend 20 or more hours a week helping their loved ones. Palliative care, which was developed to decrease the burden associated with chronic illness, emphasizes patient- and family-centered care that optimizes quality of life by anticipating, preventing, and treating suffering. Palliative care throughout the continuum of illness addresses physical, intellectual, emotional,

social, and spiritual needs while facilitating the patient’s autonomy, access to information, and choice. Palliative care is both a subspeciality and a key component of good medicine. Specialty palliative care, delivered by an interdisciplinary team, is available concurrently with or independent of curative or life-prolonging care. Palliative and nonpalliative health care providers should collaborate and communicate about care needs while focusing on peace and dignity throughout the course of illness, during the dying process, and after death. Given that most seriously ill patients are not seen by a palliative care subspecialist, every clinician should have basic competency in palliative care. For example, all primary care physicians should know the basic tenets of treating pain (Chapter 27) as well as how to discuss advance directives (Chapter 2) and give “bad” news. Specialties with a high prevalence of seriously ill patients, such as critical care and oncology, should have more advanced skills.1 Interventions to promote both specialty and primary palliative care are associated with improvements in the patient’s burden of symptoms and quality of life, although its effects on the caregiver are less consistent. A1-A3  Five points deserve special emphasis. First, palliative care can be delivered at any time during the course of an illness and is often provided concomitantly with disease-focused, life-prolonging therapy. Waiting until a patient is dying to provide palliative care is a serious error. For example, most elderly patients with chronic incurable illnesses, who might benefit from palliative care, are in the last 10 years of their lives but do not consider themselves to be dying. If palliative care is to have an impact on patients’ lives, it should be provided earlier in a patient’s illness, in tandem with other treatments. Second, prediction is an inexact science. For most illnesses, including cancer, physicians have trouble accurately predicting whether a patient is in the last 6 months of life (E-Fig. 3-1). Third, palliative care primarily focuses on the illness’s burden rather than treating the illness itself. Because these burdens can be physical, psychological, spiritual, or social, good palliative care requires a multidisciplinary approach. Fourth, palliative care takes the family unit as the central focus of care. Treatment plans must be developed for both the patient and the family. Fifth, palliative care recognizes that medical treatments are not uniformly successful and that patients die. At some point in a patient’s illness, the treatments may cause more burden than benefit. Palliative care recognizes this reality and starts with a discussion of the patient’s goals and the development of an individualized treatment plan. Many people confuse palliative care with hospice—an understandable confusion because hospices epitomize the palliative care philosophy. The two, however, are different. In the United States, hospice provides palliative care, primarily at home, for patients who have a life expectancy of 6 months or less and who are willing to forgo life-prolonging treatments. However, the requirement that patients must have a life expectancy of 6 months or less limits hospice’s availability, as does the requirement that patients give up expensive and potentially life-prolonging treatments. Moreover, because doctors often are unwilling to cease these treatments until very late in the disease course, so are most patients.

  PALLIATIVE CARE DOMAINS

Palliative care is a philosophy of care with physical, psychological, spiritual, existential, social, and ethical domains. When caring for patients with chronic life-limiting illness, good palliative care requires that the following questions be addressed:  

Is the Patient Physically Comfortable?

Across many chronic conditions, patients have a large number of inadequately treated physical symptoms (Table 3-1). The reasons are multifactorial and range from inadequate physician education, to societal beliefs regarding the inevitability of suffering in chronic illness, to public concerns regarding opioids, to the lack of evidence-based treatments in noncancer patients. The first step to improve symptom management is a thorough assessment.2 Standardized instruments such as the Brief Pain Inventory (Fig. 3-1) measure both the patient’s symptoms and the effect of those symptoms on the patient’s life. Use of standardized instruments (such as the Edmonton Symptom Assessment Scale3 [E-Fig. 3-2]) assures that physicians will identify overlooked or underreported symptoms and, as a result, will enhance the satisfaction of both the patient and family. The evidence for the treatment of end-stage symptoms continues to improve. For example, palliative care can improve quality of life in patients with endstage heart failure, A4  who often require such help.4 The use of nonsteroidal anti-inflammatory agents and opioids can result in effective pain management

CHAPTER 3  Palliative Care  

ABSTRACT

Palliative care, which was developed to decrease the burden associated with chronic illness, emphasizes patient- and family-centered care that optimizes quality of life by anticipating, preventing, and treating suffering. It is both a subspecialty and a key component of good medical care for seriously ill patients. Palliative care should be delivered throughout the course of a patient’s illness— from diagnosis through death. The specialty involves a multidisciplinary group of clinicians who work with the patients’ primary clinicians to provide an extra layer of support. When caring for patients with serious illness, good palliative care requires that the following questions be addressed: (1) is the patient’s care consistent with his or her goals; (2) is the patient physically comfortable or psychologically suffering; and (3) is the family suffering.

KEYWORDS

palliative care serious illness end-of-life supportive care advance directives living wills quality of life

10.e3

10.e4

CHAPTER 3  Palliative Care  

High Function

Mostly cancer

Death Time

Low

Short period of evident decline

High

Chronic, consistent with usual role

Mostly heart and lung failure

Death

Healthy

Time

Low

Long-term limitations with intermittent serious episodes

High

Mostly frailty and dementia

Function

Chronic, progressive, eventually fatal illness

Time

Low

Prolonged dwindling

E-FIGURE 3-1.  Different disease trajectories for different illnesses. (Permission obtained from RAND Corporation © Lynn J. Perspectives on care at the close of life. Serving patients who may die soon and their families: the role of hospice and other services. JAMA. 2001;285:925-932.)

Date:

Time:

Please circle the number that best describes your average symptom over the past 24 hours: Worst Pain

No Pain 0

1

2

3

4

5

6

7

8

9

10

0

1

2

3

4

5

6

7

8

9

10

0

1

2

3

4

5

6

7

8

9

10

0

1

2

3

4

5

6

7

8

9

10

0

1

2

3

4

5

6

7

8

9

10

0

1

2

3

4

5

6

7

8

9

10

0

1

2

3

4

5

6

7

8

9

10

0

1

2

3

4

5

6

7

8

9

10

0

1

2

3

4

5

6

7

8

9

10

0

1

2

3

4

5

6

7

8

9

10

No Fatigue

Worst Fatigue Worst Nausea

No Nausea

Not Depressed No Anxiety

Worst Depression

No Drowsiness No Shortness of Breath

Worst Drowsiness Worst Shortness of Breath Worst Possible

Best Appetite

Best Feeling or Well Being

Worst Anxiety

Best Sleep

Worst Feeling of Well Being Worst Sleep

Completed by :

Patient

Family

Assessed by (Signature/Credentials/ID#/Date/Time) Print/Stamp Name: E-FIGURE 3-2.  Edmonton Symptom Assessment System. (Hui D, Bruera E. The Edmonton Symptom Assessment System 25 years later: past, present, and future developments. J Pain Symptom Manage. 2017;53:630-643.)

CHAPTER 3  Palliative Care  

11

TABLE 3-1 APPROACHES TO THE MANAGEMENT OF PHYSICAL AND PSYCHOLOGICAL SYMPTOMS SYMPTOM

ASSESSMENT

TREATMENT

Pain

How severe is the symptom (as assessed with the use of validated instruments) and how does it interfere with the patient’s life? What is the etiology of the pain? Is the pain assumed to be neuropathic or somatic? What has the patient used in the past (calculate previous days’ equal analgesic dose)?

Prescribe medications to be administered on a standing or regular basis if pain is frequent. For mild pain: use acetaminophen or a nonsteroidal anti-inflammatory agent. For moderate pain: titrate short-acting opioids (see Table 27-4). For severe pain: rapidly titrate short-acting opioids until pain is relieved or intolerable side effects develop; start long-acting opiates once pain is controlled. Rescue doses: prescribe immediate-release opioids—10% of the 24-hour total opiate every hour (orally) or every 30 minutes (parenterally) as needed. Concomitant analgesics (e.g., corticosteroids, anticonvulsants, tricyclic antidepressants, and bisphosphonates) should be used when applicable (particularly for neuropathic pain). Consider alternative medicine and interventional treatments for pain.

Constipation

Is the patient taking opioids? Does the patient have a fecal impaction?

Prescribe laxatives for all patients on opiates. If ineffective, add drugs from multiple classes (e.g., stimulant, osmotic laxatives, and enemas). Prescribe methylnaltrexone if still constipated.

Shortness of breath

Ask the patient to assess the severity of the shortness of breath. Does the symptom have reversible causes?

Prescribe oxygen to treat hypoxia-induced dyspnea, but not if the patient is not hypoxic. Opioids relieve breathlessness without measurable reductions in respiratory rate or oxygen saturation; effective doses are often lower than those used to treat pain. Aerosolized opiates do not work. Fans or cool air may work through a branch of the trigeminal nerve. Use reassurance, relaxation, distraction, and massage therapy.

Fatigue

Is the patient too tired for activities of daily living? Is the fatigue secondary to depression? Is a disease process causing the symptom or is it secondary to reversible causes?

Provide cognitive education about conserving energy use. Treat underlying conditions appropriately.

Nausea

Which mechanism is causing the symptom (e.g., stimulation of the chemoreceptor trigger zone, gastric stimulation, delayed gastric emptying or “squashed stomach” syndrome, bowel obstruction, intracranial processes, or vestibular vertigo)? Is the patient constipated?

Prescribe an agent directed at the underlying cause (Chapter 123). If persistent, give antiemetic around the clock. Multiple agents directed at various receptors or mechanisms may be required.

Anorexia and cachexia

Is a disease process causing the symptom, or is it secondary to other symptoms (e.g., nausea and constipation) that can be treated? Is the patient troubled by the symptom or is the family worried about what not eating means?

A nutritionist may help find foods that are more appetizing (Chapter 202). Provide counseling about the prognostic implications of anorexia (Chapter 206).

Delirium

Is the cause reversible? Is the confusion acute, over hours to days? Does consciousness wax and wane? Is there a problem of attention? Does the patient have disorganized thinking? Does the patient have an altered level of consciousness—either agitated or drowsy?

Identify underlying causes and manage symptoms (Chapter 25). Recommend behavioral therapies, including avoidance of excess stimulation, frequent reorientation, and reassurance. Ensure presence of family caregivers and explain delirium to them. Prescribe haloperidol, risperidone, or olanzapine.

Depression

Over the last 2 weeks, have you been bothered (0) not at all, (1) several days, (2) more than half the days, (3) every day by: + Little interest or pleasure in doing things + Feeling down, depressed, or hopeless

Add the points for each answer. For a score >2, further evaluation is recommended, with consideration of supportive psychotherapy, cognitive approaches, behavioral techniques, pharmacologic therapies (see Table 369-5), or a combination of these interventions. Prescribe psychostimulants for rapid treatment of symptoms (within days) or selective serotonin reuptake inhibitors, which may require 3 to 4 weeks to take effect; tricyclic antidepressants are relatively contraindicated because of their side effects.

Anxiety (applicable also Over the last 2 weeks, have you been bothered (0) not at all, (1) for family members) several days, (2) more than half the days, (3) every day by: + Feeling nervous, anxious, or on edge + Not being able to stop or control worrying

Add the points for each answer. A score of >2 should lead to a more in-depth evaluation (see Chapter 369) and consideration of supportive counseling and benzodiazepines (Table 369-9).

Spiritual distress

Inquire about spiritual support.

Are you at peace?

Modified from Morrison RS, Meier DE. Palliative care. N Engl J Med. 2004;350:2582-2590.

in more than 75% of patients with cancer. Advances such as intrathecal pumps and neurolytic blocks are helpful in the remaining 25% (Chapter 27). The use of oxygen is not helpful for refractory dyspnea except when hypoxia has been documented, whereas use of medications for depression often can be helpful (Chapter 369).  

Is the Patient Psychologically Suffering?

Patients may be physically comfortable but still suffering. Psychological symptoms and syndromes such as depression, delirium, and anxiety are common in patients with life-limiting or chronic illnesses. It may be difficult to determine

whether increased morbidity and mortality are caused by the physical effects of the illness or by the psychological effects of depression and anxiety on energy, appetite, or sleep. Screening questions focusing on mood (e.g., “Have you felt down, depressed, and hopeless most of the time for the past 2 weeks?”) and anhedonism (e.g., “Have you found that little brings you pleasure or joy in the past 2 weeks?”) have been shown to help in diagnosing depression in this population. Increasing data show that treatment of depression in chronic illness is possible and improves both morbidity and mortality. For patients and families facing mortality, existential and spiritual concerns are common. Progressive illness often raises questions of love, legacy,

12

CHAPTER 3  Palliative Care  

loss, and meaning. A physician’s role is not to answer these questions or to provide reassurance, but rather to understand concerns of the patient and family, how they are coping, and what resources might help. Spirituality often is a source of comfort, and physicians can ascertain a patient’s beliefs using a brief instrument such as the FICA Spiritual Assessment Tool (Table 3-2). A single screening question such as “Are you at peace?” may identify patients who are in spiritual distress and facilitate referrals to chaplains.



Is the Family Suffering?

Families, defined broadly as those individuals who care most for the patient, are an important source of support for most patients. Families provide informal caregiving, often at the expense of their own physical, economic, and psychological health. Good palliative care requires an understanding of how the family is coping and a search for ways to provide family members with the social or clinical resources they need to improve their well-being. Comprehensive and

STUDY ID#

HOSPITAL ID# DO NOT WRITE ABOVE THIS LINE

Brief Pain Inventory (Short Form) Time:

Date: Name:

Last

First

Middle Initial

1. Throughout our lives, most of us have had pain from time to time (such as minor headaches, sprains, and toothaches). Have you had pain other than these everyday kinds of pain today? 1. Yes

2. No

2. On the diagram, shade in the areas where you feel pain. Put an X on the area that hurts the most.

Right

Left

Left

Right

3. Please rate your pain by circling the one number that best describes your pain at its worst in the last 24 hours.

0 No pain

1

2

3

4

5

6

7

8

9

10 Pain as bad as you can imagine

4. Please rate your pain by circling the one number that best describes your pain at its least in the last 24 hours.

0 No pain

1

2

3

4

5

6

7

8

9

10 Pain as bad as you can imagine

9

10 Pain as bad as you can imagine

9

10 Pain as bad as you can imagine

5. Please rate your pain by circling the one number that best describes your pain on the average.

0 No pain

1

2

3

4

5

6

7

8

6. Please rate your pain by circling the one number that tells how much pain you have right now.

0 No pain

1

2

3

4

5

6

7

8

FIGURE 3-1.  Brief Pain Inventory (short form). (Copyright 1991. Charles S. Cleeland, PhD, Pain Research Group. All rights reserved.)

CHAPTER 3  Palliative Care  

13

7. What treatments or medications are you receiving for your pain?

8. In the last 24 hours, how much relief have pain treatments or medications provided? Please circle the one percentage that most shows how much relief you have received. 0% No pain

10%

20%

30%

40%

50%

60%

70%

80%

90%

100% Complete relief

9. Circle one number that describes how, during the past 24 hours, pain has interfered with your: A. General Activity 0 1 Does not interfere

2

3

4

5

6

7

8

9

10 Completely interferes

2

3

4

5

6

7

8

9

10 Completely interferes

2

3

4

5

6

7

8

9

10 Completely interferes

B. Mood 0 1 Does not interfere C. Walking Ability 0 1 Does not interfere

D. Normal Work (includes both work outside the home and housework) 0 1 Does not interfere

2

3

4

5

6

7

8

9

10 Completely interferes

2

3

4

5

6

7

8

9

10 Completely interferes

2

3

4

5

6

7

8

9

10 Completely interferes

2

3

4

5

6

7

8

9

10 Completely interferes

E. Relations with Other People 0 1 Does not interfere F. Sleep 0 1 Does not interfere G. Enjoyment of Life 0 1 Does not interfere FIGURE 3-1, cont’d.

individually targeted interventions can reduce caregivers’ burdens, although the absolute benefits are relatively small. Because patients in palliative care often die, the palliative care team must address bereavement and postdeath family suffering. Good communication and informational brochures in an intensive care unit can decrease family members’ adverse psychological outcomes after death. A letter of condolence or a follow-up phone call to the next of kin after a patient’s death is respectful and offers the opportunity to clarify questions about the patient’s care. Some family members suffer from complicated grief—a recently described syndrome associated with separation and traumatic distress, with symptoms persisting for more than 6 months. Primary care physicians, who have ongoing relationships with the loved one, and hospices, which provide bereavement services

for a year after the patient’s death, have the opportunity to assess whether the grief symptoms persist or worsen.  

Is the Patient’s Care Consistent with the Patient’s Goals?

The sine qua non for palliative care is ensuring that the treatment plan is consistent with the patient’s values. Some patients prefer longevity over quality of life, but a large proportion of elderly, seriously ill patients are not focused on living as long as possible. Instead, they want to maintain a sense of control, relieve their symptoms, improve their quality of life, avoid being a burden on their families, and have a closer relationship with their loved ones. Ensuring that treatment is consistent with a patient’s goals requires good communication skills (Table 3-3). The approaches to giving bad news,

14

CHAPTER 3  Palliative Care  

TABLE 3-2 FICA SPIRITUAL ASSESSMENT TOOL

TABLE 3-3 CORE COMMUNICATION SKILLS

F—What is your faith/religion? Do you consider yourself a religious or spiritual person? What do you believe in that gives meaning/importance to life? I—Importance and influence of faith. Is your faith/religion important to you? How do your beliefs influence how you take care of yourself? What are your most important hopes? What role do your beliefs play in regaining your health? What makes life most worth living for you? How might your disease affect this? C—Are you part of a religious or spiritual community? Is this of support to you, and how? Is there a person you really love or is very important to you? How is your family handling your illness? What are their reactions/expectations? A—How would you like me to address these issues in your health care? What might be left undone if you were to die today? Given the severity or chronicity of your illness, what is most important for you to achieve? Would you like me to talk to someone about religious/spiritual matters?

RECOMMENDED SKILL

From Puchalski C, Romer A. Taking a spiritual history. J Palliat Med. 2000;3:129-137.

discussing goals of care, and talking about forgoing life-sustaining treatment have similar structures.5 First, the patient needs to understand the basic facts about the diagnosis, possible treatments, and prognosis. The communication skill that helps physicians communicate information is Ask-Tell-Ask—exploring what the patient knows or wants to know, then explaining or answering questions, and then providing an opportunity for the patient to ask more. In the hospital, where discontinuity of care is common and misunderstandings frequent, it is important to determine what the patient knows before providing information so as to keep everyone well coordinated. When giving bad news, knowing what the patient knows allows the physician to anticipate the patient’s reaction. Finally, information must be titrated based on the patient’s preferences. Although most patients want to hear everything about their disease, a minority do not. There is no foolproof way to ascertain what any patient wants to know other than by asking. When giving patients information, it is important to give small pieces of information, not use jargon, and to confirm that patients understand what they have been told.6 Caregivers should focus on the key message the patient should hear (the headline) rather than overwhelming the patient with biomedical information. Giving information is like dosing a medication: one gives information, checks understanding, and then gives more information based on what the patient has heard. After ensuring that the doctor and the patient have a shared understanding of the medical facts, the physician should engage in an open-ended conversation about the patient’s goals as the disease progresses. This strategy requires that the patient be asked about both hopes and fears. One might ask: “Given the medical situation, as you think about the future, what brings you joy and pleasure?” “If your time is limited, what are the things that are most important to achieve?” “What are your biggest fears or concerns?” “As you think about the future what do you want to avoid or not have the doctors do?” The clinician can use an understanding of these goals to make recommendations about which treatments to provide and which treatments would not be helpful. As a result, early palliative care can improve quality of life, mood, and even survival. Physicians find talking about prognosis particularly difficult for two reasons: first, it is hard to foretell the future accurately; and second, they fear this information will “take away patients’ hope.” Thus, they often avoid talking to patients about these issues unless specifically asked. Although some patients do not want to hear prognostic information, for many patients, this information helps them plan their lives. Patients who are told that their disease is generally terminal are more likely to spend a longer period of time in hospice and avoid aggressive treatment at the end of life, without adverse psychological consequences. Furthermore, their families usually have fewer postdeath adverse psychological outcomes. Given that one cannot guess how much information to provide, a physician can start these conversations by asking, “Are you the kind of person who wants to hear about what might happen in the future with your illness or would you rather take it day by day?” If the patient requests the latter, the physician can follow up by asking if there is someone else with whom he or she can talk about the prognosis. Second, before giving prognostic information, it is useful to inquire about the patient’s concerns in order to provide information in the most useful manner. Finally, it is appropriate when discussing prognostic information to acknowledge uncertainty: “The course of this cancer can be quite unpredictable, and physicians don’t have a crystal ball. I think you should

EXAMPLE

A.  IDENTIFYING CONCERNS AND RECOGNIZING CUES Elicit Concerns Open-ended questions “Is there anything you wanted to talk to me about today?” Active listening

Allowing patient to speak without interruption; allowing pauses to encourage patient to speak

Recognize Cues Informational concerns Patient: “I’m not sure about the treatment options” Emotional concerns

Patient: “I’m worried about that”

B.  RESPONDING TO INFORMATIONAL CONCERNS “Ask-tell-ask”

Topic: communicating information about cancer stage

Ask

“Have any of the other doctors talked about what stage this cancer is?”

Tell

“That’s right, this is a stage IV cancer, which is also called metastatic cancer…”

Ask

“Do you have questions about the staging?”

C.  RESPONDING TO EMOTIONAL CONCERNS Nonverbal Empathy: S-O-L-E-R S

Face the patient Squarely

O

Adopt an Open body posture

L

Lean toward the patient

E

Use Eye contact

R

Maintain a Relaxed body posture

Verbal Empathy: N-U-R-S-E N

Name the emotion: “You seem worried”

U

Understand the emotion: “I see why you are concerned about this”

R

Respect the emotion: “You have shown a lot of strength”

S

Support the patient: “I want you to know that I will still be your doctor whether you have chemotherapy or not”

E

Explore the emotion: “Tell me more about what is worrying you”

D. DISCUSSING GOALS OF CARE R

Reframe that the status quo is not working: “I worry that more treatment will hurt you more than help”

E

Expect emotion: “I can see this is not what you wanted to hear”

M

Map the patient values: “Have you ever filled out a living will?” “Given the medical situation, what brings you joy and meaning?” “As you look at the future what do you want to avoid?”

A

Align with the patient values: “What I am hearing you say is.…”

P

Propose a plan

Adapted from Back AL, Arnold RM, Tulsky JA. Discussing Prognosis. Alexandria, VA: American Society of Clinical Oncology; 2008.

be aware of the possibility that your health may deteriorate quickly, and you should plan accordingly. We probably are dealing with weeks to months, although some patients do better, and some do worse. Over time, the course may become clearer, and if you wish, I may be able to be a little more precise about what we are facing.” The physician must discuss these topics in an empathic way. Palliative care conversations are as much about emotions as facts. Talking about disease progression or death may elicit negative emotions such as anxiety, sadness, or frustration. These emotions decrease a patient’s quality of life and interfere with the ability to hear factual information. Empathic responses strengthen the patient-physician relationship, increase the patient’s satisfaction, and make the patient more likely to disclose other concerns. The first step is recognizing

when the patient is expressing emotions. Once the physician recognizes the emotion being expressed, he or she can respond empathically. It is also important for physicians to recognize their own emotional reactions to these conversations. The physician’s emotional reactions color impressions of the patient’s prognosis, thereby making it hard to listen to the patient, and may influence the physician to hedge bad news. The physician should become aware of her or his own emotional reactions to ensure that the conversation focuses on the patient rather than the health care provider’s needs. In addition to good communication skills, palliative care requires a basic knowledge of medical ethics and the law. For example, patients have the moral and legal right to refuse any treatment, even if refusal results in their death. There is no legal difference between withholding and withdrawing life-sustaining treatment. In many states, physicians’ aides can legally help in the dying process by clearly prescribed methods. When confronted with areas of ambiguity, the physician should know how to obtain either a palliative care or ethics consultation. During the past 10 years, there has been a societal push to encourage patients to designate health care proxies and to create advance care planning documents, typified by the use of living wills.7 These documents are meant to protect patients against unwanted treatments and to ensure that as they are dying, their wishes are followed. Unfortunately, there are few empirical data showing that these documents actually change practice. Still, discussions of the documents with health professionals and family members generally provoke important conversations about end-of-life care decisions and may help families confronted with difficult situations know they are respecting their loved one’s wishes.8  

Is the Patient Going to Die in the Location of Choice?

Most patients say that they want to die at home. Unfortunately, most patients die in institutions—either hospitals or nursing homes. Burdensome transitions decrease quality in end-of-life care. Good palliative care requires establishing a regular system of communication to minimize transitional errors. A social worker who knows about community resources is important in the development of a dispositional plan that respects the patient’s goals. Hospice programs are an important way to allow patients to die at home. In the United States, hospice refers to a specific, government-regulated form of end-of-life care, available under Medicare since 1982 but subsequently adopted by Medicaid and many other third-party insurers. Hospice care typically is given at home, a nursing home, or specialized acute care unit. Care is provided by an interdisciplinary team, which usually includes a physician, nurse, social worker, chaplain, volunteers, bereavement coordinator, and home health aides, all of whom collaborate with the primary care physician, patient, and family. Bereavement services are offered to the family for a year after the death. Hospices are paid on a per diem rate and are required to cover all the costs related to the patient’s life-limiting illness. Because of this and the fact that their focus is on comfort rather than life prolongation, many hospices will not cover expensive treatments such as inotropic agents in heart failure or chemotherapy in cancer, even if they have a palliative effect. Not surprisingly, many hospices are experimenting with different service models that do not require patients to forgo possibly life-prolonging treatments, in an attempt to enroll patients earlier in the course of their illness and increase access to their services. Hospice care for nursing home residents is associated with less aggressive care near death but an overall increase in Medicare expenditures.9 By comparison, interdisciplinary palliative care for hospitalized adults with serious illness can reduce costs.10

  Grade A References A1. Kavalieratos D, Corbelli J, Zhang D, et al. Association between palliative care and patient and caregiver outcomes: a systematic review and meta-analysis. JAMA. 2016;316:2104-2114. A2. Beernaert K, Smets T, Cohen J, et al. Improving comfort around dying in elderly people: a cluster randomised controlled trial. Lancet. 2017;390:125-134. A3. Gaertner J, Siemens W, Meerpohl JJ, et al. Effect of specialist palliative care services on quality of life in adults with advanced incurable illness in hospital, hospice, or community settings: systematic review and meta-analysis. BMJ. 2017;357:j2925. A4. Rogers JG, Patel CB, Mentz RJ, et al. Palliative care in heart failure: the PAL-HF randomized, controlled clinical trial. J Am Coll Cardiol. 2017;70:331-341.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 3  Palliative Care  

GENERAL REFERENCES 1. Ferrell BR, Temel JS, Temin S, et al. Integration of palliative care into standard oncology care: American society of clinical oncology clinical practice guideline update. J Clin Oncol. 2017;35:96-112. 2. Kelley AS, Morrison RS. Palliative care for the seriously ill. N Engl J Med. 2015;373:747-755. 3. Hui D, Bruera E. The Edmonton Symptom Assessment system 25 years later: past, present, and future developments. J Pain Symptom Manage. 2017;53:630-643. 4. Kavalieratos D, Gelfman LP, Tycon LE, et al. Palliative care in heart failure: rationale, evidence, and future priorities. J Am Coll Cardiol. 2017;70:1919-1930. 5. Bischoff K, O’Riordan DL, Marks AK, et al. Care planning for inpatients referred for palliative care consultation. JAMA Intern Med. 2018;178:48-54.

15.e1

6. Center to Advance Palliative Care. https://www.capc.org/. Accessed May 10, 2019. 7. Tolle SW, Back AL, Meier DE. Clinical decisions. End-of-life advance directive. N Engl J Med. 2015;372:667-670. 8. Epstein AS, Prigerson HG, O’Reilly EM, et al. Discussions of life expectancy and changes in illness understanding in patients with advanced cancer. J Clin Oncol. 2016;34:2398-2403. 9. Gozalo P, Plotzke M, Mor V, et al. Changes in Medicare costs with the growth of hospice care in nursing homes. N Engl J Med. 2015;372:1823-1831. 10. May P, Normand C, Cassel JB, et al. Economics of palliative care for hospitalized adults with serious illness: a meta-analysis. JAMA Intern Med. 2018;178:820-829.

15.e2

CHAPTER 3  Palliative Care  

REVIEW QUESTIONS 1. A 75-year-old man with lung cancer is admitted to the hospital with severe shortness of breath. Work-up reveals no other cause of his shortness of breath other than lymphogenic spread of his cancer. His oxygen saturation is 94%. Which of the following treatments should be instituted for his dyspnea? A . Morphine B. Benzodiazepines C. Oxygen D. A and C E. All the above Answer: A  In randomized controlled data, opioids have been shown to decrease dyspnea both in lung cancer patients and in patients with COPD. Oxygen is helpful only if the patient has hypoxia. Benzodiazepines have not been shown to decrease breathlessness. 2. Which of the following is NOT required for a patient to be in hospice? A . The patient must be DNR. B. The patient must have a life-limiting illness, which is likely to cause her death in 6 months. C. The patient wishes to focus on quality of life rather than longevity of life. D. If the patient lives at home, she must have a primary caregiver. Answer: A  The patient does not have to be DNR to be in hospice. The others are requirements of hospice. 3. Which of the following is true of depression in life-limiting illnesses? A . It is a normal reaction when people have a life-limiting illness, and it should not be treated. B. It cannot be improved because the treatments take too long to work in patients with serious illness. C. Treatment of depression decreases both morbidity and mortality. D. It requires a psychiatric consult because treatment is very complicated. Answer: C  Data show that the treatment of depression improves both quality of life and mortality.

4. Which of the following is true? A . Telling patients that they have a terminal illness will result in their losing hope. B. Telling patients they have a terminal illness has no impact on their desire for future treatment. C. Telling patients that they have terminal illnesses is associated with their choosing hospice more frequently. D. Patients have clearly stated that they do not want to be told that they have a terminal illness. Answer: C  Data suggest that telling patients that they have a life-limiting illness is associated with a lower likelihood of choosing aggressive care at the end of life and is not associated with poorer psychiatric outcomes. 5. Which of the following is NOT a key component of the definition of specialty palliative care? A . The care is interdisciplinary. B. The care focuses on physical, spiritual, and psychological components. C. The care is inconsistent with life-prolonging treatment. D. The care model includes the family. Answer: C  Specialty palliative care can be given while a patient is also receiving life-prolonging care. All the other answers are correct. 6. Pain is a common symptom in patients with cancer, and opiates are often required to control the pain. Patients who are taking oral opiates should also be assessed for which of the following problems? A . Hypogonadism B. Constipation C. HIV infection D. QT prolongation Answer: B  Constipation is a very common symptom of opiates and most patients on opiates need a laxative. On the other hand, hypogonadism is an uncommon side effect of opiates and should not routinely be screened for. QT prolongation is seen with methadone but not other opiates. HIV is not associated with oral opiate use.

15

CHAPTER 4  Disparities in Health and Health Care  

4  DISPARITIES IN HEALTH AND HEALTH CARE JOHN Z. AYANIAN Disparities in health care and health are evident in all countries around the world. Health disparities often reflect a country’s specific history, such as the legacy of conquest and colonization for American Indians and of slavery and segregation for African Americans. Socioeconomic disparities in health related to poverty or lack of education occur globally and can be reduced by better educational and economic opportunities and by effective health care systems and social services. Most efforts to understand and reduce health disparities have focused on race, ethnicity, and socioeconomic status, and have sought to disentangle the relative impact of health care, health behaviors, and biologic, social, and environmental factors as contributors to these disparities. More recently, these efforts have expanded to assess a wider range of health disparities, including those faced by sexual minorities, people with disabilities, and people in disadvantaged urban or rural areas.

  DEFINITIONS OF HEALTH DISPARITY AND

HEALTH EQUITY

The U.S. Department of Health and Human Services defines a health disparity as a “health difference that is closely linked with economic, social, or environmental disadvantage.” Conversely, health equity is defined as the “attainment of the highest level of health for all people.”1 These definitions build on U.S. National Academy of Medicine reports in which equitable health care was defined as “care that does not vary in quality due to personal characteristics, such as gender, ethnicity, geographic location, or socioeconomic status” and which identified racial and ethnic disparities in health care as an important contributor to disparities in health outcomes. However, not all differences in health care represent unacceptable disparities related to discrimination and unequal treatment of patients in the health care system. For example, differences may be related to clinical appropriateness or patients’ preferences.

  DEMOGRAPHIC CHANGES IN U.S. POPULATION

The racial and ethnic composition of the U.S. population has changed substantially over the past 50 years, growing from 193 million in 1965 to 324 million in 2015, with almost half of this growth related to nearly 60 million new immigrants. During these 50 years, the non-Hispanic white proportion of the U.S. population dropped from 84 to 62%, and the African American proportion rose slightly from 11 to 12%. In contrast, the Hispanic proportion grew substantially from 4 to 18%, and the Asian proportion increased from 1 to 6%. These trends are projected to continue through 2065, when the corresponding proportions of the U.S. population are projected to be 46% for non-Hispanic whites, 13% for African Americans, 24% for Hispanics, and 14% for Asians if current immigration patterns persist.

  RACIAL AND ETHNIC DISPARITIES IN

HEALTH OUTCOMES

Substantial differences in life expectancy between African Americans and white Americans have narrowed in the past 40 years as life expectancy has risen (Fig. 4-1). Notably, life expectancy is now about 3 years longer for Hispanic men and women relative to white men and women. Heart disease and cancer are the two leading causes of death for all five racial and ethnic groups officially designated by the U.S. federal government, but age-adjusted death rates for specific causes vary substantially by race and ethnicity (Table 4-1). African Americans have the highest age-adjusted death rates, overall and due specifically to heart disease and to cancer, followed by non-Hispanic whites. African Americans also have the highest death rates from cerebrovascular disease, diabetes mellitus, and kidney disease, but lowerthan-average death rates from chronic lung disease, poisoning, and suicide. Non-Hispanic whites, in contrast, have higher-than-average death rates from these latter three causes. Hispanics have lower-than-average death rates from all causes except diabetes mellitus (see Table 4-1). American Indians have lower-than-average death rates from most causes except diabetes mellitus and poisoning, but they also

CHAPTER 4  Disparities in Health and Health Care  

ABSTRACT

Life expectancy in the United States varies substantially by race, ethnicity, and socioeconomic status. Relative to other racial and ethnic groups, African Americans have the highest death rates overall and due to heart disease and cancer—the two most common causes of death in the United States. American adults with low incomes have substantially shorter life expectancy than more affluent adults, and the magnitude of this disparity differs widely among U.S. geographic areas. These differences arise from differences in health risk factors such as smoking and hypertension, as well as disparities in insurance coverage and access to care. The increasing diversity of the U.S. population has created new challenges and opportunities for health care providers and organizations to serve patients from diverse backgrounds. Trained interpreters improve care for patients with limited English proficiency, and coordinated care especially benefits disadvantaged patients by reducing fragmentation. Health care organizations should implement reporting systems to monitor disparities in the quality and outcomes of care, set measurable goals for reducing disparities, and encourage programs to meet these goals.

KEYWORDS

health care disparities health disparities health equity race ethnicity socioeconomic factors

15.e3

16

CHAPTER 4  Disparities in Health and Health Care  

100 2015 Hispanic or Latino

Life expectancy (years)

White, not Hispanic Black, not Hispanic 80

Male

White female

79.3

Black female

76.3 White male

71.8 Female

60 Black male

84.3 81.1 78.1

0 1975 1980 1985 1990 1995 2000 2005 2010 2015

0

20

40 60 80 Life expectancy (years)

100

FIGURE 4-1.  Life expectancy at birth. Note: Life expectancy data by Hispanic origin were available starting in 2006 and were corrected to address racial and ethnic misclassification. (Source: NCHS, Health, United States, 2016, Figure 6. Data from the National Vital Statistics System [NVSS].)

TABLE 4-1 AGE-ADJUSTED DEATH RATES FOR 10 LEADING CAUSES OF DEATH BY RACE/ETHNICITY IN UNITED STATES, 2015* ALL PERSONS

WHITE, NON-HISPANIC

AFRICAN AMERICAN

HISPANIC OR LATINO

ASIAN OR PACIFIC ISLANDER

AMERICAN INDIAN OR ALASKA NATIVE

All causes

733.1

753.2

851.9

525.3

394.8

596.9

Heart disease

168.5

171.9

205.1

116.9

86.5

118.5

Cancer

158.5

163.7

180.1

110.3

99.0

107.9

Chronic lung disease

41.6

46.9

28.9

17.7

12.2

30.9

Cerebrovascular disease

37.6

36.4

50.8

32.3

29.8

24.7

Alzheimer disease

29.4

30.8

26.6

24.2

14.7

15.4

Diabetes mellitus

21.3

18.9

37.0

25.2

15.7

34.2

Influenza & pneumonia

15.2

15.4

15.9

11.4

14.0

12.5

Poisoning

14.8

18.8

11.1

7.7

2.4

16.1

Kidney disease

13.4

12.2

25.4

11.4

8.3

12.2

Suicide

13.3

17.0

5.6

6.2

6.4

12.6

*Per 100,000 population, from Health, United States, 2016: With Chartbook on Long-term Trends in Health. Hyattsville, MD: National Center for Health Statistics; 2017:120-123.

have markedly elevated death rates from chronic liver disease (26.4 deaths per 100,000 vs. 10.8 among all persons). Asians and Pacific Islanders together have lower-than-average death rates from each of the 10 leading causes of death, including markedly lower rates for heart disease, cancer, chronic lung disease, Alzheimer disease, poisoning, and suicide. Major health risk factors that contribute to morbidity and mortality among adults vary substantially by race, ethnicity, and level of education. African American adults have the highest age-adjusted prevalence of hypertension (43%; Chapter 70), which is a major contributor to their high rates of heart disease, cerebrovascular disease, and kidney disease; whereas the prevalence of hypertension is substantially lower among non-Hispanic whites (29%), Hispanics (28%), and Asians (27%). In contrast, the prevalence of diabetes mellitus is substantially higher among African Americans (18%), Mexican Americans (18%), and Asians (16%) than among non-Hispanic whites (10%).2 Smoking rates (Chapter 29) vary widely in the United States by race/ethnicity and sex. Rates are highest among non-Hispanic white men (21%), African American men (22%), and American Indian men (28%) and women (24%). Smoking rates are intermediate among non-Hispanic white women (19%), African American women (14%), Hispanic men (16%), and Asian men (15%), and they are lowest among Hispanic women (7%) and Asian women (5%).3

  SOCIOECONOMIC DISPARITIES IN

HEALTH OUTCOMES

Socioeconomic gradients in morbidity and mortality, which are a major component of health disparities, have widened in the United States in recent

years. Adults with high incomes have experienced substantial gains in life expectancy related to their lower smoking rates and better control of hypertension, hyperlipidemia, and other chronic health conditions and risk factors. By comparison, adults with low incomes have experienced minimal gains overall.4 Among middle-aged non-Hispanic white adults without postsecondary education, life expectancy has actually decreased since 1999 as a result of rising death rates from alcohol-related liver disease, drug overdoses, and suicide.5 U.S. smoking rates have declined substantially since 1974, with the steepest drop among college graduates (Fig. 4-2). Higher smoking rates among less educated adults remain a major contributor to socioeconomic disparities in morbidity and mortality. The temporal improvement in smoking rates overall has been offset by marked increases in age-adjusted death rates due to drug overdoses—particularly among non-Hispanic whites, who have had a threefold increase in this death rate from 1999 to 2015 primarily due to opioid overdoses (Fig. 4-3).6 The magnitude of socioeconomic disparities in mortality vary widely by geographic region within the United States. When considering life expectancy at age 40, for example, adults with incomes in the lowest quartile can expect to survive to age 81 years in New York City and several California cities, but only to age 77 years in some cities in Ohio, Indiana, and Michigan. These disparities in life expectancy are primarily related to regional differences in the prevalence of behavioral and metabolic risk factors, including smoking, limited physical activity, obesity, hypertension, and diabetes mellitus.7

CHAPTER 4  Disparities in Health and Health Care  

17

60 Men

Women

No high school diploma

50

High school diploma or GED

Percent

40

No high school diploma High school diploma or GED

30

20

Some college

Some college

Bachelor’s degree or higher

10

Bachelor’s degree or higher

0 1974 1979

1985 1990 1995 2000 2005 2010 2015 1974 1979

1985 1990 1995 2000 2005 2010 2015

FIGURE 4-2.  Current cigarette smoking: Adults aged 25+. Note: Smoked 100 cigarettes in their lifetime and now smoke every day or some days. (Source: NCHS, Health, United States, 2016, Figure 10 and Table 48. Data from the National Health Interview Survey [NHIS].)

Deaths per 100,000 standard population

25

20 Non-Hispanic white1,2 15

10

Non-Hispanic black1 Hispanic1

5

0 1999

2001

2003

2005

2007

2009

2011

2013

2015

FIGURE 4-3.  Age-adjusted drug overdose death rates, by race and ethnicity: United States, 1999-2015. Significant increasing trend, p < 0.005. Rate for non-Hispanic white persons was significantly higher than for non-Hispanic black and Hispanic persons, p < 0.001. Notes: Deaths are classified using the International Classification of Diseases, Tenth Revision. Drug overdose deaths are identified using underlying cause-of-death codes X40-X44, X60-X64, X85, and Y10-Y14. Deaths for Hispanic persons may be underreported by about 5%. Access data table for Figure 3 at: https://www.cdc.gov/nchs/data/databriefs/db273_table.pdf#3. (Source: Hedegaard H, Warner M, Miniño AM. Drug overdose deaths in the United States, 1999–2015. NCHS data brief, no 273. Hyattsville, MD: National Center for Health Statistics. 2017.) 1

  INSURANCE COVERAGE AND ACCESS TO CARE

Health insurance coverage is an important contributor to racial, ethnic, and socioeconomic disparities in health care and health outcomes, especially for adults who are under age 65 years and do not have near-universal insurance coverage through Medicare as older adults do. Since the Affordable Care Act was enacted in 2010, 31 states have expanded Medicaid to non-elderly adults with annual incomes below 138% of the federal poverty level (60 years

3A. Age of delayed degenerative diseases

>60 years

3B. Age of health regression and social upheaval

50-60 years

PREDOMINANT CARDIOVASCULAR DISEASES AND RISK FACTORS

MODERN REGIONAL EXAMPLES

Infections, rheumatic heart disease, and nutritional cardiomyopathies

Rural India, sub-Saharan Africa, South America

10-35

As above plus hypertensive heart disease and hemorrhagic strokes

China

35-65

All forms of strokes; ischemic heart disease at Aboriginal communities, urban young ages; increasing obesity and diabetes India, former socialist economies

60 mm Hg, oxygen saturation >90%), is the mainstay of therapy. When oxygen alone fails, noninvasive methods for improving ventilation or tracheal intubation are required (Chapter 96). Oxygen should increase the Po2 in all patients except those who have severe right-to-left shunting (Chapter 61). Treatment of conditions that cause hypoxemia includes antibiotics (pneumonia), bronchodilators (asthma, chronic obstructive pulmonary disease), diuretics and vasodilators (pulmonary edema), anticoagulants (pulmonary embolism), hyperbaric oxygen (carbon monoxide poisoning), methylene blue (methemoglobinemia, sulfhemoglobinemia), and transfusion (anemia). GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 7  Approach to the Patient with Abnormal Vital Signs  

GENERAL REFERENCES 1. Gabayan GZ, Gould MK, Weiss RE, et al. Emergency department vital signs and outcomes after discharge. Acad Emerg Med. 2017;24:846-854. 2. Nguyen OK, Makam AN, Clark C, et al. Vital signs are still vital: instability on discharge and the risk of post-discharge adverse outcomes. J Gen Intern Med. 2017;32:42-48. 3. Jayasundera R, Neilly M, Smith TO, et al. Are early warning scores useful predictors for mortality and morbidity in hospitalised acutely unwell older patients? A systematic review. J Clin Med. 2018;7: 309. 4. Cantillon DJ, Loy M, Burkle A, et al. Association between off-site central monitoring using standardized cardiac telemetry and clinical outcomes among non-critically ill patients. JAMA. 2016;316: 519-524.

31.e1

5. Geneva II, Cuzzo B, Fazili T, et al. Normal body temperature: a systematic review. Open Forum Infect Dis. 2019;6:1-7. 6. Oka T. Stress-induced hyperthermia and hypothermia. Handb Clin Neurol. 2018;157:599-621. 7. Bentzer P, Griesdale DE, Boyd J, et al. Will this hemodynamically unstable patient respond to a bolus of intravenous fluids? JAMA. 2016;316:1298-1309. 8. Hale ZE, Singhal A, Hsia RY. Causes of shortness of breath in the acute patient: a national study. Acad Emerg Med. 2018. [Epub ahead of print.]

31.e2

CHAPTER 7  Approach to the Patient with Abnormal Vital Signs  

REVIEW QUESTIONS 1. A patient presents with malaise, cough, and shortness of breath. Vital signs include temperature 40° C, blood pressure 120/74 mm Hg, respiratory rate 18 breaths per minute, pulse 70 beats per minute, and oxygen saturation 97%. This presentation could be consistent with: A . Streptococcal pneumonia B. Pyelonephritis due to Escherichia coli C. Legionella pneumonia D. Influenza-like illness E. Mycoplasma pneumonia Answer: C  This patient is exhibiting a pulse-temperature dissociation because the pulse (70) is far lower than one would expect given that the patient is febrile to 40° C. This phenomenon is seen in a number of conditions, including typhoid fever and legionella infection. The other conditions would all be expected to produce tachycardia unless the patient could not become tachycardic because of medications (e.g., β-blockers) or cardiac conduction problems. 2. An 88-year-old man presents from a nursing home with slight agitation and vital signs that include temperature 38.7°  C, blood pressure 96/64 mm Hg, respiratory rate 22 breaths per minute, pulse 94 beats per minute, and oxygen saturation 96%. Physical examination reveals dry mucous membranes, clear lungs, a soft abdomen, an indwelling Foley catheter, and slightly cool but noncyanotic extremities. The patient should be given: A . Antipyretics (e.g., acetaminophen) B. Intravenous normal saline, 500 mL with additional boluses as tolerated C. Intravenous antibiotics D. All of the above E. Only A and B until urine culture results are available Answer: D  This case is an example of how vital signs can guide treatment in the absence of a firm diagnosis. The patient meets all three of the physical examination criteria for the systemic inflammatory response syndrome (SIRS) and is likely septic. The physician should not wait for his white blood cell count or other laboratory results to initiate antibiotic treatment because evidence suggests that early antibiotics are a crucial step in preventing morbidity and mortality. Although antibiotics should not be overused, the early provision of appropriate broad-spectrum antibiotics before the confirmation of a specific diagnosis is prudent and may be life-saving for this patient.

3. An intern is awakened at 3 am by the ward nurse regarding a patient who is postoperative day 2 from a hip replacement and is newly tachycardic. Vital signs include temperature 36° C, blood pressure 146/82 mm Hg, respiratory rate 18 breaths per minute, pulse 112 beats per minute, and oxygen saturation 97% on room air. The intern drowsily orders a 1000-mL normal saline fluid challenge for dehydration. Later that morning, the patient is acutely intubated for respiratory distress. What most likely went wrong? A . The intern failed to consider pulmonary embolism as a possible cause for the tachycardia. B. The intern failed to consider fat embolism in a patient who had recently undergone hip surgery. C. The intern failed to consider sepsis in the differential diagnosis. D. The intern failed to consider failure of the patient controlled anesthesia (PCA) pump in the differential diagnosis. E. The intern failed to realize that tachycardia can be present in both dehydration and heart failure. Answer: E  First and foremost, the intern’s main mistake was not getting out of bed to evaluate the patient in person. Vital signs alone are not sufficient data on which to base an important clinical decision. Although choices A and B are certainly possible in a postoperative orthopedic patient, heart failure is a more likely diagnosis. There is little clinical support for the other choices. 4. A patient arrives in the emergency department comatose with decreased respiratory rate in the winter. Vital signs are temperature 36° C, blood pressure 128/68 mm Hg, respiratory rate 10 breaths per minute, pulse 100 beats per minute, and oxygen saturation 100% on room air. Pupils are 6 mm and reactive, and lungs are clear. What is the single most important initial treatment? A . High-flow O2 administered by non-rebreather mask B. Intravenous normal saline, 1000 mL with additional boluses as tolerated C. Intravenous antibiotics D. Naloxone, 0.8 mg IV E. Immediate endotracheal intubation Answer: A  This patient may have carbon monoxide poisoning. It is winter, a time when people use heating devices that may have incomplete combustion. The pupillary examination is not suggestive of opiate intoxication (D), and no other diagnosis is apparent. Because oxygen is the best treatment for this condition and is generally harmless in adults, it makes sense to initiate this therapy while efforts (e.g., blood gas analysis with co-oximetry) are made to confirm the diagnosis. There is no basis for thinking this patient is dehydrated (B) or infected (C), and intubation would be premature (E). Remember that pulse oximetry is falsely elevated in carbon monoxide poisoning, so the 100% oxygen saturation means nothing.

CHAPTER 8  Statistical Interpretation of Data  

31

  INCIDENCE AND PREVALENCE

Incidence indicates the number of subjects who develop a condition over a specific time period divided by the population at risk. Incidence is usually expressed as a rate (e.g., 10% per year), but it can also be a proportion, as in lifetime incidence. Prevalence, which indicates the number of subjects who have a condition at one point in time divided by the population at risk, is always expressed as a proportion. The prevalence of a disease at any one point in time is defined by the product of incidence of the disease multiplied by the average duration of the disease.

  INTERPRETING DATA

Physicians are continuously confronted with data from many sources, including clinical measurements (e.g., vital signs), laboratory tests, and imaging studies. In fact, the work of the modern-day physician is in many respects one of information management. Although the physician may not conduct independent statistical analyses unaided, knowledge of the fundamentals of data analysis is critical for the ability to evaluate whether and how to integrate information into clinical practice. This knowledge is also essential for understanding the medical literature and incorporating it into practice. Millions of research articles are published each year, and their quality varies widely. Articles published in the leading medical journals undergo rigorous statistical review, but the same is not true for most other journals. Although it generally is inadvisable to base clinical decisions upon a single study, physicians are often confronted with new information and the dilemma of how to deal with it. Both of these goals—interpreting clinical data and understanding the medical literature—require familiarity with the laws of probability and the field of statistics.

  DATA AND THEIR CHARACTERISTICS

8  STATISTICAL INTERPRETATION OF DATA AND USING DATA FOR CLINICAL DECISIONS THOMAS H. PAYNE Statistical techniques are core to evidence-based medicine. The scientific method is based on generating a hypothesis and designing an experiment to test that hypothesis. This powerful approach has been at the root of advances in human endeavors and particularly within medicine. A clear description of a question, a formulation of a hypothesis, a careful design of an experiment to test the hypothesis, the gathering of unbiased data, and the appropriate analysis of the results of the experiment permit the determination of whether a test or treatment is useful or not. If these steps are not conducted in a rigorous manner, statistical tests are likely to produce inaccurate or unreliable results.

Variability is a constant of nature. Statistical techniques represent an effort to describe data and distinguish differences that reflect random variation from those that represent true differences. First, it is necessary to define the nature of data elements being analyzed. Numeric data can vary over a wide range, with values anywhere between the extremes of that range. Blood pressure, for example, can have values of 130, 129.3, 75, and any value in between. Such data are called continuous. Numeric data that can only have an integer value, such as number of siblings, are called discrete. Other data, which are reasonable only if they fall into one of several categories, such as whether someone is alive or dead, are categorical. Labeling a test as positive or negative, which may depend on a somewhat judgmental threshold rather than a continuum, is another example of categorical data. The next step in examining data is to assess how they are distributed. The distribution of values within a population (e.g., blood pressures) is often categorized as normal (i.e., Gaussian). A normal distribution is often characterized using both measures of central tendency (i.e., mean, median, and mode) and measures of dispersion around the center of the distribution (e.g., standard deviation). However, many phenomena in medicine produce distributions that are not normal (e.g., Poisson, binomial, etc.).

  HYPOTHESIS TESTING

In medical research, the goal is typically to discern whether two groups differ in a meaningful way such as whether an outcome in a group that received a particular treatment differs from the outcome in a group that received no treatment or a placebo. This comparison begins with a hypothesis that is stated formally as the null hypothesis and is phrased in relation to an alternative hypothesis. The two hypotheses are mutually exclusive and exhaustive. So if a study compares change in blood pressure between a group that received an intervention and one that did not (controls), the null hypothesis would be that the intervention had no effect on blood pressure and the alternative hypothesis would be that it did have an effect. A study should be designed to determine whether to accept or reject the null hypothesis. In the analysis of data collected in the study, the challenge is to determine whether differences seen in the distribution of blood pressure in people who received the treatment differed from those who did not; and if there is a difference, whether that difference is due to chance. The appropriate techniques for testing hypotheses vary depending on the nature of the data we collect (E-Table 8-1). Because blood pressure measurements vary, any true difference between intervention subjects and control subjects may be difficult to detect if there is only a small number of study subjects; the natural, biologic dispersion of

CHAPTER 8  Statistical Interpretation of Data  

ABSTRACT

An important and common element of the physician’s role is to make decisions. To make good decisions, a physician must be able to interpret clinical data and understand the medical literature based on familiarity with the laws of probability and the field of statistics. Effective decision making also includes gathering information, determining whether more information is worth the additional risk and expense, and interpreting results. All of these tasks depend on using data optimally. Information gathering begins with the most important data: those derived from a careful patient history and physical examination. However, the variety of laboratory tests and diagnostic procedures greatly increases the amount of information that must be ordered intelligently and interpreted correctly. Appropriate use of clinical information can make important differences in patients’ outcomes and in the resources required to optimize them.

KEYWORDS

probability statistical analysis decision making Bayes theorem decision analysis prediction rules

31.e3

31.e4

CHAPTER 8  Statistical Interpretation of Data  

E-TABLE 8-1 TYPES OF VARIABLES AND COMMONLY USED STATISTICAL METHODS ASSOCIATED STATISTICAL METHODS TYPE OF OUTCOME VARIABLE

EXAMPLES

BIVARIATE

MULTIVARIATE

Categorical (dichotomous)

Alive; readmission to the hospital within 30 days

2 × 2 table, chi-square analysis

Logistic regression

Categorical (nominal)

Race; cancer, tumor type

Chi-square analysis

Nominal logistic regression

Categorical (ordinal)

Glasgow Coma Scale

Mann-WhitneyWilcoxon, Kruskal-Wallis

Ordinal logistic regression

Numerical (continuous)

Cholesterol; SF-36 scales*

t Test, analysis of variance

Linear regression

Numerical (count)

Number of times pregnant; number of mental health visits in a year

Mann-WhitneyWilcoxon, Kruskal-Wallis

Poisson regression, linear models

Time to event regression

Time to breast cancer; time to viral rebound in HIV-positive subjects

Log rank

Cox proportional hazards

*Numerical scores with many values are often treated as though they were continuous. HIV = human immunodeficiency virus; SF-36 = short-form 36-item health survey. Courtesy of Thomas B. Newman, MD, MPH and Charles E. McCulloch, PhD.

32

CHAPTER 8  Statistical Interpretation of Data  

blood pressure may obscure a true difference between the two. However, if the number of subjects is increased, the ability to detect a true difference also increases, and some number should be sufficient to detect a meaningful difference if it truly exists. Below that number, the study may fail to detect a difference between blood pressure readings of the intervention and control groups even though a difference genuinely exists. This level is termed beta and represents the probability of a Type II error (i.e., failure to detect a true difference when one truly exists). When the number of subjects exceeds the minimum number, there can be greater confidence that a difference will be observed if one truly exists. The probability below which the null hypothesis can be rejected is termed the alpha level and represents the probability of rejecting the null hypothesis when the null hypothesis is actually true. A Type I error occurs when a null hypothesis is rejected although it is true. To eliminate any possibility of a Type I error would necessitate an infinite number of subjects, so by convention, a 5% or lower chance of this error is considered acceptable. Thus, at an alpha level of 0.05, a difference that is detected will be a true difference 95% of the time, but 5% of the time a random difference will be mistakenly classified as a true difference. Because falsely concluding that a difference exists (e.g., concluding that a treatment is helpful when it is actually ineffective) is considered a more serious error than the reverse (concluding that a treatment is ineffective when, in fact, it is potentially helpful), levels for beta error typically are set much lower than those for alpha (e.g., 80%). Power analysis is a method for estimating the minimum sample size necessary to detect the minimum difference between groups that is defined as clinically meaningful at specified alpha level (typically 0.05) and beta level. P values are often reported in medical studies.1 P values answer the question, “If the null hypothesis were true, what would be the probability of obtaining, by chance alone, a value of the test statistic this large or larger?” For a fixed level alpha, the power of a test increases with sample size and increases with the true difference between the null hypothesis and a specified alternative. In essence, the foundation of hypothesis testing is determining whether there is a difference and whether any difference is explained by chance. This determination hinges on probability, as well as conventions regarding tolerance for false-positive (calling a difference real when it actually is a chance effect) and false-negative (not finding a true difference) conclusions.  

Study Design

When testing the hypothesis that an intervention is effective, there may be other reasons unrelated to the intervention as to why that outcome measurement might have changed. A covariate is another factor, different than the tested intervention, that can potentially influence the outcome under study. For example, in evaluating a medication to treat hypertension, subjects also may have been exposed to a campaign to reduce dietary salt intake. The reduction in salt intake, which could have contributed to any observed changes in blood pressure, would be considered a covariate. It is essential to adjust, or control, for these covariates. The most straightforward way is to assign the intervention to one group but also to measure the change in outcome in an otherwise similar control group as well. The strongest research design is the randomized, double-blinded controlled trial in which individuals are randomly assigned either to receive the intervention to be tested or to receive an inert or inactive substitute. Subjects in both the intervention and control groups are kept uninformed as to whether they are receiving the active intervention or substitute, and the investigator observing the outcomes is also kept unaware of who is assigned to which group. Thus, neither the subject nor the person recording results knows which arm of the intervention is being observed, and so the reporting and recording of results are insulated from belief in the power of the intervention. This approach is regarded as the most accurate way to assess the effect of an intervention, because any effect of covariates would be expected to occur nearly equally in both the intervention and control groups. Furthermore, any placebo effect is minimized by masking whether or not the subject is receiving the treatment. Randomized controlled trials are, however, not always practical for many reasons, and much of what we have learned in medicine is derived from different study designs in which it is difficult to know all of the influences on the outcome. For this reason, nonrandomized studies can never guarantee adequate control for all covariates, even in very large observational studies. For example, consider whether hospital mortality rates increased or declined after a large process change, such as the installation of new computer systems to enter and communicate orders. Any changes in error rates may be due to the new computer system, but they also could be explained by heightened

awareness of the risk for errors. Investigators can attempt to control for known covariates, but other covariates may not be known. For this reason, randomized trials provide the strongest evidence for causality. In recent years, a variety of other quasi-experimental study designs have been applied. For example, an intervention might be applied to an entire practice rather than to individual patients. Although it would be ideal to assign the intervention randomly to participating practices, such an approach may be impractical or impossible. One approach would be to introduce an intervention sequentially to individual practices in a systematic manner—which is termed a stepped wedge design. If this approach cannot be done in a systematic manner or if the evaluation is being conducted post hoc, the data can be retrospectively segmented into short, consecutive time periods so that the analysis can try to distinguish temporal trends from actual effects of the intervention. This design is termed an interrupted time series. When evaluating health care interventions, it is often desirable to include both a quantitative measurement of outcomes (such as mortality or hospital admissions) and observational or qualitative methods, such as surveys or structured interviews. These data are often complementary. If, for example, a study fails to show that an intervention was effective, qualitative information may reveal whether it was properly implemented and what problems prevented implementation. Qualitative methods also are important to understand the attitudes and reactions of health care personnel and patients.  

Time-to-Event Data

A common study design is to follow a sample of patients who receive a treatment until a prespecified outcome has occurred, such as death or hospital admission. The analysis uses the time between the subject’s participation in the study until the outcome. Results are often displayed graphically as a survival distribution. Some subjects may contribute data to the study but not complete it. A Kaplan-Meier analysis allows estimation of survival in a manner that uses data from patients who drop out of the study.  

Noninferiority Trials

In some research, the goal is not to find a more effective treatment but to find one that has other advantages, such as fewer side effects or lower cost.2 In these studies, called noninferiority trials, the purpose is to evaluate a new treatment against an existing one with the goal of demonstrating that it is at least as good, for example, comparing a new treatment with a standard treatment in which the goal is not to find an approach that is more effective but to find a therapy that has lower cost or fewer adverse effects with at least similar efficacy to the standard treatment.

  CAUSALITY

There is an important difference between association and causality. The rooster’s crow is associated with, but does not cause, sunrise. Anoxia causes death. An investigator’s—and the public’s—belief that a treatment is beneficial is powerful, and the scientific method can help us determine the difference between belief and reality. Belief can influence biology. The placebo effect is the concept that effects can be due to belief and not to the intervention itself.  

Confounding

If a covariate is related to both the outcome and to the exposure or risk factor being studied, and if it is unequally distributed between the groups being compared, it becomes a confounder. Since most health outcomes have many contributing causes, there often are many possible confounders. For example, in a study of the association between obesity and heart disease, age may be a confounder. Age is related to obesity and also to heart disease. If the obese subjects are older than the nonobese subjects, then differences in the outcome of heart diseases may be due to the older age of the obese subjects rather than to their obesity. In observational studies in which randomization is not possible, steps to reduce bias due to confounding should be considered in the analysis. In an observational study of the effect of a treatment on an outcome of interest, it may be that the treatment is given to some subjects (patients) because of their characteristics, such as age or severity of illness. A variety of sophisticated statistical techniques are applied in an effort to reduce bias in observational studies. For example, a propensity score is the probability that a study subject would receive the exposure or treatment of interest, based on the subject’s characteristics and the clinical environment.3 Using the propensity score, it is possible to adjust for the effect of known confounding variables so that it is less likely the difference in outcome is due to confounding bias.

33

CHAPTER 8  Statistical Interpretation of Data  

Nevertheless, all techniques to reduce confounding depend entirely on data about relevant covariates. If such data are missing, statistical adjustment cannot produce an accurate result. A limitation of the propensity score in reducing confounding bias is that many confounders may be unknown or difficult or even impossible to measure.  

Yes

No

Positive

a

b

Negative

c

d

Machine Learning

Machine learning approaches problems by learning rules from data in contrast with expert systems. Starting with patient-level observations, algorithms sift through vast numbers of variables, looking for combinations that reliably predict outcomes. This process in some ways resembles traditional regression models: there are outcomes, covariates, and statistical functions linking the two. Machine learning can handle enormous numbers of predictors and combining them in nonlinear and highly interactive ways. Optimism for machine learning should be tempered by understanding that association is not the same as causation and that vast biomedical data available to apply to these techniques may have insufficient breadth—lacking for example social determinants of health—or insufficient quality that can hamper the results.4



Disease present

  STATISTICAL APPROACHES AND CHALLENGES Multivariable Statistics

Oftentimes a researcher is interested in the relative (as well as the absolute) importance of multiple individual predictor variables on an outcome, independent of the effects of all other variables.5 For example, in studying whether air pollution increases the risk of lung cancer, an investigator would want to take into account differences in age, race, family history, smoking history, and even radon exposure (Chapter 182). Using multivariable statistical techniques, such estimates can be made.

Multiple Comparisons

Analyses in which multiple statistical tests are conducted can lead to a false conclusion that one of them is a positive result.6 The more tests conducted, the more likely it becomes that one will be positive based on chance alone even though the P value may be set to 0.05. One way to correct for this problem is to use the Bonferroni correction, which divides the P value for rejecting the null hypothesis by the number of hypothesis tests performed.

Meta-analysis

Meta-analysis is a statistical method for summarizing the results of multiple studies to estimate an overall effect and confidence interval for a parameter measured by those studies. In meta-analysis, the formal combination of results considers variations both within the study and among studies.7

Test result

From these data, one can also calculate: Risk ratio or relative risk (RR)

a (a + b)

Relative risk reduction (RRR) Risk difference or absolute risk reduction (ARR) Number needed to treat (NNT)

÷

c c+d

1 – RR a (a + b)



c (c + d)

1 ARR

Odds ratio (OR)

ad bc

FIGURE 8-1.  A 2 × 2 table evaluating use of a test to determine if a disease is present. In this model, test sensitivity is a/(a + c), specificity is d/(b + d), and disease prevalence is (a + b)/(a + b + c+ d). Positive predictive value can then be calculated as a/(a + b) and negative predictive value as d/(c + d).

  USING DATA FOR CLINICAL DECISIONS

An important and common element of the physician’s role is to make decisions. Effective decision making includes gathering information, determining whether more information is worth the additional risk and expense, and interpreting results. All of these tasks depend on a physician’s skill in using data optimally. Information gathering begins with the most important data: those derived from a careful patient history and physical examination. However, the variety of available laboratory tests and diagnostic procedures greatly increases the amount of information that can be obtained and that must be ordered intelligently and interpreted correctly. Appropriate use of clinical information can make important differences in patients’ outcomes and in the resources required to optimize them. Evaluating a patient who presents with new breathlessness (Chapter 77) presents an example of decision-making principles. A thorough history and physical examination set the foundation, and a spectrum of choices are considered. Is this a temporary problem or a life-threatening condition for which immediate and potentially risky testing and treatment should be recommended? (E-Table 8-2) The sequence in which testing is undertaken also has implications for safety and cost. Some tests are simple and fast to perform and most helpful when results are negative; others involve substantial risk and expense but can give a definitive answer.  

Theories and Principles Useful in Decision Making

Much of medical practice involves risk and uncertainty. When discussing uncertainty, the concept of probability is useful. Probability is the likelihood that something occurs; a probability of 1.0 means it is certain to occur, and a probability of 0 means it will not occur. The term odds refers to a ratio of probabilities of an event (p): Odds = p/(1 − p). If the probability of the event

is 50% or 1 in 2, as in a coin flip, then odds of the event are 1 (or 1 : 1). If the probability is 1 in 3, or 0.33, then the odds are 0.33/0.66 or one half (1 : 2). A simple yet instructive model of decision making is a 2 × 2 table (Fig. 8-1). A group (or population) of patients is divided into 4 cells. The table shows the presence or absence of a disease across the top and test results on the left. The cells in this 2 × 2 table are labeled with letters beginning with a in the upper left and d in the lower right. The sensitivity of a test, which is defined as the probability of positive results when the disease is known to be present, is calculated in the table as a/(a + c). Test specificity, which is the probability of a negative test result when the disease is absent, is calculated as d/(b + d). Sensitivity and specificity are referred to as test characteristics because they explain how a test performs in the presence or absence of disease. In clinical practice, however, tests are usually used to determine whether a disease or condition is present or absent, or at least how likely it is based on a given test result. In the 2 × 2 table, this probability, which can be calculated as a/(a + b), is referred to as the positive predictive value. Similarly, d/(c + d), which is the negative predictive value of the test, indicates how likely it is that the disease is absent given a negative test result. Inspection of this 2 × 2 table shows another very important finding: how common the disease is in this population, shown as (a + c)/(a + b + c + d). This calculation is analogous to the prevalence of the disease in the population. A special insight from this simple 2 × 2 table is that positive and negative predictive values depend on the test characteristics, but importantly they also depend on the prevalence of the disease in this particular population of patients. Different positive and negative predictive values result when the same test is

CHAPTER 8  Statistical Interpretation of Data  

E-TABLE 8-2 PRINCIPLES OF TEST ORDERING AND INTERPRETATION The interpretation of test results depends on what is already known about the patient. No test is perfect; clinicians should be familiar with its diagnostic performance and never believe that a test “forces” them to pursue a specific management strategy. Tests should be ordered if they may provide additional information beyond that already available. Tests should be ordered if there is a reasonable chance that the data will influence the patient’s care. Two tests that provide similar information should not be ordered. In choosing between two tests that provide similar data, use the test that has lower costs or causes less discomfort and inconvenience to the patient. Clinicians should seek all of the information provided by a test, not just an abnormal or normal result. The cost-effectiveness of strategies using noninvasive tests should be considered in a manner similar to that of therapeutic strategies.

33.e1

34

CHAPTER 8  Statistical Interpretation of Data  

Sensitivity

Number of patients

A

With disease

B

Without disease

FIGURE 8-2.  For each of the vertical lines, those results to the left indicate a positive result while results on the right indicate a negative result.

used in a population in which the disease is rare compared with when it is used in a population in which the disease is common. This insight is the essence of Bayes theorem: the likelihood of the disease given a positive test result depends on both the test characteristics and on the prevalence of the disease in the population. Bayes theorem is often expressed by the formula P(A | B) = (P(B | A) * P(A))/P(B), (probability of disease A given test result B is the probability of result B given disease A times the probability of disease A, divided by probability of result B), but the essence of the greatest value of Bayes theorem to physicians can be understood from the simple 2 × 2 table. Another useful concept is the likelihood ratio, which is the prevalence of a sign (or any test result) in patients with the diagnosis of interest divided by the prevalence of the identical finding (or test result) in patients without the target diagnosis.8 The positive likelihood ratio is expressed by the formula P(B | A ) P(B | A ) (or [sensitivity] ÷ [1 − specificity]) (E-Table 8-3). Other common measures used in decision making can be derived from these simple building blocks that come from the 2 × 2 table. Pretest probability in the 2 × 2 model is the same as the prevalence of disease—how common it is in that population. Before doing any test, the probability of disease is simply the prevalence in that population. After performing the test, we may have a different probability—the post-test probability—which is why we performed the test. In the 2 × 2 model, a test is defined as either positive or negative. However, it may not be clear where to draw the line (or threshold) between positive and negative. Figure 8-2 shows 3 possible divisions between a positive and negative test result. The vertical line on the left excludes all patients who do not have the disease, but it classifies less than half of the patients with the disease as positive. The one on the right results in all patients with the disease having a positive result, but at the cost of including patients who do not have the disease in the positive result. The middle line achieves a compromise— most of the patients with the disease are defined as having a positive result and most of the patients without the disease are defined as having a negative result. Which of the three lines is chosen depends on how the test is being used. A plot of all such possible lines with the sensitivity and specificity creates a curve showing how sensitivity and specificity are related for the test: a receiver operator characteristic (ROC) curve (Fig. 8-3) is a common way to display this relationship. An ideal test would have a curve closer to A than to B—a high sensitivity and a high specificity. ROC curves can also be used to compare the ability of different tests to distinguish between the presence or absence of disease. Tests with higher areas under the curve are considered better tests. In most cases, physicians do not rely on a single test to determine whether or not a condition or disease is present. Rather, several test results and other data (which may include portions of the patient’s history and findings on the physical examination), as well as findings from past records, are incorporated into that judgment. This multiplicity of information highlights one of the limitations of Bayes theorem. Multiple test results may not be independent of each other, but Bayes theorem assumes that they are. It is also important to be aware that it is difficult to determine the accuracy of historical and examination data—particularly if not gathered first-hand but obtained from an electronic health record. Some laboratory data imaging and pathology results may vary depending on the interpreter who created the report. Astute clinicians understand that data may be inaccurate or subject to differing interpretation.

1 − Specificity FIGURE 8-3.  A receiver operator characteristic (ROC). This displays the relationship between all possible divisions between positive and negative results and test characteristics, sensitivity and specificity. The best test has very high sensitivity and specificity. In this example, A is a better test than B.

When should a test be ordered? Gathering additional information—whether from history, repeated examinations, tests, or imaging—can change the likelihood of disease, as is demonstrated by the 2 × 2 table. In general terms, additional data should be obtained only if they would change the likelihood of a disease or condition at an acceptable cost and risk, and if such a change in probability would affect what is recommended to the patient. “Will it change therapy?” is a common question that summarizes such thinking: Will the test lead to a diagnosis that would be treated differently? If not, then the value of the test is lower. Sometimes the most important choice is to defer further testing and to observe the patient over time—the “test of time.”

Utility

The results of diagnosis and treatment vary, and the importance or value of those outcomes also vary. For example, the value of surviving may depend on quality of life and side effects endured to achieve survival. Quantifying this value can be difficult, but ultimately the most important quantification is from the patient. The term utility describes the value or importance of the outcome as judged by the patient or measured in some other way. In some analyses, the length of survival is adjusted by a measure of the quality of life, described as quality-adjusted life years or QALYs.9

Decision-Making Strategies Shared Decision Making  

Physicians make many decisions, some of which are simple and many of which are complex. A good physician does not make all of these decisions independently. Especially for decisions that involve risks or preferences, the wise physician involves the patient in decision making when possible. Shared decision making considers the utility of different outcomes and the patient’s desires, which may vary widely from one person to the next. Shared decision making is an important element of physician-patient interaction and also has direct ties to the decision-making process.10

Decision Analysis

Answers to some questions involve a series of questions and probabilities that are not practical to answer with a single study or test. As a result, an approach known as decision analysis can be useful (Fig. 8-4). This analysis begins with a question, such as “should a drug be given?,” which then leads to branches stemming from the yes and no answers, each of which has its own branching from what is called a chance node. Chance nodes may lead to other branches. Each chance node has several probabilities that, by definition, must add up to 1. The end of each of the branches is assigned a utility. To determine, for example, if a drug should be given, the utility of each end branch is multiplied by the probability of ending on that branch, and this process is continued or folded back until each of the original choices is assigned a score. Using this model, the best choice is the choice with the highest score. Decision analysis can be applied to decisions in which prospective trials are impractical, such as a public health policy decision, or when a randomized controlled trial would be impractical, such as developing a strategy for screening for cervical cancer. Since probabilities assigned to some of the nodes may not be known and could plausibly be within a range, a technique known as sensitivity analysis

CHAPTER 8  Statistical Interpretation of Data  

E-TABLE 8-3 KEY DEFINITIONS FOR BAYES THEOREM* Probability

A number between 0 and 1 that expresses an estimate of the likelihood of an event

Odds

The ratio of [the probability of an event] to [the probability of the event’s not occurring]

BAYESIAN ANALYSIS Pretest (or prior) probability The probability of a disease before the information is acquired Post-test (or posterior) probability

The probability of a disease after new information is acquired

Pretest (or prior) odds

(Pretest probability of disease)/(1 − pretest probability of disease)

Likelihood ratio

(Probability of result in diseased persons)/ (Probability of result in nondiseased persons)

*Disease can mean a condition, such as coronary artery disease, or an outcome, such as postoperative cardiac complications.

34.e1

Probability 1 Outcome

Utility

Outcome

Utility

Outcome

Utility

Choice A

1 − (Probability 1) Choice node

Chance node

Probability 2

Choice B

1 − (Probability 2)

Outcome

Utility

FIGURE 8-4.  Example of decision analysis. The model begins with a choice (choice node) leading to chance nodes. The probability of branches from each choice node add up to 1. At the end of all branches is an outcome with a utility value. By folding back each chance node—adding probability × utility successively—a value is assigned to both choices.

can be used to determine if the key conclusions of the decision analysis model would change as the chance node probabilities change within that range. Including benefit and cost in decision analysis permits a cost-benefit (or costeffectiveness) analysis11 that can lead to informed decisions regarding what strategy or approach should be pursued, or whether a new technology or treatment should be adopted.  

Aids to Decision Making

As the volume of text, test results, images, and genomic data rises, the limitations of human cognition often are reached. For example, most physicians are able to keep several data elements in mind when making a decision, but decision-making aids become worth considering when the number of facts to be considered rises to include a dozen or more. Some of these aids are simple prediction rules,12 which can use clinical data to estimate the pretest probability of a condition. In other cases, automated systems, as simple as smartphone applications and as complex as cloud-based systems, can be helpful for diagnostic and therapeutic decision making. With progressively rising volumes of health care data and computing power, machine learning, artificial intelligence, and other analytical tools and techniques may increasingly provide additional aid to clinician decision making. To apply these concepts, consider a patient with breathlessness. Does this patient have a pulmonary embolism or some other condition? Although data derived from the history and physical examination individually have a low sensitivity, the pretest probability of pulmonary embolism (Chapter 74) can be classified as high, moderate or low when used together to calculate a Wells score (Chapter 74). In patients so classified as having low risk, the D-dimer test has high sensitivity and therefore a high negative predictive value; as a result, a negative test reduces the risk of pulmonary embolism substantially. The D-dimer test carries a trivial risk to the patient, and its cost is low. Based on its test characteristics, a negative D-dimer result can safely exclude venous thromboembolism and limit the number of patients requiring further evaluation with imaging techniques. Conversely, in a patient with high pretest probability,

a test with higher risk and expense may be warranted, such as computed tomographic pulmonary angiography, which has both a high sensitivity and a high specificity. By balancing the patient’s pretest probability, as determined using a prediction rule, with a test’s expenses, risks, and operating characteristics, an appropriate decision-making strategy can be developed and applied to an individual patient. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 8  Statistical Interpretation of Data  

GENERAL REFERENCES 1. Chavalarias D, Wallach JD, Li AH, et al. Evolution of reporting P values in the biomedical literature, 1990-2015. JAMA. 2016;315:1141-1148. 2. Kaji AH, Lewis RJ. Noninferiority trials: is a new treatment almost as effective as another? JAMA. 2015;313:2371-2372. 3. Haukoos JS, Lewis RJ. The propensity score. JAMA. 2015;314:1637-1638. 4. Chen JH, Asch SM. Machine learning and prediction in medicine—beyond the peak of inflated expectations. N Engl J Med. 2017;376:2507-2509. 5. Meurer WJ, Tolles J. Logistic regression diagnostics: understanding how well a model predicts outcomes. JAMA. 2017;317:1068-1069. 6. Dmitrienko A, D’Agostino RB Sr. Multiplicity considerations in clinical trials. N Engl J Med. 2018;378:2115-2122.

35.e1

7. Cargo M, Harris J, Pantoja T, et al. Cochrane qualitative and implementation methods group guidance series-paper 4: methods for assessing evidence on intervention implementation. J Clin Epidemiol. 2018;97:59-69. 8. McGee S. Teaching evidence-based physical diagnosis: six bedside lessons. South Med J. 2016;109:738-742. 9. Neumann PJ, Cohen JT. QALYs in 2018—advantages and concerns. JAMA. 2018;319:2473-2474. 10. Spatz ES, Krumholz HM, Moulton BW. Prime time for shared decision making. JAMA. 2017;317:1309-1310. 11. Sanders GD, Maciejewski ML, Basu A. Overview of cost-effectiveness analysis. JAMA. 2019. [Epub ahead of print.] 12. Alba AC, Agoritsas T, Walsh M, et al. Discrimination and calibration of clinical prediction models: users’ guides to the medical literature. JAMA. 2017;318:1377-1384.

35.e2

CHAPTER 8  Statistical Interpretation of Data  

REVIEW QUESTIONS 1. A study of 105 vegan Buddhist nuns randomly sampled from monasteries around Ho Chi Minh City found that the average femoral neck bone mineral density was 0.62 g/cm2, with a standard deviation of 0.11 g/cm2. Which of the following statements about this result is correct? A . The 95% confidence interval for the mean bone mineral density in these women is about 0.4 g/cm2 to 0.84 g/cm2. B. If bone mineral density is normally distributed, we would expect about 10% of the women in the sample to have bone mineral density outside of the interval: 0.4 g/cm2 to 0.84 g/cm2. C. The 95% confidence interval for the mean bone mineral density in these women is about 0.60 g/cm2 to 0.64 g/cm2. D. Because the women were sampled randomly, there is a 95% chance that a randomly selected woman from the population would have a bone mineral density between 0.60 g/cm2 and 0.64 g/cm2. E. Because the women were sampled randomly, there is a 95% chance that a randomly selected woman from the sample would have a bone mineral density between 0.60 g/cm2 and 0.64 g/cm2. Answer: C  We would expect about 95% of observations to be within 2 standard deviations of the sample mean, leaving 5% out, so choice B is incorrect. The 95% confidence interval for a sample mean is about mean ±2 standard SD errors of the mean (SEM), where the SEM = . In this case the SD is N 2 0.11 g/cm and the N is about 100, so the SEM will be about .11/10 = .01 g/ cm2 and the 95% CI will be about 0.60 g/cm2 to 0.64 g/cm2, as indicated in choice C. Choices D and E are incorrect because the range given is too narrow: it is ±2 SEM when it should be ±2 SD. 2. A study of data collected through the “Get with the Guidelines-Stroke Program” examined the time from onset of stroke symptoms to treatment with tissue-type plasminogen activator (tPA) among 58,353 patients with acute ischemic stroke treated within 4.5 hours of the onset of symptoms. The authors reported that “faster onset-to-treatment time, in 15-minute increments, was associated with…increased achievement of independent ambulation at discharge (OR, 1.04; 95% CI 1.03-1.05; P < 0.001)…” Which of the following is a correct interpretation of these findings? A . In this study, the effect of a 15-minute reduction in onset-to-treatment time was associated with a 4% (relative) increase in the odds of independent ambulation at discharge. B. Because the odds ratio is very close to 1.0, the results are not statistically significant. C. Because the odds ratio is very close to 1.0, the results, while highly statistically significant, are not clinically significant. D. The 4% increase in odds of independent ambulation translates into a number needed to treat (NNT) of 25. E. None of the above is correct. Answer: A  Choice A exactly expresses the meaning of the odds ratio for this study. Choices B and C are incorrect because the proximity of the odds ratio to 1 is in this case based partly on the choice of the authors to express it per 15 minutes onset-to-treatment time. This illustrates the importance of knowing the units of the predictor variable when it is not dichotomous. If the authors had expressed the difference per hour instead of per 15 minutes, the odds ratios would have been taken to the fourth power, i.e., the odds ratio for independent ambulation at discharge would have been about 1.044 = 1.17. Choice D is incorrect because estimation of the NNT requires knowing the absolute risk reduction, which was not provided in this case.

3. Assume a study of both smoking and nonsmoking mothers reports that the effect of smoking on birthweight is about a 32-gram decrease in birth weight per cigarette smoked per day during pregnancy. Which of the following statements about this finding is NOT correct? A . The result is based on a model, in which the predicted birthweight is linearly related to the number of cigarettes smoked per day. B. This model predicts that the difference in birthweight between a baby whose mother did not smoke and one who smoked 5 cigarettes per day is the same as the difference between babies of mothers who smoked 20 and 25 cigarettes per day. C. The model predicts the same effect of smoking on birthweight, regardless of the mother’s age and prepregnancy weight. D. This model predicts that cutting cigarette smoking in half will lead to a 64 gram expected weight increase in the baby. E. The model could include additional terms that would reflect the effect of mother’s age and prepregnancy weight. Answer: D  Choices A, B and C accurately describe characteristics of a linear model for the effect of smoking on birthweight. Choice D is not consistent with a linear model. Choice E reflects that the effects of other variables can be taken into account, while still maintaining a linear model for the effect of cigarettes smoked on birthweight. 4. A case-control study of the relationship of breast cancer with the use of COX-2 inhibitors found that the odds ratio and confidence interval relating cancer to use of baby aspirin was OR = 0.77, 95% CI (0.42-1.41). The authors then stated “Neither acetaminophen nor baby aspirin had any effect on the relative risk of breast cancer.” This is incorrect because: A . The confidence interval crosses 1. B. They do not give the p-value. C. The confidence interval has a lower limit of 0.42. D. Odds ratios are inappropriate for this study. E. They should have used a higher level of confidence. Answer: C  C is correct because the confidence interval allows for a 58% reduction (from (1-.42)*100%) in the chance of breast cancer associated with the use of baby aspirin, a potentially important effect. Choice A is incorrect because it merely indicates a lack of a statistically significant result and does not bear on the size of the effect. Choice B is incorrect because CIs are more useful for ruling out important effects. Odds ratios are especially useful in case-control studies, so D is incorrect. And E is incorrect because a higher level of confidence would make the CI even wider. 5. In a randomized trial of an intervention to reduce hypertension, researchers selected subjects from a single antihypertensive patient club in Shanghai. About one third of those approached agreed to be randomized. The results of this study can be safely generalized to: A . All hypertensives. B. All hypertensives in China. C. All hypertensives in Shanghai. D. All hypertensives belonging to the single club in Shanghai. E. None of the above. Answer: E  Because two thirds of the participants refused to participate, the extrapolation of the effect of the intervention might not even apply to the particular club in which the study was conducted, much less a broader population.

CHAPTER 8  Statistical Interpretation of Data  

6. A 53-year-old man presents to his primary care physician with a chief complaint of chest pain for the last two weeks. The pain is described as intermittent, usually exertional, midline, and aching in nature. His electrocardiogram is completely normal. Which of the following is the most appropriate next step? A . Watch and wait B. Exercise electrocardiography C. B-type natriuretic peptide D. Exercise nuclear scintigraphy E. Coronary angiography Answer: B  The immediate question is whether this patient’s symptoms represent new ischemic heart disease. His clinical presentation suggests a moderate probability of coronary artery disease, so watch-and-wait is probably too risky a strategy. At the same time, he does not appear to have unstable ischemic disease, so there is no need to proceed immediately to coronary angiography in preparation for coronary revascularization. Measurement of B-type natriuretic peptide would not alter management. Of the two noninvasive tests for ischemic heart disease, exercise electrocardiography is the least expensive, most convenient, and carries no radiation exposure. Guidelines would thus suggest B as the most appropriate next step. 7. A patient undergoes an exercise test, and has 2 mm of ST-segment depression on electrocardiogram. In which of these patients would this finding be most likely to change management? A . A healthy 19-year-old volunteer in a research study B. A 62-year-old woman who is completely asymptomatic C. A 62-year-old woman with frequent nonexertional aching chest pain D. A 62-year-old woman with recent myocardial infarction and chest pain at rest E. A 62-year-old woman who is completely asymptomatic four months after coronary artery bypass graft surgery

35.e3

Answer: C  Patients A and B are of sufficiently low probability for having coronary disease that the exercise test abnormality is highly likely to be a false-positive result. Patient D has an extremely high probability of coronary disease, and likely needs coronary angiography and revascularization as next steps; she does not need exercise electrocardiography, because the test is unlikely to change this management plan. Patient E’s care is also not likely to be influenced by an exercise test result because she is asymptomatic after major coronary revascularization surgery. Patient C has a low-to-moderate probability of coronary disease, and this abnormal exercise test result moves her into a midrange probability. Thus, she is likely to undergo either further testing, initiation of antianginal therapy, or both as a result of this exercise test result. 8. Which of the following is NOT an important consideration when weighing whether to order a test for a patient? A . Test may influence decision making for patient’s care B. Test is less expensive than alternative strategies C. Test is safer than alternative strategies D. Test reduces patient’s or clinician’s uncertainty E. Test is expected to be abnormal in presence of patient’s already confirmed diagnosis Answer: E  Tests should be ordered when they are expected to change care, and should be chosen on basis of safety, cost, and impact. They should not be ordered simply because they can confirm an already known diagnosis.

35

CHAPTER 9  Measuring Health and Health Care  

9  MEASURING HEALTH AND HEALTH CARE CAROLYN M. CLANCY AND ERNEST MOY Physicians routinely quantify a variety of health measures, including symptoms, vital signs, and findings on physical examination, to improve diagnosis, treatment, and prognostication. Similarly, the efficacy and quality of health care can and should be measured for several reasons. First, the quality of care delivered is often suboptimal. Persistent variations in practice for patients with the same diagnosis reflect a combination of clinical uncertainty, individualized practice styles, patients’ preferences and characteristics (age, race, ethnicity, education, income), and other factors. Both

35.e4

ABSTRACT

CHAPTER 9  Measuring Health and Health Care  

Measuring health, health care quality, and health care resources is an essential dimension of contemporary health care organizations. This chapter describes the different approaches to measurement, the evidence base for those measures, and how to match available measures to the needs and interests of patients, policymakers, and other stakeholders.

KEYWORDS

health care quality health care value evidence-based health care

36

CHAPTER 9  Measuring Health and Health Care  

suboptimal care and varied care for the same condition undermine the historical assumption that a combination of highly trained health professionals and accredited facilities is sufficient to ensure consistent high-quality care. A second major trend is attributable to the successes of biomedical science: the major challenge in health care today is the management of chronic disease for a population with increased life expectancy. For chronic conditions, health benefits are increasingly measured in improvements in functional status or quality of life, rather than simply using mortality rates or life expectancy. A third trend relates directly to how the increasing costs of health care are now threatening public budgets and investments in other social goals, such as education. Although the United States spends more per capita on health care than any other developed nation, the outcomes achieved lag far behind. Finally, advances in communication and information technologies have inspired more people to play an active role in their health and health care. These innovations have accelerated demands for transparency and shared decision making. As health insurance and health care regulation have expanded, requirements to track and justify health care services have grown. Intensifying urgency to improve the quality of health care, reduce disparities, control costs, and enhance transparency will likely lead patients and insurers to demand more data and to link quality measures to payments for services. Fortunately, modern technology can help to meet the demand for data. Patients can record and submit their health parameters using handheld devices connected to personal health records. Automated billing programs can track health care services, and electronic health records can assess the quality of physician care. Ultimately, fully integrated health information systems will allow patient information to be retrieved instantly and seamlessly whenever and wherever it is needed. In addition to assessing care quality today, these tools offer enormous promise for learning as a byproduct of care delivery.

  HOW ARE HEALTH AND HEALTH

CARE MEASURED?

Three types of measures typically assess health and health care. Measures of health quantify the sickness or well-being of a person. Measures of health care quality quantify the extent to which a patient receives needed care and does not receive unnecessary care. Health care quality is assessed using measures of structure (e.g., education and credentialing of clinicians), process (adherence to professional standards and evidence-based recommendations), and outcomes (or end results of care, including how patients experience their care and their self-reported health and function). Measures of health care resources quantify the resources used (e.g., radiographs, surgery, medication, intensive care) to improve the health of a patient. All measures can be summed up across populations within a practice or community (Table 9-1). Measures of health and health care often overlap (E-Fig. 9-1). Health measures that can be improved by health care, such as blood pressure or blood glucose levels, are often used as health care quality outcome measures. The delivery of quality health care requires the use of resources and the generation of direct health care costs, which may or may not improve health care at the margins of spending. Impaired health that reduces the ability to do work and earn wages but that could have been prevented by the delivery of health care contributes to the indirect costs of health care. At the intersection of health, health care quality, and health care resources are measures of health care value. These measures compare the health benefits of specific health care services with their costs.

  WHY MEASURE HEALTH AND HEALTH CARE?

Measures of health and health care can be used for many purposes by different stakeholders (Table 9-2). Patients can use their own health information to track their progress, adjust their lifestyle, and plan for their future health care needs. Patient-level measures are important to physicians, hospitals, health plans, and policymakers whether they are used to identify sentinel events that represent quality defects (e.g., amputating the wrong leg or giving a patient the wrong medication), to initiate root cause analyses to improve health care quality, or to assess how quality output is linked to cost input (Chapter 10). Hospitals and health plans aggregate measures over their practices to identify opportunities for raising quality, improving efficiency, and reducing care disparities. For measures for which physicians are primarily accountable, these data can be used to acknowledge and reward high-performing physicians, select physicians for inclusion on panels, and produce report cards to inform the public. Patients can use these report cards to select physicians and health plans that best match their health care needs. Regulators may examine measures to assess qualification for licensure and to identify physicians who might benefit

TABLE 9-1 MEASURES OF HEALTH AND HEALTH CARE MEASURES OF HEALTH • Mortality: rates of death typically adjusted for age and sex • Morbidity: incidence and prevalence rates of diseases and their sequelae • Functional status: assessments of a patient’s ability to perform various actions such as activities of daily living or instrumental activities of daily living as observed by a provider or reported by the patient • Self-reported health status: a patient’s assessment of his or her health and well-being MEASURES OF HEALTH CARE QUALITY • Health care outcomes: the end results or health benefits derived from good health care or the health loss attributable to poor health care • Health care processes: assessments of whether the right care was delivered at the right time and in the right way • Health care infrastructure: the availability of resources needed to deliver high-quality health care • Patient perceptions of health care: a patient’s assessment of health care received, usually emphasizing patient-provider communication and shared decision making • Access to health care: the ability of patients to gain entry into health care and navigate to needed resources MEASURES OF HEALTH CARE RESOURCES • Health care utilization: the quantity of health care services that are used • Direct costs: the costs of providers, supplies, and equipment needed to deliver health care • Indirect costs: the costs of lost wages and decreased productivity due to illness or injury that could have been prevented by appropriate health care • Nonmedical costs: the costs of health care not related to the delivery of services, such as administration, advertising, research, and profits earned by health industries TYPE OF MEASURE

EXAMPLE

HEALTH Mortality

Deaths due to colorectal cancer per 100,000 population

Morbidity

New AIDS cases per 100,000 population

Functional status

% of people unable to perform one or more activities of daily living

Self-reported health status

% of people reporting that their overall health is excellent

HEALTH CARE QUALITY Health care outcomes

Deaths per 1,000 hospitalizations with pneumonia

Intermediate outcomes

% of adults with diabetes whose blood pressure is 98%) negative predictive value to exclude coronary artery disease when it is performed and interpreted at experienced centers, can help reduce hospital stay when the findings are normal in emergency department patients at low to intermediate risk of possible ACS. However, coronary computed tomographic angiography does not have incremental value in patients

381

with a negative troponin assay and a nonischemic ECG.  A2  Conversely, the patient who is believed to be at higher risk for ACS or who continues to have typical ischemic chest pain with ECG abnormalities or elevated cardiac biomarkers should not undergo stress testing or coronary computed tomographic angiography but rather should either undergo coronary angiography or be rendered symptom free with medical therapy before stress testing. An echocardiogram may be helpful in the patient with chest pain if the ECG is nondiagnostic (i.e., minimal ST segment or T wave abnormalities). If left ventricular hypokinesis or akinesis is observed during an episode of chest pain and then improves when symptoms resolve, myocardial ischemia is likely. In the patient with anterior T wave inversion of uncertain etiology, hypokinesis of the left ventricular anterior wall suggests that the observed T wave abnormality is due to a severe stenosis of the left anterior descending coronary artery. Because echocardiography can help evaluate and identify alternative causes for the patient’s chest pain (i.e., myocarditis [Chapter 54], aortic dissection [Chapter 69], or pulmonary embolism [Chapter 74]), it is recommended in patients whose diagnosis is uncertain.

Coronary Angiography

Coronary angiography (Chapter 51) should be performed in patients who are thought to be at high risk for death, MI, or recurrent ischemia in the ensuing days, weeks, and months (see later); in patients who have spontaneous or inducible myocardial ischemia despite appropriate medical therapy; and in patients who have a confusing or difficult clinical presentation and a subsequent inconclusive noninvasive evaluation. The results of angiography help determine whether revascularization is appropriate and, if so, whether it should be attempted by coronary artery bypass grafting or percutaneous coronary intervention (PCI) (Chapter 65). In patients with NSTE ACS, coronary angiography demonstrates significant stenosis of the left main coronary artery in about 15% of patients, of all three major epicardial coronary arteries in about 30 to 35% of patients, of two of the three epicardial arteries in about 20 to 30% of patients, and of one major epicardial artery in 20 to 30% of patients. About 15% of patients have no coronary arterial narrowing of hemodynamic significance. Women with NSTE ACS are likely to have less extensive coronary artery disease than men have, and patients with non–ST segment elevation MI usually have more extensive disease than those with unstable angina. On angiography, the coronary arterial lesion responsible for NSTE ACS (the so-called culprit lesion) typically is asymmetrical or eccentric, with scalloped or overhanging edges and a narrow base or neck, features that reflect underlying plaque disruption and thrombus formation. Although obvious thrombus is visible by angiography in only one third of patients with NSTE ACS, coronary angioscopy shows plaque rupture with overlying thrombus in the majority. Interestingly, the lesion that is the nidus for ACS often is not severely stenotic when it is assessed on recently performed angiograms; in fact, two thirds of culprit lesions previously had less than 50% luminal diameter narrowing (and therefore would not have been considered appropriate for revascularization).

Risk Assessment and Triage

The initial evaluation of the patient with possible or suspected ACS should focus on an assessment of the patient’s risk of acutely sustaining a cardiac ischemic event (death, MI, or recurrent ischemia).5 Patients considered to be at low risk for a cardiac ischemic event may be observed in a chest pain evaluation unit for several hours, with repeated troponin level and ECG. If the findings of that brief evaluation are normal, the patient should be discharged home, with further evaluation performed on an outpatient basis. Conversely, patients not at low risk should be hospitalized for further evaluation and treatment. The availability of high-sensitivity cardiac troponin assays has significantly impacted the initial triage for patients with symptoms suggestive of an ACS (Fig. 63-2).6 In making this assessment, the greatest safety comes from continued observation or admission of patients who have a positive troponin level on presentation or at serial testing, have evidence of ischemia on their ECG, or have symptoms indicative of an acute exacerbation of prior coronary disease.7 After the initial triage decision is made, therapeutic interventions are based on the risk of adverse events in the ensuing hours, days, weeks, and months— estimated by either the Thrombolysis in Myocardial Infarction (TIMI) or Global Registry of Acute Coronary Events (GRACE) risk algorithm—balanced against the risk of a bleeding complication from intensive medical therapy (Table 63-1) or an adverse event from an invasive cardiac procedure. On the basis of this initial assessment, the patient’s therapy should be tailored to minimize the likelihood of adverse events.

382

CHAPTER 63  ACUTE CORONARY SYNDROME  

Acute chest pain

hs-cTn 6h

hs-cTn >ULN

Pain ULN)

Painfree, GRACE 65 years Three or more risk factors for atherosclerosis Known coronary artery disease (previous coronary arteriography or myocardial infarction) Two or more episodes of anginal chest pain at rest in the 24 hours before hospitalization Use of aspirin in the 7 days before hospitalization ST segment deviation ≥0.5 mV Elevated serum concentrations of troponin or CK-MB Global Registry of Acute Coronary Events (GRACE)† Age Heart failure (Killip) class Heart rate Systolic blood pressure ST segment deviation Cardiac arrest during presentation Serum creatinine concentration Elevated serum markers of myonecrosis RISK FACTORS FOR BLEEDING COMPLICATIONS WITH INTENSIVE THERAPY‡ Female gender Older age Renal insufficiency Low body weight Tachycardia Systolic arterial pressure (high or low) Anemia Diabetes mellitus

who complain of recurrent symptoms, nitroglycerin should be given sublingually or by buccal spray (0.3 to 0.6 mg). Patients with ongoing or recurrent chest pain should receive intravenous nitroglycerin (5 to 10 µg/minute with use of nonabsorbable tubing), with escalation of the dose in increments of 10 µg/ minute until symptoms resolve or adverse effects develop. Nitroglycerin’s most common adverse effects are headache, nausea, dizziness, hypotension, and reflex tachycardia. Nitrate tolerance can be avoided by periodically providing the patient with a nitrate-free period (i.e., a brief cessation of drug administration). Nitroglycerin should not be given to patients who have received a phosphodiesterase-5 inhibitor (i.e., sildenafil, tadalafil, or vardenafil) within the previous 24 to 48 hours because severe hypotension may ensue.

β-Adrenergic Blockers

β-Adrenergic blockers diminish symptoms and the risk of MI in ACS patients who are not already taking a β-blocker at the time of hospitalization. In the normotensive patient without ongoing chest pain or tachycardia and without contraindications to β-blockers, metoprolol should be initiated at 25 to 50 mg orally every 6 to 8 hours, with the dose increased (to 100 mg twice daily) as necessary to control heart rate, blood pressure, and symptoms. In high-risk patients and in patients with tachycardia or elevated systemic arterial pressure, metoprolol should be administered intravenously (three boluses of 5 mg each given 5 minutes apart) initially, after which an oral dose should be initiated. A reasonable target heart rate is 50 to 60 beats per minute at rest. β-Blockers should not be administered to patients with decompensated heart failure, hypotension, hemodynamic instability, or advanced atrioventricular block. Because most patients with chronic obstructive pulmonary disease or peripheral vascular disease tolerate β-blockers without difficulty, these conditions should not automatically preclude their use.

Calcium-Channel Blockers

*Individuals with three or more of these variables are considered to be at high risk, whereas those with none, one, or two are considered to be at low risk. (From Diez JG, Cohen M. Balancing myocardial ischemic and bleeding risks in patients with non-ST-segment elevation myocardial infarction. Am J Cardiol. 2009;103:1396-1402.) † Each variable is assigned a numerical score on the basis of its specific value, and the eight scores are summed to yield a total score, which is applied to a reference nomogram to determine the patient’s risk. The GRACE application tool is available online at www.outcomes-umassmed.org/grace. (From Brieger D, Fox KA, Fitzgerald G, et al. Predicting freedom from clinical events in non-ST-elevation acute coronary syndromes: the Global Registry of Acute Coronary Events. Heart. 2009;95:888-894.) ‡ The patient’s bleeding risk can be estimated with the tool available at www.crusadebleedingscore. org. (From Subherwal S, Bach RG, Chen AY, et al. Baseline risk of major bleeding in non-STsegment-elevation myocardial infarction: the CRUSADE [Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA Guidelines] Bleeding Score. Circulation. 2009;119:1873-1882.)

Calcium-channel blockers, which cause arterial vasodilation, increase coronary arterial blood flow and lower systemic arterial pressure. The non-dihydropyridine calcium-channel blockers diltiazem and verapamil slow heart rate and are recommended for the patient with a contraindication to a β-adrenergic blocker or persistent or recurrent symptoms despite treatment with nitroglycerin or a β-blocker. Oral diltiazem (30 to 90 mg four times daily of the short-acting preparation or up to 360 mg once daily of the long-acting preparation) is the preferred calcium-channel blocker because it reduces the incidence of myocardial ischemia and recurrent MI in patients with NSTE ACS. Diltiazem is contraindicated in patients with left ventricular systolic dysfunction or pulmonary vascular congestion. Caution should be exercised when combining a β-blocker with diltiazem because the two drugs may act synergistically to depress left ventricular systolic function as well as sinus and atrioventricular nodal conduction. Patients with ACS should not be prescribed short-acting nifedipine unless they are already receiving a β-blocker because it may increase the risk of death. The risks and benefits of long-acting dihydropyridines in patients with NSTE ACS are undefined.

Antiplatelet Agents

ACS patients should receive dual antiplatelet therapy (aspirin and an ADP receptor inhibitor) acutely and for up to 1 year unless the patient has an aspirin

TABLE 63-2 MANAGEMENT STRATEGIES FOR PATIENTS WITH ACUTE CORONARY SYNDROME THERAPY

INITIATION

DURATION

DOSE, ROUTE, AND DURATION

BENEFIT VS. PLACEBO (REDUCED INCIDENCE OF …)

LOW-RISK PATIENT Antianginal β-Blocker*

Within 24 hours

Hospitalization ± indefinitely

Metoprolol, 25-50 mg orally twice daily, titrated up to 100 mg twice daily; or atenolol, 50-100 mg orally daily

Recurrent ischemia

Nitroglycerin

Immediately

Hospitalization ± indefinitely

0.3-0.6 mg sublingually or 5-10 µg/min IV initially and increased by 10 µg/min every 5 minutes

Not studied

Diltiazem or verapamil†

Immediately

Hospitalization ± indefinitely

30-90 mg orally four times daily or up to 360 mg of long-acting preparation orally daily

MI, recurrent ischemia

Immediately

Indefinitely

Atorvastatin, 40-80 mg or rosuvastatin 20-40 mg orally daily

Recurrent ischemia

Aspirin

Immediately

Indefinitely

162-325 mg orally initial dose, then 81 mg orally daily

Death, MI

Clopidogrel

Immediately

1-12 months

300 mg orally initial dose, then 75 mg orally daily

MI, recurrent ischemia

Lipid Lowering Statin Antiplatelet

384

CHAPTER 63  ACUTE CORONARY SYNDROME  

TABLE 63-2 MANAGEMENT STRATEGIES FOR PATIENTS WITH ACUTE CORONARY SYNDROME—cont’d THERAPY

INITIATION

DURATION

DOSE, ROUTE, AND DURATION

BENEFIT VS. PLACEBO (REDUCED INCIDENCE OF …)

Anticoagulant Unfractionated heparin

Immediately

2 to 5 days

IV bolus of 60 U/kg, then 12 U/kg/hour IV adjusted to achieve an aPTT of 50 to 70 seconds

Death or MI (combined)

β-Blocker*

Within 24 hours

Hospitalization ± indefinitely

Metoprolol, 25-50 mg orally twice daily, titrated up to 100 mg twice daily; or atenolol, 50-100 mg orally daily

Death, MI, recurrent ischemia

Nitroglycerin

Immediately

Hospitalization ± indefinitely

0.3-0.6 mg sublingually or 5-10 µg/min IV initially and increased by 10 µg/min every 5 minutes

Not studied

Diltiazem or verapamil†

Immediately

Hospitalization ± indefinitely

30-90 mg orally four times daily or up to 360 mg of long-acting preparation orally daily

MI, recurrent ischemia

Before hospital discharge

Indefinitely

Atorvastatin 40-80 mg or rosuvastatin 20-40 mg orally daily

Recurrent ischemia

Aspirin and Clopidogrel or Prasugrel or Ticagrelor

Immediately

Indefinitely

162-325 mg orally initial dose, then 81 mg orally

Death, MI

Immediately

≥12 months

300-600 mg orally initial dose, then 75 mg orally daily

MI, recurrent ischemia

At time of PCI

≥12 months

60 mg orally initial dose, then 10 mg orally daily

Immediately

≥12 months

180 mg orally initially, then 90 mg twice daily

Cardiovascular death, MI or stroke (combined)‡ Vascular death, MI or stroke (combined)‡

Glycoprotein IIb/IIIa inhibitor (eptifibatide, tirofiban, or abciximab)

At time of PCI

12-24 hours after PCI

Abciximab, IV bolus of 0.25 mg/kg, then 0.125 µg/ kg/min IV (max. 10 µg/min) for 12 hours; or eptifibatide, IV bolus of 180 µg/kg, then 2.0 µg/ kg/min IV for 18-24 hours; or tirofiban, 0.4 µg/ kg/min IV for 30 minutes, then 0.1 µg/kg/min IV for 12 to 24 hours

MI

Immediately

2 to 5 days; discontinue after successful PCI Duration of hospitalization (up to 8 days); discontinue after successful PCI Up to 72 hours; discontinue 4 hours after PCI

IV bolus of 60 U/kg, then 12 U/kg/hour IV adjusted to achieve an aPTT of 50 to 70 seconds 1 mg/kg subcutaneously twice daily

Death or MI (combined)

0.10 mg/kg loading dose followed by 0.25 mg/kg/ hour

Bleeding#

Duration of hospitalization (up to 8 days); if used during PCI, it must be coadministered with another anticoagulant with factor IIa activity

2.5-mg subcutaneous injection once daily

Bleeding¶





MI, recurrent ischemia

HIGH-RISK PATIENT Antianginal

Lipid Lowering Statin Antiplatelet

Anticoagulants Unfractionated heparin or Enoxaparin or Bivalirudin or Fondaparinux

Immediately Immediately (only in patients managed with an early invasive strategy), Immediately

MI, recurrent ischemia§

Invasive Management Coronary angiography followed by revascularization (if appropriate)

Up to 36-80 hours after hospitalization; within 24 hours in “very high risk” patients

*Avoid in the patient with signs of decompensated heart failure, evidence of low-output state, increased risk for cardiogenic shock, or other contraindications to beta blockade (e.g., PR interval >0.24 seconds, second- or third-degree atrioventricular block without a cardiac pacemaker, active asthma, or reactive airway disease). † Avoid in patients with clinically significant left ventricular dysfunction, increased risk for cardiogenic shock, PR interval >0.24 seconds, or second- or third-degree atrioventricular block without a cardiac pacemaker. ‡ Compared with clopidogrel. § Compared with unfractionated heparin. # As monotherapy compared with heparin and glycoprotein IIb/IIIa inhibitor combination. ¶ Compared with enoxaparin. aPTT = activated partial thromboplastin time; MI = myocardial infarction; PCI = percutaneous coronary intervention. Modified from Lange RA, Hillis LD. Optimal management of acute coronary syndromes. N Engl J Med. 2009;260:2237-2240.

CHAPTER 63  ACUTE CORONARY SYNDROME  

allergy or active bleeding.9 In patients with NSTE ACS, aspirin (Chapter 76) reduces the risk of death or MI by about 50%. The recommended dose is 81 mg daily, continued indefinitely. The choice of which ADP receptor antagonist to use in combination with aspirin is determined by each patient’s characteristics (i.e., risk of bleeding), medication costs, and pharmacologic properties of the agent (see following details). The patient who is allergic to or intolerant of aspirin should be treated with an ADP receptor inhibitor (clopidogrel, ticagrelor, or prasugrel [if PCI treated]) alone. Clopidogrel (Chapter 76) is a thienopyridine that blocks the P2Y12 ADP receptor, thereby diminishing ADP-mediated platelet activation. Its antiplatelet activity is synergistic with aspirin because the two agents inhibit different plateletactivating pathways. Clopidogrel is a prodrug that must be metabolized by the cytochrome P-450 system to the active form. Polymorphisms in the cytochrome P-450 isoform CYP2C19, which are present in 15 to 20% of individuals, slow metabolism of the prodrug to the active form, thereby reducing the magnitude of platelet inhibition. Drugs that are potent inhibitors of the CYP2C19 enzyme (e.g., omeprazole, esomeprazole, cimetidine, fluconazole, ketoconazole, voriconazole, etravirine, felbamate, fluoxetine, and fluvoxamine) should not be administered with clopidogrel because they affect the metabolism to its active form and reduce its antiplatelet effects. In subjects with NSTE ACS, the addition of clopidogrel (a loading dose of 300 to 600 mg, then 75 mg daily for up to 1 year) to aspirin reduces the composite end point of cardiovascular death, nonfatal MI, or stroke by 20% (2.1% reduction in absolute risk) compared with treatment with aspirin alone. The benefit of an aspirin-clopidogrel combination is seen as early as 24 hours after drug initiation and persisted for the 12 months of the study, despite an increase in minor bleeding. Prasugrel (Chapter 76), another thienopyridine, has a greater antiplatelet effect and a more rapid onset of action than clopidogrel. In patients with ACS who are referred for PCI, prasugrel in combination with aspirin reduces ischemic events (i.e., a combination of cardiovascular death, nonfatal MI, and stroke) by 20% compared with concomitant clopidogrel and aspirin (2.2% absolute risk reduction) therapy. A5  However, this benefit is obtained at a 0.5% increased risk of life-threatening bleeding and a 0.3% increased risk of fatal bleeding. At present, prasugrel is approved for use in the ACS patient who is referred for PCI. In combination with aspirin, it is administered as a 60-mg oral loading dose followed by a 10-mg daily maintenance dose. Because prasugrel-associated bleeding complications are highest in patients with a previous stroke or transient ischemic attack, age older than 75 years, or a body weight of less than 60 kg, it should not be used in patients with any of these features. Ticagrelor (Chapter 76), a thienopyridine that does not require hepatic activation, has more rapid onset and more pronounced platelet inhibition than clopidogrel. It is a reversible inhibitor of the P2Y12 receptor, so platelet function returns more rapidly after discontinuation than with clopidogrel. In a randomized trial in NSTE ACS patients, the addition of ticagrelor to aspirin reduced the composite end point of vascular death, nonfatal MI, or stroke by about 15% compared with treatment with clopidogrel and aspirin but increased non– procedure-related bleeding by an absolute 0.7%. In combination with aspirin, ticagrelor is administered as a 180-mg oral loading dose, followed by a 90-mg twice-daily maintenance dose. In patients who receive ticagrelor, the daily aspirin maintenance dose should be 100 mg or less, and ticagrelor should not be used in patients with a history of intracranial hemorrhage. Even in patients more than 1 year after a NSTEMI, treatment with ticagrelor at 60 or 90 mg twice daily in addition to aspirin significantly reduces the risk of cardiovascular death, MI, or stroke but not of overall death because of an increased risk of major bleeding. A6  Glycoprotein IIb/IIIa inhibitors (Chapter 76) block platelet aggregation in response to all potential agonists, so they are the most potent antiplatelet agents available. Three glycoprotein IIb/IIIa inhibitors, each of which must be administered parenterally, are available: abciximab is the Fab fragment of a monoclonal antibody to the receptor; eptifibatide is a peptide; and tirofiban is a peptidomimetic molecule. Glycoprotein IIb/IIIa inhibitors reduce the incidence of recurrent ischemic events in patients with NSTE ACS who undergo PCI but not in patients who are managed with medical therapy alone. When a glycoprotein IIb/IIIa inhibitor is administered to PCI patients, it should be initiated at the time of angiography because its routine administration beforehand carries an increased bleeding risk and no improvement in outcomes. The glycoprotein IIb/IIIa inhibitor infusion (see Table 63-2) typically is continued for 12 to 24 hours after PCI.

Anticoagulants

Anticoagulant therapy should be administered to all patients with ACS unless a contraindication, such as active bleeding, is present. For the patient in whom a noninvasive, ischemia-guided management strategy is selected, treatment with unfractionated heparin, low-molecular-weight heparin (LMWH), or fondaparinux is appropriate, with fondaparinux recommended for the patient at increased risk of bleeding. For the patient in whom an invasive management strategy is selected, unfractionated heparin and LMWH are the agents of choice. Although bivalirudin may be preferred in patients undergoing PCI, it is not used in the initial management of the patient with ACS.

385

Heparin

Unfractionated heparin (Chapter 76) exerts its anticoagulant effect by accelerating the action of circulating antithrombin; it prevents thrombus propagation but does not lyse existing thrombi. In the patient with NSTE ACS, the addition of heparin to aspirin reduces the rate of in-hospital ischemic events (i.e., death or MI) by 33%. Unfractionated heparin should be initiated with an intravenous bolus of 60 U/kg, followed by a continuous infusion of approximately 12 U/kg/hour (maximum, 1000 U/hour), adjusted to maintain the activated partial thromboplastin time (aPTT) at 1.5 to 2.5 times control (i.e., 50 to 70 seconds) or a heparin concentration at 0.3 to 0.7 U/mL (by anti–factor Xa determinations). The infusion should be continued for 48 hours or until revascularization is performed, whichever occurs sooner. Frequent monitoring of the aPTT or heparin concentration is necessary because the anticoagulant response to a standard dose of unfractionated heparin varies widely among individuals; even when a weight-based nomogram (see Table 74-6) is followed, the aPTT is outside the therapeutic range more than one third of the time. Mild thrombocytopenia occurs in 10 to 20% of patients treated with unfractionated heparin. In 1 to 5% of patients, a more severe form of thrombocytopenia develops. This antibody-mediated response usually occurs 4 to 14 days after the initiation of treatment (although it may appear more quickly in patients who received heparin within the preceding 6 months) and is associated with thromboembolic sequelae in 30 to 80% of subjects (Chapter 163).

Low-Molecular-Weight Heparin

LMWHs (Chapter 76), which are fragments of unfractionated heparin, exert a more predictable anticoagulant effect, have a longer half-life, and are less likely to cause thrombocytopenia compared with unfractionated heparin. Because they provide predictable and sustained anticoagulation with once- or twice-daily subcutaneous administration, monitoring of their anticoagulant effect is not required. LMWH is superior to unfractionated heparin in preventing MI or death during hospitalization in NSTE ACS patients who have elevated serum cardiac biomarkers as well as in those considered to be at high risk for recurrent ischemia (see Table 63-1). In the low-risk subject, unfractionated heparin and LMWH have similar efficacy. Two LMWHs, enoxaparin and dalteparin, are approved for the treatment of the patient with NSTE ACS. The dose of enoxaparin is 1 mg/kg subcutaneously twice daily, and the dose of dalteparin is 120 IU/kg (maximum, 10,000 IU) subcutaneously twice daily. Therapy should be continued for the duration of the hospitalization, up to 8 days, or until revascularization is performed (whichever occurs first). In obese (>120 kg), thin (30% of R wave amplitude, or both). Chronic Phase

Resolution of ST segment elevation is variable. Resolution is usually complete within 2 weeks of inferior MI, but it can be delayed further after anterior MI. Persistent ST segment elevation, often seen with a large anterior MI, is indicative of a large area of akinesis, dyskinesis, or ventricular aneurysm. Symmetrical T wave inversions can resolve during weeks to months or can persist for an indefinite period; hence, the age of an MI in the presence of T wave inversions is often termed indeterminate. Q waves usually do not resolve after anterior MI but often disappear after inferior wall MI. Early reperfusion therapy accelerates the time course of ECG changes to minutes or hours instead of days to weeks. ST segments recede rapidly, T wave inversions and loss of R waves occur earlier, and Q waves may not develop or progress and occasionally may regress. Indeed, failure of ST segment elevation to resolve by more than 50 to 70% within 1 to 2 hours suggests failure of fibrinolysis and should prompt referral for urgent angiography and consideration of “rescue angioplasty.”

True Posterior Myocardial Infarction and Left Circumflex Myocardial Infarction Patterns

“True posterior” MI presents a mirror-image pattern of ECG injury in leads V1 to V2 to V4 (see Fig. 64-2). The location of injury of an isolated true posterior MI by magnetic resonance imaging actually involves portions of the lateral LV wall and is typically caused by occlusion of a nondominant left circumflex artery. In the precordial leads, the acute phase is characterized by ST segment depression rather than by ST segment elevation. The evolved and chronic phases show increased R wave amplitude and widening instead of Q waves. Recognition of an isolated true posterior acute MI pattern is challenging but important because the diagnosis should lead to an immediate reperfusion strategy. Extending the ECG to measure left posterior leads V7 to V9 increases sensitivity for detection of acute left circumflex–related injury patterns (i.e., ST segment elevation) with excellent specificity (Chapter 48). Other causes of prominent upright anteroseptal forces include right ventricular (RV) hypertrophy, ventricular preexcitation variants (Wolff-Parkinson-White syndrome; Chapter 58), cardiomyopathies, right bundle branch block, and normal variants with early R wave progression. New appearance of these changes or the association with an acute or evolving inferior MI usually allows the diagnosis to be made.

Right Ventricular Infarction

Proximal occlusion of the right coronary artery before the acute marginal branch can cause RV infarction as well as acute inferior MI in about 30% of cases. Because the prognosis and treatment of acute inferior MI differ in the presence of RV infarction, it is important to make this diagnosis. The diagnosis is assisted by obtaining right precordial ECG leads, which are routinely indicated for inferior acute MI (Chapter 48). Acute ST segment elevation of at least 1 mm (0.1  mV) in one or more leads V4R to V6R is both sensitive and specific (>90%) for identifying acute RV injury, and Q or QS waves effectively identify RV infarction.

Diagnosis in the Presence of Bundle Branch Block

The presence of LBBB often obscures ST segment analysis in patients with suspected acute MI. The presence of a new (or presumed new) LBBB in association with clinical (and laboratory) findings suggesting acute MI is associated with high mortality; patients with new-onset LBBB benefit substantially from reperfusion therapy and should undergo triage and treatment in the same way

391

Myocardial injury: cardiac contusion, surgery, ablation, shocks Myocardial inflammation: myocarditis, pericarditis Heart failure Cardiomyopathies: infiltrative, stress, hypertensive, hypertrophic Aortic dissection Severe aortic stenosis Tachycardias PULMONARY CAUSES Pulmonary embolism Pulmonary hypertension Respiratory failure NEUROLOGIC CAUSES Stroke Intracranial hemorrhage OTHER Shock: septic, hypovolemic, cardiogenic Renal failure Modified from Thygesen K, Alpert JS, Jaffe AS, et al. Third universal definition of myocardial infarction. Circulation. 2012;126:2020-2035.

as patients with ST elevation MI do. Certain ECG patterns, although relatively insensitive, suggest acute MI if they are present in the setting of LBBB: Q waves in two of leads I, aVL, V5, and V6; R wave regression from V1 to V4; ST segment elevation of 1 mm or more in leads with a positive QRS complex; ST segment depression of 1 mm or more in leads V1, V2, or V3; and ST segment elevation of 5 mm or more associated with a negative QRS complex. The presence of right bundle branch block (RBBB) usually does not mask typical ST-T wave or Q wave changes except in rare cases of isolated true posterior acute MI, which are characterized by tall right precordial R waves and ST segment depressions.

Differential Diagnosis

Although ST elevation MI is often an easy diagnosis to make on the basis of the presentation and test results (see later), other considerations include acute pericarditis (Chapter 68), acute myocarditis (Chapter 54), stress-induced cardiomyopathy (takotsubo syndrome) (Chapter 54), hyperkalemia, and early repolarization (see Table 64-2). All but early repolarization can be associated with abnormal biomarkers, but none are associated with a coronary occlusion. Early coronary angiography is advised when the cause of ST segment elevation is unclear (see Table 64-1).

Serum Cardiac Biomarkers of Necrosis

Cardiac-derived troponin-I (cTnI) and troponin-T (cTnT) are proteins specific to sarcomeres. Serial measurement of cardiac troponins now has become the preferred biomarker approach for differentiating acute MI from unstable angina and nonacute coronary syndromes. In the appropriate clinical setting, acute MI is indicated by a rising and/or falling pattern of troponin, with one or more value(s) above the 99th percentile upper reference limit. This risingfalling pattern has become increasingly important as more and more sensitive assays have appeared. Troponins become detectable by traditional assays 1 to 4 hours after the onset of acute MI. With highest-sensitivity assays, increased diagnostic sensitivity offers the possibility to exclude MI effectively in 1 to 2 hours.7 Indeed, investigational pathways now have been proposed that may exclude acute MI after a single sample.8 However, such testing has decreased clinical specificity for acute MI because high-sensitivity assays can detect the presence of troponin in most normal individuals, and increased levels are observed in a number of non-MI settings, including myocarditis (Chapter 54) and other causes of cardiac injury, such as cardiac, renal, and respiratory failure (Chapter 96); stroke (Chapter 379) and intracranial hemorrhage (Chapter 380); septic shock (Chapter 100); and chronic structural heart diseases (Chapter 62) (Table 64-3). As a result, the clinician always must consider the clinical context as well as the finding of a temporal rise and fall of troponin levels.8 The sensitivity and specificity of cardiac-specific TnI and TnT make them the “gold standard” for detection of myocardial necrosis. However, the

392

CHAPTER 64  ST ELEVATION ACUTE MYOCARDIAL INFARCTION  

decision to proceed with urgent reperfusion (primary angioplasty or fibrinolysis) in ST elevation MI should be based on the patient’s clinical history and the initial ECG and not be delayed by troponin testing (Chapter 45). Clinically, cTnI and cTnT are of approximately equivalent utility, except that renal failure (Chapter 122) is more likely to be associated with falsepositive elevations of cTnT than of cTnI. The troponins are maximally sensitive at 8 to 12 hours, peak at 10 to 24 hours, and persist for 5 to 14 days. However, this persistence makes more challenging the diagnosis of an early recurrent MI, for which more rapidly cleared markers (i.e., the MB isoenzyme of creatine kinase [CK-MB]) may be selectively complementary. Otherwise, however, with current troponin assays, concomitant measurement of levels of CK-MB or myoglobin is redundant and not recommended.

Other Laboratory Tests

On admission, routine assessment of complete blood count and platelet count, standard blood chemistry studies, a lipid panel, and coagulation tests (prothrombin time, partial thromboplastin time) is useful. Results assist in assessing comorbid conditions and prognosis and in guiding therapy. Hematologic tests provide a useful baseline before initiation of antiplatelet, anticoagulant, and fibrinolytic therapy or coronary angiography or angioplasty. Myocardial injury precipitates polymorphonuclear leukocytosis, commonly resulting in an elevation of the white blood cell count to up to 12,000 to 15,000/µL, which appears within a few hours and peaks at 2 to 4 days. The metabolic panel provides a useful check on electrolytes, glucose, and renal function. The lactate dehydrogenase level and aspartate aminotransferase level also may be elevated, particularly in the setting of large infarctions, but should not be ordered for diagnostic purposes. On hospital admission or the next morning, a fasting lipid panel is recommended as a baseline for lipid-lowering (statin) therapy (Chapter 195). Unless carbon dioxide retention is suspected, finger oximetry is adequate to diagnose hypoxemia and titrate oxygen therapy. The level of C-reactive protein increases with acute MI, but its incremental prognostic value in the acute setting has not been established. B-type natriuretic peptide, which increases with ventricular wall stress and relative circulatory fluid overload, may provide useful incremental prognostic information in the setting of acute MI.

Imaging

The rapid triage of ST elevation MI patients to reperfusion therapy, most often for coronary angiography with intent to perform PCI, should not be delayed by routine imaging studies. A chest radiograph is the only imaging test routinely considered in the emergency department for ST elevation MI, and its usefulness is limited to selected patients with a concern for findings that might impact initial management. Although the chest radiograph is often normal, findings of pulmonary venous congestion, cardiomegaly, or a widened mediastinum can contribute importantly to diagnosis and management decisions. For example, a history of severe, “tearing” chest and back pain in association with a widened mediastinum should raise the question of a dissecting aortic aneurysm (Chapter 69). In such cases, fibrinolytic therapy must be withheld pending more definitive diagnostic imaging of the aorta, and the approach to angiography may be modified. Other noninvasive imaging (e.g., echocardiography [Chapter 49], cardiac nuclear scanning [Chapter 50], and other testing) may be performed during the course of hospitalization to evaluate specific clinical issues, including suspected complications of acute MI, or for predischarge risk stratification. Coronary angiography (Chapter 51) is performed urgently as part of an interventional strategy for ST elevation MI or later for risk stratification in higher-risk patients who are initially managed medically.

Echocardiography

Two-dimensional transthoracic echocardiography with color flow Doppler imaging is the most generally useful noninvasive test obtained during the hospital course (Chapter 49). Echocardiography efficiently assesses global and regional cardiac function and enables the clinician to evaluate suspected complications of acute MI. The sensitivity and specificity of echocardiography for regional wall motion assessment are high (>90%), although the age of the abnormality (new vs. old) must be distinguished clinically or by ECG. Echocardiography is helpful in determining the cause of circulatory failure with hypotension (relative hypovolemia, LV failure, RV failure, or mechanical complication of acute MI). Echocardiography also can assist in differentiating pericarditis and perimyocarditis from acute MI. Doppler echocardiography is indicated to evaluate a new murmur and other suspected mechanical complications of acute MI (e.g., papillary muscle dysfunction or rupture, acute

ventricular septal defect, and LV free wall rupture with tamponade or pseudoaneurysm). Later in the course of acute MI, echocardiography may be used to assess the degree of recovery of stunned myocardium after reperfusion therapy, the degree of residual cardiac dysfunction and indications for angiotensin-converting enzyme (ACE) inhibitors and other therapies for heart failure, and the presence of LV aneurysm and mural thrombus (requiring oral anticoagulants).

Radionuclide, Magnetic Resonance, and Other Imaging Studies

Radionuclide techniques generally are too time-consuming and cumbersome for routine use in the acute setting of definite or probable acute MI. More commonly, they are used in risk stratification before or after hospital discharge to augment exercise or pharmacologic stress testing (Chapter 50). Thallium (Tl 201) or technetium (Tc 99m) sestamibi nuclear scans or rubidium (Rb 82) positron emission tomography scans can assess myocardial perfusion and viability as well as infarct size. Cardiac magnetic resonance imaging (Chapter 50) with late gadolinium enhancement also can assess infarct size as well as myocardial function during the convalescent phase. Computed tomography and magnetic resonance imaging also can be useful to evaluate patients with a suspected dissecting aortic aneurysm (Chapter 69). When a nonatherosclerotic cause of myocardial necrosis is suspected (e.g., perimyocarditis simulating acute MI), contemporary multislice (e.g., 64- to 256-slice) coronary computed tomography can assess coronary artery disease qualitatively and semiquantitatively as well as distinguish other causes of chest pain syndromes (Chapters 45 and 50).

TREATMENT  Prehospital Phase

Prehospital cardiac arrest (Chapter 57) and extensive necrosis are major causes of acute MI-associated morbidity and mortality. The first hour after the onset of symptoms represents the best opportunity for myocardial salvage with reperfusion therapy. Thus, the primary goals of prehospital care are to recognize symptoms promptly and seek medical attention, deploy an emergency medical system (EMS) trained in emergency cardiac care, and transport the patient expeditiously to a medical facility capable of advanced coronary care, including reperfusion therapy (primary PCI or fibrinolysis).9 Guidelines set an average system time goal for first medical contact-to-device of 90 minutes or less. For patients who initially arrive at or are transported to a non–PCI-capable hospital, the system time goal to device is 120 minutes or less. The greatest time lag to reperfusion therapy is the patient’s delay in calling for help. Public education efforts aimed at reducing this delay have yielded mixed results, and innovative approaches are needed. Expert EMS teams perform prehospital ECGs, communicate preliminary diagnoses, and transport patients preferentially to a PCI-capable hospital where an ST elevation MI team is on alert. This approach results in more rapid primary PCI (average savings of about 15 minutes) and superior clinical outcomes. Direct triage to the catheterization laboratory rather than to the emergency department also may decrease time to reperfusion. However, as many as one third of initial activations of the ST elevation MI team may be false positive. Aspirin (162 to 325 mg, chewed) and sublingual nitroglycerin (0.4 mg every 5 minutes for up to three doses) are administered at first medical contact where appropriate. Administering fibrinolytic therapy in the field by physician-directed EMS systems is feasible when the expected transit time is long (i.e., >120 minutes), but this practice is rare in the United States. Prehospital antiplatelet therapy of ST elevation MI patients with ticagrelor does not improve pre-PCI coronary reperfusion.

Hospital Phases

Emergency Department

The goals of emergency department care are to identify patients with acute myocardial ischemia rapidly, to stratify them into acute ST elevation MI versus other acute coronary syndromes (Fig. 64-3), to initiate a reperfusion strategy and other appropriate medical care, to assess risk (E-Fig. 64-1), and to triage them rapidly to inpatient care (patients with suspected acute coronary syndrome (ACS)) or outpatient care (patients without suspected ischemia) (see Fig. 63-2). The evaluation of patients with chest pain and other suspected acute coronary syndromes begins with a 12-lead ECG (with a goal of within 10 minutes of hospital arrival) and continues with a focused history and a targeted physical examination. Continuous ECG monitoring is started, an intravenous (IV) line is established, and admission blood tests are drawn (including for cardiac biomarkers, i.e., cTnI or cTnT). As rapidly as possible, the patient is stratified into acute ST elevation MI, likely or definite non–ST elevation ACS, possible non–ST elevation ACS, or noncardiac chest pain categories.

392.e1

CHAPTER 64  ST ELEVATION ACUTE MYOCARDIAL INFARCTION  

Risk calculator for 6-month postdischarge mortality after hospitalization for acute coronary syndrome Record the points for each variable at the bottom left and sum the points to calculate the total risk score. Find the total score on the x-axis of the nomogram plot. The corresponding probability on the y-axis is the estimated probability of all-cause mortality from hospital discharge to 6 months. Medical history 1 Age in years

Findings at initial hospital presentation Points

≤29 30–39 40–49 50–59 60–69 70–79 80–89 ≥90

0 0 18 36 55 73 91 100

2 History of congestive heart failure 3 History of myocardial infarction

4 Resting heart rate, beats/min ≤49.9 50 –69.9 70 –89.9 90 –109.9 110 –149.9 150 –199.9 ≥200

0 3 9 14 23 35 43

24 22 18 14 10 4 0

6 ST-Segment depression

11

7 Initial serum creatinine, mg/dL 0 –0.39 0 .4 –0.79 0 .8 –1.19 1.2 –1.59 1.6 –1.99 2 –3.99 ≥4

Points 1 3 5 7 9 15 20

8 Elevated cardiac enzymes

15

9 No in-hospital percutaneous coronary intervention

14

Predicted all-cause mortality from hospital discharge to 6 months

Points

0.50 0.45

2

0.40

3

0.35 Probability

4 5 6

0.30 0.25 0.20

7

0.15

8

0.10

9

0.05

Mortality risk

Points

5 Systolic blood pressure, mm Hg ≤79.9 80 –99.9 100 –199.9 120 –139.9 140 –159.9 160 –199.9 ≥200

1

Total risk score

Findings during hospitalization

(Sum of points) (From plot)

0 70

90

110

130 150 170 Total risk score

190

210

E-FIGURE 64-1.  Global Registry of Acute Coronary Events (GRACE) risk score calculator for all-cause mortality from discharge after acute coronary syndrome to 6 months. (Reprinted with permission from Eagle KA, Lim MJ, Dabbous OH, et al. A validated prediction model for all forms of acute coronary syndrome: estimating the risk of 6-month postdischarge death in an international registry. JAMA. 2004;291:2727-2733.)

CHAPTER 64  ST ELEVATION ACUTE MYOCARDIAL INFARCTION  

393

Reperfusion therapy for patients with STEMI STEMI patient who is a candidate for reperfusion Initially seen at a non-PCI-capable hospital*

Initially seen at a PCI-capable hospital

DIDO time ≤ 30 min

Transfer for primary PCI within ≤ 120 min of first medical contact

Send to cath lab for primary PCI within ≤ 90 min of first medical contact

Administer fibrinolytic agent within 30 min of arrival if time to PCI is anticipated to be >120 min after first medical contact

Diagnostic angiogram

Medical therapy only

PCI

Urgent transfer for PCI for patients with evidence of failed reperfusion or reocclusion

CABG

Transfer for angiography and revascularization within 3–24 h for other patients as part of an invasive strategy†

FIGURE 64-3.  Reperfusion therapy for patients with ST segment elevation myocardial infarction (STEMI). All eligible patients with STEMI also are treated with dual antiplatelet therapy and an anticoagulant on admission (see text). *Patients with cardiogenic shock or severe heart failure initially seen at a non–percutaneous coronary intervention (PCI)-capable hospital should be transferred for cardiac catheterization (cath lab) and revascularization as soon as possible, irrespective of time delay from MI onset (class I, level of evidence: B). † Angiography and revascularization should not be performed within the first 2 to 3 hours after administration of fibrinolytic therapy. CABG = coronary artery bypass grafting; DIDO = door in–door out. (Modified from O’Gara PT, Kushner FG, Ascheim DD, et al. 2013 ACCF/AHA guideline for the management of ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2013;127:e362-e425.)

Emergent Therapeutic Measures and Early Inpatient Care Reperfusion Therapy

Coronary reperfusion is accomplished by primary PCI (angioplasty and stenting) or by IV fibrinolytic therapy.10 Primary PCI: Prompt PCI (with a goal of 12 hours if symptoms persist) Cardiogenic shock developing within 36 hours of ST segment elevation/Q wave acute MI or LBBB acute MI in patients 400 PCIs/year with backup cardiac surgery and for operators performing >75 PCIs/year Primary PCI can be performed at centers without on-site cardiac surgery if performed on carefully selected patients by experienced operators and if arrangements are in place for rapid transfer to a surgery-capable center when needed ADVANTAGES OF PRIMARY PCI Higher initial reperfusion rates Reduced risk of intracerebral hemorrhage Less residual stenosis; less recurrent ischemia or infarction Usefulness when fibrinolysis is contraindicated Improvement in outcomes with cardiogenic shock DISADVANTAGES OF PRIMARY PCI (COMPARED WITH FIBRINOLYTIC THERAPY) Access, advantages restricted to high-volume centers and operators Longer average time to treatment Greater dependence on operators for results Higher system complexity and costs LBBB = left bundle branch block; MI = myocardial infarction; PCI = percutaneous coronary intervention (includes balloon angioplasty, stenting).

Longer-acting variants of tissue-type plasminogen activator (t-PA), given as single-bolus (tenecteplase) or double-bolus (reteplase) injections, now are more commonly used than t-PA (alteplase) in clinical practice because they are more convenient to administer, but they do not further improve survival. A8 

394

CHAPTER 64  ST ELEVATION ACUTE MYOCARDIAL INFARCTION  

TABLE 64-5 CHARACTERISTICS OF INTRAVENOUS FIBRINOLYTIC AGENTS APPROVED BY THE U.S. FOOD AND DRUG ADMINISTRATION CHARACTERISTICS

STREPTOKINASE (SK)

ALTEPLASE (t-PA)

RETEPLASE (r-PA)

TENECTEPLASE (TNK–t-PA)

Dose

1.5 MU in 30-60 minutes

100 mg in 90 minutes*

10 U + 10 U, 30 minutes apart

30-50 mg† during 5 seconds

Circulating half-life (minutes)

≅20

≅4

≅18

≅20

Antigenic/allergic reactions

Yes

No

No

No

Systemic fibrinogen depletion

Severe

Mild to moderate

Moderate

Minimal

Intracerebral hemorrhage

≅0.4%

≅0.7%

≅0.8%

≅0.7%

Patency (TIMI 2/3) rate, 90 minutes‡

≅51%

≅73-84%

≅83%

≅77-88%

Lives saved per 100 treated

≅3§

≅4‖

≅4

≅4

Cost per dose (approximate U.S. dollars)

300

1800

2200

2200

*An accelerated alteplase regimen is given as follows: 15-mg bolus, then 0.75 mg/kg during 30 minutes (maximum, 50 mg), then 0.50 mg/kg during 60 minutes (maximum, 35 mg). † TNK–t-PA is dosed by weight (supplied in 5-mg/mL vials): 90 kg = 10 mL. ‡ TIMI = Thrombolysis in Myocardial Infarction. Data from Granger CB, Califf RM, Topol EJ. Thrombolytic therapy for acute myocardial infarction: a review. Drugs. 1992;44:293-325; and Bode C, Smalling RW, Berg G, et al. Randomized comparison of coronary thrombolysis achieved with double-bolus reteplase (recombinant plasminogen activator) and front-loaded, accelerated alteplase (recombinant tissue plasminogen activator) in patients with acute myocardial infarction: the RAPID II Investigators. Circulation. 1996;94:891-898. § Patients with ST segment elevation or bundle branch block, treated 70 to 75 years), female gender, hypertension, and higher relative doses of plasminogen activators and heparin increase the risk of intracranial hemorrhage. For failed fibrinolysis, urgent transfer to a PCI-capable hospital for rescue PCI is more effective than repeated fibrinolysis. Even for patients successfully reperfused, rapid transfer to a PCI-capable facility, where angiography and PCI are performed as early as feasible within the window of 3 to 24 hours after fibrinolysis, can reduce the risk for recurrent ischemia, reinfarction, heart failure, cardiogenic shock, or death by about 35%.

Ancillary and Other Therapies General Medical Management

Beyond prompt initiation of reperfusion and other antithrombotic therapy, initial management includes bedrest (i.e., for 12 hours or until ischemia has been relieved and hemodynamic parameters have stabilized) with ECG monitoring. Routine oxygen supplementation does not benefit normoxemic patients. A9  A10  Oxygen should be used in doses just sufficient to avoid hypoxemia (e.g., initially at 2 to 4 L/minute by nasal cannula; fingertip oximetry may be used to monitor effect) in patients with hypoxemia (oxygen saturation 2-3) Old ischemic stroke (>3 months ago); intracerebral disease other than above Recent (10 minutes) cardiopulmonary resuscitation or internal bleeding Active peptic ulcer Recent noncompressible vascular punctures Pregnancy For streptokinase/anistreplase: prior exposure (especially if >5 days ago) or allergic reaction Modified from Kushner FG, Hand M, Smith SC Jr, et al. 2009 Focused updates: ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction and the ACC/ AHA/SCAI guidelines on percutaneous coronary intervention. Circulation. 2009;120:2271-2306.

candidates for the addition of an aldosterone antagonist (i.e., eplerenone; initially, 25 mg once daily) or spironolactone (12.5 to 25 mg twice daily). The optimal systolic blood pressure range during acute MI is generally 100 to 140 mm Hg. Excessive hypertension usually responds to titrated nitroglycerin, β-blocker therapy, and morphine. Relative hypotension could require discontinuation of these medications, fluid administration, and/or other measures as appropriate to the hemodynamic subset (Table 64-7). Atropine (0.5 to 1.5 mg intravenously) should be available to treat symptomatic bradycardia and hypotension related to hypervagotonia. Antiplatelet Therapy

Given the critical role of coronary thrombosis in the precipitation of acute MI, antithrombotic therapy, supplemental to primary PCI or fibrinolysis, plays a key role in management of ST elevation MI (Fig. 64-4 and see Table 64-7). Aspirin (162 to 325 mg, non–enteric-coated) should be given on presentation to all patients unless it is contraindicated (see Fig. 64-4). A daily maintenance

395

CHAPTER 64  ST ELEVATION ACUTE MYOCARDIAL INFARCTION  

TABLE 64-7 IN-HOSPITAL DOSING OF SELECTED ANTITHROMBOTIC DRUGS DRUG

INITIAL DOSE

Aspirin

MAINTENANCE DOSE

Abciximab

0.25 mg/kg IV bolus

0.125  µg/kg/min [max, 10 µg/min] for up to 12 hr

Aspirin

162-325 mg

81 mg*-325 mg/day

Clopidogrel

PCI: 300-600 mg* Fibrinolysis and age >75: 75-150 mg

75 mg/day (optional: 150 mg/day × 1 wk)

Eptifibatide

180  µg/kg IV bolus × 2†

2.0  µg/kg/min for up to 18 hr; reduce rate by 50% for CrCl 6 wk plus gentamicin 1 mg/kg IV q8h × 2 wk plus rifampin 300 mg PO/IV q8h × ≥6 wk

4. Prosthetic valve infection with methicillin-susceptible strain; use vancomycin instead of nafcillin for MRSA

5. Cefazolin 2 g IV q8h × 4-6 wk

5. PCN allergy other than immediate hypersensitivity

6. Daptomycin 6 mg/kg IV qd × 14-42 days

6. Daptomycin is FDA-approved for treatment of right-sided S. aureus infective endocarditis; for adults, some experts recommend 8-10 mg/kg IV

COAGULASE-NEGATIVE STAPHYLOCOCCI, PROSTHETIC VALVE INFECTION Vancomycin 15-20 mg/kg IV q8-12h × >6 wk plus gentamicin 1 mg/kg IV q8h × 2 wk plus rifampin 300 mg PO/IV q8h × >6 wk

Can substitute nafcillin in above doses for vancomycin if isolate is methicillin sensitive

HACEK STRAINS 1. Ceftriaxone 2 g IV qd × 4 wk; 6 wk for prosthetic valves



2. Ampicillin–sulbactam 3 g IV q6h × 4 wk; 6 wk for prosthetic valves

2. HACEK strains increasingly may produce β-lactamase

NON-HACEK GRAM-NEGATIVE BACILLI Enterobacteriaceae Extended-spectrum PCN or cephalosporin plus aminoglycosides for susceptible strains

Treat for a minimum of 6-8 wk; some species exhibit inducible resistance to thirdgeneration cephalosporins; valve surgery is required for most patients with left-sided endocarditis caused by gram-negative bacilli; consultation with a specialist in infectious diseases is recommended

Pseudomonas aeruginosa High-dose tobramycin (8 mg/kg/day IV or IM in once-daily doses) with maintenance Treat for a minimum of 6-8 wk; early valve surgery usually required for left-sided of peak and trough concentrations of 15 to 20 µg/mL and ≤2  µg/mL, respectively, in Pseudomonas endocarditis; consultation with a specialist in infectious diseases is combination with an extended-spectrum PCN (e.g., ticarcillin, piperacillin, recommended azlocillin); ceftazidime, cefepime, or imipenem in full doses; or imipenem Fungi Treatment with a parenteral antifungal agent (usually a lipid-containing amphotericin B product, 3-5 mg/kg/day IV for at least 6 weeks) and valve replacement; Fluconazole, 400 mg daily PO is an alternative for susceptible yeasts; other azoles, such as voriconazole, may be required for resistant yeasts or molds.

Long-term or lifelong suppressive therapy with PO antifungal agents often required; consultation with a specialist in infectious diseases is recommended

*Dosages are for patients with normal renal function; for those with renal insufficiency, adjustments must be made for all drugs except nafcillin, rifampin, and ceftriaxone. Gentamicin doses should be adjusted to achieve a peak serum concentration of approximately 3 µg/mL 30 min after dosing and a trough gentamicin level of 1 week after the institution of appropriate antibiotics) should prompt repeat blood cultures. If such cultures are negative, several possibilities should be considered: myocardial abscess, extracardiac infection (e.g., mycotic aneurysm, psoas or splenic abscess, vertebral osteomyelitis, septic arthritis), immune complex–mediated tissue damage, or a complication of hospitalization and therapy (e.g., drug fever, nosocomial superinfection, pulmonary embolism). Appropriate studies might include TEE, CT scan of the abdomen, bone scan, and urinalysis with microscopy (to elicit evidence of interstitial nephritis). IV line sites should be carefully examined for evidence of infection, and indwelling central lines should be changed according to published guidelines. Anticoagulation in individuals with infective endocarditis is controversial. Although new anticoagulation in the setting of native valve endocarditis does not appear to provide a benefit, continuing ongoing anticoagulation may be advisable. Some authorities recommend continuing anticoagulation in patients with mechanical prosthetic valve endocarditis. However, discontinuation of all anticoagulation for at least the first 2 weeks of antibiotic therapy is generally advised in patients with S. aureus prosthetic valve endocarditis who have experienced a recent CNS embolic event; this approach allows the thrombus to organize and potentially prevents the acute hemorrhagic transformation of embolic lesions. Reintroduction of anticoagulation in these patients must be cautious, and the international normalized ratio must be monitored carefully. The best option for patients with other indications for anticoagulation, such as deep vein thrombosis, major vessel embolization, or atrial fibrillation, is less clear and should be decided in a multidisciplinary fashion that balances the risks and benefits for each individual patient. High-dose (325 mg/day) aspirin does not prevent embolic events and tends to increase the incidence of bleeding in patients with infective endocarditis.

426

CHAPTER 67  Infective Endocarditis  

Whether a patient should remain on chronic, low-dose (81 mg) aspirin if they develop subsequent infective endocarditis is uncertain.

TABLE 67-8 INDICATIONS FOR SURGERY IN ENDOCARDITIS

Complications

INDICATION

The complications of infective endocarditis can be divided into four groups for ease of classification: direct valvular damage and consequences of local invasion, embolic complications, metastatic infections from bacteremia, and immunologic phenomena. Local damage to the endocardium or myocardium may directly erode through the involved cardiac valve or adjacent myocardial wall, resulting in hemodynamically significant valvular perforations or intra- or extracardiac fistulae. Such local complications typically present clinically with the acute onset of heart failure and carry a poor prognosis, even with prompt cardiac surgery. Valve ring abscesses also require surgical intervention and are more frequent in patients with prosthetic valves. Although a conduction defect on ECG may suggest the diagnosis, TEE is the diagnostic technique of choice for detecting paravalvular abscess, valve perforation, or intracardiac fistulae. Frank myocardial abscesses are found in up to 20% of cases on autopsy, and Aspergillus endocarditis invades the myocardium in more than 50% of cases. Pericarditis is rare and is usually associated with myocardial abscess. Myocardial infarction (MI), thought to be caused by embolism of vegetative material into the coronary arteries, is seen in 40 to 60% of cases on autopsy, although most cases are clinically silent and lack characteristic ECG changes. However, up to 15% of elderly patients may present with clinical evidence of acute MI, with potentially disastrous complications if the MI is thought to be the primary event and the patient is given thrombolytic therapy. Heart failure is the leading cause of death in infective endocarditis, usually related to direct valvular damage. Embolic events are less common now than in the preantibiotic era, but about 35% of patients have at least one clinically evident embolic event. In fungal endocarditis, the majority of patients have at least one embolic event, frequently with a large embolus. The presence of large (>10 mm), mobile vegetations on the echocardiogram, particularly when the anterior mitral valve leaflet is involved, predicts a high risk of embolic complications. In addition, patients may have frank infarction of cutaneous tissue from emboli. In addition to the skin, systemic emboli most commonly lodge in the kidneys, spleen, large blood vessels, or CNS. Vegetations of right-sided endocarditis usually embolize to the lungs and cause abnormalities on the chest radiograph, although occasionally such emboli reach the left-sided circulation via a patent foramen ovale. Renal abscesses are rare in infective endocarditis; however, bland renal infarction is a frequent asymptomatic finding on abdominal CT scanning, seen in more than 50% of cases at autopsy. Similarly, splenic infarction occurs in up to 44% of autopsy-confirmed cases. Such emboli may be asymptomatic but also can cause left upper quadrant pain radiating to the left shoulder, sometimes as the presenting symptom of infective endocarditis. A splenic infarction that progresses to form an abscess can cause persistent fever or bacteremia, so such patients should undergo abdominal CT to search for this complication. Mycotic vascular aneurysms, which frequently occur at bifurcation points, may be clinically silent until they rupture (which may be months to years after apparently successful antibiotic treatment of infective endocarditis) and have been found in 10 to 15% of cases at autopsy. Whereas peripheral mycotic aneurysms require surgical resection, intracerebral aneurysms can be resected or managed with intravascular techniques (e.g., coils) if they bleed or if they are causing a mass effect. For mycotic aneurysms of the abdominal aorta, endovascular repair may be preferable; but if endovascular therapy is used, long-term antibiotics are generally required. Many patients may have evidence of cerebrovascular emboli, which have a predilection for the middle cerebral artery distribution and may be devastating. Most emboli to the CNS occur early in the course of the disease and are evident at the time of presentation or shortly thereafter. Embolic strokes may undergo hemorrhagic transformation, with a sudden worsening of the patient’s neurologic status. Many patients with fungal endocarditis present with an embolic stroke or large emboli that occlude major vessels. Some complications of infective endocarditis result when bacteremic seeding causes metastatic infection at a distant site. Patients may present with or develop osteomyelitis, septic arthritis, or epidural abscess. Purulent meningitis (Chapter 384) is a rare complication except in pneumococcal endocarditis, although many patients with S. aureus infective endocarditis who undergo lumbar puncture have a pleocytosis. Importantly, the finding of one metastatic complication of infective endocarditis does not exclude the possibility of additional sites of hematogenous infection, particularly in S. aureus endocarditis. Thus, the need for additional diagnostic evaluations should be guided by the patient’s clinical course. The immunologic phenomena of infective endocarditis are often directly related to high levels of circulating immune complexes. Renal biopsy results nearly always are abnormal in the setting of active infective endocarditis, which classically causes a hypocomplementemic glomerulonephritis (Chapter 113). Histopathologically, the glomerular changes may be focal, diffuse, or membranoproliferative, or they may be akin to the immune complex disease found in systemic lupus erythematosus. In addition, many of the musculoskeletal conditions associated with infective endocarditis, including monoarticular and oligoarticular arthritides, are probably immune mediated. These immunologic phenomena usually abate with successful antimicrobial therapy.

CLASS*

NATIVE VALVE ENDOCARDITIS Acute aortic insufficiency or mitral regurgitation with heart failure

I

Acute aortic insufficiency with tachycardia and early closure of the mitral valve on echocardiogram

I

Fungal endocarditis

I

Evidence of annular or aortic abscess, sinus or aortic true or false aneurysm, valvular dehiscence, rupture, perforation, or fistula

I

Evidence of valve dysfunction and persistent infection after a prolonged period (7-10 days) of appropriate therapy, provided there are no noncardiac causes of infection

I

Recurrent emboli after appropriate antibiotic therapy

I

Infection with enteric gram-negative organisms or organisms with a poor response to antibiotics in patients with evidence of valve dysfunction

I

Anterior mitral leaflet vegetation (especially with size >10 mm) or persistent vegetation after systemic embolization

IIa

Increase in vegetation size despite appropriate antimicrobial therapy

IIb

Early infections of the mitral valve that can probably be repaired, especially in the presence of large vegetations and/or recurrent emboli

III

Persistent fever and leukocytosis with negative blood cultures

III

PROSTHETIC VALVE ENDOCARDITIS Early prosthetic valve endocarditis (10 mm), early surgery did not significantly reduce all-cause mortality at 6 months but markedly decreased the risk of systemic embolism, including stroke and MI. A2  Surgical management should also be considered for patients with recurrent (two or more) embolic events or those with large vegetations (>10 mm) on echocardiography and one embolic event, although the data in these situations are less convincing. The presence of S. aureus endocarditis involving the anterior mitral valve leaflet and large vegetations (>10 mm) may be a special circumstance calling for early surgical intervention to reduce the high risk of CNS emboli, especially when mitral valve repair, rather than valve replacement, can be accomplished. Unfortunately, only about 15 to 20% of these latter patients end up being good candidates for valve repair. Delaying surgery in patients with deteriorating cardiac function in an attempt to sterilize the affected valve is ill advised because the risk of progressive heart failure or further complications usually outweighs the relatively small risk of recurrent infective endocarditis after prosthetic valve implantation. Relative

CHAPTER 67  Infective Endocarditis  

contraindications to valve replacement include recent large CNS emboli (>2 cm) or bleed (because of the risk of bleeding in the perioperative period, when systemic anticoagulation is required), multiple prior valve replacements (because of the difficulty of sewing a new valve into tissue already weakened from previous surgeries), and ongoing injection drug use. On occasion, patients have both a compelling indication for valve replacement (e.g., acute heart failure) and a recent CNS embolic event. The risk of hemorrhagic transformation of such lesions during cardiac bypass–associated anticoagulation is controversial. However, it appears that the greatest risk of such transformation events is in larger (>2 cm) emboli, especially those that have exhibited a hemorrhagic component. In these latter scenarios, it is prudent to try to delay surgery for at least 2 to 4 weeks to allow organization and resolution of such emboli. However, there appears to be no survival benefit in delaying indicated valve replacement surgery (>7 days) after an ischemic stroke. After definitive surgical treatment, most patients should receive further antibiotic therapy unless a full course of antibiotics was administered before surgery and there is no evidence of ongoing infection. If the patient received antibiotics for less than 1 week before surgery or the culture from the operative site is positive, the patient should receive the equivalent of a full initial course of antibiotics appropriate for the organism. If the patient received antibiotics for 2 weeks or more and the culture result from the operative site is negative (regardless of whether valve histopathology shows inflammation or a positive Gram stain result), the patient should receive whatever remains of the originally planned course of appropriate antibiotic therapy. In patients with infective endocarditis related to implanted cardiovascular devices, complete device removal is mandatory, regardless of the pathogen, if the goal is to cure the infection. In patients who truly cannot tolerate device removal, chronic antibiotic therapy is the best alternative.12 If a replacement device needs to be implanted, the optimal timing for such a procedure is unclear. However, blood culture results should be negative, and any concomitant local or pocket site infection should be completely resolved. The duration of antimicrobial therapy after device extraction depends on the device and the infection.13 For lead-related infective endocarditis, which is usually associated with bloodstream infection, 2 weeks of therapy is recommended if there are no infection complications. For infection caused by S. aureus, therapy should be extended for up to 4 weeks. Nearly 40% of patients with infective endocarditis related to implantable cardiovascular devices have concomitant valve involvement, predominantly tricuspid valve infection, with in-hospital and 1-year mortality rates of 15% and 23%, respectively. Device removal appears to reduce the mortality rate by about 50% (from about 40% to about 20%). In such patients, concomitant 4 to 6 weeks of therapy is recommended.



PREVENTION

Despite a lack of definitive data for dental procedures, prophylactic antibiotics are recommended to prevent infective endocarditis (Table 67-9) when patients with the highest risk of adverse outcomes from endocarditis undergo dental procedures that involve manipulation of gingival tissue or the periapical region of teeth or perforation of the oral mucosa; an invasive procedure of the respiratory tract, with incision or biopsy of the respiratory mucosa, such as tonsillectomy and adenoidectomy; or invasive procedures involving infected skin, skin structures, or musculoskeletal tissue (Table 67-10). Other consensus guidelines have also narrowed the indications for antimicrobial prophylaxis. In the United Kingdom, for example, no prophylaxis is advised for any dental patient, regardless of underlying cardiac valvular conditions. In contrast, guidelines from the European Society of Cardiology are largely consistent with current AHA guidelines. Since the recent publication of these more limited recommendations from the AHA, ESC, and NICE, follow-up surveys in these countries have generated mixed results without definitive evidence of causeand-effect between reduction in dental prophylaxis use and increased endocarditis incidence due to viridans group streptococci. In the United States, however, some data suggest that streptococcal endocarditis rates have increased since the adoption of less aggressive prophylaxis strategies in 2007.14 The antibiotics chosen for preprocedure prophylaxis should be active against the organisms most likely to be released into the blood stream by the procedure of interest (Table 67-11). Data suggest that amoxicillin/clavulanate (1000 mg/200 mg) is the best therapy for preventing bacteremia, that amoxicillin alone is better than azithromycin, and that clindamycin is no better than placebo at preventing bacteremia A3  after dental procedures. Whether these data translate into preventing endocarditis, however, is unknown.15 Thus, antibiotics that cover primarily oral flora are recommended for dental and upper respiratory procedures. For patients with the conditions listed in Table 67-9 who undergo a procedure for infected skin, skin structure, or musculoskeletal tissue,

427

TABLE 67-9 HIGH-RISK CARDIAC CONDITIONS FOR WHICH ENDOCARDITIS PROPHYLAXIS WITH DENTAL PROCEDURES IS REASONABLE Prosthetic cardiac valve or prosthetic material used for cardiac valve repair Previous endocarditis Complex congenital heart disease involving unrepaired cyanotic congenital heart disease (including palliative shunts and conduits), completely repaired congenital heart disease with prosthetic material within 6 mo of the procedure, or repaired congenital heart disease with residual defects at the site or adjacent to the site of prosthetic material Cardiac transplantation recipients who develop cardiac valvulopathy Adapted from Wilson W, Taubert KA, Gewitz M, et al. Prevention of infective endocarditis guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation. 2007;116:1736-1754.

TABLE 67-10 RECOMMENDATIONS FOR ENDOCARDITIS PROPHYLAXIS PROPHYLAXIS IS RECOMMENDED* Dental: all dental procedures involving manipulation of gingival tissue or the periapical region of teeth or perforation of the oral mucosa Respiratory: procedures involving incision or biopsy of the respiratory mucosa, such as tonsillectomy and adenoidectomy Other: procedures involving infected skin, skin structures, or musculoskeletal tissue prior to incision and drainage PROPHYLAXIS IS NOT RECOMMENDED Dental: routine anesthetic injections through noninfected tissue, dental radiographs, placement of removable prosthodontic or orthodontic appliances, adjustment of orthodontic appliances, placement of orthodontic brackets, shedding of deciduous teeth, bleeding from trauma to the lips or oral mucosa Respiratory: procedures not involving incision or biopsy of the respiratory mucosa, including bronchoscopy (unless the procedure involves incision of the respiratory tract mucosa) Genitourinary: antibiotic prophylaxis solely to prevent infective endocarditis is not recommended Gastrointestinal: antibiotic prophylaxis solely to prevent infective endocarditis is not recommended *Only in patients with underlying cardiac conditions associated with the highest risk for adverse outcome from endocarditis (listed in Table 67-9). Adapted from Wilson W, Taubert KA, Gewitz M, et al. Prevention of infective endocarditis guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation. 2007;116:1736-1754.

the therapeutic regimen should contain an agent active against staphylococci and β-hemolytic streptococci. Patients with implanted cardiac devices do not require antibiotic prophylaxis for dental or other invasive procedures. However, such patients require surgical site prophylaxis at the time of device placement. The recommended regimens generally include a β-lactam (commonly cefazolin, 1 g IV 1 hour before device placement), regardless of whether a new device is being placed or a device is being revised. More aggressive prophylaxis with vancomycin, intraprocedural bacitracin pocket wash, and 2 days of postprocedural oral cephalexin does not significantly reduce infection rates. A4   

PROGNOSIS

Untreated infective endocarditis is uniformly fatal. Aggressive medical and surgical management dramatically improves the outcome. The overall mortality rate from both native and prosthetic valve endocarditis remains fairly high, ranging from 17 to 36%. Whereas certain subgroups, such as patients with viridans group streptococcal endocarditis, have a lower risk of death, patients with S. aureus, fungal, and zoonotic endocarditis have higher mortality rates. Heart failure and CNS events are the most frequent causes of death. Endocarditis recurs in about 12 to 16% of patients and is more common in injection drug users, elderly people, and patients with prosthetic valves.

TABLE 67-11 SUGGESTED ANTIBIOTICS FOR ENDOCARDITIS PROPHYLAXIS FOR DENTAL OR RESPIRATORY TRACT PROCEDURES* IN PATIENTS WITH HIGH-RISK CARDIAC CONDITIONS† PATIENT CHARACTERISTICS

REGIMEN‡

Able to take oral medications

Amoxicillin 2 g PO

Unable to take oral medications

Ampicillin 2 g IV or IM; or cefazolin or ceftriaxone 1 g IM or IV

Allergic to penicillin or ampicillin and able to take oral medications

Cephalexin 2 g PO (or other first- or second-generation oral cephalosporin in equivalent adult doses); clindamycin 600 mg PO; azithromycin 500 mg PO; or clarithromycin 500 mg PO Cephalosporins should not be used in an individual with a history of anaphylaxis, angioedema, or urticaria with penicillin or ampicillin

Allergic to penicillin or ampicillin and unable to take oral medications

Cefazolin or ceftriaxone 1 g IM or IV; or clindamycin 600 mg IM or IV

*For the applicable procedures, see Table 67-10. † For the applicable conditions, see Table 67-9. ‡ All regimens consist of a single dose 30-60 min before the procedure. IM = intramuscular; IV = intravenous; PO = oral. Adapted from Wilson W, Taubert KA, Gewitz M, et al. Prevention of infective endocarditis guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation. 2007;116:1736-1754.

The rate of relapse also varies depending on the causative organism. Easily treated infections, such as those with viridans group streptococci, have a low rate of relapse (5%), but more difficult-to-eradicate organisms may have significantly higher rates.

  Grade A References A1. Marti-Carvajal AJ, Dayer M, Conterno LO, et al. A comparison of different antibiotic regimens for the treatment of infective endocarditis. Cochrane Database Syst Rev. 2016;4:CD009880. A2. Kang DH, Kim YJ, Kim SH, et al. Early surgery versus conventional treatment for infective endocarditis. N Engl J Med. 2012;366:2466-2473. A3. Limeres Posse J, Alvarez Fernandez M, Fernandez Feijoo J, et al. Intravenous amoxicillin/clavulanate for the prevention of bacteraemia following dental procedures: a randomized clinical trial. J Antimicrob Chemother. 2016;71:2022-2030. A4. Krahn AD, Longtin Y, Philippon F, et al. Prevention of arrhythmia device infection trial: the PADIT trial. J Am Coll Cardiol. 2018;72:3098-3109.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 67  Infective Endocarditis  

GENERAL REFERENCES 1. Schranz AJ, Fleischauer A, Chu VH, et al. Trends in drug use-associated infective endocarditis and heart valve surgery, 2007 to 2017: a study of statewide discharge data. Ann Intern Med. 2019;170: 31-40. 2. Toyoda N, Chikwe J, Itagaki S, et al. Trends in infective endocarditis in California and New York State, 1998-2013. JAMA. 2017;317:1652-1660. 3. Ozcan C, Raunso J, Lamberts M, et al. Infective endocarditis and risk of death after cardiac implantable electronic device implantation: a nationwide cohort study. Europace. 2017;19:1007-1014. 4. Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis in adults: diagnosis, antimicrobial therapy, and management of complications: a scientific statement for healthcare professionals from the American Heart Association. Circulation. 2015;132:1435-1486. 5. Choi HN, Park KH, Park S, et al. Prosthetic valve endocarditis caused by HACEK organisms: a case report and systematic review of the literature. Infect Chemother. 2017;49:282-285. 6. Mohananey D, Mohadjer A, Pettersson G, et al. Association of vegetation size with embolic risk in patients with infective endocarditis: a systematic review and meta-analysis. JAMA Intern Med. 2018;178:502-510. 7. Cahill TJ, Prendergast BD. Infective endocarditis. Lancet. 2016;387:882-893. 8. Cahill TJ, Baddour LM, Habib G, et al. Challenges in infective endocarditis. J Am Coll Cardiol. 2017;69:325-344.

428.e1

9. Wang A, Gaca JG, Chu VH. Management considerations in infective endocarditis: a review. JAMA. 2018;320:72-83. 10. Anantha Narayanan M, Mahfood Haddad T, Kalil AC, et al. Early versus late surgical intervention or medical management for infective endocarditis: a systematic review and meta-analysis. Heart. 2016;102:950-957. 11. Pettersson GB, Coselli JS, Hussain ST, et al. 2016 the American Association for Thoracic Surgery (AATS) consensus guidelines: surgical treatment of infective endocarditis: executive summary. J Thorac Cardiovasc Surg. 2017;153:1241-1258. 12. Sandoe JA, Barlow G, Chambers JB, et al. Guidelines for the diagnosis, prevention and management of implantable cardiac electronic device infection. Report of a joint working party project on behalf of the British Society for Antimicrobial Chemotherapy (BSAC, host organization), British Heart Rhythm Society (BHRS), British Cardiovascular Society (BCS), British Heart Valve Society (BHVS) and British Society for Echocardiography (BSE). J Antimicrob Chemother. 2015;70:325-359. 13. Tan EM, DeSimone DC, Sohail MR, et al. Outcomes in patients with cardiovascular implantable electronic device infection managed with chronic antibiotic suppression. Clin Infect Dis. 2017;64: 1516-1521. 14. Thornhill MH, Gibson TB, Cutler E, et al. Antibiotic prophylaxis and incidence of endocarditis before and after the 2007 AHA recommendations. J Am Coll Cardiol. 2018;72:2443-2454. 15. Cahill TJ, Harrison JL, Jewell P, et al. Antibiotic prophylaxis for infective endocarditis: a systematic review and meta-analysis. Heart. 2017;103:937-944.

428.e2

CHAPTER 67  Infective Endocarditis  

REVIEW QUESTIONS 1. A 76-year-old man with known rheumatic valvular heart disease underwent elective mitral valve replacement with a Saint Jude prosthetic valve. His dentist calls you for advice regarding choice of antibiotic prophylaxis before dental extraction because the patient had developed bronchospasm and a diffuse urticarial rash after amoxicillin administration before a dental cleaning approximately 6 months ago. Which antibiotic should be administered? A . Clindamycin 600 mg orally 1 hour before the procedure B. Cefuroxime axetil 500 mg orally before the procedure C. Amoxicillin 2 g intravenously 1 hour before the procedure with corticosteroid and antihistamine coverage D. Nafcillin sodium 2 g IV 1 hour before the procedure E. Gentamicin sulfate 1 mg/kg IV 1 hour before the procedure Answer: A  The current (2007) AHA guidelines recommend clindamycin in patients who have a history of an immediate type hypersensitivity reaction to β-lactam antibiotics, which this patient demonstrated. Therefore, amoxicillin, cephalosporins, and nafcillin should be avoided. Levofloxacin and gentamicin are not recommended in these guidelines. (Wilson W, Taubert KA, Gewitz M, et al. Prevention of infective endocarditis: guidelines from the American Heart Association: a guideline from the American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation. 2007;116:1736-1754.) 2. A 68-year-old man with diabetes on chronic hemodialysis developed the acute onset of fever, chills, and left-sided abdominal pain. He had an ICD implanted 3 years ago. Blood cultures grew Staphylococcus aureus, and transesophageal echocardiography demonstrated vegetations on the mitral valve. Splenic infarctions were seen on computed tomographic scanning. Which one of the following is true regarding this presentation? A . Health care–associated infection is accounting for an increasing number of infective endocarditis cases in this country. B. Escherichia coli is a common cause of infective endocarditis in the hemodialysis population. C. Pending susceptibility testing results, gentamicin should be administered. D. To reduce health care costs, transthoracic echocardiography should have been performed instead of transesophageal echocardiography. E. For chronic hemodialysis, a tunneled catheter has a lower risk of blood stream infection compared with an arteriovenous fistula. Answer: A  Health care exposure accounts for an increasing number of cases of infective endocarditis in developed countries. Staphylococcus aureus, including methicillin-resistant strains, is a common cause of these infections. Empiric vancomycin should be administered until susceptibility results are known. (Athan E, Chu VH, Tattevin P, et al. Clinical characteristics and outcome of infective endocarditis involving implantable cardiac devices. JAMA. 2012;307:1727-1735.) 3. A 55-year-old veterinarian presents with several months of low-grade fever and night sweats. On an echocardiograph, he has evidence of endocarditis. Despite no recent antibiotic therapy, three sets of blood cultures remain negative for 7 days. Which one of the following is the most likely pathogen? A . Coagulase-negative staphylococcus B. Coxiella burnetii C. Enterococcus faecium D. Escherichia coli E. Orf virus

Answer: B  Exposure to animals, particularly sheep and goats, is a risk factor for Coxiella infection and a well-known cause of culture-negative endocarditis. Orf virus does not cause endocarditis. The other choices should result in positive blood culture results in a patient who has infective endocarditis and who has not received antibiotic therapy recently. (Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis: diagnosis, antimicrobial therapy, and management of complications: a statement for healthcare professionals from the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease, Council on Cardiovascular Disease in the Young, and the Councils on Clinical Cardiology, Stroke, and Cardiovascular Surgery and Anesthesia, American Heart Association: endorsed by the Infectious Diseases Society of America. Circulation. 2005;111:e394-434.) 4. A 78-year-old woman presents with left upper chest pain, swelling, and purulent drainage at the site where a permanent pacemaker generator was implanted. She has had no fever or chills, and her white blood cell count is normal. She underwent generator exchange 4 months ago and underwent a dental cleaning without prophylaxis 3 months ago. Which one of the following is true regarding cardiac implantable electronic device (CIED) infections? A . Surgical site prophylaxis has not been shown to reduce the risk of a CIED site infection. B. Prophylactic antibiotic should have been given before her dental cleaning. C. Device manipulation is a risk factor for device infection. D. The most likely cause of this infection is a HACEK organism. E. Antibiotic therapy for 4 weeks will likely cure the device infection without the device being removed. Answer: C  Manipulation of an implantable cardiac device is associated with the development of acute infection. Antibiotic prophylaxis before manipulation of the surgical site is beneficial, but dental prophylaxis is not. The most likely cause is Staphylococcus spp., and removal of the device is required for cure of the infection. (Baddour LM, Cha YM, Wilson WR. Clinical practice. Infections of cardiovascular implantable electronic devices. N Engl J Med. 2012;367:842-849.) 5. A 25-year-old morbidly obese man who injects heroin and cocaine presents with fever and blood cultures that grow Staphylococcus aureus. He had a past history of prior S. aureus blood stream infection 2 years ago, when he had an allergic reaction to vancomycin. At that time, he had evidence of tricuspid valve endocarditis. Which one of the following is true? A . A transthoracic echocardiography should be obtained. B. Initial empiric therapy should include daptomycin until susceptibility results are known. C. His mortality risk is high (>50%). D. Two weeks of antibiotic therapy should be curative. E. Rifampin should be administered. Answer: B  Daptomycin should be administered in case the blood culture isolate is methicillin-resistant Staphylococcus aureus. Transesophageal, rather than transthoracic, echocardiography should be obtained to evaluate for both right-sided and left-sided endocarditis. Cure rates with active antibiotic therapy for 2 weeks are high, provided there is no evidence of left-sided endocarditis or of metastatic foci of infection. Rifampin is not routinely recommended in native valve infections caused by staphylococci. (Fowler VG Jr, Boucher HW, Corey GR, et al. Daptomycin versus standard therapy for bacteremia and endocarditis caused by Staphylococcus aureus. N Engl J Med. 2006;355:653-665.)

428

CHAPTER 68  Pericardial Diseases  

slowly accumulating pericardial effusion to become quite large without compressing the cardiac chambers and to allow left ventricular remodeling to occur without excessive pericardial restraint. Conversely, a scarred or thickened pericardium can limit the filling of the heart, resulting in pericardial constriction. Despite the important functions of the normal pericardium, congenital absence or surgical resection of the pericardium does not appear to have any major untoward effects.

  ACUTE PERICARDITIS  

EPIDEMIOLOGY AND PATHOBIOLOGY

Acute inflammation of the pericardium, with or without an associated pericardial effusion, can occur as an isolated clinical problem or as a manifestation of systemic disease. Although about 85% of isolated cases of acute pericarditis are idiopathic or viral, the list of other potential causes is quite extensive (Table 68-1). Viral causes include echoviruses and group B coxsackieviruses, but obtaining specific viral titers does not alter patient management. About 6% of cases are neoplastic in origin, about 4% are caused by tuberculosis, about 3% are caused by other bacterial or fungal infections, and about 2% are caused by collagen vascular disease.1 Patients with fever greater than 38° C or a subacute course or who fail to respond promptly to therapy are most likely to have pericarditis caused by a systemic autoimmune disease, malignancy, or bacterial infection.  

CLINICAL MANIFESTATIONS

Most patients with acute pericarditis experience sharp retrosternal chest pain (see Table 45-2 in Chapter 45), which can be quite severe and debilitating. In some cases, however, pericarditis is asymptomatic, such as when it accompanies rheumatoid arthritis. Pericardial pain is usually worse with inspiration and when supine, and it is generally relieved by sitting and leaning forward. Typically, pericardial pain is referred to the scapular ridge, presumably owing to irritation of the phrenic nerves, which provide sensory innervation to the pericardium. The chest pain of acute pericarditis must be differentiated from pulmonary embolism and myocardial ischemia or infarction (Table 68-2). The pericardial friction rub is the classic finding in patients with acute pericarditis. A friction rub is a high-pitched, scratchy sound that can have one, two, or three components corresponding to ventricular ejection, rapid ventricular filling in early diastole, and atrial systole. A pericardial rub, which is differentiated from a murmur by its scratchy quality, is sometimes localized to a small area on the chest wall and may come and go spontaneously or with changes in position. To hear a rub, it may be necessary to auscultate the heart with the patient in multiple positions, especially using the diaphragm with the patient learning forward and not breathing after full expiration. The pericardial friction rub must be differentiated from a pleural rub, which is absent during suspended respiration, but the pericardial rub is unaffected.  

68  PERICARDIAL DISEASES BRIAN D. HOIT AND JAE K. OH The pericardium is a two-layered sac composed of visceral and parietal components. The visceral pericardium is a serosal cell monolayer that adheres firmly to the epicardium, reflects over the origin of the great vessels, and together with a tough, fibrous parietal layer encases the heart. The pericardial space enclosed between the two serosal layers normally contains up to 50 mL of plasma ultrafiltrate. The pericardium is well innervated, so pericardial inflammation may produce severe pain and trigger vagally mediated reflexes. As a result of its relatively inelastic physical properties, the pericardium limits acute cardiac dilation and enhances mechanical interactions of the cardiac chambers. In response to chronic stretch, the pericardium dilates to allow a

DIAGNOSIS

A targeted evaluation (Table 68-3) can help distinguish pericarditis from other conditions2 as well as help establish the cause of true pericarditis (Table 68-4).3 The diagnosis of acute pericarditis can be made with two of the following criteria: typical chest pain; pericardial friction rub; characteristic electrocardiogram (ECG) changes; and pericardial effusion.4 Early in the course of acute pericarditis, the ECG typically displays diffuse ST elevation in association with PR depression (Fig. 68-1). The ST elevation is usually present in all leads except for aVR, but the changes may be atypical in post-MI pericarditis. Classically, the ECG changes of acute pericarditis evolve over several days; resolution of the ST elevation is followed by widespread T wave inversion that subsequently normalizes, although the temporal evolution of ECG changes is highly variable. Uremic pericarditis usually occurs without the typical ECG abnormalities. Patients with acute pericarditis usually have evidence of systemic inflammation, including leukocytosis, an elevated erythrocyte sedimentation rate (ESR), and increased C-reactive protein (CRP) level. A low-grade fever is common, but a temperature greater than 38° C is unusual and suggests the possibility of bacterial pericarditis. Troponin levels typically are minimally elevated in acute pericarditis owing to some involvement of the epicardium by the inflammatory process. An elevated troponin level in acute pericarditis usually returns to normal within 1 to 2 weeks and is not associated with a worse prognosis. Although the elevated troponin level may lead to the misdiagnosis of an ST elevation MI (Chapter 64), most patients with elevated troponin levels and acute pericarditis have normal coronary angiograms. An echocardiogram (Chapter 49) can help avoid a misdiagnosis of MI.

CHAPTER 68  Pericardial Diseases  

ABSTRACT

The pericardium is a two-layered sac that encases the heart and constrains its filling. The pericardium can be affected by virtually every category of disease, including infectious, neoplastic, immune-inflammatory, metabolic, iatrogenic, traumatic, and congenital causes. Pericardial heart disease includes pericarditis, cardiac tamponade, constrictive pericarditis, and cysts. Pericarditis is an acute inflammatory condition, typically as a result of viral infection. Such cases usually resolve, occasionally recur, and infrequently progress to pericardial tamponade. Pericarditis caused by acute bacterial infection requires urgent drainage. Tuberculous pericarditis can progress to chronic constriction. The primary abnormality of cardiac tamponade is pan-cyclic compression of the cardiac chambers by increased pericardial fluid, so that all four cardiac chambers compete for a fixed intrapericardial volume. Features responsible for the pathophysiology include transmission of thoracic pressure through the pericardium and heightened ventricular interdependence. In constrictive pericarditis, the pericardium limits diastolic filling, causes dissociation of intracardiac and intrathoracic pressures, and heightens ventricular interdependence. Both conditions result in diastolic dysfunction, elevated and equal venous and ventricular diastolic pressure, respiratory variation in ventricular filling, and, ultimately, reduced cardiac output.

KEYWORDS

pericardium pericarditis pericardial effusion cardiac tamponade pericardial constriction effusive-constrictive imaging

428.e3

CHAPTER 68  Pericardial Diseases  

429

TABLE 68-1 CAUSES OF PERICARDITIS: INFECTIOUS AND NONINFECTIOUS INFECTIOUS PERICARDITIS (⅔ OF CASES) Viral (most common): echovirus and coxsackievirus (usual), influenza, EBV, CMV, adenovirus, varicella, rubella, mumps, HBV, HCV, HIV, parvovirus B19, human herpesvirus 6 (increasing reports) Bacterial: tuberculosis (4-5%)* and Coxiella burnetii (most common); other bacterial causes (rare) include pneumococcosis, meningococcosis, gonococcosis, Haemophilus, staphylococci, Chlamydia, Mycoplasma, Legionella, Leptospira, Listeria Fungal (rare): histoplasmosis more likely in immunocompetent patients; aspergillosis, blastomycosis, candidiasis more likely in immunosuppressed patients Parasitic (very rare): Echinococcus, Toxoplasma NONINFECTIOUS PERICARDITIS (⅓ OF CASES) Autoimmune Pericarditis (5000/μL (autoreactive lymphocytic) or the presence of antibodies against heart muscle tissue (antisarcolemmal) in the pericardial fluid (autoreactive antibody mediated); (2) signs of myocarditis on epicardial or endomyocardial biopsies by ≥14 cells/mm2; and (3) exclusion of infections, neoplasia, and systemic and metabolic disorders. CMV = cytomegalovirus; EBV = Epstein-Barr virus; HBV = hepatitis B virus; HCV = hepatitis C virus; HIV = human immunodeficiency virus. From Imazio M, Spodick DH, Brucato A, et al. Controversial issues in the management of pericardial diseases. Circulation. 2010;121:916-928.

TABLE 68-2 DIFFERENTIATION OF PERICARDITIS FROM MYOCARDIAL ISCHEMIA OR INFARCTION AND PULMONARY EMBOLISM FINDINGS

MYOCARDIAL ISCHEMIA OR INFARCTION

PERICARDITIS

PULMONARY EMBOLISM

CHEST PAIN Character

Pressure-like heavy, squeezing

Sharp, stabbing, occasionally dull

Sharp, stabbing

Change with respiration

No

Worsened with inspiration

In phase with respiration (absent when the patient is apneic)

Change with position

No

Worse when supine; improved when sitting up or leaning forward

No

Duration

Minutes (ischemia); hours (infarction)

Hours to days

Hours to days

Response to nitroglycerin

Improved

No change

No change

Absent (unless pericarditis is present)

Present in most patients

Pleural friction rub may occur

PHYSICAL EXAMINATION Friction rub ELECTROCARDIOGRAM ST segment elevation

Localized convex

Widespread concave

Limited to leads III, aVF, and V1

PR segment depression

Rare

Frequent

None

Modified from Little WC, Freeman GL. Pericardial disease. Circulation. 2006;113:1622-1632.

Echocardiography may demonstrate a small pericardial effusion in the presence of acute pericarditis, but a normal echocardiogram does not exclude the diagnosis of acute pericarditis. An echocardiogram is critical, however, in excluding the diagnosis of cardiac tamponade (see later). When the diagnosis of acute pericarditis is unclear, cardiac magnetic resonance imaging (MRI) can demonstrate pericardial inflammation as delayed enhancement of the pericardium (E-Fig. 68-1). Diagnostic pericardiocentesis is indicated in suspected purulent tuberculosis or malignant pericarditis or if the patient has cardiac tamponade.

TREATMENT  Patients who have manifestations of an underlying systemic process, such as an inflammatory disease, have high risk features (i.e., fever >38° C, subacute course, failure to respond to treatment, an elevated troponin level with evidence of concomitant left ventricular dysfunction, a large pericardial effusion, incipient

or established pericardial tamponade), or have suspected acute MI (Chapter 64) warrant hospitalization and further evaluation. Patients without these features can usually be followed as outpatients (Fig. 68-2). If acute pericarditis is a manifestation of an underlying disease, it often responds to treatment of the primary condition. Most cases of acute idiopathic or viral pericarditis are self-limited and respond to treatment with aspirin (650 mg every 6 hours) or another nonsteroidal anti-inflammatory drug (NSAID) such as ibuprofen (300 to 800 mg every 6 to 8 hours). The dose of NSAID should be tapered after symptoms and any pericardial effusion have resolved, but the medication should be taken for at least 3 to 4 weeks to minimize the risk of recurrent pericarditis. In addition, colchicine (0.6 to 1.2 mg/day for 3 months) should be started in all patients with acute pericarditis5 to reduce the rate of persistent symptoms at 72 hours, reduce the likelihood of recurrent pericarditis from 55 to 24% at 18 months, and reduce the rate of subsequent hospitalization. A1  The major side effect of colchicine is diarrhea. The lower dose of colchicine should be used in patients who weigh less than 70 kg or who have side effects with the higher dose. Colchicine should be avoided in patients with abnormal renal or

CHAPTER 68  Pericardial Diseases  

E-FIGURE 68-1.  Cardiac magnetic resonance image of a patient with acute pericarditis shows late gadolinium hyperenhancement of the pericardium and epicardium.

429.e1

430

CHAPTER 68  Pericardial Diseases  

hepatic function and in patients being treated with macrolide antibiotics, which alter its metabolism. A proton pump inhibitor, such as omeprazole (20 mg/day), should be considered to improve the gastric tolerability of NSAIDs. Warfarin and heparin should be avoided to minimize the risk of hemopericardium, but anticoagulation

TABLE 68-3 SELECTED DIAGNOSTIC TESTS IN ACUTE PERICARDITIS IN ALL PATIENTS Tuberculin skin test (plus control skin test to exclude anergy) BUN and creatinine to exclude uremia Erythrocyte sedimentation rate Electrocardiogram Chest radiograph Echocardiogram

may be required if the patient is in atrial fibrillation or has a prosthetic heart valve. It is prudent to avoid exercise until after the chest pain completely resolves. If pericarditis recurs, the patient can be reloaded with colchicine and intravenous ketorolac (20 mg) and then continued on an oral NSAID and colchicine for at least 3 months. Although acute pericarditis usually responds dramatically to systemic corticosteroids, observational studies strongly suggest that the use of steroids increases the probability of relapse in patients treated with colchicine. Except when needed to treat an underlying inflammatory disease, every effort should be made to avoid the use of steroids, reserving low-dose steroids for patients who cannot tolerate aspirin and other NSAIDs or whose recurrence is not responsive to colchicine and intravenous NSAIDs. If steroids are used, low-dose prednisone (0.2 to 0.5 mg/kg) appears to be as effective as higher doses and is less likely to be associated with recurrence. Steroids should be continued for at least 1 month before slow tapering, which can be guided by return of the CRP level to the normal range. Pericardiocentesis is not recommended unless purulent or tuberculous pericarditis is clinically suspected or the patient fails to respond to 2 to 3 weeks of NSAID therapy.

IN SELECTED PATIENTS Cardiac magnetic resonance imaging ANA and rheumatoid factor to exclude SLE or rheumatoid arthritis in patients with acute arthritis or pleural effusion TSH and T4 to exclude hypothyroidism in patients with clinical findings suggestive of hypothyroidism and in asymptomatic patients with unexplained pericardial effusion HIV test to exclude AIDS in patients with risk factors for HIV disease or a compatible clinical syndrome Blood cultures in febrile patients to exclude infective endocarditis and bacteremia Fungal serologic tests in patients from endemic areas or in immunocompromised patients ASO titer in children or teenagers with suspected rheumatic fever Heterophil antibody test to exclude mononucleosis in young or middle-aged patients with a compatible clinical syndrome or acute fever, weakness, and lymphadenopathy AIDS = acquired immunodeficiency virus; ANA = antinuclear antibody; ASO = antistreptolysin O; BUN = blood urea nitrogen; HIV = human immunodeficiency virus; SLE = systemic lupus erythematosus; T4 = thyroxine; TSH = thyroid-stimulating hormone. Modified from Nishimura RA, Kidd KR. Recognition and management of patients with pericardial disease. In: Braunwald E, Goldman L, eds. Primary Cardiology. 2nd ed. Philadelphia: WB Saunders; 2003:625.



PROGNOSIS

The course of viral and idiopathic pericarditis is usually self-limited, and most patients recover completely. About 25% of patients, however, have recurrent pericarditis weeks to months later, probably caused by an immune response, and some patients may have multiple debilitating episodes. Interestingly, in patients whose acute pericarditis is accompanied by myocarditis, as evidenced by elevation of serum troponin levels, the recurrence rate is closer to 10%. Recurrent pericarditis is more common in patients treated with steroids for the acute episode, especially during a rapid steroid taper. Prolonged high-dose NSAID treatment (e.g., ibuprofen 300 to 600 mg three times a day) plus colchicine (0.5 to 0.6 mg twice daily, reducing to once daily after 3 to 6 months) is effective for recurrent pericarditis. A2  A3  In patients who cannot tolerate colchicine or who have recurrent episodes despite colchicine and high-dose NSAID treatment (e.g., indomethacin 50 mg three times a day or ibuprofen 800 mg four times a day), oral steroids (e.g., prednisone 0.2 to 0.5 mg/kg/ day for 2 to 4 weeks; then slowly tapered over several months) are generally recommended. In one small randomized trial of colchicine-resistant patients ,

TABLE 68-4 PRESENTATION AND TREATMENT OF THE MOST COMMON CAUSES OF PERICARDITIS TYPE

PATHOGENESIS OR ETIOLOGY

DIAGNOSIS

TREATMENT

COMPLICATIONS

COMMENTS

Viral

Coxsackievirus B Echovirus type 8 Epstein-Barr virus

Leukocytosis Elevated ESR Mild cardiac biomarker elevation

Symptomatic relief, NSAIDs, colchicine

Tamponade Relapsing pericarditis

Peaks in spring and fall

Tuberculous

Mycobacterium tuberculosis

Isolation of organism from biopsy fluid Granulomas not specific

Triple-drug antituberculosis regimen Pericardial drainage followed by early (4-6 wk) pericardiectomy if signs of tamponade or constriction develop

Tamponade Constrictive pericarditis

1-8% of patients with tuberculosis pneumonia; rule out HIV infection

Bacterial

Group A streptococcus Staphylococcus aureus Streptococcus pneumoniae

Leukocytosis with marked left shift Purulent pericardial fluid

Pericardial drainage by catheter or surgery Systemic antibiotics Pericardiectomy if constrictive physiology develops

Tamponade in one third of patients

Very high mortality rate if not recognized early

Post–myocardial infarction

12 hours-10 days after infarction

Fever Pericardial friction rub Echo: effusion

Aspirin Prednisone

Tamponade rare

More frequent in large Q wave infarctions Anterior > inferior

Uremic

Untreated renal failure: 50% Chronic dialysis: 20%

Pericardial rub: 90%

Intensive dialysis Indomethacin: probably ineffective Catheter drainage Surgical drainage

Tamponade Hemodynamic instability on dialysis

Avoid NSAIDs ≈50% respond to intensive dialysis

Neoplastic

In order of frequency: lung cancer, breast cancer, leukemia and lymphoma, others

Chest pain, dyspnea Echo: effusion CT, MRI: tumor metastases to pericardium Cytologic examination of fluid positive in 85%

Catheter drainage Subxiphoid pericardiectomy Chemotherapy directed at underlying malignant neoplasm

Tamponade Constriction

CT = computed tomography; Echo = echocardiogram; ESR = erythrocyte sedimentation rate; HIV = human immunodeficiency virus; MRI = magnetic resonance imaging; NSAIDs = nonsteroidal anti-inflammatory drugs. Modified from Malik F, Foster E. Pericardial disease. In: Wachter RM, Goldman L, Hollander H, eds. Hospital Medicine. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2005:449.

431

CHAPTER 68  Pericardial Diseases  

I

aVR

V1

V4

II

aVL

V2

V5

III

aVF

V3

V6

FIGURE 68-1.  Electrocardiogram demonstrating typical features of acute pericarditis on presentation. There are diffuse ST elevation and PR depression except in aVR, where there is ST depression and PR elevation

Cardiac tamponade? Yes

No

TABLE 68-5 CAUSES OF MODERATE TO LARGE ASYMPTOMATIC PERICARDIAL EFFUSIONS CAUSE

Drain effusion

Evaluate per Table 68-3

Evidence of high risk? • Tamponade • Moderate or large effusion • Fever >38° C • Suspicion of systemic illness Yes

Admit to hospital NSAIDs + colchicine Search for etiology Yes

Outpatient follow-up Taper NSAIDs after 3-4 weeks Continue colchicine for 3 months

No

Trial of NSAIDs Pain relieved in 20 mm)? No

Yes

Treat pericarditis

Present for 2 cm) effusions in patients who are hemodynamically stable and in whom tamponade is not suspected, a follow-up echocardiogram should be performed in 7 days and then every month until the effusion is minimal.9 If bacterial or malignant pericarditis is suspected, diagnostic pericardiocentesis should be performed immediately even in the absence of clinical instability or suggestion of tamponade; tuberculous pericarditis is diagnosed best by pericardial biopsy. Anticoagulation with heparin or warfarin should be discontinued unless the patient has a mechanical heart valve or atrial fibrillation. In hypothyroidism (Chapter 213), the effusion and the coexistent cardiomyopathy respond to hormone replacement, sometimes over several months. Uremic pericardial effusions often respond to initiation of dialysis or more intensive dialysis (Chapter 122).

Cardiac Tamponade

The treatment of cardiac tamponade is urgent drainage of the pericardial effusion, especially when there is hemodynamic compromise. Fluid resuscitation may be of transient benefit if the patient is volume depleted (hypovolemic cardiac tamponade), but inotropic agents are usually ineffective because there is already intense endogenous adrenergic stimulation. The initiation of mechanical ventilation in a patient with tamponade may produce a sudden drop in blood pressure because the positive intrathoracic pressure further impairs cardiac filling. Echocardiographic-guided percutaneous pericardiocentesis, which can be performed at the bedside by experienced operators (Fig. 68-6), is indicated if a patient is in dire circumstances and at least 1 cm of fluid is seen anterior to the mid-right ventricular free wall throughout diastole. The ideal entry site

FIGURE 68-6.  Aspiration of pericardial fluid is indicated in cardiac tamponade or to obtain fluid for diagnostic purposes. A wide-bore needle is inserted in the epigastrium below the xiphoid process and advanced in the direction of the medial third of the right clavicle. An alternative site is over the left ventricular apex. The procedure should be performed under echocardiographic guidance, but it may need to be performed emergently for life-saving purposes in other settings. Complications of the procedure include puncture of the heart, arrhythmias, vasovagal attack, and pneumothorax. (From Forbes CD, Jackson WF. Color Atlas and Text of Clinical Medicine. 3rd ed. London: Mosby; 2003.)

(usually the apex) is defined using echocardiography as the minimal distance from the skin to pericardial fluid without intervening structures. The pericardial space is entered with a needle and then drained through a catheter. Although clinical and epidemiologic clues should tailor the extent of the diagnostic evaluation, pericardial fluid should be analyzed for lactate dehydrogenase, protein, cell count, cytology, polymerase chain reaction for tuberculosis, and culture for bacteria and tuberculosis. Continued drainage of the pericardial fluid through an indwelling catheter minimizes the risk of recurrent effusion. For hemodynamically significant effusions of less than 1 cm, organized or multiloculated effusions, and focal effusions, a limited thoracotomy-mediastinoscopy and creation of a pericardial window are advised. Surgical drainage may be the preferred treatment if pericardial tissue is required for diagnosis or in the case of recurrent effusions or bacterial pericarditis. Malignant pericardial effusions frequently recur and, similar to other recurrent pericardial effusions, may necessitate the surgical creation of a pericardial window that allows the effusion to drain into the pleural space, preventing recurrence of cardiac tamponade. An attractive alternative in these patients, especially if their overall prognosis is poor from the malignancy, is the percutaneous creation of a pericardial window by balloon dilation. Hemorrhagic effusions related to cardiac trauma or aortic dissection are best managed by emergency surgery.

434

CHAPTER 68  Pericardial Diseases  



PROGNOSIS

A pericardial effusion may recur or persist. Symptoms are usually weight loss, fatigue, dyspnea on exertion, and whatever symptoms are associated with the specific cause. Treatment of chronic or recurrent idiopathic effusions is similar to the treatment of recurrent pericarditis. If medical therapy is unsuccessful, creation of a pericardial window is indicated. A large idiopathic, asymptomatic effusion that persists for 6 months or longer can unpredictably result in tamponade in as many as 30% of patients over long-term follow-up; diagnostic pericardiocentesis occasionally detects a neoplastic or tuberculous cause. Pericardiocentesis with prolonged drainage resolves many chronic large pericardial effusions, but pericardiectomy is often required. The long-term prognosis depends on the cause of the effusion. With pericardial tamponade, the in-hospital mortality rate is less than 10%, but the subsequent mortality rate is about 75% with a malignant effusion compared with only a 3 to 5% subsequent annual mortality rate for other causes.

  PERICARDIAL CONSTRICTION  

EPIDEMIOLOGY AND PATHOBIOLOGY

Pericardial constriction, which is usually the result of long-standing pericardial inflammation, occurs when a scarred, thickened, or calcified pericardium impairs cardiac filling, thereby limiting the total cardiac volume. The most frequent causes in the developed world are previous cardiac surgery, chronic idiopathic or viral pericarditis, and mediastinal radiation. Constriction may follow cardiac surgery by several weeks to months and may occur decades after chest wall irradiation. In developing countries, tuberculous pericarditis is a more common cause of constrictive pericarditis. Other less common causes include malignant disease, especially lung cancer, breast cancer, or lymphoma; histoplasmosis; rheumatoid arthritis; and uremia. However, a specific cause may not be identified in many patients. With chronic constriction, the pericardium may thicken from its normal 2 mm or less, calcify, and adhere to the epicardium. In a subset of the patients with constriction, the pericardium may be only minimally thickened and less calcified. Fibrous scarring and adhesions of both pericardial layers obliterate the pericardial cavity. The ventricles are unable to fill beyond the rapid filling period because of physical constraints imposed by a thickened, rigid, and sometimes calcified pericardium. The pathophysiologic hallmarks of pericardial constriction are exaggerated interventricular dependence and the dissociation of intracardiac and intrathoracic pressures.  

CLINICAL MANIFESTATIONS

Patients with pericardial constriction typically present with manifestations of elevated systemic venous pressures and low cardiac output. Because there is equalization of all cardiac pressures (including right and left atrial pressures), systemic congestion is much more marked than pulmonary congestion. Typically, patients develop marked jugular venous distention, hepatic congestion, ascites, and peripheral edema, but their lungs remain clear. The limited cardiac output typically presents as exercise intolerance and may progress to cardiac cachexia with muscle wasting. In long-standing pericardial constriction, pleural effusions, ascites, and hepatic dysfunction may be prominent clinical features. Patients with pericardial constriction are much more likely to have left-sided or bilateral pleural effusions than right-sided effusions. Because of the prominent clinical symptoms of ascites and liver enzyme abnormalities, patients may be evaluated for hepatic disease before constrictive pericarditis is recognized. The jugular veins are distended with prominent x and y descents. The normal inspiratory drop in jugular venous distention may be replaced by a rise in venous pressure (Kussmaul sign). The classic auscultatory finding of pericardial constriction is a pericardial knock (Chapter 45), which is a high-pitched sound early in diastole when there is the sudden cessation of rapid ventricular diastolic filling, coinciding with the nadir of the y descent.  

DIAGNOSIS

Pericardial constriction should be considered in any patient with unexplained systemic venous congestion. Pericardial calcification, seen best on the lateral plain chest radiograph, is a classic finding but is present in only 25% of patients with constrictive pericarditis, mostly in those with long-standing constriction. Similarly, most patients with pericardial constriction have a thickened pericardium (>2 mm) that can be imaged by echocardiography, CT, and MRI (Fig. 68-7). It is important to recognize, however, that pericardial constriction can be present without pericardial calcification and, in about 20% of patients, without any obvious pericardial thickening.

FIGURE 68-7.  Computed tomography in a patient with constrictive pericarditis shows a thickened pericardium (arrow).

Transesophageal Doppler echocardiography may demonstrate pericardial thickening and calcification, but increased pericardial thickness can be missed on a transthoracic echocardiogram. Echocardiography also differentiates pericardial constriction from right heart failure caused by tricuspid valve disease or associated pulmonary hypertension. At cardiac catheterization, both cardiac tamponade and pericardial constriction impair diastolic ventricular filling and elevate venous pressure. However, the impairment in ventricular filling with constriction is minimal in early diastole until cardiac volume reaches the anatomic limit set by the noncompliant pericardium, at which time diastolic pressure rises abruptly and remains elevated until the onset of systole. The result is a prominent y descent with an elevated plateau of ventricular pressure, which has been termed the “square root” sign (E-Fig. 68-3) that differentiates constriction from tamponade, in which the y descent is absent. Stroke volume and cardiac output are reduced because of impaired filling, but the intrinsic systolic function of the ventricles can be normal.

Differential Diagnosis

The most difficult differentiation is between pericardial constriction and restrictive cardiomyopathy (Chapter 54) (E-Fig. 68-4), the clinical manifestations of which may be very similar to those of pericardial constriction (Table 68-6).10 Doppler echocardiography is the most useful method to distinguish constriction from restriction. Whereas patients with pericardial constriction usually have pronounced respiratory variation (>25%) of mitral inflow E velocity, patients with restrictive cardiomyopathies do not. In some patients with pericardial constriction and markedly elevated venous pressures, the respiratory variation may be present only after head-up tilt. The tissue Doppler measurement of early diastolic septal mitral annular velocity (e′) is almost always reduced in patients with myocardial restriction, but it remains normal or increased in patients with pericardial constriction. In addition, lateral e′, which is higher than septal or medial e′ velocity in normal and restrictive cardiomyopathy, is lower than septal e′ in most patients with constrictive pericarditis. Whereas a prominent diastolic reversal of hepatic vein flow velocity during expiration is characteristic of constriction, the reversal flow velocity occurs during inspiration in patients with right heart failure from other causes. Patients with pericardial constriction usually have only minimally elevated (115 mg/dL) Diabetes mellitus Impaired fasting glucose level (102-125 mg/dL) or abnormal glucose tolerance test result Family history of premature cardiovascular disease Abdominal obesity SUBCLINICAL TARGET ORGAN DAMAGE Left ventricular hypertrophy Carotid wall thickening or plaque Low estimated glomerular filtration rate (≤60 mL/min/1.73 m2) Microalbuminuria with urine albumin-to-creatinine ratio ≥30 mg/g Ankle-brachial blood pressure index leg BP, chest bruits, rib MR angiography; TEE; invasive angiography notching on chest radiography

Cushing syndrome (Chapter 214)

Incidental adrenal mass, truncal obesity, wide and blanching purple striae, muscle weakness

1 mg dexamethasone-suppression test; urinary cortisol after dexamethasone; adrenal CT

Pheochromocytoma (Chapter 215)

Incidental adrenal mass; paroxysms of hypertension, palpitations, perspiration, and pallor; diabetes

Plasma metanephrines; 24-hour urinary metanephrines and catecholamines; abdominal CT or MR imaging

Renal sonography

ACE = angiotensin-converting enzyme; ARB = angiotensin receptor blocker; BP = blood pressure; CT = computed tomography; GFR = glomerular filtration rate; MR, magnetic resonance; TEE = transesophageal echocardiography.

hypokalemia on initial presentation, and the diagnosis should also be considered in any patient with resistant hypertension, hypertensive heart disease, easily provoked hypokalemia on diuretic therapy, a family history of aldosteronism, or an incidentally discovered adrenal mass.  

DIAGNOSIS

Patients with suspected primary aldosteronism should be screened with a serum aldosterone level and plasma renin activity at 8 am, when aldosterone secretion is highest.8 Patients with a positive screening test (serum aldosterone ≥12 and plasma renin activity 1, serum potassium level >4.0 without any potassium supplements, and control of hypertension with a tolerable multidrug regimen.

  RENAL PARENCHYMAL HYPERTENSION

Hypertension is present in more than 85% of patients with chronic kidney disease (Chapter 121) and is a major factor causing their increased cardiovascular morbidity and mortality. The mechanisms causing the hypertension include an expanded plasma volume and peripheral vasoconstriction; the peripheral vasoconstriction is caused by both activation of vasoconstrictor pathways (renin-angiotensin and sympathetic nervous systems) and inhibition of vasodilator pathways (nitric oxide). Measurement of serum creatinine alone is an inadequate screening test for renal insufficiency. Creatinine clearance should be calculated (www.nephron.com) (Chapter 106) to screen for an estimated GFR less than 60 mL/minute per 1.73 m2. Also, a spot urine specimen should be obtained to screen for microalbuminuria, which is defined as a urine albumin-to-urine creatinine ratio of

30 to 300 mg/g (equivalent to excretion of 30 to 300 mg of albumin per 24 hours), with higher levels of albuminuria indicating more advanced chronic kidney disorder. In patients with mild (stage 2: GFR of 60 to 90 mL/minute per 1.73 m2) or moderate (stage 3: GFR of 30 to 60 mL/minute per 1.73 m2) proteinuric chronic kidney disease, stringent blood pressure control is important both to slow the progression of renal disease and to reduce the excessive cardiovascular risk. In patients with severe chronic kidney disease, hypertension often becomes difficult to treat and may require either (1) intensive medical treatment with loop diuretics, potent vasodilators (e.g., minoxidil), high-dose β-adrenergic blockers, and central sympatholytics; or (2) initiation of chronic hemodialysis as the only effective way to reduce plasma volume. In chronic hemodialysis patients, the challenge is to control interdialytic hypertension without exacerbating dialysis-induced hypotension. The annual mortality rate in the hemodialysis population (Chapter 122) is 25%, with half of the excess mortality caused by cardiovascular events related, at least in part, to hypertension.

  RENOVASCULAR HYPERTENSION

The two main causes of renal artery stenosis (Chapter 116) are atherosclerosis (85% of cases), typically in older persons with other clinical manifestations of systemic atherosclerosis, and fibromuscular dysplasia (15% of cases), typically in otherwise healthy young women. Unilateral renal artery stenosis can cause underperfusion of the juxtaglomerular cells, thereby resulting in renindependent hypertension even though the contralateral kidney maintains normal blood volume. In contrast, bilateral renal artery stenosis (or unilateral stenosis with a solitary kidney) constitutes a potentially reversible cause of progressive renal failure and volume-dependent hypertension.  

CLINICAL MANIFESTATIONS

Most patients with atherosclerotic renal artery are older persons with hypertension, hyperlipidemia, and clinically evident atherosclerosis in their coronary, peripheral arterial, or cerebrovascular circulation. Although atherosclerotic renal artery stenosis and hypertension frequently coexist, the presence of a renal artery stenosis proves neither that the patient’s hypertension is renovascular in origin nor that revascularization will improve renal perfusion and blood pressure. Three typical presentations of severe atherosclerotic renal artery are: (1) drug-refractory hypertension, (2) flash pulmonary edema, and (3) ischemic nephropathy. Fibromuscular hyperplasia may be suggested based on disease in the carotid or other arteries in young patients, especially women, with difficult-to-treat hypertension (Chapter 116) but no family history of hypertension.  

DIAGNOSIS

Patients suspected of having renal artery stenosis ( diltiazem > verapamil. Clinical Use.  Calcium-channel blockers are generally well tolerated, do not require monitoring with blood tests, and are proven to be safe and effective. They also are useful antianginal drugs (Chapter 362) and provide better stroke protection than do other antihypertensive agents. Amlodipine is equivalent to chlorthalidone (a potent thiazide-like diuretic) and lisinopril (an ACE inhibitor) in protecting against nonfatal coronary events, stroke, and death but provides less protection against heart failure. Advantages of amlodipine include predictable dose-dependent potency, once-daily dosing because of its long half-life, tolerability, and cost (≤$10 per month for generic amlodipine). Unlike diuretics, ACE inhibitors and ARBs, a high-salt diet or concurrent nonsteroidal antiinflammatory drug (NSAID) therapy does not compromise the effectiveness of dihydropyridine calcium-channel blockers. These drugs, which have some diuretic action, lower blood pressure and prevent hypertensive complications equally in black and nonblack patients. Amlodipine and other dihydropyridine calciumchannel blockers are less renoprotective than ACE inhibitors or ARBs in patients with proteinuric chronic kidney disease. Such patients should not receive amlodipine as first-line therapy but often benefit from it as an adjunct after initiation of appropriate first-line therapy with either an ACE inhibitor or ARB, as well as a diuretic. Diltiazem is a usually well-tolerated alternative in patients who cannot tolerate amlodipine or would benefit from its other effects. Verapamil is not recommended because it is a weak antihypertensive medication and causes constipation. Side Effects.  Short-acting dihydropyridines are not to be used to treat hypertension. By triggering an abrupt fall in blood pressure with reflex sympathetic activation, these rapidly acting arterial vasodilators can precipitate myocardial ischemia/infarction and death. The principal side effect of the dihydropyridines is dose-dependent ankle edema. With amlodipine, ankle edema is far more common with a 10-mg dose than with 2.5- or 5-mg doses. The edema, which is mainly vasogenic because of selective arterial dilation, can be improved by concomitant therapy with an ACE inhibitor or ARB that causes balanced arterial and venous dilation. Long-acting dihydropyridine calcium-channel blockers are rarely associated with flushing and headache. All calcium-channel blockers

TABLE 70-5 SELECTED ORAL ANTIHYPERTENSIVE AGENTS DRUG

DOSE RANGE, TOTAL, MG/DAY (DOSES PER DAY)

USUAL STARTING DOSE, MG/DAY (DOSES PER DAY) DRUG

DOSE RANGE, TOTAL, MG/DAY (DOSES PER DAY)

USUAL STARTING DOSE, MG/DAY (DOSES PER DAY)

CALCIUM-CHANNEL BLOCKERS

Metoprolol XL

50-200 (1-2)

50 (1)

Dihydropyridines

Nadolol

20-320 (1)

40 (1)

Amlodipine

2.5-10 (1)

2.5 (1)

Penbutolol

10-80 (1)

10 (1)

Felodipine

2.5-20 (1-2)

2.5 (2)

Pindolol

10-60 (2)

10 (1)

Isradipine CR

2.5-20 (2)

2.5 (2)

Propranolol

40-180 (2)

40 (2)

Nicardipine SR

30-120 (2)

30 (2)

Propranolol LA

60-180 (1-2)

60 (1)

Nifedipine XL

30-120 (1)

30 (1)

Timolol

20-60 (2)

20 (2)

Nisoldipine

10-40 (1-2)

10 (2)

VASODILATING β-BLOCKERS Carvedilol

6.25-50 (2)

6.25 (2)

Diltiazem CD

120-540 (1-2)

180 (1)

Carvedilol CR

10-80 (1)

20 (1)

Verapamil HS

120-480 (1)

180 (1)

Labetalol

100-2400 (2)

200 (2)

Nebivolol

5-40 (1)

5 (1)

150-300 (1)

150 (1)

Nondihydropyridines

ANGIOTENSIN-CONVERTING ENZYME INHIBITORS Benazepril

10-80 (1-2)

20 (1)

DIRECT RENIN INHIBITOR

Captopril

25-150 (2)

25 (2)

Aliskiren

Enalapril

2.5-40 (2)

5 (2)

α-BLOCKERS

Fosinopril

10-80 (1-2)

20 (2)

Doxazosin

1-16 (1-2)

1 (1)

Lisinopril

5-80 (1-2)

40 (2)

Prazosin

1-40 (2-3)

1 (2)

Moexipril

7.5-30 (1)

7.5 (1)

Terazosin

1-20 (1)

1 (1)

Perindopril

4-16 (1)

4 (1)

20-120 (2)

20 (2)

Quinapril

5-80 (1-2)

40 (2)

Phenoxybenzamine for pheochromocytoma

Ramipril

2.5-20 (1)

2.5 (1)

Trandolapril

1-8 (1)

2 (1)

Clonidine

0.3-1.2 (3)

0.3 (3)

Clonidine patch

0.1-0.6 (weekly)

0.1 (weekly)

Guanabenz

2-32 (2)

2 (2)

Guanfacine

1-3 (1) (qhs)

1 (1)

Methyldopa

250-1000 (2)

250 (2)

Reserpine

0.05-0.25 (1)

0.05 (1)

Hydralazine

10-200 (3)

25 (3)

Minoxidil

2.5-100 (1)

2.5 (1)

Aliskiren/HCTZ

75-300/12.5-25 (1)

150/12.5 (1)

Amiloride/HCTZ

5/50 (1)

5/50 (1)

Amlodipine/benazepril

2.5-5/10-20 (1)

2.5/10 (1)

Amlodipine/olmesartan

5-10/20-40 (1)

5/20 (1)

Amlodipine/telmisartan

5/20-10/80 (1)

5/20 (1)

Amlodipine/valsartan

5-10/160-320 (1)

5/160 (1)

Atenolol/chlorthalidone

50-100/25 (1)

50/25 (1)

Azilsartan/chlorthalidone

40-80/12.5-25 (1)

40/12.5 (1)

Benazepril/HCTZ

5-20/6.25-25 (1)

20/6.25 (1)

Bisoprolol/HCTZ

2.5-10/6.25 (1)

2.5/6.25 (1)

Candesartan/HCTZ

16-32/12.5-25 (1)

16/12.5 (1)

Enalapril/HCTZ

5-10/25 (1-2)

5/25 (1)

Eprosartan/HCTZ

600/12.5-25 (1)

600/12.5 (1)

Fosinopril/HCTZ

10-20/12.5 (1)

10/12.5 (1)

Irbesartan/HCTZ

150-300/12.5-25 (1)

150/12.5 (1)

Losartan/HCTZ

50-100/12.5-25 (1)

50/12.5 (1)

Olmesartan/HCTZ

20-40/12.5 (1)

20/12.5 (1)

Olmesartan/amlodine/HCTZ

20-40/5-10/12.5-25 (1)

20/5/12.5 (1)

Spironolactone/HCTZ

25/25 ( 1 2 -1)

25/25 (1/2)

Telmisartan/HCTZ

40-80/12.5-25 (1)

40/12.5 (1)

Telmisartan/amlodipine/HCTZ

40-80/2.5-10/12.5-25 (1)

40/5/12.5 (1)

Trandolapril/verapamil

2-4/180-240 (1)

2/180 (1)

Triamterene/HCTZ

37.5/25 ( 2 -1)

37.5/25 (1/2)

Valsartan/HCTZ

80-160/12.5-25 (1)

160/12.5 (1)

Valsartan/Amlodipine/HCTZ

80-160/5-10/12.5-25 (1)

160/5/12.5 (1)

ANGIOTENSIN RECEPTOR BLOCKERS Azilsartan

40-80 (1)

40 (1)

Candesartan

8-32 (1-2)

8 (1)

Eprosartan

400-800 (1-2)

400 (1)

Irbesartan

75-300 (1)

150 (1)

Losartan

25-100 (2)

50 (1)

Olmesartan

5-40 (1)

20 (1)

Telmisartan

10-80 (1)

40 (1)

Valsartan

80-320 (1-2)

160 (2)

DIURETICS Thiazide-Type Diuretics Indapamide

0.625-2.5 (1)

1.25 (1)

Chlorthalidone

6.25-50 (1)

12.5 (1)

HCTZ*

12.5-50 (1)

12.5 (1)

Metolazone

2.5-5 (1)

2.5 (1)

Loop Diuretics Bumetanide

0.5-2 (2)

1 (2)

Ethacrynic acid

25-100 (2)

25 (2)

Furosemide

20-160 (2)

20 (2)

Torsemide

2.5-20 (1-2)

5 (2)

Potassium Sparing Eplerenone

25-100 (1-2)

25 (1)

Spironolactone

12.5-100 (1-2)

12.5 (1)

Triamterene

25-100 (1)

37.5 (1)

Amiloride

5-20 (1)

10 (2)

STANDARD β-BLOCKERS Acebutolol

200-800 (2)

200 (2)

Atenolol

25-100 (1)

25 (1)

Betaxolol

5-20 (1)

5 (1)

Bisoprolol

2.5-20 (1)

2.5 (1)

Carteolol

2.5-10 (1)

2.5 (1)

Metoprolol

50-450 (2)

50 (2)

CENTRAL SYMPATHOLYTICS

DIRECT VASODILATORS

FIXED-DOSE COMBINATIONS

*HCTZ = hydrochlorothiazide; there is no evidence to support the use of 12.5-25 mg of HCTZ to reduce the risk of cardiovascular events.

1

450

CHAPTER 70  Arterial Hypertension  

TABLE 70-6 MAJOR CONTRAINDICATIONS AND SIDE EFFECTS OF ANTIHYPERTENSIVE DRUGS DRUG CLASS Diuretics Thiazides

MAJOR CONTRAINDICATIONS Gout

Loop diuretics

Hepatic coma

Potassium-sparing diuretics

Serum potassium concentration >5.5 mEq/L GFR 50 years; (2) systolic blood pressure 130-180 mm Hg; (3) high cardiovascular risk defined as clinical coronary disease, estimated glomerular filtration rate 20-59 mL/min/1.73 m2, 10-year score >15%*, or age ≥75 years. SPRINT exclusion criteria: (1) diabetes†; (2) history of stroke; (3) proteinuria >1 g in 24 hours; (4) heart failure; (5) estimated glomerular filtration rate 60. (From Wright JT, Jr., Williamson JD, Whelton, PK, et al. A randomized trial of intensive versus standard bloodpressure control. N Engl J Med. 2015;373:2103-2116.)

CHAPTER 70  Arterial Hypertension  

452.e1

E-TABLE 70-1 DIFFERENCES AMONG CURRENT TREATMENT GUIDELINES FOR ADULTS WITH HYPERTENSION GUIDELINE 2017 AACE/ACE1

POPULATION Diabetes

THRESHOLD OFFICE BP (mm Hg) TO START OR INTENSIFY THERAPY

INITIAL DRUG OPTIONS

≥130/80 for most ≥120/80 if high CVD risk and low risk for adverse

ACE-I or ARB, CCB, thiazides, BB

≥140/90 ≥130/80 if high CVD risk (especially for stroke)

ACE-I or ARB, CCB, Thiazide-type diuretic

effects of intensive BP therapy Less stringent goals for frail patients or those with adverse medication effects 2017 ADA2

Diabetes

2017 ACP/AAFP3

≥60 y:

General Stroke or TIA High CV risk

≥150 ≥140 ≥140

2017 Hypertension Canada’s CHEP4

Low risk Macrovascular target organ damage or other risk factors Selected high risk (including ≥75 y)  Diabetes  CKD

≥160/100 ≥140/90

Thiazide-type diuretic or BB (6 mo may be reasonable

12 mo No high risk of bleeding and no significant overt bleeding on DAPT Class IIb: >12 mo may be reasonable

FIGURE 76-1.  Treatment algorithm for duration of P2Y12 inhibitor therapy in patients with CAD treated with DAPT. Colors correspond to Class of Recommendation. Clopidogrel is the only currently used P2Y12 inhibitor studied in patients with SIHD undergoing PCI. Aspirin therapy is almost always continued indefinitely in patients with CAD. Patients with a history of ACS more than 1 year prior who have since remained free of recurrent ACS are considered to have transitioned to SIHD. In patients treated with DAPT after DES implantation who develop a high risk of bleeding (e.g., treatment with oral anticoagulant therapy), are at high risk of severe bleeding complication (e.g., major intracranial surgery), or develop significant overt bleeding, discontinuation of P2Y12 inhibitor therapy after 3 months for SIHD or after 6 months for ACS may be reasonable. Arrows at the bottom of the figure denote that the optimal duration of prolonged DAPT is not established. ACS indicates acute coronary syndrome; BMS, bare metal stent; CABG, coronary artery bypass graft surgery; CAD, coronary artery disease; DAPT, dual antiplatelet therapy; DES, drug-eluting stent; Hx, history; Lytic, fibrinolytic therapy; NSTE-ACS, non–ST-elevation acute coronary syndrome; PCI, percutaneous coronary intervention; SIHD, stable ischemic heart disease; S/P, status post; and STEMI, ST-elevation myocardial infarction. (Reproduced from Levine GN, Bates ER, Bittl JA, Brindis RG, et al. 2016 ACC/AHA guideline focused update on duration of dual antiplatelet therapy in patients with coronary artery disease. J Am Coll Cardiol. 2016;68:1082-115.)

in patients 75 years of age or older or less than 60 kg, so a different P2Y12 inhibitor should be used in such patients.

juice) or CYP2C19 (e.g., omeprazole), the dose should be reduced to 50 mg twice daily.

Laboratory Monitoring of Aspirin and P2Y12 Inhibitor Therapy

Clinical Uses

The platelet inhibitory effects of aspirin or a P2Y12 blocker can be monitored in any given individual receiving such therapy by a variety of in vitro methods, including point-of-care blood testing. However, randomized trials have been unable to demonstrate that personalized antiplatelet therapy based on pointof-care testing of platelet function is effective in reducing ischemic events.17

Phosphodiesterase Inhibitors Mechanism of Action

Dipyridamole inhibits the enzyme phosphodiesterase-5 in platelets, thereby increasing intraplatelet cyclic AMP and cyclic GMP levels, and also blocks the uptake of adenosine, thereby inhibiting platelet aggregation. Dipyridamole also leads to increased vascular smooth muscle cell cyclic GMP levels and coronary artery vasodilation. Cilostazol is a phosphodiesterase-3 inhibitor that inhibits platelet activation and leads to smooth muscle cell relaxation.

Pharmacology

The absorption and bioavailability of dipyridamole is variable, and multiple formulations of the drug are commercially available. In current clinical practice, more extended- or sustained-release formulations of dipyridamole are used. Dipyridamole is metabolized in the liver, and it should be avoided in patients with severe hepatic or renal disease. An extended release combination dipyridamole (200 mg) and aspirin (25 mg), used for prevention of stroke, is taken twice daily. Cilostazol is rapidly absorbed; its antiplatelet effects are observed as early as 3 hours after a dose and persist for up to 12 hours. Cilostazol is metabolized in the liver by hepatic cytochrome P-450 enzymes 3A4 and 2C19. The usual dose is 100 mg twice daily, taken at least half an hour before or two hours after breakfast and dinner. In patients concomitantly taking strong or moderate inhibitors of CYP3A4 (e.g., ketoconazole, eryrothmycin, diltiazem, grapefruit

Dipyridamole alone exerts little clinically relevant antithrombotic effect, and the drug is used primarily in combination with aspirin for platelet inhibition and reduction in cardiovascular ischemic events. Combination therapy with aspirin and dipyridamole is at least as effective as aspirin alone for the secondary prevention of stroke, but is less well tolerated. In current clinical practice, dipyridamole is not used for the prevention of cardiac ischemic events. However, extended-release dipyridamole plus aspirin is an effective option for the secondary prevention of noncardioembolic ischemic stroke (Chapter 379). Because of its vasodilatory effects and potential for a “coronary steal” phenomenon, dipyridamole should be used with caution in patients with a recent ACS or severe coronary artery disease. Cilostazol, presumably owing to its vasodilatory action, improves claudication symptoms and is recommended as an effective therapy to improve symptoms and increase walking distance in patients with claudication (Chapter 71).18,19 Cilostazol is contraindicated in patients with heart failure.

Thrombin Receptor Antagonists Mechanism of Action

Thrombin is a potent platelet activator that acts via the platelet protease-activated receptor (PAR). The thrombin receptor antagonist vorapaxar inhibits thrombinmediated platelet activation by binding to the platelet PAR-1 receptor.

Pharmacology

Vorapaxar is rapidly and almost completely absorbed in the gastrointestinal tract. Peak antiplatelet activity occurs within 1 to 2 hours. Vorapaxar is metabolized in the liver via the P450 (CYP) 3A4 enzyme. Vorapaxar should not be used concomitantly with drugs that are strong inhibitors or inducers of CYP3A4. Although vorapaxar reversibly inhibits the PAR-1 receptor, the drug’s long half-life makes it effectively irreversible. The dose of vorapaxar is 2.08 mg once daily.

CHAPTER 76  Antithrombotic and Antiplatelet Therapy  

Clinical Uses

Concerns about bleeding complications have limited the development, approval, and incorporation of thrombin receptor antagonists into clinical practice. Vorapaxar has been studied only as an addition to aspirin and/or clopidogrel therapy, not as a sole antiplatelet agent. It is FDA approved for the reduction of future thrombotic cardiovascular events in patients with a history of MI (Chapters 63 and 64) or with peripheral arterial disease (Chapter 71).

  Grade a References A1. Belley-Cote EP, Hanif H, D’Aragon F, et al. Genotype-guided versus standard vitamin K antagonist dosing algorithms in patients initiating anticoagulation. Thromb Haemost. 2015;114:768-777. A2. Hakoum MB, Kahale LA, Tsolakian IG, et al. Anticoagulation for the initial treatment of venous thromboembolism in people with cancer. Cochrane Database Syst Rev. 2018;1:CD006649. A3. Li A, Garcia DA, Lyman GH, et al. Direct oral anticoagulant (DOAC) versus low-molecular-weight heparin (LMWH) for treatment of cancer associated thrombosis (CAT): a systematic review and meta-analysis. Thromb Res. 2019;173:158-163. A4. Brandao GM, Junqueira DR, Rollo HA, et al. Pentasaccharides for the treatment of deep vein thrombosis. Cochrane Database Syst Rev. 2017;12:CD011782. A5. Eikelboom JW, Connolly SJ, Bosch J, et al. Rivaroxaban with or without aspirin in stable cardiovascular disease. N Engl J Med. 2017;377:1319-1330. A6. Buller HR, Prins MH, Lensin AW, et al. Oral rivaroxaban for the treatment of symptomatic pulmonary embolism. N Engl J Med. 2012;366:1287-1297. A7. Weitz JI, Lensing AWA, Prins MH, et al. Rivaroxaban or aspirin for extended treatment of venous thromboembolism. N Engl J Med. 2017;376:1211-1222. A8. Agnelli G, Buller HR, Cohen A, et al. Oral apixaban for the treatment of acute venous thromboembolism. N Engl J Med. 2013;369:799-808. A9. Agnelli G, Buller HR, Cohen A, et al. Apixaban for extended treatment of venous thromboembolism. N Engl J Med. 2013;368:699-708. A10. Büller HR, Decousus H, Grosso MA, et al. Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism. N Engl J Med. 2013;369:1406-1415.

501

A11. Giugliano RP, Ruff CT, Braunwald E, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2013;369:2093-2104. A12. Goette A, Merino JL, Ezekowitz MD, et al. Edoxaban versus enoxaparin-warfarin in patients undergoing cardioversion of atrial fibrillation (ENSURE-AF): a randomised, open-label, phase 3b trial. Lancet. 2016;388:1995-2003. A13. Cohen AT, Harrington RA, Goldhaber SZ, et al. Extended thromboprophylaxis with betrixaban in acutely ill medical patients. N Engl J Med. 2016;375:534-544. A14. Bruins Slot KM, Berge E. Factor Xa inhibitors versus vitamin K antagonists for preventing cerebral or systemic embolism in patients with atrial fibrillation. Cochrane Database Syst Rev. 2018;3:CD008980. A15. Sterne JA, Bodalia PN, Bryden PA, et al. Oral anticoagulants for primary prevention, treatment and secondary prevention of venous thromboembolic disease, and for prevention of stroke in atrial fibrillation: systematic review, network meta-analysis and cost-effectiveness analysis. Health Technol Assess. 2017;21:1-386. A16. Sharma M, Cornelius VR, Patel JP, et al. Efficacy and harms of direct oral anticoagulants in the elderly for stroke prevention in atrial fibrillation and secondary prevention of venous thromboembolism: systematic review and meta-analysis. Circulation. 2015;132:194-204. A17. Raskob GE, van Es N, Verhamme P, et al. Edoxaban for the treatment of cancer-associated venous thromboembolism. N Engl J Med. 2018;378:615-624. A18. Meyer G, Vicaut E, Danays T, et al. Fibrinolysis for patients with intermediate-risk pulmonary embolism. N Engl J Med. 2014;370:1402-1411. A19. Zheng SL, Roddick AJ. Association of aspirin use for primary prevention with cardiovascular events and bleeding events: a systematic review and meta-analysis. JAMA. 2019;321:277-287. A20. Bowman L, Mafham M, Wallendszus K, et al. Effects of aspirin for primary prevention in persons with diabetes mellitus. N Engl J Med. 2018;379:1529-1539. A21. McNeil JJ, Nelson MR, Woods RL, et al. Effect of aspirin on all-cause mortality in the healthy elderly. N Engl J Med. 2018;379:1519-1528. A22. Sharma A, Hai O, Garg A, et al. Duration of dual antiplatelet therapy following drug-eluting stent implantation: a systematic review and meta-analysis of randomized controlled trials. Curr Probl Cardiol. 2017;42:404-417. A23. Raheja H, Garg A, Goel S, et al. Comparison of single versus dual antiplatelet therapy after TAVR: a systematic review and meta-analysis. Catheter Cardiovasc Interv. 2018;92:783-791.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 76  Antithrombotic and Antiplatelet Therapy  

GENERAL REFERENCES 1. Vestergaard AS, Skjøth F, Larsen TB, et al. The importance of mean time in therapeutic range for complication rates in warfarin therapy of patients with atrial fibrillation: a systematic review and meta-regression analysis. PLoS ONE. 2017;12:1-17. 2. Karimi K, Odhav A, Kollipara R, et al. Acute cutaneous necrosis: a guide to early diagnosis and treatment. J Cutan Med Surg. 2017;21:425-437. 3. Shen YM, Wolfe H, Barman S. Evaluating thrombocytopenia during heparin therapy. JAMA. 2018;319:497-498. 4. Zeitouni M, Kerneis M, Nafee T, et al. Anticoagulation in acute coronary syndrome—state of the art. Prog Cardiovasc Dis. 2018;60:508-513. 5. Mumoli N, Mastroiacovo D, Tamborini-Permunian E, et al. Dabigatran in nonvalvular atrial fibrillation: from clinical trials to real-life experience. J Cardiovasc Med (Hagerstown). 2017;18:467-477. 6. Calkins H, Willems S, Gerstenfeld EP, et al. Uninterrupted dabigatran versus warfarin for ablation in atrial fibrillation. N Engl J Med. 2017;376:1627-1636. 7. Pollack CV Jr, Reilly PA, van Ryn J, et al. Idarucizumab for dabigatran reversal—full cohort analysis. N Engl J Med. 2017;377:431-441. 8. Tornkvist M, Smith JG. Labaf A. Current evidence of oral anticoagulant reversal: a systematic review. Thromb Res. 2018;162:22-31. 9. Powers WJ, Rabinstein AA, Ackerson T, et al. 2018 guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2018;49:e46-e110. 10. McFadyen JD, Schaff M, Peter K. Current and future antiplatelet therapies: emphasis on preserving haemostasis. Nat Rev Cardiol. 2018;15:181-191. 11. Patrono C, Morais J, Baigent C, et al. Antiplatelet agents for the treatment and prevention of coronary atherothrombosis. J Am Coll Cardiol. 2017;70:1760-1776.

501.e1

12. Rothwell PM, Cook NR, Gaziano JM, et al. Effects of aspirin on risks of vascular events and cancer according to bodyweight and dose: analysis of individual patient data from randomised trials. Lancet. 2018;392:387-399. 13. Levine GN, Bates ER, Bittl JA, et al. 2016 ACC/AHA guideline focused update on duration of dual antiplatelet therapy in patients with coronary artery disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2016;68:1082-1115. 14. Guirguis-Blake JM, Evans CV, Senger CA, et al. Aspirin for the primary prevention of cardiovascular events: a systematic evidence review for the U.S. Preventive Services Task Force. Ann Intern Med. 2016;164:804-813. 15. Valgimigli M, Bueno H, Byrne RA, et al. 2017 ESC focused update on dual antiplatelet therapy in coronary artery disease developed in collaboration with EACTS: the task force for dual antiplatelet therapy in coronary artery disease of the European Society of Cardiology (ESC) and of the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J. 2018;39:213-260. 16. Vos CG, Vahl A. Anticoagulation and antiplatelet therapy in patients with peripheral arterial disease of the femoro-popliteal arteries. J Cardiovasc Surg (Torino). 2018;59:164-171. 17. Michelson AD, Bhatt DL. How I use laboratory monitoring of antiplatelet therapy. Blood. 2017;130:713-721. 18. Gerhard-Herman MD, Gornik HL, Barrett C, et al. 2016 AHA/ACC guideline on the management of patients with lower extremity peripheral artery disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2017;69:e71-e126. 19. Tsigkou V, Siasos G, Rovos K, et al. Peripheral artery disease and antiplatelet treatment. Curr Opin Pharmacol. 2018;39:43-52.

501.e2

CHAPTER 76  Antithrombotic and Antiplatelet Therapy  

REVIEW QUESTIONS 1. Based on the half-life and onset of action of warfarin, which of the following is the optimal management of this drug in the case of uncomplicated major surgery? A . Warfarin should be stopped 5 days before surgery and restarted as soon as the patient can take oral medications after surgery. B. Warfarin should be stopped 2 days before surgery and restarted as soon as the patient can take oral medications after surgery. C. Warfarin should be stopped 5 days before surgery and restarted at least 5 days after surgery. D. Warfarin should be stopped 2 days before surgery and restarted at least 5 days after surgery. E. Warfarin should be stopped 1 day before surgery and a dose of vitamin K given 1 day before surgery to reverse warfarin, which then is restarted as soon as the patient can take oral medications after surgery. Answer: A  With a half-life of 40 hours, warfarin has to be stopped 5 days before surgery to eliminate the anticoagulant effect. It takes 5 to 7 days for warfarin to achieve therapeutic anticoagulant effect, and thus it can be started very shortly after surgery provided that there is no bowel paralysis or active bleeding. 2. For a patient requiring treatment for pulmonary embolism but with a high risk for bleeding, for whom quick elimination of the anticoagulant effect if needed is desirable, which one of the heparins is preferable? A . Unfractionated heparin (as intravenous infusion) B. Fondaparinux C. Danaparoid D. Low-molecular-weight heparin E. No difference—they are all equivalent Answer: A  With a half-life of 1 hour at therapeutic concentration, unfractionated heparin is the heparin that will be eliminated fastest. Low-molecular-weight heparins have half-lives of 2 to 3 hours, and fondaparinux and danaparoid about 20 hours.

3. For which one of the new oral anticoagulants would screening with a thrombin time analysis be sensitive to identify clinically important plasma concentrations of the drug? A . Dabigatran B. Rivaroxaban C. Apixaban D. Edoxaban E. All of the above Answer: A  Dabigatran is a direct thrombin inhibitor, for which a thrombin time is a sensitive test. The others are factor Xa inhibitors, for which thrombin time is not sensitive. 4. A 59-year-old male with hypertension and obesity is seen in the emergency department with subsequent chest pain. The ECG shows ST-elevations and sinus rhythm. How should aspirin best be given—dose and duration—to this patient? A . 81 mg chewable aspirin immediately and then daily for the rest of his life B. 325 mg chewable aspirin immediately and then daily for one year, after which the dose is reduced to 81 mg daily C. 325 mg chewable aspirin immediately and then 81 mg daily for 1 year D. 162 to 325 mg chewable aspirin immediately and then 162 mg daily for 1 year E. 162 to 325 mg chewable aspirin immediately and then 81 mg daily for the rest of his life Answer: E  For patients with non–ST-elevation ACS or ST-elevation MI 162 to 325 mg chewable aspirin should be given immediately. Subsequently, an 81-mg dose gives comparable ischemic protection but lower risk of bleeding compared with a dose of 162 to 325 mg and is therefore preferable for longterm use.

504

CHAPTER 77  Approach to the Patient with Respiratory Disease  

77  APPROACH TO THE PATIENT WITH RESPIRATORY DISEASE MONICA KRAFT

Respiratory symptoms, which are among the most common reasons why patients seek medical care, are responsible for approximately 20% of office visits to a primary care physician. In addition to a careful history, a systematic physical examination is critical for accurate diagnosis. Even in young adults, persistent respiratory symptoms are associated with a greater likelihood of developing chronic lung disease.1 A careful pulmonary examination complements the cardiac physical examination (Chapter 45). Inspection may reveal an elevated jugular pressure, indicative of right heart failure owing to cor pulmonale (Chapter 75). Cervical or supraclavicular adenopathy (Chapter 159) may be the first clue to suggest a thoracic malignancy (Chapter 182) or mycobacterial infection (Chapter 308). Unilateral arm swelling can be caused by venous thrombosis (Chapter 74), whereas venous engorgement of the head and neck can be caused by a tumor that results in superior vena cava syndrome (see Fig. 92-8 in Chapter 92). On the cardiac examination, a loud pulmonic second heart sound is suggestive of pulmonary hypertension, which also can result in a murmur of tricuspid (see Table 45-7 in Chapter 45) or pulmonic valve insufficiency. Inspection of the chest may show hyperinflation and reduced diaphragmatic excursion, typical of chronic obstructive pulmonary disease (COPD; Chapter 82), chest wall abnormalities such as kyphoscoliosis (Chapter 92), or diaphragmatic muscle wall weakness as in many hypoventilation syndromes (Chapter 80). Percussion may reveal dullness in patients with pleural effusions or with lung that has been consolidated by pneumonia. Auscultation of the lungs includes listening at both apices and over both upper and lower lobes, anteriorly and posteriorly, and during inspiration and respiration. Normal lung sounds are heard during inspiration and early expiration as soft and non-musical sounds (Table 77-1).2 Bronchial breath sounds, which sound similar to but often somewhat harsher than normal lung sounds, are heard throughout expiration as well as inspiration, similar to what would be heard by placing a stethoscope over the trachea. The term rales is no longer used and has been replaced by the term crackles. Fine crackles are non-musical and heard typically in late inspiration; they are most commonly a sign of heart failure (Chapter 52) or interstitial lung disease (Chapter 86). By comparison, coarse crackles, which unlike fine crackles tend to be transmitted through the mouth and cleared by coughing, are typical of bronchitis (Chapter 90) and COPD (Chapter 82). Wheezes are high-pitched, musical sounds heard during expiration and sometimes inspiration, most commonly in asthma (Chapter 81) and sometimes in COPD (Chapter 82). When these diseases are severe, however, the degree of airflow may be insufficient to produce wheezes. A rhonchus is a musical, low-pitched sound typically heard in expiration and sometimes during inspiration; it often resolves with coughing. Like coarse crackles, rhonchi are common in bronchitis (Chapter

TABLE 77-1 DIAGNOSTIC UTILITY OF LUNG AUSCULTATION AUSCULTATORY FINDING

CLINICAL CORRELATION

Bronchial breathing

Pneumonia or interstitial lung disease

Fine crackle

Heart failure, interstitial lung disease, alveolar filling disorders

Coarse crackle

Bronchitis

Wheeze

Asthma, COPD

Rhonchus

Bronchitis, COPD

Stridor

Upper-airway obstruction from laryngeal or tracheal inflammation, mass lesions, or external compression

Pleural friction rub

Pleural inflammation or tumors

COPD = chronic obstructive pulmonary disease.

90) and COPD (Chapter 82). A pleural friction rub, which classically occurs during inspiration but sometimes also during expiration, is heard in patients with inflammatory diseases or malignancies involving the pleura (Chapters 92 and 182). Stridor is a musical, high-pitched sound that may be audible without a stethoscope and that indicates upper airway obstruction, such as found with acute inflammatory or chronic degenerative diseases of the larynx (Chapter 401) or obstruction of the trachea, as may be caused by intrathoracic malignant diseases (Chapter 182). An absence of breath sounds would be noted if the lung is not ventilated because of a complete bronchial obstruction or if it is displaced by a pleural effusion. Tactile fremitus, which is a vibratory sensation noted during breathing, is increased in patients who have consolidated lung from pneumonia, because the vibratory sensation conducts better through such lung tissue and is diminished in patients with pleural effusion. Egophony, by which a patient’s recitation of the long E sound is heard on auscultation as a long A sound, is another indication of consolidation typical of pneumonia. Evaluation of the abdomen may show a readily palpable liver, sometimes mistaken for hepatomegaly, in patients with COPD and low diaphragm. Examination of the extremities may reveal cyanosis in patients who are hypoxemic, usually with a partial pressure of oxygen less than 55 mm Hg, although it also may be observed in patients with methemoglobinemia (Chapter 149). Clubbing (Chapter 45) is indicative of chronic hypoxemia, as seen in patients with chronic right-to-left-shunting from congenital heart disease (Chapter 61) or other causes of long-standing hypoxemia (Chapters 82 and 86), but it also may be indicative of pleural-based diseases (Chapter 92) as part of the syndrome of hypertrophic pulmonary osteoarthropathy (Chapters 169 and 259). In patients with suspected hypoxemia, careful analyses of arterial blood gases can help determine its severity and guide therapy (Chapter 95). In patients in whom it is difficult to distinguish heart failure from a pulmonary cause of hypoxemia, an elevated brain natriuretic peptide level may point to a cardiac cause (Chapter 52). Chest imaging (Chapter 78) is a crucial part of the evaluation of many potential pulmonary complaints, and pulmonary function testing (Chapter 79) can be extremely helpful in distinguishing among causes of acute and chronic lung disease. Among the most common respiratory complaints are cough, wheezing, dyspnea, and hemoptysis. Each can and should be approached in a systematic way.

  APPROACH TO THE PATIENT WITH COUGH

Cough is the single most common respiratory complaint for which patients seek care. Referrals of patients with persistently troublesome chronic cough of unknown cause account for 10 to 38% of outpatient visits to respiratory specialists. For acute cough, defined as coughing that has been present for less than 8 weeks, a careful medical history and physical examination will usually reveal the diagnosis3 (Table 77-2). Although most acute coughs are of minor consequence, cough can occasionally be a sign of a potentially life-threatening illness, such as pulmonary embolism (Chapter 74), pneumonia (Chapter 91), or heart failure (Chapter 52). Up to 98% of all cases of chronic cough, defined as a cough that persists for more than 8 weeks, in immunocompetent adults are caused by eight common conditions: postnasal drip syndrome from a variety of rhinosinus conditions often termed the “upper airway cough syndrome” (UACS, asthma (Chapter 81), gastroesophageal reflux disease (GERD) (Chapter 129), chronic bronchitis (Chapter 82), eosinophilic bronchitis, bronchiectasis (Chapter 84), use of angiotensin-converting enzyme (ACE) inhibitors, and postinfectious cough. Postinfectious cough is usually nonproductive and generally persists for 3 to 8 weeks after an upper respiratory tract infection; patients have a normal chest radiograph. Environmental and occupational factors can also contribute to chronic cough and should be assessed.4 Cough can also be seen in athletes, which requires a different diagnostic approach.5 Uncommon causes of chronic cough include bronchogenic carcinoma (Chapter 182), chronic interstitial pneumonia (Chapter 86), sarcoidosis (Chapter 89), left ventricular failure (Chapter 52), and aspiration (Chapter 88).  

DIAGNOSIS

In chronic cough (Fig. 77-1), the character and timing are not of diagnostic help. A chest radiograph should be obtained in all patients, but other tests should not be ordered in current smokers or patients taking ACE inhibitors until the response to smoking cessation or discontinuation of the drug for at least 4 weeks can be assessed. Sinus radiographs, barium esophagography,

CHAPTER 77  Approach to the Patient with Respiratory Disease  

ABSTRACT

Respiratory symptoms, which are among the most common reasons why patients seek medical care, are responsible for approximately 20% of office visits to a primary care physician. In addition to a careful history, a systematic physical examination is critical for accurate diagnosis and subsequent evaluation. Among the most common respiratory complaints are cough, wheezing, dyspnea, and hemoptysis. Each can and should be approached in a systematic way. This chapter discusses the necessary history, differential diagnosis, and evaluation to determine a diagnosis and guide therapy for these very common complaints.

KEYWORDS

cough wheeze dyspnea hemoptysis

504.e1

505

CHAPTER 77  Approach to the Patient with Respiratory Disease  

TABLE 77-2 SPECTRUM OF CAUSES AND FREQUENCIES OF COUGH IN IMMUNOCOMPETENT ADULTS COMMON

LESS COMMON

TABLE 77-3 TESTING CHARACTERISTICS OF DIAGNOSTIC PROTOCOL FOR EVALUATION OF CHRONIC COUGH

ACUTE COUGH

POSITIVE PREDICTIVE VALUE, %

NEGATIVE PREDICTIVE VALUE, %

Common cold

Asthma exacerbation

TESTS

Acute bacterial sinusitis

Pneumonia

Pertussis

Heart failure

Sinus radiograph

Sinusitis

57-81

95-100

Exacerbations of COPD

Aspiration syndromes

Methacholine inhalation challenge

Asthma

60-82

100

Allergic rhinitis

Pulmonary embolism

GERD, esophageal stricture

63-93

Exacerbation of bronchiectasis

Modified barium esophagography

38-63

Environmental irritant rhinitis

Esophageal pH*

GERD

89-100

Bronchoscopy

Endobronchial mass/lesion

50-89

SUBACUTE COUGH Postinfectious cough Pertussis

DIAGNOSIS

100

*24-Hour esophageal pH monitoring. GERD = gastroesophageal reflux disease.

CHRONIC COUGH Rhinosinus conditions/UACS

Bronchiolitis

Asthma

Occupational and environmental

Non-asthma eosinophilic bronchitis

Aspiration Bronchogenic carcinoma

Bronchiectasis

Drug-induced cough

Gastroesophageal reflux

Tuberculosis

Chronic bronchitis/COPD

Interstitial lung disease



COPD = chronic obstructive pulmonary disease; UACS = upper airway cough syndrome.

methacholine challenge, esophageal pH, and bronchoscopy can be ordered as part of the initial evaluation, depending on the history and physical examination (Table 77-3; see Fig. 77-1). If a test points toward a possible diagnosis, a trial of treatment for that condition is needed to confirm the diagnosis. The diagnosis of psychogenic cough is often a diagnosis of exclusion.6

TREATMENT  The specific cause of cough can be diagnosed and treated successfully 84 to 98% of the time, so nonspecific therapy aimed to suppress the cough per se is rarely indicated.7 There is no strong evidence that nonspecific therapies such as antitussives, mucolytics, decongestants, or antihistamine-decongestant combinations are efficacious for acute cough in the setting of an upper respiratory tract infection. A1  For nonspecific persistent cough,8 effective treatment of chronic gastroesophageal reflux disease with a proton pump inhibitor (Chapter 129)9 provides no more than modest benefit, with approximately one in five patients improving. Inhaled corticosteroids can reduce cough but should be used only after evaluation by chest radiography and often spirometry. A2  Dextromethorphan and codeine-containing cough suppressants can reduce chronic cough by approximately 40%. In adults with refractory chronic cough without active respiratory disease or infection, gabapentin (up to a maximum daily dose of 1800 mg) or pregabalin (300 mg daily) significantly improves coughspecific quality of life compared with placebo. A3  For chronic refractory cough despite comprehensive evaluation and opioid therapy, a combination of education, speech and language intervention, which includes cough suppression techniques, and counseling can significantly reduce cough and its negative impact on quality of life. A4  Coughing can also be reduced by training patients to focus externally rather than internally.

  APPROACH TO THE PATIENT WITH WHEEZING

Wheeze is a continuous musical sound that lasts longer than 80 to 100 msec, likely generated by flow through critically narrowed collapsible bronchi. Although expiratory wheezing is a common physical finding in asthma (Chapter 81), the many causes of wheezing (Table 77-4) (e.g., COPD [Chapter 82], pulmonary edema [Chapter 52], bronchiolitis [Chapter 86], bronchiectasis [Chapter 84], and less common entities such as carcinoid [Chapter 219] and parasitic infections) often can be distinguished based on the history, physical examination, and pulmonary function testing (Chapter 79).

DIAGNOSIS

On pulmonary function testing, the shape of inspiratory and expiratory flowvolume loops provide key information about the presence of airway obstruction and whether the obstruction is extrathoracic or intrathoracic (E-Fig. 77-1). An important cause of extrathoracic obstruction is vocal cord lesions (Chapter 181). Variable intrathoracic obstruction can be caused by tracheomalacia, whereas fixed upper airway obstruction can be caused by a proximal tracheal tumor.

TREATMENT  Treatment of the specific cause will usually lead to complete or at least partial resolution of wheezing. However, treatment of associated asymptomatic or minimally symptomatic gastroesophageal reflux disease is not beneficial.

  APPROACH TO THE PATIENT WITH DYSPNEA

Dyspnea is the sensation of difficult, labored, or unpleasant breathing. The word unpleasant is very important to this definition because the labored or difficult breathing encountered by healthy individuals while exercising does not qualify as dyspnea because it is at the level expected for the degree of exertion. The sensation of dyspnea is often poorly or vaguely described by the patient. The physiology of dyspnea remains unclear, but multiple neural pathways can be involved in processes that lead to dyspnea. In acute dyspnea, or shortness of breath of sudden onset, the history, physical examination, and laboratory testing must first focus on potential life-threatening conditions, including pulmonary embolism (Chapter 74), pulmonary edema (Chapters 52 and 53), acute airway obstruction from anaphylaxis or foreign bodies, pneumothorax (Chapter 92), or pneumonia (Chapter 91). For chronic dyspnea, specific conditions to consider include COPD (Chapter 82), asthma (Chapter 81), interstitial lung disease (Chapter 86), heart failure (Chapter 52), cardiomyopathy (Chapter 54), GERD (Chapter 129), other respiratory diseases, or hyperventilation syndrome (Table 77-5).10  

DIAGNOSIS

In addition to an appropriate history and physical examination, a chest radiograph, electrocardiogram (ECG), pulmonary function testing, and an exercise test with electrocardiographic monitoring and pulse oximetry at rest and during exercise are key tests to assess patients with unexplained dyspnea (Fig. 77-2). For acute dyspnea, B-type natriuretic peptide testing can be extremely helpful in distinguishing heart failure from other causes.11 The utility of more detailed pulmonary testing with maximal inspiratory and expiratory pressures, flowvolume loops, with or without methacholine challenge (Chapter 79), computed tomographic screening of the chest, and echocardiography depends on history and physical examination and the results of these tests. When GERD is a suspected cause of dyspnea, a modified barium esophagogram or 24-hour esophageal pH monitoring, or both, should be considered (Chapter 129). Other more invasive tests such as cardiac catheterization or lung biopsy may be indicated when the results of less invasive tests have not been conclusive.12

CHAPTER 77  Approach to the Patient with Respiratory Disease V (L/S) Inspiration Expiration





A

7 6 5 4 3 2 1 0 1 2 3 4 5 6 100

B

0100

C

0100 0100 Vital capacity (%)

D

E

0 100

0

E-FIGURE 77-1.  Schematic flow-volume loop configurations in a spectrum of airway lesions. A is normal; B is variable extrathoracic upper airway obstruction; C is variable intrathoracic upper airway lesion; D is fixed upper airway obstruction; and E is small airway obstruction. L/S = liters per second; = ventilation

505.e1

506

CHAPTER 77  Approach to the Patient with Respiratory Disease  

Chronic cough

Investigate and treat

A cause of cough is suggested

Inadequate response to optimal Rx

History, physical examination, chest radiograph

Smoking ACE-inhibitor

Discontinue

No response Post-nasal drip/rhinitis/sinusitis Empiric treatment (Chapter 398) with anti-histamine/decongestant, nasal saline irrigation Asthma (Chapter 81) Evaluate via spirometry, bronchodilator reversibility, methacholine challenge; then treat with inhaled corticosteroids, beta-adrenergic inhalers, leukotriene receptor antagonists (Chapter 81); Empiric treatment as a second option Gastroesophageal Reflux Disease (GERD) Empiric treatment (Chapter 129) with protein pump inhibitor, diet/lifestyle

Inadequate response to optimal Rx Further investigations to consider if empiric treatments partially effective or ineffective (see Table 77-3 for testing regarding specific diagnoses): • 24h esophageal pH monitoring • endoscopic or videofluoroscopic swallow evaluation • barium esophagram • sinus imaging • HRCT • bronchoscopy • echocardiogram • environmental assessment • polysomnogram

Important general considerations Optimize therapy for each diagnosis Check adherence with medications Due to the possibility of multiple causes, maintain all partially effective treatments

FIGURE 77-1.  Algorithm for the management of chronic cough lasting longer than 8 weeks. ACE = angiotensin-converting enzyme; HRCT = high-resolution computed tomography; Rx = prescription.

TABLE 77-4 DIAGNOSIS OF SELECTED WHEEZING ILLNESSES OTHER THAN ASTHMA DISEASES

DISTINGUISHING FEATURES

UPPER AIRWAY DISEASES Postnasal drip syndrome

History of postnasal drip, throat clearing, nasal discharge; physical examination shows oropharyngeal secretions or cobblestone appearance to mucosa.

Epiglottitis

History of sore throat out of proportion to pharyngitis. Evidence of supraglottitis on endoscopy or lateral neck radiographs.

Vocal cord dysfunction syndrome

Lack of symptomatic response to bronchodilators, presence of stridor plus wheeze in absence of increased P(A-a)o2; extrathoracic variable obstruction on flow-volume loops; paradoxical inspiratory and/or early expiratory adduction of vocal cords on laryngoscopy during wheezing. This syndrome can masquerade as asthma, be provoked by exercise, and often coexists with asthma.

Retropharyngeal abscess

History of stiff neck, sore throat, fever, trauma to posterior pharynx; swelling noted by lateral neck or CT radiographs.

Laryngotracheal injury due to tracheal cannulation

History of cannulation of the trachea by endotracheal or tracheostomy tube; evidence of intrathoracic or extrathoracic variable obstruction on flow-volume loops, neck and chest radiographs, laryngoscopy, or bronchoscopy.

Neoplasms

Bronchogenic carcinoma, adenoma, or carcinoid tumor is suspected when there is hemoptysis, unilateral wheeze, or evidence of lobar collapse on chest radiograph or combinations of these; diagnosis is confirmed by bronchoscopy.

Anaphylaxis

Abrupt onset of wheezing with urticaria, angioedema, nausea, diarrhea, and hypotension, especially after insect bite, in association with other signs of anaphylaxis such as hypotension or hives, or administration of drug or IV contrast, or family history.

LOWER AIRWAY DISEASES COPD

History of dyspnea on exertion and productive cough in cigarette smoker. Because productive cough is nonspecific, it should only be ascribed to COPD when other cough-phlegm syndromes have been excluded, forced expiratory time to empty more than 80% of vital capacity is >4 sec, and there is decreased breath sound intensity, unforced wheezing during auscultation, and irreversible, expiratory airflow obstruction on spirometry.

Pulmonary edema

History and physical examination consistent with passive congestion of the lungs, ARDS, impaired lung lymphatics; abnormal chest radiograph, echocardiogram, radionuclide ventriculography, cardiac catheterization, or combinations of these.

Aspiration

History of risk for pharyngeal dysfunction or gastroesophageal reflux disease; abnormal modified barium swallow, 24-hr esophageal pH monitoring, or both.

CHAPTER 77  Approach to the Patient with Respiratory Disease  

507

TABLE 77-4 DIAGNOSIS OF SELECTED WHEEZING ILLNESSES OTHER THAN ASTHMA—cont’d DISEASES

DISTINGUISHING FEATURES

Pulmonary embolism

History of risk for thromboembolic disease, positive confirmatory tests.

Bronchiolitis

History of respiratory infection, connective tissue disease, transplantation, ulcerative colitis, development of chronic airway obstruction over months to a few years rather than over many years in a nonsmoker; mixed obstructive and restrictive pattern on PFTs and hyperinflation; may be accompanied by fine nodular infiltrates on chest radiograph.

Cystic fibrosis

Combination of productive cough, digital clubbing, bronchiectasis, progressive COPD with Pseudomonas sp colonization and infection, obstructive azoospermia, family history, pancreatic insufficiency, and two sweat chloride determinations of >60 mEq/L; some patients are not diagnosed until adulthood, in one instance as late as age 69 yr; when sweat test is occasionally normal, definitive diagnosis may require nasal transepithelial voltage measurements and genotyping.

Carcinoid syndrome

History of episodes of flushing and watery diarrhea; elevated 5-hydroxyindoleactic acid level in 24-hr urine specimen.

Bronchiectasis

History of episodes of productive cough, fever, or recurrent pneumonias; suggestive chest radiographs or typical chest CT findings; ABPA should be considered when bronchiectasis is central.

Lymphangitic carcinomatosis

History of dyspnea or prior malignancy; reticulonodular infiltrates with or without pleural effusions; suggestive high-resolution chest CT scan; confirmed by bronchoscopy with biopsies.

Parasitic infections

Consider in a nonasthmatic patient who has traveled to an endemic area and complains of fatigue, weight loss, fever; peripheral blood eosinophilia; infiltrates on chest radiograph; stools for ova and parasites for nonfilarial causes; blood serologic studies for filarial causes.

ABPA = allergic bronchopulmonary aspergillosis; ARDS = acute respiratory distress syndrome; COPD = chronic obstructive pulmonary disease; CT = computed tomography; IV = intravenous; P(A-a)o2 = alveolar-arterial oxygen tension gradient; PFTs = pulmonary function tests.

TABLE 77-5 DISEASES THAT CAUSE DYSPNEA GROUPED BY PHYSIOLOGICAL MECHANISMS OF ACTION* INCREASED RESPIRATORY DRIVE Stimulation of Chemoreceptors Conditions leading to acute hypoxemia Impaired gas exchanger (e.g., asthma, pulmonary embolism, pneumonia, congestive heart failure†) Environmental hypoxia (e.g., altitude, contained space with fire) Conditions leading to increased dead space, acute hypercapnia Impaired gas exchanger (e.g., acute, severe asthma; exacerbation of COPD; severe pulmonary edema) Impaired ventilator pump (e.g., muscle weakness, airflow obstruction) Metabolic acidosis Renal disease (e.g., renal failure, renal tubular acidosis) Decreased oxygen carrying capacity (e.g., anemia) Decreased release of oxygen to tissues (e.g., hemoglobinopathy) Decreased cardiac output Stimulation of Pulmonary Receptors (Irritant, Mechanical, Vascular)‡ Interstitial lung disease Pleural effusion (compression atelectasis) Pulmonary vascular disease (e.g., thromboembolism, idiopathic pulmonary hypertension) Heart failure Mild asthma Behavioral Factors Hyperventilation syndrome, anxiety disorders, panic attacks VENTILATORY PUMP: INCREASED EFFORT OR WORK OF BREATHING Muscle Weakness Myasthenia gravis, Guillain-Barré syndrome, spinal cord injury, myopathy, postpoliomyelitis syndrome Decreased compliance of the chest wall Severe kyphoscoliosis, obesity, pleural effusion Airflow Obstruction (Including increased resistive load from narrowing of the airways and increased elastic load from hyperinflation) Asthma, COPD, laryngospasm, aspiration of foreign body, bronchitis *Some diseases appear in more than one category, because they act via several physiologic mechanisms. † Heart failure includes both systolic and diastolic dysfunction. Systolic dysfunction may produce dyspnea at rest and with activity. Diastolic dysfunction typically leads to symptoms primarily with exercise. In addition to the mechanisms noted above, systolic heart failure may also produce dyspnea via metaboreceptors, which are postulated to exist in muscles and be stimulated by changes in the metabolic milieu when oxygen delivery does not meet oxygen demand. ‡ These conditions probably produce dyspnea by a combination of increased ventilator drive and primary sensory input from the receptors. COPD = chronic obstructive pulmonary disease.

TREATMENT  Whenever possible, the final determination of the cause of dyspnea is made by observing which specific therapy eliminates it. Dyspnea may be simultaneously the result of more than one condition, each of which may need to be treated. In cases of refractory dyspnea despite maximally treated chronic heart and lung disease, judicious use of opioids can curb air hunger.13

  APPROACH TO THE PATIENT WITH HEMOPTYSIS

Hemoptysis is the expectoration of blood from the lung parenchyma or airways. Hemoptysis may be scant, with just the appearance of streaks of bright red blood in the sputum, or massive, with the expectoration of a large volume of blood. Massive hemoptysis, which is defined as the expectoration of at least 600 mL of blood in 24 to 48 hours, may occur in 3 to 10% of patients with hemoptysis. Dark red clots also may be expectorated when the blood has been present in the lungs for days. Pseudohemoptysis, which is the expectoration of blood from a source other than the lower respiratory tract, may cause diagnostic confusion when patients cannot clearly describe the source of the bleeding. Pseudohemoptysis can occur when blood from the oral cavity, nares, pharynx, or tongue clings to the back of the throat and initiates the cough reflex, or when patients who have hematemesis aspirate into the lower respiratory tract. When the oropharynx is colonized with Serratia marcescens, a red-pigment–producing aerobic gram-negative rod, the sputum can also be red and be confused with hemoptysis. Hemoptysis can be caused by a wide variety of disorders. Virtually all causes of hemoptysis (Table 77-6) may result in massive hemoptysis, but massive hemoptysis is most frequently caused by infection (e.g., tuberculosis [Chapter 308], bronchiectasis, lung abscess [Chapter 84], and cancer [Chapter 182]). Infections with aspergilloma (Chapter 319) and in patients with cystic fibrosis (Chapter 83) also are associated with massive hemoptysis. Iatrogenic causes of massive hemoptysis include rupture of a pulmonary artery after less than 0.2% of cases of balloon-guided flotation catheterization and tracheal artery fistula as a complication of tracheostomy. In nonmassive hemoptysis, the cause is bronchitis in more than one third of cases (Chapter 90), bronchogenic carcinoma (Chapter 182) in one fifth of cases, tuberculosis (Chapter 308) in 7%, pneumonia (Chapter 91) in 5%, and bronchiectasis in 1% (Chapter 84). Using a systematic diagnostic approach (see later), the cause of hemoptysis can be found in 68 to 98% of cases. The remaining 2 to 32% have idiopathic or central hemoptysis, which occurs most commonly in men between 30 and 50 years of age. Prolonged follow-up of idiopathic hemoptysis almost always fails to reveal the

508

CHAPTER 77  Approach to the Patient with Respiratory Disease  

Evaluation of Patients with Chronic Dyspnea Patient with suspected chronic dyspnea

Conduct detailed history and physical examination. Conduct appropriate level 1 testing as needed to confirm diagnosis. Is the diagnosis evident? Yes

Possible diagnoses: Asthma Chronic obstructive pulmonary disease Heart failure Pleural effusion Anemia Kyphoscoliosis

No Level 1: Complete blood count Metabolic profile Chest radiograph Electrocardiogram Spirometry Pulse Oximetry

Conduct appropriate Level 2 testing

Is the diagnosis evident?

Yes

Level 2: Echocardiogram Brain natriuretic peptide Pulmonary function testing Arterial blood gas High-resolution computed tomography Holter monitor Radionuclide study Ventilation-perfusion (V/Q) scan Exercise treadmill testing

No

Possible diagnoses: Chronic pulmonary embolism Restrictive lung disease Interstitial lung disease Pericardial disease Heart failure Valvular heart disease Coronary artery disease Cardiac dysrhythmia

Conduct appropriate level 3 testing (specialty consultation for these tests) Is the diagnosis evident?

Yes Possible diagnoses: Gastroesophageal reflux disease Primary pulmonary hypertension Coronary artery disease Deconditioning

No Consider: Psychogenic dyspnea Specialty consultation

Level 3: Bronchoscopy Esophageal pH probe testing Lung biopsy Cardiac catheterization Cardiopulmonary exercise testing Bronchoscopy Esophageal pH probe testing Lung biopsy

FIGURE 77-2.  Algorithm outlining the approach to chronic dyspnea. (Modified from Karnani NG, Reisfield GM, Wilson GR. Evaluation of chronic dyspnea. Am Fam Phys. 2005;71:1529-1537.)

TABLE 77-6 COMMON CAUSES OF HEMOPTYSIS Cardiovascular Arteriovenous malformation Congenital heart disease Pulmonary embolism (fat, septic, thrombotic) Heart failure, especially from mitral stenosis Pulmonary vascular disease Pulmonary veno-occlusive disease Pulmonary artery rupture following catheterization Tricuspid endocarditis Pulmonary Infection Anthrax Lung abscess Mycetoma/fungal infection Necrotizing pneumonia Parasitic (e.g., Paragonimus westermani) Tuberculosis or nontuberculous mycobacterial disease Tularemia Viral (e.g., Herpes simplex) Yersinia pestis (plague) Rheumatic Disease Amyloid Anti-glomular basement membrane disease (Goodpasture) Behçet disease Genetic collagen defect (Ehlers-Danlos) Granulomatosis with polyangiitis Idiopathic pulmonary hemosiderosis Primary anti-phospholipid syndrome Systemic lupus erythematosus

Tracheobronchial/Airway Diseases Bronchogenic carcinoma Bronchiectasis, including cystic fibrosis Bronchitis, acute and chronic Bullous emphysema Broncholithiasis Bronchovascular fistula Dieulafoy disease (subepithelial bronchial artery) Foreign body Metastatic cancer to bronchus or trachea Drugs and Toxins Argemone alkaloid (e.g., contaminated cooking oil) Bevacizumab Cocaine use Nitrogen dioxide toxicity Trimellitic anhydride Trauma Blunt chest trauma (bronchial rupture, lung contusion) Penetrating lung injury Iatrogenic Airway stent Aortobronchial fistula due to aortic graft or stent Erosion of tracheal tube into innominate artery Transthoracic needle aspiration Vascular injury from pulmonary artery catheter Miscellaneous and Rare Causes Systemic coagulopathy or thrombolytic agents Platelet dysfunction/antiplatelet medications/thrombocytopenia von Willebrand disease Catamenial hemoptysis (pulmonary endometriosis) Leukemia/bone marrow transplant

509

CHAPTER 77  Approach to the Patient with Respiratory Disease  

source of bleeding, even though 10% continue to have occasional episodes of hemoptysis.  

DIAGNOSIS

The diagnostic evaluation for hemoptysis begins with a detailed medical history and a complete physical examination.14 Information on the amount of bleeding should be obtained, as well as details about the frequency, timing, and duration of hemoptysis. For example, repeated episodes of hemoptysis occurring over a period of months to years suggest a bronchial adenoma or bronchiectasis as the cause, whereas small amounts of hemoptysis occurring every day for weeks are more likely to be caused by bronchogenic carcinoma. A travel history can suggest coccidioidomycosis (Chapter 316) and histoplasmosis (Chapter 316) in the United States, paragonimiasis and ascariasis (Chapter 334) in the Far East, and schistosomiasis (Chapter 334) in South America. Orthopnea and paroxysmal nocturnal dyspnea suggest heart failure (Chapter 52), especially from mitral stenosis (Chapter 66). In patients who have occupational exposure to trimellitic anhydride, which occurs when heated metal surfaces are sprayed with a corrosion-resistant epoxy resin, hemoptysis can be part of the postexposure syndrome. In a patient with the triad of upper airway disease, lower airway disease, and renal disease, granulomatosis with polyangiitis (Chapter 254) should be suspected. Pulmonary hemorrhage also may be a presenting manifestation of systemic lupus erythematosus (Chapter 250). Goodpasture syndrome, which typically occurs in young men, is also associated with renal disease (Chapter 113). Diffuse alveolar hemorrhage occurs in 20% of cases during autologous bone marrow transplantation (Chapter 168) and should be suspected in patients who have undergone recent bone marrow transplantation when they present with cough, dyspnea, hypoxemia, and diffuse pulmonary infiltrates. On physical examination, inspection of the skin and mucous membranes may show telangiectasias suggesting hereditary hemorrhagic telangiectasia (Chapter 164) or ecchymoses and petechiae, suggesting a hematologic abnormality (Chapter 163). Pulsations transmitted to a tracheostomy cannula should

heighten suspicion of a tracheal artery fistula. Inspection of the thorax should show evidence of recent or old chest trauma, and unilateral wheeze or crackles may herald localized disease such as a bronchial adenoma or carcinoma. Although pulmonary embolism (Chapter 74) cannot be definitively diagnosed on physical examination, tachypnea, phlebitis, and pleural friction rub suggest this disorder. If crackles are heard on the chest examination, heart failure as well as other diseases causing diffuse pulmonary hemorrhage (see earlier) or idiopathic pulmonary hemosiderosis (Chapter 86) should be considered. Careful cardiovascular examination may help diagnose mitral stenosis (Chapter 66), pulmonary artery fistulas, or pulmonary hypertension (Chapter 75). Routine laboratory studies should include a complete blood count, urinalysis, and coagulation studies. The complete blood count may suggest an infection, hematologic disorder, or chronic blood loss. Urinalysis may reveal hematuria and suggest the presence of a systemic disease (e.g., Wegener granulomatosis, Goodpasture syndrome, systemic lupus erythematosus) associated with renal disease. Coagulation studies may uncover a hematologic disorder that is primarily responsible for hemoptysis or that contributes to excessive bleeding from another disease. The ECG may help suggest the presence of a cardiovascular disorder. Although as many as 30% of patients with hemoptysis have a normal chest radiograph, a routine chest radiograph is the beginning of the diagnostic process (Fig. 77-3). Bronchoscopy can localize the bleeding site in up to 93% of patients by fiberoptic bronchoscopy and in up to 86% with rigid bronchoscopy. It may establish sites of bleeding different from those suggested by the chest radiograph. The best results are obtained when bronchoscopy is performed during or within 24 hours of active bleeding, and rates of diagnosis fall to approximately 50% by 48 hours after bleeding. When there is no active bleeding, bronchoscopy with bronchoalveolar lavage can be helpful in patients thought to have diffuse intrapulmonary hemorrhage. Typical findings include bright red or blood-tinged lavage fluid from multiple lobes in both lungs or a substantial number of hemosiderin-laden macrophages (i.e., at least 20% of the total number of alveolar macrophages).

Evaluation of Non-Massive Hemoptysis History and physical examination

Exclude pseudohemoptysis and hematemesis

Chest radiography

Infiltrate

Mass

Antibiotics

Chest CT

Resolution

Repeat chest radiography in 6-8 weeks

Normal

No resolution

Chest CT, pulmonary consult

Bronchoscopy; pulmonary consult

Other parenchymal disease Chest CT

No specific diagnosis

Specific diagnosis

Condition-specific evaluation and treatment

Normal Consider oral antibiotics

No further evaluation

Recurrence

Chest CT

Abnormal

Chest CT FIGURE 77-3.  Algorithm for evaluation of non-massive hemoptysis. CT = computed tomographic scan. (Adapted from Earwood JS, Thompson TD. Hemoptysis: evaluation and management. Am Acad Fam Physicians. 2015;91;243-249.)

TABLE 77-7 EXAMPLES OF SPECIAL EVALUATIONS FOR HEMOPTYSIS ACCORDING TO CATEGORY OF DISEASE* TRACHEOBRONCHIAL DISORDERS Expectorated sputum for TB, parasites, fungi, and cytology Bronchoscopy (if not done) High-resolution chest CT scan LOCALIZED PARENCHYMAL DISEASES Expectorated sputum for TB, parasites, fungi, and cytology Chest CT scan Lung biopsy with special stains DIFFUSE PARENCHYMAL DISEASES Expectorated sputum for cytology Blood for BUN, creatinine, ANA, RF, complement, cryoglobulins, ANCA, anti-GBM antibody High-resolution chest CT scan Lung or kidney biopsy with special stains CARDIOVASCULAR DISORDERS Echocardiogram Arterial blood gas on 21% and 100% oxygen Ventilation-perfusion scans Chest CT with contrast Aortogram, HEMATOLOGIC DISORDERS Coagulation studies Bone marrow *This table is not meant to be all inclusive. ANA = antinuclear antibody; ANCA = antineutrophil cytoplasmic antibody; BUN = blood urea nitrogen; CT = computed tomography; GBM = glomerular basement membrane; RF = rheumatoid factor; TB = tuberculosis.

Depending on the results of the initial evaluation and the likely categories of hemoptysis, additional diagnostic tests can be helpful (Table 77-7). Bronchoscopy may not be needed in patients who have stable chronic bronchitis (Chapter 82) with one episode of blood streaking or who have acute tracheobronchitis (Chapter 82). Bronchoscopy also may not be needed with obvious cardiovascular causes of hemoptysis, such as heart failure and pulmonary embolism.

TREATMENT  Treatment is targeted toward the cause of hemoptysis. Bronchoscopic approaches (Chapter 93) are increasingly used for endobronchial lesions as is bronchial artery embolization for the emergency treatment of hemoptysis and the treatment of hemoptysis associated with bronchiectasis.15

  Grade A References A1. Smith SM, Schroeder K, Fahey T. Over-the-counter (OTC) medications for acute cough in children and adults in community settings. Cochrane Database Syst Rev. 2014;11:CD001831. A2. Johnstone KJ, Chang AB, Fong KM, et al. Inhaled corticosteroids for subacute and chronic cough in adults. Cochrane Database Syst Rev. 2013;3:CD009305. A3. Vertigan AE, Kapela SL, Ryan NM, et al. Pregabalin and speech pathology combination therapy for refractory chronic cough: a randomized controlled trial. Chest. 2016;149:639-648. A4. Chamberlain Mitchell SA, Garrod R, Clark L, et al. Physiotherapy, and speech and language therapy intervention for patients with refractory chronic cough: a multicentre randomised control trial. Thorax. 2017;72:129-136.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 77  Approach to the Patient with Respiratory Disease  

GENERAL REFERENCES 1. Kalhan R, Dransfield MT, Colangelo LA, et al. Respiratory symptoms in young adults and future lung disease. The CARDIA lung study. Am J Respir Crit Care Med. 2018;197:1616-1624. 2. Sarkar M, Madabhavi I, Niranjan N, et al. Auscultation of the respiratory system. Ann Thorac Med. 2015;10:158-168. 3. Irwin RS, French CL, Chang AB, et al. Classification of cough as a symptom in adults and management algorithms: CHEST guideline and expert panel report. Chest. 2018;153:196-209. 4. Tarlo SM, Altman KW, Oppenheimer J, et al. Occupational and environmental contributions to chronic cough in adults: chest expert panel report. Chest. 2016;150:894-907. 5. Boulet LP, Turmel J, Irwin RS. Cough in the athlete: CHEST guideline and expert panel report. Chest. 2017;151:441-454. 6. Vertigan AE. Somatic cough syndrome or psychogenic cough—what is the difference? J Thorac Dis. 2017;9:831-838. 7. Smith JA, Woodcock A. Chronic cough. N Engl J Med. 2016;375:1544-1551. 8. Gibson P, Wang G, McGarvey L, et al. Treatment of unexplained chronic cough: CHEST guideline and expert panel report. Chest. 2016;149:27-44.

510.e1

9. Kahrilas PJ, Altman KW, Chang AB, et al. Chronic cough due to gastroesophageal reflux in adults: CHEST guideline and expert panel report. Chest. 2016;150:1341-1360. 10. Hale ZE, Singhal A, Hsia RY. Causes of shortness of breath in the acute patient: a national study. Acad Emerg Med. 2018;25:1227-1234. 11. Januzzi JL Jr, Chen-Tournoux AA, Christenson RH, et al. N-terminal pro-B-type natriuretic peptide in the emergency department: the ICON-RELOADED study. J Am Coll Cardiol. 2018;71: 1191-1200. 12. Huang W, Resch S, Oliveira RK, et al. Invasive cardiopulmonary exercise testing in the evaluation of unexplained dyspnea: insights from a multidisciplinary dyspnea center. Eur J Prev Cardiol. 2017;24:1190-1199. 13. Takahashi K, Kondo M, Ando M, et al. Effects of oral morphine on dyspnea in patients with cancer: response rate, predictive factors, and clinically meaningful change (CJLSG1101). Oncologist. 2019. [Epub ahead of print.] 14. Gavelli F, Patrucco F, Statti G, et al. Mild-to-moderate hemoptysis: a diagnostic and clinical challenge. Minerva Med. 2018;109:239-247. 15. Panda A, Bhalla AS, Goyal A. Bronchial artery embolization in hemoptysis: a systematic review. Diagn Interv Radiol. 2017;23:307-317.

510.e2

CHAPTER 77  Approach to the Patient with Respiratory Disease  

REVIEW QUESTIONS 1. Which of the following is true about eosinophilic bronchitis? A . The diagnosis improves with bronchodilator therapy. B. It is not associated with airways hyperresponsiveness. C. Sputum eosinophils are absent. D. The cough is not responsive to inhaled corticosteroids. E. Hemoptysis has been present on two occasions. Answer: B  Eosinophilic bronchitis is associated with eosinophilic infiltration of airway tissue but is it NOT associated with airways hyperresponsiveness noted in asthma. It responds to inhaled corticosteroids. 2. A 53-year-old woman returns for reevaluation of cough of one year in duration. The cough is not productive, occurs during the day and night, and is triggered by talking, laughing, and cold air. Her pulmonary function tests are normal. The cough has not responded to appropriate therapy for asthma with inhaled corticosteroids and short-acting β-agonists. She has a history of hypertension, hyperlipidemia, cigarette smoking (10 pack years, currently smokes half a pack per day). She denies any history of lung disease, atopy, rhinitis, gastroesophageal reflux disease, dysphagia, or obstructive sleep apnea. The physical examination reveals bibasilar crackles and trace pedal edema. Her cardiac and abdominal examinations were unremarkable. Further evaluation including esophageal impedance and manometry was unrevealing. What is the appropriate next step in this patient’s management? A . High-resolution computed tomography of the chest B. Computed tomography of the sinuses C. Bronchoscopy D. Echocardiogram E. Video swallowing study Answer: D  With a cough that is associated with bibasilar crackles and pedal edema, a cardiac cause is likely. A brain natriuretic peptide level should be checked, but an echocardiogram will be necessary to determine cardiac function and to exclude valvular abnormalities that may have been missed on physical examination. A chest radiograph would be warranted before chest computed tomography to exclude other causes of crackles, such as interstitial lung disease. The patient has no history of atopy or rhinitis, so sinus computed tomography is not indicated. A cough due to aspiration is not usually associated with pedal edema, and she has no history of dysphagia, so video swallow study would not be the next step in the evaluation. While cardiac causes are being considered, bronchoscopy would not be an appropriate choice at this point. 3. A 60-year-old man presents with dyspnea on exertion beginning approximately 2 months ago. He denies chest pain, cough, and wheezing but does admit to occasional chest tightness, particularly at night. He has no history of cardiovascular disease, lung disease or diabetes. He has a 30 pack-year history of cigarette smoking but quit five years ago. He takes medication daily for hypertension and gastroesophageal reflux disease. He does not exercise regularly. His BMI is 36 kg/m2. On physical examination his blood pressure is 140/88, and his other vital signs and examination are normal. Pulmonary function tests reveal a reduced FEV1 and preserved FVC, with an FEV1/FVC ratio of 65%, consistent with airflow limitation. Chest radiograph, complete blood count, electrocardiogram, and brain natriuretic peptide levels are normal. What is the next appropriate step in this patient’s evaluation? A . Echocardiogram B. Pulse oximetry with ambulation (6-minute walk test) C. Exercise treadmill test D. High-resolution chest computed tomography E. 24-hour Holter monitoring

Answer: B  This patient likely has chronic obstructive pulmonary disease (COPD) manifesting as dyspnea on exertion. His main risk factor is cigarette smoking. The presence of hypertension, obesity, and lack of exercise can suggest cardiovascular disease as a second and possibly contributing cause. A normal brain natriuretic peptide level and electrocardiogram make primary cardiac disease less likely but do not exclude it. Anemia is excluded by a normal complete blood count. COPD is the likely diagnosis, and oxygen testing to exclude hypoxemia is the best first test, and this test may also explain his hypertension. Additional testing likely will be required if the oxygen titration test is unrevealing. 4. A 17-year-old woman with a history of atopy with allergic rhinitis presents with wheezing of 3 months duration. The wheezing occurs primarily during the day while playing soccer, is triggered by cold air and exercise, and has not responded to a 6-week course of an inhaled corticosteroid and shortacting β-agonist. She denies chest tightness, chest pain, or a history of lung or cardiovascular disease. Her history is notable for allergies to dust, multiple weeds, grasses, and animal dander. She has persistent rhinitis for which she performs nasal saline rinses and uses intranasal corticosteroids daily. Her physical examination reveals normal vital signs and erythema of the upper airway without nasal polyps. The remainder of her physical examination is normal without wheezing. Pulmonary function tests, chest radiograph, and methacholine challenge testing are normal. What is the next step in this patient’s evaluation? A . Sinus computed tomography B. High-resolution chest computed tomography C. Direct laryngoscopy after exercise D. Echocardiogram E. Bronchoscopy with biopsy Answer: C  This presentation suggests vocal cord dysfunction presenting as wheezing associated with exercise. Wheezing in the setting of rhinitis and exercise suggests an airway process (either upper, lower or both) as the most likely diagnosis. Because the patient has not responded to therapy for asthma and a methacholine challenge is negative, asthma is highly unlikely. In the setting of rhinitis, an evaluation of the upper airway to exclude vocal cord dysfunction during an episode of wheezing is the next appropriate step. 5. A 46-year-old man presents with a cough that has produced blood-streaked sputum for the past two days. Associated symptoms include rhinorrhea, congestion, and subjective fever. He estimates the total amount of blood loss to be less than one tablespoon. The medical history is unremarkable. He does not smoke cigarettes and has not recently traveled, lost weight, or experienced night sweats. His vital signs are within normal limits, and the patient appears to be breathing comfortably, except for an intermittent cough. No blood is produced during the office visit. Pulmonary examination reveals increased tactile fremitus with increased breath sounds and egophony over the right lower chest. Nasal, oropharyngeal, cardiovascular, and abdominal examinations are unremarkable. CBC is normal, and chest radiography reveals a right lower lobe pneumonia. What is the next step in this patient’s evaluation? A . High-resolution chest CT B. Bronchial artery embolization C. Treatment with inhaled bronchodilators D. Treatment with antibiotics E. Echocardiogram Answer: D  This patient presents with non-massive hemoptysis and no evidence of underlying heart or lung disease. His hemoptysis is likely due to an infectious community-acquired pneumonia. He should receive antibiotic therapy with a follow-up chest radiograph in 6 to 8 weeks. If the hemoptysis or infiltrate persists, high-resolution chest CT is warranted to look for other causes, such as malignancy or bronchiectasis.

510

CHAPTER 78  Imaging in Pulmonary Disease  

78  IMAGING IN PULMONARY DISEASE PAUL STARK

  IMAGING OF THE LUNGS, MEDIASTINUM,

AND CHEST WALL



EPIDEMIOLOGY

Worldwide, chest radiography is the most commonly performed imaging procedure; more than 75 million chest radiographs are performed every year in the United States alone. Chest radiographs provide useful information about the patient’s anatomy and disease at a minimal monetary cost and with radiation exposure that most experts agree is negligible (0.05 to 0.1 mSv) (Chapter 17). Although many novel imaging techniques are available, the conventional chest radiograph remains invaluable in the initial assessment of disorders of the lung, pleura, mediastinum, and chest wall.  

Imaging Techniques

The standard chest radiograph is performed at 2 m from the x-ray tube focal spot to the image detector, in frontal and lateral projections. If possible, the radiographs should be obtained with the patient inhaling to total lung capacity. These images, which provide views of the lungs, mediastinum, and chest wall simultaneously, are typically acquired, stored, and distributed digitally.

Bedside Radiography

Although bedside radiography accounts for a large number of chest radiographs, especially in the intensive care unit (ICU), the images obtained are generally of lower technical quality, cost more, and are more difficult to interpret. Lung volumes are low, thereby leading to crowding of vascular structures, and the low kilovoltage technique required for the mobile equipment yields radiographs with overexposed lungs and an underpenetrated mediastinum. The anteroposterior projection and the slightly lordotic angulation of the x-ray beam combine to distort the basal lung structures and magnify the cardiac silhouette. Recumbent studies also make recognition of pleural effusions or pneumothoraces more difficult. In the ICU, chest radiography can be ordered selectively rather than as a daily routine, without compromising care.

Computed Tomography

Computed tomography (CT) has multiple advantages over conventional radiography. It displays cross-sectional anatomy free of superimposition, with a 10-fold higher contrast resolution. Multislice CT scanners acquire a continuous, volumetric, near-isotropic data set with possibilities for high-quality two-dimensional or three-dimensional reformatting (volume rendering) in any plane. High-resolution CT of the lung parenchyma is an important application; narrow collimation of the beam combined with an edge-enhancing high spatial frequency algorithm results in exquisite detail of normal and abnormal lungs, and correlation with pathologic anatomy is high. CT angiography is also a key component of the evaluation of suspected pulmonary embolism (Chapter 74).1

Magnetic Resonance Imaging

Magnetic resonance imaging (MRI) depends on the magnetic properties of hydrogen atoms. Magnetic coils and radio frequency coils lead to induction, excitation, relaxation, and eventual readout of magnetized protons. The molecular environment of hydrogen atoms will affect the rate at which they release energy; this energy yields a spatial distribution of signals that is converted into an image by computer algorithms, similar to CT. Because of its soft tissue specificity, MRI has applications in the assessment of chest wall invasion, mediastinal infiltration, and diaphragmatic involvement by lung cancer or malignant mesothelioma.

Positron Emission Tomography

Fluorodeoxyglucose positron emission tomography (FDG-PET) uses labeled fluorodeoxyglucose to image the glycolytic pathway of tumor cells or other metabolically active tissues with affinity for glucose. This technique has proved

CHAPTER 78  Imaging in Pulmonary Disease  

ABSTRACT

Thoracic imaging relies primarily on chest radiography and computed tomography. Important radiologic features that allow for detection and localization of abnormalities on chest radiographs include the following signs: 1. Silhouette sign indicates obscuration of a contour or soft tissue border normally outlined by aerated lung owing to adjacent pulmonary or pleural opacification. 2. Spine sign relies on an increase in basal opacification over the lower thoracic spine as seen on the lateral chest radiograph, thereby raising the possibility of a subtle pulmonary opacity in the lower lobes. 3. Incomplete border sign in the context of an intrathoracic nodule or mass indicates that the opacity is not located in the lung parenchyma but rather represents a pleural, mediastinal, or chest wall lesion. It can also be described as a “one-edged” lesion; it can form obtuse angles of interface with the chest wall or the mediastinum and can display tapered borders. 4. Hilum overlay sign distinguishes a mediastinal mass from a hilar mass; only a hilar mass obscures the ipsilateral pulmonary artery. 5. Mach bands or Mach effect rely on retinal illusions due to edge enhancement and facilitate visualization of contours on chest radiographs but can also mimic pathologic findings such as a pneumothorax. The task of a thoracic radiologist consists of detection and description of abnormal findings followed by a differential diagnosis and discussion, which include comparison with prior imaging studies, recommendations for further imaging and follow up. In this way, the radiologist aims to decrease diagnostic uncertainty and increase the overall probability of a correct diagnosis.

KEYWORDS

chest radiograph CT scanning lung pleura mediastinum

510.e3

CHAPTER 78  Imaging in Pulmonary Disease  

helpful in studying intrathoracic tumors and has facilitated the work-up of solitary pulmonary nodules. Integrated PET-CT scans have improved the diagnosis and staging of intrathoracic tumors.2

Ultrasonography

Outside the heart, ultrasonography traditionally has played only a limited role in thoracic imaging. Its primary use has been to localize pleural effusions and guide their drainage (Chapter 92). However, some data suggest that lung ultrasound can be used as a preliminary screen for pneumonia, pulmonary edema,3 or pneumothorax. In the intensive care setting, ultrasound also may help with the diagnosis of ventilator-associated pneumonia (Chapter 91), pneumothorax (Chapter 92), and diffuse alveolar damage (Chapter 85).  

Evaluation of Chest Images

Images of the chest are best evaluated by examining regions of the lung for specific findings and relating these findings to known diagnostic groups. A number of critical radiographic features should be considered, with an appreciation for the known causes of these changes.

511

disease, can be identified in a small percentage of patients with predominantly interstitial lung disease, such as sarcoidosis, pulmonary lymphoma, and pulmonary calcinosis. Because of such limitations, a graphically descriptive approach that combines analysis of predominant opacities, assessment of lung expansion, and distribution and profusion of disease yields a differential diagnosis. The term infiltrate should be avoided; instead, the term pulmonary opacities should be used, with opacities further classified as large (i.e., >1 cm in largest dimension) or small (i.e., 10% difference). The usual cause is the presence of more than one restrictive process, such as a parenchymal restrictive disorder plus obesity, respiratory muscle weakness, atelectasis, or occult obstruction. Grading the severity of such a “complex restrictive disorder”6 requires additional consideration. Some patients have a mixed disorder with evidence of both obstruction and restriction. Common causes include cystic fibrosis (Chapter 83), sarcoidosis (Chapter 89), and heart failure (Chapters 52 and 53) as well as cases in which the causes of the obstructive disorder and the restrictive disorder are unrelated. Disorders of the central airways can cause characteristic patterns of abnormality. In a “fixed airway obstruction” such as tracheal stenosis (see Fig. 79-1H), flow is typically reduced on both inspiration and expiration. In contrast, in a variable extrathoracic (upper) airway obstruction (see Fig. 79-1F), inspiration is disproportionately reduced; however, expiration is often abnormal, merely less so. Likewise, in variable intrathoracic obstruction (e.g., relapsing polychondritis, tracheomalacia, or a dynamic intrathoracic tracheal tumor), the expiratory flow-volume curve is reduced but in a pattern unlike that seen in asthma or COPD (see Fig. 79-1G). These central airway obstructive patterns are often mistaken for COPD but may signify a locally treatable cause of obstruction. In patients with heart disease, a decline in FEV1/FVC ratio is associated with underfilling of the left heart and low cardiac output. By comparison, a decline in FVC with preserved FEV1/FVC ratio is associated with left ventricular hypertrophy and diastolic dysfunction.7

  PROVOCATIVE TESTING Assessing Airway Responsiveness Hyperresponsiveness of airways to the smooth muscle–contracting effect of pharmacologic agents such as methacholine, as well as to cold air, dry air, and other physical stimuli, is characteristic of asthma (Chapter 81). It is also observed in COPD and other obstructive airway diseases. Bronchoprovocation studies, in which graded doses of a stimulus are used to elicit airway constriction, are performed to measure airway responsiveness. A responsive airway, that is, one in which a small stimulus leads to a fall in FEV1, may be used to confirm the diagnosis of asthma (Chapter 81). Exhaled nitric oxide is a marker of eosinophilic airway inflammation and can be used to predict the likelihood that airway obstruction will improve with corticosteroid treatment. However, the utility of exhaled nitric oxide levels for asthma management is controversial.

  CARDIOPULMONARY EXERCISE TESTS

Some patients have dyspnea (Chapter 77) or exercise limitation that is not adequately explained by the clinical examination, standard pulmonary function

523

testing, and chest imaging. For such patients, laboratory testing of physiologic performance during exercise can be enlightening. Cardiopulmonary exercise testing, which is usually performed on a cycle ergometer or treadmill, includes monitoring of the heart rate, electrocardiography, and pulse oximetry as well as breath-by-breath measurement of tidal volume, breathing rate, oxygen consumption, and carbon dioxide production. Optional measurements include arterial blood gases and noninvasive cardiac output. Outcomes include maximal oxygen uptake (V̇ O2max), maximal workload, maximal heart rate, ventilation parameters during exercise, and measurements of gas exchange. Results are analyzed to determine if anaerobic metabolism occurs when the study subject reaches maximal effort and to determine what limits the ability of a patient to exercise—a gas exchange abnormality, ventilatory limitation, cardiac limitation, or deconditioning. Simple tests of exercise performance, such as the 6-minute walk test, can quantify and serially assess exercise performance.

  BRONCHOALVEOLAR LAVAGE

Bronchoalveolar lavage is useful for evaluation of opportunistic infections in immunocompromised hosts (Chapter 265),8 but its utility in the evaluation of interstitial lung disease is controversial. The procedure is generally safe, although provision must be made for the transient deterioration in gas exchange after the procedure. Oxygen supplementation is usually necessary, and intubation and mechanical ventilation are sometimes needed. The differential cell count on a normal bronchoalveolar lavage specimen includes 85% macrophages or more, 10 to 15% lymphocytes, 3% neutrophils or less, 1% eosinophils or less, 1% mast cells or less, and less than 5% squamous epithelial cells (which are an indicator of contamination from the upper airway). Smokers may have higher cell counts and a higher percentage of neutrophils. Increased lymphocyte counts are seen in sarcoidosis (Chapter 89), hypersensitivity pneumonitis (Chapter 88), nonspecific interstitial pneumonitis (Chapter 86), collagen vascular diseases (Chapter 86), radiation pneumonitis (Chapter 88), cryptogenic organizing pneumonia (Chapter 86), and lymphoproliferative disorders. Increased neutrophil counts are seen in idiopathic pulmonary fibrosis (Chapter 86), collagen vascular diseases (Chapter 86), infectious pneumonia (Chapter 91), aspiration pneumonia (Chapter 91), acute respiratory distress syndrome (Chapter 96), diffuse alveolar damage (Chapter 85), acute interstitial pneumonia (Chapter 86), and asbestosis (Chapter 87). Increased eosinophils can be seen in asthma (Chapter 81), bronchitis (Chapter 90), allergic bronchopulmonary aspergillosis (Chapter 319), eosinophilic granulomatosis with polyangiitis (Chapter 254), Hodgkin lymphoma (Chapter 177), and drug-induced lung disease (Chapter 88). If eosinophils are more than 25%, eosinophilic pneumonia is likely (Chapter 161). If lymphocytes are increased and the clinical differential diagnosis includes sarcoidosis or hypersensitivity pneumonitis, analysis of T-cell populations may be helpful; the CD4:CD8 ratio is typically increased in sarcoidosis but reduced in hypersensitivity pneumonitis. If more than 20% of macrophages stain positive for hemosiderin, diffuse alveolar hemorrhage is considered likely (Chapter 85), particularly if lavage fluid is progressively bloody in successive aliquots of lavage fluid. Cellular constituents of bronchoalveolar lavage are usually stained for cytologic analysis for malignant cells and viral inclusions. If Langerhans cell histiocytosis (Chapter 86) is considered possible, 5% or more CD1a–positive cells support the diagnosis. If chronic beryllium disease or beryllium sensitization is possible, a lymphocyte proliferation test in response to exposure to beryllium salts can be helpful (Chapter 87). Staining of solid material from the bronchoalveolar lavage with periodic acid–Schiff (PAS) stain for the presence of PAS-positive material is essential to the diagnosis of pulmonary alveolar proteinosis (Chapter 85). A diagnosis of lipoid pneumonia (Chapter 88), caused by the aspiration of oil, can be confirmed by an excess of lipid-laden macrophages from bronchoalveolar lavage. The presence of asbestos bodies or silica is not diagnostic of lung disease related to these substances (Chapter 87) but does indicate significant exposure. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 79  Respiratory Testing and Function  

  COMPLEX DISORDERS

Many pulmonary function tests have more than one physiologic defect, such as a combination of restriction plus obstruction, or more than one cause of restriction (e.g., pulmonary fibrosis plus obesity or heart failure). These cases do not fit neatly into standard interpretive patterns of typical obstruction or restriction, yet they can be described, and their patterns suggest a differential diagnosis. A mixed disorder is a combination of both restriction, as indicated by a reduced TLC, plus obstruction, as indicated by a reduced FEV1/FVC ratio. Although it is widely recognized, it only occurs in 1 to 2% of pulmonary function tests. A mixed disorder may be due to a combination of a restrictive disorder plus an unrelated obstructive disorder, such as an interstitial disease plus COPD, but there are several disorders that typically produce a mixed pattern including cystic fibrosis (Chapter 83), sarcoidosis (Chapter 89), Langerhans cell histiocytosis (Chapter 160), and heart failure (Chapters 52 and 53). In a mixed pattern, the degree of severity of restriction can be determined from the TLC percent predicted. The overall impairment can be determined from the FEV1 percent predicted. The severity of the obstructive component can be inferred from the FEV1 percent predicted divided by the TLC percent predicted. Reduced lung volumes are the sine qua non of the diagnosis of restriction. However, about half of patients whose spirometry suggests restriction (reduced vital capacity with normal FEV1/FVC ratio, also called preserved ratio impaired spirometry or PRISM) have a normal total lung capacity, so they do not have true restriction but rather what is called the nonspecific pattern. The nonspecific pattern is very common, occurring in 9 to 10% of all complete pulmonary function tests. It is approximately as frequent as true restriction. Patients with the nonspecific pattern commonly have evidence of an obstructive disorder, not restriction, as indicated either by increased airway resistance, a response to a bronchodilator, or other clinical indicator of obstruction. It can be argued that the clinical utility of the measurement of airway resistance is limited to patients with the nonspecific pattern of whom about half have an increased airway resistance. Some patients with a nonspecific pattern do not have evidence of airway obstruction but are obese or have other chest wall limitations, neuromuscular weakness, poor performance, heart failure, pleural effusion, or a variety of other conditions. In typical cases of restrictive lung disease, including most cases of interstitial lung disease, lung volumes are reduced in proportion to the severity of the interstitial disease. For example, TLC might be reduced to 60% of the predicted value and FVC would be similarly reduced to about 60% of the predicted value. In such a case, grading the severity of restriction is easy. One should be cautious not to overlook the fact that the gas exchange abnormality may be more or less severe, depending on the underlying pathology. In about one third of cases of restriction, the reductions in FEV1 and FVC are disproportionate (>10% of predicted value) to the reduction in TLC. The discrepancy may be large, for example TLC may be 70% predicted while FVC is only 25% predicted. Interpreting physicians sometimes disagree whether to grade severity of restriction based on TLC percent predicted or FVC percent predicted. In such a case, the impairment could be called either mild or very severe. In a study of such cases, the term “complex restriction” was proposed. Whereas cases of “typical restriction” were seen in association with interstitial lung disease, and hence weighted toward older men, patients with “complex restriction” were more often women, younger, underweight, had slightly less severe reductions in Dlco, and more often had atelectasis, a mosaic pattern on computed tomography, obesity, diaphragm dysfunction, or neuromuscular disease.

  PULMONARY FUNCTION IN OBESITY

523.e1

The epidemic of obesity is manifested in many organ systems, including the respiratory system. Dyspnea, exercise limitation, and respiratory failure are more common in obese persons than in the nonobese. Asthma is more common and more severe in obese patients. The effects of obesity on lung function are usually relatively modest among ambulatory patients with a body mass index (BMI) less than 40. The most commonly observed effect of obesity on lung function is a reduction in expiratory reserve volume (the amount of air exhaled between FRC and residual volume), which is substantially reduced even in persons who are overweight (BMI 25 to 30) or mildly obese (BMI 30 to 35). Vital capacity is reduced in obesity, but the effect is modest, usually within the normal range, and highly variable. In large studies, vital capacity or FVC is reduced on average by 0.5 to 0.8% for each unit increase in BMI above 25. Effects of obesity on total lung capacity and FEV1 are somewhat smaller. The FEV1/FVC ratio and Dlco actually increase slightly with increasing BMI. In exercise studies, the effects of obesity among ambulatory outpatients are likewise modest. Such patients have an increased work of breathing and decreased external work efficiency related to the work of moving their own body mass, but maximal oxygen uptake is often normal.

523.e2

CHAPTER 79  Respiratory Testing and Function  

GENERAL REFERENCES 1. Godfrey MS, Jankowich MD. The vital capacity is vital: epidemiology and clinical significance of the restrictive spirometry pattern. Chest. 2016;149:238-251. 2. Thacher JD, Schultz ES, Hallberg J, et al. Tobacco smoke exposure in early life and adolescence in relation to lung function. Eur Respir J. 2018;51:1-9. 3. Dempsey TM, Scanlon PD. Pulmonary function tests for the generalist: a brief review. Mayo Clin Proc. 2018;93:763-771. 4. Cid-Juárez S, Thirión-Romero I, Torre-Bouscoulet L, et al. Inspiratory capacity and vital capacity of healthy subjects 9-81 years of age at moderate-high altitude. Respir Care. 2019;64:153-160.

5. Graham BL, Brusasco V, Burgos F, et al. 2017 ERS/ATS standards for single-breath carbon monoxide uptake in the lung. Eur Respir J. 2017;49:1-31. 6. Clay RD, Iyer VN, Reddy DR, et al. The “complex restrictive” pulmonary function pattern: clinical and radiologic analysis of a common but previously undescribed restrictive pattern. Chest. 2017;152:1258-1265. 7. Cuttica MJ, Colangelo LA, Shah SJ, et al. Loss of lung health from young adulthood and cardiac phenotypes in middle age. Am J Respir Crit Care Med. 2015;192:76-85. 8. Sakata KK, Klassen CL, Bollin KB, et al. Microbiologic yield of bronchoalveolar lavage specimens from stem cell transplant recipients. Transpl Infect Dis. 2017;19:1-17.

CHAPTER 79  Respiratory Testing and Function  

523.e3

REVIEW QUESTIONS 1. A 58-year-old man with exercise-related cough and a body mass index of 42 has a hemoglobin level of 14 g/dL, normal spirometry without a significant bronchodilator response, and a diffusing capacity (Dlco) that is 144% of the reference value. The most appropriate next test is: A . Bronchoalveolar lavage for hemosiderin-laden macrophages B. Echocardiography to exclude intracardiac shunt C. Quantitative assay for JAK2 mutation D. Methacholine challenge E. Measurement of hemoglobin P50 Answer: D  Obesity and asthma are the most likely causes of an increased Dlco, and a search for rare causes of an increased Dlco usually is not indicated. Bronchoalveolar lavage can be useful if clinical information suggests pulmonary hemorrhage. An echocardiogram may demonstrate a left to right shunt, which is a rare cause of an increased Dlco. JAK2 mutations are associated with polycythemia vera, but the patient is not polycythemic. 2. A 61-year-old male former smoker (40 pack-years) complains of dyspnea and cough. Pulmonary function testing shows normal spirometry and lung volumes; there is an isolated reduction in diffusing capacity (Dlco). The most useful next test is: A . Echocardiography B. Right-sided heart catheterization C. High-resolution computed tomography of the chest D. Maximal respiratory pressures E. Bronchoalveolar lavage for hemosiderin-laden macrophages Answer: C  An isolated reduction in Dlco is most often associated with emphysema or fibrosis (or both), which are seen best with computed tomography. An isolated reduction in Dlco is less often due to pulmonary vascular disorders such as pulmonary hypertension, so echocardiography and right-sided heart catheterization may be valuable in some cases but would have a lower yield. Muscle weakness can reduce the Dlco, but it also reduces lung volumes. More than 20% hemosiderin-laden macrophages on bronchoalveolar lavage is suggestive of diffuse alveolar hemorrhage, which is a rare cause of an increased Dlco. 3. A 53-year-old never-smoker with a saddle nose deformity has severe dyspnea and dry cough. His pulmonary function test results are as follows: 53 yo M Ht = 177 cm Wt = 89 kg BMI = 28 Never-smoker CONTROL

%PRED

FVC

4.51

93

FEV1

1.52

40

FEV1/FVC

33.7

43

MVV

55

36

FEF50/FIF50

0.3

30

FEFmax

2.7

32

FIFmax

4.5

75

Maximal Expiratory Flow, liters/sec

%PRED = percentage of predicted value

8

6

4

2

0

–2

–4

–6 –8

Flow Volume Curve Predicted Control Post dilator

523.e4

CHAPTER 79  Respiratory Testing and Function  

He reports episodes of ear pain and erythema, refractory to antibiotics but responsive to steroids. What is the next most appropriate test? A . Methacholine challenge B. Maximal respiratory pressures C. Airway resistance D. Imaging of the central airways (bronchoscopy or dynamic computed tomography) E. Measurement of exhaled nitric oxide Answer: D  He has relapsing polychondritis. His main respiratory issue is dynamic central airway collapse due to chondromalacia of the tracheal and bronchial cartilage. The flow-volume curve shows characteristic flattening, as opposed to the “scooped out” pattern of asthma and COPD. Inspiratory flows are normal. He does not have a disorder of airway reactivity, so methacholine challenge adds little useful information and may not be safe with this degree of obstruction. Maximal respiratory pressures are not likely to be abnormal. Airway resistance will be abnormal but will add nothing diagnostically. Exhaled nitric oxide is abnormal in patients with eosinophilic airway inflammation and would not be expected to be abnormal in this case. 4. A patient with mild obstruction on spirometry has a maximal voluntary ventilation that is reduced out of proportion to the FEV1. Which of the following is least likely to be helpful? A . Maximal respiratory pressures B. Inspiratory flow-volume curve C. Cardiopulmonary exercise challenge D. Airway resistance measurement E. Careful scrutiny of test for repeatability of measures and technician comments on patient performance Answer: C  A disproportionate reduction in maximal voluntary ventilation may be due to inspiratory obstruction, muscle weakness, or poor performance. Cardiopulmonary exercise testing is likely to be abnormal regardless of the cause of the abnormality. The other four options would yield more specific diagnostic information. 5. A 34-year-old man is being evaluated for dyspnea and lack of energy. Results are as follows: TLC, 62% predicted; FVC, 40%; FEV1, 36%; FEV1/FVC, 0.90%; Dlco, 60%. The expiratory flow-volume curve is as shown: 0

Flow Volume Curve FET Predicted Control 3.9

10

2

Maximal Inspiratory Flow, liters/sec

Maximal Expiratory Flow, liters/sec

12

8

4

6

6

4

8

2

10

0

12 0

1

2

3 4 5 Expired Volume, liters

6

7

8

What test is likely to be most helpful? A . Maximal respiratory pressures B. Airway resistance C. Methacholine challenge D. Cardiopulmonary exercise test E. Arterial blood gases Answer: A  The convex shape of the flow-volume curve in an adult suggests muscle weakness or poor performance. In a patient with restriction, the disproportionate reduction in FVC compared with TLC may be due to muscle weakness. This patient has a myopathy. Alternative considerations include chest wall limitation, poor performance, and occult airflow obstruction. The most helpful measurements on this patient will be maximal respiratory pressures, which will likely result in referral to a neurologist. Airway resistance is unlikely to be abnormal with this flow-volume curve. There is little to suggest asthma, and an exercise study is likely to be abnormal but may not reveal the cause of the abnormality. Arterial blood gases are usually normal in neuromuscular disorders until the FEV1 and FVC are severely reduced, after which hypercapnia develops as an indicator of respiratory failure.

CHAPTER 79  Respiratory Testing and Function  

523.e5

6. A 40-year-old woman is referred for second opinion about her “asthma.” She has never smoked and has been symptomatic since a hospitalization after a motor vehicle accident 10 years ago. She has not responded to bronchodilators and inhaled steroids. This is her first spirometry test. 40 yo F Ht = 178 cm Wt = 79 kg BMI = 25 CONTROL

%PRED

TLC

5.73

83

RV

2.32

124

RV/TLC

0.41

148

FVC

3.40

68

FEV1

1.51

38

FEV1/FVC

44.5

MVV

11

FEF50/FIF50

56 7

1.1

110

DLCO (hb adj)

18.58

61

SpO2

98

%PRED = percentage of predicted value.

Expiratory Flow, liters/sec

6

4

2

Inspiratory Flow, liters/sec

0

1

2

3

4

Expired Volume (L) Inspired

2

4

6

What test is most likely to be helpful? A . Computed tomography of the chest B. Oral exhaled nitric oxide C. Methacholine challenge D. Laryngoscopic examination of the upper airways E. Sputum examination for Charcot Leyden crystals and Curschmann spirals Answer: D  She has tracheal stenosis, which resulted from prolonged intubation or tracheostomy after her motor vehicle accident. CT of the chest does not always identify tracheal stenosis. Oral exhaled NO, methacholine challenge, and examination of the sputum for Charcot Leyden crystals and Curschmann spirals all are manifestations of asthma, but her flow-volume curve has the characteristic appearance of tracheal stenosis, not asthma. The tracheal stenosis is obvious on examination of the flow-volume curve and reinforces the need to perform spirometry for evaluation of patients thought to have common obstructive disorders such as asthma and COPD. Suggested Interpretation: Abnormal. Severe fixed airway obstruction is indicated by the reduced FEV1 and MVV and shape of the inspiratory and expiratory flow-volume curves. There is no immediate response to bronchodilator. Dlco is mildly reduced, consistent with a pulmonary parenchymal or vascular process. Lung volumes and oxygen saturations are normal.

523.e6

CHAPTER 79  Respiratory Testing and Function  

7. An 80-year-old man who underwent right pneumonectomy 16 years ago for lung cancer has severe aortic stenosis and severe coronary disease. He has pulmonary function testing before aortic valvuloplasty.

Maximal Expiratory Flow, liters/sec

12

Flow Volume Curve PRED mayo Control Post dilator

10 8 6 4 2 0 0

1

2

3 4 5 Expired Volume, liters

6

7

80 yo M Ht = 185 cm Wt = 66 kg BMI = 19 CONTROL

%PRED

TLC

5.37

73

RV

3.53

142

RV/TLC

0.66

194

FVC

1.86

38

FEV1

1.25

35

FEV1/FVC

67.5

91

MVV

33

27

Max Insp Press Max Exp Press

25 29

25 16

DLCO (hb adj) SpO2

9.5 100

38 99

%PRED = percentage of predicted value.

How would you interpret his results? A. Mild restriction B. Severe restriction C. Mild-to-severe restriction D. Severe mixed obstruction/restriction E. Severe complex restrictive disorder Answer: E  This illustrates the dilemma posed by conventional thinking about grading restriction. Should one call this mild based on TLC or severe based on FVC or split the difference and call it mild-to-severe? When the reductions in TLC and FVC are discordant by more than 10%, there is usually a second process contributing to the restriction. In this case, the processes are a combination of pneumonectomy, heart failure, and weakness. Suggested Interpretation: Abnormal. Complex restriction. A restrictive process is indicated by the mild reduction in TLC. The disproportionately severe reductions in vital capacity and FEV1, relative to TLC, suggests an additional process, which might include chest wall limitation, muscle weakness, poor performance, heart failure, or occult obstruction. Dlco (adjusted for hemoglobin) is severely reduced, consistent with a pulmonary parenchymal or vascular process or anemia. Oximetry is normal at rest and during exercise.

CHAPTER 79  Respiratory Testing and Function  

523.e7

8. A 52-year-old woman with primary biliary cirrhosis, type 2 diabetes, and moderate persistent asthma was evaluated for increasing dyspnea. 52 yo F Ht = 168 cm Wt = 103 kg BMI = 37 CONTROL

%PRED

TLC

5.22

97

RV

2.62

141

.50

69

2.45

69

RV/TLC FVC FEV1

1.98

POST DILATOR

%CHANGE

3.10

26

2.37

20

69

FEV1/FVC

80.6

100

MVV

74

Raw

18.6

400

DLCO (hb adj)

25

107

SPO2

97

96

76.7

70

%PRED = percentage of predicted value.

Maximal Expiratory Flow, liters/sec

12

Flow Volume Curve PRED Control Post dilator

10

8

6

4

2

0

0

1

2

3 4 Expired Volume, liters

5

6

7

How would you interpret her results? A. Moderate restriction B. Moderate obstruction C. Poor test performance D. Nonspecific abnormality E. Complex restrictive disorder Answer: D  This pattern fits neither obstruction, because of the normal FEV1/FVC ratio, nor restriction, because of the normal TLC. The patient’s successive efforts were highly repeatable, within less than 150 mL, arguing against poor performance. The findings do not fit the description of complex restriction, because although the FVC % predicted is less than TLC % predicted, the TLC is not abnormal. This nonspecific pattern was described in 2009 and further characterized in 2011. It was previously thought that this pattern represented a variant of obstruction, and it is frequently associated with obstructive disorders, such as this patient’s asthma. However, 30 to 40% of cases have no evidence of obstruction but rather some form of chest wall limitation, such as obesity, muscle weakness, or chest wall deformity. For this patient, airway resistance was measured to evaluate the nonspecific pattern and proved to be very high. In addition, the shape of the flow-volume curve and the response to bronchodilator suggest an obstructive process. Suggested Interpretation: Abnormal. FVC and FEV1 are moderately reduced in a nonspecific pattern with a normal TLC and FEV1/FVC ratio. The shape of the flow-volume curve, the increased airway resistance and the improved flows after bronchodilator all suggest a partly reversible obstructive process. Dlco and oximetry at rest and during exercise are all normal.

523.e8

CHAPTER 79  Respiratory Testing and Function  

9. A 62-year-old man is being evaluated for hematuria, coronary artery disease, peripheral arterial disease with an ischemic foot ulcer, and type 2 diabetes. He has smoked a pack of cigarettes per day for 40 years and quit 7 years ago. He complains of dyspnea on exertion. 62 yo M Ht = 189 cm Wt = 125 kg BMI = 35 %PRED

POST DILATOR

%CHANGE

FVC

CONTROL 5.08

92

5.53

9

FEV1

4.12

98

4.11

0

107

74.3

FEV1/FVC

81.2

MVV

127

DLCO

29.8

SPO2

94

84 100 93

%PRED = percentage of predicted value.

Maximal Expiratory Flow, liters/sec

12

Flow Volume Curve PRED Control Post dilator

10 8 6 4 2 0 0

1

2

3 4 Expired Volume, liters

5

6

7

What is the most appropriate next test? A . CT of the chest B. Overnight oximetry C. Cardiopulmonary exercise test D. Arterial blood gas Answer: B  Chest CT is not indicated by findings from pulmonary function testing. Nevertheless, it was done for lung cancer screening purposes. Like many U.S. Midwesterners, this patient has several indeterminate lung nodules that require follow-up. Cardiopulmonary exercise testing, if performed, might show evidence of deconditioning and also might identify evidence of myocardial ischemia. It was not performed because of the patient’s ischemic foot. A nuclear stress test was performed instead. It showed no evidence of stress-induced ischemia or infarction. Left ventricular size and function appeared to be normal. The patient had no evidence of hypoxemia, and hypercapnia would be unlikely with normal spirometry, so arterial blood gases were not obtained. An overnight oximetry showed frequent nocturnal desaturations with a pattern suggesting REM accentuation. The patient has severe obstructive sleep apnea as evidenced by the sawtooth abnormality, which indicates a two-fold increase in the likelihood of obstructive sleep apnea compared with normal subjects. Suggested Interpretation: Numerical results of spirometry, DLCO and oximetry are normal, however, the sawtooth configuration of the flow-volume curve indicates redundant tissue in the upper airway. This correlates with snoring and may be predictive of obstructive sleep apnea. Bourne MH, Jr., Scanlon PD, Schroeder DR, et al. The sawtooth sign is predictive of obstructive sleep apnea. Sleep Breath. 2017;21:469-474.

CHAPTER 79  Respiratory Testing and Function  

523.e9

10. A 58-year-old woman is a current smoker with a 40 pack-year smoking history. She is severely dyspneic and is oxygen dependent. 58 yo F Ht = 162 cm Wt = 60 kg BMI = 23 CONTROL

%PRED

TLC

5.19

103

RV

1.39

75

RV/TLC

POST DILATOR

%CHANGE

.27

73

FVC

3.40

107

3.33

−2

FEV1

2.56

100

2.62

2

FEV1/FVC

75.3

93

MVV

96

99

DLCO

7.0

32

SPO2

91 (rest)

73 (exercise)

78.7

%PRED = percentage of predicted value.

Maximal Expiratory Flow, liters/sec

12

Flow Volume Curve Predicted Control Post dilator

10 8 6 4 2 0 0

1

2

3 4 Expired Volume, liters

5

6

7

What is the most likely cause of this abnormality? A . Emphysema B. Pulmonary fibrosis C. Both A and B D. Primary pulmonary hypertension Answer: A  This is an isolated reduction in Dlco, which is often found in patients with emphysema or pulmonary fibrosis or both. Combined pulmonary fibrosis and emphysema is a recognized entity, mostly in current and former smokers. The curious aspect of this dual disease entity is that the increased lung recoil caused by the fibrosis can counterbalance the loss of recoil from emphysema. In some cases, the two are matched, thereby preserving airway patency and resulting in normal airflows and lung volumes as in this case. In other cases, however, either the restrictive or the obstructive physiology may predominate. Both processes impair gas exchange, however, often resulting in a very low Dlco as in this case. This patient had severe diffuse fibrosis with honeycombing plus emphysematous changes, particularly in the upper lungs. Suggested Interpretation: Abnormal. TLC, FEV1, FVC, FEV1/FVC and MVV are within accepted ranges of normal. Dlco is severely reduced, consistent with emphysema or other pulmonary vascular or parenchymal process. Oxygen saturation is slightly reduced at rest and decreases markedly during exercise. Tzilas V, Bouros D. Combined pulmonary fibrosis and emphysema, a clinical review. COPD Research and Practice. 2016;2:2. DOI: 10.1186/s40749-016 -0018-1.

524

CHAPTER 80  Disorders of Ventilatory Control  

80  DISORDERS OF VENTILATORY CONTROL ATUL MALHOTRA AND FRANK POWELL



DEFINITIONS AND PATHOGENESIS

Ventilatory Control

exercise, breathing becomes dependent primarily on metabolic stimuli. Patients may complain of fatigue or sleepiness because arousals from sleep tend to occur during the hyperpneic phase. Paroxysmal nocturnal dyspnea, a classic symptom of heart failure (Chapter 52), most commonly reflects underlying Cheyne-Stokes breathing. Patients often are diagnosed in the sleep laboratory while undergoing investigation for possible obstructive sleep apnea. However, in contrast to obstructive sleep apnea, Cheyne-Stokes breathing usually resolves during rapid eye movement (REM) sleep, and arousals on the electroencephalogram typically occur during the hyperpneic phase. Furthermore, CheyneStokes breathing generally does not resolve immediately when nasal continuous positive airway pressure (CPAP) is applied.

Ventilation is controlled by complex interactions between central chemoreceptors, which respond mainly to arterial carbon dioxide tensions, and peripheral chemoreceptors, which respond to arterial carbon dioxide and oxygen tensions. These reflexes modulate the underlying respiratory rhythm, and disorders of ventilatory control are caused by derangements in these control systems.

TREATMENT  Medical management of Cheyne-Stokes breathing most often is treatment of the underlying heart failure (Chapter 53). After optimization of medical management, the Cheyne-Stokes breathing pattern frequently resolves. CPAP can improve breathing indices but is no better than standard medical therapy from the standpoint of mortality. Results have been similar with other ventilator techniques. A1 ,3 An implantable phrenic nerve stimulator remains experimental.4

  HYPOVENTILATION SYNDROMES

Hypoventilation syndromes are defined by a lack of adequate alveolar ventilation to maintain a normal arterial carbon dioxide tension of 40 mm Hg. The two most common clinical settings that result in chronic hypoventilation are severe chronic obstructive pulmonary disease (COPD; Chapter 82) and morbid obesity (Chapters 377 and 207); less common causes are chronic opiate therapy, neuromuscular weakness (Chapters 393 and 394), and severe kyphoscoliosis (Chapter 92). The epidemiology of these hypoventilation syndromes is poorly studied, but about 15% of patients with severe COPD or morbid obesity have an elevated Paco2. Regardless of the cause, patients with hypoventilation frequently have further worsening of their ventilation at the onset of sleep due to loss of the wakefulness stimulus, which is the normal drive to breathe while awake, and some degree of upper airway collapse after the onset of sleep (Chapter 377). Patients with central sleep apnea (Chapter 377), which is a group of conditions in which cessation of airflow occurs because of a lack of respiratory effort, are classified into those with inadequate ventilatory drive and those with excessive drive (Table 80-1).1 The apparent paradox of how excessive drive leads to central apnea is explained by the concept of loop gain. A negative feedback control system with a high loop gain is prone to instability that leads to periods of excessive breathing followed by periods of apnea. The prototype of a condition with high loop gain is periodic breathing or Cheyne-Stokes breathing (Fig. 80-1). Hypercapnic diseases include acquired diseases and the central congenital hyperventilation syndrome (Table 80-2).  

Cheyne-Stokes Breathing

Cheyne-Stokes breathing is a waxing and waning pattern of breathing, which is classically described as crescendo-decrescendo and often includes periods of central apnea. Cheyne-Stokes is seen most commonly during sleep in patients with heart failure.  

EPIDEMIOLOGY

Cheyne-Stokes breathing is a form of ventilatory instability that occurs in 20 to 40% of patients with left ventricular systolic dysfunction.2 Male sex, advanced age, low baseline Paco2, and atrial fibrillation are risk factors for Cheyne-Stokes breathing among patients with heart failure. Controversy remains regarding whether this breathing pattern itself is deleterious or whether it is simply a marker of the underlying severity of cardiac disease. Cheyne-Stokes breathing represents about 5 to 10% of all cases of sleep apnea (Chapter 377) and is uncommon among patients who do not have heart failure.  

PATHOBIOLOGY

Individuals with Cheyne-Stokes breathing have robust chemosensitivity as evidenced by marked increases in ventilation with small increases in Paco2. The drive to breathe may be further increased by neural reflexes that are triggered by extravascular lung fluid and an elevated left atrial pressure. Intermittent hypoxemia and catecholamine surges, which are frequent in these patients, contribute to oxidative stress and neuroendocrine activation, both of which are thought to contribute to worsening of the underlying heart failure.  

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Patients with Cheyne-Stokes breathing can sometimes be diagnosed at the bedside by careful observation of their breathing pattern. During sleep or



Central Congenital Hypoventilation Syndrome  

DEFINITION AND EPIDEMIOLOGY

Central congenital hypoventilation syndrome is a rare congenital condition, previously referred to as Ondine curse, characterized by a diminished ventilatory response to carbon dioxide.5 The central congenital hypoventilation syndrome was traditionally diagnosed in neonates, but more subtle forms of disease are increasingly noted in older children and adults.  

PATHOBIOLOGY

The syndrome is now defined by a mutation in the PHOX2B gene, located on chromosome 4p12. The PHOX2B gene is a highly conserved homeobox gene that is expressed in cardiopulmonary reflex pathways, including central CO2-sensitive chemoreceptors that are located in the retrotrapezoid nucleus and that provide excitatory input to the respiratory pattern generator. Abnormalities in PHOX2B genes have also been associated with Hirschsprung disease (Chapter 127), neural crest tumors, cardiac asystole (Chapter 57), and other abnormalities of the autonomic nervous system (Chapter 390). Because most parents of affected children with the central congenital hypoventilation syndrome do not carry a PHOX2B mutation, the mutations are de novo. About 90% of patients are heterozygous for a polyalanine repeat expansion mutation, in which the affected allele has 24 to 33 alanines rather than the normal 20 alanines. The remaining 10% of central congenital hypoventilation syndrome patients have missense, nonsense, or frameshift mutations in the PHOX2B gene.6  

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Neonates can present with cyanosis at birth, recurrent central apneas, or both. Adults can present with idiopathic central sleep apnea, unexplained hypercapnia, or autonomic abnormalities (Chapter 390). Confirmation of the diagnosis requires the demonstration of an abnormality in the PHOX2B gene.

TREATMENT  There are currently no specific therapies for central congenital hypoventilation syndrome beyond supportive care. Genetic counseling is required for afflicted individuals and their families, given the autosomal dominant pattern of inheritance. Patients must be cautioned against the use of sedatives, which could precipitate respiratory failure. Mechanical ventilation during sleep either invasively (through tracheostomy) or noninvasively (through bilevel positive airway pressure support [Chapter 377]) is required in most patients. Some patients remain fully ventilator dependent. Diaphragmatic pacing can sometimes be effective7 but ventilatory stimulants are generally ineffective.

CHAPTER 80  Disorders of Ventilatory Control  

ABSTRACT

Ventilation is controlled by complex interactions between central chemoreceptors, which respond mainly to arterial carbon dioxide tensions, and peripheral chemoreceptors, which respond to arterial carbon dioxide and oxygen tensions. Disorders of ventilatory control are caused by derangements in this control system. Hypoventilation syndromes, defined by a lack of adequate alveolar ventilation to maintain a normal arterial carbon dioxide tension of 40 mm Hg, are observed with morbid obesity, opiate overdoses, muscular weakness, and in some cases of severe chronic obstructive pulmonary disease (COPD). Central apneas can occur when the “loop gain” in the negative feedback ventilatory control system is too high and becomes prone to instability, thereby leading to periods of excessive breathing followed by apnea. The prototype of this condition is Cheyne-Stokes breathing, which is common during sleep in patients with heart failure. Treatment of Cheyne-Stokes breathing is controversial but starts with addressing any underlying medical causes such as optimizing medical therapy for heart failure. Finally, central congenital hypoventilation syndrome is a rare condition that is defined by a mutation in the PHOX2B gene on chromosome 4p12 and characterized by a diminished ventilatory response to carbon dioxide and sleep apnea.

KEYWORDS

arterial chemoreceptor central chemoreceptor central apnea central congenital hypoventilation syndrome Cheyne-Stokes breathing loop gain

524.e1

525

CHAPTER 80  Disorders of Ventilatory Control  

TABLE 80-1 CLASSIFICATION OF CENTRAL SLEEP APNEA CENTRAL SLEEP APNEA SYNDROME

MECHANISM

THERAPY

Sleep transition apneas

Carbon dioxide fluctuations during transitions from sleep to wake to sleep

Reassurance, occasionally hypnotics or oxygen

Chronic narcotic therapy

Lack of central drive

Reduce narcotic dose Consider positive-pressure device

Cheyne-Stokes breathing

High loop gain from robust chemosensitivity and ventilatory drive

Optimize medical therapy for heart failure; consider PAP device

Idiopathic central apnea

Unknown

Supportive, bilevel PAP; consider ventilatory stimulants

Treatment of emergent central apnea or “complex apnea”

Lowering upper airway resistance at CPAP initiation improves efficiency of carbon dioxide excretion

Reassurance, generally resolves spontaneously

Sleep hypoventilation syndromes

Fall in drive with loss of wakefulness stimulus, loss of accessory muscle activity during REM sleep

Noninvasive ventilation

CPAP = continuous positive airway pressure; PAP = positive airway pressure; REM = rapid eye movement.

Cheyne-Stokes Respiration

FIGURE 80-1.  Cheyne-Stokes breathing with crescendo-decrescendo pattern of breathing. The thermistor detects air temperature changes at the mouth and nose. Note absences in airflow without respiratory effort seen in the abdominal belts. This breathing pattern leads to intermittent desaturations, arousals from sleep, and bursts of tachycardia. The loop gain concept can be understood by considering the thermostat analogy in which a control system is working to regulate a stable room temperature (e.g., 20° C). By analogy, the respiratory control system is working primarily to maintain a stable PaCO2 of 40 mm Hg and stable pH. Situations in which marked fluctuations in room temperature might occur would include one in which the thermostat is excessively sensitive (i.e., furnace turns on if room temperature falls to 19.999° C); if the furnace is too powerful, a marked overshoot in room temperature will be followed by a prolonged period when the furnace does not run. In the analogy to Cheyne-Stokes breathing, carbon dioxide is equated to room temperature and would be predicted to be unstable if chemosensitivity (i.e., the thermostat) were excessively robust (i.e., a marked increase in ventilation for a small change in carbon dioxide) or if the efficiency of carbon dioxide excretion were high (i.e., marked fall in PaCO2 with increased ventilation). Situations that increase the propensity for carbon dioxide fluctuations lead to elevated loop gain and thus increase the risk for Cheyne-Stokes breathing.



Acquired Hypoventilation Syndromes  

DEFINITION AND EPIDEMIOLOGY

Patients with hypoventilation syndromes cannot maintain adequate minute ventilation to keep their Paco2 at 40 mm Hg. Patients can be classified into those who lack central ventilatory drive and those who have a pulmonary mechanical or neuromuscular abnormality that prevents adequate gas exchange (see Table 80-2). The case frequency is unknown, but hypercapnic respiratory failure is one of the more common admission diagnoses in intensive care units.  

PATHOBIOLOGY

Patients with conditions characterized by the lack of central drive have reasonably normal lungs and respiratory muscle function but lack adequate response to carbon dioxide and hypoxia. In contrast, most patients with mechanical or neuromuscular abnormalities have a larger work of breathing compared with

normal individuals; the most common underlying conditions are severe COPD (Chapter 82) and morbid obesity (Chapter 207) with the obesityhypoventilation syndrome. Such individuals have diminished but not absent chemoresponsiveness, which may be an acquired trait or may relate to an as yet unknown genetic predisposition. Another cause of inadequate gas exchange is neuromuscular disease; common causes include disorders of neuromuscular transmission (Chapter 394), severe muscle weakness (Chapter 393), the residua from poliovirus infection (Chapter 355), Guillain-Barré syndrome (Chapter 392), and acute poisoning (Chapter 102).  

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Patients with hypoventilation have myriad presentations ranging from asymptomatic abnormalities in laboratory testing (e.g., elevated Paco2, unexplained low Sao2, or elevated serum bicarbonate level) to respiratory failure in the intensive care unit (e.g., respiratory infection with laboratory evidence of chronic abnormalities, such as acute-on-chronic respiratory acidosis). Patients

526

CHAPTER 80  Disorders of Ventilatory Control  

TABLE 80-2 CLASSIFICATION OF HYPERCAPNIC DISEASES MECHANISM

DIAGNOSIS

TREATMENT

ACQUIRED DISEASE Narcotic overdose

Reduced central drive

History, narcotized pupils, toxicology

Supportive care, naloxone

Acute severe asthma

Severe airflow obstruction, high dead space

Typical history, wheezing on examination, low FEV1/FVC

Bronchodilators, anti-inflammatories, mechanical ventilation (usually invasive)

Acute exacerbation of COPD

Airflow obstruction, high dead space

History, cigarette smoking, low FEV1/FVC, infectious etiology

Bronchodilators, anti-inflammatories, noninvasive ventilation

Obesity-hypoventilation syndrome

Low respiratory system compliance, high upper airway resistance, low central drive

High BMI, lack of other diagnoses; blunted carbon dioxide response

Weight loss, nocturnal bilevel positive airway pressure

Neuromuscular disease (e.g., myasthenia gravis, ALS, polymyositis, GBS/AIDP)

Lack of respiratory muscle force

Immediate orthopnea, low VC, low MIPs/MEPs

Underlying cause, nocturnal noninvasive ventilation, supportive care

Severe parenchymal lung disease, e.g., COPD

Lack of alveolar surface area; high pulmonary dead space and work of breathing

Typical history, smoking, low FEV1 and FEV1/ FVC

Bronchodilator, anti-inflammatory therapy, possible nocturnal noninvasive ventilation, smoking cessation

Kyphoscoliosis

Low respiratory system compliance

Physical examination

Supportive care, noninvasive ventilation

PHOX2B mutation, lack of central drive

Genetic testing

Supportive care, mechanical ventilation (usually noninvasive)

CONGENITAL DISEASE Central congenital hypoventilation syndrome

AIDP = acute inflammatory demyelinating polyneuropathy; ALS = amyotrophic lateral sclerosis; BMI = body mass index; COPD = chronic obstructive pulmonary disease; FEV1 = forced expiratory volume in 1 second; FVC = forced vital capacity; GBS = Guillain-Barré syndrome; MEPs = maximal expiratory pressures; MIPs = maximal inspiratory pressures; VC = vital capacity.

PaCO2

pH > 7.4 Consider metabolic alkalosis if a 10-mEq rise in HCO3 yields 7-mm Hg rise in PaCO2 Treat underlying cause

pH < 7.4

Acute if 10-mm Hg rise in PaCO2 yields 1-mEq rise in HCO3 Abrupt presentation See Chapter 110

Chronic if 10-mm Hg rise in PaCO2 yields 4-mEq rise in HCO3

Consider cause Careful history Can’t breathe

Neuromuscular • Immediate orthopnea • Diaphragmatic percussion MIPs/ MEPs

Parenchymal √ Breath sounds √ PFTs Consider severe COPD

Won’t breathe: Check duration Low P0.1 Chest wall • Kyphoscoliosis • Obesity-hypoventilation

Symptoms since birth: √ PHOX2B gene

Acquired: Consider brain stem lesions, drugs, etc.

FIGURE 80-2.  A flow chart of a systematic approach to hypercapnia and various causes of hypoventilation. The change in pH can help determine the cause and chronicity. A careful history and physical examination, coupled with pulmonary function testing, can help classify patients into those who “can’t breathe” because of neuromuscular or mechanical abnormalities of the respiratory system compared with those who “won’t breathe” because of central nervous system disease. COPD = chronic obstructive pulmonary disease; MEPs = maximal expiratory pressures; MIPs = maximal inspiratory pressures; P0.1 = the negative mouth pressure generated during the first 100 msec of an occluded inspiration; PFTs = pulmonary function tests.

who acutely overdose on sedative-hypnotic or narcotic agents may present with acute respiratory acidosis and loss of consciousness. Patients who take chronic narcotics may present with central sleep apnea-hypopnea or otherwise unexplained oxygen desaturation at night. Once it is suspected, the diagnosis of hypoventilation is confirmed by the finding of Paco2 higher than 42 mm Hg on analysis of an arterial blood sample.

If the increase in Paco2 is of short duration so that renal compensation has not yet occurred (Chapter 110), the serum bicarbonate level is increased by 1 mEq/L for every rise of 10 mm Hg in Paco2. By comparison, if the respiratory acidosis is of sufficient duration for renal compensation to occur, the serum bicarbonate level will be increased by 4 mEq for every rise of 10 mm Hg in Paco2 (Fig. 80-2).

Once an elevated Paco2 is established, it is appropriate to distinguish patients who “can’t breathe” from those who “won’t breathe.” “Can’t breathe” implies that a respiratory mechanical problem or neuromuscular weakness is causing the elevation in Paco2. Abnormalities in pulmonary function testing (e.g., a very low vital capacity) suggest a parenchymal or chest wall disorder. Ultrasound can identify phrenic neuropathy causing diaphragmatic dysfunction. Patients who “won’t breathe” have central nervous system abnormalities that affect central drive, chemosensitivity, or both.

TREATMENT AND PROGNOSIS  The treatment of hypoventilation should focus on the underlying cause. Acute poisonings can be managed supportively or, in some cases, with specific antidotes (Chapter 102). Chronic conditions can be treated by addressing the underlying cause, such as weight loss in obesity-hypoventilation syndrome or cholinesterase inhibitors in myasthenia gravis (Chapter 394).8 For parenchymal lung disease, treatment is directed at the underlying cause, if possible (Chapters 82 and 86). Sedative medications should be used cautiously because they can occasionally precipitate acute respiratory failure. Although profound hypoxemia can clearly be deleterious, oxygen occasionally can precipitate severe acute respiratory acidosis, particularly in patients with acute exacerbations of COPD (Chapter 82). As a result, hypoventilating patients with COPD require cautious management including the careful administration of supplemental oxygen, which should be titrated to an arterial oxygen saturation of 90% or an arterial oxygen tension of 60 mm Hg. Severe hypoventilation requires mechanical ventilation (Chapter 97), such as noninvasive ventilation for an acute exacerbation of COPD. For other presentations in which the PaCO2 is believed to be acutely elevated, endotracheal intubation and mechanical ventilation are frequently used, especially in patients with impaired consciousness. For chronic hypoventilation in hypercapnic COPD, noninvasive bilevel positive airway pressure through a face mask during sleep can maintain alveolar ventilation, but there is no definitive evidence that noninvasive positive-pressure ventilation can prolong life or reduce hospitalizations in patients with COPD and chronic respiratory failure. In addition, the considerable difficulty of adhering to nocturnal bilevel therapy in COPD emphasizes the need for discussions with patients and families regarding its risks and benefits. For obesity hypoventilation syndrome without severe obstructive sleep apnea, noninvasive ventilation improves PaCO2 and reduces daytime sleepiness, as well as improving health-related quality of life. A2  In a recent randomized trial of patients with central sleep apnea and heart failure, adaptive servo-ventilation, which provides inspiratory pressure support in addition to expiratory positive airway, was not helpful. A3  Other chronic hypoventilation syndromes are also commonly treated with bilevel positive airway pressure, although data are not compelling. In some chronic conditions, such as motor neuron disease (Chapter 391), tracheostomy should be discussed, although the impact of such interventions on quality of life should be carefully considered. Regardless of the underlying cause, an elevation in the PaCO2 level is considered a poor prognostic sign. End-of-life discussions are also important in such cases because the prognosis of patients with chronic respiratory failure is generally poor.

  Grade A References A1. Yang H, Sawyer AM. The effect of adaptive servo ventilation (ASV) on objective and subjective outcomes in Cheyne-Stokes respiration (CSR) with central sleep apnea (CSA) in heart failure (HF): a systematic review. Heart Lung. 2016;45:199-211. A2. Masa JF, Corral J, Caballero C, et al. Non-invasive ventilation in obesity hypoventilation syndrome without severe obstructive sleep apnoea. Thorax. 2016;71:899-906. A3. O’Connor CM, Whellan DJ, Fiuzat M, et al. Cardiovascular outcomes with minute ventilationtargeted adaptive servo-ventilation therapy in heart failure: the CAT-HF trial. J Am Coll Cardiol. 2017;69:1577-1587.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 80  Disorders of Ventilatory Control  

GENERAL REFERENCES 1. Orr JE, Malhotra A, Sands SA. Pathogenesis of central and complex sleep apnoea. Respirology. 2017;22:43-52. 2. Grimm W, Kesper K, Cassel W, et al. Cheyne-Stokes respiration during wakefulness in patients with chronic heart failure. Sleep Breath. 2017;21:419-426. 3. Mansukhani MP, Kolla BP, Naessens JM, et al. Effects of adaptive servoventilation therapy for central sleep apnea on health care utilization and mortality: a population-based study. J Clin Sleep Med. 2019;15:119-128. 4. DiMarco AF. Diaphragm pacing. Clin Chest Med. 2018;39:459-471.

527.e1

5. Weese-Mayer DE, Rand CM, Zhou A, et al. Congenital central hypoventilation syndrome: a bedsideto-bench success story for advancing early diagnosis and treatment and improved survival and quality of life. Pediatr Res. 2017;81:192-201. 6. Bishara J, Keens TG, Perez IA. The genetics of congenital central hypoventilation syndrome: clinical implications. Appl Clin Genet. 2018;11:135-144. 7. Diep B, Wang A, Kun S, et al. Diaphragm pacing without tracheostomy in congenital central hypoventilation syndrome patients. Respiration. 2015;89:534-538. 8. Piper A. Obesity hypoventilation syndrome: weighing in on therapy options. Chest. 2016;149: 856-868.

527.e2

CHAPTER 80  Disorders of Ventilatory Control  

REVIEW QUESTIONS 1. Which of the following is currently the treatment of choice for CheyneStokes breathing in heart failure? A . Optimize medical therapy B. Nasal continuous positive airway pressure C. Nasal bilevel positive airway pressure D. Carotid body resection E. Uvulopalatopharyngoplasty Answer: A  The treatment of choice for Cheyne-Stokes breathing is currently optimization of medical therapy for the underlying heart failure. Trials of nasal continuous positive airway pressure have failed to improve outcome compared with usual care. Bilevel therapy has not been rigorously studied but may make the situation worse. Although the upper airway can sometimes narrow or collapse in central apnea, there is no role for uvulopalatopharyngoplasty in the absence of obstructive sleep apnea. 2. In which of the following conditions would the “loop gain” for reflexes that control breathing be expected to be greater than normal? A . Central congenital hypoventilation syndrome B. Obesity hypoventilation C. An acute exacerbation of chronic obstructive pulmonary disease D. Congestive heart failure E. During rapid eye movement sleep Answer: D  Loop gain refers to the overall instability in a negative feedback control system, such as the ventilatory control system, whose role is to maintain stable Paco2 levels. High loop gains (>1) lead to unstable breathing such as the waxing and waning pattern of Cheyne-Stokes breathing that is common in heart failure. The other factors listed all tend to decrease loop gain.

3. Which of the following is the gene associated with the central congenital hypoventilation syndrome? A . CFTR B. PHOX2B C. A1AT D. MECP2 E. BMP Answer: B The PHOX2B gene is the hallmark for the diagnosis of central congenital hypoventilation syndrome. The other genes have been associated with various respiratory conditions but not with the central congenital hypoventilation syndrome. 4. Which of the following is true regarding obesity-hypoventilation syndrome? A . Serum bicarbonate is a useful screening test. B. Leptin deficiency is generally seen in afflicted humans. C. Diaphragm pacing is first-line therapy. D. It is present in roughly 50% of patients with obstructive sleep apnea. E. Hypercapnia usually persists despite major weight loss. Answer: A  Serum bicarbonate is a useful screening test because it is elevated owing to a compensatory metabolic alkalosis in patients with chronic respiratory acidosis caused by chronic hypoventilation. Leptin is deficient in some animal models but rare in human obesity. Obstructive sleep apnea is common in the obesity-hypoventilation syndrome, and obesity-hypoventilation syndrome is seen in roughly 10% of obstructive sleep apnea. Weight loss usually improves gas exchange in these patients.

CHAPTER 81 Asthma  

527

81  ASTHMA JEFFREY M. DRAZEN AND ELISABETH H. BEL



DEFINITION

Asthma is a clinical syndrome of unknown etiology characterized by recurrent episodes of airway obstruction that resolve spontaneously or as a result of treatment. These changes occur in the setting of various types of airway inflammation that are thought to reflect specific endotypes of this clinical syndrome. Although airway obstruction is largely reversible, some changes in the asthmatic airway may be irreversible.  

EPIDEMIOLOGY

Asthma is an extremely common disorder affecting boys more commonly than girls and, after puberty, women slightly more commonly than men; approximately 8% of the adult population of the United States has signs and symptoms consistent with a diagnosis of asthma. Although most cases begin before the age of 25 years, new-onset asthma may develop at any time throughout life. An estimated 300 million people worldwide suffer from asthma, with 250,000 annual deaths attributed to the disease.1 The worldwide prevalence of asthma increased more than 50% in the latter half of the 20th century. In the first decade of the 21st century, the prevalence of wheezing in children increased by about 0.1% per year, although it seems to have reached a plateau by 2015. During the period from 1980 to 2010, the greatest increases in the prevalence of asthma have occurred in countries that adopted an “industrialized” lifestyle without contact with farm animals. In contrast, being raised in a farming environment in close contact with cows is associated with a much lower risk of asthma, independent of genetic factors.  

PATHOBIOLOGY

Genetics

In twin studies, asthma has about 60% heritability, indicating that both genetic and environmental factors are important in its etiology. Multiple genome-wide association studies have shown that there are multiple genetic loci associated with asthma. Each contributes only a small fraction of the total disease burden, and even when considered together, they can explain only a small fraction of the prevalence of asthma.

Pathology

The asthma syndrome is characterized by marked heterogeneity in clinical expression, environmental triggers, and immunopathologic mechanisms. Traditionally, clinicians have identified only two forms of asthma, atopic and non-atopic asthma, but it has become clear that this division into two subtypes is an oversimplification, and that multiple mechanisms also referred to as “endotypes” can lead to the “classic” signs and symptoms of asthma. Atopic or allergic asthma represents the most common and most studied subtype of asthma. It is characterized by the presence of immunoglobulin E (IgE) mediated sensitization to environmental allergens and can be observed in almost all school-aged children with asthma and about half of adult asthmatics. Inflammation and remodeling of the airways are characteristic pathologic features of atopic asthma. Remodeling changes include epithelial fragility, edema and hyperemia of the mucosa, and thickening of the subepithelial reticular basement due to deposition of collagen type III and IV. In more severe chronic asthma, the airway wall thickens as a result of hyperplasia of airway smooth muscle, hypertrophy of mucus secreting glands, and subepithelial angiogenesis. Infiltration of the airway wall with inflammatory cells including adaptive T helper-2 (Th2) cells, mast cells, and eosinophils is a general finding, and occurs even in the mildest cases of atopic asthma. Th2 cells initiate and propagate the inflammatory cascade associated with allergy by producing a variety of cytokines. These cytokines include interleukin (IL)-4, IL-5, and IL-13. IL-4 induces B-cell proliferation and is capable of stimulating these cells to produce IgE. IL-13 has similar effects of those to IL-4 with respect to B-cell stimulation and IgE production but, in addition,

CHAPTER 81 Asthma  

ABSTRACT

Asthma is the most common lung disease in the world. This chapter reviews the epidemiology, pathobiology, and treatment of asthma. Specific treatment scenarios are reviewed based on the severity of asthma symptoms and airflow derangements manifested by the patient.

KEYWORDS

asthma reversible airway obstruction β-agonist inhalers mild persistent asthma severe asthma asthma emergencies asthma in pregnancy

527.e3

528

CHAPTER 81 Asthma  

induces airway hyperresponsiveness, goblet cell metaplasia, and mucus hypersecretion. IL-5 plays a major role in the regulation of eosinophil formation, maturation, recruitment, and survival. Increased production of IL-5 may be related to the pathogenesis of severe eosinophilic asthma. Mast cells are found in close association with airway smooth muscle cells and are classically activated by allergens when IgE is bound to the high affinity IgE receptor. IgE-mediated mast cell degranulation and activation results in the release of histamine and in the generation of cysteinyl leukotrienes and prostaglandins, which contribute to bronchoconstriction and airway hyperresponsiveness. Eosinophils accumulate in the airways after allergen exposure in sensitized persons. Upon activation these cells release toxic granule products (e.g., major basic protein and eosinophil cationic protein) that can damage airway epithelium and nerves. These cells also produce lipid mediators such as leuko­ trienes and platelet activating factor, and a range of cytokines, growth factors, and chemokines. Non-atopic or intrinsic asthma develops in adulthood and accounts for less than 5% of cases of asthma. It is not associated with allergies, is often accompanied by chronic rhinosinusitis and nasal polyposis, and is poorly responsive to inhaled glucocorticoids. The eosinophilic airway inflammation observed in this type of asthma is not dependent on adaptive immunity but is likely caused by type 2 innate lymphoid cells (ILC2) that produce high amounts of IL-5 and IL-13. The growth and maturation of ILC2 cells are stimulated by epithelial cytokines (“alarmins”), which are induced by chronic exposure to pollutants, viruses, or fungi. ILC2 cells produce little IL-4, so there is no associated IgE response from B-cells. Patients with intrinsic asthma often have a severe form of asthma with elevated eosinophil counts in peripheral blood, and generally require high doses of systemic glucocorticoids to keep their asthma under control. Both atopic and intrinsic asthma are so called “Type-2-high” asthma phenotypes. However, cluster analyses of large cohorts of adult patients with asthma have identified at least three subtypes of asthma that are not characterized by the presence of Type 2 cytokines or eosinophilic airway inflammation. These “Type-2-low” subtypes of asthma are characterized by increased numbers of neutrophils in the airway or no airway inflammation at all. One increasingly recognized “Type-2-low” asthma phenotype is obesity-associated asthma. This type of asthma occurs later in life, usually in female patients without prior airway disease. The mechanism of obesity associated asthma is not well understood. Other “Type-2-low” phenotypes of asthma include certain forms of occupational asthma (Chapter 87) (e.g., isocyanate-induced asthma), neutrophilic asthma with subclinical microbial infection, and asthma in athletes (e.g., in endurance aquatic sports). In clinical practice in adults, different asthma phenotypes often coexist and interact within the same patient, thereby underscoring the need for a personalized and targeted management approach.

Physiological Changes in Asthma

Increased resistance to airflow is the hallmark of asthma; it is due to airway obstruction resulting from smooth muscle constriction, thickening of the airway epithelium, and free liquid within the airway lumen. Obstruction to airflow is manifested by increased airway resistance and decreased flow rates throughout the vital capacity. At the onset of an asthma attack, obstruction occurs at all airway levels; as the attack resolves, these changes are reversed— first in the large airways (i.e., mainstem, lobar, segmental, and subsegmental bronchi) and then in the more peripheral airways. This anatomic sequence of onset and reversal is reflected in the physiological changes observed during resolution of an asthmatic episode. Specifically, as an asthma attack resolves, flow rates first normalize at volumes high in the vital capacity and only later at volumes low in the vital capacity. Because asthma is largely an airway disease rather than an air space disease, no primary changes occur in the static pressurevolume curve of the lungs. However, during an acute attack of asthma, airway narrowing may be so severe as to result in airway closure, with individual lung units closing at a volume that is near their maximal volume. This closure results in a change of the pressure-volume curve such that for a given contained gas volume within the thorax, elastic recoil is decreased, which in turn further depresses expiratory flow rates. Exercise-induced bronchoconstriction, which is the transient narrowing of the airways after exercise, occurs frequently among athletes who may not have asthma or even have any respiratory symptoms. The mechanism is uncertain, but the key stimulus is probably airway dehydration because of increased ventilation.

Additional factors influence the mechanical behavior of the lungs during an acute attack of asthma. During inspiration in an asthma attack, the maximal inspiratory pleural pressure becomes more negative than the subatmospheric pressure of 4 to 6 cm H2O usually required for tidal airflow. The expiratory phase of respiration also becomes active as the patient tries to force air from the lungs. As a consequence, peak pleural pressures during expiration, which normally are, at most, only a few centimeters of water above atmospheric pressure, may be as high as 20 to 30 cm H2O above atmospheric pressure. The low pleural pressures during inspiration tend to dilate airways, whereas the high pleural pressures during expiration tend to narrow airways. During an asthma attack, the wide pressure swings, coupled with alterations in the mechanical properties of the airway wall, lead to a much higher resistance to expiratory airflow than to inspiratory airflow. The respiratory rate is usually rapid during an acute asthmatic attack. This tachypnea is driven not by abnormalities in arterial blood gas composition but rather by stimulation of intrapulmonary receptors with subsequent effects on central respiratory centers. One consequence of the combination of airway narrowing and rapid airflow rates is a heightened mechanical load on the ventilatory pump. During a severe attack, the load can increase the work of breathing by a factor of 10 or more and can predispose to fatigue of the ventilatory muscles. With respect to gas exchange, the patchy nature of asthmatic airway narrowing results in a maldistribution of ventilation (V) relative to pulmonary perfusion (Q). A shift occurs from the normal preponderance of V/Q units, with a ratio of near unity, to a distribution with a large number of alveolar-capillary units, with a V/Q ratio of less than unity. The net effect is to induce arterial hypoxemia. In addition, the hyperpnea of asthma is reflected as hyperventilation with a low arterial Pco2.  

CLINICAL MANIFESTATIONS

History

During an acute asthma attack, patients seek medical attention for shortness of breath accompanied by cough, wheezing, and anxiety. The degree of breathlessness experienced by the patient is not closely related to the degree of airflow obstruction but is often influenced by the acuteness of the attack. Dyspnea may occur only with exercise (exercise-induced asthma), after treatment with agents inhibiting the actions of cyclooxygenase 1 (aspirinexacerbated respiratory disease),2 after exposure to a specific known allergen (extrinsic asthma), or for no identifiable reason (intrinsic asthma). Variants of asthma exist in which cough, hoarseness, or inability to sleep through the night is the only symptom. Identification of a provoking stimulus through careful questioning helps establish the diagnosis of asthma and may be therapeutically useful if the stimulus can be avoided. Most patients with asthma complain of shortness of breath when they are exposed to rapid changes in the temperature and humidity of inspired air. For example, during the winter months in less temperate climates, patients commonly become short of breath on leaving a heated house; in warm humid climates, patients may complain of shortness of breath on entering a cold dry room, such as an air-conditioned theater. An important factor to consider in taking a history from a patient with asthma is the potential for occupational exposures in asthma (Chapter 87). Asthma that is brought on by occupational exposures is termed occupational asthma; preexisting asthma that is exacerbated by workplace exposures is termed workplace-exacerbated asthma. In reactive airway dysfunction syndrome, a single large exposure leads to a persistent asthma-like phenotype in a previously normal individual.

Physical Examination Vital Signs

Common features noted during an acute attack of asthma include a rapid respiratory rate (often 25 to 40 breaths per minute), tachycardia, and pulsus paradoxus (an exaggerated inspiratory decrease in the systolic pressure). The magnitude of the pulsus is related to the severity of the attack; a value greater than 15 mm Hg indicates an attack of moderate severity. Pulse oximetry, with the patient respiring ambient air, commonly reveals an oxygen saturation near 90%.

Thoracic Examination

Inspection may reveal that patients experiencing acute attacks of asthma are using their accessory muscles of ventilation; if so, the skin over the thorax may be retracted into the intercostal spaces during inspiration. The chest is usually hyperinflated, and the expiratory phase is prolonged relative to the

CHAPTER 81 Asthma  

inspiratory phase. Percussion of the thorax demonstrates hyperresonance, with loss of the normal variation in dullness due to diaphragmatic movement; tactile fremitus is diminished. Auscultation reveals wheezing, which is the cardinal physical finding in asthma but does not establish the diagnosis (Chapter 77). Wheezing, commonly louder during expiration but heard during inspiration as well, is characterized as polyphonic in that more than one pitch may be heard simultaneously (Video 81-1). Accompanying adventitious sounds may include rhonchi, which are suggestive of free secretions in the airway lumen, or rales, which should raise the suspicion of an alternative diagnosis and are indicative of localized infection or heart failure. The loss of intensity or the absence of breath sounds in a patient with asthma is an indication of severe airflow obstruction.  

DIAGNOSIS

Laboratory Findings Pulmonary Function Findings

A decrease in airflow rates throughout the vital capacity is the cardinal pulmonary function abnormality during an asthmatic episode.3 The peak expiratory flow rate (PEFR), the forced expiratory volume in the first second (FEV1), and the maximal mid-expiratory flow rate (MMEFR) are all decreased in asthma (Chapter 79). In severe asthma, dyspnea may be so severe as to prevent the patient from performing a complete spirogram. In this case, if 2 seconds of forced expiration can be recorded, useful values for PEFR and FEV1 can be obtained. Gradation of attack severity (Table 81-1) must be assessed by objective measures of airflow; no other methods yield accurate and reproducible results. As the attack resolves, the PEFR and the FEV1 increase toward normal together while the MMEFR remains substantially depressed; as the attack resolves further, the FEV1 and the PEFR may normalize while the MMEFR remains depressed (Fig. 81-1). Even when the attack has fully resolved clinically, residual depression of the MMEFR is not uncommon; this depression may resolve during a prolonged course of treatment. If the patient is able

TABLE 81-1 RELATIVE SEVERITY OF AN ASTHMATIC ATTACK AS INDICATED BY PEFR, FEV1, AND MMEFR TEST

PREDICTED VALUE (%)

SEVERITY OF ASTHMA

PEFR FEV1 MMEFR

>80 >80 >80

No spirometric abnormalities

PEFR FEV1 MMEFR

>80 >70

55-75

PEFR FEV1 MMEFR

>60 45-70 30-50

PEFR FEV1 MMEFR

100 IU/L E. All of the above Answer: C  Patients with severe asthma who have elevated eosinophils in peripheral blood, so called severe eosinophilic asthmatics, have shown to respond well to anti-eosinophil drugs such as monoclonal antibodies against interleukin-5. These drugs have been shown to reduce the exacerbation rate by about 50%. Lung function, PEF variability, or serum IgE are not good predictors of response to any of the new biologic agents against type-2 inflammation.

CHAPTER 82  Chronic Obstructive Pulmonary Disease  

535

82  CHRONIC OBSTRUCTIVE PULMONARY DISEASE JOHN REILLY



DEFINITION

Chronic obstructive pulmonary disease (COPD) is an umbrella term for a number of conditions that result in fixed airway obstruction and dyspnea on exertion. COPD is characterized by persistent respiratory symptoms and airflow limitation that is due to airway or alveolar abnormalities usually caused by significant exposure to noxious particles or gases. Nearly all patients have both the air space destruction associated with emphysema as well as the pathologic airway changes that are consistent with chronic bronchitis. However, subsets of the COPD population can differ in terms of natural history and response to therapeutic intervention. Daily cough and sputum for 3 months for two or more years is the qualifying definition for chronic bronchitis. Airflow limitation is defined as a ratio of forced expiratory volume in 1 second (FEV1) and the forced vital capacity (FVC) of less than 0.7 (FEV1/ FVC 90%, Fio2 ≤ 0.6 • If infiltrates on chest radiograph (COPD, asthma, PTE), start at Fio2 0.4 and adjust according to Spo2 (consider starting higher if pulmonary embolism is strongly suspected) 3. Ventilation • Tidal volume: begin with 8 mL/kg PBW (see Fig. 97-1 for formulas); decrease to 6 mL/kg PBW over a few hours if ARDS present (see Fig. 97-1); inspiratory flow rate of 50-60 L/min • Rate: begin with 10-20 breaths/min (10-15 if not acidotic; 20-30 if acidotic or ARDS and small volumes); adjust for pH; goal pH > 7.3 with maximal rate of 35; may accept higher Paco2 and lower pH goal if minute ventilation high 4. Secondary modifications • Triggering: in spontaneous modes, minimize diaphragm inactivity • Assessment of auto-PEEP, especially in patients with increased airways obstruction (e.g., asthma, COPD) • I/E ratio: 1 : 2-4, either set or as function of flow rate; and auto-PEEP 5. Monitoring • Clinical: blood pressure, ECG, observation of ventilatory pattern including assessment of dyssynchrony, effort or work by the patient (P0.1 below 3-4 cm H2O); assessment of airflow throughout expiratory cycle • Ventilator: tidal volume, minute ventilation, airway pressures (including auto-PEEP), total compliance • Arterial blood gases, pulse oximetry *Decisions within this algorithm will be influenced by the specific conditions of the individual patient. ARDS = acute respiratory distress syndrome; COPD = chronic obstructive pulmonary disease; ECG = electrocardiogram; Fio2 = fraction of inspired oxygen; HFOV = high-frequency oscillatory ventilation; I/E ratio = inspiratory-to-expiratory ratio; NIV = noninvasive ventilation; Pao2 = partial pressure of oxygen in arterial blood; PBW = predicted body weight; PCV = pressure-controlled ventilation; P0.1 = negative pressure generated against an occluded airway after 0.1 seconds, PEEP = positive end-expiratory pressure; PSV = pressure-support ventilation; PTE = pulmonary thromboembolism; Spo2 = arterial oxygen saturation by pulse oximetry.

most patients have a dependent zone that is consolidated, atelectatic, or fluid filled; a nondependent, often small, zone that looks normal; and a middle zone that has some collapsed regions that can be recruited to resemble the nondependent regions if high levels of airway pressure are transiently used (these approaches are called recruitment maneuvers). Gas exchange can often be improved by high tidal volumes, but at the expense of regional overdistention of those lung units that were not affected by the disease process itself—a treatment strategy that can lead to worse lung injury and poorer clinical outcomes. Ventilator-induced lung injury can be minimized by strategies that avoid or diminish regional lung overdistention by using smaller tidal volumes. For example, the use of lower tidal volumes (6 mL/kg predicted body weight) decreases mortality by 22% (from an absolute value of 40 to 31%) compared with 12 mL/kg predicted body weight despite the fact that the larger tidal volume provides higher values of PaO2 (Fig. 97-1). Lung protective strategies may also be useful in ventilated patients who are intubated for reasons other than ARDS (e.g., intubated ICU patients, patients undergoing major abdominal surgery, and patients with brain death whose lungs may be donated for transplantation). A7  A lung-protective strategy with limitation of tidal

CHAPTER 97  Mechanical Ventilation  

638.e1

Mechanical Ventilation

Biochemical injury (biotrauma) Epithelium/ interstitium

Cytokines, complement, PGs, LTs, ROS, proteases mφ

Bacteria

Biophysical injury • Shear • Overdistention • Cyclic stretch • Intrathoracic pressure

Alveolar-capillary permeability Cardiac output Organ perfusion

Neutrophils Distal Organ Dysfunction E-FIGURE 97-3.  Mechanisms by which mechanical ventilation may lead to distal organ dysfunction. LTs = leukotrienes; mφ = macrophages; PGs = prostaglandins; ROS = reactive oxygen species. (Modified from Slutsky AS, Tremblay LN. Multiple system organ failure: is mechanical ventilation a contributing factor? Am J Respir Crit Care Med. 1998;157: 1721-1725.)

639

CHAPTER 97  Mechanical Ventilation  

of mortality in ARDS patients than Vt, PEEP, or Pplat.11 The reason for this is that ΔP is equal to Vt/Respiratory system compliance(CRS) since CRS = Vt/ (Pplat − PEEP). Thus, ΔP is the tidal volume normalized to lung volume available for ventilation, since CRS is a reasonable surrogate for lung volume.

volume should be considered in patients who are at high risk for development of ARDS.

Positive End-Expiratory Pressure

PEEP traditionally has been used to improve oxygenation while reducing Fio2. Within the context of the current paradigm of trying to minimize iatrogenic complications of mechanical ventilation, PEEP is a therapy that can potentially minimize the injury caused by ventilation at low lung volumes by recruiting lung units and keeping them open. The critical issues are how to set the optimum PEEP level in an individual patient and how to determine whether the procedures to recruit the lung units and keep them open are less harmful than allowing the lung units to remain de-recruited. One experimental option is chest computed tomography to assess whether areas of the lung are recruited, but this technique is not practical for routine assessment. Oxygenation response to PEEP is also an indirect but imperfect assessment of the lung recruitability. Data are inconclusive regarding the benefits of higher (≈13 cm H2O) compared with lower (≈8 cm H2O) PEEP levels, and PEEP levels often are individualized on the basis of a PEEP/Fio2 table (Fig. 97-1). Higher PEEP levels are probably associated with decreased mortality in ARDS patients with Pao2/ Fio2 of less than 200 mm Hg but not in patients with higher Pao2/Fio2 ratios. PEEP guided by esophageal pressure measured by an esophageal balloon to keep PL at end expiration positive can significantly increase Po2 levels and respiratory compliance compared with treatment guided by a standard protocol. A recent study in patients with moderate to severe ARDS found that a strategy using a lung recruitment maneuver and titrating PEEP to relatively high levels increased mortality compared to a strategy with lower PEEP. A8 

Adjunctive Approaches for Ventilating ARDS Patients

Although the neuromuscular blocking agent cisatracurium was shown to decrease mortality in moderate-severe ARDS in a study in 2010 and a subsequent small meta-analysis, A9  a recent study A10  found no change in mortality. Given that the more current study was more than twice as large as all prior studies, and used a less heavily sedated control group, which is more in keeping with current practice, the routine use of neuromuscular blockers is not recommended in patients with moderate-severe ARDS. The use of prone position in patients with ARDS can improve oxygenation compared with the supine position by permitting a more even distribution of pleural pressure and ventilation, thereby reducing ventilator-induced lung injury and decreasing Fio2. Use of the prone position decreases mortality by about 15 percentage points in patients with PaO2/FIO2 below 150 mm Hg. A11  A critical factor in the use of the prone position is proper training of medical personnel in how to place patients safely in the prone position.  

Obstructive Airways Diseases

The major physiologic abnormality in patients with obstructive airways diseases (e.g., COPD, asthma) is an increase in airway resistance with tidal expiratory airflow limitation; patients may also have a concomitant increase in minute ventilation. These factors lead to dynamic hyperinflation, which is associated with numerous complications. Thus, the main goals in the ventilatory support of patients with obstructive airway diseases are to minimize auto-PEEP, to rest the respiratory muscles, to maintain adequate gas exchange, and to decrease the oxygen cost of breathing while simultaneously minimizing the iatrogenic complications of mechanical ventilation. These strategies allow time for the diagnosis and treatment of the primary cause of the exacerbation (Chapters 81

Driving Pressure (ΔP)

The optimal approach to picking the correct lung protective strategy is not known, but a recent individual patient data meta-analysis suggested that the driving pressure (ΔP = Plateau pressure (Pplat) − PEEP) was a better predictor

Ventilatory Strategy for Patients with ARDS* Goal 1: Low Vt /Pplat Initiation: Calculate PBW —Male: 50 + 2. 3 (height [inches] – 60) —Female: 45.5 + 2.3 (height [inches] – 60) Initiate volume assist control —start with 8 mL/kg, and to 6 mL/kg over a few hours

Keep Pplat (based on 0.5-sec pause) < 35 cm H2O If Pplat > 30 cm H2O, Vt by 1 mL/kg to 5 or 4 mL/kg If Pplat < 25 AND Vt < 6 mL/kg, Vt by 1 mL/kg until Pplat > 25 cm H2O OR Vt = 6 mL/kg If patient severely distressed and/or breath stacking, consider Vt to 7 or 8 mL/kg, as long as Pplat ≤ 30 cm H2O †

Goal 2: Adequate Oxygenation

Goal 3: Arterial pH Goal: pH: 7.30–7.45 Acidosis algorithm If pH 7.15–7.30 • set rate until pH > 7.30 or PaCO2 < 25 mm Hg (max RR = 35) • if RR = 35 & pH < 7.30 NaHCO3 may be given If pH < 7.15 • set RR to 35 • if set RR = 35 & pH < 7.15, Vt may be in 1 mL/kg steps until pH > 7.15 (Pplat target may be exceeded) Alkalosis algorithm If pH > 7.45 • set RR until patient RR > set RR (minimum set RR = 6/min)

Specific goal: PaO2 55-80 mm Hg or SpO2 88-95% Use only FIO2 /PEEP combinations shown below to achieve this target • if oxygenation is low, choose FIO2 /PEEP combination (from FIO2 /PEEP table) to the right • if oxygenation is high, choose FIO2 /PEEP combination to the left

FIO2/PEEP Table FIO2 PEEP

0.3

0.4

0.4

0.5

0.5

0.6

0.7

0.7

0.7

0.8

0.9

0.9

0.9

5

5

8

8

10

10

10

12

14

14

14

16

18

1.0 18–24

*Based on ARDS Network Algorithm † If compliance of the chest wall is markedly decreased (e.g., massive ascites), it may be reasonable or necessary (if the patient is very hypoxemic) to allow a Pplat >30 cm H2O.

FIGURE 97-1.  Ventilatory strategy for patients with the acute respiratory distress syndrome (ARDS) as proposed by the ARDSNetwork. Several caveats should be considered in using the low tidal volume strategy. (1) Tidal volume (Vt) is based on predicted body weight (PBW), not actual body weight; PBW tends to be about 20% lower than actual body weight. (2) The protocol mandates decreases in the Vt lower than 6 mL/kg of PBW if the plateau pressure (Pplat) is greater than 30 cm H2O and allows small increases in Vt if the patient is severely distressed or if there is breath stacking, as long as Pplat remains at 30 cm H2O or lower. (3) Because arterial carbon dioxide (CO2) levels will rise, pH will fall; acidosis is treated with increasingly aggressive strategies dependent on the arterial pH. (4) The protocol has no specific provisions for the patient with a stiff chest wall, which in this context refers to the rib cage and abdomen; in such patients, it seems reasonable to allow Pplat to increase to more than 30 cm H2O, even though it is not mandated by the protocol; in such cases, the limit on Pplat may be modified on the basis of analysis of abdominal pressure, which can be estimated by measuring bladder pressure. RR = respiratory rate; SpO2 = oxygen saturation based on pulse oximeter. FIO2 = fraction of inspired oxygen; PaCO2 = arterial partial pressure of carbon dioxide; PaO2 = arterial partial pressure of oxygen; PBW = predicted body weight; Pplat = plateau pressure; PEEP = positive end-expiratory pressure; Vt = tidal volume.

640

CHAPTER 97  Mechanical Ventilation  

and 82). When the patient is being assisted by the ventilator, adding external PEEP to match the patient’s intrinsic PEEP is essential to decrease work of breathing.

Noninvasive Ventilation

For patients who have acute respiratory failure resulting from an exacerbation of COPD and who require ventilatory support, the preferred approach is NIV if the patient is hemodynamically stable and does not need to be intubated to protect the airway. It is important to choose a comfortable mask and to reassure the patient because some patients find the mask difficult to tolerate. This strategy may be applied with several ventilation modes, including pressure support and positive end-expiratory airway pressure. The ventilation settings are adjusted to improve gas exchange and to ensure the patient’s comfort. Despite this approach, some patients with COPD require intubation and ventilation because of cardiac or respiratory arrest, agitation, increased sputum, worsening respiratory failure, or other concomitant severe disorders.

Intubation and Ventilation

The key goal after intubation of patients with airway obstruction is to minimize the detrimental effects of dynamic hyperinflation. The most effective way is to decrease the minute ventilation, even if this approach results in an increased Paco2—a strategy known as permissive hypercapnia (or controlled hypoventilation). Judicious use of sedation may decrease carbon dioxide production and improve patient-ventilator synchrony, although the avoidance of sedation can reduce the duration of ventilation and hospitalization. Care must be taken when using paralytic agents in patients who have asthma and who are also receiving corticosteroids, because their use may lead to prolonged muscle weakness, and difficulty with extubation, and slower post–intensive care unit recovery. Increasing expiratory time by use of a higher peak inspiratory flow may be somewhat helpful, but it is not nearly as effective as decreasing minute ventilation. What level of Paco2 (and pH) should be tolerated is not known with

certainty, but maintaining pH higher than approximately 7.20 is a reasonable target if the patient is not having side effects (e.g., arrhythmias, increasing right-sided heart failure), although lower values have been reported in clinical studies. In patients with COPD who are spontaneously breathing, the addition of external (set) PEEP at a level that is just less than what is necessary to overcome the auto-PEEP fully may not increase Pplat and may decrease the inspiratory effort that the patient needs to generate to initiate inspiratory airflow. This strategy does not appear to be as effective in patients with status asthmaticus, in whom it may cause an increase in Pplat because expiratory flow limitation is not the major mechanism of obstruction.

  DISCONTINUATION OF

MECHANICAL VENTILATION

Discontinuation of ventilatory support and extubation should occur as expeditiously as possible to minimize the iatrogenic consequences of intubation and mechanical ventilation.12 However, the advantages of early discontinuation must be balanced against the detrimental consequences if patients deteriorate and require urgent reintubation. Patients who fail extubation and require reintubation (around 15%) have a high mortality, which to some degree may be precipitated by the failed extubation itself. From the moment that mechanical ventilation is instituted, it is important that the clinician start planning for eventual discontinuation of ventilatory support. A key aspect of this approach is serial evaluation, with aggressive treatment of the factors contributing to the patient’s ventilatory dependence. The first reason for delayed extubation is the lack of systematic screening to determine if the patient is ready to be extubated. This can be performed by routine spontaneous breathing trials as soon as the patient meets a number of criteria (Fig. 97-2). The majority of patients (between 60 and 75%) will be successfully extubated on the first attempt. Several trials have demonstrated that protocols implemented by nonphysician health care professionals improved care and were associated with substantial cost savings compared with standard

Approach to Discontinuing Ventilation/Extubation 1. Daily assessment: Is patient ready for a spontaneous breathing trial? • General: resolving process, patient alert, no continuous sedation • Gas exchange: P/F > 200; FIO2 ≤ 50% • Hemodynamics: no vasopressors • Respiratory: PEEP ≤ 5-7 cm H2O

Evaluate and treat reversible causes of failure • Sedation, fluid status, myocardial ischemia, pain control, bronchodilator need, etc.

Yes 2. Initiate screening for spontaneous breathing trial (SBT): • Monitor patient with ECG, oximetry • Patient breathes spontaneously on T-piece, or on PSV of 5-7 cm H2O • Monitor physiological variables (RR, gas exchange, hemodynamics, subjective comfort) If patient physiologically stable, continue 3. Continue SBT for 30-120 minutes: Discontinue if any of the following occurs: • General: anxiety or sweating • Gas exchange: SpO2 < 88%; PaCO2 by >10 mm Hg • Hemodynamics: sustained HR changes of >± 20% OR HR > 140/min; SBP < 90 OR > 180 mm Hg • Respiratory: RR > 35/min for > 5 min; signs of WOB (paradoxical breathing, accessory muscles…)

If patient physiologically unstable

Reinstitute ventilation • Stable, nonfatiguing, comfortable

Failed criterion

Reintubate

Failure No failure criterion met 4. Extubate

5. Monitor

FIGURE 97-2.  Algorithm for assessing whether a patient is ready to be liberated from mechanical ventilation and extubated. ECG = electrocardiogram; HR = heart rate; PaCO2 = arterial partial pressure of carbon dioxide; PEEP = positive end-expiratory pressure; P/F = PaO2/FIO2 ratio; PSV = pressure support ventilation; RR = respiratory rate; SBP = systolic blood pressure; SpO2 = oxygen saturation based on pulse oximeter; WOB = work of breathing.

management approaches, even though the specifics of the protocols were different. A strategy that pairs spontaneous awakening, based on the interruption of sedatives, with spontaneous breathing trials improves extubation rates, reduces ICU length of stay, and decreases mortality by about one third. Apart from new complications or persistence of the initial insult, three factors contribute to inability to wean and prolong the time on the ventilator: (1) respiratory muscle dysfunction at the time of weaning, which in part may be caused by the ventilatory process; (2) cardiovascular factors and volume status that may lead to weaning-induced pulmonary edema via diastolic dysfunction or ischemia; (3) neurologic factors including residual sedation or the inability to protect the airways. The global approach to weaning suggests that screening should be performed using relatively simple criteria of stability (e.g., the patient has shown improvement in the underlying process), and that the patient should be hemodynamically stable with no or minimal need for vasopressors, and with oxygen requirements that can be met by face mask once the patient is extubated. If the patient meets these general criteria, a spontaneous breathing trial is recommended (Fig. 97-2); if the patient passes the trial, the patient can be extubated. Gradual weaning is not necessary; instead, patients should be assessed on a daily basis regarding their suitability for removal from ventilatory support, and if they are not ready, a comfortable, nonfatiguing form of mechanical ventilation should be used between the assessments. High-flow nasal cannula oxygen reduces the risk of reintubation within 72 hours compared with conventional oxygen therapy in patients at low risk for reintubation A12  and is as good as NIV for high risk patients. A13  Despite repeated attempts, a small group of patients (about 10 to 15%) will continue to be ventilator-dependent even after one week of weaning attempts.13 In these patients, a global approach including mobilization, nutrition, and therapies focused on psychological factors may be important14; specialized weaning centers may be useful.15 It is possible that respiratory muscle rehabilitation may help, but the evidence for such an approach is weak. Trials of simple tracheostomy masks, during which patients breathe unassisted through their tracheostomy for varying periods of time before being reconnected to the ventilator, may be the preferred approach for weaning.

  Grade A References A1. Girardis M, Busani S, Damiani E, et al. Effect of conservative vs conventional oxygen therapy on mortality among patients in an intensive care unit: the oxygen-ICU randomized clinical trial. JAMA. 2016;316:1583-1589. A2. Chu DK, Kim LH, Young PJ, et al. Mortality and morbidity in acutely ill adults treated with liberal versus conservative oxygen therapy (IOTA): a systematic review and meta-analysis. Lancet. 2018;391:1693-1705. A3. Xu XP, Zhang XC, Hu SL, et al. Noninvasive ventilation in acute hypoxemic nonhypercapnic respiratory failure: a systematic review and meta-analysis. Crit Care Med. 2017;45:e727-e733. A4. Frat JP, Thille AW, Mercat A, et al. High-flow oxygen through nasal cannula in acute hypoxemic respiratory failure. N Engl J Med. 2015;372:2185-2196. A5. Reade MC, Eastwood GM, Bellomo R, et al. Effect of dexmedetomidine added to standard care on ventilator-free time in patients with agitated delirium: a randomized clinical trial. JAMA. 2016;315:1460-1468. A6. Young D, Harrison DA, Cuthbertson BH, et al. Effect of early vs late tracheostomy placement on survival in patients receiving mechanical ventilation: the TracMan randomized trial. JAMA. 2013;309:2121-2129. A7. Futier E, Constantin JM, Paugam-Burtz C, et al. A trial of intraoperative low-tidal-volume ventilation in abdominal surgery. N Engl J Med. 2013;369:428-437. A8. Cavalcanti AB, Suzumura EA, Laranjeira LN, et al. Effect of lung recruitment and titrated positive end-expiratory pressure (PEEP) vs low PEEP on mortality in patients with acute respiratory distress syndrome: a randomized clinical trial. JAMA. 2017;318:1335-1345. A9. Alhazzani W, Alshahrani M, Jaeschke R, et al. Neuromuscular blocking agents in acute respiratory distress syndrome: a systematic review and meta-analysis of randomized controlled trials. Crit Care. 2013;17:1-10. A10. Moss M, Huang DT, Brower RG, et al. Early neuromuscular blockade in the acute respiratory distress syndrome. N Engl J Med. 2019;380:1997-2008. A11. Guerin C, Reignier J, Richard JC, et al. Prone positioning in severe acute respiratory distress syndrome. N Engl J Med. 2013;368:2159-2168. A12. Hernández G, Vaquero C, González P, et al. Effect of postextubation high-flow nasal cannula vs conventional oxygen therapy on reintubation in low-risk patients: a randomized clinical trial. JAMA. 2016;315:1354-1361. A13. Hernandez G, Vaquero C, Colinas L, et al. Effect of postextubation high-flow nasal cannula vs noninvasive ventilation on reintubation and postextubation respiratory failure in high-risk patients: a randomized clinical trial. JAMA. 2016;316:1565-1574.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 97  Mechanical Ventilation  

GENERAL REFERENCES 1. Pham T, Brochard LJ, Slutsky AS. Mechanical ventilation: state of the art. Mayo Clin Proc. 2017;92:1382-1400. 2. Goligher EC, Ferguson ND, Brochard LJ. Clinical challenges in mechanical ventilation. Lancet. 2016;387:1856-1866. 3. Narendra DK, Hess DR, Sessler CN, et al. Update in management of severe hypoxemic respiratory failure. Chest. 2017;152:867-879. 4. Joshi N, Estes MK, Shipley K, et al. Noninvasive ventilation for patients in acute respiratory distress: an update. Emerg Med Pract. 2017;19:1-20. 5. Wedzicha JA, Miravitlles M, Hurst JR, et al. Management of COPD exacerbations: a European Respiratory Society/American Thoracic Society Guideline. Eur Respir J. 2017;49:1-16. 6. Mekontso Dessap A, Boissier F, Charron C, et al. Acute cor pulmonale during protective ventilation for acute respiratory distress syndrome: prevalence, predictors, and clinical impact. Intensive Care Med. 2016;42:862-870. 7. Nieman GF, Satalin J, Kollisch-Singule M, et al. Physiology in medicine: understanding dynamic alveolar physiology to minimize ventilator-induced lung injury. J Appl Physiol. 2017;122:1516-1522. 8. Fan E, Del Sorbo L, Goligher EC, et al. An official American Thoracic Society/European Society of Intensive Care Medicine/Society of Critical Care Medicine Clinical Practice Guideline: mechanical

641.e1

ventilation in adult patients with acute respiratory distress syndrome. Am J Respir Crit Care Med. 2017;195:1253-1263. 9. Brochard L, Slutsky A, Pesenti A. Mechanical ventilation to minimize progression of lung injury in acute respiratory failure. Am J Respir Crit Care Med. 2017;195:438-442. 10. Goligher EC, Dres M, Fan E, et al. Mechanical ventilation-induced diaphragm atrophy strongly impacts clinical outcomes. Am J Respir Crit Care Med. 2018;197:204-213. 11. Amato MB, Meade MO, Slutsky AS, et al. Driving pressure and survival in the acute respiratory distress syndrome. N Engl J Med. 2015;372:747-755. 12. Schmidt GA, Girard TD, Kress JP, et al. Official executive summary of an American Thoracic Society/ American College of Chest Physicians Clinical Practice Guideline: liberation from mechanical ventilation in critically ill adults. Am J Respir Crit Care Med. 2017;195:115-119. 13. Beduneau G, Pham T, Schortgen F, et al. Epidemiology of weaning outcome according to a new definition. The WIND study. Am J Respir Crit Care Med. 2017;195:772-783. 14. Dupuis S, Brindamour D, Karzon S, et al. A systematic review of interventions to facilitate extubation in patients difficult-to-wean due to delirium, agitation, or anxiety and a meta-analysis of the effect of dexmedetomidine. Can J Anaesth. 2019;66:318-327. 15. Jubran A, Grant BJ, Duffner LA, et al. Long-term outcome after prolonged mechanical ventilation: a long-term acute-care hospital study. Am J Respir Crit Care Med. 2019. [Epub ahead of print.]

641.e2

CHAPTER 97  Mechanical Ventilation  

REVIEW QUESTIONS 1. Which one of the following effects of positive end-expiratory pressure (PEEP) is NOT usually expected when used for a patient with the acute respiratory distress syndrome (ARDS)? A . Reduced right ventricle afterload B. Recruit alveolar regions C. Decrease ventilator-induced lung injury D. Increase oxygenation E. Increase functional residual capacity Answer: A  PEEP is expected to recruit the lung and increase lung volume, with beneficial effects on oxygenation and potentially a reduced risk of ventilatorinduced lung injury if the patient is recruitable. By increasing intrathoracic pressures, PEEP usually increases right ventricle afterload. In situations of major derecruitment, PEEP may recruit part of the pulmonary vasculature, which can decrease pulmonary resistance. However, under most clinical settings, increasing PEEP from a baseline level increases right ventricle afterload. 2. Which one of the following benefits is NOT to be expected in a patient with an exacerbation of chronic obstructive pulmonary disease (COPD) who is being treated with noninvasive ventilation (NIV): A . Unload the respiratory muscles B. Increase tidal volume C. Reduce the need for intubation D. Reduced infectious complications E. Reduced readmission rates Answer: E  NIV improves gas exchange by increasing tidal volume and unloading the respiratory muscles, thereby allowing the patient to avoid endotracheal intubation, with a subsequent reduction in infectious complications. NIV is therefore associated with less intubation and improved hospital survival, but there is no evidence that NIV reduces the rate of subsequent readmissions.

3. High-flow nasal cannula can provide short-term or outcome benefits in all of the following conditions except which one? A . Poor tolerance of face mask oxygen B. Postextubation period in low-risk patients C. Postextubation period in high-risk patients D. Exacerbation of COPD E. Hypoxemic respiratory failure due to pneumonia Answer: D  High flow oxygen is usually better tolerated than a classic Venturi mask to improve oxygenation and has been shown to have clinical benefits postextubation and in hypoxemia due to pneumonia. There is no good evidence concerning its role in obstructive lung disease.

CHAPTER 98  Approach to the Patient with Shock  

641

98  APPROACH TO THE PATIENT WITH SHOCK DEREK C. ANGUS



DEFINITION

Shock, which is an acute circulatory dysfunction that results in inadequate tissue perfusion, is a medical emergency requiring prompt diagnosis and intervention to prevent circulatory collapse, multisystem organ failure, and death. The four major categories of shock are cardiogenic, distributive, hypovolemic, and obstructive (Fig. 98-1). Cardiogenic shock (Chapter 92) is due to pump failure caused most commonly by acute myocardial ischemia, but also by acute valvular dysfunction, arrhythmias, myocarditis, or ventricular wall rupture. Distributive shock is caused by loss of peripheral vasomotor tone, most commonly due to sepsis (Chapter 100). Rarer causes of distributive shock include neurogenic shock, typically following major spinal cord injury (Chapter 371), and anaphylaxis (Chapter 238). Hypovolemic shock is caused by loss of circulating volume from dehydration, profound vascular leakage (e.g., secondary to the toxinmediated endothelial leak of dengue fever) or hemorrhage (e.g., secondary to trauma, perioperative, or gastrointestinal bleeding, or rupture of an aortic aneurysm). Obstructive shock, which is less common, is caused by intrinsic blockage of a major vessel, such as saddle pulmonary embolus (Chapter 74), or extrinsic compression, such as cardiac tamponade, status asthmaticus, or tension pneumothorax. Patients also can present with combined forms of shock. For example, septic shock can include loss of autonomic tone (distributive), myocardial depression (cardiogenic), and loss of vascular volume (hypovolemic) owing to dehydration and vascular leak.  

EPIDEMIOLOGY

In the United States, an estimated 500,000 incident cases of shock occur each year. Septic shock, which is the most common cause, accounts for more than 250,000 cases per year. Of the 1.6 million adults hospitalized with major trauma (Chapter 103) in the United States each year, about 100,000 (6%) present with shock, predominantly owing to hemorrhage.1 Of the 800,000 cases per year of acute myocardial infarction, about 40,000-50,000 (5%) develop cardiogenic shock (Chapter 99).2 Of the 200,000 or so patients who suffer a clinically significant pulmonary embolus each year, about 25,000 (13%) present in shock (or cardiac arrest). Thus, over half of all shock is due to sepsis (primarily distributive), one fifth is due to trauma (primarily hypovolemic), one tenth is due to coronary artery disease (cardiogenic), and the remaining (approximately 20%) is due to other causes. Shock occurs in both sexes, all age groups, and all racial and ethnic groups, with its incidence depending on the frequency of precipitating conditions within these populations.  

PATHOBIOLOGY

Shock is classically described in terms of cardiovascular pathophysiology when oxygen delivery to tissues is inadequate because of reduced cardiac output (hypovolemic, cardiogenic, or obstructive) or impaired vasomotor tone (distributive). However, these macrocirculatory changes are often accompanied by important pathobiologic changes in the cells and tissue beds of vital organs.

Cardiovascular Pathophysiology

In hypovolemic shock a decrease in circulating blood volume decreases venous return and thus cardiac preload. As preload decreases, the heart shifts left on the Starling curve to the point that stroke volume and, hence, both cardiac output and oxygen delivery fall. Decreases in cardiac output and oxygen delivery are detected by a complex system of baroreceptors and chemoreceptors that stimulate autonomic and endocrine-mediated compensatory changes in the heart and vasculature. Peripheral vasomotor tone is increased in arterial and venous capacitance vessels, thereby reducing total capacitance and effectively increasing the circulating volume, venous return, and preload. Selective autoregulatory changes also redistribute blood flow preferentially to vital organs (e.g., heart and brain) at the expense of the splanchnic, muscle, and skin vascular

CHAPTER 98  Approach to the Patient with Shock  

ABSTRACT

Shock, which is acute circulatory dysfunction that results in inadequate tissue perfusion, is a medical emergency requiring prompt diagnosis and intervention to prevent circulatory collapse, multisystem organ failure, and death. The four major categories of shock are cardiogenic, distributive, hypovolemic, and obstructive. Cardiogenic, hypovolemic, and obstructive shock are a result of reduced cardiac output, whereas distributive shock is caused by loss of vasomotor tone. Septic shock, which is predominantly distributive, accounts for over half of the 500,000 or so cases of shock per year in the United States. A presumptive diagnosis is based on hypotension, signs of hypoperfusion, and lactic acidosis. A rapid history, clinical examination, and bedside ultrasound help establish the probable type of shock, which can guide therapy as well as help identify the inciting cause and any concomitant problems. Respiratory support is provided with supplemental oxygen and mechanical ventilation. Obstructive shock requires specific procedures to relieve obstruction, such as needle thoracostomy for tension pneumothorax. Rapid intravenous fluid resuscitation with crystalloids, accompanied by vasopressors as necessary, is key for all forms of shock except cardiogenic shock. In cardiogenic shock, a small fluid bolus may still be necessary, but inotropic support is the primary strategy. The adequacy of resuscitation is determined by ensuring adequate cardiac preload via dynamic measures, such as straight leg raises, and monitoring blood pressure, lactate clearance, and clinical signs of hypoperfusion. Even when the initial resuscitation is successful, patients often will develop multisystem organ dysfunction that requires admission to intensive care for ongoing management.

KEYWORDS

shock hypotension hypoperfusion cardiac output vasomotor tone sepsis trauma

641.e3

642

CHAPTER 98  Approach to the Patient with Shock  

Shock

Type Etiology

Hypovolemic Major trauma Gastrointestinal bleed Dehydration

Cardiogenic Myocardial infarction Myocarditis Arrhythmia

Obstructive Pulmonary embolus Cardiac tamponade Tension pneumothorax Status asthmaticus

Distributive Sepsis Spinal cord injury Anaphylaxis

Cardiovascular findings Preload (filling pressures) Cardiac contractility Cardiac output Afterload (peripheral tone) Hemodynamic support Relieve obstruction Volume expansion Inotropes Vasopressors FIGURE 98-1.  The causes, cardiovascular findings, and hemodynamic support for different types of shock. Under cardiovascular findings, bidirectional arrows indicate variation in findings among patients with the particular type. Under hemodynamic support, the number of + signs indicate the importance of therapy. A combined + and − indicates that the intervention could help some patients but must be used with caution.

beds. In addition, sympathetic stimulation of the heart generates tachycardia and increases inotropy to increase cardiac output. The tradeoff initially sustains effective oxygen delivery. However, as blood volume decreases, compensation fails, and tissue hypoperfusion of vital organs ensues. In cardiogenic shock (Chapter 99), cardiac output falls because of primary pump failure rather than reduced preload. When left ventricular failure predominates, accompanying pulmonary edema (Chapter 52) can lead to arterial hypoxemia, which further compromises oxygen delivery. Peripheral and central compensatory mechanisms are similar to those in hypovolemic shock, with the caveat that increasing sympathetic drive to increase myocardial contractility has limited effect. In obstructive shock, the mechanism that reduces cardiac output depends on the cause: massive pulmonary embolus (Chapter 74) increases right ventricular afterload but reduces left ventricular preload, whereas tension pneumothorax (Chapter 92) and cardiac tamponade (Chapter 68) principally reduce cardiac output via reduced preload, akin to hypovolemic shock. As with cardiogenic shock, obstructive shock can be accompanied by impaired oxygenation, thereby further compromising oxygen delivery. In distributive shock, the autoregulation of peripheral vasomotor tone is impaired. For example, neurogenic shock following acute spinal cord injury (Chapter 371) arises owing to loss of sympathetic tone, thereby leading to dilatation of capacitance vessels, increased arterial-venous shunting, and loss of selective autoregulation. Preload is reduced, but afterload is also reduced. Compensatory mechanisms, such as reflex tachycardia, will depend on the level of the cord injury. Thus, the effects on cardiac output are variable (and cardiac output can even be increased), but effective oxygen delivery to vital organs can still be compromised. Septic shock (Chapter 100) is primarily a form of distributive shock triggered initially by the release of circulating mediators that have local effects on peripheral vessels, thereby causing both vasodilation and vascular leak. These effects impair autoregulation, increase capacitance, and reduce the absolute and effective circulating blood volume, thereby reducing preload and afterload, with variable effects on cardiac output. Inflammatory mediators released in sepsis also have direct myocardial depressant effects. Importantly, because of both impaired redistribution of flow and impaired tissue oxygen extraction (discussed below), distributive shock can persist

despite a seemingly adequate oxygen delivery, and blood returning from the peripheral beds can have a normal or elevated oxygen content.

Cellular and Organ Pathobiology

As oxygen delivery falls, hypoperfused tissue beds increase oxygen extraction, which decreases venous oxygen content but preserves aerobic metabolism. With further reduction in oxygen delivery, cells switch to glycolysis, thereby generating lactic acid and a base deficit. An acid environment helps oxygen dissociate from hemoglobin, partially offsetting the consequences of reduced oxygen delivery. Cells also decrease mitochondrial activity, a form of protective hypometabolism to decrease oxygen demand. However, these compensatory mechanisms can be overwhelmed, thereby resulting in deepening ischemia and acidosis. When cells are stressed or injured by ischemia, they release molecules that signal via damage-associated molecular pattern (DAMP) receptors, which stimulate innate immune cells (Chapters 32 and 33) to release a cascade of inflammatory mediators that alter endothelial function and permeability, coagulation and fibrinolysis, and leukocyte recruitment and activity. These changes disrupt microvascular flow and increase tissue edema and hypoxia. Tissue hypoxia also directly stimulates release of vasoactive mediators, such as nitric oxide and adenosine, thereby further contributing to vasomotor dysregulation. Reduced mitochondrial activity can also be maladaptive, with impaired oxygen utilization exacerbating cellular injury. After prolonged ischemia, reperfusion can aggravate these effects via reactive oxygen species, which induces local tissue ischemia-reperfusion injury. Thus, regardless of the initial cause of shock, prolonged tissue ischemia generates local changes characteristic of distributive shock, and these changes can persist for hours or days after resuscitation. Untreated, prolonged ischemia eventually exhausts ATP stores; as a result, tissue cells undergo bioenergetic failure, which triggers widespread apoptotic and necrotic cell death.

Genetic Susceptibility

Many conditions that predispose to shock have a multifactorial etiology that includes a genetic component. For example, genetic variants at over 50 loci are associated with coronary artery disease, the principal cause of cardiogenic

CHAPTER 98  Approach to the Patient with Shock  

shock. Individuals with a high genetic risk load (a composite measure of variation across the 50 genetic loci) have twice the risk of serious coronary events, such as acute myocardial infarction or cardiac-related death.3 Similarly, asthma (Chapter 81; a risk factor for obstructive shock) has a strong genetic basis, and recent genome-wide association studies show an association between single nucleotide polymorphisms (SNPs) in 5 genes (gasdermin B, interleukin-33, DNA repair protein RAD50, interleukin 1 receptor-like 1, and encoding cadherin-related family member 3) and the likelihood of being hospitalized for severe exacerbations of asthma, including status asthmaticus. Susceptibility and outcome of septic shock have also been linked to genetic variations, such as variable number tandem repeats, in genes involved in the detection of microbial products (e.g., mannose binding lectin-2 and Toll-like receptor (TLR)-4), activation of the inflammatory cascade (e.g., tumor necrosis factor-alpha), and downstream events, such as leukocyte recruitment (e.g., macrophage migration inhibitory factor-1), and coagulation activation (e.g., plasminogen activator inhibitor-1). For example, these relationships are welldocumented in sibling and familial studies of meningococcal disease (Chapter 282). Genetic variation has also been described in other components of the host response to shock, such as the sympathetic receptors in the peripheral vasculature that are responsible for regulating vasomotor tone (e.g., alpha 1B adrenergic receptor gene), the innate immune response to DAMPs (e.g., TLR signaling), and activation of adaptive immunity.4 These variations may affect the trajectory and outcome of shock, as well as the therapeutic response to various agents, such as adrenergic vasopressors, used to treat shock. However, none of the described variation has yet led to a specific clinical approach for the management of shock.

643

TABLE 98-1 PHYSICAL EXAMINATION AND SELECTED LABORATORY SIGNS IN SHOCK Central nervous system

Acute delirium, restlessness, disorientation, confusion, and coma, which may be secondary to decreased cerebral perfusion pressure (mean arterial pressure minus intracranial pressure). Patients with chronic hypertension or increased intracranial pressure may be symptomatic at normal blood pressures. Cheyne-Stokes respirations may be seen with severe decompensated heart failure. Blindness can be a presenting complaint or complication.

Temperature

Hyperthermia results in excess tissue respiration and greater systemic oxygen delivery requirements. Hypothermia can occur when decreased systemic oxygen delivery or impaired cellular respiration decreases heat generation.

Skin

Cool distal extremities (combined low serum bicarbonate and high arterial lactate levels) aid in identifying patients with hypoperfusion. Pallor, cyanosis, sweating, and decreased capillary refill and pale, dusky, clammy or mottled extremities indicate systemic hypoperfusion. Dry mucous membranes and decreased skin turgor indicate low vascular volume. Low toe temperature correlates with the severity of shock.

General cardiovascular

Neck vein distention (e.g., heart failure, pulmonary embolus, pericardial tamponade) or flattening (e.g., hypovolemia), tachycardia, and arrhythmias. Decreased coronary perfusion pressures can lead to ischemia, decreased ventricular compliance, and increased left ventricular diastolic pressure. A “mill wheel” heart murmur may be heard with an air embolus.

Before the overt expression of shock, the dominant clinical features are those of the inciting event, such as the chest pain of an acute myocardial infarction (Chapter 64) or the fever, cough, and purulent sputum production of a worsening pneumonia (Chapter 91). During these early phases, the body’s compensatory mechanisms to ensure adequate oxygen delivery will produce only subtle nonspecific signs and symptoms (Chapter 7). For example, the respiratory rate will typically rise to generate a compensatory respiratory alkalosis to the incipient metabolic acidosis. Sympathetic autonomic responses to absolute or relative hypovolemia include tachycardia, sometimes a temporary increase in blood pressure, and peripheral vasoconstriction, which can result in cool skin, peripheral cyanosis, and sluggish capillary refill. However, in patients with impending distributive shock, impaired vasomotor control may paradoxically present as warm, flushed peripheries.

Heart rate

Usually elevated. However, paradoxical bradycardia can be seen in patients with preexisting cardiac disease and severe hemorrhage. Heart rate variability is associated with poor outcomes.

Systolic blood pressure

May actually increase slightly when cardiac contractility increases in early shock and then fall as shock advances. A single episode of undifferentiated hypotension with a systolic blood pressure 10 mm Hg with inspiration) seen in asthma, cardiac tamponade, and air embolus.

Mean arterial blood pressure

Diastolic blood pressure + [pulse pressure/3]

Shock index

Heart rate/systolic blood pressure. Normal = 0.5 to 0.7. A persistent elevation of the shock index (>1.0) indicates impaired left ventricular function (as a result of blood loss or cardiac depression) and is associated with increased mortality.

Respiratory

Tachypnea, increased minute ventilation, increased dead space, bronchospasm, hypocapnia with progression to respiratory failure, acute lung injury, and adult respiratory distress syndrome.

Abdomen

Low-flow states may result in abdominal pain, ileus, gastrointestinal bleeding, pancreatitis, acalculous cholecystitis, mesenteric ischemia, and shock liver.

Renal

Because the kidney receives 20% of cardiac output, low cardiac output reduces the glomerular filtration rate and redistributes renal blood flow from the renal cortex toward the renal medulla, thereby leading to oliguria. Paradoxical polyuria in early sepsis may be confused with adequate hydration.

Metabolic

Respiratory alkalosis is the first acid-base abnormality, but metabolic acidosis occurs as shock progresses. Hyperglycemia, hypoglycemia, and hyperkalemia may develop.



CLINICAL MANIFESTATIONS

Early Manifestations

The cardinal features of overt shock are hypotension together with evidence of inadequate perfusion (Table 98-1). Typically, a systolic blood pressure lower than 90 mm Hg or mean blood pressure lower than 60 mm Hg overwhelms autoregulatory mechanisms so that selective vasoconstriction no longer can preserve adequate blood flow to vital organs. With decreased cerebral perfusion, confusion, disorientation, altered consciousness, or coma develop; focal or localizing signs are atypical and suggest an alternative diagnosis or preexisting cerebrovascular disease (Chapter 378) sensitive to low flow states. Oliguria develops but may not be noted immediately. Skin changes will be more apparent, and the patient will typically appear cold, clammy, and cyanotic, with loss or diminution of peripheral pulses (Chapter 7). Hypothermia is common, even in septic shock. Reliance of hypoperfused tissue beds on anaerobic metabolism leads to hyperlactatemic metabolic acidosis (Chapter 110) which stimulates increased respiratory effort; as the patient fatigues or mental status worsens, however, air hunger may give way to agonal breathing. Similarly, although tachycardia is typical, worsening hypotension and myocardial ischemia in advanced shock can result in paroxysmal bradycardia. Although peripheral cyanosis is a common early feature, central cyanosis is also common in cardiogenic and obstructive shock. Even in hypotensive or septic shock, profound oxygen extraction coupled with compromised respiration may promote central cyanosis. All organs are compromised, even if not clinically evident. For example, splanchnic and hepatic flow is severely diminished, often resulting in ileus and hepatic ischemia (Chapter 134).

Cryptic Shock

In some patients, signs of hypoperfusion and hyperlactatemia develop despite blood pressure that is temporarily normal, either because of pronounced

Courtesy of Emanuel P. Rivers, MD.

644

CHAPTER 98  Approach to the Patient with Shock  

vasoconstriction at the expense of compromised flow and hypoperfusion, or because the patient’s preexisting hypertension is masking an important decline relative to baseline blood pressure.

Untreated Shock

The length of time that a patient can survive in shock before diagnosis and intervention depends on the patient’s underlying fitness and the magnitude of the inciting event. For example, a patient with shock from major arterial hemorrhage may only survive minutes. A patient who suffers a massive acute myocardial infarction with ensuing cardiogenic shock may survive only a few hours. Similarly, gram-negative septic shock in patients with chemotherapyinduced neutropenia may also be fatal within hours. However, other patients may survive for several days with altered mental status, borderline hypotension, and oliguria. Regardless of the rate of decline, untreated shock progresses to catastrophic ischemia, acidemia, vital organ failure, and death. The patient will lapse into coma, potentially with seizures typical of anoxic brain injury; worsening myocardial ischemia will accelerate pump failure and potentially cause fatal arrhythmias; and ventilation will degenerate to agonal breaths or apnea.

Typical Progression of Treated or Partially Treated Shock

A period of prolonged tissue hypoperfusion can compromise the function of any vital organ or tissue bed. Thus, even if the patient is resuscitated, sequelae that may blossom over the following hours and days include acute kidney injury (Chapter 112), shock liver (Chapters 137 and 138), intestinal ileus and ischemia (Chapter 134), anoxic brain injury (Chapter 371), and acute respiratory distress syndrome (Chapter 96).  

DIAGNOSIS

should be presumed to have shock. Although the magnitude of hypoperfusion broadly correlates with the serum lactate level and even a minor elevation within the normal range (typically 120

>140

Blood pressure

Normal

Normal

Decreased

Decreased

Pulse pressure

Normal or increased

Decreased

Decreased

Decreased

Respiratory rate (per minute)

14-20

20-30

30-40

>35

Urine output (mL/hr)

>30

20-30

5-15

Negligible

Mental status

Slightly anxious

Mildly anxious

Anxious, confused

Confused, lethargic

Fluid replacement

Crystalloid

Crystalloid

Crystalloid and blood

Crystalloid and blood

*Estimates based on a 70-kg patient. From Committee on Trauma of the American College of Surgeons. Advanced Trauma Life Support for Doctors. Chicago: American College of Surgeons; 1997:108.

Although the primary cause of inadequate oxygen delivery is cardiovascular failure, supplemental oxygen can raise arterial oxygen content, thereby ameliorating tissue hypoxia. Supplemental oxygen by facemask is encouraged for all patients in suspected shock, with a greater emphasis placed on invasive mechanical ventilation (Chapter 97) via endotracheal intubation for patients with significant respiratory distress or severe hypotension or acidosis. A patient who is in shock and who is struggling to breathe despite supplemental oxygen should be intubated without a trial of noninvasive ventilation because it can be difficult to implement effectively, does not provide airway protection, and is less effective than invasive ventilation at reducing the work of breathing, facilitating gas exchange, and recruiting alveoli (Chapter 96). Invasive mechanical ventilation (Chapter 97) protects the airway, reduces the risk of aspiration, decreases the work of breathing (and therefore decreases oxygen demand), and improves gas exchange (to increase oxygen delivery and facilitate compensatory respiratory alkalosis). In cardiogenic shock (Chapter 99), mechanical ventilation provides the added benefit of improving alveolar recruitment and increasing transmural pressure on the heart, thereby reducing cardiac afterload and facilitating ventricular function. The major complication of initiating mechanical ventilation in patients with shock is inadvertent worsening of cardiac output and hypotension because the positive intrathoracic pressure associated with mechanical ventilation retards right ventricular filling. This complication is most likely in patients with hypovolemic shock (who are most dependent on preload), but distributive shock, obstructive shock due to extrinsic compression, and some instances of cardiogenic shock can all be preload-dependent. Many sedation agents typically used during intubation, such as etomidate or short-acting benzodiazepines, cause vasodilation and hypotension, thereby potentially exacerbating this problem.

647

CHAPTER 98  Approach to the Patient with Shock  

TABLE 98-4 VASOPRESSOR AGENTS PERIPHERAL VASCULATURE AGENT

DOSE RANGE

CARDIAC EFFECTS

VASOCONSTRICTION VASODILATION HEART RATE CONTRACTILITY DYSRHYTHMIAS

1-4  µg/kg/min

TYPICAL USE

0

1+

1+

1+

1+

5-10  µg/kg/min 11-20  µg/kg/min

1-2+ 2-3+

1+ 1+

2+ 2+

2+ 2+

2+ 3+

Vasopressin

0.04-0.1 units/min

3-4+

0

0

0

1+

Septic shock, post– cardiopulmonary bypass shock state, no outcome benefit in sepsis

Phenylephrine

20-200  µg/min

4+

0

0

0

1+

Vasodilatory shock; best for supraventricular tachycardia

Norepinephrine 1-20  µg/min

4+

0

2+

2+

2+

First-line vasopressor for septic shock, vasodilatory shock

Epinephrine

1-20  µg/min

4+

0

4+

4+

4+

Refractory shock, shock with bradycardia, anaphylactic shock

Dobutamine

1-20  µg/kg/min

1+

2+

1-2+

3+

3+

Cardiogenic shock, septic shock

Milrinone

37.5-75  µg/kg bolus followed by 0.375-0.75  µg/min

0

2+

1+

3+

2+

Cardiogenic shock, right-sided heart failure, dilates pulmonary artery; caution in renal failure

Dopamine

“Renal dose” does not improve renal function; may be used with bradycardia and hypotension Vasopressor range

Courtesy of Emanuel P. Rivers, MD.

A bolus of intravenous fluids (e.g., 500 mL normal saline) before or during intubation will frequently be necessary, together with judicious use of sedation agents, avoidance of excessive positive-pressure ventilation, and a potential increase in the dose of vasopressors.

Body Positioning

If the patient is very hypotensive, cardiac filling and cerebral perfusion can be augmented by having the patient lie flat or with the legs elevated (Trendelenburg position), especially patients in hypovolemic or distributive shock. Intubation and central line access are also facilitated by these positions. However, patients in severe respiratory distress, especially owing to pulmonary edema, usually cannot tolerate lying flat, and their hypoxemia may worsen. Similarly, until the airway is secured, a patient at risk of vomiting blood or gastric contents will be at higher risk of aspiration when managed flat. Furthermore, if a patient is suspected of having increased intracranial pressure (e.g., following major intracranial hemorrhage [Chapter 380], fulminant hepatic failure [Chapter 145], or traumatic brain injury [Chapter 371]), lying flat or in the Trendelenburg position could paradoxically worsen cerebral perfusion.

Other Emergency Procedures

A variety of other procedures must be initiated immediately in specific situations. For example, obstructive shock owing to cardiac tamponade (Chapter 68) will require emergency pericardial drainage, and a tension pneumothorax (Chapter 92) will require an emergency chest tube. A trauma patient (Chapter 103) may require a pelvic binder or other tourniquet, a bedside ultrasound to evaluate possible intra-abdominal bleeding, and an emergency computed tomographic scan or exploratory laparotomy, depending on the findings. Similarly, a patient with an obvious major upper gastrointestinal bleed (Chapter 126) requires emergent endoscopy. Patients with septic shock and right upper quadrant pain require intravenous antibiotics and an urgent ultrasound followed by endoscopic retrograde cholangiopancreatography if dilated biliary ducts or stones are seen (Chapter 146). Patients with cardiogenic shock (Chapter 99) may require emergent percutaneous coronary intervention (Chapter 65). In addition, problems that can frequently co-occur with shock, such as generalized seizures (Chapter 375), acute alcohol intoxication (Chapter 30), drug overdose (Chapters 31 and 102), and hyperglycemia (Chapter 216) require emergency evaluation and treatment.

Monitoring the Adequacy of Resuscitation

After the initial appraisal and management, the subsequent hours are crucial to confirm the diagnosis and cause of shock, as well as to monitor the effectiveness of resuscitation. Simple measures of the adequacy of resuscitation include restoration of normal blood pressure, improvement in the clinical signs of hypoperfusion (e.g., sustained production of urine output and improved mental status), and improvement in base deficit and hyperlactatemia. Although blood pressure can be measured by cuff manometry, the unreliability of this method in patients who are in profound shock commonly

requires placement of an arterial line if the patient does not improve rapidly. However, restoration of blood pressure does not mean that oxygen delivery has improved or that tissue perfusion has been restored. Clinical and biochemical improvement may lag improvement in oxygen delivery by hours, or it may be obfuscated either by medical treatment, such as the need for sedating agents, or the blossoming of acute organ dysfunction despite successful resuscitation. A pulmonary artery catheter can monitor right and pulmonary capillary wedge pressures, stroke volume, cardiac output, and mixed venous oxygen saturation. In shock states, inadequate oxygen delivery leads to increased peripheral oxygen extraction, and low venous oxygen saturation for venous blood returning to the right heart. However, multiple clinical trials failed to demonstrate better outcomes based on therapy guided by a pulmonary artery catheter, and it is no longer recommended in the routine management of shock.13 One way to assess the adequacy of fluid resuscitation is to determine the responses to maneuvers that increase preload. Examples include passive leg raising or administering a fluid challenge14 over 30 minutes. Several multicenter trials of early goal-directed therapy, which specifically tries to improve measures of hemoglobin and oxygenation, have failed to demonstrate benefit beyond standard care. For example, a low central venous pressure suggests hypovolemia, but treatment of septic shock based on central venous pressure readings and blood pressure targets does not improve outcomes. A9  Current recommendations therefore emphasize: an early suspicion for shock; a low threshold to check a serum lactate level; prompt antibiotics if sepsis is a possible cause (Chapter 100); and an initial bolus of 20 to 30 mL/kg of intravenous fluids, followed by close monitoring of the adequacy of resuscitation using the clinical examination, serial blood lactate assessments, or measures of preload dependency, such as a passive leg raise test.

Ongoing Supportive Care

Unless a patient has advanced directives to limit care, shock should be treated with the goal of saving the patient’s life. Patients who do not rapidly improve are typically admitted to an intensive care unit, where vital organ function can be monitored and appropriate support can be instituted. As soon as the patient is able or family members are available, conversations to understand preferences for cardiopulmonary resuscitation (Chapter 7), mechanical ventilation (Chapter 97), and other aspects of intensive care are extremely important (Chapter 3). The physician should provide the best possible information about the diagnosis, potential course, and prognosis.



PREVENTION

Shock can be prevented by preventing the events that cause it (e.g., trauma, sepsis, myocardial infarction) or intervening promptly in sick patients to prevent

their progression to shock. Many hospitals have instituted early warning systems to deploy rapid response teams for patients with abnormal vital signs (Chapter 7). Such programs can facilitate early assessment and treatment, and they may also improve outcomes.  

PROGNOSIS

Mortality from shock is now about 25%. The degree to which blood pressure, cardiac output, and measures of hypoperfusion, such as serum lactate, are responding to resuscitation is prognostic, and early evidence of multiorgan dysfunction is predictive of the need for prolonged intensive care and higher mortality. For patients who survive, the long-term prognosis is variable but is a function of age, chronic health, the natural history of the inciting event, and the duration and intensity of the hospital course. Interestingly, acute features of the shock itself (e.g., degree of hypotension) predict short-term mortality but are less predictive of long-term outcomes.

  Grade A References A1. Brass P, Hellmich M, Kolodziej L, et al. Ultrasound guidance versus anatomical landmarks for subclavian or femoral vein catheterization. Cochrane Database Syst Rev. 2015;1:CD011447. A2. Brass P, Hellmich M, Kolodziej L, et al. Ultrasound guidance versus anatomical landmarks for internal jugular vein catheterization. Cochrane Database Syst Rev. 2015;1:CD006962. A3. Rochwerg B, Alhazzani W, Sindi A, et al. Fluid resuscitation in sepsis: a systematic review and network meta-analysis. Ann Intern Med. 2014;161:347-355. A4. Semler MW, Self WH, Wanderer JP, et al. Balanced crystalloids versus saline in critically ill adults. N Engl J Med. 2018;378:829-839. A5. Holcomb JB, Tilley BC, Baraniuk S, et al. Transfusion of plasma, platelets, and red blood cells in a 1:1:1 vs a 1:1:2 ratio and mortality in patients with severe trauma: the PROPPR randomized clinical trial. JAMA. 2015;313:471-482. A6. Holst LB, Haase N, Wetterslev J, et al. Lower versus higher hemoglobin threshold for transfusion in septic shock. N Engl J Med. 2014;371:1381-1391. A7. Gordon AC, Mason AJ, Thirunavukkarasu N, et al. Effect of early vasopressin vs norepinephrine on kidney failure in patients with septic shock: the VANISH randomized clinical trial. JAMA. 2016;316:509-518. A8. Hajjar LA, Vincent JL, Barbosa Gomes Galas FR, et al. Vasopressin versus norepinephrine in patients with vasoplegic shock after cardiac surgery: the VANCS randomized controlled trial. Anesthesiology. 2017;126:85-93. A9. Rowan KM, Angus DC, Bailey M, et al. Early, goal-directed therapy for septic shock—a patient-level meta-analysis. N Engl J Med. 2017;376:2223-2234.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 98  Approach to the Patient with Shock  

GENERAL REFERENCES 1. Spaite DW, Hu C, Bobrow BJ, et al. The effect of combined out-of-hospital hypotension and hypoxia on mortality in major traumatic brain injury. Ann Emerg Med. 2017;69:62-72. 2. Benjamin EJ, Blaha MJ, Chiuve SE, et al. Heart disease and stroke statistics-2017 update: a report from the American Heart Association. Circulation. 2017;135:e146-e603. 3. Khera AV, Emdin CA, Drake I, et al. Genetic risk, adherence to a healthy lifestyle, and coronary disease. N Engl J Med. 2016;375:2349-2358. 4. Adefurin A, Ghimire LV, Kohli U, et al. Genetic variation in the alpha1B-adrenergic receptor and vascular response. Pharmacogenomics J. 2017;17:366-371. 5. Singer M, Deutschman CS, Seymour CW, et al. The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA. 2016;315:801-810. 6. Arshed S, Pinsky MR. Applied physiology of fluid resuscitation in critical illness. Crit Care Clin. 2018;34:267-277. 7. Messina A, Longhini F, Coppo C, et al. Use of the fluid challenge in critically ill adult patients: a systematic review. Anesth Analg. 2017;125:1532-1543.

648.e1

8. Perner A, Cecconi M, Cronhjort M, et al. Expert statement for the management of hypovolemia in sepsis. Intensive Care Med. 2018;44:791-798. 9. Cannon JW. Hemorrhagic shock. N Engl J Med. 2018;378:370-379. 10. Wise R, Faurie M, Malbrain M, et al. Strategies for intravenous fluid resuscitation in trauma patients. World J Surg. 2017;41:1170-1183. 11. Moller MH, Claudius C, Junttila E, et al. Scandinavian SSAI clinical practice guideline on choice of first-line vasopressor for patients with acute circulatory failure. Acta Anaesthesiol Scand. 2016;60:1347-1366. 12. Khanna A, English SW, Wang XS, et al. Angiotensin II for the treatment of vasodilatory shock. N Engl J Med. 2017;377:419-430. 13. Sionis A, Rivas-Lasarte M, Mebazaa A, et al. Current use and impact on 30-day mortality of pulmonary artery catheter in cardiogenic shock patients: results from the CardShock Study. J Intensive Care Med. 2019. [Epub ahead of print.] 14. Bentzer P, Griesdale DE, Boyd J, et al. Will this hemodynamically unstable patient respond to a bolus of intravenous fluids? JAMA. 2016;316:1298-1309.

648.e2

CHAPTER 98  Approach to the Patient with Shock  

REVIEW QUESTIONS 1. A 72-year-old woman weighing 60 kg presents to the emergency department with 2 days of fever, rigors, and difficulty breathing. Her blood pressure is 85/50, her heart rate is 110/min, and she appears uncomfortable and somewhat confused. A chest radiograph shows a lobar pneumonia. Her blood lactate level is 4.2 mmol/L. In addition to treating her with supplemental oxygen, additional steps should include: A . Insert a peripheral IV catheter, obtain a sputum culture, and start normal saline at 100 cc/hr. B. Insert a peripheral IV catheter, obtain a sputum culture, and give a bolus of 1.5 L of normal saline. C. Insert a peripheral IV catheter, obtain blood and sputum cultures, start broad-spectrum antibiotics, and give a bolus of 1.5 L of lactated Ringer solution. D. Insert a peripheral IV catheter, obtain blood and sputum cultures, start broad-spectrum antibiotics, and give normal saline at 100 cc/hr. E. Insert a central venous catheter, obtain blood and sputum cultures, start broad-spectrum antibiotics, and give normal saline at 100 cc/hr. Answer: C  See clinical features, diagnosis, management and treatment sections of “Approach to the patient with shock.” This patient has pneumonia and features of septic shock (hypotension, altered mental status suggesting hypoperfusion, an elevated serum lactate level). Priorities for care, in addition to providing respiratory support, include establishing intravenous access, initiating treatment for the pneumonia (which is the inciting cause of sepsis), and starting with a 20 to 30 cc/kg bolus of lactated Ringer solution, which is preferred over normal saline. Central venous access is not required if adequate peripheral access can be established. 2. Regardless of the original type of shock, persistent tissue hypoperfusion can promote features typical of which form of shock: A . Hypovolemic shock B. Cardiogenic shock C. Distributive shock D. Dissociative shock E. Obstructive shock Answer: C  See pathobiology section of “Approach to the patient with shock.” Prolonged hypoperfusion of vital tissue beds results in activation of an innate immune response, with direct release of local vasoactive substances, that together disrupt microvascular flow and interfere with regional vasomotor control, thereby producing a distributive shock component. 3. When resuscitating a patient with massive bleeding caused by trauma, coagulopathic bleeding can be reduced by the following strategy: A . Avoid all crystalloid fluids. B. Give 1 to 2 L crystalloid followed by plasma to platelets to packed cells in a ratio of 1 : 1 : 1. C. Give 1 to 2 L crystalloid followed by plasma to platelets to packed cells in a ratio of 1 : 1 : 3. D. Give 1 to 2 L crystalloid followed by plasma to platelets to packed cells in a ratio of 1 : 3 : 1. E. Avoid all crystalloids and give plasma to platelets to packed cells in a ratio of 1 : 1 : 3. Answer: B  See blood and blood products section of “Approach to the patient with shock.” Current recommendations for the management of trauma-related hemorrhage are to commence with 1 to 2 L crystalloid followed by plasma, platelets, and packed cells. The ratio of 1 : 1 : 1 reduces the likelihood of coagulopathy.

4. After six hours of resuscitating a 65-year-old man in septic shock with several liters of intravenous fluids and high-dose norepinephrine, his blood pressure is currently 90/65 mm Hg. However, his nurse reports that his urine output is poor, and he is requiring increasing FiO2 to maintain an arterial oxygen saturation greater than 90%. At this point, you should: A . Add low-dose dopamine to increase urine output. B. Conduct a straight leg raise test to assess his preload dependency. C. Insert a pulmonary artery catheter to determine venous oxygen saturation. D. Add dobutamine to increase cardiac output. E. Stop intravenous fluids to avoid worsening pulmonary edema. Answer: B  See management and treatment section of “Approach to the patient with shock.” This patient still requires vasopressors to maintain his blood pressure, so he is still in shock. He also has signs suggestive of worsening organ dysfunction, including oliguria and poor oxygenation. However, even though many liters of IV fluid have been administered, it is unclear if he is adequately fluid-resuscitated. In particular, it is important to establish that he is preload independent. A straight leg raise can be used as a dynamic measure of preload dependency. Low-dose dopamine is not helpful in septic shock. A pulmonary artery catheter provides many measures of central hemodynamics, but a static measure of venous oxygen saturation will not help determine preload dependency; furthermore, trials using the pulmonary artery catheterization in this setting have not shown improved outcome. Myocardial depression can occur in sepsis, so the patient may benefit from the judicious use of dobutamine. However, randomized trials of the routine use of inotropic agents have not shown improved outcomes, and dobutamine would be appropriate only if it was clear that he had impaired myocardial contractility (e.g., by a bedside echocardiogram). Until preload independency or fluid overload has been demonstrated, it would be very unwise to discontinue intravenous fluids. 5. The most common cause of shock in the United States is: A . Sepsis B. Major trauma C. Acute myocardial infarction D. Pulmonary embolus E. Status asthmaticus Answer: A  See epidemiology section of “Approach to the patient with shock.” Septic shock accounts for over half of all cases of shock in the United States, followed by trauma and acute myocardial infarction. Other causes are considerably more rare.

648

CHAPTER 99  Cardiogenic Shock  

TABLE 99-1 DIAGNOSIS OF CARDIOGENIC SHOCK CLINICAL SIGNS Hypotension Oliguria Clouded sensorium Cool and mottled extremities HEMODYNAMIC CRITERIA Systolic blood pressure < 90 mm Hg or >30 mm Hg decrease from baseline for >30 minutes Cardiac index < 2.2 L/min/m2 Pulmonary capillary wedge pressure > 18 mm Hg OTHER Documented myocardial dysfunction Exclusion of hypovolemia, hypoxemia, and acidosis

Rupture/tamponade, 1.4% Acute MR, 6.9% VSD, 3.9% RV shock, 2.8% Other, 6.5%

LV failure, 78.5%

FIGURE 99-1.  Causes of cardiogenic shock in patients with myocardial infarction in the SHOCK trial registry. LV = left ventricular; MR = mitral regurgitation; RV = right ventricular; VSD = ventricular septal defect. (Modified from Hochman JS, Buller J, Sleeper LA, et al. Cardiogenic shock complicating acute myocardial infarction—etiologies, management and outcome: a report from the SHOCK Trial Registry Investigators. J Am Coll Cardiol. 2000; 36:1063-1070).

99  CARDIOGENIC SHOCK STEVEN M. HOLLENBERG



DEFINITION

Cardiogenic shock occurs when the heart is unable to deliver enough blood to maintain adequate tissue perfusion. Cardiogenic shock is a hemodynamic syndrome defined by sustained systemic hypotension (systolic BP < 90 mm Hg), pulmonary capillary wedge pressure (PCWP) greater than 18 mm Hg, and cardiac index less than 2.2 L/minute/m2 (Table 99-1). The diagnosis of cardiogenic shock is often made on clinical grounds—hypotension combined with signs of poor tissue perfusion, including oliguria, clouded sensorium, and cool extremities, all in the setting of myocardial dysfunction. To make the diagnosis, it is important to document myocardial dysfunction and to exclude or correct factors such as hypovolemia, hypoxemia, and acidosis.  

EPIDEMIOLOGY

The predominant cause of cardiogenic shock (Fig. 99-1) is left ventricular failure secondary to an extensive acute myocardial infarction (MI) or cumulative loss of myocardial function in a patient with previous MI. However, any cause of severe left ventricular (LV) or right ventricular (RV) dysfunction can lead to cardiogenic shock, including fulminant myocarditis (Chapter 54), end-stage cardiomyopathy (Chapter 54), a mechanical complication of an

acute MI (Chapter 64), or prolonged cardiopulmonary bypass (Table 99-2). Stress-induced (takotsubo) cardiomyopathy may also present with cardiogenic shock (Chapter 54). Acute valvular regurgitation from endocarditis (Chapter 67) or chordal rupture (Chapter 66) can lead to shock, as can physiologic stress in the setting of severe valvular stenosis. Cardiac tamponade (Chapter 68) and massive pulmonary embolism (Chapter 74) with acute RV failure can cause shock without pulmonary edema. An important consideration is that some cardiogenic shock may have an iatrogenic component because of medications that exacerbate hypotension. Early diagnosis of impending shock and identification of patients at high risk for development of shock is essential, both to speed intervention and to avoid therapies that may worsen hemodynamics. After a decline over the past two decades, the incidence of cardiogenic shock complicating acute MI appears to be increasing for unclear reasons. However, the mortality associated with cardiogenic shock continues to fall as effective early treatment and more widespread adoption of early revascularization have improved outcomes.1 Only about 25% of patients who develop cardiogenic shock are in shock when they initially present to the hospital; in the others, shock usually evolves over several hours. Patients with early and late shock show similar demographic, historical, clinical, and hemodynamic characteristics. Risk factors for the development of cardiogenic shock in MI parallel those for LV dysfunction and the severity of coronary artery disease (CAD). Patient characteristics include older age, anterior MI, diabetes, hypertension, multivessel CAD, previous MI, and peripheral vascular or cerebrovascular disease. Clinical risk factors include decreased ejection fraction, larger infarctions, and lack of compensatory hyperkinesis in myocardial territories remote from the infarction. Clinical harbingers of impending shock include the degree of

CHAPTER 99  Cardiogenic Shock  

ABSTRACT

Cardiogenic shock is the syndrome that ensues when the heart is unable to deliver enough blood to maintain adequate tissue perfusion. Acute myocardial infarction is the leading cause, but other potential etiologies need to be considered. The pathogenesis of cardiogenic shock is a “downward spiral” in which myocardial dysfunction reduces stroke volume, cardiac output, and blood pressure; these changes compromise myocardial perfusion, exacerbate ischemia, and further depress myocardial function, cardiac output, and systemic perfusion. Evaluation and therapy must begin simultaneously. Echocardiography should be performed promptly to assess overall and regional systolic function and to exclude mechanical causes of shock. Invasive hemodynamic monitoring is often useful in patients who do not respond to initial therapy because clinical estimates of filling pressures can be unreliable, hemodynamic status can change precipitously, and concomitant right ventricular dysfunction is often under-recognized. Vasopressor support may be needed to break the vicious circle of progressive hypotension and further myocardial ischemia. Patients who do not respond rapidly to inotropic agents may be considered for mechanical support. Routine use of an intra-aortic balloon pump (IABP) is not effective. Percutaneous cardiac assist devices provide better hemodynamics compared with IABPs but have not been shown to improve mortality. Nevertheless, survival rates are improving as advances in supportive therapy and reperfusion are applied in clinical practice.

KEYWORDS

cardiogenic shock coronary revascularization vasopressors inotropic therapy hemodynamic monitoring IABP mechanical circulatory support

648.e3

CHAPTER 99  Cardiogenic Shock  

TABLE 99-2 CAUSES OF CARDIOGENIC SHOCK ACUTE MYOCARDIAL INFARCTION Pump failure Large infarction Smaller infarction with preexisting left ventricular dysfunction Infarct extension Reinfarction Infarct expansion Mechanical complications Acute mitral regurgitation due to papillary muscle rupture Ventricular septal defect Free wall rupture Pericardial tamponade Right ventricular infarction CARDIOMYOPATHY Myocarditis Peripartum cardiomyopathy End-stage low-output heart failure Hypertrophic cardiomyopathy with outflow tract obstruction Stress cardiomyopathy VALVULAR HEART DISEASE Acute mitral regurgitation (chordal rupture) Acute aortic regurgitation Aortic or mitral stenosis with tachyarrhythmia or other comorbid condition causing decompensation Prosthetic valve dysfunction TACHYARRHYTHMIA OTHER CONDITIONS Prolonged cardiopulmonary bypass Septic shock with severe myocardial depression Penetrating or blunt cardiac trauma Orthotopic transplant rejection Massive pulmonary embolism Pericardial tamponade

hypotension and tachycardia at hospital presentation. The factors that predict mortality after cardiogenic shock reflect the severity of the acute insult as well as comorbid conditions. Coronary angiography most often demonstrates multivessel CAD, with left main stenosis in 30% of patients and three-vessel coronary disease in 60%. Multivessel CAD may help explain failure to develop compensatory hyperkinesis in remote myocardial segments.  

PATHOBIOLOGY

Cardiogenic shock is characterized by a downward cascade in which myocardial dysfunction reduces stroke volume, cardiac output, and blood pressure; these changes compromise myocardial perfusion, exacerbate ischemia, and further depress myocardial function, cardiac output, and systemic perfusion.2 Concomitant diastolic dysfunction increases left atrial pressure, which leads to pulmonary congestion and hypoxemia that can exacerbate myocardial ischemia and impair ventricular performance. Compensatory mechanisms include sympathetic stimulation, which increases heart rate and contractility, thereby raising cardiac output but also increasing myocardial oxygen demand. Compensatory vasoconstriction can increase blood pressure, but it also increases myocardial afterload, further impairing cardiac performance and increasing myocardial oxygen demand. In the face of inadequate perfusion, this increased demand can worsen ischemia and perpetuate a vicious circle that, if unbroken, may culminate in death. Interruption of this cycle of myocardial dysfunction and ischemia is the basis for therapeutic regimens for cardiogenic shock. Patients with cardiogenic shock do not always have severe LV dysfunction, so mechanisms other than primary pump failure are often operative. Furthermore, systemic vascular resistance is not always elevated, suggesting that compensatory vasoconstriction is not universal. Inflammatory responses may contribute to vasodilation and myocardial dysfunction in cardiogenic shock. Patients in cardiogenic shock may have areas of nonfunctional but viable myocardium because of myocardial stunning or hibernation. Myocardial stunning represents postischemic dysfunction that persists despite restoration of normal blood flow. Hibernating myocardial segments have persistently impaired function at rest owing to severely reduced coronary blood flow. Although

649

TABLE 99-3 CLINICAL SIGNS OF VOLUME STATUS AND PERFUSION SIGNS AND SYMPTOMS OF CONGESTION Orthopnea, paroxysmal nocturnal dyspnea Jugular venous distention Abdominojugular reflux Rales Hepatomegaly Edema Right upper quadrant tenderness POSSIBLE EVIDENCE OF LOW PERFUSION Narrow pulse pressure Obtundation Cool extremities Decreased exercise tolerance Renal/hepatic dysfunction

conceptually distinct, these two conditions may overlap. Repetitive episodes of myocardial stunning can occur in areas of viable myocardium subtended by a critical coronary stenosis. Contractile function of hibernating myocardium improves with revascularization, and the severity of the antecedent ischemic insult determines the intensity of stunning, thereby providing a rationale for reestablishing coronary patency in cardiogenic shock. The notion that some myocardial tissue may recover function emphasizes the importance of measures to provide hemodynamic support and minimize myocardial necrosis in patients with shock.  

CLINICAL MANIFESTATIONS

The physical examination should be geared toward characterizing the patient’s hemodynamic profile (Table 99-3) by evaluating congestion (“wet” or “dry”) and systemic perfusion (“cold” or “warm”). Signs of left-sided congestion (Chapter 52) include pulmonary rales, whereas jugular venous distention (see Fig. 45-1), peripheral edema, and ascites may indicate right-sided congestion. Most patients with cardiogenic shock present wet and cold. Patients with shock are usually ashen or cyanotic, and they have cool skin and mottled extremities. Jugular venous distention and pulmonary rales are usually present, although their absence does not exclude the diagnosis. A precordial heave resulting from LV dyskinesis may be palpable. The heart sounds may be distant, and third and fourth heart sounds are usually present. A systolic murmur of mitral regurgitation or a ventricular septal defect may be heard, but either complication can occur without an audible murmur (Chapter 64). Signs or symptoms of kidney, liver, intestinal, and cognitive dysfunction may be observed.3  

DIAGNOSIS

After recognizing the clinical manifestations of apparent cardiogenic shock, the clinician must confirm its presence and assess its cause while simultaneously initiating supportive therapy before irreversible damage to vital organs ensues. The clinician must balance overzealous pursuit of an etiologic diagnosis before achieving stabilization with overzealous empirical treatment without adequate characterization of the underlying pathophysiologic process. An electrocardiogram (ECG) should be performed immediately. In cardiogenic shock caused by acute MI, the ECG most commonly shows ST elevation, but ST depression or nonspecific changes are found in 25% of cases. If RV infarction is suspected, ST elevation in modified right-sided leads may be diagnostic (Chapter 64). An ECG showing Q waves or bundle branch block may suggest extensive disease. Other initial diagnostic tests include a chest radiograph, complete blood count, and measurement of arterial blood gases, electrolytes, and cardiac biomarkers. A chest film may demonstrate pulmonary edema or suggest an alternative diagnosis, as when a widened mediastinum indicates aortic dissection (Chapter 69).

Echocardiography

Echocardiography is an indispensable tool for confirming the diagnosis of cardiogenic shock and should be performed as early as possible, preferably with Doppler (Chapter 49). Echocardiography provides information about overall and regional systolic function, diastolic function, and valvular disease, and it can rapidly diagnose mechanical causes of shock such as

650

CHAPTER 99  Cardiogenic Shock  

papillary muscle rupture, acute ventricular septal defect, free wall rupture, and tamponade.

Right-Sided Heart Catheterization

If the history, physical examination, chest radiograph, and echocardiogram demonstrate systemic hypoperfusion, low cardiac output, and elevation of venous pressures, right heart catheterization may not be necessary for diagnosis. If there is any uncertainty, however, invasive monitoring can be quite useful to characterize hemodynamics and to exclude volume depletion, right ventricular infarction, and mechanical complications Right-sided heart catheterization is most useful, however, to optimize therapy in unstable patients. In such patients, clinical estimates of filling pressures can be unreliable, and optimal filling pressures may be even higher in individual patients with LV diastolic dysfunction. Changes in myocardial performance or therapeutic interventions, including revascularization, can change cardiac output and filling pressures precipitously. Concomitant right ventricular dysfunction is often underrecognized in patients with cardiogenic shock, and its importance underappreciated; right heart catheterization is the best and most expeditious way to assess right-sided hemodynamics in these patients. Measurement of cardiac output and mixed venous oxygen saturation can assess cardiac performance and help select patients for inotropic and/or mechanical support.

should be corrected. Relief of pain and anxiety can reduce excessive sympathetic activity, and decrease oxygen demand, preload, and afterload. Arrhythmias (Chapter 58) can reduce cardiac output and should be corrected promptly with antiarrhythmic drugs (see Table 58-6), cardioversion, or pacing (Chapter 60). If the cause is likely to be an acute MI, aspirin and heparin should be given immediately (Chapter 64).5 Some therapies routinely used in acute MI (e.g., nitrates, β-blockers, angiotensin-converting enzyme inhibitors) have the potential to exacerbate hypotension in cardiogenic shock and are associated with poorer outcomes in hypotensive patients. Consequently, these agents should be avoided in patients with a tenuous hemodynamic status until they stabilize. An initial assessment of fluid status and systemic perfusion should be performed.6 Ischemia produces diastolic dysfunction, so high filling pressures may be necessary to maintain stroke volume in some patients. Some patients may benefit from rapid bolus infusions of 100 to 200 mL of crystalloid titrated to clinical end points. Patients who do not respond rapidly to initial treatment should be considered for invasive hemodynamic monitoring to identify the filling pressure at which cardiac output is maximized. Maintenance of adequate preload is particularly important in patients with RV infarction. Following initial stabilization and restoration of adequate blood pressure, tissue perfusion should be assessed. If tissue perfusion is adequate but significant pulmonary congestion remains, diuretics (e.g., intravenous furosemide as a 20 to 40 mg bolus) may be employed. If tissue perfusion remains inadequate, inotropic support and/or mechanical support should be initiated.

Vasopressors and Inotropes

P

Maintenance of adequate blood pressure is essential to break the vicious circle of progressive hypotension and further myocardial ischemia. When arterial pressure remains inadequate, therapy with vasopressor agents, titrated not only to blood pressure but also to clinical indices of perfusion and mixed venous oxygen saturation, may be required.7 Norepinephrine (0.02 to 1.0 µg/kg/minute) acts primarily as a vasoconstrictor, has only a relatively mild inotropic effect, and increases coronary flow. It is preferable to dopamine, A1  A2  which acts as both an inotrope (at 3 to 10 µg/kg/minute) and vasopressor (at 10 to 20 µg/ kg/minute). Vasopressor infusions must be titrated carefully in patients with cardiogenic shock to maximize coronary perfusion pressure with the least possible increase in inotropy myocardial oxygen demand. Invasive hemodynamic monitoring with an arterial line and temporary right heart catheterization are advisable during initial titration of vasoactive agents. If tissue perfusion remains inadequate despite norepinephrine, inotropic therapy should be initiated. Dobutamine (2.5 to 20 µg/kg/minute), a selective β1-adrenergic receptor agonist, can improve myocardial contractility and increase cardiac output. Dobutamine is the initial agent of choice in patients with systolic blood pressures greater than 90 mm Hg, but it may exacerbate hypotension in some patients and can precipitate tachyarrhythmias. Milrinone (0.125 to 0.75 µg/kg/minute without loading dose), a phosphodiesterase inhibitor, has fewer chronotropic and arrhythmogenic effects than catecholamines, but it has a long half-life and can cause hypotension; it is usually reserved for situations in which all other agents have proved ineffective.

a

Intra-aortic Balloon Pumps (IABP)

TREATMENT  Initial Management

Initial stabilization of the patient with suspected cardiogenic shock includes venous access, supplemental oxygen, and continuous ECG monitoring (Fig. 99-2).4 Many patients require endotracheal intubation and mechanical ventilation (Chapter 97), not only to improve arterial blood gasses, but also to reduce the work of breathing and facilitate sedation. Electrolyte abnormalities

MANAGEMENT OF CARDIOGENIC SHOCK PATHWAY Resuscitation and medical therapy Inotropes/vasopressors Mechanical ventilation Cause-specific medical therapy

R

Emergent reperfusion (for Mi/ACS) PCI CABG Fibrinolysis

e

l

c o v e r

l Temporary mechanical support* IABP Peripheral VAD ECMO Implantable VAD

y

i a t i o n

Durable VAD

Transplant

Destination VAD

FIGURE 99-2.  Potential cardiogenic shock care pathway, care location, and care providers. ACS = acute coronary syndrome; CABG = coronary artery bypass graft; ECMO = extracorporeal membrane oxygenation; IABP = intra-aortic balloon pump; MI = acute myocardial infarction; PCI = percutaneous coronary intervention; VAD = ventricular assist device. *Consider temporary mechanical support before reperfusion in cases of refractory cardiac shock. (Adapted from van Diepen S, Katz JN, Albert NM, et al. Contemporary management of cardiogenic shock: a scientific statement from the American Heart Association. Circulation. 2017;136:e232-e268.)

,

IABPs reduce systolic afterload and augment diastolic perfusion pressure without increasing oxygen demand, but they do not significantly improve blood flow distal to a critical coronary stenosis. Despite a convincing hemodynamic rationale for their use, randomized trials have shown no improvement in 30-day, 1-year, or 6-year mortality with IABP insertion in patients who have cardiogenic shock and who undergo early revascularization for MI. A3  A4  Whether IABP insertion may still be reasonable to support occasional patients through a critical period of shock until definitive therapy is undertaken is not known. Failure of IABP to reverse hypoperfusion is a poor prognostic sign, and such patients should be considered for more aggressive mechanical support. ,

Reperfusion

Supportive therapy may improve blood pressure and cardiac output in cardiogenic shock, but rapid restoration of myocardial blood flow is the cornerstone of therapy for patients with cardiogenic shock due to MI (Chapter 64). Fibrinolytic therapy can restore patency of the infarcted artery and decreases the likelihood of progression to cardiogenic shock (see Table 64-6 in Chapter 64), but it is ineffective after cardiogenic shock has already developed. Prompt revascularization is the only intervention that consistently reduces mortality rates in patients with cardiogenic shock, and randomized trials suggest that about 13 patients will be saved at 1 year for each 100 patients treated. A5  Furthermore, most survivors will have good functional status. Outcomes are best when PCI is performed within 6 hours after the onset of symptoms, but survival benefits are still demonstrable up to 48 hours after the onset of MI and 18 hours after the onset of shock. In patients with multivessel disease, data suggest that stenting of the culprit lesion only is as good as immediate multivessel PCI, with a lower risk that the patient will require renal replacement therapy. A6  Patients over 75 years of age who are suitable for aggressive therapy also appear to benefit.

CABG surgery is more likely to provide complete revascularization and achieves long-term survival rates comparable to those of PCI, often despite worse coronary anatomy and a higher prevalence of diabetes. In practice, however, emergency CABG is performed less than 10% of the time. When cardiogenic shock results from mechanical complications of MI (Chapter 64), surgery is recommended when feasible. For acute mitral regurgitation due to papillary muscle rupture, vasoactive and mechanical support are temporizing measures; definitive therapy requires expeditious surgical valve repair or replacement (Chapter 66). Although mortality is 20 to 40%, surgical results are improving, and both survival and ventricular function are improved compared with medical therapy. Timely surgery is also critical in patients whose cardiogenic shock is caused by ventricular septal or free wall rupture. Because perforations are exposed to shear forces, the rupture site can expand abruptly. Repair can be technically difficult owing to the need to suture in areas of necrosis. Surgical mortality is 20 to 50% and is especially high for serpiginous inferoposterior ruptures, which are typically less well circumscribed than anteroapical ruptures. RV function is an important determinant of outcome in this setting. Timing of surgery has been controversial, but guidelines now recommend that operative repair be undertaken early, within 48 hours of the rupture. Placement of a septal occluding device may be helpful in selected patients.

Circulatory Support

Percutaneous mechanical circulatory support devices can potentially interrupt the downward spiral of myocardial dysfunction, hypoperfusion, and ischemia in cardiogenic shock, thereby allowing time for myocardial recovery.8 These devices, to different extents, increase arterial pressure and cardiac output, reduce LV filling pressures and afterload, and support coronary perfusion. They can provide short-term support as a bridge to recovery or to transplantation,9 and sometimes are used as chronic therapy when transplantation cannot be performed.10 For cardiogenic shock in acute MI, these devices improve hemodynamic function but have not been shown to reduce morbidity. A7  Extracorporeal membrane oxygenation (ECMO) provides cardiopulmonary support for both heart and lungs using a membrane oxygenator and an arterial return catheter. ECMO reduces RV and LV preload but increases LV afterload, and it can be useful for refractory cardiogenic shock.11 A key issue in choosing among mechanical support options is right ventricular function. Some patients with both right and left heart failure benefit from initial biventricular support. The left ventricle rarely fails alone for long, and assessment of right ventricular hemodynamics with invasive hemodynamic monitoring can be important to optimize mechanical support strategies.

Management of Special Conditions

At the end stage of a dilated or restrictive cardiomyopathy (Chapter 54), low cardiac output can result in cardiogenic shock. A search for reversible precipitating causes should be undertaken. Some patients will respond to inotropic therapy and will have a brief period of relative improvement. Appropriate candidates should be referred for evaluation for possible cardiac transplantation (Chapter 53) or mechanical support. A fully magnetically levitated circulatory pump can provide an 85% survival rate free of device surgery or disabling stroke in appropriately selected patients at 6 months. A8  LVADs can be used either as a bridge to transplantation or as destination therapy. A discussion about end-of-life care is also warranted. Acute myocarditis (Chapter 54) can take a fulminant course leading to shock in 10 to 15% of cases. Patients with acute myocarditis are typically younger than those with cardiogenic shock due to MI, and they more commonly present with dyspnea rather than chest pain. Echocardiography usually shows global LV dysfunction. Supportive therapy is indicated; some patients may require circulatory support and even consideration of cardiac transplantation. Immunosuppressive therapy has not been shown to improve outcome in fulminant myocarditis. Patients with hypertrophic cardiomyopathy (Chapter 54) may sometimes present with severe outflow tract obstruction and shock. In such patients, diuretics and inotropic therapy typically worsen the obstruction. Careful volume resuscitation and use of a pure α-agonist, such as phenylephrine (0.1 to 0.3 mg/ kg/minute), can reduce obstruction by increasing afterload and cavity size. Stress (takotsubo) cardiomyopathy (Chapter 54) presents with chest pain and ECG changes similar to acute MI; exclusion of significant coronary obstruction and characteristic apical hypokinesis with basal sparing establish the diagnosis. Some patients may have LV dysfunction severe enough to produce shock. Treatment is supportive. Most patients recover LV function within days to weeks, and the long-term prognosis is excellent. Acute mitral regurgitation (Chapter 66) presents with pulmonary edema and decreased forward cardiac output. Causes include papillary muscle rupture in acute MI, spontaneous chordal rupture, infective endocarditis (Chapter 67), and trauma (Chapter 103). The diagnosis is best made by echocardiography. Immediate stabilization may include inotropic or vasopressor therapy to support cardiac output and blood pressure. Definitive therapy, however, consists of surgical valve repair or replacement (Chapter 66). Acute aortic regurgitation most commonly results from infective endocarditis (Chapter 67) with leaflet destruction, but it may also be due to traumatic injury

(Chapter 103) or acute aortic dissection (Chapter 69). The pulse pressure is usually narrow, indicating decreased forward stroke volume, and the bounding pulsations seen with chronic aortic regurgitation are usually absent. Temporizing measures include afterload reduction, with vasopressor and inotropic support as needed. IABP is contraindicated, and excessive slowing of the heart rate may worsen hemodynamics by prolonging diastole. Definitive therapy is surgical.



PROGNOSIS

Cardiogenic shock is still the most common cause of death in acute MI. Survival rates are improving as a result of advances in supportive therapy and reperfusion in appropriately selected patients. Hemodynamics predict shortterm but not long-term mortality.12 Age and time to revascularization independently predict survival, but the benefits of revascularization are seen at every level of risk. Average 1-year survival after early revascularization is 50 to 55%, and the survival benefit is maintained at 6-year follow-up, with 5-year survival approaching 45%. The quality of life in survivors is usually excellent; 83% of patients are either asymptomatic or have only mildly symptomatic heart failure. For patients with end-stage nonischemic myocardial disease, the prognosis is very poor in the absence of heart transplantation or long-term mechanical support.13

  Grade A References A1. Schumann J, Henrich EC, Strobl H, et al. Inotropic agents and vasodilator strategies for the treatment of cardiogenic shock or low cardiac output syndrome. Cochrane Database Syst Rev. 2018;1:CD009669. A2. Rui Q, Jiang Y, Chen M, et al. Dopamine versus norepinephrine in the treatment of cardiogenic shock: a PRISMA-compliant meta-analysis. Medicine (Baltimore). 2017;96:1-8. A3. Unverzagt S, Buerke M, de Waha A, et al. Intra-aortic balloon pump counterpulsation (IABP) for myocardial infarction complicated by cardiogenic shock. Cochrane Database Syst Rev. 2015;3:CD007398. A4. Thiele H, Zeymer U, Thelemann N, et al. Intraaortic balloon pump in cardiogenic shock complicating acute myocardial infarction: long-term 6-year outcome of the randomized IABP-SHOCK II trial. Circulation. 2019;139:395-403. A5. Jeger RV, Urban P, Harkness SM, et al. Early revascularization is beneficial across all ages and a wide spectrum of cardiogenic shock severity: a pooled analysis of trials. Acute Card Care. 2011;13:14-20. A6. Thiele H, Akin I, Sandri M, et al. One-year outcomes after PCI strategies in cardiogenic shock. N Engl J Med. 2018;379:1699-1710. A7. Ouweneel DM, Eriksen E, Sjauw KD, et al. Percutaneous mechanical circulatory support versus intra-aortic balloon pump in cardiogenic shock after acute myocardial infarction. J Am Coll Cardiol. 2017;69:278-287. A8. Mehra MR, Naka Y, Uriel N, et al. A fully magnetically levitated circulatory pump for advanced heart failure. N Engl J Med. 2017;376:440-450.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 99  Cardiogenic Shock  

GENERAL REFERENCES 1. Kalavrouziotis D, Rodes-Cabau J, Mohammadi S. Moving beyond SHOCK: new paradigms in the management of acute myocardial infarction complicated by cardiogenic shock. Can J Cardiol. 2017;33:36-43. 2. Furer A, Wessler J, Burkhoff D. Hemodynamics of cardiogenic shock. Interv Cardiol Clin. 2017;6:359-371. 3. Harjola VP, Mullens W, Banaszewski M, et al. Organ dysfunction, injury and failure in acute heart failure: from pathophysiology to diagnosis and management. A review on behalf of the acute heart failure committee of the heart failure association (HFA) of the European Society of Cardiology (ESC). Eur J Heart Fail. 2017;19:821-836. 4. van Diepen S, Katz JN, Albert NM, et al. Contemporary management of cardiogenic shock: a scientific statement from the American Heart Association. Circulation. 2017;136:e232-e268. 5. Mebazaa A, Combes A, van Diepen S, et al. Management of cardiogenic shock complicating myocardial infarction. Intensive Care Med. 2018;44:760-773. 6. Chakravarthy M, Tsukashita M, Murali S. A targeted management approach to cardiogenic shock. Crit Care Clin. 2018;34:423-437. 7. Moller MH, Claudius C, Junttila E, et al. Scandinavian SSAI clinical practice guideline on choice of first-line vasopressor for patients with acute circulatory failure. Acta Anaesthesiol Scand. 2016;60: 1347-1366.

651.e1

8. Bonello L, Delmas C, Schurtz G, et al. Mechanical circulatory support in patients with cardiogenic shock in intensive care units: a position paper of the “unite de soins intensifs de Cardiologie” group of the French Society of Cardiology, endorsed by the “groupe atherome et cardiologie Interventionnelle” of the French Society of Cardiology. Arch Cardiovasc Dis. 2018;111:601-612. 9. den Uil CA, Akin S, Jewbali LS, et al. Short-term mechanical circulatory support as a bridge to durable left ventricular assist device implantation in refractory cardiogenic shock: a systematic review and meta-analysis. Eur J Cardiothorac Surg. 2017;52:14-25. 10. Miller LW, Rogers JG. Evolution of left ventricular assist device therapy for advanced heart failure: a review. JAMA Cardiol. 2018;3:650-658. 11. Khorsandi M, Dougherty S, Bouamra O, et al. Extra-corporeal membrane oxygenation for refractory cardiogenic shock after adult cardiac surgery: a systematic review and meta-analysis. J Cardiothorac Surg. 2017;12:1-13. 12. Mahmoud AN, Elgendy IY, Mojadidi MK, et al. Prevalence, causes, and predictors of 30-day readmissions following hospitalization with acute myocardial infarction complicated by cardiogenic shock: findings from the 2013-2014 national readmissions database. J Am Heart Assoc. 2018;7:1-13. 13. Shah M, Patel B, Tripathi B, et al. Hospital mortality and thirty day readmission among patients with non-acute myocardial infarction related cardiogenic shock. Int J Cardiol. 2018;270:60-67.

651.e2

CHAPTER 99  Cardiogenic Shock  

REVIEW QUESTIONS 1. Which of the following is the most common cause of cardiogenic shock? A . Pump failure in acute myocardial infarction B. Myocarditis C. Mechanical complications of acute myocardial infarction D. Right ventricular infarction E. Valvular heart disease Answer: A  The predominant cause of cardiogenic shock is left ventricular failure secondary to acute myocardial infarction. The other listed answers can lead to cardiogenic shock but are less common. 2. Coronary angiography in patients with cardiogenic shock in the setting of acute myocardial infarction most commonly shows which of the following? A . Left main coronary artery disease B. Single-vessel left anterior descending coronary artery disease C. Multivessel disease D. Extensive collateralization E. Two-vessel disease involving the right coronary artery Answer: C  In cardiogenic shock resulting from acute myocardial infarction, coronary angiography most often demonstrates multivessel disease. About 30% of patients have a left main coronary artery occlusion, about 60% have three-vessel coronary disease, and only about 20% have single-vessel disease. 3. Which of the following is true concerning the use of vasopressor agents in cardiogenic shock? A . Norepinephrine is more arrhythmogenic than dopamine. B. Norepinephrine is associated with decreased 28-day mortality compared with dopamine in patients with cardiogenic shock. C. Dopamine is preferable because it improves renal function. D. Dobutamine is more effective at raising blood pressure. E. Blood pressure should be monitored noninvasively to avoid vascular complications. Answer: B  In a randomized trial of patients with shock, norepinephrine reduced 28-day mortality in a prespecified subgroup of patients with cardiogenic shock compared with dopamine. Dopamine was more arrhythmogenic and does not improve renal function in patients with shock, although some patients have increased urine output. Dobutamine has vasodilatory effects and can worsen hypotension. Noninvasive blood pressure monitoring can be unreliable in patients with shock; arterial line placement is recommended.

4. Complete the following statement correctly: Percutaneous left ventricular assist devices for cardiogenic shock: A . Improve hemodynamics compared with intra-aortic balloon pumping. B. Improve 30-day mortality compared with intra-aortic balloon pumping. C. Should be reserved for patients eligible for cardiac transplantation. D. Are associated with decreased vascular complication rates compared with intra-aortic balloon pumping. E. Provide support independent of right ventricular function. Answer: A  Percutaneous left ventricular assist devices (LVADs) provide short-term hemodynamic support after cardiogenic shock. By allowing time for left ventricular recovery, they are usually intended as a bridge to definitive therapy, such as transplantation. Percutaneous LVADs provide better hemodynamics compared with an intra-aortic balloon pump, with higher cardiac indices and mean arterial pressures as well as lower filling pressures; however, they have not been shown to improve mortality at 30 days. Known complications of percutaneous LVADs include limb ischemia and bleeding. The available percutaneous devices support the left ventricle and thus require adequate right ventricular function. 5. Which of the following is true concerning prognosis after cardiogenic shock in the setting of myocardial infarction? A . Hemodynamics predict long-term mortality among patients undergoing revascularization. B. The early survival benefits of revascularization are lost by 5-year follow-up. C. Benefits of revascularization are seen only in younger patients. D. Prognosis is worse than in patients with end-stage nonischemic myocardial disease. E. Quality of life in survivors is usually excellent. Answer: E  The quality of life in survivors of cardiogenic shock complicating acute myocardial infarction is usually excellent, with 83% of patients either asymptomatic or having only mildly symptomatic heart failure. Hemodynamics predict short-term but not long-term mortality. The survival benefit of early revascularization is maintained at 6-year follow-up, with 5-year survival approaching 45%. Among patients undergoing revascularization, age and time to revascularization predict survival, but the benefits of revascularization are seen at every level of risk, with an average 1-year survival of 50 to 55%. For patients with end-stage nonischemic myocardial disease, the prognosis is very poor in the absence of heart transplantation.

CHAPTER 100  Shock Syndromes Related to Sepsis  

651

100  SHOCK SYNDROMES RELATED TO SEPSIS JAMES A. RUSSELL



DEFINITION

Sepsis is life-threatening organ dysfunction caused by a dysregulated host response to infection. Organ dysfunction is defined as an acute change of two or more points in the total Sequential Organ Failure Assessment (SOFA) score1 (Table 100-1). In addition to the full SOFA, a briefer quick SOFA— defined by respiratory rate of 22/min or less, altered mentation, and systolic blood pressure higher than 100 mm Hg—provides simple bedside criteria to screen quickly for sepsis. Bacteremia is defined as the growth of bacteria in blood cultures, but infection does not have to be proved to diagnose sepsis at the onset. Septic shock was also redefined in 2016 as a subset of sepsis with persisting hypotension requiring vasopressors to maintain a mean arterial pressure of 65 mm Hg or higher accompanied by a serum lactate level greater than 18 mg/dL (2 mmol/L), despite adequate volume resuscitation.

CHAPTER 100  Shock Syndromes Related to Sepsis  

ABSTRACT

The definition of septic shock is life-threatening organ dysfunction caused by a dysregulated host response to infection. Endotoxins, which are innate immune receptors, are shuttled from HDL to LDL and are cleared mainly by the liver via the LDL receptor. Widespread endothelial injury increases endothelial permeability. Septic shock activates the coagulation system and causes sepsisassociated coagulopathy. An immunosuppressed state, which can exceed the pro-inflammatory response starting about 3 to 4 days after the onset of sepsis, can be fatal. Cardiovascular features include vasodilation, hypovolemia, decreased ventricular contractility, and increased vascular permeability. Management includes oxygen (many patients require mechanical ventilation), early antibiotics (within 1 hour), fluid resuscitation (30 mL/kg crystalloid over first 3 hours followed by additional fluids), vasopressors (e.g., norepinephrine, 1 to 50 µg/minute if the mean arterial pressure is less than 65 mm Hg despite adequate fluid resuscitation), consideration of vasopressin and hydrocortisone if the response to vasopressors is inadequate, transfusion if the hemoglobin level is less than 7 g/dL, and dobutamine if there is ventricular dysfunction. Early goal-directed therapy is no better than usual care. Continuous renal replacement therapy may be preferable to intermittent hemodialysis in patients with hemodynamic instability. The 28-day mortality is about 20 to 25%. Early deaths are usually due to refractory septic shock. Later deaths (after day 3) are usually due to multiple organ dysfunction, nosocomial infection, or both.

KEYWORDS

septic shock antibiotics fluids norepinephrine vasopressin corticosteroids dobutamine

651.e3

652

CHAPTER 100  Shock Syndromes Related to Sepsis  

TABLE 100-1 SCREENING FOR SEPSIS AND SEPTIC SHOCK Step 1. Screen for sepsis using Quick SOFA (qSOFA): One point is given for each adverse finding as follows: 1. Respiratory rate ≥22/minute 2. Altered mentation—Glasgow Coma Score less than 15 3. Systolic blood pressure ≤100 mm Hg Step 2. Diagnose sepsis as a 2-point increase in SOFA as follows: Sequential (Sepsis-related) Organ Failure Assessment (SOFA). Step 3. Diagnose septic shock by the following: Use of vasopressors (e.g., norepinephrine, epinephrine, phenylephrine) to maintain a mean arterial pressure ≥65 mm Hg PLUS Serum lactate >18 mg/dL (2 mmol/L) despite adequate volume resuscitation SEQUENTIAL (SEPSIS-RELATED) ORGAN FAILURE ASSESSMENT SCORE SCORE SYSTEM

0

Respiratory (Pao2/Fio2)

≥400

Coagulation (platelet count × 103/µL) Liver (bilirubin mg/dL [µmol/L])

1

2

3

4

2 mmol/L and vasopressors to maintain a mean arterial pressure ≥65 mm Hg despite adequate volume resuscitation are required in the latest definition of septic shock (Sepsis-3). Tachypnea >22/min and tachycardia >100/min are components of the quick SOFA, not septic shock. There is no need for a second agent such as vasopressin or hydrocortisone. Both should be considered in patients who do not respond adequately to norepinephrine.

659

CHAPTER 101  Disorders Due to Heat and Cold  

101  DISORDERS DUE TO HEAT AND COLD MICHAEL N. SAWKA AND FRANCIS G. O’CONNOR

  TEMPERATURE REGULATION

Body temperature is regulated through two parallel processes that modify body heat balance: behavioral (clothing, shelter, physical activity) and physiologic (skin blood flow, sweating, shivering). Both peripheral (skin) and central (core) thermal receptors provide afferent input to a central nervous system integrator (hypothalamic thermoregulatory center), and any deviation between the controlled variable (body temperature) and a theoretical reference variable (“set point” temperature) results in a heat loss or conservation response. Humans normally regulate body (core) temperature at about 37° C (98.6° F), and fluctuations within the narrow range of 35° C to 41° C (95° F to 105.8° F) can be tolerated by healthy acclimatized persons; core temperatures outside this range can induce morbidity and mortality. By one estimate, about 8% of worldwide mortality can be attributed to high or low ambient environmental temperatures, with about 90% of the fatalities due to cold.1 There is no single core temperature because it varies at different deep body sites and during rest and physical exercise. Arterial blood temperature, which provides the best invasive measurement of core temperature, is slightly lower than brain temperature. The most accurate noninvasive index of core temperature is esophageal temperature, followed in order of preference by rectal, gastrointestinal tract (telemetry pill), and oral temperature. Ear (tympanic and auditory meatus) or scanned temporal artery temperature should not be relied on for clinical judgment. Rectal temperatures are most commonly recommended because they are easy to measure and are not biased by environmental conditions.

  HEAT ILLNESS  

DEFINITION

Minor heat-related illnesses include miliaria rubra, heat syncope, and heat cramps. Serious heat illness represents a spectrum from heat exhaustion to heat injury and heatstroke.

CHAPTER 101  Disorders Due to Heat and Cold  

ABSTRACT

Humans normally regulate body (core) temperature at about 37° C (98.6° F), and fluctuations within the narrow range of 35° to 41° C can be tolerated by healthy acclimatized persons. Rectal, esophageal, and oral measures of core temperature can be used for clinical judgment but not ear or scanned temporal artery temperature. Serious heat illness includes heat exhaustion, heat injury, and heatstroke. Heat illness accounts for substantial morbidity and mortality in the world and appears to be increasing in the United States. Management of serious heat illness should stress aggressive whole-body cooling (e.g., cool/ cold water immersion or skin soaking with accelerated evaporation), rehydration, and monitoring. Cold injuries are classified as hypothermia (mild, moderate, and profound) and peripheral cold injuries (nonfreezing and freezing), which often occur simultaneously. Moderate (core temperature < 32° C) and profound (core temperature < 26° C) cold require active rewarming with common complications including ventricular fibrillation. Frostbitten tissues should be protected from trauma and not thawed until confident that warmth can be maintained because refreezing causes additional injury. Gentle rewarming in a water-bath (38° to 43° C) with ibuprofen administration is recommended. Imaging can help predict likelihood of tissue viability, and surgical consultation is recommended. Therapeutic hyperthermia and therapeutic hypothermia are experimental.

KEYWORDS

temperature regulation heatstroke heat exhaustion hypothermia freezing and nonfreezing peripheral cold injuries frostbite therapeutic hypothermia and hyperthermia

659.e3

660

CHAPTER 101  Disorders Due to Heat and Cold  

TABLE 101-1 FACTORS PREDISPOSING TO SERIOUS HEAT ILLNESS INDIVIDUAL FACTORS Lack of acclimatization Low physical fitness Excessive body weight Dehydration Advanced age Young age Toll-like receptor-4 polymorphisms HEALTH CONDITIONS Inflammation and fever Viral or bacterial infection Cardiovascular disease Diabetes mellitus Gastroenteritis Rash, sunburn, and previous burns to large areas of skin Seizures Thyroid storm Neuroleptic malignant syndrome Malignant hyperthermia Sickle cell trait Cystic fibrosis DRUGS Anticholinergic properties (atropine) Antiepileptic (topiramate) Antihistamines Glutethimide (Doriden) Phenothiazines Tricyclic antidepressants Amphetamines, cocaine, Ecstasy [3,4-methylenedioxy-methamphetamine (MDMA)] Ergogenic stimulants (e.g., ephedrine, ephedra) Lithium Diuretics β-Blockers Ethanol Nonsteroidal anti-inflammatory drugs ENVIRONMENTAL FACTORS High temperature High humidity Little air motion Lack of shade Heat wave Physical exercise Heavy clothing Prior compromised heat exposures



EPIDEMIOLOGY

Heat illness accounts for considerable morbidity and mortality in the world today. Serious heat illness is associated with a variety of individual factors, health conditions, drugs, and environmental factors (Table 101-1). Exertional heat illness is among the leading causes of death in young athletes, and its incidence appears to be increasing in the United States. Classic heat illness caused by high environmental temperatures remains a problem especially in homebound elderly persons without air conditioners.2 Anticholinergic and sympathomimetic poisoning (Chapter 102) can induce hyperthermia. Malignant hyperthermia (Chapter 404) is a rare disorder occurring in genetically predisposed individuals. Rapid and massive skeletal muscle contraction from exposure to certain volatile anesthetic agents (most commonly halothane, sevoflurane, desflurane, isoflurane, or enflurane) or depolarizing muscle agents relaxants (e.g., succinylcholine) can trigger core temperature elevations well above 43° C (110° F). However, some data suggest that heat disorders with extreme exercise may represent a similar syndrome.3 Neuroleptic malignant syndrome (Chapter 406) is an idiosyncratic hyperthermic reaction caused by skeletal muscle rigidity from treatment with neuroleptic medications (e.g., antipsychotics, antidepressants, antiemetics). Both malignant hyperthermia and neuroleptic malignant syndrome are potentially fatal without prompt recognition and early intervention. Heat illness can also occur in low-risk individuals who have taken appropriate precautions relative to situations to which they have been exposed in the past. Historically, such unexpected cases were attributed to dehydration (which

impairs thermoregulation and increases hyperthermia and cardiovascular strain), but it is now suspected that a previous heat exposure or a concurrent event (e.g., sickness or injury) might make these individuals more susceptible to serious heat illness. One theory is that previous heat injury or illness primes the acute phase response and augments the hyperthermia of exercise, inducing unexpected serious heat illness. Another theory is that previous infection produces proinflammatory cytokines that deactivate the cells’ ability to protect against heat shock.  

PATHOBIOLOGY

Body temperature can increase from a number of mechanisms: exposure to environmental heat (impeded heat dissipation); physical exercise (increased heat production); fever from systemic illness (elevated set point with subsequent activation of shivering); and medications (neuroleptic malignant syndrome and malignant hyperthermia). In addition, febrile persons have accentuated elevations in core temperature when they are exposed to high ambient temperature, physical exercise, or both. Environmental temperature and humidity, medications, and exercise heat stress in turn challenge the cardiovascular system to provide high blood flow to the skin, where blood pools in warm, compliant vessels such as those found in the extremities. When blood flow is diverted to the skin, reduced perfusion of the intestines and other viscera can result in ischemia, endotoxemia, and oxidative stress. Several common mutations in toll-like receptor 4 are associated with endotoxin hyporesponsiveness. In addition, excessively high tissue temperatures (heat shock: >41° C [105.8° F]) can produce direct tissue injury; the magnitude and duration of the heat shock influence whether cells respond by adaptation (acquired thermal tolerance), injury, or death (apoptotic or necrotic). Heat shock, ischemia, and systemic inflammatory responses can result in cellular dysfunction, disseminated intravascular coagulation, and multiorgan dysfunction syndrome (E-Fig. 101-1). Furthermore, reduced cerebral blood flow, combined with abnormal local metabolism and coagulopathy, can lead to dysfunction of the central nervous system.  

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Minor heat illness is common and can be recognized by its clinical features. Miliaria rubra (heat rash) results from the occlusion of eccrine sweat gland ducts and can be complicated by secondary staphylococcal infection. Heat syncope (fainting) is caused by temporary circulatory insufficiency as a result of blood pooling in the peripheral veins, especially the cutaneous and lower extremity veins. Skeletal muscle cramps most commonly occur during and after intense exercise and are probably related to dehydration, loss of sodium or potassium, and neurogenic fatigue rather than to overheating itself. Serious heat illness includes heat exhaustion, heat injury, and heatstroke, with some individuals progressing along this spectrum. Patients who exhibit symptoms (e.g., dizziness, un-steady gait, ataxia, headache, confusion, weakness, fatigue, nausea, vomiting, diarrhea) should have an immediate assessment of their mental status, core (rectal) temperature, and other vital signs. The most common causes of hospital admission are fluid and electrolyte disorders, renal failure, urinary tract infection, and heatstroke. Until proven otherwise, heatstroke should be the initial working diagnosis in anyone who is a heat casualty and has an altered mental status. Heat exhaustion is defined as a syndrome of hyperthermia (temperature at time of event usually ≤40° C or 104° F) and debilitation that occur during or immediately after exertion in the heat, accompanied by no more than minor central nervous system dysfunction (headache, dizziness, mild confusion), which resolves rapidly with intervention. It is primarily a cardiovascular event (insufficient cardiac output) frequently accompanied by sweaty hot skin, dehydration, and collapse. Heat injury is a moderate to severe illness characterized by evidence of damage to end organs (e.g., liver, renal, gut) and tissues (e.g., rhabdomyolysis) without sufficient neurologic symptoms to be diagnosed as heatstroke. It is usually associated with body temperatures above 40° C (104° F). Heatstroke is a severe illness characterized by profound mental status changes with high body temperatures, usually but not always higher than 40° C (104° F). However, patients with a core temperature higher than 40° C do not universally have a heat injury or heatstroke, and core temperatures this high can be seen transiently after stressful exercise in the heat. To establish the diagnosis of heatstroke, the entire clinical picture, including mental status and laboratory results, must be considered. Heatstroke is often categorized as classic or exertional; classic heatstroke is observed primarily in otherwise sick and compromised individuals, and exertional heatstroke is observed primarily in apparently healthy and physically fit individuals during or after vigorous exercise

CHAPTER 101  Disorders Due to Heat and Cold  

660.e1

Heat stroke High body temperature with CNS dysfunction

Tissue thermal injury Necrotic/apoptotic cell death

Coagulopathies Fibrin deposition Excessive bleeding

Immune modulators Endotoxin, cytokines

Systemic inflammatory response syndrome

Multi-organ failure and death E-FIGURE 101-1.  Schematic of the sequence of events occurring in response to heat stroke that stimulate a systemic inflammatory response syndrome that leads to multiorgan dysfunction and death. (From Leon LR, A. Bouchama. Heat stroke. Compr Physiol. 2015;5:611-647.)

CHAPTER 101  Disorders Due to Heat and Cold  

TABLE 101-2 COMPARISON OF CLASSIC AND EXERTIONAL HEATSTROKE PATIENT CHARACTERISTICS

CLASSIC

EXERTIONAL

Age

Young children or elderly

15-55 years

Health

Chronic illness

Usually healthy

Fever

Unusual

Common

Prevailing weather

Frequent in heat waves

Variable

Activity

Sedentary

Strenuous exercise

Drug use

Diuretics, antidepressants, anticholinergics, phenothiazines

Ergogenic stimulants or cocaine

Sweating

Often absent

Common

Acid-base disturbances

Respiratory alkalosis

Lactic acidosis

Acute renal failure

Uncommon

Common (≈15%)

Rhabdomyolysis

Uncommon

Common (≈25%)

CK

Mildly elevated

Markedly elevated (500-1000 U/L)

ALT, AST

Mildly elevated

Markedly elevated

Hyperkalemia

Uncommon

Common

Hypocalcemia

Uncommon

Common

DIC

Mild

Marked

Hypoglycemia

Uncommon

Common

ALT = alanine aminotransferase; AST = aspartate aminotransferase; CK = creatine kinase; DIC = disseminated intravascular coagulation.

(Table 101-2). In heatstroke, neuropsychiatric impairments (e.g., marked confusion, disorientation, combativeness, and seizures) develop early and universally but are readily reversible with early cooling. In addition, heatstroke can be complicated by liver damage, rhabdomyolysis, disseminated intravascular coagulation, water and electrolyte imbalance, and renal failure. In fulminant heatstroke, patients have the full spectrum of abnormalities associated with the systemic inflammatory response syndrome (Chapter 100).

PREVENTION AND TREATMENT  Heat illness can be prevented by heat acclimatization and acquired thermal tolerance, maintenance of adequate hydration, and avoidance of overwhelming heat exposure.4 Adequate fluid intake is critical, and oral rehydration solutions should contain sodium and other electrolytes to restore both intracellular and extracellular fluid. Management of serious heat illness, which should begin in the field setting, includes cooling, rehydration, and monitoring (Table 101-3). The first priority should be immediately to initiate whole body cooling and to continue cooling until the core temperature falls below 38.8° C (102° F). Body cooling lowers skin temperatures, thereby facilitating conduction and convection from the core to the shell, and reduces cardiovascular stress by causing arterial and venous constriction that redirects blood back to the heart. Immersion or soaking of the skin in cool or ice water with skin massage is the most effective method, but other effective methods include soaking of the skin followed by accelerated evaporation with fans or the use of ice sheets and ice packs. These noninvasive treatments can be supplemented with the infusion of chilled (≈4° C) normal saline.5 Cooling can induce shivering, which is usually not sufficient to increase body temperature, so shivering need not be treated. In the hospital, the highest priority for patient care remains urgent cooling, including cold intravascular fluid.6 Patients who are unconscious are at risk of poor airway control and may require endotracheal intubation to prevent aspiration. Fluid and electrolyte deficits should be corrected; restoration of plasma volume with isotonic fluids (e.g., normal saline) sufficient to sustain adequate perfusion, as judged by carefully monitored urine output, is also a priority. Rapid overcorrection of serum electrolytes (e.g., sodium) should be avoided. If rhabdomyolysis (Chapter 105) and myoglobinuria are present, maintaining urine flow helps minimize renal injury. For exercise-induced and environmental heat illness, no pharmacologic interventions have been proved to augment cooling. For patients with malignant hyperthermia, however, dantrolene should be administered as a loading bolus of 2.5 mg/kg intravenously, with subsequent bolus doses of 1 mg/kg intravenously until the signs have abated.7

661

TABLE 101-3 MANAGEMENT OF HEAT ILLNESS HEAT EXHAUSTION Rest and shade Loosen and remove clothing Supine position and elevate legs Actively cool skin Fluids by mouth Monitor core temperature Monitor mental status HYPERTHERMIA Protect the airway Insert at least two large-bore intravenous lines Monitor core temperature; options include rectal, pulmonary artery, esophageal probe Actively cool the skin until core temperature reaches 350 g/dL

Lead

25 g/dL

Lithium

0.6-1.2 mEq/L

>1.2 mEq/L††

Methanol

None measured

>25 mg/dL

Methemoglobin

1-2%

>15%

Phenobarbital

15-40  µg/mL

>40 g/mL

Phenytoin

10-20  µg/mL

>20 g/mL

Salicylates

≤30 mg/dL

>30 mg/dL

Theophylline

8-20  µg/mL

>20 g/mL

Valproic acid

50-100  µg/mL

>100 g/mL

NORMAL

TOXIC*

SOURCE: URINE Arsenic

100 g/24-hr urine††

Mercury

20 g/L††

Thallium

200 g/L††

*The “toxic” level is provided for perspective. For many toxicants, simply being above this value does not imply a specific need for therapy or a necessarily poor prognosis. It does, however, generally suggest a need for additional evaluation, observation, or monitoring. Similarly, depending on the clinical context, levels below the “toxic” range may still be consequential. † False-positive levels of 16 to 28 µg/mL have been reported in patients with bilirubin levels greater than 17 mg/dL. ‡ Levels drawn more than 4 hours after ingestion should be plotted on the nomogram provided by Rumack and Matthew (Rumack BH, Matthew H. Acetaminophen poisoning and toxicity. Pediatrics. 1975;55:871-876) to assess the potential for toxicity. § Lower levels may be toxic in pregnant patients and in those with prolonged exposure to carbon monoxide. | Consult a reference laboratory for normal values; results are assay dependent. ¶ Some patients may require levels above the therapeutic range to control symptoms. **The value of 80 mg/dL (0.08 g/dL) for ethanol is the statutory limit for operating a motor vehicle. Consequential clinical effects other than inebriation are uncommon with concentrations below 200 mg/dL. †† Lower values may indicate toxicity if appropriate clinical findings are present.

672

CHAPTER 102  Acute Poisoning  

agents, solvents (e.g., carbon tetrachloride, trichloroethylene, tetrachloroethylene, toluene), and sulfonamides. Agents that decrease glomerular perfusion by reducing renal blood flow include amphotericin, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, cocaine, cyclosporine, mannitol (excessive chronic doses), methotrexate, and nonsteroidal anti-inflammatory drugs.

Imaging

A computed tomographic scan of the head can detect life-threatening cerebral edema secondary to toxicant-induced hepatic failure, ethylene glycol, and methanol. It also detects intracranial bleeding caused by anticoagulants, scorpion venom, and sympathomimetics (e.g., amphetamines, cocaine, phenylpropanolamine). An abdominal radiograph can reveal radiopaque ferrous sulfate tablets, drug-filled packets in illicit drug smugglers (body packers), or metals such as arsenic, lead, mercury, and thallium.

Diagnostic Syndromes

Given the myriad combinations of signs, symptoms, and laboratory findings, making the correct diagnosis in a noncommunicative patient can be daunting. A thorough history from bystanders, friends, and prehospital medical personnel may yield crucial information. In addition, the diagnostic possibilities can be narrowed by findings that can narrow the differential diagnosis with modest certainty. For example, consider a patient with sudden loss of consciousness, anion gap metabolic acidosis, and bradycardia without hypoxemia. Among the relevant possible causes from the list of toxicants that cause an anion gap metabolic acidosis (see earlier) along with sudden loss of consciousness are hydrogen sulfide, cyanide, and sodium azide.

TREATMENT  Initial Stabilization

Intubation and Respiratory Support

Appropriate airway management should be instituted to correct hypoxemia and respiratory acidosis and to protect against pulmonary aspiration (Fig. 1022); intubation should be considered if the patient has depressed consciousness and a decreased gag reflex. Rapid-sequence intubation facilitates airway management. Anatomic difficulties should be anticipated in patients with caustic ingestions (e.g., hypopharyngeal burns that may perforate); angioedema caused by angiotensin-converting enzyme inhibitor therapy or envenomation by some rattlesnakes, such as the canebrake (Crotalus horridus atricaudatus) and eastern diamondback (Crotalus adamanteus; Chapter 104); and swelling secondary to direct tissue injury (e.g., huffing compressed hydrocarbons, smoking crack) or secondary to anaphylactoid and anaphylactic reactions. Endotracheal intubation by flexible fiberoptic nasopharyngoscopy may be indicated in these cases. Hypoxemia can occur with toxicants that produce CNS depression, such as opioids, antidepressants, barbiturates, sedative-hypnotics, and central α2-adrenergic receptor agonists (clonidine), or agents causing peripheral neuromuscular impairment, such as nicotine, organophosphorus compounds, tetrodotoxin (puffer fish, blue-ringed octopus), botulinum, or envenomation from elapids (coral snake), Mojave rattlesnakes, or certain coelenterates (box jellyfish; Chapter 104). Respiratory acidosis can rapidly worsen the toxicities of cyclic antidepressants and salicylates; sedation of these patients should be accompanied by immediate airway placement and ventilatory support. Intoxicated patients may have an increased risk for pulmonary aspiration because of concomitant CNS depression, attenuated airway reflexes, full stomachs, and delayed gastric emptying. Succinylcholine can cause prolonged paralysis in patients with organophosphorus poisoning and can exacerbate hyperkalemia from cardioactive steroids (e.g., digoxin), hydrofluoric acid, or rhabdomyolysis (Chapter 105). Rhabdomyolysis has been reported with adrenergic agents, doxylamine, phencyclidine, heroin, carbon monoxide, Tricholoma equestre mushrooms, and envenomation by crotaline snakes, scorpions, or widow spiders (Latrodectus sp); short-acting nondepolarizing agents, such as vecuronium and rocuronium, are preferable in these cases.

Advanced Life Support

Standard emergency cardiovascular care algorithms (Chapter 57) must be modified for effects caused by specific poisons. Atropine often does not reverse bradycardia secondary to β-adrenergic receptor antagonists, or L-type calciumchannel antagonists and occasionally with cardioactive steroids, and it may actually impair the ability to do adequate gastrointestinal decontamination. In these cases, more specific therapy with intravenous calcium (calcium-channel antagonists), high doses of glucagon (β-adrenergic receptor antagonists, calciumchannel antagonists), or digoxin-specific Fab antibody (cardiac glycosides) is

indicated. High-dose insulin-glucose therapy can successfully reverse myocardial depression and conduction abnormalities in humans poisoned with β-adrenergic receptor antagonists10 and calcium-channel antagonists. Intravenous sodium bicarbonate may reverse cardiac conduction delays caused by antiarrhythmic drugs with sodium-channel blockade recovery rates of greater than 1 second (Vaughn-Williams classification IA and IC), cocaine, cyclic antidepressants, diphenhydramine, and quinine. β-Adrenergic receptor antagonists are contraindicated in patients with cocaine-induced myocardial syndromes because they can result in unopposed α-adrenergic–mediated vasoconstriction, but phentolamine can reverse the agonistic effects of cocaine on α-adrenergic receptors. Benzodiazepines can reverse significant sinus tachycardia from sympathomimetic agents. Calcium may also be life-saving in systemic hydrofluoric acid poisoning and severe hypermagnesemia, and it is indicated for symptomatic hypocalcemia caused by ethylene glycol toxicity. Drug-induced hypertension may be transitory; nitroprusside, phentolamine, or labetalol should be used if treatment is clinically indicated. In patients with toxicant-induced circulatory collapse refractory to maximal therapy, intravenous lipid emulsion therapy should be administered to those poisoned by lipophilic toxicants (e.g., calcium channel blockers) and circulatory assist devices may support the patient until sufficient toxicant is eliminated (see Table 102-6 for dosing details) (Chapter 99).

Decontamination Activated Charcoal

Single-dose activated charcoal without prior gastric emptying has been the preferred method of treatment for the ingestion of substances that have the potential to cause moderate to life-threatening toxicity and are known to adsorb to activated charcoal. The absence of clinical signs and symptoms does not preclude administration of activated charcoal because toxicant absorption and toxicity can be delayed. Activated charcoal can also be administered when the ingested toxicant cannot be identified but significant toxicity is a concern. Activated charcoal consists of pyrolysis products that have been specially cleaned to produce an internal pore structure to which substances can adsorb, thereby limiting their systemic absorption. Activated charcoal can be administered with antiemetic drugs or given through a nasogastric tube, when necessary. The oral dose is approximately 1 g/kg body weight, with a maximum single dose of 100 g. Efficacy in preventing toxicant absorption declines with time, so activated charcoal should be given as soon as possible after ingestion. However, the documented efficacy of activated charcoal for reducing toxicant blood levels has not translated into reduced mortality in reports or in randomized trials. A2  The decision to administer activated charcoal should be based on a risk/benefit assessment that includes nature of the exposure, clinical effects displayed during evaluation, and abilities of the medical facility and staff. For patients likely to have a good outcome, the risk and effort associated with activated charcoal administration are not worthwhile. Its use is justified in patients who present early (1 to 2 hours) after exposures to a large amount of a concerning toxin that is likely to be adsorbed to charcoal. Activated charcoal should not be used in patients at risk for aspiration until the airway is secure to minimize aspiration; the patient’s head should also be elevated unless it is contraindicated. Activated charcoal is contraindicated in patients with a perforated bowel, functional or mechanical bowel obstruction, ingestion of a pure aliphatic hydrocarbon such as gasoline or kerosene (no benefit and increased risk for aspiration), and ingestion of caustic acid and alkali (no benefit and obscures endoscopy). Certain agents, such as lithium, iron, and metal salts, and ethanol, do not adsorb significantly to activated charcoal, but its use is not precluded if the patient has ingested other toxicants that do adsorb to activated charcoal. Pulmonary aspiration and bowel obstruction from inspissated activated charcoal are the most common complications; both occur more frequently when multidose activated charcoal is administered, but they can be avoided by withholding treatment in patients who have suboptimal bowel function or decreased fecal elimination.

Gastric Emptying

Two methods of gastric emptying, syrup of ipecac and orogastric lavage through a large-bore tube, are no longer routinely used. Both are relatively ineffective therapies that potentially increase the risk for aspiration. No welldesigned study has documented any benefit of gastric emptying, either by lavage or by syrup of ipecac, compared with the use of activated charcoal alone. Gastric emptying by lavage or, rarely, by syrup of ipecac may be of benefit and should be performed in patients who have ingested toxicants that do not adsorb to activated charcoal and are known to produce significant morbidity or for which aggressive decontamination may offer the best chance for survival (e.g., colchicine, sodium azide, sodium fluoroacetate). Removal of a liquid toxicant, such as ethylene glycol, may be accomplished by aspiration of gastric contents through a nasogastric tube. Contraindications to gastric emptying include those for activated charcoal, a bleeding diathesis, and the ingestion of sharp objects. Placement of an endotracheal tube before gastric lavage may be necessary to protect the airway in patients who

673

CHAPTER 102  Acute Poisoning  

Patient stable? Yes

No Assess airway Intubate to correct or avoid: Hypoxemia Respiratory acidosis Pulmonary aspiration

Initiate ALS Modifications: Atropine: often ineffective for bradycardia due to BARAs, CCAs, cardiac glycosides Benzodiazepines: for cocaine-induced tachycardia Calcium: for CCAs, HF, hypermagnesemia Glucagon: for BARAs, CCAs Digoxin-specific fab: for cardiac glycosides High-dose insulin-glucose: for BARAs, CCAs Nitroprusside: for drug-induced hypertension NaHCO3: for myocardial sodium-channel blockers Phentolamine: reverses cocaine-induced α-adrene rgicagonism Avoid BARAs: in cocaine-induced ischemia

Decontamination can be performed simultaneously with stabilization therapies

Administer antidote Indicate for specific toxins

Correct hypovolemia Initiate/continue vasopressors Consider circulatory assist, e.g., balloon pump, heart-lung bypass

No

Patient unstable Continue resuscitation

Patient hemodynamically stable?

Decontaminate Oral 1. AC 1g/kg (maximum 100 g) Indications: Toxin with potential for serious toxicity Toxin adsorbs to AC Contraindications: Nonprotected airway Bowel obstruction/perforation Ingestion of pure aliphatic hydrocarbon or caustics 2. Gastric emptying (large-bore orogastric tube lavage; nasogastric tube aspiration of liquid toxin) Indications: Toxins nonadsorbent to AC and with potential for consequential toxicity; ideally performed ≤1 hour post-ingestion Contraindications: Same as for AC; also ingestion of sharp objects or presence of bleeding diathesis 3. Other: Whole bowel irrigation with PEG Surgical removal of drug packets Dermal Wash with soap and water Ocular Irrigate with NS

Yes

Patient stable Is toxin eliminated by kidneys?

Yes

Is enhanced elimination desired? Hemodynamic instability may prevent use of extracorporeal modalities

Is toxin removed by extracorporeal device?

Yes

Consider use of MDAC for toxins with known or potential enhanced elimination Indications: Definite–carbamazepine, dapsone, phenobarbital, quinine, salicylates, theophylline Potential–amitriptyline, dextropropoxyphene, digitoxin, digoxin, disopyramide, nadolol, phenylbutazone, phenytoin, piroxicam, sotalol Contraindications: Same as for single-dose AC Consider urinary alkalinization Indications: Chlorpropamide, 2,4-dichlorophenoxyacetic acid, formic acid, methotrexate, phenobarbital, salicylates Contraindications: Volume overload, pulmonary or cerebral edema Institute appropriate extracorporeal modality (see Table 102-7)

FIGURE 102-2.  Algorithm for the management of acute poisoning. AC = activated charcoal; ALS = advanced life support; BARAs = β-adrenergic receptor antagonists; CCAs = L-type calcium-channel antagonists; HF = hydrofluoric acid; MDAC = multidose activated charcoal; NS = 0.9% saline solution; PEG = nonabsorbable polyethylene glycol solution.

have a decreased level of consciousness and impaired gag reflex but is not required in all cases. Major complications of gastric emptying include pulmonary aspiration, esophageal tears and perforations, and laryngospasm (with lavage).

Whole Bowel Irrigation

Whole bowel irrigation with a nonabsorbable polyethylene glycol solution has been recommended for iron and sustained-release medications, for agents not adsorbed to activated charcoal, and for body packers (smugglers who swallow packets of illicit drugs, usually heroin or cocaine). The most common complication is vomiting, and whole bowel irrigation is contraindicated in patients with bowel perforation, obstruction, hemorrhage, or hemodynamic or respiratory instability. The initial recommended dose is 500 mL/hour given orally or by nasogastric tube, with titration to 2000 mL/hour as tolerated; treatment continues until the rectal effluent clears. Rarely, surgery may be necessary to remove packets in smugglers who have symptoms of cocaine toxicity or are obstructed; heroin toxicity is usually adequate managed with naloxone.

Endoscopic removal of cocaine packets should never be attempted because of the risk of packet rupture.

Antidotes

Few toxicants have specific therapies (Table 102-6). Although antidotes may be essential in treating patients exposed to certain toxicants, their use does not preclude the need for ongoing supportive care and, in some cases, extracorporeal elimination.

Enhanced Elimination

Methods to accelerate the elimination of toxicants or drugs from the body include multiple doses of activated charcoal, urinary alkalinization, and extracorporeal removal. Another method, using the oral ion exchange resins sodium polystyrene sulfonate and cholestyramine, has experimentally enhanced the elimination of lithium, digoxin, digitoxin, and organochlorines but has limited clinical usefulness. Text continued on p. 678

674

CHAPTER 102  Acute Poisoning  

TABLE 102-6 ANTIDOTES AND INDICATIONS FOR USE§ ANTIDOTE

INDICATION FOR USE

DOSE*

TREATMENT END POINT

COMMENTS

Antivenom, Crotalidae (Fab)†¶ A3 

Crotaline snake (e.g., 4-6 vials; repeat for persistent or Halt in progression of rattlesnakes, copperhead) worsening clinical condition; circumferential and proximal repeated doses of 2 vials at 6, 12, swelling and 18 hours after initial Resolving systemic effects antivenom dose are recommended

Antivenom, Latrodectus (equine)†¶

Black widow spider (Latrodectus sp)

1 vial diluted in 50-100 mL NS, infused over 1 hour; can repeat

Resolution of symptoms, vital signs Dilution and slow infusion rate are critical normal to avoid anaphylactoid reaction Indications include severe pain unresponsive to opioids and severe hypertension Serum sickness can occur IV calcium is ineffective

Atropine

Carbamates Nerve agents Organophosphorus compounds

2 mg IV; double the dose every 5 minutes to achieve atropinization and hemodynamic stability; then start continuous infusion of 10-20% of total stabilizing dose per hour

Cessation of excessive oral and pulmonary secretions, >80 beats/min, systolic blood pressure >80 mm Hg

Doubling of the dose every 5 minutes (e.g., 2 mg, 4 mg, 8 mg, 16 mg) estimated to achieve atropinization within 30 minutes Stop infusion when patient develops concerning signs or symptoms of anticholinergic toxidrome (see Table 102-1); restart infusion at lower rate when signs or symptoms abate

Calcium salt‡

Calcium-channel antagonists

Calcium chloride 10%, 10 mL (1 g) over 10 minutes; can be given in 1 minute if critically ill Calcium gluconate 10%, 30 mL (3 g) over 10 minutes; can be given in 1 minute if critically ill

Reversal of hypotension; may not reverse bradycardia

Hydrofluoric acid

Systemic toxicity: calcium gluconate 10%, 1-3 g (10-30 mL) per dose IV over 10-minute period; repeat as needed every 5-10 minutes Topical toxicity: calcium chloride 10%, 1 g (10 mL) mixed into water soluble lubricant Calcium gluconate 10%, 1 g (10 mL) per dose IV over 10-minute period; repeat as needed every 5-10 minutes Calcium gluconate 10%, 1-2 g (10-20 mL) per dose IV over 10-minute period; repeat as needed every 5-10 minutes Calcium gluconate 10%, 0.5-1.0 g (5-10 mL) per dose over 10-minute period; repeat as needed every 10 minutes

Reversal of life-threatening manifestations of hypocalcemia and hyperkalemia Topical: reversal of severe neuropathic pain from dermal exposure

All indications: Monitor ionized calcium levels IV extravasation causes tissue necrosis, especially with calcium chloride Can administer at faster than stated rates for immediate life-threatening conditions (i.e., in 1 minute) Calcium chloride contains three times more elemental calcium than calcium gluconate does Can dilute and give intra-arterially or IV with a Bier block for extremity exposures and burns Topical: apply to skin under occlusive dressing

Hyperkalemia (except cardiac glycosides) Hypermagnesemia

Hypocalcemia (e.g., ethylene glycol) L-Carnitine

Valproate-induced hyperammonemia or hepatotoxicity

Better safety profile than historical equine-derived antivenom Repetitive dosing indicated for recurrent soft tissue swelling Less effective at correcting hematologic (i.e., coagulation and platelet) disorders

Reversal of myocardial depression and conduction delays

May precipitate ventricular arrhythmias

Reversal of respiratory depression, hypotension, and cardiac conduction blocks

Simultaneous therapies to increase magnesium elimination should be instituted

Reversal of tetany

Correct symptomatic hypocalcemia; avoid excessive administration that may increase production of calcium oxalate crystals in ethylene glycol poisoning

100 mg/kg (maximum 6 g) IV over Treat until clinical improvement 30 minutes, then 15 mg/kg IV occurs over 30-minute period q4h (maximum 6 g/day)

Levocarnitine is active form Adjust dose for end-stage renal disease

Cyanide antidote kit Cyanide Amyl nitrite Sodium nitrite Sodium thiosulfate [Hydroxocobalamin is preferred if available, see below]

Amyl nitrite: 0.3-mL pearls, crush and inhale over 30-second period Sodium nitrite 3%: 10 mL IV over 10-minute period Sodium thiosulfate 25%: 50 mL (12.5 g) IV over 10-minute period

Resolution of lactic acidosis and moderate to severe clinical signs and symptoms: seizures, coma, dyspnea, apnea, hypotension, bradycardia

Coordinate amyl nitrite with continued oxygenation and give only until sodium nitrite infusion is begun; nitrites may produce hypotension and excess methemoglobinemia Sodium nitrite dose must be adjusted if patient has hemoglobin 350 g/dL Prolonged therapy can cause pulmonary toxicity

Iron salts

675

CHAPTER 102  Acute Poisoning  

TABLE 102-6 ANTIDOTES AND INDICATIONS FOR USE§—cont’d ANTIDOTE

INDICATION FOR USE

DOSE*

TREATMENT END POINT

COMMENTS

Digoxin-specific antibody Digoxin fragments (Fab) Digitalis and related plants (e.g., oleander, lily of the valley) Other cardiac glycosides (e.g., bufadienolides [Bufo toads])

Unknown digoxin dose or serum Resolution of hyperkalemia, level, or for plant or toad source: symptomatic bradyarrhythmias, acute toxicity—10-20 vials; ventricular arrhythmias, Mobitz chronic toxicity—3-6 vials II or third-degree heart block Digoxin dose known: number of vials = (mg ingested × 0.8) ÷ 0.5 Digoxin serum level known: number of vials = [serum level (ng/mL) × weight (kg)] ÷ 100 Infuse dose over 30 minutes

Each vial binds 0.5 mg of digoxin or digitoxin Monitor ECG and potassium levels Digoxin serum levels unreliable after antidote administered unless test is specific for free serum digoxin

Dimercaprol (BAL)

Inorganic arsenic: 3-5 mg/kg IM q4h for 48 hours and then q12h for 7 to 10 days. Lead: 75 mg/m2 (4 mg/kg) IM q4h for 5 days Inorganic mercury: 5 mg/kg IM, then 2.5 mg/kg IM q12h for 10 days or until patient is clinically improved

Arsenic: 24-hour urinary arsenic 100,000

Scorpions

>1 million

1 million

>1000 (mostly due to

Jellyfish

>1 million

100,000

100,000

1000

0.54 mL blood/L urine)

Foamy

Proteinuria

Blue or green

Pseudomonas urinary tract infection Bilirubin Methylene blue

Pink or red

Aniline dyes in sweets Porphyrins (on standing) Blood, hemoglobin, myoglobin Drugs: phenindione, phenolphthalein Anthocyaninuria (beetroot, “beeturia”)

Orange

Drugs: anthraquinones (laxatives), rifampicin Urobilinogenuria

Yellow

Mepacrine Conjugated bilirubin Phenacetin Riboflavin

Brown or black

Melanin (on standing) Myoglobin (on standing) Alkaptonuria

Green or black

Phenol Lysol

Brown

Drugs: phenazopyridine, furazolidone, l-dopa, niridazole Hemoglobin and myoglobin (on standing) Bilirubin

From Forbes CD, Jackson WF. Color Atlas and Text of Clinical Medicine. 3rd ed. London: Mosby; 2003.

mortality.5 Evidence for pan–proximal tubular dysfunction (e.g., glycosuria, aminoaciduria, phosphaturia) indicates that Fanconi syndrome is present. The dipstick for protein is a sensitive assay based on color change induced by the presence of proteins at a given pH. It is most sensitive to the presence of albumin and is much less sensitive to other proteins, such as the light chains of Bence Jones protein (Chapter 178). The presence of 1+ protein correlates with about 30 mg/dL of albuminuria, and 3+ protein correlates with more than 500 mg/dL of proteinuria. Because the dipstick is not a quantitative measurement, small amounts of proteinuria in an oliguric patient may give the false appearance of high-grade proteinuria. The excretion of abnormal quantities of albumin below the level detectable by the urine dipstick is called microalbuminuria. Normal albumin excretion, which is less than 30 mg/day, is best detected by radioimmunoassay or enzyme immunoassay. Microalbuminuria is the earliest clinically detectable stage of diabetic nephropathy (Chapter 115). Proteinuria of increasing severity is associated with a more rapid decline in the GFR, regardless of the GFR, except in minimal change disease (Chapter 113). The dipstick for heme uses the peroxidase-like activity of hemoglobin and myoglobin molecules to detect the presence of heme pigment. The reaction occurs on exposure to hemoglobin, myoglobin, or intact RBCs. Urine dipstick testing is very sensitive (97%) and reasonably specific (75%) for the detection of hematuria. The presence of myoglobin, which is found in patients with rhabdomyolysis (Chapter 105), or free hemoglobin, which is seen in patients with intravascular hemolytic anemias (Chapter 151), is suspected if the heme reaction is intensely positive and there is a paucity of cellular elements in the sediment. Since RBCs lyse rapidly on storage, especially in a hypotonic urine, their absence on routine analysis of the urinary sediment is likely to provide false reassurance in the presence of a positive dipstick. Persistent, isolated, asymptomatic, microscopic hematuria is an independently significant risk factor for the subsequent development of end-stage renal disease.6 Although gross asymptomatic hematuria is associated with cancer in 8 to 25% of cases, only about 2.5% of patients with dipstick-positive hematuria have cancer, including 0.5)

Non-glomerular hematuria

CT scan/MRI Urine cytology x 3

Nephrology referral Negative findings and patient with low risk of cancer

Positive findings or negative findings in a patient with high risk of cancer

Observe

Urology referral

FIGURE 106-3.  Algorithm for the evaluation of asymptomatic hematuria. BPH = benign prostatic hyperplasia; CT = computed tomography; GFR = glomerular filtration rate; MRI = magnetic resonance imaging; RBC = red blood cell; UTI = urinary tract infection. (Courtesy Ali Gharavi, MD. Modified from Cohen RA, Brown RS. Microscopic hematuria. N Engl J Med. 2003;348:2330-2338).

FIGURE 106-4.  Dysmorphic erythrocytes. These dysmorphic erythrocytes vary in size, shape, and hemoglobin content and reflect glomerular bleeding. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

WBC casts can be a dominant feature of many diseases that traditionally are thought of as glomerular diseases, such as lupus nephritis (Chapter 250) and antineutrophil cytoplasmic antibody (ANCA)–associated glomerulonephritis (Chapter 254). Tubular cell casts are seen with any acute tubular injury and are the dominant cellular casts in ischemic acute tubular necrosis (Chapter 112). They also can be seen with nephrotoxic injury, such as with aminoglycosides and cisplatin. Some casts may contain both leukocytes and tubule cells. Crystals can be a normal finding in the urine or serve as clues to pathophysiologic processes. Certain crystals, such as the hexagonal crystals seen with cystinuria (Chapter 119), are always abnormal (Fig. 106-9). Others, such as the octahedral calcium oxalate crystals (Fig. 106-10), may be a normal finding or may be evidence for ethylene glycol intoxication (Chapter 102). Coffin lid-shaped triple phosphate crystals, which are composed of ammonium magnesium phosphate (Fig. 106-11), are seen in urinary tract infections with urea-splitting organisms (Chapter 268). Uric acid crystals, sodium urate crystals

FIGURE 106-5.  Isomorphic erythrocytes. These erythrocytes are similar in size, shape, and hemoglobin content. Isomorphic cells reflect nonglomerular bleeding from lesions such as calculi and papillomas or hemorrhage from cysts in polycystic renal disease. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

(Fig. 106-12), and calcium phosphate amorphous crystals are common and do not usually have pathologic significance.

Other Elements

Bacteria may be seen in the urine sediment. A spun urine sediment may show rods or cocci in chains, but bacteria are identified best by Gram staining of the urine sediment. Budding yeast forms (which are highly refractile), trichomonads, and spermatozoa also may be seen in the urinary sediment.

Disorders of Electrolytes

The kidneys mediate the homeostatic balance of a variety of electrolytes. Electrolyte disorders may accompany renal failure or may reflect isolated defects

704

CHAPTER 106  Approach to the Patient with Renal Disease  

TABLE 106-3 CAUSES OF STERILE PYURIA INFECTIOUS Current use of antibiotics Past treated urinary tract infection (within past 2 weeks) Tuberculosis (Chapter 308) Gonorrhea (Chapter 283) and chlamydia (Chapter 302) Mycoplasma and ureaplasma (Chapter 301) Genital herpes (Chapter 350) Trichomoniasis (Chapter 332) Fungal infections (Chapter 322) Schistosomiasis (Chapter 334) Prostatitis (Chapter 120) Balanitis Appendicitis (Chapter 133) NONINFECTIOUS Interstitial nephritis (e.g., analgesic nephropathy; Chapter 114) Urinary tract stones (Chapter 117) Polycystic kidney (Chapter 118) Current or recent urinary tract manipulation Catheter cystoscopy Endoscopy Foreign body such as surgical mesh in the urethra or a retained stent Urinary tract neoplasm (Chapter 187) Pelvic irradiation Urinary fistula Renal transplant rejection (Chapters 43 and 122) Renal vein thrombosis (Chapter 116) Papillary necrosis Interstitial cystitis (Chapter 268) Inflammatory diseases (e.g., systemic lupus erythematosus [Chapter 250] or Kawasaki disease [Chapter 254]) Adapted from Wise GJ, Schlegel PN. Sterile pyuria. N Engl J Med. 2015;372:1048-1054.

FIGURE 106-7.  Number and type of granules and their density in the cast vary in different casts. The presence of erythrocytes in this cast may mean that the granules are derived partly from disrupted erythrocytes. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

FIGURE 106-8.  A cast composed entirely of erythrocytes reflects heavy hematuria and active glomerular disease. Crescentic nephritis is likely to be present if erythrocyte cast density is greater than 100/mL. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

FIGURE 106-9.  Typical hexagonal cystine crystal. A single crystal provides a definitive diagnosis of cystinuria. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

FIGURE 106-6.  Hyaline cast of the type seen in small numbers in normal urine. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

in tubular function or specific defects in ion transporters. The functional integrity of the renal tubules can be assessed by calculating the fractional excretion of an electrolyte X. FE ÷ =

(urine X ) (plasma X ) (urine Cr ) (plasmaCr )

× 100 =

(urine X ) × (plasmaCr ) × 100 (urineCr ) × (plasma X )

Imbalances in renal sodium and water handling can lead to extracellular volume depletion or to the formation of edema (Chapter 108). Potassium homeostasis is regulated primarily at the distal nephron, but dietary intake, transcellular shifts, and gastrointestinal losses also contribute to balance (Fig. 106-13; Chapter 109). Calcium homeostasis is regulated by the actions of parathyroid hormone, calcitonin, and vitamin D on bone, intestine, kidney, and parathyroid tissue. Although Ca++ is also filtered and excreted by the kidney, the important role of these hormones generally places disorders of

FIGURE 106-10.  Oxalate crystals. A pseudocast of calcium oxalate crystals accompanied by crystals of calcium oxalate dehydrate. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

CHAPTER 106  Approach to the Patient with Renal Disease  

hypo- and hypercalcemia within endocrinology (Chapter 232). Magnesium deficiency (Chapter 111) may be caused by decreased intake (either orally or by impaired intestinal absorption) or by increased losses via the gastrointestinal tract, kidneys, or skin. Conversely, hypermagnesemia (Chapter 111) can rarely occur in individuals with a normal GFR when the rate of intake exceeds capacity of renal clearance (e.g., excessive use of Mg++-based antacids or large parenteral doses given for preeclampsia). Similar to potassium, hyperphosphatemia (Chapter 111) can arise from reduced excretion (primarily chronic kidney disease), excessive intake, or redistribution of cellular phosphorus. Hypophosphatemia also can be caused by shifts into the intracellular compartment, or may occur from malnutrition or from renal losses. High urine phosphorus in the setting of hypophosphatemia suggests a renal tubule defect, hyperparathyroidism (Chapter 232), or a form of rickets (Chapter 231) and can be assessed by determining the fractional excretion.

FIGURE 106-11.  Coffin-lid crystals of magnesium ammonium phosphate (struvite).

(From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

705

  SPECIFIC RENAL SYNDROMES

This chapter considers the approach to the patient with acute kidney injury (Chapter 112), glomerular syndromes (nephrotic vs. nephritic; Chapter 113), tubulointerstitial disease (Chapter 114), vasculitis and vascular diseases of the kidney (Chapter 116), papillary necrosis, and chronic kidney disease (Chapter 121).  

Acute Kidney Injury

Acute kidney injury (Chapter 112) is a syndrome in which glomerular filtration declines during a period of hours to days. The serum creatinine level is elevated in both acute and chronic kidney disease, but an actively rising serum creatinine level confirms an acute or acute-on-chronic insult to kidney function. As a blood filtration organ, the kidney is susceptible to an acute compromise of renal arterial perfusion (Chapter 116), such as prerenal kidney injury, or blockage in urine outflow, such as urinary obstruction due to benign prostatic hypertrophy (Chapter 120). Thus, the patient with acute renal failure is best approached by evaluation for prerenal, renal, and postrenal causes. The

FIGURE 106-12.  Urate crystals. Complex crystals suggestive of acute urate nephropathy or urate nephrolithiasis. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

Potassium balance

Hyperkalemia (Chapters 109 and 214)

Hypokalemia (Chapter 109)

GI losses (Chapter 131)

Acidemic: Diabetic ketoacidosis Renal tubular acidosis (Chapters 110 and 115)

Renal losses

Alkalemic: Diuretics Vomiting Bartter syndrome Gitelman syndrome (Chapters 109 and 110)

Renal failure (decreased GFR) (Chapters 112 and 121)

Elevated arterial blood pressure (Chapter 70)

Low renin

Low aldosterone (Chapter 108)

Low cortisol

ACE-inhibitor Angiotensin II receptor blockers Heparin

Primary hyperaldosteronism Secondary hyperaldosteronism Non-aldosterone mineralocorticoid (Chapters 108 and 214) FIGURE 106-13.  Regulation of potassium balance. ACE = angiotensin-converting enzyme; AKI = acute kidney injury; GI = gastrointestinal.

Decreased K+ secretion

Sickle cell disease Drugs (Chapter 114)

706

CHAPTER 106  Approach to the Patient with Renal Disease  

intrarenal causes of acute kidney injury include acute tubular necrosis (Chapter 112), acute interstitial nephritis (Chapter 114), acute glomerulonephritis (Chapter 113), and acute vasculitis and vascular disease (Chapters 113 and 116). The careful and systematic evaluation of the patient should start with a thorough history and physical examination, which should be followed by selected laboratory tests and often an imaging test, such as renal ultrasonography. Most cases of acute renal failure in the hospital have hemodynamic or toxic causes, so prerenal azotemia and acute tubular necrosis must be considered carefully and distinguished from one another.  

DIAGNOSIS

Laboratory Testing

The normal concentration of blood urea nitrogen (BUN), which is a product of protein catabolism, is about 10-fold higher than the creatinine concentration. Because the BUN-to-creatinine ratio commonly rises with arterial underfilling, BUN typically is used as a marker of effective volume status. Classically, the BUN-to-creatinine ratio will be higher than 15 to 20 in prerenal azotemia but 10 or close to it in acute tubular necrosis. However, the BUN concentration (and hence its ratio to creatinine concentration) may be inappropriately high in other circumstances, such as with high protein intake, gastrointestinal bleeding, or the use of steroids or tetracyclines. The BUN concentration and its ratio to creatinine concentration may be low in patients who have a poor dietary intake of protein, malnutrition, or liver disease. The excretion of sodium in the setting of oliguria and acute kidney injury (Chapter 112) often gives insight into the appropriateness of tubular function. The fractional excretion of sodium (FeNa) is calculated as follows: FE Na = (urine Na plasma Na ) (urineCr plasma Cr ) × 100 where Na is the sodium concentration (in mmol/L) and Cr is the creatinine concentration (in mmol/L or mg/dL). In the setting of oliguria, FeNa below 1% often denotes prerenal azotemia, whereas FeNa above 1% suggests intrinsic renal damage. Although this measurement is generally useful, FeNa below 1% may be seen without evidence of a prerenal component, including contrast nephropathy (Chapter 112), hepatorenal syndrome (Chapter 145), obstructive uropathy (Chapter 120), interstitial nephritis (Chapter 114), glomerulonephritis (Chapter 113), and rhabdomyolysis (Chapter 105). Conversely, a high FeNa can be seen in cases in which there is a prerenal component, including diuretic use, adrenal insufficiency (Chapter 214), cerebral salt wasting, and salt-wasting nephropathy (Chapter 108). The FeNa must be evaluated in the context of the clinical situation because it can be low or high in a normal patient or in a patient with chronic kidney disease. Ultimately, a patient’s volume status is best evaluated at the bedside and should not be deduced solely from a measurement of electrolytes.

Imaging

Ultrasonography, which is the most commonly used renal imaging study (Fig. 106-14), provides reliable information about obstruction, kidney size, presence of masses, and renal echotexture. Ultrasonography has only a 90% sensitivity for the detection of hydronephrosis and hence is not sufficient to exclude obstruction (Chapter 120) with certainty. In addition, its inability to detect stones in the ureters and bladder limits its utility in the evaluation for kidney stones (Chapter 117). Ultrasonography can detect vascular disease,

FIGURE 106-14.  Normal findings on sagittal renal ultrasound. The cortex is hypoechoic compared with the echogenic fat containing the renal sinus. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

and Doppler imaging permits evaluation of the renal vessels with resistive indices. Resistive indices are crucial in ascribing renal dysfunction to the detected vascular disease (Chapter 116). A high resistive index reflects parenchymal disease with scarring and indicates that intervention on the vascular disease itself is unlikely to improve renal function. A CT scan stone protocol to assess the kidneys, ureters, and bladder is the study of choice for detecting kidney stones (Chapter 117) because of its ability to detect all types of stones, nonobstructing stones, and stones in the ureters (Fig. 106-15). Masses in the kidney can be evaluated with either contrast CT or a renal ultrasound examination. CT angiography with iodinated contrast material can assess possible renal artery stenosis (Chapter 116) with an accuracy comparable to that of MR angiography.  

Glomerular Syndromes: Nephrotic versus Nephritic

The nephrotic syndrome (Chapter 113) is characterized by the presence of proteinuria of more than 3.5 g/day/1.73 m2, with accompanying edema, hypertension, and hyperlipidemia. Other consequences include a predisposition to infection and hypercoagulability. In general, the diseases associated with nephrotic syndrome do not cause acute kidney injury, although acute kidney injury may be seen with minimal change disease, human immunodeficiency virus (HIV)–associated nephropathy, and bilateral renal vein thrombosis (Chapter 116). The causes of primary idiopathic nephrotic syndrome, in decreasing order of prevalence, are focal and segmental glomerulosclerosis, membranous nephropathy, minimal change disease, and membranoproliferative glomerulonephritis. Membranous nephropathy has been associated with antibodies to the M-type phospholipase A2 receptor. Secondary causes of the nephrotic syndrome include diabetic nephropathy (Chapter 115), amyloidosis (Chapter 179), and membranous lupus nephritis (Chapters 113 and 250). The acute nephritic syndrome is an uncommon but dramatic presentation of an acute glomerulonephritis (Chapter 113). The hallmark of the acute nephritic syndrome is the presence of dysmorphic RBCs and RBC casts, but their absence does not exclude the syndrome. The acute nephritic syndrome can be caused by any of the rapidly progressive glomerulonephropathies with ANCA-associated vasculitis (granulomatosis with polyangiitis, microscopic polyangiitis, and eosinophilic granulomatosis with polyangiitis), anti–glomerular basement membrane (anti-GBM) glomerulonephritis, and immune complex–mediated glomerulonephritis (including systemic lupus erythematosus, cryoglobulinemia, postinfectious glomerulonephritis, endocarditis, IgA nephropathy, and Henoch-Schönlein purpura). The rapid decline in renal function often warrants urgent and usually inpatient evaluation.  

DIAGNOSIS

Laboratory Testing

Proteinuria (as albuminuria) of more than 3.5 g in 24 hours generally indicates glomerular disease (Chapter 113). Lesser quantities do not preclude glomerular disease, and electrophoresis gives valuable insight into the composition of the proteinuria (Chapter 178). On occasion, overflow proteinuria of a lowmolecular-weight protein, such as light chains in Bence Jones proteinuria, can

FIGURE 106-15.  Delayed excretion in the left kidney secondary to a distal calculus. Contrast-enhanced computed tomography scan shows dilated left renal pelvis. (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

CHAPTER 106  Approach to the Patient with Renal Disease  

be higher than 3.5 g/day without any of the manifestations or implications of the nephrotic syndrome; a urine protein electrophoresis study is important in making the distinction. A comparison of the microalbumin-to-creatinine ratio with the protein-to-creatinine ratio will give an insight into the presence of Bence Jones protein because of the absence of albuminuria despite significant proteinuria. Collection must be done by discarding the first morning void and collecting all urine output for the next 24 hours, including the first morning void the next day. The 24-hour urine collection for protein excretion is cumbersome and subject to inaccuracies. Instead, a spot urine sample for protein and creatinine can be used to estimate the amount of protein excreted. A protein-to-creatinine ratio of 3 translates to a 24-hour protein excretion of about 3 g. The ratio is most accurate when the first morning urine collection is used and may be inaccurate in patients with orthostatic proteinuria. The evaluation of proteinuric renal dysfunction, particularly when glomerular diseases are suspected, should follow a stepwise progression from noninvasive serologic evaluation to a definitive or confirmatory diagnostic evaluation, such as a renal biopsy.12 Sometimes an expeditious diagnosis is needed, and a biopsy may be done relatively early in the evaluation.

Serologies and Genetic Testing

An antinuclear antibody (ANA) titer can be useful to evaluate glomerular disease in either nephrotic or nephritic presentations. A high ANA titer (e.g., 1 : 320), especially if it is accompanied by a more specific finding such as anti–doublestranded DNA antibody or anti-Smith antibody, can be highly specific for the diagnosis of lupus nephritis (Chapter 250), which usually requires a renal biopsy. Lower titers (e.g., 1 : 80 or 1 : 40) are nonspecific. A rheumatoid factor titer will usually be elevated in patients with rheumatoid arthritis (Chapter 248), but vasculitis is a relatively late and rare event. Rheumatoid factor can be detected in some forms of cryoglobulinemia (Chapter 178); for example, IgM and occasionally IgG, which is present in type II and type III cryoglobulinemia, has rheumatoid factor activity. Rheumatoid factor also can be seen as a nonspecific finding in bacterial endocarditis (Chapter 67) and systemic vasculitis (Chapter 254). The levels of complement components C3 and C4 and the 50% hemolyzing dose of complement (CH50) usually are measured to evaluate suspected rapidly progressive glomerulonephritis (Chapter 113). Complement levels are usually low in active systemic lupus erythematosus (Chapter 250), post-streptococcal glomerulonephritis (Chapter 113), endocarditis (Chapter 67), membrano­ proliferative glomerulonephritis, cryoglobulinemia (Chapter 178), shunt nephritis with infection of a ventriculoatrial shunt, and glomerulonephritis associated with visceral abscesses. A particularly depressed C4 compared with C3 should raise the suspicion of cryoglobulinemia. Serum immunoelectrophoresis will detect elevated polyclonal IgA levels in about 50% of cases of IgA nephropathy (Chapter 113) and Henoch-Schönlein purpura (Chapter 113). Polyclonal elevation of IgG may occur in a variety of systemic diseases and is a nonspecific finding. The presence of a monoclonal protein in the serum should raise the suspicion for a monoclonal gammopathy– associated disease (Chapter 178). The differential diagnosis includes monoclonal gammopathy of uncertain significance, myeloma kidney, lymphomas (Chapter 176), amyloidosis (Chapter 179), light chain deposition disease, heavy chain deposition disease, immunotactoid glomerulonephritis, and cryoglobulinemia. These conditions, with the exception of monoclonal gammopathy of undetermined significance, have been collectively called monoclonal gammopathy of renal significance when they affect the kidney. The concentration of the monoclonal protein is higher when the diagnosis of multiple myeloma is made, but even small quantities of Bence Jones proteins in the serum can have clinical significance. A urine immunoelectrophoresis always should be obtained if myeloma is suspected. Because a substantial fraction of multiple myelomas can have no heavy chain excretion and small quantities of light chains may be difficult to detect by serum immunoelectrophoresis, a urine immune electrophoresis test for Bence Jones protein should be obtained. In light chain myeloma, patients may have Bence Jones proteinuria even in the absence of an M component in the serum immunoelectrophoresis. Bence Jones proteinuria may be present in myeloma kidney, amyloidosis, light chain deposition disease, lymphoma, or, occasionally, monoclonal gammopathy of uncertain significance. However, some patients with systemic AL (light chain) amyloidosis have a normal serum immunoelectrophoresis and no Bence Jones proteinuria (Chapter 178). More sensitive assays for serum free light chains and an assessment of the ratio of κ to λ lights chains increase the sensitivity for detection of monoclonal gammopathies.

707

The antineutrophil cytoplasmic antibody (ANCA) assay has allowed earlier and more definitive recognition of vasculitic causes of rapidly progressive glomerulonephritis (Chapter 254), especially granulomatosis with polyangiitis, microscopic polyangiitis, and eosinophilic granulomatosis with polyangiitis, when it is confirmed by enzyme-linked immunosorbent assay. The antibodies cause two different patterns of staining: perinuclear staining (p-ANCA) and cytoplasmic staining (c-ANCA). Both antigens actually have a cytoplasmic distribution, and the perinuclear staining pattern is an artifact of the fixation method. In most cases, the antigen for p-ANCA is myeloperoxidase (MPO), whereas the antigen for c-ANCA is proteinase 3 (PR3). Anti-MPO antibodies are associated with microscopic polyangiitis, idiopathic crescentic glomerulonephritis, or Churg-Strauss syndrome (eosinophilic granulomatosis with polyangiitis; Chapter 254). Anti-PR3 antibodies often correlate with the classic disease of granulomatosis with polyangiitis (formerly known as Wegener granulomatosis) (Chapter 254). Anti–glomerular basement membrane (anti-GBM) antibodies are autoantibodies to the Goodpasture antigen (Chapter 113), which resides in a domain of the α chain of type 4 collagen. An early and accurate diagnosis of Goodpasture syndrome can be made by immunofluorescence and confirmed by Western blot analysis. Anti-GBM antibody staining also may occur in the presence of a positive ANCA. In these cases, the theory is that exposure of the Goodpasture antigen, as a result of the glomerular injury, leads to antiGBM antibody formation as a secondary process. Cryoglobulins (Chapter 178) are thermolabile immunoglobulins. They are a single monoclonal type in type I cryoglobulinemia. In type II and type III cryoglobulinemia, however, the mixture of immunoglobulins includes one with rheumatoid factor activity against IgG. Type I and type II cryoglobulins are more likely to be associated with clinical disease, especially at higher titers. Type III cryoglobulinemia is often of less clinical significance. Type I cryoglobulinemia is seen with Waldenström macroglobulinemia and multiple myeloma (Chapter 178); type II, with hepatitis C infection (Chapters 139 and 140), Sjögren syndrome (Chapter 252), lymphomas (Chapters 176 and 177), and systemic lupus erythematosus (Chapter 250); and type III, with hepatitis C (Chapters 139 and 140), chronic infections, and inflammatory conditions. When cryoglobulinemia is associated with hepatitis C, the hepatitis C virus (HCV) RNA is concentrated in the cryoprecipitate as the antigen; the diagnosis can be made by an RNA assay of the cryoprecipitate at 37° C. Membranous nephropathy is associated with chronic hepatitis B infection with hepatitis B surface antigenemia (Chapter 140). Classic polyarteritis nodosa (Chapter 254) occasionally is seen with chronic hepatitis B infection, often with surface antigenemia and hepatitis B e antigenemia. M-type phospholipase A2 receptor antibodies also have been detected as autoantibodies in idiopathic membranous nephropathy. Hepatitis C serology is associated with a variety of renal diseases, including cryoglobulinemia, membranoproliferative glomerulonephritis, and membranous nephropathy. The evaluation may include the antibody test and an assay for HCV RNA. On occasion, the HCV RNA analysis may have to be conducted on the cryoprecipitate at 37° C. HIV-associated nephropathy (Chapter 113) is associated with nephrotic syndrome and acute kidney injury. In the appropriate clinical setting, HIV serology and viral titers are warranted tests for both clinical syndromes. Streptococcal infection can be confirmed as the cause of postinfectious glomerulonephritis (Chapter 113) with an anti-DNase or antistreptolysin O assay. Acute and convalescent serology assays are used to confirm recent infection. The erythrocyte sedimentation rate (ESR) is a relatively nonspecific test in the evaluation of renal disease. However, a high ESR often points to systemic vasculitis (Chapter 254), multiple myeloma (Chapter 178), or malignant disease as the underlying cause. However, the ESR often is elevated in the nephrotic syndrome (Chapter 113), including diabetic nephropathy (Chapter 115). Exome sequencing can detect pathogenic findings and establish a genetic diagnosis in nearly 10% of patients with chronic kidney disease.13 However, healthy adults also may have potentially pathogenic variants,14 so careful interpretation by expert clinicians is critical for proper diagnosis.

Renal Biopsy

No formal guidelines exist for the indications to perform a renal biopsy. Most nephrologists will perform a biopsy for adults with idiopathic nephrotic syndrome and for children with steroid-dependent or steroid-resistant nephrotic syndrome. In addition, acute kidney injury without an identifiable inciting

708

CHAPTER 106  Approach to the Patient with Renal Disease  

TABLE 106-4 MAJOR CAUSES OF TUBULOINTERSTITIAL DISEASE Ischemic and toxic acute tubular necrosis Allergic interstitial nephritis Interstitial nephritis secondary to immune complex–related collagen vascular disease, such as Sjögren disease or systemic lupus erythematosus Granulomatous diseases: sarcoidosis, tubulointerstitial nephritis with uveitis IgG4-related interstitial nephritis Pigment-related tubule injury: myoglobinuria, hemoglobinuria Hypercalcemia with nephrocalcinosis Tubular obstruction: drugs such as indinavir, uric acid in tumor lysis syndrome Myeloma kidney or cast nephropathy Infection-related interstitial nephritis: Legionella, Leptospira species Infiltrative diseases, such as lymphoma

cause is a clear indication for biopsy. Notably, patients with hospital-acquired kidney failure rarely meet this indication. Other abnormal clinical findings, such as gross or microscopic hematuria or subnephrotic proteinuria, often but not always lead to a kidney biopsy. Renal biopsy usually is performed percutaneously with real-time ultrasound or CT guidance. About 1 to 2% of patients without an underlying coagulopathy will develop bleeding that requires a transfusion. The transjugular approach can be used in patients in whom the risks for bleeding are high.  

Tubulointerstitial Diseases

Tubulointerstitial diseases (Chapter 114) vary in presentation from acute kidney injury to chronic kidney dysfunction that initially is manifested as asymptomatic mild renal insufficiency (Table 106-4). The urine sediment often contains small to moderate amounts of proteinuria, usually less than 1 g/day, as well as WBCs, RBCs, tubule cells, and WBC casts. RBC casts are rare in acute interstitial nephritis and more characteristic of glomerular disease.  

Vasculitis and Vascular Diseases of the Kidney

Vascular diseases of the kidney can be divided into large-vessel obstruction and medium- to small-vessel diseases (Chapter 116). Renovascular disease is a common cause of hypertension, heart failure, and renal insufficiency. About 90% of renal artery stenosis is atherosclerotic in origin, with most of the remaining caused by fibromuscular dysplasia, which is more common in women 20 to 50 years of age. Medium-sized arterial vessel diseases include polyarteritis nodosa, which is seen in patients with hepatitis B (Chapters 139 and 140), HIV infection (Chapter 113), or, rarely, hepatitis C (Chapters 139 and 140). Symptoms include abdominal pain, hypertension, and mild renal insufficiency, often with a benign sediment; diagnostic findings include microaneurysms at the bifurcation of medium-sized arteries. Other diseases involving small vessels include atheroembolic disease (Chapter 116), which is seen after arteriography or surgery or, rarely, spontaneously. This syndrome typically affects the kidneys, gastrointestinal tract, and lower extremities, but it can also involve the central nervous system when the aortic arch is affected. The thrombotic microangiopathies include hemolytic-uremic syndrome and thrombotic thrombocytopenic purpura (Chapter 163). Thrombotic thrombocytopenic purpura is associated with an acquired inhibitor to or the congenital inherited absence of a protease that cleaves large-molecular-weight von Willebrand multimers. The hemolytic-uremic syndrome is caused by endothelial injury. In diarrhea-positive (or typical) hemolytic-uremic syndrome, the endothelial injury is induced by Shiga toxin from Escherichia coli O157:H7 infection. In diarrhea-negative (atypical) hemolytic-uremic syndrome, dysregulation of the alternative complement pathway is the underlying cause of endothelial injury. The antiphospholipid antibody syndrome (Chapter 73) can cause largevessel thrombosis and stenosis as well as a thrombotic microangiopathy with proteinuria, hypertension, and renal insufficiency. Scleroderma renal crisis, which is a manifestation of systemic sclerosis (Chapter 251), often leads to an inexorable progression to end-stage renal insufficiency if untreated. A systemic vasculitis may be manifested in a variety of ways, including skin manifestations such as petechial rash, purpura, digital gangrene, and splinter hemorrhages. Otitis, sinusitis, epistaxis, hemoptysis, and nasal septal ulcers are common manifestations of granulomatosis with polyangiitis (Chapter 254). Pulmonary hemorrhage can be a catastrophic manifestation of Goodpasture syndrome (Chapter 113), anti-GBM disease, or ANCA-associated vasculitis (Chapter 254). Abdominal pain and tenderness and gastrointestinal

FIGURE 106-16.  Magnetic resonance angiography. Coronal three-dimensional image shows right renal artery stenosis (arrow). (From Johnson RJ, Feehally J. Comprehensive Clinical Nephrology. London: Mosby; 2000.)

hemorrhage may be observed in Henoch-Schönlein purpura and classic polyarteritis nodosa (Chapter 254). Neurologic symptoms may be a manifestation of vasculitis, such as microscopic polyangiitis (Chapter 254) and cryoglobulinemia (Chapter 178).  

DIAGNOSIS

Radiologic Evaluation

Magnetic resonance imaging (MRI) with MR angiography (Fig. 106-16) is highly sensitive for detecting atherosclerotic renovascular disease (Chapter 116), but it tends to overestimate the degree of stenosis. Its accuracy in detecting fibromuscular dysplasia, however, is less well validated. MRI also can be used to evaluate renal masses. MRI does not require iodinated contrast material, but gadolinium-based contrast agents for vascular studies are associated with the syndrome of nephrogenic systemic fibrosis in patients with advanced renal failure (Chapter 251). Renal arteriography, which is the “gold standard” in the evaluation of renal artery stenosis (Chapter 116), also is used for the evaluation of arteriovenous malformations, polyarteritis nodosa, and other vascular lesions of the kidneys. This invasive study uses iodinated contrast material and incurs a small risk for atheroembolic disease (Chapter 116). Therapeutic angioplasty and stenting can be done at the time of angiography.  

Papillary Necrosis

Acute necrosis of the renal papilla is associated with sickle cell anemia (Chapter 154), analgesic nephropathy (Chapter 114), diabetic nephropathy (Chapter 115), and obstructive pyelonephritis (Chapter 268). In sickle cell disease (Chapter 154),15 the hypoxic and hypertonic milieu of the inner medulla promotes sickling, and chronic sickling at the vasa recta results in medullary ischemia. Massive and prolonged consumption of analgesics, particularly the combination of aspirin, caffeine, and acetaminophen, is associated with chronic interstitial nephritis and a predisposition to papillary necrosis (Chapter 114); medullary ischemia is thought to be caused by inhibition of synthesis of vasodilatory prostaglandins by aspirin, and direct toxicity is attributed to metabolites of phenacetin. Similarly, medullary perfusion is thought to be compromised in diabetic nephropathy (Chapter 115) and obstructive pyelonephritis (Chapter 120). The clinical manifestations of papillary necrosis can include flank pain and hematuria. If the papilla is sloughed, obstruction may occur at the renal pelvis or ureter of the affected kidney, with referred pain migrating from the flank to the groin. A sloughed papilla may precipitate frank renal failure if the function of the contralateral kidney is impaired or if obstruction occurs at the level of the bladder or urethra. Classically, papillary necrosis is diagnosed on an excretory pyelogram as a calyceal defect after sloughing of a papilla, but CT with contrast enhancement is as good for advanced lesions. If the necrotic papilla is retained, however, the defect will be more subtle. Transitional cell carcinoma (Chapter 187) can occur in the setting of papillary necrosis or can mimic its appearance.

Obstruction, if present, must be relieved, but treatment otherwise is limited to pain control and hydration.  

Chronic Kidney Disease

Chronic kidney disease, which is defined as either kidney damage or a GFR of less than 60 mL/min/1.73 m2 for longer than 3 months, includes five stages (Table 106-5). Kidney damage is defined as pathologic abnormalities or markers of kidney damage, including abnormalities in the composition of blood or urine or abnormalities on imaging tests. The excretion of 30 to 300 mg of albumin in a 24-hour period defines microalbuminuria. An estimated 12% of the adult U.S. population has abnormal albumin excretion in the urine, and the frequency increases with age. Kidney failure is defined as either a GFR of less than 15 mL/min/1.73 m2 that is accompanied by signs and symptoms of uremia or a need for initiation of kidney replacement therapy for treatment of complications of decreased GFR (Fig. 106-17). End-stage renal disease includes all cases requiring treatment by dialysis or transplantation regardless of the level of GFR. Patients with chronic kidney disease warrant referral to a nephrologist. Care of these patients should focus on efforts to slow disease progression, to optimize medical management, and to make a seamless transition to renal replacement therapy (Chapter 121). The care should include optimal blood pressure control, use of angiotensin-converting enzyme inhibitors and angiotensin receptor blockers if indicated, dietary counseling, careful management of calcium and phosphorus levels, control of the parathyroid hormone level, and management of anemia with the use of erythropoietin and iron supplements. A1 

TABLE 106-5 STAGES OF CHRONIC KIDNEY DISEASE* DESCRIPTION

GFR (mL/min/1.73 m2)

1

Kidney damage with normal or ↑GFR

≥90

2

Kidney damage with mild or ↓GFR

60-89

3

Moderate ↓GFR

30-59

4

Severe ↓GFR

15-29

5

Kidney failure

20

U[Na] < 20

Glucocorticoid deficiency Hypothyroidism Stress Drugs SIADH

Acute or chronic renal failure

Nephrotic syndrome Cirrhosis Cardiac failure

FIGURE 108-3.  Diagnostic approach to hyponatremia. RTA = renal tubular acidosis; SIADH = syndrome of inappropriate antidiuretic hormone secretion. (Modified from Halterman R, Berl T. Therapy of dysnatremic disorders. In: Brady H, Wilcox C, eds. Therapy in Nephrology and Hypertension. Philadelphia: Saunders; 1999:256; and Data from Cohen DM, Ellison DH. Evaluating hyponatremia. JAMA. 2015;313:1260-1261.)

A careful evaluation, including a review of prior plasma sodium values, will help determine the rate of decline and provide important clues to the cause (Fig. 108-3). The physical examination should assess for hypervolemia (Chapter 52). Conversely, orthostatic hypotension and tachycardia suggest hypovolemia. The BUN, plasma electrolytes, glucose level, and osmolality should be checked to allow comparison of the measured with the calculated plasma osmolality according to the following equation: Plasma osmolality (mOsm kg) = 2 Na + (mmol L) + (blood urea nitrogen[mg dL] 2.8) + (glucose[mg dL] 18) Other laboratory tests in selected patients should include liver function tests and measurement of plasma creatinine, uric acid, thyroid-stimulating hormone, and cortisol concentrations; if indicated, an adrenocorticotropic hormone stimulation test should be performed (Chapter 214). Symmetrically elevated levels of both BUN and creatinine point to intrinsic renal disease, whereas a disproportionate elevation of BUN over creatinine suggests hypovolemia with a tendency to prerenal azotemia (Chapter 112). In contrast, very low levels of BUN and uric acid are typical of both SIADH and the cerebral salt-wasting syndrome (see Normovolemic and Hypovolemic Hyponatremia). About 85% of inpatients with hyponatremia have true hyponatremia, about 25% of whom are hypovolemic, about 25% of whom have an edematous state, about one third of whom are normovolemic, and the remainder of whom usually have renal failure. Since the plasma sodium concentration declines by approximately 1.6 mmol/L for each increase of 100 mg/dL (5.5 mmol/L) in plasma glucose concentration, marked elevation in the plasma glucose concentration can cause hypertonic hyponatremia. In contrast to hyperglycemia, an elevated BUN does not alter the plasma sodium concentration, even though urea contributes to the laboratory measurement of plasma osmolality; thus, a hyponatremic patient with a normal or elevated laboratory measurement of plasma osmolality that can be fully attributed to an increased BUN should be considered as having hypotonic hyponatremia.

A discrepancy in which measured plasma osmolality exceeds calculated plasma osmolality, even after accounting for glucose and urea, indicates the presence of an unidentified small solute (osmolar gap), including alcohols (e.g., ethanol, methanol, ethylene glycol, and isopropyl alcohol) and the organic anions of weak acids, which raise the plasma anion gap. Because these small molecules do not affect the movement of water, the patient’s water balance is determined by the plasma sodium concentration. However, an osmolar gap should prompt a thorough investigation for poisoning, intoxication, or an organic acidosis (Chapter 110). As soon as true hypotonic hyponatremia is established, the evaluation aims at classifying the cause into one of three categories based on the patient’s volume status (see Fig. 108-3). Abnormal liver function test results can provide adjunctive support for hepatic disease and a hypervolemic hyponatremic state. The diagnosis of heart failure should be made clinically, but it can be assisted by a BNP level, chest radiograph, or echocardiography (Chapter 52). In the absence of a clinically obvious edema, a low urine sodium concentration ( 40

Steady state in 12-36 hours Expected Pco2 = 1.5 (measured HCO3) + 8 ± 2 (Winter equation)

Metabolic alkalosis

Less predictable Expected Pco2 increases 0.5 mm Hg per 1-mEq/L increase in HCO3

Alkalemia Primary respiratory acidosis

Primary metabolic acidosis

Respiratory acidosis Acute

Assess compensation using Winter equation (see Table 110-4) • PCO2 appropriately low—well-compensated primary metabolic acidosis • PCO2 higher than predicted—superimposed respiratory acidosis • PCO2 lower than predicted—superimposed primary respiratory alkalosis

Chronic, after 24-36 hours Respiratory alkalosis Acute Chronic, after 24-36 hours

Calculate anion gap • Normal—hyperchloremic acidosis • High—anion gap metabolic acidosis—look for cause: toxic ingestion (see osmolar gap), uremia, lactic acidosis, or ketoacidosis (check levels) Compare change in gap and change in HCO3 (delta/delta) • 1:1 simple anion gap metabolic acidosis • < 1:1 suspect additional hyperchloremic acidosis • > 1:1 suspect additional metabolic alkalosis FIGURE 110-1.  Evaluation of acidemia.

TABLE 110-3 LABORATORY STEPS IN IDENTIFYING ACID-BASE DISORDERS EVALUATE pH

ACIDEMIC

ALKALEMIC

Elevated Pco2

Respiratory acidosis

Metabolic alkalosis

Elevated HCO3

Respiratory acidosis

Metabolic alkalosis

Decreased Pco2

Metabolic acidosis

Respiratory alkalosis

Decreased HCO3

Metabolic acidosis

Respiratory alkalosis

EVALUATE FOR EXPECTED COMPENSATION Meets expectation: simple disorder with compensation or could be offsetting metabolic alkalosis and acidosis Does not meet expectation: complex disorder, but pH indicates whether acidosis or alkalosis is dominant If a metabolic disorder is dominant, a Pco2 greater than predicted indicates an additional respiratory acidosis. A Pco2 less than predicted indicates an additional respiratory alkalosis. If a respiratory disorder is dominant, an HCO3 concentration greater than predicted indicates additional metabolic alkalosis. An HCO3 concentration less than predicted indicates an additional metabolic acidosis. ASSESS ANION GAP Elevated: metabolic acidosis is present whether acidemic or alkalemic. If alkalemic, an additional metabolic or respiratory alkalosis is present. If the gap is greater than the fall in HCO3, consider an additional metabolic alkalosis or respiratory acidosis. If the gap is less than the fall in HCO3, consider an additional nongap acidosis or respiratory alkalosis.

Compensatory Changes

Few patients have an isolated acid-base disturbance. In nearly all cases, a respiratory or renal compensation (or both) occurs in response to counteract a primary acid-base process. When functioning normally, the lungs may maintain a normal pH and Pco2 during changes in volatile acid production. The kidneys will also maintain normal acid-base balance during changes in fixed acid production. Only excesses beyond the capacity to eliminate an acid or alkali load will lead to clinical disturbances. Patients with renal or lung disease may do less well in response to metabolic and respiratory disorders.

EXPECTED COMPENSATION

Metabolic acidosis

Expected 1-mEq/L increase in HCO3 per 10-mm Hg rise in Pco2 Expected 3- to 5-mEq/L increase in HCO3 per 10-mm Hg rise in Pco2 Expected 1- to 2-mEq/L fall in HCO3 per 10-mm Hg fall in Pco2 Expected 5-mEq/L fall in HCO3 per 10-mm Hg fall in Pco2

When an acid-base disturbance develops, the initial response to modulate its severity depends on the titration of various body buffer pairs. For example, phosphate, hemoglobin, and albumin change their protonated and unprotonated concentrations. The body will further attempt to correct the extracellular pH toward normal but usually not to normal. For metabolic disturbances caused by increased or decreased nonvolatile acid, the response is respiratory; for primary respiratory acidosis and alkalosis, the compensation is renal (Table 110-4). The direction of change in HCO3− and Pco2 is the same when the primary disturbance is compensated; the ratio of HCO3− to Pco2 and thus pH become more normal. These compensations tend to take time, so acidbase disturbances, particularly the respiratory conditions, are classified as acute (lasting less than 24 to 48 hours) or chronic. Peripheral blood does not demonstrate complete compensation for most acid-base disturbances, with the occasional exception of chronic respiratory alkalosis. Full compensation for metabolic acidosis would expend large amounts of respiratory muscle energy, which could limit a prolonged response. Full compensation for metabolic alkalosis would result in excessive hypoventilation and adverse effects on oxygenation. In contrast, the CNS closely regulates its pH, with nearly full correction within 1 to 2 days. Before this compensation occurs, acute alkalemia may be associated with cerebral vasoconstriction and ischemia, whereas acidemia may result in vasodilation and cerebral edema. Rapid changes in blood Pco2 affect the CNS chemosensors more quickly than do changes in HCO3− because of more rapid movement of nonionic CO2 across the blood-brain barrier. Increases in CNS CO2 lead to acidification of the medullary center interstitial fluid and an increased ventilatory drive. Decreases in CNS CO2 (alkalinization of the respiratory center) lead to hypoventilation. Acid-base changes are reflected in the composition of the cerebrospinal fluid (CSF). In metabolic acidosis, peripheral chemosensors in the carotid body stimulate the CNS to increase ventilation to reduce Pco2. The fall in peripheral Pco2 will lead to dissolved CO2 leaving the CNS ahead of HCO3−; the alkalinization of the medullary center interstitial fluid will then slow the hyperventilatory response until a new steady state of hypocapnia is achieved. Patients may sense dyspnea or air hunger acutely with rapid and shallow respirations. In severe cases of metabolic acidemia, the respirations are deep and gasping, typical of Kussmaul breathing. When the bicarbonate concentration increases as a result of metabolic alkalosis, a hypoventilatory response, signaled from the peripheral chemosensors, raises Pco2. As Pco2 rises, the dissolved CO2 will enter the CSF and will acidify the medullary respiratory center. The stimulus to breathe will, in part, antagonize the peripheral signal until a steady state of hypoventilation is reached. The acute stimulus of hypercapnia to increase net renal acid excretion disappears when the stable hypercapnia of chronic respiratory acidosis allows carbonic acid production and elimination to become equal. However, the hypochloremia, brought about by the compensatory early excretion of NH4Cl, and elevated serum HCO3− , maintained by the high Pco2, persist. In respiratory alkalosis, the primary event is a fall in Pco2 because of increased alveolar ventilation. On transition from acute to chronic respiratory alkalosis, the compensatory mechanisms that initially maintained a more normal

735

CHAPTER 110  Acid-Base Disorders  

systemic pH are no longer required as CO2 production and elimination become equal. Thus, the initial compensatory decrease in renal acid excretion brought about by increased loss of filtered NaHCO3 ceases, but low serum HCO3− and high serum Cl− concentrations are maintained. In identifying whether an acid-base disturbance is simple (a single disturbance with its compensation) or complex (multiple primary processes simultaneously present), it is useful to compare the expected compensation for a simple process with the observed parameters of the blood gases (see Table 110-3). For example, if Pco2 is lower than would be predicted in a patient with a simple, compensated metabolic acidosis, an additional respiratory alkalosis must be driving the Pco2 down. If Pco2 is higher than would be predicted for a low bicarbonate level in a patient with metabolic acidosis, a coexistent respiratory acidosis is present.

  METABOLIC ACIDOSIS  

EPIDEMIOLOGY AND PATHOBIOLOGY

In metabolic acidosis, the primary change is a fall in serum bicarbonate. The compensatory response is to increase ventilation to reduce Pco2. Worsening acidosis elicits increasing alveolar ventilation. Primary metabolic acidosis results from an imbalance between net acid production and net acid excretion (NAE) in the form of urinary ammonium excretion and acid phosphate excretion. Consider the following relationship, where Ux represents the urinary concentration and the urinary flow rate V̇ :  + (U phos × V)  − (U − × V)  NAE = (U NH 4 × V) HCO3

In a normal steady-state condition, the rate of excretion of net acid must be equal to the rate of production. The normal production rate depends on diet. If net acid production is normal, metabolic acidosis could occur because of a failure to reabsorb bicarbonate or a failure to elaborate enough urinary buffers, as is the case in renal failure and renal tubular acidosis. An inequality also could develop if net acid production were excessive or if large extrarenal bicarbonate losses were unable to be matched by maximal adaptive increases in net acid excretion. Endogenous sources of acid include ketoacidosis and lactic acidosis, whereas exogenous sources might be acid metabolic products of ingested ethylene glycol or methanol. On occasion, strong inorganic acids may be ingested. When net acid is retained, the serum bicarbonate concentration falls. However, maintenance of a constant low serum HCO3− concentration does not guarantee that there is a new steady state in which net acid production is equal to net acid excretion because body buffers such as carbonate salts of bone may become depleted by relentless acid retention, as in chronic kidney disease and distal renal tubular acidosis. The causes of metabolic acidosis are usually categorized according to the presence of either hyperchloremia or an elevated serum anion gap. The serum anion gap is the net charge difference when the sum of chloride and bicarbonate is subtracted from the serum sodium concentration. Anion gap = [Na + ] − ([Cl − ] + [HCO3 − ]) The normal anion gap is due to the net unmeasured anionic charge associated predominantly with albumin. When acidemia is present, albumin is in a more protonated form, which lowers the normal gap. In alkalemia, the effect of pH is to increase the gap attributed to albumin. Each 1 g/dL of albumin contributes approximately 2.5 mEq/L to the normal anion gap. The anion gap may be low with hypoalbuminemia or with an increase in unmeasured cations, such as immunoglobulin G myeloma paraproteins, calcium, lithium, or magnesium. The anion gap may be high in the presence of unmeasured anions including sulfates, bromides, iodides, and immunoglobulin A myeloma light chains. When the anion gap is increased above the normal value of 10 to 12 mEq/L by a non-chloride acid anion, an anion gap metabolic acidosis exists.6 The accompanying proton is responsible for lowering the serum bicarbonate concentration. The degree of increase in the anion gap, sometimes referred to as the delta anion gap, may be estimated by the difference between the observed anion gap and a normal value of 10 to 12 mEq/L. A similar calculation for a change in serum HCO3− can be made by subtracting the observed HCO3− from the normal value of about 25 mEq/L (the delta HCO3−). Comparison of the two values (the delta-delta) may help identify more complicated acid-base disorders. If the increase in the anion gap is larger than the decrease in serum HCO3− , an additional process is raising the HCO3− level. The patient may have a coexisting metabolic alkalosis or be compensating for chronic respiratory acidosis. If the decrease in serum HCO3− is larger than the increase in the anion gap, it is a sign of another process that raises the Cl− while lowering the HCO3− level, such as an additional hyperchloremic acidosis or respiratory

alkalosis. In most anion gap acidoses, the increase in anion gap and the decrease in HCO3− is not 1 : 1 because the excretion of urinary anions with Na+ results in a hyperchloremic component to the acidosis. Conversely, any buffering of H+ with non-HCO3− buffers will decrease the drop in HCO3− compared with the increase in anion gap. In severe cases of anion gap acidosis, Cl− may be displaced into cells, thereby resulting in a higher anion gap compared with the decrease in HCO3−—a hypochloremic anion gap acidosis.  

CLINICAL MANIFESTATIONS

The effects of metabolic acidosis depend on its rapidity of onset and severity. Patients often complain of fatigue, nausea, vomiting, shortness of breath, and dyspnea on exertion. Acutely, deep respirations, often labored with the use of accessory muscles, may be detected, but hyperventilation may be less notable with chronic metabolic acidemia. Metabolic acidemia also may be associated with vasodilation, tachycardia, and hypotension (Chapter 98). The negative inotropic effect of acidemia on the heart can exacerbate septic shock (Chapter 100). The stress of an underlying illness or an increase in adrenergic and corticosteroid activity associated with acidemia may elevate the peripheral white blood cell count and cause hyperglycemia. Other findings can include hyperkalemia (Chapter 109), hyperphosphatemia (Chapter 111), and hyperuricemia (Chapter 257) as well as hypocalcemia and markers of bone injury7 as a result of decreased renal synthesis of 1,25-dihydroxyvitamin D.  

Anion Gap Metabolic Acidoses

A variety of abnormalities can cause anion gap acidoses. One mnemonic for the common ones is gold mark, for glycols (ethylene and propylene), oxoproline, l-lactate, d-lactate, methanol, aspirin, renal failure, and ketoacidosis. Because some causes are life-threatening, a rapid diagnosis is required. The osmolar gap should be calculated in all cases of anion gap acidosis (Table 110-5) because unmeasured toxic, nonionic alcohols that contribute to body osmolality but not to acidity oxidize to dangerous unmeasured organic acid anions that contribute only to the anion gap. The osmolar gap is defined as the difference between the measured and the calculated serum osmolality. The serum osmolality should be measured by a freezing point depression technique and compared with the calculated osmolality. Calculated osmolality = 2 (Na + ) + (Glucose [mg dL] ÷ 18) + (Blood urea nitrogen [mg dL] ÷ 2.8)  

UREMIC ACIDOSIS

The metabolic acidosis of advanced chronic kidney disease (Chapter 121) may be due to tubular leakage of HCO3− , but it is often present when inadequate ammonia production is unable to facilitate excretion of the normal metabolic acid load.8 Many patients with renal failure can acidify their urine, but the lack of buffering capacity diminishes net acid excretion. Many organic and inorganic anions, such as phosphate and sulfates, are retained at glomerular filtration rates of less than 25 mL/minute and constitute an increased anion gap in association with the metabolic acidosis. The magnitude of the gap is usually less than 20 mEq/L consisting of poorly filtered sulfates and phosphates. The renal patient who is maximally producing NH3 to stay in balance with daily acid production may be unable to accommodate any further acid production, such as a metabolic or respiratory acidosis, that would require increased ammoniagenesis. The patient with poor glomerular filtration will retain HCO3−,

TABLE 110-5 CAUSES OF INCREASED ANION AND OSMOLAR GAPS ANION GAP METABOLIC ACIDOSIS

OSMOLAR GAP

Uremia

No

Lactic acidosis

Variable/no

d-Lactic acidosis

No

Diabetic ketoacidosis

No

Starvation ketoacidosis

No

Alcoholic ketoacidosis

If ethanol is present

Ethylene glycol

Yes

Methanol

Yes

Salicylates

No

5-Oxoprolinuria (acetaminophen)

No

736

CHAPTER 110  Acid-Base Disorders  

thereby worsening both metabolic and respiratory alkalosis. The systemic acid-base disturbance in renal diseases with prominent tubular dysfunction is attributable to the kidney’s inability to secrete hydrogen and to reabsorb and generate HCO3− . It is particularly pronounced in oliguric acute kidney injury and is exacerbated by hypercatabolic states such as infection. A significant metabolic acidosis in a patient with chronic kidney disease of unknown cause should raise the possibility of urinary tract obstruction or chronic tubulointerstitial diseases (Chapter 114), including amyloidosis (Chapter 179), myeloma (Chapter 178), autoimmune disorders, and analgesic nephropathy (Chapter 114). It is important to treat the metabolic acidosis of chronic kidney disease.9 Maintaining the serum HCO3− concentration above 20 to 22 mEq/L, by administering NaHCO3 at a rate of 1 mEq HCO3−/kg/day, will slow the progression of chronic kidney disease, delay end-stage renal failure, A1  and improve nutritional status.10  

PROGNOSIS

In population studies, a low serum bicarbonate level is associated with higher all-cause mortality. The relative risk of death is about 2.6-fold higher in patients with chronic kidney disease and about 1.7-fold higher even without it.  



OVERPRODUCTION OF ENDOGENOUS ACIDS

Lactic Acidosis  

EPIDEMIOLOGY AND PATHOBIOLOGY

Overproduction of lactate may occur with severe exertion, but true lactic acidosis is frequently associated with critical illness, multiorgan failure, and increased mortality. Lactate, which is the final product in the anaerobic pathway of glucose metabolism, is produced from pyruvate in a reaction catalyzed by lactate dehydrogenase:

anaerobic metabolism. Other causes include thiamine deficiency (Chapter 205), hypophosphatemia (Chapter 111), isoniazid toxicity (Chapter 102), and hypoglycemic states (Chapter 217). Metformin may cause lactic acidosis, particularly in elderly patients with cardiac, hepatic, or renal dysfunction. Nucleoside antivirals (Chapter 364), including zidovudine, may cause lactic acidosis and abnormal liver function as a result of toxic mitochondrial effects. Abnormal mitochondrial function is also a feature of aspirin overdose (Chapter 76) or toxicity with hypoglycin from ingestion of the unripe ackee fruit ( Jamaican vomiting sickness). The antibiotic linezolid is another cause of lactic acidosis. Lactic acidosis can also be caused by the overproduction of lactate, which may occur with severe exertion and malignant neoplasms, particularly with a large tumor burden from lymphoma or widely metastatic cancer. Malignant cells can upregulate glycolytic activity, which may increase their uptake of glucose and decrease their dependence on mitochondrion-derived energy. These tumors can use large amounts of available glucose and inorganic phosphate, thereby leading to a syndrome of hypoglycemia, hypophosphatemia, and lactic acidosis.  

TREATMENT  Treatment of lactic acidosis is aimed at correction of the underlying cause. Central venous oxygen saturation should be increased, with a goal of at least 70%, by restoring tissue perfusion and ventilation. In general, the mean arterial pressure should be maintained at 65 to 70 mm Hg, the heart rate below 100 beats/minute, and the hemoglobin level over 7 g/dL. However, specific therapy to increase the clearance of lactate is not of significant incremental value. Even though it can temporarily improve the pH, it does not improve hemodynamics; and furthermore, it adversely lowers ionized serum calcium compared with saline. Sodium bicarbonate can be considered when the arterial pH is below 7.0 or when acidemia has resulted in decreased cardiac inotropy or systemic vasodilation and shock. It is preferable to give NaHCO3 as an isotonic mixture in 5% dextrose and water rather than as a hypertonic bolus, because the latter carries the risk of pulmonary edema and hypernatremia. The quantity of administered sodium bicarbonate to raise arterial pH to 7.2 should be estimated by multiplying the desired minus observed bicarbonate concentration by 50% of body weight. Full correction should be avoided. In patients with a metabolic acidosis after seizures (Chapter 375), the lactate is quickly metabolized to HCO3− by the liver and kidneys, and the acidosis often resolves within 60 minutes. The administration of HCO3− is usually unnecessary and may precipitate an overshoot metabolic alkalosis as the lactate is metabolized, which lowers the seizure threshold. In patients with intestinal bacterial overgrowth (Chapter 131), a syndrome of disorientation, ataxia, and anion gap metabolic acidosis may develop after a carbohydrate meal because of bacterial production of D-lactate. This isomer of the mammalian L-lactate can be measured only by a specific D-lactate assay. The condition is treated with oral antibiotics and appropriate diet.

NADH + pyruvate + H + → lactate + NAD+ A high reduced nicotinamide adenine dinucleotide (NADH)/NAD ratio will favor the formation of lactate. Conversion of ethanol to acetaldehyde and conversion of β-hydroxybutyrate to acetoacetate use NAD and produce NADH. Alcohol metabolism may be associated with excessive β-hydroxybutyrate and lactic acidosis. Lactic acidosis is caused by an imbalance in the rates of lactate production and its clearance, primarily in the liver. Lactic acidosis, which increases the anion gap, is most often due to circulatory failure, hypoxia, and mitochondrial dysfunction that each increase anaerobic glycolysis and the rate of conversion of pyruvate to lactate (Table 110-6). Sepsis (Chapter 100) is associated with an elevated lactate level because of poor clearance and impaired gluconeogenesis. Lactic acidosis can also result from seizure activity (Chapter 375) when lactate is released from muscle cells that have sustained a period of

TABLE 110-6 CAUSES OF LACTIC ACIDOSIS Shock (Chapter 98): septic (Chapter 100), cardiogenic (Chapter 99), or hypovolemic Advanced heart failure (Chapters 52 and 53) Severe trauma (Chapter 103) Severe hypoxemia (Chapter 96) with Pao2 5.3 Fanconi syndrome

Type 4 RTA Tubular disorder

Classic distal RTA, type 1

Type 4 RTA Hypoaldosteronism

Look for cause

Increase blood HCO3 to 18-20 mEq/L Urine pH > 5.3 FEHCO3 < 3%

Urine pH < 5.3

Urine pH > 7.5 FEHCO3 > 15-20%

Proximal RTA

Look for cause FIGURE 110-2.  Evaluation of the patient with suspected rental tubular acidosis, hyperchloremic metabolic acidosis of renal origin. FeHCO3 = fractional excretion of HCO3−; RTA = renal tubular acidosis.

proximal tubule dysfunction. Causes (Table 110-7) include genetic diseases such as glucose-6-phosphatase deficiency (Chapter 152), cystinosis (Chapter 119), hereditary fructose intolerance (Chapter 194), and Wilson disease (Chapter 200). Multiple myeloma (Chapter 178) and Sjögren syndrome (Chapter 252) should be considered in an adult patient. Primary hyperparathyroidism (Chapter 232) results in proximal renal tubular acidosis and hypophosphatemia secondary to inhibition of Na+/H+ exchange and sodium phosphate cotransport in the proximal tubule by parathyroid hormone through cyclic adenosine monophosphate. Hyperparathyroidism is one of the few causes of metabolic acidosis with hypercalcemia. The Cl−/phosphate ratio in plasma may be elevated. Drug toxicity with aminoglycosides, cisplatin, and ifosfamide may cause proximal tubule dysfunction. The antiretroviral drug tenofovir, a nucleotide analogue reverse transcriptase inhibitor, is a cause of the Fanconi syndrome. The syndrome also may be seen after kidney transplantation (Chapter 122).  

Distal Renal Tubular Acidosis

In distal renal tubular acidosis (type 1), failure to produce ammonia leads to an inability to excrete adequate net acid, thereby leading to continuous retention of acid in the body. The degree of acidemia is often severe, with pH reaching values as low as 7.2, whereas urine pH usually exceeds 5.3. Kindreds have been described in which mutations in genes for the distal vacuolar H+-ATPase cause an autosomal recessive distal renal tubular acidosis with deafness. Mutations resulting in defective Cl−/HCO3− exchange protein (AE1) have been linked to an autosomal dominant form of distal renal tubular acidosis.12,13 Distal renal tubular acidosis (see Table 110-7) is also associated with autoimmune disorders, including systemic lupus erythematosus (Chapter 250) and Sjögren syndrome (Chapter 252), and genetic diseases, including sickle cell anemia (Chapter 154), Wilson disease (Chapter 200), Fabry disease (Chapter 197), cystic kidney diseases (Chapter 118), and hereditary elliptocytosis (Chapter 152). Hypercalciuria and hyperoxaluria may cause distal renal tubular acidosis; nephrocalcinosis and nephrolithiasis may be present. Increased proximal tubular citrate reabsorption, as a consequence of the chronic acidosis, also leads to hypocitraturia, which is a risk factor for calcium nephrolithiasis (Chapter 117). A chronically alkaline urine is a risk for pure CaHPO4 stones (brushite). Amyloidosis (Chapter 179) may be manifested as severe acidemia and other tubular dysfunction, including nephrogenic diabetes

insipidus. Chronic tubulointerstitial diseases of the kidney (Chapter 114), including reflux nephropathy (Chapter 119) and urinary obstruction, may result in renal tubular acidosis with hypokalemia or hyperkalemia. Acute tubulointerstitial nephritis also may result in renal tubular acidosis. Drugs such as amphotericin B can cause hypokalemic distal renal tubular acidosis. Topiramate, used for migraines, is a carbonic anhydrase inhibitor that may cause mixed proximal and distal renal tubular acidosis.  

HYPERCHLOREMIC METABOLIC ACIDOSIS OF RENAL ORIGIN ASSOCIATED WITH HYPERKALEMIA

Hyperkalemic, hyperchloremic acidosis (type 4) suggests dysfunction of the cortical collecting duct, where acidification of urine and disorders in potassium secretion may occur. Some patients with high blood potassium and hyperchloremic acidosis can lower urinary pH below 5.3, whereas others appear to have defects in both potassium balance and urinary acidification. Hyperkalemia itself may worsen metabolic acidosis by decreasing NH3 accumulation by countercurrent multiplication in the medullary interstitium. Causes include hyporenin-hypoaldosteronism, as seen in diabetic renal disease (Chapter 115); other tubulointerstitial diseases (Chapter 114), usually with some renal impairment; sickle cell anemia (Chapter 154); and the use of drugs such as β-blockers and nonsteroidal anti-inflammatory drugs. Low renin and aldosterone levels can also be found in cases of volume expansion with hypertension. Cyclosporine and tacrolimus may lead to decreased electrical driving forces for K+ and H+ secretion. Hyperkalemic acidosis with elevated renin and low aldosterone is found in adrenal insufficiency (Chapter 214), in isolated hypoaldosteronism (Chapter 214), and with the use of angiotensinconverting enzyme inhibitors, renin inhibitors, and angiotensin II receptor blockers. High renin and aldosterone levels are anticipated when the renal collecting duct cell is insensitive to aldosterone, as in urinary tract obstruction, sickle cell anemia, amyloidosis, and systemic lupus erythematosus. Inhibition of aldosterone action with spironolactone or eplerenone may cause hyperkalemic acidosis, as does ENaC inhibition by amiloride, triamterene, trimethoprim, and lithium. Autosomal recessive pseudohypoaldosteronism type 1 is due to inactivating mutations of the sodium channel ENaC, whereas autosomal dominant pseudohypoaldosteronism type 1 is due to mutations of the mineralocorticoid receptor. Both cause hypovolemia, metabolic acidosis, and hyperkalemia with secondary increases in renin and aldosterone. In Gordon syndrome

740

CHAPTER 110  Acid-Base Disorders  

TABLE 110-7 CAUSES OF RENAL TUBULAR ACIDOSIS* HYPOKALEMIC DISTAL (TYPE 1) RTA Hereditary tubule disorders Vacuolar H+-ATPase β-subunit gene mutations Carbonic anhydrase type II deficiency Cl−/HCO3− exchanger (AE1) mutations Genetic causes Sickle cell anemia Fabry disease Wilson disease Elliptocytosis Paroxysmal nocturnal hemoglobinuria Medullary cystic kidneys Autoimmune disorders Systemic lupus erythematosus Sjögren syndrome Multiple myeloma and amyloidosis Drugs: amphotericin, cisplatinum, aminoglycosides Nephrocalcinosis and hypercalcemic disorders Tubulointerstitial diseases Acute tubulointerstitial nephritis Reflux nephropathy Analgesic nephropathy PROXIMAL (TYPE 2) RTA Hereditary tubule disorders NaHCO3 cotransport (NBC) mutations Carbonic anhydrase deficiency Generalized proximal tubular dysfunction Hereditary Fanconi syndrome Genetic diseases: cystinosis, glycogen storage disease (glucose-6-phosphatase deficiency), Wilson disease Hormonal: hyperparathyroidism, vitamin D deficiency Multiple myeloma Lysozymuria Sjögren syndrome Renal transplantation Heavy metals: cobalt, mercury, lead Drugs: ifosfamide, outdated tetracycline, tenofovir, tacrolimus, aminoglycosides HYPERKALEMIC (TYPE 4) RTA Renal diseases—aldosterone resistance Diabetes mellitus Amyloidosis Systemic lupus erythematosus Urinary tract obstruction Hyporeninism Autonomic neuropathy (diabetic) Sickle cell anemia Primary hypoaldosteronism Adrenal insufficiency: Addison disease Tubular mutations: pseudohypoaldosteronism Drugs: potassium-sparing diuretics, amiloride, triamterene, spironolactone, nonsteroidal anti-inflammatory drugs, lithium, trimethoprim, cyclosporine, tacrolimus, renin inhibitors, angiotensin-converting enzyme inhibitors, angiotensin II receptor antagonists

with glycosuria, phosphaturia, aminoaciduria, and uricosuria. In proximal renal tubular acidosis, the steady-state urine pH is usually less than 5.3, the acidosis is not severe (i.e., HCO3− usually not less than 16), and acid excretion may balance acid production at this new steady state. In contrast to proximal renal tubular acidosis, distal renal tubular acidosis (type 1) is generally a more severe metabolic disorder that may be accompanied by hypercalciuria, nephrocalcinosis, calcium phosphate kidney stones (Chapter 117), and bone disease that includes rickets in children and osteomalacia in adults. Proximal and distal renal tubular acidoses usually can be distinguished by a careful clinical evaluation (Fig. 110-2). Helpful findings include the presence of a urine pH greater than 5.3 in distal but not in proximal renal tubular acidosis during acidemia; a fractional excretion of bicarbonate as high as 10 to 15% in proximal renal tubular acidosis; and the lowering of serum potassium on correction of proximal but not of distal tubular acidosis. In patients with an elevated serum anion gap, unmeasured anions such as keto acids and lactate, rather than NH4+, are present in urine, so a positive urinary anion gap does not indicate renal tubular acidosis. On occasion, the prompt renal excretion of organic anions with sodium and potassium may minimize the increase in the serum anion gap. In the metabolic acidosis of glue sniffers, hippurate, which is a product of toluene, is rapidly excreted, thus giving the appearance of a nongap metabolic acidosis with a positive urinary anion gap. Similarly, if keto acids are completely cleared into the urine, ketoacidosis may be manifested as a hyperchloremic acidosis rather than as an anion gap acidosis.

TREATMENT  If possible, treatment of metabolic acidosis should focus on correction of the underlying cause, such as discontinuation of an offending drug, permitting the body’s homeostatic mechanisms to correct the acid-base disturbance. Patients whose pH is less than 7.2 are typically treated with infusions of sodium bicarbonate, guided by the estimated base deficit in milliequivalents, calculated by the serum HCO3− concentration in milliequivalents per liter: Amount of HCO3 − = (25 − [HCO3 − ]) × wt (kg) 2 Such treatment appears to be especially beneficial in patients with acute kidney injury. A2  In general, the correction of metabolic acidemia should be based on a calculated amount, with not more than 50% of the estimate given before recalculation. Moreover, this equation is used for deficit correction only; the ongoing losses of 1 to 2 mEq/kg per day, equivalent to the daily acid load, should be replaced in distal renal tubular acidosis with NaHCO3, KHCO3, or citrate salts in divided doses. Hypokalemia may accompany distal renal tubular acidosis and may improve with treatment. Citrate should be avoided as an alkalinizing salt in patients with low glomerular filtration rate. Proximal renal tubular acidosis in children may affect growth and require large quantities of bicarbonate in excess of 1 to 2 mEq/kg per day to correct the acidosis because ingested alkali is promptly excreted in alkaline urine. In adults, treatment is often deferred because the steady-state acidosis allows a normal acid excretion rate. Hypokalemia may worsen with bicarbonate treatment of proximal tubular acidosis. In type 4 renal tubular acidosis, treatment of hyperkalemia with a lowpotassium diet, thiazide, or loop diuretics or sodium polystyrene sulfonate often improves urinary acidification without the use of bicarbonate salts.

*Type 3 renal tubular acidosis (RTA) is not listed separately because it is an overlap of proximal and distal dysfunction.



(pseudohypoaldosteronism type 2), increases in Na+ and Cl− reabsorption through increased activity of the distal thiazide-sensitive NaCl transporter lead to hypertension, hyperkalemic acidosis, volume expansion, and consequently low renin and aldosterone.  

CLINICAL MANIFESTATIONS AND DIAGNOSIS

The urinary anion, or charge, gap helps distinguish renal tubular acidosis from extrarenal bicarbonate loss (e.g., from diarrhea).14 Because the normal renal response to metabolic acidosis is an increase in ammoniagenesis, the urine should contain large amounts of NH4Cl while the kidney retains sodium and potassium; the urinary charge gap, which is (Na+ + K+) − Cl− , should be strongly negative because of the unmeasured NH4+. In renal diseases such as distal renal tubular acidosis, however, the urinary anion gap will be zero or positive because of either the failure of ammoniagenesis or the excretion of sodium plus potassium with bicarbonate. With type 2 (proximal) renal tubular acidosis, patients often have Fanconi syndrome

PROGNOSIS

The prognosis of renal tubular acidosis generally depends on the presence of an underlying systemic disease, such as myeloma (Chapter 178). In children, disorders such as medullary cystic kidney disease (Chapter 118) and cystinosis (Chapter 119) usually result in renal failure by the teenage years. These patients are candidates for renal replacement therapy, including transplantation. Chronic metabolic acidosis in children, if not well treated, is associated with rickets (Chapter 231) and short stature.

  METABOLIC ALKALOSIS  

EPIDEMIOLOGY AND PATHOBIOLOGY

In metabolic alkalosis, the primary event is elevation of the plasma bicarbonate concentration. In response to increased systemic pH, alveolar ventilation is decreased to increase Pco2 and thereby decrease pH. However, respiratory compensation is generally less effective in cases of metabolic alkalosis than in cases of metabolic acidosis. Contributing factors may include the fact that hypoventilation also decreases Po2, which is a potent stimulus for

CHAPTER 110  Acid-Base Disorders  

the peripheral chemoreceptors to increase alveolar ventilation when Po2 falls below about 60 mm Hg. A second mechanism that may blunt respiratory compensation is intracellular acidosis in the brain in the setting of hypokalemia. In acute metabolic alkalosis, an initial paradoxical acidotic shift in CSF pH secondary to a sudden increase in Pco2, analogous to the alkaline shift in CSF pH in acute metabolic acidosis, may activate central chemoreceptors and increase ventilatory drive despite peripheral stimulation to decrease alveolar ventilation. In chronic metabolic alkalosis, CSF pH may return to normal, so respiratory drive is controlled entirely by the peripheral chemoreceptors. The result is that the ventilatory response to metabolic alkalosis is highly varied: many patients with metabolic alkalosis maintain nearly normal Pco2 levels, and the level rarely rises above 60 mm Hg. Metabolic alkalosis requires a generation phase, in which new HCO3− is added to the extracellular fluid, and a maintenance phase, in which the new elevated serum HCO3− concentration is sustained. Without the maintenance phase, a kidney with normal filtration and tubular function has a high capacity to excrete HCO3−, thereby preventing alkalosis. Maintenance of a high HCO3− concentration usually occurs because of volume depletion, reduced glomerular filtration rate, hypokalemia, or low chloride levels.

Metabolic Alkalosis of Renal Origin Associated with Volume Depletion  

Metabolic alkalosis of renal origin may be the result of excessive urinary chloride excretion, most commonly related to diuretics that inhibit reabsorption of Cl− . The Cl− loss results in hypochloremia, with a compensatory increase in plasma HCO3− to maintain electroneutrality. Extracellular volume depletion stimulates the renin-angiotensin-aldosterone pathway, and high aldosterone levels superimposed on increased distal urinary flow rates result in increased K+ excretion and hypokalemia. The volume depletion and hypokalemia enhance proximal HCO3− reabsorption, thereby maintaining the alkalosis, and the prerenal fall in the glomerular flow rate limits HCO3− filtration. Important but rare genetic syndromes characterized by urinary chloride wasting include Bartter syndrome and Gitelman syndrome. Bartter syndrome is an autosomal recessive salt-losing state associated with extracellular volume depletion and excessive urinary chloride loss that results in hypokalemia and hypochloremic metabolic alkalosis. Secondary increases of plasma renin and aldosterone occur, as does renal juxtaglomerular cell hyperplasia. The syndrome resembles the effects of furosemide on the thick ascending limb of Henle; gene mutations in the Na-K-2Cl cotransporter, the renal outer medullary potassium channel (ROMK), and chloride channels have been described. Because calcium reabsorption occurs in the thick ascending limb of Henle, Bartter syndrome (Chapter 119), like furosemide, causes hypercalciuria and nephrocalcinosis as well as polyuria due to decreased urinary concentrating ability. Gitelman syndrome is an autosomal recessive cause of extracellular volume depletion, urinary chloride wasting, and hypokalemic metabolic alkalosis. It is due to inactivating mutations in the SLC12A3 gene encoding the thiazidesensitive NaCl cotransporter of the renal distal tubule. Urinary concentrating ability is preserved, and patients are hypocalciuric because decreased NaCl reabsorption in the distal tubule is associated with a decrease in calcium excretion. Hypomagnesemia may also be severe.

Metabolic Alkalosis of Nonrenal Origin with Extracellular Volume Depletion  

Metabolic alkalosis may develop as a result of gastrointestinal Cl− loss from vomiting, nasogastric suctioning, or secretory diarrhea. In such cases, extracellular volume is usually contracted, hypochloremia develops, and the urinary chloride level is usually less than 20 mEq/L. In Zollinger-Ellison syndrome (Chapter 219), excessive gastrin-induced gastric acid secretion may result in an acidic stool with high chloride content. Diarrhea does not cause metabolic alkalosis unless the electrolyte relationship [(Na+ + K+) − Cl−] in the stool is less than plasma HCO3− . Infectious gastroenteritis, congenital chloridorrhea, and villous adenomas also cause chloride losses in stool. Congenital chloridorrhea is an autosomal recessive disorder of defective intestinal apical Cl−/HCO3− exchange associated with the downregulated adenoma (DRA) gene. With vomiting, the initiating event is loss of HCl. This secretion of HCl into the stomach lumen by the parietal cell is coupled to the absorption of HCO3− in exchange for chloride at the basolateral membrane. When gastric acid is normally secreted, a mild increase in serum HCO3− spills into urine

741

and causes an “alkaline tide.” With vomiting, however, the net loss of HCl generates the alkalosis. Initially, this increased HCO3− is filtered by the glomeruli and excreted in urine accompanied by Na+ and K+; volume depletion begins to develop. As vomiting continues, extracellular volume depletion worsens, glomerular filtration falls, HCO3− filtration is limited, volume depletion increases the renin–angiotensin II–aldosterone system, proximal fluid and HCO3− reabsorption increase, distal Na+ reabsorption increases under the influence of aldosterone, and greater H+ secretion enhances HCO3− reabsorption. These effects reduce renal Na+ loss but at the expense of maintaining the metabolic alkalosis. Significant K+ losses, which occur as a result of the bicarbonaturia and hyperaldosteronism, lead to hypokalemia, which is actually due to renal, not gastrointestinal, losses as a consequence of attempts to maintain extracellular volume. The hypokalemia further increases proximal NaHCO3 reabsorption, distal H+ secretion, and K+ reabsorption, all at the expense of further reabsorption of HCO3− . At the new steady state after vomiting or nasogastric suctioning ceases, the paradoxical aciduria of metabolic alkalosis develops as HCO3− reabsorption is complete, and the urine contains low levels of Na+, K+, and Cl− . The patient may be hypovolemic, hypokalemic, and alkalemic, but because Na+, K+, and acid-base balance are intrinsically linked, lifethreatening volume depletion, potassium depletion, and alkalemia are usually avoided. Most nonrenal metabolic alkaloses with volume depletion are due to gastrointestinal losses. However, some patients with cystic fibrosis (Chapter 83) may develop hypochloremic alkalosis as a consequence of excessive sweat chloride content related to the CFTR gene mutation. The sweat gland, like the principal cell in the kidney, contains the aldosteronesensitive epithelial sodium channel, so Na+ absorption from the glandular duct renders the lumen electronegative. When Cl− absorption is decreased in cystic fibrosis (Chapter 83), the lumen becomes more negative, thereby decreasing Na+, Cl− , and fluid absorption and also leading to salty sweat; the proportionally large Cl− loss generates a hypochloremic metabolic alkalosis.

Metabolic Alkalosis of Renal Origin with Volume Expansion and Hypertension  

The renal conditions that cause metabolic alkalosis and volume expansion are due to a proportionately greater increase in Na+ reabsorption above what is required to maintain a steady state of Na+ balance, rather than primary loss of the Cl− anion. As Na+ is reabsorbed, electroneutrality is maintained by an increase in plasma HCO3− ; Cl− balance is normal, Cl− appears in urine, and hypochloremia is not present. In the kidney, the loss of net acid as NH4Cl in excess of the acid produced generates a metabolic alkalosis, in which the new bicarbonate generated is due to proton secretion by the distal nephron through H+-ATPases. The H+ then combines with NH3 to form NH4+ in urine. Na+ is reabsorbed independently of Cl− in the cortical collecting duct through the aldosterone-sensitive cells containing the ENaC. When Na+ is reabsorbed by the principal cells of the cortical collecting duct, the tubule lumen becomes electronegative and stimulates both K+ and H+ secretion by the electrogenic H+-ATPases. To the extent that HCO3− remains in the lumen, the secreted protons complete HCO3− reabsorption. Additional secreted protons combine with NH3 and phosphates and lead to net acid excretion. Any increase in the distal H+ secretory mechanism will produce more urinary net acid; more new HCO3− will be generated and returned to the now expanded extracellular fluid, and metabolic alkalosis will develop. The increased plasma HCO3− will be filtered, but in the absence of a stimulus to increase proximal HCO3− reabsorption, the HCO3− will flow distally to be reabsorbed by the increased H+ secretion of the collecting duct. At first, the alkalosis is mild, but increased cortical collecting duct Na+ reabsorption will also lead to increased K+ secretion and hypokalemia. Hypokalemia increases the capacity for proximal HCO3− reabsorption, thereby opposing the effect of volume expansion, so that distal delivery of HCO3− decreases. The higher than normal distal H+ secretion titrates urinary buffers, so further new HCO3− is formed and the alkalosis worsens. Metabolic alkaloses in the hypermineralocorticoid syndromes are sustained by hypokalemia.

Metabolic Alkalosis of Nonrenal Origin Associated with Normal or Expanded Volume  

If an alkalotic patient is not hypochloremic, electroneutrality must be maintained either by depletion of an alternative anion or by an excessive concentration of a cation. An example of a metabolic alkalosis associated with depletion of a non-chloride anion is hypoproteinemic alkalosis, with hypoalbuminemia and a small anion gap. Chloride balance is normal and chloride appears in urine.

742

CHAPTER 110  Acid-Base Disorders  

Alkalosis also may result from the addition of alkali salts of organic anions. The normal response to the ingestion of NaHCO3 is rapid urinary alkalinization because of an unaltered threshold for HCO3− reabsorption. However, a marked excess of HCO3− , as may be administered in an attempt to alkalinize a patient’s urine, expands volume and causes an alkalemia, especially in the presence of volume depletion or low glomerular filtration. Milk-alkali syndrome, usually seen when patients in renal failure ingest milk or calcium antacids, is associated with hypercalcemia, alkalemia, and normal chloride concentration. Other situations in which intake of alkali salts results in metabolic alkalosis include infusion of large quantities of sodium salts of metabolizable organic compounds, such as acetate, citrate, lactate, or bicarbonate; hyperalimentation with acetate salts; chronic peritoneal dialysis with acetate or lactate dialysate; and excessive transfusions or plasmapheresis, in which large quantities of citrate, used as an anticoagulant, are delivered.  

above 25 mEq/L may be taking diuretics such as furosemide or thiazides surreptitiously; a diuretic screen can document the presence of the drug. If the screen is negative, Bartter or Gitelman syndrome (Chapter 119) should be considered. Bartter syndrome is less common, is usually more severe, and presents in young patients. The presence of hypercalciuria favors Bartter syndrome, whereas hypocalciuria and hypomagnesemia suggest Gitelman syndrome. Specific causes of renal alkalosis with volume expansion and hypertension can be classified according to levels of renin and aldosterone. Primary increases in renin with secondary increases in aldosterone can be seen in patients with unilateral renal artery stenosis (Chapter 116), renin-secreting tumors of the kidney (Chapter 70), and malignant hypertension (Chapter 70). Low renin and elevated aldosterone levels are characteristic of primary hyperaldosteronism from adrenal adenoma or hyperplasia (Chapter 214). Cortisol activation of mineralocorticoid receptors is normally limited by its conversion to inactive cortisone by the intracellular enzyme 11β-hydroxysteroid dehydrogenase Type 2. However, a high cortisol level with volume expansion is seen in hypercortisolism and adrenocorticotropic hormone–secreting tumors (Chapter 214). Inhibition of this enzyme will also result in low renin levels, low aldosterone levels, and hypokalemic alkalosis. Both genetic mutations (the apparent mineralocorticoid excess syndrome) and an excess consumption of glycyrrhizic acid found in licorice and anisette are causes of this enzyme block. Another cause of hypertension with hypokalemic alkalosis but with low renin and aldosterone levels is Liddle syndrome (Chapter 119), in which an activating mutation in the cortical collecting duct sodium channel (ENaC) leads to increased Na+ reabsorption. Metabolic alkalosis may also develop without volume expansion when a non-reabsorbable anion is presented to the cortical collecting duct lumen. Nitrates, sulfates, and certain antibiotics such as nafcillin, carbenicillin, and ticarcillin may obligate K+ and H+ secretion as Na+ is reabsorbed. Topical administration of silver nitrate to burn victims may result in alkalosis.

CLINICAL MANIFESTATIONS

Mild metabolic alkalosis up to a pH of 7.50 is usually asymptomatic. When the pH exceeds 7.55, however, the alkalosis itself and the compensatory hypoventilation are frequently associated with metabolic encephalopathy. Symptoms include confusion, obtundation, delirium, and coma. The seizure threshold is lowered; tetany, paresthesias, muscle cramping, and other symptoms of low calcium are seen. In patients with hypocalcemia, these signs may be seen at pH values above 7.45. Other findings include cardiac tachyarrhythmias and hypotension. Lactate production increases as a result of the increased anaerobic glycolysis.  

DIAGNOSIS

In diagnosis of the cause of metabolic alkalosis, it is important to distinguish whether the condition is chloride responsive or chloride unresponsive. Metabolic alkalosis is generally divided into two categories on the basis of its responsiveness to chloride (see Table 110-2). Chloride-responsive metabolic alkalosis is associated with extracellular fluid and chloride depletion and is seen in cases of gastric fluid loss and diuretic use. A diagnostic clue comes from the serum electrolytes. HCO3− is increased with a corresponding fall in serum chloride (hypochloremic alkalosis). Chloride-unresponsive metabolic alkalosis is seen in patients with extracellular fluid expansion in conditions such as primary aldosteronism and hypokalemia. Entry of hydrogen ions into cells can also lead to metabolic alkalosis in patients with hypokalemia. Vomiting, nasogastric suction, and diarrhea are usually obvious sources of metabolic alkalosis. However, the Zollinger-Ellison syndrome (Chapter 219), villous adenomas (Chapter 184), and VIPomas (Chapter 219) may be more difficult to diagnose unless the index of suspicion is high. Patients who present with hypokalemic metabolic alkalosis (Fig. 110-3) with normal or low blood pressure and have urinary chloride concentrations

TREATMENT  In chloride-responsive patients (see Table 110-2), treatment is directed at increasing urinary excretion of bicarbonate. In patients with mild to moderate alkalosis, liberalizing salt intake and administering potassium chloride is effective in increasing renal HCO3− excretion. The K+ deficit is likely to be at least 100 mEq for every decrease of 1 mEq/L in serum potassium. Unless potassium chloride is also replenished, the improvement in filtration and proximal reabsorption will result in severe potassium wasting as bicarbonaturia develops and aldosterone’s effects remain. In addition, complete resolution of alkalosis will not occur until K+ is normalized. In a patient with renal failure and vomiting, the elevation in HCO3− may be more severe because of poor HCO3− filtration.

Hypokalemic metabolic alkalosis

Normotension or hypotension

Hypertension

Low renin

Using diuretics?

Yes

UCI > 20 mEq/L

Low aldosterone

No

UCI > 20 Vomiting Nasogastric suctioning Secretory diarrhea

Consider Bartter or Gitelman syndrome

High cortisol level Hypercortisolism ACTH-secreting tumor

HSD11B2 inhibition Glycyrrhizic acid Apparent mineralocorticoid excess syndrome

Elevated renin

High aldosterone

Liddle syndrome

Malignant hypertension Reninoma Renal artery stenosis

Primary hyperaldosteronism Adrenal adenoma Adrenal hyperplasia

FIGURE 110-3.  Evaluation of the patient with hypokalemic metabolic alkalosis. ACTH = adrenocorticotropic hormone; HSD11B2 = 11β-hydroxy-steroid dehydrogenase Type 2.

CHAPTER 110  Acid-Base Disorders  

In cases of volume expansion and alkalosis, acetazolamide may be administered carefully while monitoring its potential for losing K+. If this agent fails to work, dilute solutions of HCl (0.1N HCl) may be cautiously administered. The amount of H+, in milliequivalents, to be given may be calculated as the product of the desired change in serum HCO3− concentration (mEq/L) times 0.5 of body weight in kilograms. It is likely that this calculation will overestimate the amount of acid needed for correction, so no more than one third of the amount should be given before recalculating to avoid metabolic acidosis. Full correction of HCO3− should not be the goal. In the absence of renal failure, intravenous acetazolamide (250 to 500 mg every 8 hours) may be effective but may greatly increase K+ losses. Chloride-unresponsive patients (see Table 110-2) include those with mineralocorticoid excess. In these patients, the metabolic alkalosis can be lessened by potassium replacement or by blocking Na+ reabsorption with aldosterone antagonists such as spironolactone, starting with 25 mg orally, or amiloride, beginning with 5 mg orally. Indomethacin effectively treats Bartter syndrome (Chapter 119) by interfering with prostaglandin E2 to allow greater NaCl reabsorption in the thick ascending limb. Gitelman and Bartter syndromes are best treated with combinations of potassium chloride, a potassium-sparing diuretic, and magnesium if needed.

  RESPIRATORY ACIDOSIS

Respiratory acidosis is characterized by a primary elevation in Pco2 as reflected by reduced arterial pH with variable elevation in the HCO3− concentration. It is most frequently caused by a decrease in alveolar ventilation due to pulmonary disease (Chapter 96), respiratory muscle fatigue, musculoskeletal abnormalities of the chest wall, or abnormalities in ventilatory control (Chapter 80).  

CLINICAL MANIFESTATIONS

Clinical findings in respiratory acidosis are related to the degree and duration of the respiratory acidosis and whether hypoxemia is present. A precipitous rise in Pco2 can lead to confusion, anxiety, psychosis, asterixis, seizures, and myoclonic jerks, with progressive depression of the sensorium and coma at an arterial Pco2 greater than 60 mm Hg (CO2 narcosis). Hypercapnia, which increases cerebral blood flow and volume, can lead to symptoms and signs of elevated intracranial pressure, including headaches and papilledema. Other findings in acute respiratory acidosis include signs of catecholamine release, such as skin flushing, diaphoresis, and increased cardiac contractility and output. Symptoms of chronic hypercapnia include fatigue, lethargy, and confusion, in addition to the findings seen in acute hypercapnia. The slow time course of many of these diseases allows the kidney to compensate adequately by increasing its excretion of hydrogen ion as ammonium and generating and reabsorbing bicarbonate to restore systemic pH toward normal values. This compensatory process is not maximal until 3 to 5 days after the onset of respiratory acidosis. Chronic NaHCO3 retention and edema often accompany chronic respiratory acidosis.  

DIAGNOSIS

Disorders that cause a respiratory acidosis include central effects of drugs, stroke, and infection; airway obstruction; primary parenchymal processes, such as chronic obstructive pulmonary disease (Chapter 82) and acute respiratory distress syndrome (Chapter 96); disorders of ventilation (Chapter 80); and neuromuscular diseases, such as myasthenia gravis (Chapter 394) and muscular dystrophies (Chapter 393). Permissive hypercapnia has been used clinically in patients with acute respiratory distress syndrome to limit pulmonary damage secondary to mechanical ventilation (Chapter 97).

TREATMENT 

pulmonary arteriolar vasodilation increases perfusion to poorly ventilated alveoli. Patients recovering from an acute-on-chronic respiratory acidosis should be monitored carefully to correct hypokalemia, hypochloremia, and hypovolemia so that adequate renal excretion of bicarbonate can occur. Bicarbonate therapy is not indicated for respiratory acidosis unless the pH falls below 7.0 and the patient is about to be intubated. There is a role for bicarbonate therapy in patients with renal failure (Chapter 121), in whom adequate compensatory acid excretion cannot take place.

  RESPIRATORY ALKALOSIS  

EPIDEMIOLOGY AND PATHOBIOLOGY

In respiratory alkalosis, a primary decrease in Pco2 is reflected by increases in arterial pH and variable decreases in plasma bicarbonate concentration. The most common cause is alveolar hyperventilation, not underproduction of CO2. Acute hypocapnia results in an initial increase in the pH of both the CSF and the brain’s intracellular environment. However, this increase is quickly offset by a decrease in bicarbonate levels. In acute respiratory alkalosis, one of the primary mechanisms of this fall in bicarbonate appears to be the generation of lactate as a result of vasoconstriction, hypoxia, and increased hemoglobin affinity for oxygen. The combination of increased oxygen demand and decreased oxygen delivery may contribute to adverse clinical outcomes in hypocapnic alkalosis. Cerebral blood flow is significantly decreased by hypocapnia, which is a potent vasoconstrictor. As in respiratory acidosis, the CNS is immediately affected because of the blood-brain barrier’s permeability to CO2. In addition, as in respiratory acidosis, CSF and intracellular pH show an initial short-lived response that parallels the systemic increase in pH. Renal compensation for sustained hypocapnia is complete in 36 to 72 hours. The mechanism rests primarily in the kidney’s net reduction of hydrogen ion excretion, which it accomplishes largely by decreasing ammonium and titratable acid excretion. The threshold for bicarbonate excretion is also lowered, and bicarbonaturia develops. As a result, systemic bicarbonate levels decrease, and arterial pH returns toward normal values. Acute exposure to high altitude (Chapter 88) results in hypoxia-induced hyperventilation. Compensation requires several days and is characterized by a gradual further increase in hyperventilation, a steadily decreasing Pco2, and a recovering Po2. The effect of the hypoxic stimulus to ventilate is initially modulated by the effects of alkalosis, both peripherally and centrally. However, as HCO3− falls in the CSF, inhibition of the central stimulus to ventilate decreases. Once a steady state is achieved, the drive to ventilate is determined by the effects of hypoxemia and alkalemia on the peripheral chemoreceptors.  

CLINICAL MANIFESTATIONS

The clinical manifestations of respiratory alkalosis depend on the degree and duration of the condition but are primarily those of the underlying disorder. Chronic hypocapnia does not appear to be associated with any significant clinical symptoms. Symptoms of acute hypocapnia with Pco2 below 30 mm Hg are largely attributable to the alkalemia and include dizziness, perioral or extremity paresthesias, confusion, asterixis, hypotension, seizures, and coma. These can be related to decreased cerebral blood flow or reduced free calcium because alkalosis increases calcium’s protein-bound fraction. Shortness of breath and chest wall pain, which frequently accompany hyperventilation because of pain or anxiety, do not appear to be related to hypocapnia.  

Treatment of both chronic and acute respiratory acidosis aims primarily to correct the underlying cause and to ensure adequate ventilation. In acute respiratory acidosis, measures to relieve severe hypoxemia and acidemia should be instituted immediately, including intubation and assisted mechanical ventilation (Chapter 97) if necessary. Patients with myxedema coma require thyroid replacement (Chapter 213). In patients with compensated chronic respiratory acidosis, rapid and complete correction of hypercapnia can result in post-hypercapnic metabolic alkalosis. Patients who are chronically hypercapnic and hypoxemic should receive necessary oxygen even though their PCO2 will rise. The rise is not necessarily due to loss of a hypoxic drive to ventilation but may instead be because of release of CO2 from hemoglobin in the presence of oxygen or because oxygen-induced

743

DIAGNOSIS

Alveolar hyperventilation leading to respiratory alkalosis is seen with hypoxemia from pulmonary disease (Chapter 77), heart failure (Chapter 52), high altitudes (Chapter 88), or anemia. Mechanical ventilation (Chapter 97) is also a common cause of respiratory alkalosis. Another common cause of respiratory alkalosis is primary stimulation of the central chemoreceptor, as seen in sepsis (Chapter 100), hepatic cirrhosis (Chapter 144), salicylate intoxication (Chapters 76 and 102), correction of metabolic acidosis, hyperthermia (Chapter 101), and pregnancy, as well as cortical hyperventilation from anxiety and pain. In these situations, central signals override peripheral chemoreceptors until the primary stimulus is removed.

Primary neurologic diseases that can stimulate alveolar hyperventilation include acute stroke, infection, trauma, and tumors. Two patterns of respiration are seen: central hyperventilation and Cheyne-Stokes respiration (Chapter 80). Central hyperventilation, which is associated with lesions at the pontinemidbrain level, is regular, but with an increased rate and tidal volume. CheyneStokes breathing, which is characterized by periods of hyperventilation alternating with apnea, is seen in patients with bilateral cortical and upper pontine lesions and in patients with heart failure.

TREATMENT  Treatment of respiratory alkalosis must address the underlying cause of the disturbance. Hyperventilation syndrome is a diagnosis of exclusion, but patients who exhibit symptoms, such as tetany and syncope, and who do not have more serious causes of hyperventilation can be treated with a rebreathing mask. Hypophosphatemia can be seen in these patients, but it usually improves with treatment of the alkalosis. Patients with respiratory alkalosis associated with mountain sickness can be pretreated with acetazolamide to induce a metabolic acidosis, thereby preventing extreme elevations in pH (Chapter 88).

  Grade A References A1. Susantitaphong P, Sewaralthahab K, Balk EM, et al. Short- and long-term effects of alkali therapy in chronic kidney disease: a systematic review. Am J Nephrol. 2012;35:540-547. A2. Jaber S, Paugam C, Futier E, et al. Sodium bicarbonate therapy for patients with severe metabolic acidaemia in the intensive care unit (BICAR-ICU): a multicentre, open-label, randomised controlled, phase 3 trial. Lancet. 2018;392:31-40.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 110  Acid-Base Disorders  

GENERAL REFERENCES 1. Chen W, Abramowitz MK. Epidemiology of acid-base derangements in CKD. Adv Chronic Kidney Dis. 2017;24:280-288. 2. Ring T, Nielsen S. Whole body acid-base modeling revisited. Am J Physiol Renal Physiol. 2017;312:F647-F653. 3. Weiner ID, Verlander JW. Ammonia transporters and their role in acid-base balance. Physiol Rev. 2017;97:465-494. 4. Seifter JL, Chang HY. Disorders of acid-base balance: new perspectives. Kidney Dis (Basel). 2017;2:170-186. 5. Berend K. Diagnostic use of base excess in acid-base disorders. N Engl J Med. 2018;378: 1419-1428. 6. Liamis G, Filippatos TD, Liontos A, et al. Serum osmolal gap in clinical practice: usefulness and limitations. Postgrad Med. 2017;129:456-459. 7. Wesson DE, Pruszynski J, Cai W, et al. Acid retention with reduced glomerular filtration rate increases urine biomarkers of kidney and bone injury. Kidney Int. 2017;91:914-927.

744.e1

8. Nagami GT, Hamm LL. Regulation of acid-base balance in chronic kidney disease. Adv Chronic Kidney Dis. 2017;24:274-279. 9. Goraya N, Wesson DE. Clinical evidence that treatment of metabolic acidosis slows the progression of chronic kidney disease. Curr Opin Nephrol Hypertens. 2019. [Epub ahead of print.] 10. Goraya N, Wesson DE. Management of the metabolic acidosis of chronic kidney disease. Adv Chronic Kidney Dis. 2017;24:298-304. 11. Kamel KS, Halperin ML. Acid-base problems in diabetic ketoacidosis. N Engl J Med. 2015;372: 546-554. 12. Palazzo V, Provenzano A, Becherucci F, et al. The genetic and clinical spectrum of a large cohort of patients with distal renal tubular acidosis. Kidney Int. 2017;91:1243-1255. 13. Mohebbi N, Wagner CA. Pathophysiology, diagnosis and treatment of inherited distal renal tubular acidosis. J Nephrol. 2018;31:511-522. 14. Alexander RT, Bitzan M. Renal tubular acidosis. Pediatr Clin North Am. 2019;66:135-157.

744.e2

CHAPTER 110  Acid-Base Disorders  

REVIEW QUESTIONS 1. A young patient with type 1 diabetes mellitus and normal renal function presents with a febrile illness and a blood glucose level of 345 mg/dL. The arterial pH is 7.30 and the pCO2 is 30 mm Hg. Serum laboratory values include sodium 135 meq/L; potassium 5.5 meq/L; chloride 110 meq/L; bicarbonate 15 meq/L. The urine sodium is 50 meq/L; urine potassium 50 meq/L and urine chloride 20 meq/L. The urine pH is 5.0. Urine glucose 2+. Which of the following statements is most likely to be true? A . The patient has diabetic ketoacidosis. B. The patient has a proximal renal tubular acidosis. C. The patient has an excess of total body potassium. D. There is bicarbonaturia. E. The normal serum anion gap is an error. Answer: A  The patient with well-preserved glomerular filtration rate will spill large amounts of keto acids in the urine, with sodium and potassium giving a large urinary charge gap but a normal serum anion gap. Since the urinary pH is acid, the urine gap is not bicarbonaturia. Loss of keto acid anions and osmotic polyuria will likely lead to total body potassium deficits despite the shift of potassium from cells to extracellular water. In this case, ketoacidosis is the cause of a non-gap acidosis.

2. A 36-year-old woman returns from an overseas trip with gastroenteritis and is found to have a urinary Na+ of 34 mEq/L, a urinary K+ of 41 mEq/L, and a urinary Cl− of 6 mEq/L. Which one of the following statements is most likely to be correct? A . She has severe metabolic acidosis from the diarrhea. B. She has both metabolic acidosis from diarrhea and bicarbonaturia from an unsuspected renal tubular acidosis. C. She may have acidosis from diarrhea but must also have more significant metabolic alkalosis from vomiting. Her blood is alkalemic. D. She has not been vomiting but has metabolic alkalosis from congenital chloridorrhea. E. Her urinary NH4+ must be very high. Answer: C  She has chloride losses that are most likely gastric losses of HCl. The urine has very little chloride, consistent with hypochloremia. Congenital chloridorrhea is very unlikely to produce this syndrome in an otherwise healthy 36 year old. Diarrhea causing metabolic acidosis would be compensated by increased ammoniagenesis and high chloride excretion.

744

CHAPTER 111  Disorders of Magnesium and Phosphorus  

111  DISORDERS OF MAGNESIUM AND PHOSPHORUS

substantial depletion of total body magnesium can occur before serum magnesium levels drop appreciably. Magnesium deficiency may be due to nutritional deficiency, intestinal malabsorption, redistribution into bone, or losses via cutaneous, lower gastrointestinal, or renal routes (Table 111-1). The recommended daily allowance of magnesium is 420 mg for males and 320 mg for females. Approximately 25% of alcoholics are chronically hypomagnesemic because of a combination of poor nutritional intake and increased renal loss. Magnesium deficiency can occur, rarely, in protein-calorie malnutrition and may be associated with acute hypomagnesemia during refeeding because of rapid cellular magnesium uptake. Fat malabsorption in conditions such as celiac disease, Crohn disease, and small intestinal resection causes magnesium deficiency because free fatty acids accumulate in the intestinal lumen, where they combine with magnesium to form insoluble soaps. Proton pump inhibitors also can cause hypomagnesemia, primarily in patients concurrently using diuretics. This is thought to be due to inhibition of intestinal absorption. Lower gastrointestinal tract secretions are rich in magnesium, so diarrhea of colonic origin is a common cause of hypomagnesemia. Sweat contains significant amounts of magnesium, and transient hypomagnesemia can occur after prolonged, intense exercise such as marathon runs. Magnesium is also lost from burned skin surfaces, and 40% of patients with severe burns (Chapter 103) are hypomagnesemic. In patients with severe hyperparathyroidism (Chapter 232) and high bone turnover, continued sequestration of minerals within bone after parathyroidectomy causes transient hypocalcemia, hypomagnesemia, and hypophosphatemia. Renal magnesium losses can occur in the recovery phase of acute tubular necrosis or urinary tract obstruction. Hypomagnesemia is common in diabetes mellitus (Chapter 216), in which it is thought to be due to a combination of poor intestinal absorption, osmotic diuresis, and decreased renal tubule reabsorption. Inhibition of sodium reabsorption in the thick ascending limb of Henle by loop diuretics and in the distal convoluted tubule by thiazide diuretics inhibits tubular magnesium reabsorption and leads to urinary magnesium wasting. Tubular toxins that are common causes of renal magnesium wasting include cisplatin, carboplatin, amphotericin B, and aminoglycosides, which are often associated with hypokalemia and rarely with renal tubule acidosis, as well as calcineurin inhibitors such as cyclosporine and tacrolimus, which also cause hyperkalemia. Antibodies to the epidermal growth factor receptor, such as cetuximab and panitumumab, which are used to treat metastatic colorectal cancer, downregulate a distal tubule magnesium channel and cause isolated severe hypomagnesemia. Inherited hypomagnesemia is usually caused by renal magnesium loss and can be subdivided into three main types, depending on the coexistence of other electrolyte disturbances: Bartter and Gitelman syndromes, which are associated with renal salt wasting and hypokalemic metabolic alkalosis; familial

ALAN S.L. YU TABLE 111-1 CAUSES OF MAGNESIUM DEFICIENCY   MAGNESIUM METABOLISM

Magnesium is an important mineral component of the bony skeleton, a cofactor for many metabolic enzymes, and a regulator of ion channels and transporters in excitable tissues.1,2  

Normal Magnesium Metabolism

The majority of total body magnesium is intracellular or in bone, with only 1% in extracellular fluid. The normal serum magnesium concentration is 1.8 to 2.3 mg/dL (1.5 to 1.9 mEq/L). The average daily intake of magnesium is 300 mg, the main sources of which are green vegetables, nuts, whole grain cereals, milk, and seafood. Magnesium is absorbed mainly in the jejunum and ileum. In the kidney, 70 to 80% of serum magnesium is filtered at the glomerulus, with the majority being reabsorbed along the length of the tubule, particularly in the thick ascending limb of Henle. In states of magnesium deficiency or excess, renal tubule reabsorption is tightly regulated so that magnesium excretion is adjusted accordingly.

  MAGNESIUM DEFICIENCY PATHOBIOLOGY  

Magnesium deficiency is usually detected when hypomagnesemia becomes evident. However, because magnesium is stored primarily intracellularly,

Nutritional deficiency  Alcoholism*  Malnutrition   Refeeding syndrome Intestinal malabsorption* Proton pump inhibitors Lower gastrointestinal losses   Colonic diarrhea*   Intestinal fistula   Laxative abuse Cutaneous losses  Burns*   Exercise-induced sweating Redistribution into bone   Hungry bone syndrome Renal losses   Polyuria (including diabetes mellitus)*   Volume expansion  Hyperaldosteronism   Bartter and Gitelman syndromes  Hypercalcemia   Loop and thiazide diuretics*   Nephrotoxins (cisplatin, amphotericin, aminoglycosides, pentamidine, cyclosporine)* Epidermal growth factor monoclonal antibodies (cetuximab, panitumumab)* *Common causes.

CHAPTER 111  Disorders of Magnesium and Phosphorus  

ABSTRACT

Magnesium and phosphate are important components of bone, but they also play important roles within cells. Disorders of magnesium and phosphate homeostasis—especially hypomagnesemia, hypophosphatemia, and hyperphosphatemia in the setting of kidney disease—are common and most often asymptomatic. This chapter reviews the causes of magnesium and phosphate disorders, tests and algorithms to diagnose the underlying cause, and treatment options, including replacement regimens.

KEYWORDS

magnesium hypomagnesemia phosphate hypophosphatemia hyperphosphatemia

744.e3

CHAPTER 111  Disorders of Magnesium and Phosphorus  

hypomagnesemia with hypercalciuria and nephrocalcinosis; and isolated hypomagnesemia, which is usually associated with hypocalcemia.  

CLINICAL MANIFESTATIONS

Mild-to-moderate hypomagnesemia or magnesium deficiency is frequently asymptomatic. Manifestations of increased neuronal excitability are the most common symptoms, including paresthesias, tetany, and seizures. These may be associated with Chvostek sign (twitching of the cheek muscles in response to tapping the facial nerve in front of the ear) or Trousseau sign (carpal spasm induced by compressing the upper arm with a tourniquet or blood pressure cuff). Cardiac disturbances range from mild electrocardiographic abnormalities (nonspecific T wave changes, U waves, prolonged QT interval, and repolarization alternans) to ventricular tachycardia, torsades de pointes, and ventricular fibrillation (Chapter 59). Coexistent hypokalemia is very common for two reasons: many of the causes of hypomagnesemia are also causes of potassium loss, and hypomagnesemia itself causes renal potassium wasting. Severe hypomagnesemia also impairs parathyroid hormone secretion and induces tissue resistance to its actions, thereby leading to hypocalcemia.  

is enhanced, such as inflammatory bowel disease and intestinal obstruction. However, the kidney has a very large capacity to excrete excess magnesium. Thus, persistent hypermagnesemia is seen almost exclusively in patients who have chronic renal insufficiency (Chapter 121) who are also taking excessive amounts of magnesium in the form of antacids, cathartics, or enemas.  

TREATMENT  Mild hypermagnesemia in a patient with good renal function usually requires no treatment because renal clearance is rapid and the normal serum half-life of magnesium is approximately 1 day. In the event of serious toxicity, the effects of magnesium can be temporarily antagonized by the administration of intravenous calcium salts (5 to 10 mL of 10% calcium chloride). Renal magnesium excretion can be enhanced by administering furosemide (20 to 40 mg every 4 hours) together with a saline infusion (0.9% NaCl at 150 mL/hour, titrated to replace urinary losses). In patients with advanced renal insufficiency, the most effective method of magnesium removal is hemodialysis.

DIAGNOSIS

Urine magnesium × Serum creatinine 0.7 × Serum magnesium × Urine creatinine

With extrarenal magnesium loss (usually diarrhea, malabsorption or laxative abuse), the FEMg is appropriately suppressed ( 4 to 6 mg/dL) causes hypotension, nausea, vomiting, facial flushing, urinary retention, and ileus. Above serum levels of 8 to 12 mg/dL, flaccid skeletal muscle paralysis and hyporeflexia may ensue, along with bradyarrhythmias, respiratory depression, coma, and cardiac arrest. A low or even negative serum anion gap is sometimes seen.

The cause of the magnesium deficiency is often obvious from the history. In difficult diagnostic cases, a random urine sample should be collected and the fractional excretion of magnesium (FEMg) determined. FE Mg =

745



PROGNOSIS

Severe hypermagnesemia is potentially fatal. Lesser degrees of hypermagnesemia usually respond well to treatment.

  PHOSPHORUS METABOLISM

TREATMENT  It is unclear whether mild, asymptomatic hypomagnesemia needs to be treated.3 Magnesium repletion is recommended in hypomagnesemic patients if they are symptomatic, have underlying cardiac or seizure disorders, exhibit concurrent severe hypocalcemia or hypokalemia, or have severe hypomagnesemia (15%.

disease. After the underlying cause is identified and adequately treated, acid suppressive therapy can be withdrawn if there are no additional risk factors for ulcer disease, such as NSAID therapy or H. pylori infection. If the cause of the idiopathic ulceration is not clarified and there is doubt about the adequacy of the diagnostic testing for H. pylori, empirical eradication therapy can be considered, especially when there is histologic evidence of chronic active gastritis without further explanation. If related to unidentified H. pylori, gastritis should slowly disappear after successful eradication therapy.



PREVENTION

Primary Prevention

A test-and-treat strategy for H. pylori colonization is sometimes considered for patients with dyspeptic symptoms, but there is no specific way to prevent H. pylori–associated ulcer disease. By contrast, primary prevention of NSAIDassociated ulcer disease is widely advocated for patients at high risk because of a prior ulcer, severe concomitant disease, or use of warfarin or high-dose corticosteroids as well as for critically ill patients in intensive care units. H2 blockers (at a dose equivalent to ranitidine 300 mg twice daily; Table 129-1) partially prevent duodenal ulcer disease during NSAID therapy but have no effect on preventing gastric ulcers unless a higher dose (equivalent to famotidine 40 mg twice daily) is given. Proton pump inhibitors A8  (at a dose equivalent to omeprazole 20 mg once daily; Table 129-1) and misoprostol (in doses varying between 400 and 800 mg/day) partially protect against both gastric and duodenal ulcers during NSAID use. Misoprostol and proton pump inhibitors are equally effective, but adherence with therapy is lower with misoprostol owing to its side effects. Patients should be advised of the importance of adherence because less than 80% adherence to gastroprotection is associated with a more than two-fold increased risk for ulcer disease compared with those who are fully adherent. During low-dose aspirin therapy, primary prevention of ulcers is advocated for the same risk groups, using a proton pump inhibitor or an H2-receptor antagonist. A9  For the prevention of stress ulcers in patients in intensive care units, proton pump inhibitors are preferred. A10 

Secondary Prevention

Secondary prevention of H. pylori–associated ulcer disease is mandatory and consists of successful bacterial eradication. Testing to ascertain H. pylori status after eradication therapy is indicated in patients with prior complicated ulcer

disease or with persistent or recurrent symptoms after therapy, as well as in patients who fail to complete the therapeutic course. Secondary prevention of NSAID-related ulcer disease is preferentially achieved by the withdrawal of NSAIDs. In patients who must continue taking NSAIDs, a change to a selective COX2 inhibitor A11  in combination with a proton pump inhibitor at a dose equivalent to esomeprazole 20 mg twice daily is advocated, especially for patients with complicated ulcer disease. This combination is associated with a lower risk for secondary peptic ulcer complications than treatment with a COX2 inhibitor alone. Secondary prevention of recurrent ulcers in patients who use aspirin may depend on H. pylori status. In H. pylori–positive patients, H. pylori eradication is as effective as a proton pump inhibitor for the prevention of recurrent ulcers. H. pylori–negative patients require additional acid suppressive therapy at a dose equivalent to esomeprazole 20 mg twice daily. A12  Secondary prevention of idiopathic ulcer disease consists primarily of maintenance therapy with a proton pump inhibitor and treatment of the underlying condition. When there is doubt about the accuracy of the diagnostic assessment for H. pylori, an empirical course of eradication treatment can be considered.

Complications Hemorrhage

Hemorrhage (Chapter 126), which is the most common complication of peptic ulcer disease, occurs in about one in six patients with ulcers over the course of their ulcer activity. Ulcers caused by NSAIDs account for a larger proportion of these hemorrhages. Peptic ulcer is thus the most common cause of nonvariceal upper gastrointestinal bleeding, accounting for 40 to 60% of cases in most populations. Bleeding is associated with a 5 to 15% risk for rebleeding and up to a 10% risk for mortality. Hemorrhage may occur along a continuum from a serious acute event associated with hemodynamic shock and high mortality to slow or intermittent blood loss leading to chronic anemia. Approximately 80% of patients with bleeding ulcers describe a prior history of symptomatic disease, and about 20 to 30% have suffered a previous hemorrhage. Assessment of the magnitude of bleeding is of paramount importance in determining the need for transfusion and subsequent management (Table 130-5). Initial hematocrit levels may be misleading and are likely to fall because of hemodilution. Rapid bleeding is usually apparent on the basis of clinical signs (pallor, systolic blood pressure ≤100 mm Hg, pulse ≥100 beats/minute); immediate fluid resuscitation is indicated to prevent circulatory collapse. Transfusion is needed in patients with marked anemia. Restrictive transfusion

879

CHAPTER 130  Acid Peptic Disease  

TABLE 130-5 PARAMETERS OF THE ROCKALL, BLATCHFORD, AND AIMS65 SCORING SYSTEMS FOR UPPER GASTROINTESTINAL BLEEDING* SCORING CATEGORIES

SCORING PARAMETERS

ROCKALL SYSTEM†

BLATCHFORD SYSTEM‡

PARAMETER

SCORE

PARAMETER

SCORE

AIMS65§ PARAMETER

SCORE

Age

Age (yr)

60-79 ≥80

1 2

N/A



>65

1

Clinical status

Systolic blood pressure

100 N/A N/A N/A Any major comorbidity Renal or liver failure, or disseminated malignancy

1 — — — 2 3

1 2 3 1 1 1 — 2 2

≤90

Pulse Melena Syncope Altered mental status Comorbidities

100-109 90-99 100 Present Present N/A Hepatic disease Cardiac failure

Blood urea, mmol/L

N/A



Hemoglobin, g/dL

N/A



INR Albumin

N/A N/A

— —

6.5-7.9 8.0-9.9 10-24.9 >25.0 Men: 12.0-13.0 Women: 10.0-12.0 Men: 10.0-12.0 Men and women: 1.5 4 weeks)

History/Physical Examination

Specific Clue

Functional/IBS-Diarrhea

• Stop medication/diet agent when found • Proceed to specific test for diagnosis

• CBC, ESR, CMP • Consider celiac antibody test • Treat

High Volume, Dehydration, Elderly

Alarm Symptoms, No Clue Initial Blood Test:

CBC (differential), electrolytes, liver tests, protein/albumin, calcium, folate, vitamin B12, consider celiac antibodies

Stool Studies: Failed therapy

• IV hydration • Hospitalization

O&P, Giardia Ag or PCR, WBC or calprotectin, fat, electrolytes/osmolality (if watery) Consider stool culture or PCR panel, laxative screen* Consider stool α-1 antitrypsin if hypoproteinemic

Endoscopic Studies:

Colonoscopy with biopsies EGD with duodenal biopsies

No Diagnosis (Elusive) Watery (see Fig. 131-3)

Malabsorptive (see Fig. 131-3)

(Common causes: bile salt or carbohydrate malabsorption, fecal incontinence, drug/laxatives, microscopic colitis) • Repeat history/physical examination • Review studies done • Further testing when cause

FIGURE 131-2.  Initial approach to chronic diarrhea. *Perform stool culture in those who are immunosuppressed; perform laxative screen if laxative abuse is suspected. Ag = antigen; CBC = complete blood count; CMP = comprehensive metabolic panel; EGD = esophagogastroduodenoscopy; ESR = erythrocyte sedimentation rate; IV = intravenous; O&P = ova and parasites; PCR = polymerase chain reaction; WBC = white blood cells.

Endoscopy and Biopsy

Upper endoscopy with distal duodenal biopsies should be undertaken if serologic tests for celiac disease are positive or diagnostic clues suggest small bowel mucosal malabsorption (Chapter 125). Small bowel biopsy is virtually always abnormal when the tTG immunoglobulin A (IgA) antibody level is very high (more than ten-fold the normal range), and antiendomysial antibody (EMA) is positive. Some patients with celiac disease may have patchy mucosal disease and require enteroscopy with jejunal biopsies for diagnosis. Wireless video capsule endoscopy (Chapter 125) and balloon-assisted enteroscopy are used to diagnose diseases that reside deep in the small bowel. Patients with severe watery or elusive diarrhea should have a colonoscopy to assess for villous adenomas, microscopic colitis, mastocytosis, or early inflammatory bowel disease. Colonoscopy also may show brown pigmentation suggestive of melanosis coli due to chronic use of anthracene laxatives. Terminal ileal ulcers may indicate infectious or inflammatory bowel disease.

Other Laboratory Tests Malabsorption

If chronic diarrhea is the presenting symptom, a stool examination for ova and parasites and a PCR test for Giardia or stool antigen-capture enzymelinked immunosorbent assay (ELISA) should be obtained. A stool test for fat on a high-fat diet (70 to 100 g/day) is the best available screening test for malabsorption (Table 131-6). If the fecal fat test result is negative, selective carbohydrate malabsorption or other causes of watery diarrhea should be considered. If the fecal fat test result is positive, further testing should be based on clinical suspicion for particular diseases. If pancreatic insufficiency is suspected, imaging studies of the pancreas should be performed. If bacterial overgrowth is suspected, culture of an intestinal aspirate or a breath test should be obtained. Small bowel contrast imaging is useful in detecting structural abnormalities that predispose to bacterial overgrowth (Table 131-7). If proximal mucosal damage is suspected, multiple duodenal biopsies should

be obtained. If there are no clues, CT or MR enterography may help to detect middle and distal small bowel mucosal diseases. Some individuals with celiac disease present with selective nutrient deficiencies without diarrhea. In these cases, tTG antibody tests and intestinal biopsy should be performed. In patients hospitalized for severe diarrhea or malnutrition, a more streamlined evaluation usually includes a stool for culture, ova and parasites, and fat; an abdominal imaging study; and a biopsy of the small intestine and colon.

Watery Diarrhea

Breath tests to measure the respiratory excretion of H2 and methane after administration of carbohydrates can assess carbohydrate malabsorption or bacterial overgrowth (see Table 131-6). The diagnosis of endocrine tumors, such as carcinoids, gastrinoma, VIPoma, medullary carcinoma of the thyroid, glucagonoma, somatostatinoma, and systemic mastocytosis, is made by showing elevated blood levels of serotonin, chromogranin A, or urinary 5-hydroxyindoleacetic acid and serum levels for gastrin, vasoactive intestinal peptide, calcitonin, glucagon, somatostatin, histamine, or prostaglandins (Chapter 219). Somatostatin receptor scintigraphy has proved to be sensitive and useful in the diagnosis and evaluation of carcinoid tumors and gastrinomas (Chapter 219).

Inflammatory Diarrhea

Stool occult blood, white blood cells, and calprotectin are helpful tests for bowel inflammation. Video capsule endoscopy of the small bowel may detect ulcerations deep in the small bowel not reachable by standard upper or lower endoscopy and not detected with conventional barium contrast radiography. However, the risk for capsule retention in the small bowel is high in patients with Crohn disease or NSAID use, particularly when there is a history of obstructive symptoms. The most sensitive test for protein-losing enteropathy is measurement of intestinal protein loss by 24-hour stool excretion or clearance of α1-antitrypsin.

888

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  

Chronic Watery Diarrhea

Noninflammatory If high volume or elusive:

Inflammatory

Bloody

Bloody or Watery Stool hemoccult positive Lactoferrin, calprotectin, WBCs

Watery

Ulcerative colitis Chronic ischemia

Crohn disease Ischemic enteritis Food allergy Eosinophilic gastroenteritis§ Radiation

Celiac disease Tropical sprue Microscopic colitis (requires colon biopsy for diagnosis)

>1 L (often >3 L) (persists with fast)

Tumor Hormone* VIPoma Carcinoid Gastrinoma Thyroid medullary cancer

Laxatives† women > men +/– osmotic gap Mg, PO4

3 L output daily “pancreatic cholera,” elevated VIP level. Carcinoid: elevated urine 5-hydroxyindole acetic acid, positive OctreoScan. Gastrinoma: elevated gastrin level, positive secretin stimulation test, diarrhea due to high volume of acid secretion. Thyroid medullary cancer: elevated calcitonin level. †Stool volume may be high or low volume depending on dose ingested, may respond to fast. ‡Carbohydrate malabsorption (CHO) may be due to lactase deficiency, dietary fructose, sorbitol in diabetic candies or liquid medications. §Full-thickness biopsy may be needed for diagnosis. FGF = fibroblast growth factor; IBS = irritable bowel syndrome; WBCs = white blood cells.

TABLE 131-6 TESTS FOR THE EVALUATION OF MALABSORPTION* TEST

COMMENTS

GENERAL TESTS OF ABSORPTION Quantitative stool fat test

Gold standard test of fat malabsorption. Requires ingestion of a high-fat diet (70-100 g) for 2 days before and during the collection. Stool is collected for 2-3 days. Normally, 20 ppm of exhaled H2 after lactose ingestion suggests lactose malabsorption. Absorption of other carbohydrates (e.g., sucrose, glucose, fructose) also can be tested.

SPECIFIC TESTS FOR MALABSORPTION Tests for Pancreatic Function Secretin stimulation test

The gold standard test of pancreatic function. Requires duodenal intubation with a double-lumen tube and collection of pancreatic juice in response to intravenous secretin. Allows measurement of bicarbonate (HCO3−) and pancreatic enzymes. A sensitive test of pancreatic function, but labor intensive and invasive.

Fecal elastase-1 test

Stool test for pancreatic function. Equal sensitivity to the secretin stimulation test for the diagnosis of moderate to severe pancreatic insufficiency. Unreliable with mild insufficiency. False-positive results occur with increased stool volume and intestinal mucosal diseases.

Tests for Bacterial Overgrowth Quantitative culture of small intestinal aspirate

Gold standard test for bacterial overgrowth. Greater than 105 colony-forming units (CFU)/mL in the jejunum suggests bacterial overgrowth. Requires special anaerobic sample collection, rapid anaerobic and aerobic plating, and care to avoid oropharyngeal contamination. False-negative results occur with focal jejunal diverticula and when overgrowth is distal to the site aspirated.

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  

889

TABLE 131-6  TESTS FOR THE EVALUATION OF MALABSORPTION—cont’d TEST Hydrogen breath test

COMMENTS The 50-g glucose breath test has a sensitivity of 90% for growth of 105 colonic-type bacteria in the small intestine. If bacterial overgrowth is present, increased H2 is excreted in the breath. An early hydrogen level (20 ppm suggests bacterial overgrowth. Specificity is increased when performed with simultaneous nuclear scintigraphy to define transit time to the cecum. False-positive results occur in patients with rapid transit. False-negative results occur with non-hydrogen-producing organisms. Concomitant measurement of breath methane improves test sensitivity.

Tests for Mucosal Disease Small bowel biopsy

Obtained for a specific diagnosis when there is a high index of suspicion for small intestinal disease. Several biopsy specimens (4-5) must be obtained to maximize the diagnostic yield. Distal duodenal biopsy specimens are usually adequate for diagnosis, but occasionally enteroscopy with jejunal biopsy specimens is necessary. Small intestinal biopsy provides a specific diagnosis in some diseases (e.g., intestinal infection, Whipple disease, abetalipoproteinemia, agammaglobulinemia, lymphangiectasia, lymphoma, amyloidosis). In other conditions, such as celiac disease and tropical sprue, the biopsy specimens show characteristic findings, but the diagnosis is made by positive celiac serology or improvement after specific treatment.

Test of Ileal Function 75

SeHCAT test

This is a test of bile acid absorption. Seven days after ingestion of radiolabeled synthetic selenium–homocholic acid conjugated with taurine (75SeHCAT), whole body retention is measured by a gamma-counting device. The result is expressed as a fraction of baseline ingestion. Retention values of 50 mmol/L). In osmotic diarrhea, the presence of uncharged solute or unmeasured cation in the colonic lumen draws in water, depresses stool Na+ (usually 90 mmol/L) and there is no osmotic gap; this diarrhea may be diagnosed by determining stool Cl− concentration because these anions displace stool Cl− , resulting in a depressed stool Cl− value (usually 90 mmol/L, K+ concentrations usually 2 weeks) may be due to persistent or recurrent infections. These diarrheas occur most commonly in children exposed to unsafe drinking water in developing countries, patients who have acquired immunodeficiency syndrome (AIDS) or are immunosuppressed for other reasons, and recent travelers. Recurrent or prolonged infectious diarrhea may lead to severe malnutrition and death (mortality rate, 50%). Treatment for children includes nutrition support with supplemental vitamin A (200,000 IU orally twice yearly) and zinc (20 mg elemental orally daily for 14 days). Severe disease may require total parenteral nutrition. In patients with AIDS, protracted diarrhea may be caused by treatable agents such as E. histolytica, Giardia, or Strongyloides or by organisms such as Cryptosporidium, Isospora belli, and microsporidia that are difficult to treat or untreatable. The most effective treatment is retroviral therapy to improve the immune system (Chapter 364). In travelers returning from developing countries with infectious diarrhea that persists for longer than 3 to 4 weeks, stool should be examined for culture and for ova and parasites. In patients with a recent history of antibiotic use, stool also should be sent for C. difficile toxin or PCR. Any specific organisms that are identified should be treated. If treatment with trimethoprimsulfamethoxazole or a fluoroquinolone has been unsuccessful, tetracycline (250 mg orally four times daily for 7 to 10 days), doxycycline (100 mg orally two times a day for 7 to 10 days), or metronidazole (250 mg orally three times daily for 7 to 10 days) can be tried. After documented infectious diarrhea, 25% of patients experience pain, bloating, urgency, a sense of incomplete evacuation, and loose stools for 6 months or longer; some of these patients have celiac disease, so screening (see later) is warranted in this setting. When no other cause is found, these patients are deemed to have postinfectious irritable bowel syndrome.

890

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  

Sporadic outbreaks of severe, prolonged diarrhea, often greater than 1 year in duration, occasionally have been reported. This form of prolonged diarrhea is called Brainerd diarrhea. The organism has yet to be identified. The diarrhea is difficult to treat; cholestyramine (4 g orally three times daily) may be helpful.  

Malabsorptive Syndromes

Malabsorption is caused by many different diseases, drugs (e.g., the lipase inhibitor orlistat; Chapter 207), and nutritional products (the nonabsorbable fat olestra) that impair intraluminal digestion, mucosal absorption, or delivery of the nutrient to the systemic circulation (E-Fig. 131-2; see Table 131-4). Dietary fat is the nutrient most difficult to absorb. Fatty stools (steatorrhea) are the hallmark of malabsorption; a stool test for fat is the best screening test. Malabsorption does not always cause diarrhea. Clinical signs of vitamin or mineral deficiencies may occur in the absence of diarrhea. A careful focused history is crucial in guiding further testing to confirm the suspicion of malabsorption and to make a specific diagnosis (Fig. 131-4). The goals of treatment are to correct or treat the underlying disease and replenish losses of water, electrolytes, and nutrients.  

Conditions That Impair Intraluminal Digestion

Most digestion and absorption of nutrients occur in the small intestine (see E-Fig. 131-2). Carbohydrates and most dietary proteins are water soluble and readily digested by pancreatic enzymes. Pancreatic proteases (trypsinogen, chymotrypsinogen, procarboxypeptidases) are secreted from acinar cells in inactive forms. The cleavage of trypsinogen to trypsin by the duodenal brushborder peptidase enteropeptidase (enterokinase) allows trypsin to cleave the remaining trypsinogen and other proteases to their active form. Most dietary lipids (long-chain triglycerides, cholesterol, and fat-soluble vitamins) are water insoluble and must undergo lipolysis and incorporation into mixed micelles before they can be absorbed across the intestinal mucosa. Pancreatic lipase, in the presence of its cofactor, colipase, cleaves long-chain triglycerides into fatty acids and monoglycerides. The products of lipolysis interact with bile salts and phospholipids to form mixed micelles, which also incorporate cholesterol and fat-soluble vitamins (D, A, K, and E) in their hydrophobic centers. Bicarbonate secreted from pancreatic duct cells is physiologically important to neutralize gastric acid because pancreatic enzyme activity and bile salt micelle formation are optimum at a luminal pH of 6 to 8.



IMPAIRED MIXING

Surgical alterations, such as partial gastrectomy with gastrojejunostomy (Billroth II anastomosis) or gastrointestinal bypass surgeries for obesity, result in the release of biliary and pancreatic secretions into the intestine at a site remote from the site of entry of gastric contents. This imbalance can result in impaired lipolysis and impaired micelle formation, with subsequent fat malabsorption. Bypass of the duodenum also impairs absorption of iron, folate, and calcium. Rapid transit through the jejunum contributes to the malabsorption of nutrients. Individuals with these conditions also have surgical anastomoses that predispose to bacterial overgrowth.  

DUMPING SYNDROME

After esophageal (distal esophagectomy, myotomy for achalasia), gastric (Nissan wrap, hiatal hernia repair, gastrojejunostomy), and bariatric (Rouxen-Y and duodenal switch gastric bypass) surgeries, the unregulated delivery of concentrated sugars and food into the duodenum and jejunum results in altered insulin regulation, maldigestion, osmotic movement of fluid into the intestinal lumen, and rapid transit such that intestinal contact time is insufficient for absorption of nutrients. Patients may present with severe diarrhea, malabsorption, abdominal cramping, gas, and weight loss. Some patients have associated sweatiness, dizziness, and altered cognition because of postprandial hypoglycemia. A modified oral glucose tolerance test that shows late (120 to 180 minutes) hypoglycemia and an early (30 minutes) rise in hematocrit with an increased pulse rate suggests the dumping syndrome in patients with consistent symptoms. A small bowel barium study to assess transit time may be helpful in the diagnosis. Treatment is with a diet that is low in concentrated sugars divided into six small meals. Administration of pectin (15 g with each meal) may slow gastric emptying. In patients who are refractory to dietary measures, a short-acting somatostatin analogue (e.g., octreotide, 25 to 200 µg SC three times daily) or the better tolerated intramuscular preparation (10 to 20 mg monthly) improves dumping symptoms. In patients with predominant reactive hypoglycemia 1 to 3 hours after a meal (late dumping), an α-glycosidase hydrolase inhibitor (e.g., acarbose, 50 to 100 mg orally three times daily) that blocks carbohydrate absorption in the small bowel may be beneficial. Continuous tube feeding is also effective.

Unexplained iron or folate deficiency Unexplained bone disease

Suspected malabsorption

Stool for ova and parasites Stool for Giardia antigen test Positive

Treat infection

Negative

Stool fat test

Normal

Carbohydrate malabsorption Evaluation of watery and inflammatory diarrhea (Fig. -3)

Abnormal

Suspicion of pancreatic disease (alcoholic, pancreatic surgery), fecal fat > 30 g/day

Proximal

Suspicion of rapid transit (altered gastric, small bowel, esophageal anatomy)

Suspicion of (elderly, DM, small bowel stricture or dysmotility) (Table 1 -7)

Suspicion of mucosal disease (weight loss, multiple nutrient malabsorption) Distal

Image pancreas Secretin test

Small bowel for transit time

Culture intestinal aspirate H2 breath test Small bowel

CT or MR enterography

FIGURE 131-4.  Approach to the diagnosis of malabsorption. CT = computed tomography; DM = diabetes mellitus; MR = magnetic resonance; PCR = polymerase chain reaction.

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  

Intraluminal Digestion

Lipase Colipase pH 6-7

MG + FA

MG, FA,Chol, vitamin D, A, K, E

Starch

Proteases

Amylase

ApoB, A MG + FA

Peptides a.a. Di-, tripeptides Oligosaccharides Disaccharides

TG, Chol, vitamins D, A, K, E

TG

Apo proteins Mixed micelle

Protein

Removal

Bile salts

Pancreatic enzymes

TG

Mucosal Uptake

890.e1

Chylomicron

Lymphatics

Phospholipid a.a. Di-, tripeptides

Capillaries

Sugar monomers

Disaccharidases, peptidases E-FIGURE 131-2.  Phases of intestinal digestion and absorption of dietary fat, protein, and carbohydrate. a.a. = Amino acids; ApoB, A = apolipoproteins B and A; Chol = cholesterol; FA = fatty acids; MG = monoglycerides; TG = triglycerides.

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  



IMPAIRED LIPOLYSIS

A deficiency in pancreatic lipase may be caused by the congenital absence of pancreatic lipase or by destruction of the pancreatic gland as a result of alcoholrelated pancreatitis, cystic fibrosis, or pancreatic cancer. Pancreatic lipase also can be denatured by excess secretion of gastric acid (e.g., gastrinoma; Chapter 219). In such cases, lipase denaturation can be offset by treatment with a high-dose proton pump inhibitor (e.g., omeprazole 60 mg/day orally) to block acid secretion.  

CHRONIC PANCREATITIS

Chronic pancreatitis (Chapter 135) is the most common cause of pancreatic insufficiency and impaired lipolysis. In the United States, chronic pancreatitis most commonly results from alcohol abuse; in contrast, tropical (nutritional) pancreatitis is most common worldwide. Malabsorption of fat does not occur until more than 90% of the pancreas is destroyed.  

CLINICAL MANIFESTATIONS

Individuals with pancreatic causes of malabsorption typically present with bulky, fat-laden stools (usually >30 g of fat daily), abdominal pain, and diabetes, although some present with diabetes in the absence of gastrointestinal symptoms. Stools usually are not watery because undigested triglycerides form large emulsion droplets with little osmotic force and, in contrast to fatty acids, do not stimulate water and electrolyte secretion in the colon. Deficiency of fat-soluble vitamins is seen only rarely, presumably because gastric and residual pancreatic lipase generates enough fatty acids for some micelle formation. In severe disease, subclinical protein malabsorption, manifested by the presence of undigested meat fibers in the stool, and subclinical carbohydrate malabsorption, manifested by gas-filled, floating stools, can occur. Weight loss, when it occurs, is most often caused by decreased oral intake to avoid abdominal pain or diarrhea and less commonly by malabsorption. Pancreatic enzyme replacement and analgesics are the mainstays of treatment for chronic pancreatitis (see Table 135-5 in Chapter 135).  

DIAGNOSIS

Between 30 and 40% of individuals with alcohol-related chronic pancreatitis have calcifications on abdominal radiographs. A qualitative or quantitative test for fecal fat is positive in individuals whose pancreas is more than 90% destroyed. Noninvasive tests of pancreatic function are not sensitive enough to detect mild to moderate insufficiency, so the secretin stimulation test is preferred (see Table 131-6) if it can be obtained.  

bacteria in the upper small bowel low. Any condition that produces local stasis or recirculation of colonic luminal contents allows development of a predominantly “colonic” flora (coliforms and anaerobes, such as Bacteroides and Clostridium) in the small intestine (see Table 131-7). Anaerobic bacteria cause impaired micelle formation by releasing cholylamidases, which deconjugate bile salts. The unconjugated bile salts, with their higher pKa, are more likely to be in the protonated form at the normal upper small intestinal pH of 6 to 7 and can be absorbed passively. As a result, the concentration of bile salts decreases in the intestinal lumen and can fall to less than the critical micellar concentration, causing malabsorption of fats and fat-soluble vitamins. Vitamin B12 deficiency and carbohydrate malabsorption also can occur with generalized bacterial overgrowth. Anaerobic bacteria ingest vitamin B12 and release proteases that degrade brush-border disaccharidases. Although anaerobic bacteria use vitamin B12, they synthesize folate. Individuals with bacterial overgrowth usually have low serum vitamin B12 levels but normal or high folate levels; this helps distinguish bacterial overgrowth from tropical sprue, in which vitamin B12 and folate levels are usually low because of decreased mucosal uptake.  





TREATMENT  The goals of treatment are to correct the structural or motility defect, if possible; to eradicate offending bacteria; and to provide nutritional support. Acid-reducing agents should be stopped, if possible. Treatment with antibiotics should be based on culture results whenever possible; otherwise, empirical treatment is given. Rifaximin (400 mg orally three times daily) is effective, A8  but less so in individuals with an excluded (blind) intestinal loop. Tetracycline (250 to 500 mg orally four times daily), doxycycline (100 mg orally two times daily), or a broad-spectrum antibiotic against aerobes and enteric anaerobes (ciprofloxacin, 500 mg orally twice daily; amoxicillin-clavulanic acid, 250 to 500 mg orally three times daily; cephalexin, 250 mg orally four times daily with metronidazole, 250 mg orally three times daily) should be given for 14 days. Prokinetic agents such as metoclopramide (10 mg orally four times daily) or erythromycin (250 to 500 mg orally four times daily) can be tried to treat small bowel motility disorders, but often they are not efficacious. Octreotide (50 µg SC every day) may improve motility and reduce bacterial overgrowth in individuals with scleroderma. If the structural abnormality or motility disturbance cannot be corrected, the patient is at risk for malnutrition and deficiencies of vitamin B12 and fat-soluble vitamins. Cyclic treatment (1 to 3 weeks of every 4 to 6 weeks) with rotating antibiotics may be required in these patients to prevent recurrent bouts of bacterial overgrowth. If supplemental calories are needed, medium-chain triglycerides should be given because they do not depend on micelle formation for their absorption. Monthly treatment with vitamin B12 should be considered, along with supplemental vitamins D, A, K, and E and calcium.

Decreased Bile Salt Synthesis and Delivery

Intestinal Bacterial Overgrowth

In health, only small numbers of lactobacilli, enterococci, gram-positive aerobes, or facultative anaerobes can be cultured from the upper small bowel lumen. Motility and acid are the most important factors in keeping the number of

DIAGNOSIS

The diagnosis of bacterial overgrowth should be considered in elderly people and in individuals with predisposing underlying disorders (see Table 131-7).10 Bacterial overgrowth may be associated with the irritable bowel syndrome. The identification of greater than 105 CFU/mL in a culture of small intestinal aspirate is the gold standard in diagnosis but is not readily available. The noninvasive tests with a sensitivity and specificity comparable to those of intestinal culture are the glucose hydrogen and methane breath test; in individuals with low vitamin B12 levels, a Schilling test before and after antibiotic therapy can be diagnostic if available (Chapter 155).

IMPAIRED MICELLE FORMATION

Malabsorption can occur in individuals with cholestatic liver disease or bile duct obstruction, as well as in occasional patients who have undergone prior cholecystectomy.9 The clinical consequences of malabsorption are seen most often in women with primary biliary cirrhosis because of the prolonged nature of the illness. Although these individuals can present with steatorrhea, osteoporosis or, less commonly, osteomalacia is the most common presentation. The cause of bone disease in these patients is poorly understood and often is not related to vitamin D deficiency. Bone disease is treated with calcium supplements (and vitamin D if a deficiency is documented), weight-bearing exercise, and a bisphosphonate (e.g., alendronate, 10 mg/day orally or 70 mg orally once weekly).

CLINICAL MANIFESTATIONS

Individuals with bacterial overgrowth can present with diarrhea, abdominal cramps, gas and bloating, anorexia, weight loss, and signs and symptoms of vitamin B12 and fat-soluble vitamin deficiency. Watery diarrhea occurs because of the osmotic load of unabsorbed carbohydrates and stimulation of colonic secretion by unabsorbed fatty acids.

Bile salt concentrations in the intestinal lumen can fall to less than the critical concentration (2 to 3 mmol/L) needed for micelle formation because of decreased bile salt synthesis (severe liver disease), decreased bile salt delivery (cholestasis), or removal of luminal bile salts (bacterial overgrowth, terminal ileal disease or resection, cholestyramine therapy, acid hypersecretion). Fat malabsorption resulting from impaired micelle formation is generally not as severe as malabsorption resulting from pancreatic lipase deficiency, presumably because fatty acids and monoglycerides can form lamellar structures, which to a certain extent can be absorbed. Malabsorption of fat-soluble vitamins (D, A, K, and E) may be marked, however, because micelle formation is required for their absorption.  

891



ILEAL DISEASE

Disease of the terminal ileum is most commonly due to Crohn disease, which also may lead to ileal resection, but it also can be caused by radiation enteritis, tropical sprue, tuberculosis, Yersinia infection, or idiopathic bile salt malabsorption. These diseases cause bile salt wasting in the colon. The clinical consequences of bile salt malabsorption are related directly to the length of the diseased or resected terminal ileum. In an adult, if less than

892

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  

100 cm of ileum is diseased or resected, watery diarrhea results because of stimulation of colonic fluid secretion by unabsorbed bile salts. Bile acid diarrhea responds to cholestyramine (2 to 4 g taken at breakfast, lunch, and dinner). A9  If more than 100 cm of ileum is diseased or resected, bile salt losses (>3 g/day) in the colon exceed the capacity for increased bile salt synthesis in the liver, the bile salt pool shrinks, and micelle formation is impaired. As a result, steatorrhea ensues, and fatty acid–induced intestinal secretion synergizes with the bile acid–induced secretion to cause diarrhea. Treatment is with a low-fat diet, vitamin B12 (300 to 1000 µg SC once every month or 2 mg/day orally), dietary supplements of calcium (500 mg orally two or three times daily, monitor 24-hour urine calcium for adequacy of dose), and a multiple vitamin and mineral supplement. An antimotility agent should be given for diarrhea. Bile salt binders may worsen diarrhea. Screening for fat-soluble vitamin deficiencies (vitamins A and E, 25-OH vitamin D, and prothrombin time) and bone disease (bone densitometry, serum calcium, intact parathyroid hormone, 24-hour urine for calcium) should be done. Three long-term complications of chronic bile salt wasting and fat malabsorption are renal stones, bone disease (osteoporosis and osteomalacia), and gallstones. Oxalate renal stones occur as a consequence of excess free oxalate absorption in the colon. Free oxalate is generated when unabsorbed fatty acids bind luminal calcium, which is then unavailable for binding oxalate. Renal oxalate stones sometimes can be avoided with a low-fat, low-oxalate diet and calcium supplements. Bone disease is caused by impaired micelle formation with a resulting decrease in absorption of vitamin D; year-round sun exposure reduces this complication. Vitamin D (50,000 U orally one to three times per week) and calcium supplements (500 mg orally two to three times daily day) should be given to susceptible individuals, but vitamin D levels and serum and urinary calcium must be monitored for response to treatment because excess vitamin D can be toxic. The mechanism of gallstone formation in these individuals is unclear; pigmented gallstones are most common.  

Conditions That Impair Mucosal Absorption  

PATHOBIOLOGY

Nutrients are absorbed along the entire length of the small intestine, with the exception of iron and folate, which are absorbed in the duodenum and proximal jejunum, and bile salts and cobalamin (vitamin B12), which are absorbed in the distal ileum. The efficiency of nutrient uptake at the mucosa is influenced by the number of villus absorptive cells, the presence of functional hydrolases and specific nutrient transport proteins on the brush-border membrane, and transit time. Transit time determines the contact time of luminal contents with the brush-border membrane and influences the efficiency of nutrient uptake across the mucosa. Mucosal malabsorption can be caused by specific (usually congenital) brush-border enzyme or nutrient transporter deficiencies or by generalized diseases that damage the small intestinal mucosa or result in surgical resection or bypass of the small intestine. The nutrients malabsorbed in these general malabsorptive diseases depend on the site of intestinal injury (proximal, distal, or diffuse) and the severity of damage. The main mechanism of malabsorption in these conditions is a decrease in surface area available for absorption. Some conditions (infection, celiac disease, tropical sprue, food allergies, and GVHD) are characterized by intestinal inflammation and villus flattening; others are characterized by ulceration (ulcerative jejunitis, NSAIDs, Crohn disease), infiltration (amyloidosis), or ischemia (radiation enteritis, mesenteric ischemia). Long-chain fatty acids are transported across the microvillus membrane of villus epithelial cells by the fatty acid transport protein FATP4. The bile salts from mixed micelles remain in the intestinal lumen and are absorbed in the distal ileum by sodium-dependent cotransport. Oligosaccharides and larger oligopeptides (products of pancreatic enzyme digestion), sucrose, and lactose are hydrolyzed further by enzymes present in the brush-border membrane of villus epithelial cells before they are absorbed. Although only sugar monomers (glucose, galactose, fructose) can be taken up at the apical epithelial cell membrane, dipeptides and tripeptides are readily taken into the cell. Water-soluble vitamins are readily absorbed throughout the small intestine. Fat-soluble vitamins, minerals, and cobalamin are more difficult to absorb because of the requirement for micelle formation (vitamins D, A, K, and E), a divalent charge (magnesium, calcium, iron), or selected sites of uptake in the intestine (iron, cobalamin). Calcium is absorbed best in the proximal small intestine by a vitamin D–dependent calcium channel (TRPV6). Magnesium is absorbed in the small intestine by members of the transient receptor potential family (TRPM6 and TRPM7). Mutations in TRPM6 have been

identified in the rare disorder hereditary hypomagnesemia. Ferrous iron is transported into intestinal epithelial cells by a proton-coupled metal-ion transporter (Nramp2) that has specificity for Fe2+ and other divalent cations (Zn2+, Mn2+, Co2+, Cd2+, Cu2+, Ni2+, and Pb2+). The absorption of calcium and nonheme iron is enhanced by solubilization with hydrochloric acid. Intraluminal compounds such as oxalate, phytate, and long-chain fatty acids bind to calcium and magnesium, decreasing their absorption. Individuals with severe mucosal disease or short-bowel syndrome with high fecal fluid outputs lose magnesium and zinc from endogenous secretions. Folates (Chapters 155 and 205) are both taken in the diet and produced by bacteria in the colon. Dietary folates are absorbed in the proximal small intestine through a reduced folate carrier (RFC1). Deficiency can be caused by poor intake or malabsorption secondary to intestinal disease or drugs. The cobalamins (Chapters 155 and 205) are abundant in foods containing animal proteins (e.g., meat, seafood, eggs, milk). Cobalamin (vitamin B12) deficiency in industrialized countries is rarely due to poor dietary intake but rather reflects the inability to absorb cobalamin. This inability may be caused by a lack of intrinsic factor, consumption of cobalamin by overgrowth of anaerobic bacteria in the small bowel lumen, ileal disease or resection, or defective transcobalamin II. Large amounts of cobalamin are present in the liver (2 to 5 mg), and cobalamin is reabsorbed from bile through the enterohepatic circulation, thereby limiting daily losses to less than 1 µg. It usually takes 10 to 12 years for cobalamin deficiency to develop after it is eliminated from the diet, but deficiency can occur more rapidly (2 to 5 years) with malabsorptive syndromes. If lack of gastric acid causes food-cobalamin malabsorption, treatment with oral cyanocobalamin supplementation (Chapter 155) is curative.  

LACTASE DEFICIENCY EPIDEMIOLOGY  

Acquired lactase deficiency is the most common cause of selective carbohydrate malabsorption. Most individuals, except those of northern European descent, begin to lose lactase activity by the age of 2 years. The prevalence of lactase deficiency is highest (85 to 100%) in persons of Asian, African, and Native American descent.  

PATHOBIOLOGY

The persistence or nonpersistence of lactase activity is associated with a single nucleotide polymorphism C/T−13910 that is found upstream of the lactase gene on chromosome 2q21-22. Hypolactasia is associated with the C/C−13910 genotype in diverse ethnic groups. The mechanism by which this variant downregulates the lactase gene is not known, but functional studies suggest genotype-dependent alterations in levels of messenger RNA.  

CLINICAL MANIFESTATIONS

Adults with lactase deficiency typically complain of gas, bloating, and diarrhea after the ingestion of milk or dairy products but do not lose weight. Unabsorbed lactose is osmotically active, drawing water followed by ions into the intestinal lumen. On reaching the colon, bacteria metabolize lactose to shortchain fatty acids, carbon dioxide, and hydrogen gas. Short-chain fatty acids are transported with sodium into colonic epithelial cells, facilitating the reabsorption of fluid in the colon. If the colonic capacity for the reabsorption of short-chain fatty acids is exceeded, an osmotic diarrhea results (see later discussion of carbohydrate malabsorption in watery diarrheas).  

DIAGNOSIS

The diagnosis of acquired lactase deficiency can be made by empirical treatment with a lactose-free diet, which results in resolution of symptoms; by the hydrogen breath test after oral administration of lactose; or by genetic testing. Many intestinal diseases cause secondary reversible lactase deficiency, including viral gastroenteritis, celiac disease, giardiasis, and bacterial overgrowth.  

CELIAC DISEASE DEFINITION AND EPIDEMIOLOGY  

Celiac disease is a chronic inflammatory condition of the small intestine precipitated by the ingestion of wheat, rye, and barley in individuals with certain genetic predispositions.11 The prevalence in general populations worldwide is approximately 1%. High-risk groups for celiac disease include first-degree relatives and individuals with type 1 diabetes mellitus, autoimmune thyroid disease, autoimmune liver disease, and Down, Turner, or Williams syndromes. Approximately 5% of patients diagnosed with irritable bowel syndrome or with microscopic (lymphocytic) colitis have celiac disease.

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  



PATHOBIOLOGY

Environmental and genetic factors are important in the development of celiac disease, with some recent data suggesting a relationship with prior reovirus infection. Approximately 15% of first-degree relatives of affected individuals are found to have celiac disease. Predisposition has been mapped to the human leukocyte antigen (HLA)-D region on chromosome 6. More than 90% of northern Europeans with celiac disease have the DQ2 heterodimer encoded by alleles DQA1*0501 and DQB1*0201, compared with 20 to 30% of controls. A smaller celiac group carries HLA-DQ8. Many non-HLA alleles identified in genome-wide association studies account for a small portion of genetic risk. Most such genes are involved in adaptive and innate immune responses. Overlap variants have been identified in diabetes, rheumatoid arthritis, and Crohn disease. The alcohol-soluble protein fraction of wheat gluten, the gliadins, and similar prolamins in rye and barley trigger intestinal inflammation in susceptible individuals. Oat grains, which have prolamins rich in glutamine but not proline, are rarely toxic. Gliadins and similar prolamins with high proline content are relatively resistant to digestion by human proteases. Many gliadin and prolamin peptides have been identified that stimulate HLA-DQ2- and HLA-DQ8restricted intestinal T-cell clones from individuals with celiac disease. In blood lymphocyte comprehensive screen assays, three immunodominant prolamin peptides have been identified from wheat, barley, and rye in DQ2 celiac individuals; dominant peptides differ in those with DQ8. The DQ2 protein expressed on antigen-presenting cells has positively charged binding pockets; tTG (the autoantigen recognized by EMA) may enhance intestinal inflammation by deamidation of select glutamine residues in gliadin and similar prolamins to negatively charged glutamic acid. In the deamidated form, most gliadin peptides have a higher binding affinity for DQ2 and are more potent stimulants of gluten-sensitized T cells. Villous atrophy may be caused by inflammation that is triggered by γ-interferon released from DQ2- or DQ8restricted CD4 T cells in the lamina propria. Intraepithelial lymphocytes also directly kill intestinal epithelial cells under the influence of IL-15 released from stressed enterocytes.  

CLINICAL MANIFESTATIONS

Celiac disease usually manifests early in life, at approximately 2 years of age (after wheat has been introduced into the diet), or later in the second to fourth decades of life, but it can occur at any age. It may first manifest clinically after abdominal surgery or an episode of infectious diarrhea.12 Adults with celiac disease in the United States often present with anemia or osteoporosis without diarrhea or other gastrointestinal symptoms. Other extraintestinal manifestations of celiac disease include rash (dermatitis herpetiformis), neurologic disorders (peripheral neuropathy, ataxia, epilepsy), psychiatric disorders (depression, paranoia), reproductive disorders (infertility, spontaneous abortion), short stature, dental enamel hypoplasia, chronic hepatitis, and pancreatitis. Individuals with significant mucosal involvement present with watery diarrhea, weight loss or growth retardation, and the clinical manifestations of vitamin and mineral deficiencies. Cobalamin deficiency is more common (10% of patients) than previously thought and usually corrects itself on a gluten-free diet. Symptomatic individuals require supplementation of vitamin B12. Diarrhea is caused by many mechanisms, including a decreased surface area for water and electrolyte absorption, the osmotic effect of unabsorbed luminal nutrients, an increased surface area for chloride secretion (crypt hyperplasia), and the stimulation of intestinal fluid secretion by inflammatory mediators and unabsorbed fatty acids. Some individuals have impaired pancreatic enzyme secretion caused by decreased mucosal cholecystokinin release or bacterial overgrowth that may contribute to diarrhea. Individuals with nonceliac gluten sensitivity have wheat-related intestinal and extraintestinal symptoms similar to those of celiac disease but lack intestinal inflammation or celiac serologic markers. Fermentable oligosaccharides, disaccharides, monosaccharides, and polyols (FODMAPs) and gluten or nongluten wheat proteins (amylase trypsin inhibitors) may be the offending agents.  

893

DIAGNOSIS

Routine screening for celiac disease is not recommended.13 In symptomatic high-risk groups, however, screening with anti-tTG IgA antibody testing is a cost-effective strategy.14 High titers (>10-fold) of the anti-tTG IgA and EMA antibodies are virtually diagnostic of celiac disease. EMA IgA antibodies, detected by indirect immunofluorescence, are highly sensitive (90%) and specific (95 to 99%) for active celiac disease in skilled laboratory testing. An

FIGURE 131-5.  Intestinal biopsy appearance of flattened villi, hyperplastic crypts, and increased intraepithelial lymphocytes. (Courtesy John Hart, MD.)

FIGURE 131-6.  Regeneration of villi after initiation of a gluten-free diet. (Courtesy of John Hart, MD.)

ELISA test to detect antibodies against tTG has equal sensitivity to the EMA test but is less specific. The anti-deamidated gliadin (a biotinylated synthetic γ-gliadin peptide with glutamic acid substituted for glutamine) IgA and IgG antibody immunofluorometric assay has a sensitivity and specificity that approaches that of anti-tTG IgA antibody. Patients with mild disease may have negative antibody studies. Anti-tTG, gliadin peptide, and EMA IgA antibodies tests are negative in individuals with selective IgA deficiency (present in up to 3% of individuals with celiac disease). In these patients, the anti-deamidated gliadin IgG antibody, which has a sensitivity and specificity close to that of the tTG IgA antibody, is helpful in diagnosis. In equivocal cases (negative serologic findings and equivocal biopsy result or positive serologic findings and normal biopsy result), HLA genotyping is useful to exclude the diagnosis of celiac disease in persons who lack the DQ2 or DQ8 gene. The diagnosis of celiac disease is confirmed by characteristic abnormalities seen on duodenal biopsies and improving when a gluten-free diet is instituted (Figs. 131-5 and 131-6). In children, biopsy is not required if the patient has a greater than 10-fold increase in anti-tTG IgA, a positive EMA test, a positive HLA genetic test, and typical gastrointestinal symptoms.15 However, this strategy has not been validated for adults, for whom biopsy is recommended for diagnosis. Mucosal flattening may be observed endoscopically as scalloped or reduced duodenal folds. Characteristic features found on intestinal biopsy include blunted or absent villi, crypt hyperplasia, increased intraepithelial lymphocytes, and infiltration of the lamina propria with plasma cells and lymphocytes. In some individuals, the only abnormal biopsy finding is increased intraepithelial lymphocytes. A hypoplastic mucosa indicates irreversible (endstage) intestinal disease.

TREATMENT  Treatment consists of a lifelong gluten-free diet. Wheat, rye, and barley grains should be excluded from the diet. Rice and corn grains are tolerated. Oats (if not contaminated by wheat grain) are tolerated by most. Early referral to a registered dietitian knowledgeable in the gluten-free diet and to a reputable celiac support group or website is crucial in maintaining dietary

894

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  

TABLE 131-8 VITAMIN AND MINERAL DOSES USED IN THE TREATMENT OF MALABSORPTION VITAMIN

ORAL DOSE

Vitamin A*

Water-soluble A, 25,000 U/day†

Vitamin E

Water-soluble E, 400-800 U/day† ‡

Vitamin D

25,000-50,000 U/day

Vitamin K

5 mg/day

Folic acid

1 mg/day

Calcium§

1500-2000 mg elemental calcium/day Calcium citrate, 500 mg calcium/tablet† Calcium carbonate, 500 mg calcium/tablet†

Magnesium

Liquid magnesium gluconate† 1-3 tbsp (12-36 mEq magnesium) in 1-2 L of ORS or sports drink sipped throughout the day Magnesium chloride hexahydrate† 100-600 mg elemental magnesium/day

Zinc

Zinc gluconate† 20-50 mg elemental zinc/day‖

Iron

150-300 mg elemental iron/day Polysaccharide-iron complex† Iron sulfate or gluconate

PARENTERAL DOSE

2 mL of a 50% solution (8 mEq), total dose, in both buttocks IM

Iron sucrose¶ Sodium ferric gluconate complex¶ Iron dextran (as calculated for anemia) (IV or IM¶; Chapter 150)

B-complex vitamins 1 megadose tablet/day Vitamin B12

2 mg/day

1 mg IM or SC/mo**

Copper

Copper sulfate 2-3 mg/day

1-2 mg IV/day

negative celiac serologic findings, agammaglobulinemia [diagnosed by hypogammaglobulinemia and lack of plasma cells on small bowel biopsy], autoimmune enteritis [diagnosed by a positive antienterocyte antibody and crypt apoptosis or loss of goblet cells on small bowel biopsy]), bacterial overgrowth, pancreatic insufficiency, microscopic colitis, or other food allergies (cow’s milk, soy protein). Up to 40% of patients with celiac disease with symptomatic improvement have incomplete histologic recovery on a gluten-free diet; in such patients, a stricter diet may further improve symptoms and histology. A small percentage of patients with severe symptoms and enteropathy do not improve despite a strict gluten-free diet. In such patients, repeat intestinal biopsy is indicated. Some patients will have collagen deposition beneath the surface epithelium (collagenous sprue) or a polyclonal population of intraepithelial lymphocytes (refractory celiac disease type I). Others will have ulcerative jejunitis or a monoclonal population of intraepithelial T cells with an aberrant phenotype or clonal T-cell receptor-γ gene rearrangements (refractory celiac disease type II), which are predictive of enteropathy-associated T-cell lymphoma (Chapter 176) that portends a poor prognosis. Video capsule endoscopy and device-assisted enteroscopy may be helpful in establishing these diagnoses. Patients with collagenous sprue, autoimmune enteritis, or refractory celiac disease type I often respond to prednisone (20 to 40 mg/day orally) or budesonide (9 mg orally daily). Patients with celiac disease have a higher likelihood of having other autoimmune conditions, such as type 1 diabetes, thyroiditis, rheumatoid arthritis, inflammatory bowel disease, systemic lupus erythematosus, Sjögren syndrome, primary biliary cirrhosis, autoimmune hepatitis, vitiligo, and pancreatitis. Interestingly, approximately one third of patients with idiopathic sporadic ataxia have transglutaminase 6 antibodies, consistent with gluten-induced disease. Individuals with celiac disease are at increased risk for B-cell lymphoma (Chapter 176), gastrointestinal tract carcinomas (esophageal, small bowel, and colonic adenocarcinomas), and increased mortality; a strict gluten-free diet for life may lessen these risks. Intestinal T-cell lymphoma is rare and should be suspected in individuals who have abdominal pain, recurrence of symptoms after initial response to a gluten-free diet, or refractory celiac disease.  

IM, intramuscularly; IV, intravenously; ORS = oral rehydration solution; SC = subcutaneously. *Monitor serum vitamin A level to avoid toxicity, especially in patients with hypertriglyceridemia. † Form best absorbed or with least side effects. ‡ Monitor serum calcium and 25-OH vitamin D levels to avoid toxicity. § Monitor 24-hr urine calcium to assess adequacy of dose. ‖ If intestinal output is high, additional zinc should be given. Monitor for copper deficiency with high doses. ¶ Parenteral therapy should be given in a supervised outpatient setting because of the risk for fatal reactions. Decreased risk for fatal reactions when compared with iron dextran. **For vitamin B12 deficiency, 1 mg IM or SC twice per week for 4 wk, then once per month.

compliance. Owing to secondary lactase deficiency, a lactose-free diet should be recommended until symptoms improve. Celiac disease is associated with an increased risk for hospitalization with influenza and pneumococcal infection,16 so vaccination against both is recommended. Bone densitometry should be performed on individuals diagnosed with celiac disease who are older than 30 years of age because up to 70% have osteopenia or osteoporosis. Patients with diarrhea and weight loss should be screened for vitamin and mineral deficiencies. Documented deficiencies of vitamins and minerals should be replenished (Table 131-8), and women of childbearing age should take folic acid supplements. Bone mass often improves on a gluten-free diet alone. Patients with vitamin D or calcium deficiency should receive supplements (Chapter 205), with the dose monitored by 25-OH vitamin D levels and a 24-hour urine test for calcium. Most patients with seronegative villous atrophy do not have celiac disease. Most such patients will have spontaneous improvement despite continuing to ingest gluten.17



PROGNOSIS

Of patients with celiac disease treated with a gluten-free diet, 90% experience symptomatic improvement within 2 weeks. The most common cause of a poor dietary response is continued ingestion of gluten. Other possibilities include a missed intestinal infection (see later), an alternative diagnosis (e.g., ARB use particularly in elderly patients with a sprue-like enteropathy but

TROPICAL SPRUE

Tropical sprue is an inflammatory disease of the small intestine associated with the overgrowth of predominantly coliform bacteria. It occurs in residents or travelers to the tropics, especially India, Southeast Asia, Puerto Rico, and parts of the Caribbean. With the expansion of tourism and the global economy, this may be an under-recognized cause of enteropathy or mistaken for celiac disease. Individuals classically present with diarrhea and megaloblastic anemia secondary to vitamin B12 and folate deficiency, but some have anemia only. Intestinal biopsy characteristically shows subtotal and patchy villous atrophy in the proximal and distal small intestine, which may be caused by the effect of bacterial toxins on gut structure or by the secondary effects of vitamin B12 deficiency on the gut (megaloblastic gut). Diagnosis is based on history, documentation of vitamin B12 or folate deficiency, and the presence of an abnormal small intestinal biopsy report. Treatment is a prolonged course of tetracycline (250 mg orally four times daily) or doxycycline (100 mg orally two times daily), folic acid (5 mg/day orally), and, with coexistent deficiency, vitamin B12 injections (1000 µg weekly) until symptoms resolve. Relapse or reinfection occurs in 20%, mainly in natives of the tropics.  

GIARDIA LAMBLIA

Giardia lamblia infection (Chapter 330), the most common protozoal infection in the United States, can cause malabsorption in individuals infected with many trophozoites, especially the immunocompromised or IgA-deficient hosts. Malabsorption occurs when many organisms cover the epithelium and cause mucosal inflammation, which results in villous flattening and a decrease in absorptive surface area. Stool for ova and parasites at this stage of infection is often negative because of the attachment of organisms in the proximal small intestine. Diagnosis can be made by PCR, stool antigen-capture ELISA test, or duodenal biopsy.  

HUMAN IMMUNODEFICIENCY VIRUS

Diarrhea, malabsorption, and wasting are common in individuals with AIDS but are seen less frequently with improved antiretroviral therapy (Chapter 366). In patients who are receiving highly active antiretroviral therapy, diarrhea is more likely to be due to protease inhibitors than to enteric infection. Malabsorption is usually due to infection with cryptosporidia, Mycobacterium avium-intracellulare complex, I. belli, or microsporidia. AIDS enteropathy (a

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  

term used if no organism is identified) also can cause malabsorption. Mechanisms of malabsorption and diarrhea include villous atrophy, increased intestinal permeability, rapid small bowel transit (in patients with protozoal infection), and ultrastructural damage of enterocytes (in AIDS enteropathy). Among individuals with AIDS and diarrhea, results of fecal fat absorption are frequently abnormal. Serum albumin, vitamin B12, and zinc levels are often low. Vitamin B12 deficiency is caused mainly by ileal disease, but low intrinsic factor and decreased transcobalamin II may be contributing factors. Management of malabsorption should focus on restoring the immune system by treating the underlying HIV infection with antiviral therapy. If possible, the offending organism should be treated with antibiotics. If the organism cannot be eradicated, chronic diarrhea and malabsorption result; treatment in these cases consists of antimotility agents and a lactose-free, low-fat diet. Pancreatic enzyme replacement therapy can be tried in HIV-infected individuals who are taking highly active antiretroviral therapy or nucleoside analogues and who have fat malabsorption of obscure origin. If supplemental calories are needed, liquid oral supplements that are predigested and high in medium-chain triglycerides (semi-elemental) are tolerated best. Vitamin and mineral deficiencies should be screened for and treated.  

WHIPPLE DISEASE

Whipple disease (Chapters 133 and 259), a rare cause of malabsorption, manifests with gastrointestinal complaints in association with systemic symptoms, such as fever, joint pain, or neurologic manifestations.18 Approximately one third of patients have cardiac involvement, most commonly culture-negative endocarditis. Occasionally, individuals present with ocular or neurologic disease without gastrointestinal symptoms. Men are affected more commonly than women, particularly white men. The organism responsible for causing Whipple disease is a gram-positive actinomycete, Tropheryma whippelii. The epidemiology and pathogenesis of Whipple disease are poorly understood. The prevalence of the disease is higher in farmers than in other workers, which suggests that the organism lives in the soil. Using PCR, T. whippelii has been detected in sewage and in duodenal biopsy specimens, gastric juice, saliva, and stool of individuals without clinical disease. Whether the latter represents a carrier state or the presence of nonpathogenic organisms is not known. Immunologic defects, IL-16, and an association with the HLA-B27 gene may be disease factors. Small intestinal biopsy shows villous blunting and infiltration of the lamina propria with large macrophages that stain positive with the periodic acid–Schiff method and are filled with the organism. It is important to distinguish these macrophages from macrophages infected with M. aviumintracellulare complex, which stain positive on acid-fast staining and are found in individuals with AIDS. Treatment is with a prolonged course of broadspectrum antibiotics (e.g., ceftriaxone 2 g/day IV or meropenem 1 g IV three times daily; then trimethoprim 160 mg and sulfamethoxazole 800 mg orally two times daily for 1 year or trimethoprim 160 mg and sulfamethoxazole 800 mg orally two times daily for 1 year). Relapses occur, but initial treatment with parenteral ceftriaxone or meropenem appears to be associated with a low relapse rate.  



Crohn disease and radiation enteritis now is less common because of improved medical and radiation therapies. The severity of malabsorption depends on the site and extent of resection; the capacity for adaptive hyperplasia, dilation, and elongation; and the function of the residual bowel. Mechanisms of malabsorption after small bowel resection include a decreased absorptive surface area, decreased luminal bile salt concentration, rapid transit, and bacterial overgrowth. Limited jejunal resection usually is tolerated best because bile salt and vitamin B12 absorption remain normal. Ileal resection is less well tolerated because of the consequences of bile salt wasting and the limited capacity of the jejunum to undergo adaptive hyperplasia. Adaptive hyperplasia in residual small bowel after resection depends on nutrients, endogenous secretions (pancreatic and biliary juice), local factors (trefoil peptides, prostaglandins, polyamines), growth hormone, and growth factors (epidermal growth factor [EGF], insulin-like growth factor-1 [IGF1], transforming growth factor-α [TGFα], IL-11). The glucagon-like peptide-2 (GLP2) produced in L cells in the terminal ileum and colon is a potent stimulant of adaptive hyperplasia in the jejunum in response to a meal. Using intestinal stem cell technology, epithelial organoids have been successfully grown in culture systems, but bowel regeneration has not yet been accomplished. When less than 100 cm of jejunum remains, the colon takes on an important role in caloric salvage and fluid reabsorption. Malabsorbed carbohydrates are digested by colonic bacteria to short-chain fatty acids, which are absorbed in the colon.

TREATMENT  Parenteral nutrition may be avoided by a diet rich in complex carbohydrates, oral rehydration solutions, and acid-reducing and antimotility agents.19 In comparison, individuals with fewer than 100 cm of jejunum and no colon have high jejunostomy outputs and often require intravenous fluids or parenteral nutrition to survive. These individuals waste sodium, chloride, bicarbonate, magnesium, zinc, and water in their ostomy effluent. Dietary modifications should include a high-salt, nutrient-rich diet given in small meals. An oral rehydration solution with a sodium concentration 75 to 90 mmol/L is absorbed best. Oral vitamin and mineral doses higher than the usual U.S. recommended daily allowances are required (see Table 131-8). Vitamin B12 should be given parenterally (500 to 1000 µg SC every month). Magnesium deficiencies are often difficult to replenish with oral magnesium because of its osmotic effect in the intestinal lumen. A liquid magnesium preparation added to an oral rehydration solution and sipped throughout the day may minimize magnesiuminduced fluid losses. Potent antimotility agents, such as tincture of opium (0.5 to 1 mL orally four times daily) or liquid morphine 20 mg/mL (1 mL orally four times daily), often are needed to slow transit and maximize contact time for nutrient absorption. High-volume jejunostomy outputs can be lessened by inhibiting endogenous secretions with a proton pump inhibitor (e.g., omeprazole, 40 mg orally one or two times daily, or lansoprazole, 30 mg orally one or two times daily) and, in severe cases, octreotide (100 to 250 µg SC three times daily; if effective, convert to an equivalent long-acting monthly dosage). The benefit of octreotide may be offset by its potential to inhibit intestinal adaptation and impair pancreatic enzyme secretion with doses greater than 300 µg/day. In the most severe cases, supplemental calories must be provided by nocturnal tube feeding or parenteral nutrition. Treatment with growth hormone (0.1 mg/kg/day SC) with or without glutamine (30 g/day orally) for 4 weeks may reduce parenteral nutrition requirements in patients who have had massive intestinal resections. Teduglutide (0.05 mg/kg/day SC), a glucagon-like peptide-2 analogue that stimulates adaptive hyperplasia in remnant intestine after resection, reduces parenteral nutrition requirements. Small bowel transplantation should be considered for individuals who require parenteral nutrition to survive and then develop progressive liver disease or venous access problems.20

GRAFT-VERSUS-HOST DISEASE

Diarrhea occurs frequently after allogeneic bone marrow or stem cell transplantation (Chapter 168). Immediately after transplantation, diarrhea is caused by the toxic effects of cytoreductive therapy on the intestinal epithelium. From 20 to 100 days after transplantation, diarrhea is usually due to GVHD or infection. Patients with GVHD present clinically with a skin rash, hepatic cholestasis, buccal mucositis, anorexia, nausea, vomiting, abdominal cramps, and diarrhea. The diagnosis of GVHD in the gastrointestinal tract can be made on biopsy of the stomach, small intestine, or colon. In mild cases, the mucosa appears normal on inspection at endoscopy, but apoptosis of gastric gland or crypt cells can be found on biopsy. In severe cases, denudation of the intestinal epithelium results in diarrhea and malabsorption and often requires parenteral nutritional support. Octreotide (50 to 250 µg SC three times daily) may be helpful in controlling voluminous diarrhea. Treatment of GVHD is with steroids and antithymocyte globulin combined with parenteral nutritional support until intestinal function returns.

SHORT-BOWEL SYNDROME

Malabsorption caused by small bowel resection or surgical bypass is called the short-bowel syndrome. The most common causes in the United States are massive resection of the jejunum, owing to strangulated bowel, volvulus, or ischemia (mesenteric or after intra-abdominal surgery), and jejunal exclusion, owing to gastric bypass surgery. Short-bowel syndrome resulting from

895



PROGNOSIS

Long-term complications include bone disease, renal stones (oxalate stones if the colon is present, urate stones with a jejunostomy), gallstones, bacterial overgrowth, fat-soluble vitamin deficiencies, essential fatty acid deficiency, and d-lactic acidosis.

Conditions That Impair Nutrient Delivery to the Systemic Circulation  

Insoluble lipids (present in chylomicrons) are exocytosed across the basolateral membrane of epithelial cells into the intestinal lymphatics. From there, they enter the mesenteric lymphatics and the general circulation through the

896

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  

thoracic duct. Sugar monomers, amino acids, and medium-chain fatty acids are transported across the basolateral membrane of intestinal epithelial cells into capillaries and into the portal circulation. Sugar monomers are transported across the basolateral membrane by the facilitative glucose transporter isoform (GLUT2) and amino acids by facilitative amino acid carriers (see E-Fig. 131-1C).  

IMPAIRED LYMPHATIC DRAINAGE

Diseases that cause intestinal lymphatic obstruction, such as primary congenital lymphangiectasia (malunion of intestinal lymphatics), and diseases that result in secondary lymphangiectasia (lymphoma, tuberculosis, Kaposi sarcoma, retroperitoneal fibrosis, constrictive pericarditis, severe heart failure) result in fat malabsorption. The increased pressure in the intestinal lymphatics leads to leakage and sometimes rupture of lymph into the intestinal lumen, with the loss of lipids, γ-globulins, albumin, and lymphocytes. The diagnosis of lymphangiectasia can be made by intestinal biopsy, but the specific cause may be more difficult to identify. Individuals with lymphangiectasia malabsorb fat and fat-soluble vitamins and have protein loss into the intestinal lumen. The most common presentation is hypoproteinemic edema. Nutritional management includes a low-fat diet and supplementation with medium-chain triglycerides, which are absorbed directly into the portal circulation. Fat-soluble vitamins should be given if deficiencies develop.

  CONGENITAL MALABSORPTION AND DIARRHEA

A number of congenital disorders, mostly autosomal recessive, alter nutrient digestion, epithelial cell transport, or villus epithelial architecture. These disorders usually present within the first few weeks of life and may become targets for stem cell therapy.21 Mutations in sugar disaccharidases include lactase (rare), sucrose-isomaltase, maltase-glucoamylase, and trehalase (digests trehalose in mushrooms) deficiencies. These disorders are characterized by osmotic diarrhea when the offending sugar is introduced into the diet. Diagnosis is made by a low stool pH (owing to conversion of malabsorbed sugar to short-chain fatty acids by colonic bacteria), reduced enzyme activity on duodenal biopsy, and genetic testing. The treatment is to feed alternative sources of carbohydrates. Congenital enterokinase (enteropeptidase) deficiency causes severe protein malabsorption owing to the inability to activate trypsinogen and other pancreatic proteases in the intestinal lumen. Treatment is with supplemental pancreatic enzymes. Defects in epithelial transport proteins include congenital glucose-galactose malabsorption (mutation in the SGLT1 gene) that results in severe carbohydrate malabsorption. Congenital chloride and sodium diarrheas are rare autosomal recessive disorders caused by mutations in the epithelial Cl−/HCO3− (SLC26) and the Na+/H+ exchanger (SCL9, GUCY2C), respectively. These disorders cause maternal polyhydramnios and fluid-filled bowel loops in utero as well as severe diarrhea at birth. Diagnosis in the newborn is made by high stool chloride levels and metabolic alkalosis in congenital chloride diarrhea, high stool sodium levels and metabolic acidosis in congenital sodium diarrhea, and genetic testing. Abetalipoproteinemia is due to defects in chylomicron formation and exocytosis at the basolateral membrane of intestinal epithelial cells that are necessary for the delivery of lipids to the systemic circulation. Mutations in the microsomal triglyceride transfer protein (MTTP), apolipoprotein B gene, and SAR1B (triggers coat protein of vesicles in the endoplasmic reticulum) result in chylomicron retention in epithelial cells. Children with this disorder have fat malabsorption and the consequences of vitamin E deficiency (retinopathy and spinocerebellar degeneration). Biochemical tests show low plasma levels of apoprotein B, triglyceride, and cholesterol. Membrane lipid abnormalities result in red blood cell acanthosis (burr cells). Intestinal biopsy, which is diagnostic, is characterized by engorgement of epithelial cells with lipid droplets. Treatment is with a low-fat diet containing medium-chain triglycerides and high doses of fat-soluble vitamins, especially vitamin E. Homozygous loss-of-function mutations in the gene encoding CD55 cause protein-losing enteropathy, often accompanied by primary intestinal lymphangiectasia, edema due to hypoproteinemia, and malabsorption. Affected individuals may often have recurrent infections and angiopathic thromboembolic disease. Defects in villus epithelial cell structure presents at birth with severe malabsorption. Microvillus inclusion disease is due to defects in intraepithelial cellular protein transport and epithelial polarization. Mutations have been identified in myosin type 5 motor protein (MYO5B) and syntaxin-3 (STX3) genes. Tufting enteropathy is characterized by disorganized enterocytes at

villus tips (tufts). Mutations have been identified in EpCAM and SPINK2 to suggest disorders in tight junction formation. Most infants with these disorders require parenteral nutrition or small bowel transplantation for survival.

  WATERY DIARRHEA

Watery diarrhea may be due to osmotic, secretory, inflammatory, or often combined mechanisms (see Fig. 131-3).  



Ingestion of Nonabsorbable or Poorly Absorbable Solutes

MAGNESIUM AND SODIUM PHOSPHATE AND SULFATE DIARRHEAS

Magnesium, phosphate, and sulfate are poorly absorbed minerals. Individuals who ingest significant amounts of magnesium-based antacids or high-potency multimineral and multivitamin supplements or those who surreptitiously ingest magnesium-containing laxatives or nonabsorbable anion laxatives, such as Na2PO4 (neutral phosphate) or Na2SO4 (Glauber or Carlsbad salt), may develop osmotically induced, watery diarrhea that may be high volume.  

SORBITOL AND FRUCTOSE DIARRHEA

Dietetic food, chewing gum, candies, and medication elixirs that are sweetened with sorbitol, which is an unabsorbable carbohydrate, can cause diarrhea. Excessive consumption of pears, prunes, peaches, and apple juice, which also contain sorbitol and fructose, a poorly absorbable sugar, can result in diarrhea. Most soft drinks are now sweetened with fructose-containing corn syrup and may be a cause of diarrhea when ingested in high concentrations.  

Rapid Intestinal Transit

A small amount of carbohydrate in the diet is unabsorbed by the normal small intestine. Diets that are high in carbohydrate and low in fat may allow rapid gastric emptying and rapid small intestinal motility, thereby leading to carbohydrate malabsorption and osmotic diarrhea. Rapid transit time also occurs in thyrotoxicosis (Chapter 213). Because of the production of H2 and carbon dioxide gas by colonic bacteria, abdominal gas and cramping may be the predominant symptoms.  

Bile Acid Diarrhea

Ileal malabsorption of bile salts results in the stimulation of colonic fluid secretion and watery diarrhea. Three types of bile acid malabsorption induce diarrhea. Type 1 results when severe disease (e.g., Crohn disease), resection, or bypass of the distal ileum allows bile salts to escape absorption (see earlier). Type 2 may be congenital: rarely owing to a defect in the intestinal sodium bile acid transporter but more commonly owing to increased bile salt production (associated with decreased levels of fibroblast growth factor 19 (FGF19), an intestinal fibroblast growth factor that normally downregulates bile salt synthesis in the liver, and increased levels of 7-α-hydroxy-4-cholesten-3-one [C4], which is a marker of bile acid synthesis, in blood). The result is increased bile salt production that overwhelms reabsorption in the ileum. Type 3 is caused by various conditions, including prior cholecystectomy, celiac disease, pancreatic insufficiency, microscopic colitis, bacterial overgrowth, gastric surgery, or vagotomy. Postulated mechanisms include a bile salt storage problem, increased production, decreased recycling, or saturation of absorption. Primary (type 2) bile acid diarrhea is now thought to account for up to 30% of cases of functional or irritable bowel syndrome with chronic diarrhea (Chapter 128).22 The diagnosis is made by selenium homocholic acid taurine (75SeHCAT) testing (Table 131-6), a 48-hour stool collection for bile acids on a high-fat diet, or blood tests that measure FGF19 and C4 levels.23

TREATMENT  Diarrhea due to types 1 and 2 often responds to cholestyramine (2 to 4 g orally two to four times daily), colestipol (5 g orally two or three times daily), or the more potent and better tolerated bile salt binder, colesevelam (625-mg tablet orally two to six times daily). Obeticholic acid (25 mg orally daily), a farnesoid X receptor agonist that stimulates FGF19 production, may lessen the frequency of bile acid diarrhea.24 Fat-soluble vitamin deficiency is a potential risk with chronic use of bile salt binders. Although many patients with type 3 respond to cholestyramine or colesevelam, some do not. In these patients, motility-altering drugs such as opiates (e.g., loperamide, 2 to 4 mg orally two to four times daily) and anticholinergics (e.g., hyoscyamine sulfate, 0.125 to 0.250 mg orally two to four times daily) may be of benefit.

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  

  TRUE SECRETORY DIARRHEAS

Endocrine diseases that can cause secretory diarrheas (see Fig. 131-3) include carcinoid tumors (Chapter 219), gastrinomas (Chapter 219), VIPomas of the pancreas (Chapter 219), and medullary carcinoma of the thyroid (Chapter 213). Diarrhea is also seen in 60 to 80% of patients with systemic mastocytosis (Chapter 240). Diarrhea resulting from gastrinoma is distinct in that it is caused by high volumes of hydrochloric acid secretion that overwhelm the reabsorptive capacity of the colon and by maldigestion of fat owing to pH inactivation of pancreatic lipase and precipitation of bile salts.

times daily) or opiates (e.g., loperamide, 2 mg orally four times daily, maximal dose 16 mg daily) is warranted.

  INFLAMMATORY DIARRHEAS

Diarrhea resulting from inflammation is characterized by watery or bloody stools, fecal leukocytes, and loss of protein in the stool (see Fig. 131-3).  

Villous Adenomas

Large (4 to 18 cm) villous adenomas, particularly in the rectum or occasionally the sigmoid colon, may cause secretory diarrhea of 500 to 3000 mL/24 hours characterized by hypokalemia, chloride-rich stool, and metabolic alkalosis. Increased numbers of goblet cells and increased prostaglandin E2 are responsible for the diarrhea. Chloride wasting in the stool and metabolic alkalosis are also found in congenital chloridorrhea, which is caused by a defect in the intestinal Cl−/HCO3− transporter. The metabolic alkalosis distinguishes these two diarrheas from most other diarrheas that cause metabolic acidosis. A villous adenoma is usually diagnosed by colonoscopy. The prostaglandin antagonist indomethacin (25 to 100 mg/day orally) reduces the diarrhea in some patients; resection is curative.  

Diabetes Mellitus–Related Diarrhea

Constipation is more common than diarrhea in patients with diabetes. Highvolume, watery diarrhea, often with nocturnal incontinence, occurs in 20% of patients with poorly controlled type 1 diabetes. These patients usually have concomitant neuropathy, nephropathy, and retinopathy. The diarrhea may be due to several causes, including celiac disease, anal incontinence, bacterial overgrowth related to dysmotility, medications (metformin, acarbose), and autonomic neuropathy. If no specific cause is found, clonidine (initial dose 0.1 mg orally twice daily and titrated slowly to a maximal dose of 0.5 to 0.6 mg orally twice daily) may be helpful. Patients with neuropathy frequently have impaired anal sphincter function, and high-dose loperamide (4 mg orally four times daily) may improve the incontinence.  

Alcoholic Diarrhea

Diarrhea related to alcohol ingestion (Chapter 30) may be due to rapid intestinal transit, decreased bile and pancreatic secretion, nutritional deficiencies such as folate or vitamin B12, or alcohol-related enteric neuropathy. Diarrhea may be acute with binge drinking, or it may be chronic and watery and persist for days or weeks. The diarrhea slowly resolves with abstinence from alcohol, proper nutrition, and the repletion of vitamin deficiencies.  

Factitious Diarrhea

Approximately 30% of patients referred to tertiary centers have chronic diarrhea from laxative abuse. The diarrhea is usually severe and watery, often with nocturnal symptoms. Some patients may have abdominal pain, weight loss, nausea, vomiting, hypokalemic myopathy, and acidosis. Stool volumes range from 300 to 3000 mL per day depending on the dose of laxative ingested. In the United States, bisacodyl is the most common cause. Other culprits include anthraquinone (senna, cascara, aloe, rhubarb) or osmotic laxatives (neutral phosphate, Epsom salts, and magnesium citrate). Some patients abuse other agents that cause diarrhea, such as the diuretics furosemide and ethacrynic acid. More than 90% of laxative abusers are women who have underlying eating disorders such as anorexia nervosa or bulimia (Chapter 206) or middle-aged women who have complicated medical histories and who often work in health care. In patients with unexplained diarrhea, laxative screening of stool and urine (see later) should be performed to exclude this syndrome before an extensive medical evaluation is performed for other causes of chronic diarrhea.  

Chronic Idiopathic Secretory Diarrhea

In a small subset of patients with secretory diarrhea, no cause is found despite an extensive evaluation. These cases are labeled as chronic idiopathic secretory diarrhea. In most patients, the diarrhea resolves within 6 to 24 months, which suggests a possible postinfectious or Brainerd diarrhea. If no diagnosis is found after thorough testing and a search for surreptitious laxative abuse, a therapeutic trial with bile salt–binding drugs (e.g., cholestyramine, 4 g orally before meals three times daily, or the more potent colesevelam, 625-mg tablet two to six

Inflammatory Bowel Disease

See Chapter 132.  



897

Eosinophilic Gastroenteritis

Eosinophilic gastroenteritis is an increasingly recognized condition of unknown etiology characterized by infiltration of eosinophils in the mucosa, muscle, or serosal layers of the gastrointestinal tract.25 Approximately 50% of patients have atopic histories. Infestation with nematodes (Chapter 335) must be excluded before this diagnosis is made. Diarrhea occurs in 30 to 60% of patients with mucosal disease. Patients with involvement of the muscle layer often present with abdominal pain, nausea, and vomiting indicative of gastric outlet or intestinal obstruction. Peripheral eosinophilia is present in most patients. The disease may involve the entire gastrointestinal tract from esophagus to anus, or it may be isolated to a segment. With diffuse involvement, patients may have steatorrhea, protein-losing enteropathy, and blood loss. Management includes an elimination diet based on allergy testing or an elemental diet. Budesonide (9 mg orally daily) or prednisone (0.5 to 1 mg/ kg orally daily) is highly effective, but relapses are common when corticosteroids are tapered. Targeted anti-eosinophil agents (e.g., anti-IL-5 therapies) are under investigation.  

Microscopic (Collagenous and Lymphocytic) Colitis

These two conditions, collectively known as microscopic colitis, may or may not be the same disease or variants of the same disease.26 Lymphocytic colitis is equally prevalent in men and women, whereas collagenous colitis occurs 10 times more often in middle-aged or elderly women. These conditions may be associated with autoimmune disease or with NSAID use. There is an increased prevalence (5%) of microscopic colitis among individuals with celiac disease. These diseases may be categorized as either inflammatory or secretory diarrheas. An epidemiologic relationship to medications such as NSAIDs, H2-receptor blockers, proton pump inhibitors, selective serotonin reuptake inhibitors, and smoking has been reported, and increased luminal prostaglandin levels may cause the diarrhea. Enteric infections, food hypersensitivity, and intraluminal bile have been proposed as triggers for prostaglandin release from lymphocytes. Antidiarrheal agents such as loperamide (2 mg orally four times daily) are the mainstay of therapy for mild diarrhea, which usually has a benign and self-limited course. Budesonide (4.5 to 9 mg orally daily) is the most effective therapy. A10  In patients who do not tolerate or respond to it, an alternative is bismuth subsalicylate therapy (8 chewable 262-mg tablets orally daily). Patients with refractory disease may require corticosteroids (e.g., prednisone, 40 mg/ day orally), a trial of azathioprine or anti-TNF-α antibodies.  

Food Allergy

Food allergies or sensitivities, especially to cow’s milk and soy protein, are a well-established cause of enterocolitis in children, with an estimated frequency of 5%. Symptoms of abdominal cramps, diarrhea, and sometimes vomiting occur shortly after ingestion of the allergen (Chapter 238). The role of food allergy in causing diarrhea in adults is less clear owing to the lack of a reliable diagnostic test. Allergy testing correlates poorly with intestinal allergy. The most common food allergens are milk, soy, eggs, seafood, nuts, and wheat. Sequential elimination diets can be diagnostic and therapeutic.

  RADIATION ENTERITIS

Patients who receive pelvic radiation for malignancies of the female urogenital tract or the male prostate may develop chronic radiation enterocolitis 6 to 24 months after total doses of radiation greater than 40 to 60 Gy (Chapters 17 and 133), but symptoms can develop as late as 20 years after treatment. Early abnormalities include an increase in inflammatory mediators, an increase in cholinergic stimulation of intestinal tissue, and endothelial cell apoptosis that precedes epithelial cell apoptosis. The last finding suggests that vascular injury is the primary event. Diarrhea may be caused by bile acid malabsorption if the ileum is damaged, by bacterial overgrowth if radiation causes small intestinal strictures or bypass, or by radiation-induced chronic inflammation of the small intestine and colon. Rapid transit also may contribute to malabsorption and diarrhea.

TREATMENT  Treatment is often unsatisfactory. Anti-inflammatory drugs (sulfasalazine, corticosteroids) and antibiotics have been tried with little success. Cholestyramine (4 g orally three times daily) and NSAIDs (e.g., naproxen, 250 to 500 mg orally twice daily) may help, as may opiates (loperamide, 2 mg orally four times daily, or loperamide-N-oxide, 3 mg orally two times daily).

  PROTEIN-LOSING GASTROENTEROPATHY

Severe protein loss through the gastrointestinal tract can be caused by mucosal diseases such as lymphangiectasia, lymphatic obstruction, bacterial or parasitic infection, gastritis (Chapter 130), gastric cancer, collagenous colitis, inflammatory bowel disease (Chapter 132), celiac disease, sarcoidosis (Chapter 89), lymphoma (Chapter 176), tuberculosis (Chapter 308), Ménétrier disease (Chapter 183), eosinophilic gastroenteritis, and food allergies. A variety of extraintestinal diseases, including systemic lupus erythematosus (Chapter 250), heart failure (Chapter 52), and constrictive pericarditis (Chapter 68), also can be causative. Patients with systemic lupus erythematosus (Chapter 250) may present with protein-losing enteropathy as the only manifestation of their disease. Treatment focuses on the underlying disease.

  MISCELLANEOUS DISEASES

Although acute mesenteric arterial or venous thrombosis manifests as an acute bloody diarrhea, chronic mesenteric vascular ischemia (Chapter 134) may manifest as watery diarrhea. Gastrointestinal tuberculosis (Chapter 308) and histoplasmosis (Chapter 316) manifest as diarrhea that may be either bloody or watery, as do certain immunologic diseases, such as Behçet syndrome and Churg-Strauss syndrome. All of these diseases may be misdiagnosed as inflammatory bowel disease (Chapter 132). Neutropenic enterocolitis, an ileocolitis that occurs in patients with neutropenia and leukemia, sometimes is caused by C. difficile infection.

TREATMENT OF CHRONIC DIARRHEA  Antidiarrheal Therapy

Antidiarrheal agents are of two types: those used for mild to moderate diarrheas and those used for severe secretory diarrheas. A major shortcoming of opiates, the most commonly prescribed antidiarrheal agents, is that they have no antisecretory effect. Rather, they act by decreasing intestinal motility, thereby allowing longer contact time with the mucosa for improved fluid absorption. The exception is racecadotril, an enkephalinase inhibitor, that blocks intestinal fluid secretion without affecting motility. Bulk-forming agents (psyllium, 7 g in 8 oz water orally up to five times daily and methylcellulose, 3 to 6 tablets twice daily with 300 mL of water) act by binding water and increasing the consistency of stool. Pectin has been shown to have proabsorptive activity. These agents may be useful in patients with fecal incontinence. Bismuth subsalicylates (524 mg orally every hour up to eight doses daily) have mild antisecretory and antimotility effects and are effective and safe in mild diarrheas. The opiates may be symptomatically useful in mild to moderate diarrheas. Paregoric, deodorized tincture of opium, codeine, and diphenoxylate with atropine largely have been supplanted by loperamide. Loperamide does not pass the blood-brain barrier and has a high first-pass metabolism in the liver; it has a high therapeutic-to-toxic ratio and is essentially devoid of addiction potential. It is safe in adults, even in total doses of 24 mg/day. The usual dose is 2 to 4 mg two to four times daily. Opiates may be harmful in patients with severe diarrheas because large volumes of fluid may pool in the intestinal lumen (third space), and stool output is no longer a reliable gauge for replacing fluid losses. The antimotility effects are a problem in infectious diarrheas because stasis may enhance bacterial invasion and delay clearance of microorganisms from the bowel. Opiates and anticholinergics also are dangerous in severe inflammatory bowel disease or severe C. difficile infection, where they may precipitate megacolon. Antidiarrhea agents that are used for the treatment of severe secretory and inflammatory diarrheas generally have profiles with more serious side effects. The somatostatin analogue octreotide (initial dose, 100 to 600 µg SC in two to four divided doses daily; maximal dose, 1500 µg daily) lessens diarrhea in the carcinoid syndrome and in neuroendocrine tumors because it inhibits hormone secretion by the tumor. It is also effective in the treatment of dumping syndrome

and chemotherapy-related diarrheas. Long-acting octreotide (20 to 30 mg intramuscularly every month) is available for once-a-month dosing. Octreotide can suppress pancreatic enzyme secretion and make diarrhea worse; it also may be of only limited usefulness in short-bowel syndrome and AIDS diarrhea. Agents such as phenothiazine and calcium-channel blockers have mild antisecretory effects, but side effects limit their use. Clonidine (initial dose, 0.1 mg orally twice daily, titrated slowly to a maximal dose of 0.5 to 0.6 mg twice daily) is most useful in opiate withdrawal diarrhea and is sometimes useful in diabetic diarrhea; postural hypotension may limit its use, particularly in patients with diabetes. Alosetron (0.5 mg orally twice daily for 4 weeks, maximal dose 1 mg orally twice daily) may be justified for severe diarrhea-predominant irritable bowel syndrome; associations with ischemic colitis and severe constipation have limited its use. Indomethacin (250 to 500 mg orally twice daily), a cyclooxygenase blocker that inhibits prostaglandin production, is useful in the treatment of diarrheas caused by acute radiation, AIDS, or villous adenomas of the rectum or colon; occasionally, it may be useful in neuroendocrine tumors and food allergy. For eosinophilic gastroenteritis, corticosteroids (prednisone, 20 to 40 mg/day orally for 7 to 10 days) are the mainstay of therapy, but disodium cromoglycate (200 mg orally four times daily) also may be useful; food elimination diets are not usually effective. Treatment of inflammatory bowel disease is described in Chapter 132.

  Grade A References A1. Freedman SB, Williamson-Urquhart S, Farion KJ, et al. Multicenter trial of a combination probiotic for children with gastroenteritis. N Engl J Med. 2018;379:2015-2026. A2. Schnadower D, Tarr PI, Casper TC, et al. Lactobacillus rhamnosus GG versus placebo for acute gastroenteritis in children. N Engl J Med. 2018;379:2002-2014. A3. Wilcox MH, Gerding DN, Poxton IR, et al. Bezlotoxumab for prevention of recurrent Clostridium difficile infection. N Engl J Med. 2017;376:305-317. A4. Kelly CR, Khoruts A, Staley C, et al. Effect of fecal microbiota transplantation on recurrence in multiply recurrent Clostridium difficile infection: a randomized trial. Ann Intern Med. 2016;165:609-616. A5. Soares-Weiser K, Maclehose H, Bergman H, et al. Vaccines for preventing rotavirus diarrhoea: vaccines in use. Cochrane Database Syst Rev. 2012;11:CD008521. A6. Severe K, Rouzier V, Anglade SB, et al. Effectiveness of oral cholera vaccine in Haiti: 37-month follow-up. Am J Trop Med Hyg. 2016;94:1136-1142. A7. Riddle MS, Connor BA, Beeching NJ, et al. Guidelines for the prevention and treatment of travelers’ diarrhea: a graded expert panel report. J Travel Med. 2017;24:S57-S74. A8. Gatta L, Scarpignato C. Systematic review with meta-analysis: rifaximin is effective and safe for the treatment of small intestine bacterial overgrowth. Aliment Pharmacol Ther. 2017;45:604-616. A9. Wilcox C, Turner J, Green J. Systematic review: the management of chronic diarrhoea due to bile acid malabsorption. Aliment Pharmacol Ther. 2014;39:923-939. A10. Münch A, Bohr J, Miehlke S, et al. Low-dose budesonide for maintenance of clinical remission in collagenous colitis: a randomised, placebo-controlled, 12-month trial. Gut. 2016;65:47-56.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  

GENERAL REFERENCES 1. Bányai K, Estes MK, Martella V, et al. Viral gastroenteritis. Lancet. 2018;392:175-186. 2. Giddings SL, Stevens AM, Leung DT. Traveler’s diarrhea. Med Clin North Am. 2016;100:317-330. 3. Cybulski RJ Jr, Bateman AC, Bourassa L, et al. Clinical impact of a multiplex gastrointestinal PCR panel in patients with acute gastroenteritis. Clin Infect Dis. 2018;67:1688-1696. 4. Feher C, Mensa J. A comparison of current guidelines of five international societies on Clostridium difficile infection management. Infect Dis Ther. 2016;5:207-230. 5. DuPont HL. Persistent diarrhea: a clinical review. JAMA. 2016;315:2712-2723. 6. Camilleri M, Sellin JH, Barrett KE. Pathophysiology, evaluation, and management of chronic watery diarrhea. Gastroenterology. 2017;152:515-532. 7. Schiller LR. Evaluation of chronic diarrhea and irritable bowel syndrome with diarrhea in adults in the era of precision medicine. Am J Gastroenterol. 2018;113:660-669. 8. Arasaradnam RP, Brown S, Forbes A, et al. Guidelines for the investigation of chronic diarrhoea in adults: British Society of Gastroenterology, 3rd edition. Gut. 2018;67:1380-1399. 9. Ruiz-Campos L, Gisbert JP, Ysamat M, et al. Systematic review with meta-analysis: the prevalence of bile acid malabsorption and response to colestyramine in patients with chronic watery diarrhoea and previous cholecystectomy. Aliment Pharmacol Ther. 2019;49:242-250. 10. Ghoshal UC, Ghoshal U. Small intestinal bacterial overgrowth and other intestinal disorders. Gastroenterol Clin North Am. 2017;46:103-120. 11. Lindfors K, Ciacci C, Kurppa K, et al. Coeliac disease. Nat Rev Dis Primers. 2019;5:1-18. 12. Lebwohl B, Sanders DS, Green PHR. Coeliac disease. Lancet. 2018;391:70-81. 13. Chou R, Bougatsos C, Blazina I, et al. Screening for celiac disease: evidence report and systematic review for the US Preventive Services Task Force. JAMA. 2017;317:1258-1268. 14. Bibbins-Domingo K, Grossman DC, Curry SJ, et al. Screening for celiac disease: US Preventive Services Task Force recommendation statement. JAMA. 2017;317:1252-1257.

898.e1

15. Werkstetter KJ, Korponay-Szabo IR, Popp A, et al. Accuracy in diagnosis of celiac disease without biopsies in clinical practice. Gastroenterology. 2017;153:924-935. 16. Simons M, Scott-Sheldon LAJ, Risech-Neyman Y, et al. Celiac disease and increased risk of pneumococcal infection: a systematic review and meta-analysis. Am J Med. 2018;131:83-89. 17. Aziz I, Peerally MF, Barnes JH, et al. The clinical and phenotypical assessment of seronegative villous atrophy; a prospective UK centre experience evaluating 200 adult cases over a 15-year period (20002015). Gut. 2017;66:1563-1572. 18. Hujoel IA, Johnson DH, Lebwohl B, et al. Tropheryma whipplei infection (Whipple disease) in the USA. Dig Dis Sci. 2018;64:213-223. 19. Buchman AL. Intestinal failure and rehabilitation. Gastroenterol Clin North Am. 2018;47:327-340. 20. Loo L, Vrakas G, Reddy S, et al. Intestinal transplantation: a review. Curr Opin Gastroenterol. 2017;33:203-211. 21. Hong SN, Dunn JC, Stelzner M, et al. Concise review: the potential use of intestinal stem cells to treat patients with intestinal failure. Stem Cells Transl Med. 2017;6:666-676. 22. Oduyebo I, Camilleri M. Bile acid disease: the emerging epidemic. Curr Opin Gastroenterol. 2017;33:189-195. 23. Vijayvargiya P, Camilleri M, Carlson P, et al. Performance characteristics of serum C4 and FGF19 measurements to exclude the diagnosis of bile acid diarrhoea in IBS-diarrhoea and functional diarrhoea. Aliment Pharmacol Ther. 2017;46:581-588. 24. Walters JR, Johnston IM, Nolan JD, et al. The response of patients with bile acid diarrhoea to the farnesoid X receptor agonist obeticholic acid. Aliment Pharmacol Ther. 2015;41:54-64. 25. Walker MM, Potter M, Talley NJ. Eosinophilic gastroenteritis and other eosinophilic gut diseases distal to the oesophagus. Lancet Gastroenterol Hepatol. 2018;3:271-280. 26. Pardi DS. Diagnosis and management of microscopic colitis. Am J Gastroenterol. 2017;112:78-85.

898

CHAPTER 132  Inflammatory Bowel Disease  

132  INFLAMMATORY BOWEL DISEASE GARY R. LICHTENSTEIN



DEFINITION

Inflammatory bowel disease refers to two chronic idiopathic inflammatory disorders, ulcerative colitis and Crohn disease. Characteristic clinical, endoscopic, and histologic features are critical for the diagnosis of these disorders, but no single individual finding is absolutely diagnostic for one disease or the other. Mucosal ulceration in Crohn disease may be transmural and may occur anywhere in the gastrointestinal tract, most commonly in the distal ileum and proximal colon. The hallmark of ulcerative colitis is continuous ulceration starting in the rectum and limited to the colon. Approximately 10% of patients with inflammatory bowel disease have what is called “inflammatory bowel disease-undetermined,” a term used when Crohn colitis cannot be distinguished from ulcerative colitis.

898.e2

ABSTRACT

CHAPTER 132  Inflammatory Bowel Disease  

Inflammatory bowel disease is an idiopathic disorder with etiology related to a dysregulated immune response to the host intestinal microbiota. Ulcerative colitis and Crohn disease represent the two primary types of inflammatory bowel disease, with ulcerative colitis having inflammation limited to the colon and Crohn disease involving any segment of the gastrointestinal tract from the mouth to the anus. Patients with inflammatory bowel disease have a genetic predisposition, but these diseases spare no socioeconomic class and occur equally in men and women. Both ulcerative colitis and colonic Crohn disease increase the risk for colorectal cancer. With the recent advent of biologic therapy, the treatment of these disorders has evolved and improved prognosis.

KEYWORDS

Crohn disease inflammatory bowel disease ulcerative colitis ileitis colitis

CHAPTER 132  Inflammatory Bowel Disease  



EPIDEMIOLOGY

Inflammatory bowel disease occurs worldwide, but the highest incidence is found in North America, the United Kingdom, and northern Europe. Data suggest an increasing incidence and prevalence over time and in different regions around the world, although ulcerative colitis remains slightly more prevalent than Crohn disease.1 The incidence of ulcerative colitis in North America is estimated to be 19.3 per 100,000 person years and 24.3 per 100,000 person years in Europe, with a prevalence of approximately 285 per 100,000 persons in North America and up to 500 per 100,000 persons in Europe. The incidence of Crohn disease in North America is estimated to be 20.2 per 100,000 person years and 12.7 per 100,000 person years in Europe, with a prevalence of approximately 320 per 100,000 in North America and in Europe.2 Crohn disease and ulcerative colitis may occur at any age, but both have their peak incidence in the second to fourth decade, with a second peak in the seventh decade. The female-to-male ratio for both ulcerative colitis and Crohn disease suggests no gender preference. Crohn disease and ulcerative colitis are complex polygenic disorders, for which having a family history of inflammatory bowel disease is a risk factor. Crohn disease and ulcerative colitis occur in all ethnic and socioeconomic groups, but their incidence is highest in white Caucasians and Jewish people of Eastern European (Ashkenazi) descent. In North America and the United Kingdom, however, the incidence of Crohn disease in African Americans and African Caribbeans appears to be approaching that of whites. Studies of migrants from certain developing countries in South Asia to the United Kingdom suggest an increased prevalence of inflammatory bowel disease in subsequent generations, presumably because of environmental influences. Cigarette smoking is associated with a worse prognosis in patients with Crohn disease but an improved course in ulcerative colitis. Nonsteroidal antiinflammatory drugs (NSAIDs) appear to be associated with new onset of inflammatory bowel disease and with exacerbations of disease. Appendectomy for an inflammatory indication has been suggested as protective against the development of ulcerative colitis. Diet does not clearly affect the course of inflammatory bowel disease.  

PATHOBIOLOGY

Although the trigger for inflammatory bowel disease is not known, three major pathways likely activate the disease: a genetic predisposition, immune dysregulation, and an environmental trigger. A possible explanation is that the inability of the innate immune system to clear microbial antigens, combined with increased intestinal epithelial permeability to antigens, eventually leads to an overactive adaptive immune response.

Genetics

Of patients with inflammatory bowel disease, 5 to 20% have another family member with inflammatory bowel disease. First-degree relatives have a 10- to 15-fold increased risk for developing inflammatory bowel disease. The concordance rate of developing Crohn disease in identical twins, siblings, and first-degree relatives is 50%, 0 to 3%, and 5 to 10%, respectively. Ulcerative colitis follows similar genetic patterns but with slightly lower risk rates. Twenty percent of patients with a positive family history of inflammatory bowel disease will have discordant disease type: one family member with Crohn disease and another with ulcerative colitis. More than 200 gene susceptibility loci have been linked to inflammatory bowel disease, with at least 30 specific for Crohn disease and more than 20 specific for ulcerative colitis.3 Some of these genes also may correlate with the severity of disease. The first gene discovered to be associated with Crohn disease was NOD2/CARD15, which is located on chromosome 16 (16q12) and expressed in intestinal epithelial Paneth cells, macrophages, and dendritic cells. This gene is involved in the expression of an intracellular receptor that senses muramyl dipeptide, a peptidoglycan component of gram-positive bacteria. Activation of NOD2 leads to activation of nuclear factor κ-B (NF-κB), which mediates transcription of numerous proinflammatory cytokines. A mutation in the leucine-rich domain of the NOD2 protein, which interacts with bacterial lipopolysaccharide, leads to failure in activation of NF-κB and is associated with the development of Crohn disease. The ATG16L1 gene on chromosome 2 and the IRGM gene on chromosome 5 also have been associated with increased susceptibility to Crohn disease. Both are members of a family of genes involved in autophagy, an autonomous process that involves the maintenance of cellular homeostasis and organelle turnover, as well as the processing of intracellular pathogens, the subsequent presentation of antigens, and the regulation of cell signaling. Toll-like receptor-4

899

gene polymorphisms are associated with both Crohn disease and ulcerative colitis. Polymorphisms of the interleukin-23 (IL-23) receptor gene are associated with ulcerative colitis and a varied risk for Crohn disease. The presence of an adenylate cyclase 7 (ADCY7) missense mutation confers a higher risk for ulcerative colitis.4 The protein encoded by ADCY7 is a member of a family of ten enzymes that convert ATP to cAMP. Each member has unique tissuespecific expression patterns. ADCY7 is expressed in hematopoietic cells. In these cells, cAMP modulates innate and adaptive immune functions, including inhibition of the proinflammatory cytokine tumor necrosis factor (TNF)-α. The loss of function of ADCY7 reduces production of cAMP, thereby leading to an excessive inflammatory response that predisposes to ulcerative colitis. Human leukocyte antigen (HLA) class II polymorphisms, especially in HLA-DR molecules, may confer increased risk for ulcerative colitis and possibly Crohn as well. The OCTN1 gene (located on chromosome 5q31) and the DLG5 gene (located on chromosome 10) are associated with Crohn disease. DLG5, which encodes a scaffolding protein that is important for maintaining epithelial integrity in various organs, may interact with the NOD2/CARD15 gene to increase susceptibility to Crohn disease. OCTN1 encodes for an ion channel and also increases the risk for Crohn disease; mutations in this gene may disrupt ion channels through altered function of cation transporters and cellto-cell signaling in the intestinal epithelium. Whether susceptibility genes can be linked to clinical phenotypes is an area of intense investigation. One large-scale study found that three loci (NOD2, MHC, and MST1 3p21) were associated with whether inflammatory bowel disease manifested as ileal Crohn disease, colonic Crohn disease, or ulcerative colitis rather than by the classic definitions of Crohn disease or ulcerative colitis.5 Inflammatory bowel disease also has been associated with Turner syndrome (Chapter 220), glycogen storage type Ib (Chapter 196), and the Hermansky-Pudlak syndrome (triad of albinism, platelet aggregation defect, and accumulation of ceroid-like pigment in tissue) (Chapter 164). Inflammatory bowel disease is associated with various diseases that have known genetic predisposition, including ankylosing spondylitis (Chapter 249), psoriasis (Chapter 409), atopy (Chapter 235), eczema (Chapter 409), celiac sprue (Chapter 131), cystic fibrosis (Chapter 83), primary sclerosing cholangitis (Chapter 146), multiple sclerosis (Chapter 383), autoimmune thyroid disease (Chapter 213), autoimmune hemolytic anemia (Chapter 151), primary biliary cirrhosis (Chapter 146), myasthenia gravis (Chapter 394), and Cogan syndrome (Chapter 254).  

PATHOPHYSIOLOGY

Microbes likely play a part in the development of inflammatory bowel disease. In several animal models of colitis, colitis does not develop in a sterile environment but can be induced after the introduction of commensal bacteria. For instance, helminth infection protects mice with Nod2 mutations, which otherwise make them susceptible to Crohn disease, by inhibiting colonization by an inflammatory Bacteroides species.6 Diverting the fecal stream away from active mucosal inflammation, such as in an ileostomy, also helps alleviate inflammation in Crohn disease. Crohn disease and ulcerative colitis preferentially occur in the terminal ileum and colon, which contain the highest concentration of bacteria, on the order of approximately 1012 organisms per gram of luminal contents. Antibiotics, particularly antibiotics with broadspectrum anaerobic coverage, are helpful in the treatment of Crohn disease. More recently, several genetic polymorphisms associated with sensing the intestinal microbial environment and triggering an immune response have been linked to inflammatory bowel disease. Both Crohn disease and ulcerative colitis are products of a dysregulated innate immune system that triggers T cells and a humoral response. TH17 cells, which are activated in Crohn disease and ulcerative colitis, are stimulated by IL-23, which is produced by antigen-presenting cells.

Pathology Crohn Disease

As a result of a dysregulated immune system, patients with Crohn disease develop aphthous ulcers, which are superficial mucosal ulcers. As the disease progresses, the ulceration becomes deeper, transmural, and discrete; it may form a serpiginous pattern and may occur anywhere from the mouth to the anus in a noncontinuous pattern. The most common location for ulceration is the ileocecal region. In some patients, chronic disease leads to the formation of fibrotic strictures, and approximately 30% of patients may develop fistulas in the course of their disease. In early Crohn disease, the histopathologic findings are characterized by an acute inflammatory infiltrate in the lamina propria, with cryptitis, and crypt

900

CHAPTER 132  Inflammatory Bowel Disease  

abscesses. Later in the disease process, the crypt architecture becomes distorted, with a lymphocytic infiltrate and a resulting branching and shortening of the crypts. Noncaseating granulomas, which may be found in up to 15% of endoscopic biopsy specimens and up to 70% of surgical specimens, are not unique to Crohn disease but help confirm the diagnosis when other classic features are present. Surgical specimens also may show transmural intestinal wall inflammation and fat creeping on the serosal surface.

Ulcerative Colitis

In mild ulcerative colitis, the mucosa is granular, hyperemic, and edematous in appearance. As the disease becomes more severe, the mucosa ulcerates, and the ulcers may extend into the lamina propria. Ulcerative colitis starts in the rectum and may extend proximally in a continuous pattern, but it affects only the colon. Pseudopolyps may form owing to epithelial regeneration after recurrent acute attacks. With chronic disease, the colonic mucosa may lose the normal fold pattern, the colon may shorten, and the colon may appear narrowed. In early ulcerative colitis, the histopathologic findings are characterized by epithelial necrosis, an acute inflammatory infiltrate in the lamina propria, cryptitis, and crypt abscesses. In chronic disease, a predominant lymphocytic infiltrate and distortion of crypt architecture are seen.  

CLINICAL MANIFESTATIONS

Symptoms of inflammatory bowel disease are varied and may be a consequence of the location of the disease, the duration of disease, and any anatomic complications of the disease, such as strictures and fistulas in Crohn disease (Table 132-1).

Symptoms Crohn Disease

The terminal ileum is affected in about 70% of patients with Crohn disease.7 Primary ileal disease occurs in 30% of patients, whereas ileocolonic disease occurs in 40%. Symptoms may include abdominal pain, typically in the right lower quadrant, diarrhea, hematochezia, and fatigue. With more severe disease, fever and weight loss may be present. Some patients may present with obstructive symptoms, such as abdominal pain, abdominal distention, and nausea. Only approximately 5% of patients develop Crohn disease in the upper gastrointestinal tract, and esophageal Crohn disease occurs in fewer than 2% of patients. Subjects with upper gastrointestinal Crohn disease may present with dysphagia, odynophagia, chest pain, or heartburn. Gastroduodenal disease occurs in 0.5 to 4% of patients and commonly occurs along with distal disease. Symptoms may include upper abdominal pain. Isolated jejunal disease is rare; if the jejunum is involved, there is also distal small bowel involvement. Up to 30% of patients have perianal disease (Chapter 136) that may include the development of fistulas, abscesses, fissures, and skin tags. Symptoms of perianal disease include pain and discharge. Fever may be present if there is an abscess. Fistulas, which are internal tracts that can occur anywhere in the gastrointestinal tract and connect to various sites, occur in 20 to 40% of Crohn patients. Penetrating Crohn disease also may cause intra-abdominal and perianal

TABLE 132-1 CLINICAL CHARACTERISTICS OF CROHN DISEASE AND ULCERATIVE COLITIS

abscesses owing to a fistula with a blind end or intestinal perforation. External fistulas, which present with symptoms of fluid discharge from the cutaneous opening, can be enterocutaneous or perianal. Internal fistulas can be enteroenteric, rectovaginal, or enterocolonic. Patients may present with persistent abdominal pain and fever with an abscess in this location.

Ulcerative Colitis

As with Crohn disease, symptoms and signs of ulcerative colitis depend on the extent and severity of disease. At the time of diagnosis, 14 to 37% of patients have pancolitis, 36 to 41% have disease extending beyond the rectum, and 44 to 49% have proctosigmoiditis. Symptoms include hematochezia, diarrhea, tenesmus, production of excessive mucus, urgency to defecate, and abdominal pain. In the setting of proctitis or proctosigmoiditis, patients may have constipation with difficulty defecating. With more extensive and severe colonic involvement, patients also may have weight loss and fever. They also may have nausea and vomiting because of abdominal pain, fatigue because of anemia, and peripheral edema because of hypoalbuminemia.

Physical Examination

Signs on physical examination are representative of the type of disease as well as its location and severity. Oral ulcers may be present in Crohn disease. The location of abdominal tenderness usually reflects the location of intestinal involvement. In Crohn disease, abdominal tenderness is classically in the right lower quadrant and may include fullness or a mass depending on the severity of inflammation. Peritoneal signs may occur when penetrating Crohn disease causes intestinal perforation. Rectal examination may reveal skin tags, hemorrhoids, fissure, and fistulae.

Extraintestinal Manifestations

Arthropathy, the most common extraintestinal manifestation (Table 132-2), affects up to 10 to 20% of subjects.8 Peripheral arthralgias, arthritis, ankylosing spondylitis (Chapter 249), and sacroiliitis may exacerbate with gastrointestinal symptoms. Dermatologic disorders, such as erythema nodosum (10 to 15%; see Fig. 411-24) and pyoderma gangrenosum (1 to 2%; Chapter 245), develop in up to 15% of patients. Eye disorders, especially uveitis and episcleritis (Chapter 395), may occur in 5 to 15%. Patients with inflammatory bowel disease also have up to a 10% risk for renal calculi, especially calcium oxalate stones (Chapter 117), in the setting of fat malabsorption with Crohn disease in the small bowel. Uric acid stones can occur in the setting of severe volume depletion. Patients with inflammatory bowel disease, especially patients with ulcerative colitis, are at increased risk for primary sclerosing cholangitis—2 to 7.5% of patients develop this disorder, and 70 to 80% of patients with this disorder have inflammatory bowel disease (Chapter 146).

Extraintestinal Complications

Patients with inflammatory bowel disease are susceptible to extraintestinal complications from the disease itself or medications used to treat disease. These complications include osteoporosis, osteomalacia, arthritic complications, thromboembolic events, pulmonary disease, and renal, dermatologic, and neurologic complications. Osteoporosis occurs in approximately 15% of patients, and steroid therapy (Chapter 32) is the major risk factor; avascular

TABLE 132-2 EXTRAINTESTINAL COMPLICATIONS OF INFLAMMATORY BOWEL DISEASE CROHN DISEASE

ULCERATIVE COLITIS

Ocular disorders (uveitis, episcleritis)

+

+

Arthropathy

+

+

Oral ulcers

+



+

+



Skin disorders (pyoderma gangrenosum, erythema nodosum)



Nephrolithiasis

+

+

Usually discrete

Continuous

Primary sclerosing cholangitis

+

+

+



Bone disorders (osteoporosis, osteomalacia)

+



Thromboembolic disease

+

+

B12 deficiency

+



CHARACTERISTICS

CROHN DISEASE

ULCERATIVE COLITIS

Peak age of onset (years of age)

15-30, 2nd peak in the 7th decade

20-40, 2nd smaller peak beyond the 7th decade

Sex distribution (F/M)

1.2/1

1/1

Potential sites of gastrointestinal involvement

Esophagus to anus

colon

Skipped areas of involvement

+

Transmural inflammation

+

Type of ulceration Fistula Stricture

+



Perianal disease (fissure, skin tags)

+



COMPLICATIONS

CHAPTER 132  Inflammatory Bowel Disease  

necrosis of the hip and septic arthritis are unusual complications of steroids or other immunosuppressive therapies. Cheilitis may be a result of iron deficiency anemia (Chapter 150). Patients with inflammatory bowel disease are at an increased risk for thromboembolic disease, especially in the setting of active intestinal disease, even when compared with other autoimmune diseases such as rheumatoid arthritis and celiac disease. Secondary amyloidosis with renal involvement can be a consequence of chronic inflammation. Asthma is the most common pulmonary disorder observed in association with Crohn disease. Patients also are at risk for optic neuritis and multiple sclerosis (Chapter 383), as well as for peripheral neuropathy (Chapter 392) from vitamin B12 deficiency, which may occur as a result of poor absorption owing to active small bowel disease or surgical resection.  

DIAGNOSIS

When diarrhea (Chapter 131) is the predominant symptom, the initial evaluation should include a thorough medical history, testing for infectious colitis (Chapter 131), and screening for endocrine-metabolic disorders such as hyperthyroidism (Chapter 213) and hypocalcemia (Chapter 232). Infections with organisms such as Salmonella (Chapter 292) Shigella (Chapter 293), Amoeba (Chapter 331), Giardia (Chapter 330), Escherichia coli O157:H7 (Chapter 288), and Campylobacter (Chapter 287) can be accompanied by bloody diarrhea, abdominal cramps, and an endoscopic mucosal appearance identical to that of ulcerative colitis. Stool studies are needed to diagnose or exclude these infections. If hematochezia and abdominal pain are the predominant symptoms, the differential diagnosis is broad (Table 132-3).

Diagnostic Evaluation Endoscopic Evaluation

In a patient with symptoms suggestive of inflammatory bowel disease and no evidence for an infection to explain the symptoms, endoscopic evaluation is essential. Colonoscopy is the initial endoscopic test for patients who present with lower gastrointestinal symptoms such as diarrhea and hematochezia, except in the presence of acute severe peritoneal symptoms. Colonoscopy to the terminal ileum is important if there is a potential diagnosis of inflammatory bowel disease. Small bowel imaging (such as small bowel follow-through or computed tomography [CT] enterography) also may be needed to determine whether there is small bowel disease or to determine the distribution of disease.

901

Capsule endoscopy is useful if all other endoscopic and radiologic testing is nondiagnostic, but Crohn disease of the small bowel is still suspected. Findings on capsule endoscopy should be followed by endoscopy to obtain biopsies. Capsule endoscopy should not be performed if Crohn disease is complicated by a known small bowel stricture.

Crohn Disease

Early endoscopic findings in Crohn disease include superficial small mucosal ulcers, also called aphthous ulcers. As the severity of Crohn disease progresses, the ulcerations become deeper and may become round, linear, or serpiginous. Intersecting longitudinal and transverse ulcers cause a cobblestone mucosal appearance, with “stone” areas representing normal mucosa (Fig. 132-1). Areas of ulceration, which are typically interspersed with normal “skip” areas, may occur anywhere from the esophagus to anus but are most common in the ileocecal region. Isolated colonic disease occurs in 25% of patients, and 60% will have rectal involvement, thereby making it at times difficult to differentiate from ulcerative colitis. The diagnosis of inflammatory bowel disease is contingent upon accurate histopathologic results, so biopsy of the affected area(s) is key. Findings of an inflammatory infiltrate in the lamina propria and distortion of the crypt architecture support the diagnosis (Fig. 132-2). The diagnosis of Crohn disease may be made by histopathologic examination alone if noncaseating granulomas are seen, but granulomas are rarely found on endoscopic biopsies. The diagnosis of Crohn disease is usually based on a combination of information gleaned from histopathologic findings, colonoscopy, and small bowel imaging. A skip pattern of ulceration, ulceration in the small bowel or upper gastrointestinal tract, or the presence of fistulas support the diagnosis of Crohn disease. Colonic and small bowel ulceration occur in several other disorders, including infections that may not be detected by routine stool studies (such as enterohemorrhagic

TABLE 132-3 DIFFERENTIAL DIAGNOSIS OF ILEITIS AND COLITIS INFECTIONS Bacterial   Aeromonas   Campylobacter jejuni   Chlamydia (proctitis)   Clostridium difficile   Mycobacterium tuberculosis   Salmonella   Shigella  Enterohemorrhagic Escherichia coli   Yersinia Viral  Cytomegalovirus   Herpes simplex virus (proctitis)   Human immunodeficiency virus Fungal   Histoplasma capsulatum Parasitic   Entamoeba histolytica   Helminths VASCULAR Collagen Vascular Disease   Behçet disease   Churg-Strauss syndrome   Henoch-Schönlein purpura   Systemic lupus erythematosus   Polyarteritis nodosa Ischemia

MEDICATIONS/TOXINS Nonsteroidal anti-inflammatory drugs Pancreatic enzyme supplements— fibrosing colopathy Phosphosoda bowel preparations Radiation INFLAMMATORY Appendicitis Diverticular disease Eosinophilic gastroenteritis Nongranulomatous ulcerative jejunoileitis (celiac disease) NEOPLASIA Carcinoid Carcinoma primary or metastatic Lymphoma Mycosis fungoides Malignant histiocytosis MISCELLANEOUS Amyloidosis Sarcoidosis Endometriosis Tubo-ovarian abscesses

Modified from Aberra FN, Lichtenstein GR. Crohn disease. In: Talley NJ, Kane SV, Wallace MD, eds. Practical Gastroenterology and Hepatology: Small and Large Intestine. Wiley-Blackwell; 2010:225-235.

FIGURE 132-1.  Endoscopic appearance of Crohn disease with cobblestoning.

FIGURE 132-2.  In ulcerative colitis, histopathology from colonic biopsies reveals features of crypt distortion and lymphocytic infiltration in the mucosa. (Modified from AGA Institute GastroSlides 2010.)

902

CHAPTER 132  Inflammatory Bowel Disease  

E. coli), vascular disorders, immune-related enterocolitis, neoplasia, diverticulitis, radiation, and medications such as NSAIDs (see Table 132-3).

Ulcerative Colitis

The diagnosis of ulcerative colitis is based on endoscopic findings (Video 132-1) and histopathology. Early in the disease process, patients develop diffuse mucosal erythema with loss of the normal mucosal vascular pattern. In mild disease, the mucosa may have a granular and edematous appearance. As the disease becomes more severe, the mucosa becomes more friable, bleeds easily when the mucosa is touched, and may eventually ulcerate (Fig. 132-3). Endoscopic findings, which start in the rectum and may extend proximally in a continuous pattern, affect only the colon. The term “backwash ileitis” describes a spillover effect from ulcerative colitis and should not be construed as actual involvement of the terminal ileum by ulcerative colitis. Pseudopolyps may form owing to epithelial regeneration after recurrent attacks in patients with long-standing disease. The presence of pseudopolyps is thought to be a marker of more severe disease. With chronic disease, the colonic mucosa may lose its normal fold pattern, and the colon may shorten and appear narrowed. Careful endoscopic observation—including the observed vascular pattern, bleeding, and ulceration—can assess the severity of ulcerative colitis and monitor its response to treatment. Features such as crypt distortion, continuous mucosal inflammation starting from the rectum, absence of granulomas, and absence of small bowel disease are consistent with ulcerative colitis. Early in the disease process, chronic inflammatory findings, such as crypt distortion, may not be present, and the diagnosis may be more difficult to establish.

Radiology

Radiologic imaging is vital and almost always should be obtained when inflammatory bowel disease, particularly Crohn disease, is suspected. CT enterography and magnetic resonance imaging (MRI) enterography have supplanted barium studies (such as an upper gastrointestinal series, small bowel follow-through, and barium enema) to diagnose the extent, severity, and type of disease (strictures and fistulas) of Crohn disease. CT and MRI enterography are also useful for the detection of abdominal abscesses and perineal abscesses.

Laboratory Findings

Anemia may result from chronic disease, blood loss, or nutritional deficiencies of iron, folate, or vitamin B12. A modestly elevated leukocyte count is indicative of active disease, but a marked elevation suggests an abscess or another suppurative complication. The erythrocyte sedimentation rate and C-reactive protein are nonspecific serum inflammatory markers that are commonly used to monitor the activity of disease. Hypoalbuminemia is an indication of malnutrition and is common with active disease. Ileal disease or resection of more than 100 cm of distal ileum results in a diminished serum vitamin B12 level because of malabsorption.

Serologic Markers

Serologic markers are supportive but may not be used independently to diagnose inflammatory bowel disease. Anti–Saccharomyces cerevisiae antibodies (ASCA), which are antibodies to yeast, are present in 40 to 70% of patients with Crohn disease and in less than 15% of patients with ulcerative colitis. The combination of elevated ASCA immunoglobulin A (IgA) and IgG titers is highly specific for Crohn disease, ranging from 89 to 100%. Perinuclear antineutrophil cytoplasmic antibodies (pANCA) are present in 20% of Crohn patients, primarily in colon-predominant disease, and in 55% of patients with ulcerative colitis. ASCA-positive and pANCA-negative disease are associated with a sensitivity of 55% and specificity of 93% for Crohn disease. The antimicrobial antibodies anti-I2 (Crohn disease–related protein from Pseudomonas fluorescens), anti-Cbir1 (flagellin-like antigen), and anti-OmpC (E. coli outer membrane porin C) are also associated with Crohn disease.

TREATMENT  The aim of medical therapy is to reduce inflammation and subsequently induce and maintain clinical remission while simultaneously healing the mucosa. Medications used to treat inflammatory bowel disease include the categories of 5-aminosalicylate (5-ASA), antibiotics, corticosteroids, immunomodulators, and biologics (infliximab, adalimumab, certolizumab pegol, golimumab, vedolizumab, natalizumab, and ustekinumab; Table 132-4). The specific medical therapy selected is based on the location, extent (nonpenetrating and nonstricturing, stricturing, and penetrating and fistulizing disease), severity of disease, and future prognosis (Fig. 132-4; see Fig. 132-3). Patients with high-risk disease (high potential for future surgery or complications such as fistula and abscess) or high disease activity are considered candidates for biologic therapy.9 Supportive medical therapy, such as antidiarrheal and antispasmodic medications, also may be used.

Categories of Medical Therapy 5-Aminosalicylate

5-ASA, which acts as a topical anti-inflammatory within the lumen of the intestine, is used to treat mild to moderate ulcerative colitis and as maintenance therapy for patients in remission. A1  A2  Sulfasalazine is the combination of a sulfapyridine with 5-ASA; 5-ASA is responsible for the anti-inflammatory property of this drug, whereas sulfapyridine is the carrier that allows 5-ASA to be delivered to the colon. Other oral formulations of 5-ASA allow it to be delivered to the intestine by different mechanisms. Mesalamine is released in the intestine based on a pH delivery model, whereas sulfasalazine, olsalazine, and balsalazide are released in the intestine by bacterial cleavage of a covalent bond between 5-ASA and a prodrug. For rectal and sigmoid disease, 5-ASA suppository and enema preparations are also effective for induction and maintenance of remission in patients with ulcerative colitis. A3  Adverse events associated with 5-ASAs are uncommon but may include nausea, dyspepsia, hair loss, headache, worsening diarrhea, and hypersensitivity reactions ,

Corticosteroids

Corticosteroids are primarily used to treat flares of ulcerative colitis and Crohn disease. Oral formulations may be used for mild to moderate disease, whereas systemic corticosteroids are used for moderate to severe disease. Enteric-coated budesonide, a pH-dependent ileal release formulation, is an oral corticosteroid with high topical activity and low systemic bioavailability (10%). Enteric-coated budesonide is indicated for treatment of active mild to moderate ileocecal Crohn disease. Budesonide MMX is released in the colon and is available for treatment of mild to moderately active ulcerative colitis. Budesonide foam is administered rectally with an applicator in the colon for mild to moderately active ulcerative colitis limited to the rectum and sigmoid colon. Oral corticosteroids such as prednisone and methylprednisolone are used for moderate to severe disease, starting at doses ranging from 40 to 60 mg/day for 1 to 2 weeks, then typically tapered over a period of 8 to 12 weeks. Intravenous methylprednisolone is used for severe disease, with dosing ranging from 40 to 60 mg/day. When symptoms are controlled, systemic corticosteroids are not recommended because of their substantial side effects and their lack of efficacy for maintaining remission (Chapter 32).

Immunomodulatory therapy

FIGURE 132-3.  Endoscopic appearance of ulcerative colitis.

In patients who remain symptomatic despite 5-ASA therapy or who have moderate to severe Crohn disease or ulcerative colitis, the thiopurine analogues (6-mercaptopurine and azathioprine) may be used. A4  Methotrexate also may be used for moderate to severe Crohn disease. A5  These immunomodulators have a slow onset of action, often requiring several months of administration before their benefits are achieved, so they are used principally for maintenance of remission. Azathioprine, the prodrug of 6-mercaptopurine, typically is

903

CHAPTER 132  Inflammatory Bowel Disease  

TABLE 132-4 MEDICAL THERAPIES FOR INFLAMMATORY BOWEL DISEASE DRUG

DOSE

RELEASE SITE

5-AMINOSALICYLATES Sulfasalazine (Azulfidine)

2-6 g/day

Colon

Mesalamine (Asacol, Lialda, Apriso)

2.4-4.8 g/day

Distal ileum, colon

Olsalazine (Dipentum)

1-3 g/day

Colon

Balsalazide (Colazal)

6.25 g/day

Colon

Mesalamine (Pentasa)

2-4 g/day

Duodenum, jejunum, ileum, colon

Mesalamine (Rowasa), enema, suppository

4 g/day (enema) 1 g/day (suppository)

Rectum/sigmoid Rectum

Mesalamine (Canasa), suppository

1 g/day (suppository)

Rectum

Budesonide (Entocort EC)

Induction: 9 mg PO daily Maintenance: 6 mg PO daily

Small intestine

Budesonide (MMX, UCERIS)

Induction 9 mg PO daily

Colon

Prednisone

0.25-0.75 mg/kg PO daily

Systemic

Methylprednisolone

40-60 mg IV daily

Systemic

6-Mercaptopurine

1.5 mg/kg/day

Systemic

Azathioprine

2.5 mg/kg/day

Systemic

Methotrexate

Induction: 25 mg SC weekly × 4 mo. Maintenance: 15-25 mg SC weekly

Systemic

Tofacitinib

10 mg PO twice daily × 8 weeks then 5 to 10 mg PO twice daily × 8 weeks further then 10 mg twice daily for at least 8 weeks; followed by 5 or 10 mg twice daily, depending on therapeutic response

Systemic

Cyclosporine

2-4 mg/kg/day IV

Systemic

Infliximab

Induction: 5 mg/kg IV weeks 0, 2, 6 Maintenance: 5-10 mg/kg IV every 8 weeks

Systemic

Adalimumab

Induction: 160 mg SC week 0, 80 mg week 2 Maintenance: 40 mg SC every other week

Systemic

Golimumab

Induction: 200 mg SC week 0 and 100 mg week 2 Maintenance: 100 mg SC every 4 weeks

Systemic

Certolizumab pegol

Induction: 400 mg SC weeks 0, 2, 4 Maintenance: 400 mg SC every 4 weeks

Systemic

Natalizumab

300 mg IV every 4 weeks

Systemic

Vedolizumab

Induction: 300 mg IV at 0, 2, and 6 weeks Maintenance: 300 mg IV every 8 weeks

Systemic

Ustekinumab

Induction: intravenous: Up to 55 kg: 260 mg; Greater than 55 kg to 85 kg: 390 mg; Greater than 85 kg: 520 mg. Maintenance: 90 mg SC every 8 weeks

CORTICOSTEROIDS

IMMUNOMODULATORS

BIOLOGICS

IV = intravenously; PO = orally; SC = subcutaneously.

Treatment Options

Mild to moderate disease activity • 5-ASA (rectal preparation for proctitis) for induction • Prednisone (rectal preparation for proctitis) for induction • Budesonide MMX for induction

FIGURE 132-4.  Ulcerative colitis treatment algorithm.

Moderate to severe disease activity • Steroid for induction (oral or intravenous) • Immunomodulator • Anti-TNF therapy Infliximab, adalimumab or golimumab • Vedolizumab • Cyclosporine for induction (intravenous)

Disease refractory to medical therapy, colonic dysplasia, or cancer • Proctocolectomy

904

CHAPTER 132  Inflammatory Bowel Disease  

prescribed at a dose of 2 to 3 mg/kg/day; the equivalent dose of 6-mercaptopurine is 1.5 mg/kg/day. A disadvantage of the thiopurine analogues is the slow clinical response that may not be evident for as long as 12 weeks. Their side effects include allergic reactions, pancreatitis, myelosuppression, nausea, infections, hepatotoxicity, and malignancy, especially lymphoma. The white blood cell count and liver chemistries must be monitored routinely. Methotrexate, which is a folic acid antagonist, is given as 25 mg intramuscularly (IM) or subcutaneously (SC) once per week for 16 weeks for active Crohn disease and, 15 mg to 25 mg IM or SC once per week for maintaining remission. Tofacitinib (an oral, non-selective inhibitor of the Janus kinase family) has a more rapid onset of activity and is FDA-approved for treatment of moderate to severe ulcerative colitis A6  at 10 mg orally twice daily for at least 8 weeks, then 5 or 10 mg twice daily with the medication discontinued if an adequate therapeutic benefit is not achieved by 16 weeks. Side effects include activation of nonsystemic herpes zoster, an abnormal lipid profile, and the potential risk of lymphoma.

Antibiotics

The exact mechanism for the beneficial effect of broad-spectrum antibiotics in the treatment of inflammatory bowel disease is not known. Potential mechanisms include eliminating small intestinal bacterial overgrowth, eradicating a bacterially mediated antigenic trigger, and potential immunosuppressive properties (e.g., metronidazole). The primary role of antibiotics is in Crohn disease, where metronidazole (10 to 20 mg/kg/day for 4 to 8 weeks), ciprofloxacin (500 mg orally (PO) twice daily for 4 to 8 weeks), or both are primary inductive therapies for perianal fistulae. Metronidazole also may be a helpful adjunctive treatment for colonic Crohn disease and to prevent postoperative recurrence in Crohn disease as well. In addition, a novel enteric form of rifaximin may be of benefit for mild to moderate Crohn disease.

Biologics

Anti–Tumor Necrosis Factor-α Agents

Monoclonal antibody therapy directed against tumor necrosis factor-α (anti–TNF-α) include infliximab (Remicade), which is a chimeric mouse-human IgG1 monoclonal antibody that is approved to treat moderate to severe Crohn disease, fistulizing Crohn disease, and moderate to severe ulcerative colitis that has failed to respond to conventional therapy. Adalimumab (Humira) and certolizumab pegol (Cimzia) have been approved to treat moderate to severe Crohn disease that has failed to respond to conventional therapy, and adalimumab (Humira) and golimumab (Simponi) have been approved to treat moderate to severe ulcerative colitis that has failed to respond to conventional therapy. Adalimumab and golimumab are fully human IgG1 antibodies that are self-administered subcutaneously. Certolizumab pegol, which is a chimeric pegylated Fab fragment to TNF-α, also is administered subcutaneously.

UGI Disease

Other Molecules

Natalizumab (Tysabri), a humanized IgG4 monoclonal antibody, binds to the α4 subunit of α4β1 and α4β7 integrins expressed on all leukocytes except neutrophils. Natalizumab inhibits the interactions between α4 integrins on the surface of leukocytes and adhesion molecules on vascular endothelial cells in the gastrointestinal tract, thereby preventing adhesion and recruitment of leukocytes. Natalizumab is approved for the treatment of moderate to severe Crohn disease that is refractory to other therapies, but there are strict guidelines for prescribing natalizumab because of its associated risk for progressive multifocal leukoencephalopathy (Chapter 346). Another small adhesion molecule, vedolizumab (Entyvio), is approved for patients with moderate to severe ulcerative colitis and adult patients with moderate to severe Crohn disease when one or more standard therapies (corticosteroids, immunomodulators, or TNF blocker medications) have not provided an adequate response. Because this agent is gut-selective and is not associated with impairment of central nervous system immunosurveillance, the risk for progressive multifocal leukoencephalopathy appears to be very low in this molecule. Ustekinumab is a monoclonal antibody to the p40 subunit of interleukin-12 and interleukin-23. When used as intravenous induction therapy with patients with moderately to severely active Crohn disease and as subcutaneous maintenance therapy, it can induce and maintain clinical remission. A8 

Crohn Disease Medical Therapy Mild to Moderate Crohn Disease

Sulfasalazine (3 to 6 g/day) is superior to placebo for treating active ileocolonic and colonic Crohn disease, with response rates ranging from 45 to 55% for mild to moderate disease, but is not clearly effective for small bowel disease alone (Fig. 132-5). Mesalamine is not thought to be effective for induction of remission or for maintaining remission in Crohn disease. In a phase 2 randomized trial of patients with moderately active Crohn disease (800 mg of extended intestinal release rifaximin twice daily for 12 weeks) induced remission in 63% of patients compared with 43% of controls, with few adverse events. For mild to moderate Crohn disease involving the distal small intestine or proximal colon, eneteric-coated budesonide (9 mg/day) provides approximately a 70% response rate after 8 weeks and is significantly more effective than mesalamine (4 g/

Ileocolitis/ colitis

Perianal disease

Fistulizing disease*

Mild to moderate Extended release Budesonide†

Mild to moderate Immunomodulator Anti–TNF-α agents Antibiotics

Mild to moderate Immunomodulator Anti–TNF-α agents Antibiotics**

Ileitis

Mild to moderate Mild to moderate Extended release Immunomodulator Budesonide Anti–TNF-α agents Ustekinumab Vedolizumab

In patients with early Crohn disease, timely use and escalation of anti–TNF-α treatment based on symptoms and biomarkers (e.g., a fecal calprotectin ≥250 µg/g of stool, a C-reactive protein level ≥5 mg/L, a Crohn Disease Activity Index ≥150 (E-Table 132-1), or prednisone use in the previous week) results in better clinical and endoscopic outcomes than does symptom-driven decisions alone. A7  Before anti-TNF therapy is considered, however, risk versus benefit needs to be assessed in each individual patient, given the potential risk for infection and malignancy.

Moderate to severe Corticosteroids induction Immunomodulator: Anti– TNF-α agents, vedolizumab, tofacitinib, natalizumab, ustekinumab Supportive therapy for induction (NPO, TPN)

Moderate to severe Temporary diversion Surgery Supportive therapy for induction (NPO, TPN)

Persistent fistula Temporary diversion Surgery Supportive therapy for induction (NPO, TPN)

Refractory to medical therapy or perforation Surgery *Abscess should be excluded before initiating medical therapy. † Proximal colon disease involvement. **Perianal location. FIGURE 132-5.  Crohn disease treatment algorithm. NPO = nothing by mouth; TNF-α = tumor necrosis factor-α; TPN = total parenteral nutrition; UGI = upper gastrointestinal.

CHAPTER 132  Inflammatory Bowel Disease  

E-TABLE 132-1 CROHN DISEASE ACTIVITY INDEX CLINICAL OR LABORATORY VARIABLE

WEIGHTING FACTOR

Number of liquid or soft stools each day × 7 days

×2

Abdominal pain (graded from 0-3 on severity) each day × 7 days

×5

General well being, subjectively assessed from 0 (well) to 4 (terrible) each day × 7 days

×7

Presence of complications (see below)

× 20

Taking diphenoxylate-atropine or opiates for diarrhea

× 30

Presence of an abdominal mass (0 = none, 2 = questionable, 5 = definite)

× 10

Hematocrit of 2 × 103 IU/mL

Inactive carrier

+





+



+

2 × 103 IU/mL

Ab = antibodies; Ag = antigen; HBc = hepatitis B core; HBe = hepatitis B e; HBs = hepatitis B surface; HBV = hepatitis B virus; IgM = immunoglobulin M. *Anti-HBc IgM can be detected at low titers.

968

CHAPTER 140  Chronic Viral and Autoimmune Hepatitis  

B

A

C FIGURE 140-1.  Liver biopsies in patients with chronic hepatitis C. A, Lymphoid nodule with germinal center; minimal interface hepatitis (hematein-eosin, magnification 200×). B, Mild fibrosis, Metavir score F1 (picrosirius-hemalun, magnification 100×). C, Extensive fibrosis, Metavir score F3 (picrosirius-hemalun, magnification 20×). (Courtesy Prof. Elie-Serge Zafrani, Department of Pathology, Henri Mondor Hospital, Créteil, France.)

TABLE 140-3 RESULTS OF MAJOR STUDIES FOR THE TREATMENT OF HBeAg-POSITIVE CHRONIC HEPATITIS B AT 6 MONTHS FOLLOWING 48 OR 52 WEEKS OF PEGYLATED IFN-α (PEGIFN-α) AND AT 48 OR 52 WEEKS OF NUCLEOSIDE/ NUCLEOTIDE ANALOGUE THERAPY PegIFN-α PLACEBO/CONTROL GROUPS

NUCLEOSIDE/NUCLEOTIDE ANALOGUE

PegIFN-α2a 180 µg ONCE WEEKLY

PegIFN-α2b 100 µg ONCE WEEKLY

ENTECAVIR 0.5 mg DAILY

TENOFOVIR DISOPROXIL FUMARATE 245 mg DAILY

TENOFOVIR ALAFENAMIDE 25 mg DAILY

HBe seroconversion

4-6%

32%

29%

21%

21%

10%

HBV DNA 3.5

2.8-3.5

control) or INR

2.3

CTP classification: Child A: score of 5-6; Child B: score of 7-9; Child C: score of 10-15 2. MODEL OF END-STAGE LIVER DISEASE (MELD) SCORE (RANGE, 6-40) [0.957 × LN (creatinine in mg/dL) + 0.378 × LN (bilirubin in mg/dL) + 1.12 × LN (INR) + 0.643] × 10 MELD-Na is calculated first by determining the traditional MELD (MELD(0)); if the initial MELD(i) score is 12 or greater, the score is adjusted by incorporating the serum sodium value [MELD(0) + 1.32 × (137-Na) − [0.033 × MELD(i) × (137-Na)] INR = international normalized ratio; LN = natural logarithm.

Hepatopulmonary syndrome is associated with exertional dyspnea, which can lead to extreme debilitation. Clubbing of the fingers, cyanosis, and vascular spiders may be seen on physical examination. Hepatopulmonary syndrome is present in approximately 5 to 10% of patients awaiting liver transplantation. Portopulmonary hypertension is manifested as exertional dyspnea, syncope, and chest pain. On examination, an accentuated second sound and right ventricular heave are prominent (Chapter 75).  

DIAGNOSIS

The diagnosis of cirrhosis should be considered in any patient with chronic liver disease. In asymptomatic patients with compensated cirrhosis, typical signs of cirrhosis may not be present, and the physical examination and laboratory test findings may be entirely normal. The diagnosis may often require histologic confirmation by liver biopsy, which is the “gold standard” for the diagnosis of cirrhosis. However, liver biopsy is an invasive procedure subject to sampling error, and the presence of cirrhosis can often be confirmed noninvasively by a combination of serum biomarkers, imaging techniques, and measurements of liver stiffness.

Physical Examination

On physical examination, stigmata of cirrhosis consist of muscle atrophy, mainly involving the bitemporal muscle regions and the thenar and hypothenar eminences; spider angiomas, mostly on the trunk, face, and upper limbs; and palmar erythema involving the thenar and the hypothenar eminences and the tips of the fingers. Although muscle atrophy is a marker of liver insufficiency, spider angiomas and palmar erythema are markers of vasodilation and a hyperdynamic circulation. Men may have hair loss on the chest and abdomen,

994

CHAPTER 144  Cirrhosis and Its Sequelae  

gynecomastia, and testicular atrophy. Petechiae and ecchymoses may be present as a result of thrombocytopenia or a prolonged prothrombin time. Dupuytren contracture, which is a thickening of the palmar fascia, occurs mostly in alcoholic cirrhosis. A pathognomonic feature of cirrhosis is the finding on abdominal examination of a small right liver lobe, with a span of less than 7 cm on percussion, and a palpable left lobe that is nodular with increased consistency. Splenomegaly may also be present and is indicative of portal hypertension. Collateral circulation on the abdominal wall (caput medusae) may also develop as a consequence of portal hypertension. Absence of any of these physical findings does not exclude cirrhosis.

Laboratory Tests

Laboratory test results suggestive of cirrhosis include even subtle abnormalities in serum levels of albumin or bilirubin or elevation of the international normalized ratio. The most sensitive and specific laboratory finding suggestive of cirrhosis in the setting of chronic liver disease is a low platelet count (1.1 g/dL) in patients in whom the source of ascites is the hepatic sinusoid (e.g., cirrhosis or cardiac ascites). Protein levels in ascitic fluid are an indirect marker of the integrity of the hepatic sinusoids: normal sinusoids are permeable structures that “leak” protein, whereas sinusoids in cirrhosis are “capillarized” and do not leak as much protein. The three main causes of ascites—cirrhosis, peritoneal malignant disease or tuberculosis, and heart failure—can easily be distinguished by combining the results of both the serum-ascites albumin gradient and ascites total protein content. Cirrhotic ascites typically has a high serum-ascites albumin gradient and low protein, cardiac ascites has a high serum-ascites albumin gradient and high protein, and ascites secondary to peritoneal malignant disease typically has a low serum-ascites albumin gradient and high protein. A high serum B-type natriuretic peptide has a high diagnostic accuracy in the diagnosis of ascites due to heart failure.

Hepatorenal Syndrome

Hepatorenal syndrome is a diagnosis of exclusion, but patients typically have ascites that is unresponsive to diuretics and often have hyponatremia. The differential diagnosis includes conditions that worsen vasodilation, such as sepsis, the use of vasodilators, and large-volume paracentesis not accompanied by albumin infusion; conditions that decrease effective arterial blood volume, such as gastrointestinal hemorrhage, overdiuresis, or diarrhea (often induced by overdoses of lactulose); conditions that induce renal vasoconstriction, such as nonsteroidal anti-inflammatory drugs; and nephrotoxic insults, such as from aminoglycosides. As a result, the diagnosis of hepatorenal syndrome can only be made after diuretics have been discontinued, any condition that leads to worsening of the hemodynamic status of the cirrhotic patient has been excluded or treated, and intravascular volume has been expanded with albumin.

Spontaneous Bacterial Peritonitis

A high index of suspicion and early diagnosis are key in the management of spontaneous bacterial peritonitis. Diagnostic paracentesis should be performed in any patient with symptoms or signs of spontaneous bacterial peritonitis, including unexplained encephalopathy and renal dysfunction. Because spontaneous bacterial peritonitis is often asymptomatic and frequently community acquired, diagnostic paracentesis should be performed when any patient with cirrhosis is admitted to the hospital, regardless of the cause for admission. The diagnosis of spontaneous bacterial peritonitis is established by an ascitic fluid PMN count greater than 250/µL. Bacteria can be isolated from ascitic fluid in only 40 to 50% of cases, even with sensitive methods such as inoculation directly into a blood culture bottle. Spontaneous bacterial peritonitis is mostly a monobacterial infection, usually with a gram-negative enteric organism. However, the widespread use of antibiotic prophylaxis in cirrhosis has led to an increased prevalence of infections with multidrug-resistant organisms. Anaerobes and fungi very rarely cause spontaneous bacterial peritonitis; their presence, as well as a polymicrobial infection, should raise suspicion of secondary bacterial peritonitis.

Hepatic Encephalopathy

The diagnosis of overt hepatic encephalopathy is mainly clinical and based on a history and physical examination that shows alterations in consciousness and behavior as well as the presence of asterixis. Ammonia levels are unreliable, and there is poor correlation between the grade of hepatic encephalopathy and ammonia blood levels. High levels (>150 µmol/L) are, however, indicative of hepatic encephalopathy and can be useful in the evaluation of a patient with neurocognitive disturbances of unknown origin. Psychometric tests and an electroencephalogram, which typically shows generalized slow waves and the presence of triphasic waves, are commonly used in research but are not generally used for clinical diagnosis. Minimal or subclinical hepatic encephalopathy, which is present in up to 80% of patients with cirrhosis, is diagnosed solely on the basis of abnormal results of psychometric and neuropsychological tests of attention (e.g., number connection test, digit symbol test) and psychomotor function (e.g., grooved pegboard). Screening of cirrhotic patients for minimal hepatic encephalopathy is not widely recommended because diagnostic tests are not standardized and the benefits of treatment are uncertain.

Hepatopulmonary Syndrome and Portopulmonary Hypertension

The diagnostic criteria for hepatopulmonary syndrome are arterial hypoxemia with a Pao2 of less than 80 mm Hg or an alveolar arterial oxygen gradient of greater than 15 mm Hg, along with evidence of pulmonary vascular shunting on contrast echocardiography (Chapter 49) or a 99mTc-labeled macroaggregated albumin scan demonstrating abnormal shunting of radioactivity to the brain.

995

Portopulmonary hypertension is diagnosed by the presence of mean pulmonary arterial pressure higher than 25 mm Hg on right-sided heart catheterization, provided pulmonary capillary wedge pressure is less than 15 mm Hg.

TREATMENT  Treatment of cirrhosis should ideally be aimed at interruption or reversal of fibrosis. Although antifibrotic drugs have not been shown to reverse fibrosis consistently or to improve outcomes in cirrhotic patients, eradication of the hepatitis C or the hepatitis B virus has been associated with regression of fibrosis. Treatment of compensated cirrhosis is currently directed at preventing the development of decompensation by (1) treating the underlying liver disease (e.g., antiviral therapy for hepatitis C or B) to reduce fibrosis and to prevent decompensation; (2) avoiding factors that could worsen liver disease, such as alcohol, hepatotoxic drugs, and superimposed viral infections; and (3) screening for varices (to prevent variceal hemorrhage) and for hepatocellular carcinoma (to treat it at an early stage) (Fig. 144-6).6 Probiotics may reduce the risk of hospitalization for hepatic encephalopathy and the clinical worsening of advanced but stable cirrhosis. A2  Treatment of decompensated cirrhosis focuses on specific decompensating events and the option of liver transplantation. Increasingly, data reveal that different therapies for the same complication may be applicable to patients with different risk profiles, mainly based on severity of the disease (see Table 144-2).

Varices and Variceal Bleeding

Reducing portal pressure decreases the risk for the development of varices and variceal hemorrhage as well as the risk for ascites and death.7 Nonselective β-adrenergic blockers (propranolol, nadolol) reduce portal pressure by producing splanchnic vasoconstriction and decreasing portal venous inflow. In patients with cirrhosis and medium or large varices that have never bled, nonselective β-blockers significantly reduce the risk for first variceal hemorrhage. A3  Treatment options have included propranolol (initiated at a dose of 20 mg orally twice a day) and nadolol (initiated at a dose of 20 mg orally every day), with the dose titrated to produce a resting heart rate of about 50 to 55 beats per minute, provided that the systolic blood pressure does not decrease below 90 mm Hg. A3  β-Adrenergic blockers can also lower the risk of developing ascites. Endoscopic variceal ligation (see Fig. 125-3), a therapy that aims to obliterate varices by placing rubber rings on variceal columns, is at least as useful as traditional nonselective β-blockers to prevent a first variceal hemorrhage. Ligation has no effect on portal pressure and can lead to hemorrhage from ligationinduced ulcers. Carvedilol (a nonselective β-blocker with vasodilating properties at a dose of 12.5 mg/day) has been shown to be superior to ligation and to lower portal pressure in patients who do not respond to propranolol A4 ; however, its use in patients with ascites may not be advisable because of its vasodilating effect. A rational approach is to start therapy with propranolol or nadolol and to use ligation in patients who cannot tolerate or have contraindications to β-blockers. In the compensated patient, carvedilol could be used in patients who cannot tolerate propranolol or nadolol. Screening endoscopy should be repeated every 2 to 3 years in patients with no varices, every 1 to 2 years in patients with small varices, and sooner in patients with decompensated disease so that effective therapy can be instituted before the varices grow in size and bleed. Patients with cirrhosis and variceal hemorrhage require resuscitation in an intensive care unit. However, overtransfusion should be avoided because it can precipitate rebleeding. A5  Hemoglobin values should be maintained at about 8 g/dL. Prophylactic antibiotics should be used in this setting not only to prevent bacterial infections but also to decrease rebleeding and death. The recommended antibiotic is oral norfloxacin at a dose of 400 mg twice daily for 5 to 7 days, although intravenous ceftriaxone at a dose of 1 g/day for 5 to 7 days is preferable in patients with advanced liver disease (malnutrition, ascites, encephalopathy, and jaundice) or in patients who already are receiving norfloxacin prophylaxis. The most effective specific therapy for the control of active variceal hemorrhage is the combination of a vasoconstrictor with endoscopic therapy. Safe vasoconstrictors include terlipressin, somatostatin, and the somatostatin analogues octreotide and vapreotide; they can be initiated at admission to the hospital and continued for 2 to 5 days. The vasoconstrictor currently available in the United States is octreotide, which is used as a 50-µg intravenous bolus followed by an infusion at 50 µg/hour. The preferred endoscopic therapy to control variceal hemorrhage is esophageal variceal ligation. In cases in which bleeding continues despite vasoconstrictors and ligation, an esophageal stent appears to be more effective as a bridge to a more definitive therapy and is safer than balloon tamponade, although it may not reduce mortality. A6  The transjugular intrahepatic portosystemic shunt (TIPS) is generally recommended for patients who fail to respond to standard therapy (Fig. 144-7). In patients at high risk of failure, Child C (score 10 to 13) patients and Child B patients who have actively bleeding varices at endoscopy, preemptive TIPS placement (24 to 48 hours after admission) is associated with a reduced failure rate and a significant improvement in survival. Therefore, preemptive “early”

996

CHAPTER 144  Cirrhosis and Its Sequelae  

Cirrhosis

Compensated

Decompensated

HCC surveillance US, AFP q6mos

Varices surveillance

No varices Repeat EGD in 3 years

Small varices Repeat EGD in 2 years

Variceal Hemorrhage • Early vasoconstrictors + endoscopic therapy (ligation) • Do not overtransfuse • Short-term antibiotic prophylaxis • Prevent rebleeding with propranolol + ligation

Monitor INR, albumin, bilirubin

Monitor creatinine

Large varices • Nonselective β-blockers • Ligation in those intolerant

Ascites • Salt restriction • Diuretics (spironolactone-based) • Avoid NSAIDs • No water restriction unless serum Na 30 mg/dL or creatinine >1 mg/dL, or both) or serum bilirubin concentration higher than 4 mg/dL at the time of diagnosis. Albumin at a dose of 1.5 g per kilogram of body weight at diagnosis, repeated on the third day at an intravenous dose of 1 g per kilogram of body weight, can reduce the risk of renal failure and acute mortality. However, this dosing is empirical, should not exceed 100 g per dose, and does not reduce 3-month mortality in cirrhotic patients with infections other than spontaneous bacterial peritonitis. A13 

997

The administration of nonabsorbable (or poorly absorbable) antibiotics can prevent the development of spontaneous bacterial peritonitis and other infections in cirrhosis by selectively eliminating gram-negative organisms in the gut. However, the widespread use of prophylactic norfloxacin is associated with a higher rate of infections by antibiotic-resistant organisms. Long-term antibiotic prophylaxis with oral norfloxacin at a dose of 400 mg/day may be justified only in two groups: patients who have recovered from a previous episode of spontaneous bacterial peritonitis and patients who have an ascites protein level of less than 1 g/L with advanced liver and circulatory dysfunction as evidenced by the presence of jaundice, hyponatremia, or renal dysfunction.

Hepatic Encephalopathy

Treatment of overt hepatic encephalopathy starts by exclusion of alternative causes of altered mental status. Once the diagnosis of hepatic encephalopathy has been made, treatment involves identifying and treating the precipitating factor and reducing the ammonia level.10 Precipitating factors include infections, overdiuresis, gastrointestinal bleeding, high oral protein load, and constipation. Narcotics and sedatives contribute to hepatic encephalopathy by directly depressing brain function. TIPS is a common precipitant of hepatic encephalopathy, and shunt reduction or occlusion may be required. Among agents aimed at decreasing ammonia production in the gut, lactulose (15 to 30 mL orally twice daily adjusted to obtain two or three soft bowel movements per day) is the first choice for the treatment of episodic overt encephalopathy.11 Polyethylene glycol 3350-electrolyte solution (4 L orally or by nasogastric tube over 4 hours), however, may lead to a more rapid clinical response. Other agents include orally administered nonabsorbable antibiotics, such as rifaximin (550 mg two times per day), neomycin (500 mg to 1 g three times per day), and metronidazole (250 mg two to four times per day). Drugs that may increase ammonia fixation in the liver, such as L-ornithine-L-aspartate, benzoate, and glycerol phenylbutyrate, are being studied. Once an episode of overt encephalopathy has resolved, secondary prophylaxis with lactulose is recommended. If a precipitating factor has been identified and is well controlled or when liver function or nutritional status has improved, prophylactic therapy may be discontinued. In patients with recurrent encephalopathy, rifaximin together with lactulose is useful in preventing further recurrence. Switching dietary protein from an animal source to a vegetable source may be beneficial in recurrent or persistent encephalopathy, but protein restriction is not necessary and should not be used chronically. Experimental approaches can engineer gut microbiota to reduce urease activity in mice, but whether this approach can be extended to humans is unknown.12

Pulmonary Complications

Hepatopulmonary syndrome rarely resolves spontaneously, and medical therapy is disappointing.13 TIPS is not generally recommended. The only viable treatment is liver transplantation (Chapter 145). By comparison, liver transplantation is indicated only in a subset of patients with portopulmonary hypertension. In fact, a mean pulmonary arterial pressure higher than 45 mm Hg is an absolute contraindication to liver transplantation. The use of vasodilators should be considered in these patients (Chapter 75).

Surgical Therapy

Liver Transplantation

Orthotopic liver transplantation (Chapter 145), which is the definitive therapy for cirrhosis, is indicated when the risk for dying of liver disease is greater than the risk for dying of transplantation, as determined by a Child-Pugh score of 7 or higher (see Table 144-2) or a Model for End-Stage Liver Disease (MELD) score of 15 or higher. MELD and MELD-Na (see Table 144-2), which estimate the risk for 3-month mortality, are used to determine the priority for liver transplantation. The number of available deceased donor organs is lower than the number of patients awaiting liver transplantation; as a result, 15 to 20% of patients awaiting liver transplantation in the United States die before an organ becomes available.



PRIMARY PREVENTION

Treatment of the underlying liver disease, before the development of cirrhosis, is a primary prevention strategy. Because the major causes of cirrhosis are related to lifestyle choices such as injection drug use (Chapter 31), alcohol consumption (Chapter 30), obesity, and unprotected sex, primary prevention programs that focus on encouraging alcohol abstinence, reducing high-risk behavior for hepatitis virus infection, weight reduction, and vaccination for hepatitis B are even better prevention strategies.  

PROGNOSIS

The outcome of cirrhosis depends on the patient’s stage. Patients with compensated cirrhosis die of liver disease only after transition to a decompensated stage. The 10-year survival rate of patients who remain in a compensated stage

is approximately 90%, whereas their likelihood of decompensation is 50% at 10 years. Inception cohort studies of patients with compensated cirrhosis show a median survival of all patients, including those in whom decompensation develops over time, of about 10 years. Overall, the median survival after decompensation is about 2 years. Survival is even lower in patients with refractory ascites, hyponatremia, or recurrent variceal hemorrhage and is lowest in patients who are hospitalized with an acute decompensating event, in whom 28-day mortality is about 30% and correlates with the number of organ failures present. Hepatocellular carcinoma develops at a fairly constant rate of 3% per year and is associated with a worse outcome at whatever stage it develops. Predictors of survival are somewhat different in compensated and decompensated patients. Parameters of portal hypertension (varices, splenomegaly, platelet count, γ-globulin) assume greater importance in compensated patients, whereas renal dysfunction, hemorrhage, and hepatocellular carcinoma are important predictive factors in patients with decompensated cirrhosis.14 In clinical practice, the Child-Pugh score is applicable to all cirrhotic patients, and the MELD score is used in decompensated patients to determine priority for liver transplantation (see Table 142-2).

  Grade A References A1. Villanueva C, Graupera I, Aracil C, et al. A randomized trial to assess whether portal pressure guided therapy to prevent variceal rebleeding improves survival in cirrhosis. Hepatology. 2017;65:1693-1707. A2. Viramontes Horner D, Avery A, Stow R. The effects of probiotics and symbiotics on risk factors for hepatic encephalopathy: a systematic review. J Clin Gastroenterol. 2017;51:312-323. A3. Kerbert AJ, Chiang FW, van der Werf M, et al. Hemodynamic response to primary prophylactic therapy with nonselective beta-blockers is related to a reduction of first variceal bleeding risk in liver cirrhosis: a meta-analysis. Eur J Gastroenterol Hepatol. 2017;29:380-387. A4. Sinagra E, Perricone G, D’Amico M, et al. Systematic review with meta-analysis: the haemodynamic effects of carvedilol compared with propranolol for portal hypertension in cirrhosis. Aliment Pharmacol Ther. 2014;39:557-568. A5. Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013;368:11-21. A6. Escorsell A, Pavel O, Cardenas A, et al. Esophageal balloon tamponade versus esophageal stent in controlling acute refractory variceal bleeding: a multicenter randomized, controlled trial. Hepatology. 2016;63:1957-1967. A7. Albillos A, Zamora J, Martinez J, et al. Stratifying risk in the prevention of recurrent variceal hemorrhage: results of an individual patient meta-analysis. Hepatology. 2017;66:1219-1231. A8. Holster IL, Tjwa ET, Moelker A, et al. Covered transjugular intrahepatic portosystemic shunt versus endoscopic therapy + beta-blocker for prevention of variceal rebleeding. Hepatology. 2016;63:581-589. A9. Caraceni P, Riggio O, Angeli P, et al. Long-term albumin administration in decompensated cirrhosis (ANSWER): an open-label randomised trial. Lancet. 2018;391:2417-2429. A10. Bureau C, Thabut D, Oberti F, et al. Transjugular intrahepatic portosystemic shunts with covered stents increase transplant-free survival of patients with cirrhosis and recurrent ascites. Gastroenterology. 2017;152:157-163. A11. Bureau C, Adebayo D, Chalret de Rieu M, et al. Alfapump® system vs. large volume paracentesis for refractory ascites: a multicenter randomized controlled study. J Hepatol. 2017;67:940-949. A12. Sridharan K, Sivaramakrishnan G. Vasoactive agents for hepatorenal syndrome: a mixed treatment comparison network meta-analysis and trial sequential analysis of randomized clinical trials. J Gen Intern Med. 2018;33:97-102. A13. Thévenot T, Bureau C, Oberti F, et al. Effect of albumin in cirrhotic patients with infection other than spontaneous bacterial peritonitis. A randomized trial. J Hepatol. 2015;62:822-830.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 144  Cirrhosis and Its Sequelae  

GENERAL REFERENCES 1. Hadjihambi A, Arias N, Sheikh M, et al. Hepatic encephalopathy: a critical current review. Hepatol Int. 2018;12:135-147. 2. D’Amico G, Morabito A, D’Amico M, et al. Clinical states of cirrhosis and competing risks. J Hepatol. 2018;68:563-576. 3. Bosch J, Iwakiri Y. The portal hypertension syndrome: etiology, classification, relevance, and animal models. Hepatol Int. 2018;12:1-10. 4. Zardi EM, Zardi DM, Giorgi C, et al. Portopulmonary hypertension and hepatorenal syndrome. Two faces of the same coin. Eur J Intern Med. 2017;43:22-27. 5. Na HY, Kim JH, Choe WH, et al. Clinical features of spontaneous bacterial peritonitis: a 10-year experience from a single center. Korean J Gastroenterol. 2017;69:129-134. 6. Ge PS, Runyon BA. Treatment of patients with cirrhosis. N Engl J Med. 2016;375:767-777. 7. Garcia-Tsao G, Abraldes JG, Berzigotti A, et al. Portal hypertensive bleeding in cirrhosis: risk stratification, diagnosis, and management: 2016 practice guidance by the American Association for the Study of Liver Diseases. Hepatology. 2017;65:310-335.

998.e1

8. Conejo I, Guardascione MA, Tandon P, et al. Multicenter external validation of risk stratification criteria for patients with variceal bleeding. Clin Gastroenterol Hepatol. 2018;16:132-139. 9. Amin AA, Alabsawy EI, Jalan R, et al. Epidemiology, pathophysiology, and management of hepatorenal syndrome. Semin Nephrol. 2019;39:17-30. 10. Wijdicks EF. Hepatic encephalopathy. N Engl J Med. 2016;375:1660-1670. 11. Kornerup LS, Gluud LL, Vilstrup H, et al. Update on the therapeutic management of hepatic encephalopathy. Curr Gastroenterol Rep. 2018;20:1-6. 12. Nardelli S, Ridola L, Gioia S, et al. Management of hepatic encephalopathy not responsive to first-line treatments. Curr Treat Options Gastroenterol. 2018;16:253-259. 13. Soulaidopoulos S, Cholongitas E, Giannakoulas G, et al. Review article: update on current and emergent data on hepatopulmonary syndrome. World J Gastroenterol. 2018;24:1285-1298. 14. Haj M, Rockey DC. Predictors of clinical outcomes in cirrhosis patients. Curr Opin Gastroenterol. 2018;34:266-271.

998.e2

CHAPTER 144  Cirrhosis and Its Sequelae  

REVIEW QUESTIONS 1. A 55-year-old man with chronic hepatitis C complains of fatigue. On physical examination, he has vascular spiders, palmar erythema, and a palpable left lobe of his liver. He has no ascites or asterixis. Laboratory analysis demonstrates aspartate aminotransferase 100, alanine aminotransferase 67, alkaline phosphatase 145, and platelet count of 120,000. Abdominal computed tomography (CT) scan shows a nodular liver contour, portosystemic collaterals, but no masses. Which of the following would you recommend as the next step? A . Liver biopsy B. Upper endoscopy C. Start nonselective β-blockers D. Start spironolactone E. Magnetic resonance imaging of the liver Answer: B  Although the gold standard in the diagnosis of cirrhosis is liver biopsy, this procedure is not necessary when clinical, laboratory, and imaging results are sufficient to establish the diagnosis. In the setting of chronic liver disease, a palpable left lobe of the liver is almost diagnostic of cirrhosis. The most sensitive and specific laboratory finding suggestive of cirrhosis in the setting of chronic liver disease is a low platelet count (1.1 g/dL), indicating ascites that is secondary to hepatic sinusoidal hypertension. The fluid also has a high protein content (>2.5 g/dL) indicative of normal “leaky” sinusoids, so the problem is unlikely to be in the liver. A liver biopsy would not be the most indicated test. In this clinical setting, the most likely possibility is right-sided heart failure, which should be evaluated, initially with an echocardiogram. A high serum B-type natriuretic peptide has a high diagnostic accuracy in the diagnosis of ascites due to heart failure. Although the ascites total white blood cell count is somewhat elevated, the fluid contains only 77 neutrophils (640 × 0.12), and therefore it is not consistent with bacterial peritonitis, which is based on more than 250 neutrophils/µL. In peritoneal causes of ascites (carcinomatosis, tuberculosis), the ascites protein is high and the serum-to-ascites albumin gradient is low (45 mm Hg despite pharmacologic interventions) Active, incurable malignant disease Metastatic nonhepatic malignant disease Hepatoma with macrovascular invasion, extrahepatic metastases, or exceeding Milan or UCSF criteria and not able to be downstaged Cholangiocarcinoma with percutaneous transperitoneal biopsy (likely seeding from the biopsy), tumor diameter >3 cm, or extrahepatic spread Active, uncontrolled, and untreatable sepsis or other serious infectious disease Active HIV infection with AIDS-defining illness, high-titer HIV RNA, very low CD4 count, or unresponsive to HAART Anatomic anomaly or extensive vascular thromboses precluding hepatic transplantation AIDS = acquired immunodeficiency syndrome; HAART = highly active antiretroviral therapy; HIV = human immunodeficiency virus; PAP = pulmonary artery pressure; UCSF = University of California, San Francisco.

donors or living donors, three fundamental questions must be addressed at the time of evaluation.2,3 1. Is liver transplantation indicated? The patient should have liver failure, complications of liver disease, or a metabolic condition that is best treated by liver transplantation and for which there are no alternative treatments. 2. Are there contraindications to transplantation (Table 145-2)? Comorbid conditions that could severely compromise graft or patient outcome must be identified. 3. Will the patient be able to tolerate and comply with immunosuppression and post-transplantation management? Inability to comply with the rigors of post-transplantation care and management could lead to graft loss and recipient death.  

Medical Assessment

Certain components of the medical evaluation apply to all potential recipients. Viral serologies characterize the status of ongoing infection, prior exposure, or vaccination related to the hepatitis viruses (Chapter 140), Epstein-Barr virus (Chapter 353), cytomegalovirus (Chapter 352), and human immunodeficiency virus (HIV; Chapter 364). Colonoscopy is recommended for all patients older than 50 years to exclude colon cancer or polyps (Chapter 184) and in all patients with primary sclerosing cholangitis (Chapter 146) to detect possible ulcerative colitis or Crohn disease (Chapter 132). Cardiopulmonary assessment (Chapters 45, 62, and 79) is performed in patients older than 50 years and in any patient with risk factors for coronary artery disease, such as hypertension, hypercholesterolemia, a significant family history, cigarette smoking, or diabetes. Cardiac catheterization (Chapter 51) is performed in patients with positive cardiac stress test results to confirm and to delineate the extent of coronary disease. Ultrasonography, computed tomography, and magnetic resonance imaging are useful to examine the biliary tract (Chapter 138), to exclude hepatocellular cancer (Chapter 186) or cholangiocarcinoma (Chapter 186), and to determine the patency of major vascular structures, such as the hepatic artery, celiac axis, splenic artery, portal vein, and superior mesenteric vein (Chapter 134). The primary principle for consideration of liver transplantation is that a patient’s predicted survival with transplantation must exceed expected survival without transplantation. For deceased donor liver transplantation, this principle is fulfilled by a Model for End-Stage Liver Disease (MELD) score of approximately 15 (see Table 144-2 in Chapter 144). In addition, certain complications of cirrhosis (Chapter 144), such as ascites, spontaneous bacterial peritonitis, encephalopathy, uncontrolled variceal hemorrhage, nutritional wasting, failure to thrive, and development of hepatoma (Chapter 186), shorten survival. Immediate evaluation and listing for transplantation are indicated for any patient with a MELD score of 15 or more or any of these complications. Other complications that are not considered life-threatening may also be indications for consideration of liver transplantation. Examples include metabolic bone disease resulting in osteopenia and bone fractures (Chapters 230

999

and 231), inadequate nutrition, muscle wasting, severe fatigue, poor concentrating ability, and intractable pruritus. Because the supply of donor livers is limited, transplant centers must select patients who have the greatest chance for a successful outcome. Elderly, obese, and deconditioned patients as well as patients with underlying vascular disease or long-standing diabetes mellitus are poor candidates for liver transplantation.  

Surgical Assessment

The three phases of surgery are native liver dissection, the anhepatic phase, and revascularization of the graft. Native liver dissection, which is characterized by meticulous dissection and prompt control of bleeding vessels, can be complicated by morbid obesity, significant portal hypertension, portal thromboses, prior portosystemic shunt surgery, or prior abdominal surgery. The length of this phase is usually 1 to 2 hours, with blood losses ranging from 0 to 5 units of blood. During the anhepatic phase, which usually lasts 1.5 to 3 hours and is associated with blood loss ranging from 0 to 5 units of blood, the vascular supply of the liver is completely interrupted, and the native liver is excised. Toward the end of the anhepatic phase, vessels are anastomosed, with the donor hepatic veins typically grafted to the recipient vena cava with only one caval anastomosis. Recipients with severe coagulopathy, cachexia, and nutritional deficiencies may be prone to excessive bleeding and metabolic complications. When the venous clamps are removed, patients are at risk for primary fibrinolysis or consumptive coagulopathy. The arterial anastomosis is typically performed after unclamping to shorten the warm ischemia period.  

Psychosocial Assessment

A patient referred for liver transplantation may have a history of alcohol, drug, or substance abuse. Psychosocial evaluation is important to help determine the risk for recidivism after transplantation. For patients with alcoholic cirrhosis, nearly all transplant centers require a minimum 6-month period of documented abstinence and enrollment in education or rehabilitation programs. A patient with severe alcoholic hepatitis, poor prognosis, adequate social support, and willingness to comply with alcohol abstinence and rehabilitation might be an exception to the 6-month guideline. Other key aspects of the psychosocial assessment include (1) evaluating the patient for any underlying psychiatric illness (Chapter 369), its severity, and how it should be managed; (2) defining and establishing social and psychological support for the potential recipient, including family, friends, and significant others; and (3) in living donor liver transplantation, evaluating the potential living donors.  

Assigning Priority: Model for End-Stage Liver Disease

The prioritization of U.S. liver transplantation candidates and the allocation of deceased donor livers are based on the MELD or MELD-Na score (see Table 144-2 in Chapter 144), which ranges from 6 (best prognosis) to 40 (worst prognosis). The 3-month mortality rate increases from 10% with a MELD score of 20 to 60% at a MELD score of 35, and essentially to 100% with a MELD score above 40. A patient’s survival is not improved by deceased donor liver transplantation if the MELD score is below about 15. As a result, deceased donor livers must be made available more broadly if no local patient has a MELD score of 15 or higher. In addition, the SHARE 35 rule allows for regional sharing of a donor liver to a regional patient with a score of 35 or higher. Patients with acute liver failure or post-transplantation patients with graft failure due to hepatic artery thrombosis have a MELD score of 40 and top priority status. This allocation system has improved transplantation rates and has not altered waiting list or post-transplantation mortality. In patients with cirrhosis (Chapter 144), hyponatremia (Chapter 108) is associated with hepatorenal syndrome, ascites, and death from liver disease. Hyponatremia is an independent predictor of mortality up to a MELD score of 30. As a result, the MELD score is now adjusted for serum sodium concentration in allocating deceased donor livers (MELD-Na). Other exceptions can include patients with the hepatorenal syndrome (Chapter 144), the portopulmonary syndrome (Chapter 144), cholangiocarcinoma (Chapter 146), cystic fibrosis (Chapters 83 and 137), familial amyloid polyneuropathy (Chapter 179), and primary hyperoxaluria (Chapter 194) (E-Table 145-1).

  DISEASE-SPECIFIC INDICATIONS FOR TRANSPLANTATION



Acute Liver Failure

Acute liver failure is defined as hepatic injury of fewer than 26 weeks in duration, an international normalized ratio of 1.5 or greater, and altered mental

CHAPTER 145  Hepatic Failure and Liver Transplantation  

E-TABLE 145-1 LIVER CANDIDATES WITH EXCEPTIONAL CASES NOT REQUIRING REVIEW BY THE REGIONAL REVIEW BOARD IN THE UNITED STATES Hepatopulmonary syndrome. The patient must have evidence of intrapulmonary shunt and Pao2 below 60 mm Hg on room air. These patients are listed with MELD 22 with 10% increase every 3 months if the Pao2 remains under 60 mm Hg. Cholangiocarcinoma. The liver center requesting the MELD exception must have a written United Network for Organ Sharing–approved protocol that includes neoadjuvant chemoradiation and pretransplantation operative staging to exclude metastatic disease. MELD 22 with 10% increase every 3 months as long as tumor is controlled by the treatment protocol and there is no evidence of metastases. Cystic fibrosis. The patient must have FEV1 below 40% of normal range. MELD 22 with 10% increase every 3 months. Familial amyloid polyneuropathy. The patient must lack significant cardiac involvement with echocardiogram showing ejection fraction greater than 40% and have ambulatory status, appropriate TTR gene mutation, and biopsy-proven amyloid. MELD 22 with 10% increase every 3 months. Primary hyperoxaluria (type I). The patient must have renal failure due to oxalate stone disease and alanine:glyoxylate aminotransferase deficiency proven by liver biopsy (sample analysis with measurement of alanine:glyoxylate aminotransferase enzyme activity and/or genetic analysis of the alanine:glyoxylate aminotransferase gene). MELD 28 with 10% increase every 3 months. Portopulmonary syndrome. The patient must have pulmonary hypertension diagnosed by appropriate measurements of mean pulmonary artery pressure (MPAP), pulmonary vascular resistance (PVR), and transpulmonary gradient. With treatment, the MPAP should be less than 35 mm Hg and PVR should be less than 400 dynes/sec/cm−5. MELD 22 with 10% increase every 3 months. MELD = Model for End-Stage Liver Disease.

999.e1

1000

CHAPTER 145  Hepatic Failure and Liver Transplantation  

status in the absence of chronic liver disease except for Wilson disease (Chapter 200), vertically acquired hepatitis B, or autoimmune hepatitis. The main causes of acute liver failure in the United States are acetaminophen toxicity (Chapter 141), dietary supplements, herbal medications (Chapter 141), acute viral hepatitis (Chapter 139), autoimmune hepatitis (Chapter 140), Wilson disease, mushroom poisoning (Chapters 102 and 141), acute hepatic ischemia (Chapter 134), Budd-Chiari syndrome, acute fatty liver of pregnancy (sometimes associated with the HELLP syndrome—hemolysis, elevated liver enzymes, low platelets; Chapters 151 and 226), and malignant infiltration of the liver (Chapter 186). However, about 5% of cases are of uncertain cause despite comprehensive assessment.4 Severe acute liver injury without hepatic encephalopathy is less likely to progress to liver failure requiring transplantation if it is related to acetaminophen toxicity compared with other causes.5

Patients with acute liver failure have a low systemic vascular resistance and a low mean arterial pressure. In patients with elevated ICP, the decrease in mean arterial pressure may further compromise cerebral perfusion pressure and blood flow. A goal of vasopressor therapy is to maintain cerebral perfusion pressure from 60 to 80 mm Hg. Treatment of hypotension to maintain a mean arterial pressure above 75 mm Hg may include dextrose-containing crystalloid, albumin, norepinephrine, and vasopressin (Chapter 98). Careful monitoring of volume status is critical, and care should be taken to avoid volume overload, which may worsen cerebral edema. Acute renal failure (Chapter 112) is common in patients with acute liver failure, particularly when it is caused by acetaminophen, mushroom poisoning, or Wilson disease. Renal replacement therapy, when required, should be administered as continuous venovenous dialysis (Chapter 122). Intermittent modes of hemodialysis should be avoided because of their adverse hemodynamic effects. In acute liver failure, hypoglycemia is a common complication that should be avoided and managed by continuous glucose infusions. Supplementation with phosphate, magnesium, and potassium (Chapter 109) may be required. Enteral feedings (Chapter 204) should be initiated early, and parenteral nutrition (Chapter 204) should be started if enteral feedings are contraindicated.

TREATMENT  Patients with acute liver failure criteria should be transferred to the intensive care unit for enhanced monitoring, and a liver transplant center should be contacted for potential transfer of the patient.6 Specific treatments include N-acetylcysteine for acetaminophen hepatotoxicity (Chapters 102 and 141), N-acetylcysteine plus either silibinin or penicillin G for mushroom poisoning (Chapters 102 and 141), N-acetylcysteine and removal of the offending drug for drug-induced liver injury (Chapter 141), nucleos(t)ide analogues for hepatitis B (Chapter 140), acyclovir for herpes or varicella-zoster hepatitis (Chapters 350 and 351), dialysis and copper chelation for Wilson disease (Chapter 200), corticosteroids for autoimmune hepatitis (Chapter 140), delivery of the infant for acute fatty liver of pregnancy, and transjugular intrahepatic portosystemic shunting for the Budd-Chiari syndrome. In addition, all causes of acute liver failure can cause a range of systemic complications. A key issue in the general management of patients with acute liver failure is monitoring and treatment of encephalopathy. Progressive encephalopathy in acute liver failure is characteristically associated with cerebral edema and an elevated intracranial pressure (ICP). Elevated ICP is rare in patients with grade I (mild confusion) or grade II (agitated state) encephalopathy but occurs in 25 to 35% of patients with grade III encephalopathy (stuporous) and 65 to 75% of patients with grade IV (coma, but responsive to deep pain) encephalopathy. Irreversible brain injury is likely when the ICP is above 50 mm Hg and the cerebral perfusion pressure (mean arterial pressure minus ICP) is below 40 mm Hg for more than 2 hours. Liver transplantation may be contraindicated under these circumstances. Because ammonia may play a role in the pathogenesis of cerebral edema, lactulose (20 g every 4 to 6 hours), A1  rifaximin (550 mg twice daily), or both are recommended, especially for lower grades of encephalopathy.7 Patients with grade III/IV encephalopathy require intubation, mechanical ventilation, elevation of the head of the bed, sedation (e.g., propofol, initial dose of 0.005 mg/kg/ minute intravenously [IV], maintenance dose of 0.005 to 0.05 mg/kg/minute IV, which can be increased in increments of 0.005 mg/kg/minute every 5 minutes), and paralysis (cisatracurium, initial dose of 0.1 to 0.2 mg/kg IV, maintenance dose of 1 to 3 µg/kg/minute IV). Seizures (Chapter 375) should be treated with phenytoin (initial dose of 15 mg/kg IV, maintenance dose of 3 mg/kg every 12 hours IV), propofol (1 mg/kg IV loading dose, 3 to 7 mg/kg/hour IV maintenance), pentobarbital (13 mg/kg IV loading dose, 2 to 3 mg/kg/hour IV maintenance), or midazolam (0.2 mg/kg IV loading dose, 0.1 to 0.25 mg/kg/hour IV maintenance). Levetiracetam (500 [up to 1500] mg every 12 hours IV) may be used if renal function is preserved. Placement of an intracranial transducer for monitoring of ICP is desirable, but severe underlying coagulopathy may prohibit its placement. Elevated ICP may require treatment with mannitol (0.5 to 1.0 mg/kg every 4 to 6 hours IV), hyperventilation (target pH 7.45), pentobarbital (dose described before), and hypothermia (32° C to 34° C). The goal for these interventions is to maintain ICP below 20 mm Hg. Corticosteroids do not improve cerebral edema or lower ICP in acute liver failure, and their use is not recommended. Patients with acute liver failure are at increased risk for bacterial and fungal infections. Common infections include line sepsis, pneumonia (Chapter 91), and urinary tract infections (Chapter 268). Prolonged ventilation, dialysis, invasive procedures, and use of multiple antibiotics increase the risk for fungal infection (Chapter 266). Surveillance cultures of blood, urine, and sputum and periodic chest radiography are required every 2 days. Prophylactic antibiotic or antifungal therapy is not currently recommended. Plasma exchange may provide benefit by reducing the systemic inflammatory response. A2  The infusion of hypertonic saline and the administration of mild hypothermia have not been shown to be useful. In addition to an elevated international normalized ratio, many patients develop thrombocytopenia because of consumptive coagulopathy (Chapter 166). However, prophylactic use of transfusion of clotting factors or platelets should be restricted to the treatment of hemorrhage or in preparation for invasive procedures.



Selecting for Liver Transplantation

Clinical variables associated with reduced likelihood for spontaneous recovery include severe encephalopathy, advanced age, certain causes (hepatitis B, druginduced hepatitis, Wilson disease), long duration of jaundice, massive necrosis on liver biopsy, and markedly diminished liver volume on radiologic imaging. Transplant-free survival in patients with acute liver failure is approximately 50% for acetaminophen toxicity, hepatitis A, shock liver, and acute fatty liver of pregnancy/HELLP but less than 25% for all other causes. Indications for listing for transplantation are based on the severity of disease but vary somewhat among countries (Table 145-3). Transplantation for acute liver failure accounts for only 2% of all liver transplants. About 45% of patients with acute liver failure are listed for transplantation; about 10% die on the waiting list, about 5% improve without transplantation, and about 30% undergo liver transplantation. Overall 1- and 5-year survival rates after liver transplantation are 79% and 71%, respectively, but post-transplantation survival is better for patients with acute acetaminophen overdose and worse for patients with severe encephalopathy before transplantation. Living donor and auxiliary liver transplantation may be considered, but their use is controversial.  

Chronic Liver Failure

Alcoholic cirrhosis (Chapter 144), which is now the leading indication for liver transplantation in the United States, remains controversial because many patients who totally abstain from alcohol may recover to the point that transplantation is no longer indicated. Candidates for transplantation must typically adhere to at least 6 months of rehabilitation and document abstinence from

TABLE 145-3 CRITERIA FOR SELECTION OF PATIENTS WITH ACUTE LIVER FAILURE FOR LIVER TRANSPLANTATION KING’S COLLEGE, LONDON, UNITED KINGDOM Acetaminophen toxicity Acidosis (pH 6.5 plus creatinine >3.4 mg/dL Other causes of acute liver failure INR >6.5, or Any three of the following: Age 40 years Non-A, non-B hepatitis or drug-induced disease Duration of jaundice before encephalopathy >7 days INR >3.5 Bilirubin >17.5 mg/dL HÔPITAL PAUL-BROUSSE, VILLEJUIF, FRANCE For all causes of acute liver failure: Hepatic encephalopathy, and Factor V 12 g/dL without hemolysis.

have had no or little response to corticosteroids and for elderly patients with comorbidities such as diabetes who are at high risk for developing corticosteroid-induced side effects. When rituximab is administered before splenectomy, vaccination against Streptococcus pneumoniae, Haemophilus influenzae type B, and/or Neisseria meningitidis should be administered whenever possible 2 weeks before rituximab because splenectomy may be required thereafter (see Chapter 159 for specific recommendations).

Splenectomy

Splenectomy has long been the preferred second-line option for primary wAIHAs. The rate of sustained response after splenectomy is approximately 60 to 70%, but clear predicting factors for response are still lacking. The perioperative risk of laparoscopic splenectomy is low and acceptable, with a mortality rate of less than 1%. The most feared complications, even after the laparoscopic procedure, remain the rare but unpredictable risk of overwhelming sepsis and the risk of postoperative thromboembolic complications, especially in the splanchnic veins. Perioperative low-molecular-weight heparin is therefore recommended in wAIHA patients who undergo splenectomy and especially in those who also have positive antiphospholipid antibodies. The best time for splenectomy is controversial now that alternatives such as rituximab are available at least in some countries. In children aged less than 5 to 7 years, this procedure should be avoided or delayed as long as possible. In adults, it must be considered early in the course of the disease in patients who have failed to respond to both corticosteroids (or need high and unacceptable doses to maintain at least a partial remission) and rituximab.

Danazol

In patients dependent on corticosteroids, the use of danazol, an attenuated androgen analogue, at 400 to 800 mg/day can be helpful as a corticosteroidsparing strategy in adults requiring a dose of daily prednisone greater than 15 mg to maintain a remission. However, because of its common androgenic side effects, its use is usually limited in female patients, and its potential liver toxicity makes long-term use difficult also in men.

Other Treatment Lines

Immunosuppressive and Cytotoxic Agents

In patients with refractory wAIHA in whom splenectomy and rituximab have failed, the management is mainly based on experience and mostly retrospective data in the literature, as opposed to controlled, prospective, comparison trials. The efficacy of azathioprine, cyclophosphamide, and to a lesser extent cyclosporine and mycophenolate mofetil has been reported in small cases series. The choice depends on the efficacy/safety ratio individualized for each patient. These drugs should be reserved for patients who have failed to respond to rituximab and to splenectomy or are not suitable candidates for splenectomy.

  PAROXYSMAL NOCTURNAL HEMOGLOBINURIA  

DEFINITION

PNH is a rare and potentially life-threatening clonal blood disorder with protean manifestations caused by an acquired somatic mutation in the phosphatidylinositol glycan (PIG-A) gene.9 In pluripotent hematopoietic stem cells, the mutation in PIG-A leads to a deficiency of glycosylphosphatidylinositol (GPI) anchors and GPI-anchored proteins, including the complement regulatory proteins CD55 and CD59 normally expressed on the surface of RBCs. PNH RBCs are therefore highly vulnerable to activation of complement on their surfaces (especially in case of fever, acidosis, or hypoxia) and the formation of the membrane attack complex (MAC) (Chapter 44).10 The resulting chronic intravascular hemolysis is the hallmark of the classic hemolytic form of the disease, and the release of free hemoglobin contributes to most of its clinical manifestations (dysphagia, recurrent abdominal pain, erectile dysfunction).  

EPIDEMIOLOGY

PNH can occur at any age but is most common between 10 and 50 years. The mean age at diagnosis is about 34 years (median age is about 40) and the female/male ratio is close to 1. Median survival after diagnosis is approximately 20 years. It is a rare disorder with an estimated prevalence in the population of one in 105 to one in 106. There is no sex preference, and a family history of PNH is highly unusual.  

PATHOBIOLOGY

PNH is due to a somatic mutation that causes a defect in the red cell membrane. The disease begins in a single hematopoietic stem cell when the PIG-A gene on the short arm of the active X chromosome acquires a mutation. The PIG-A gene encodes PIG-A, an enzyme that is essential for the synthesis of GPI. The location of PIG-A on the X chromosome allows a single mutation in the gene to affect GPI synthesis. This lipid normally forms a peptide link with the C-terminal amino acid of numerous proteins, anchoring them to the red cell membrane. Molecular genetic studies indicate that the PNH clone upregulates genes that make the PNH clone resistant to apoptosis and immune attack. The mutation in a hematopoietic stem cell affects PIG-A in blood cells of all lineages. Almost 150 different mutations of PIG-A have been identified. Most of them inactivate PIG-A and cause total loss of GPI in the descendants of the carrier stem cell. Red cells with complete deficiency of GPI are termed PNH III erythrocytes, and those with partial deficiency are called PNH II erythrocytes. The coexistence of PNH III and PNH II red cells in the same patient indicates the presence of two mutant clones. A small number of

CHAPTER 151  Autoimmune and Intravascular Hemolytic Anemias  

hematopoietic stem cells in normal people bear the PIG-A mutation; they have no proliferative advantage and persist in small numbers. In normal blood, the frequency of PIG-A–deficient cells is about one in 50,000 red cells. In contrast to those with PNH, however, the deficient cells in normal subjects arise from committed hematopoietic cells. The presence of PIG-A–deficient cells in normal subjects suggests that PNH involves not only the PIG-A mutation but also a second step, perhaps a mutation, that allows expansion of the mutated clone.

Functional Consequences of Deficiency of GPI

The membrane inhibitor of reactive hemolysis (CD59, or protectin) and CD55, an inhibitor of C3 convertase, are two of the many proteins that GPI normally anchors to the red cell. They prevent polymerization of C9, the final step in assembly of the MAC that begins with cleavage of C5 to C5b. Deficiencies of CD59 and CD55 allow unimpeded assembly of the MAC on the erythrocyte surface, thereby initiating intravascular hemolysis. A variety of nonspecific factors, such as a reduction in the pH of blood, can activate complement. The morning hemoglobinuria that occurs in some cases of PNH is probably the result of subtle acidification of blood during sleep.  

CLINICAL MANIFESTATIONS

Classically, some patients with PNH arise in the morning and pass dark urine. The typical paroxysms of hemoglobinuria occur on a background of chronic, low-grade intravascular hemolysis that causes constant hemosiderinuria in PNH. About one third of cases evolve into aplastic anemia (Chapter 156).11 Transformation to acute myelogenous leukemia is a rare event. Abdominal pain, dysphagia, and erectile dysfunction are additional clinical features. The basis of these symptoms is probably scavenging by free plasma hemoglobin of endogenous nitric oxide, a regulator of vasomotor and smooth muscle tone. In about one third of cases, venous thrombosis occurs in unusual sites and can cause intra-abdominal splanchnic vein thrombosis involving the hepatic veins (Budd-Chiari syndrome) or the portal veins, or less frequently cerebral vein thrombosis. Splenomegaly is uncommon; hepatomegaly and ascites point to the complication of intra-abdominal venous thrombosis. Hemosiderinuria is the result of chronic intravascular hemolysis. Subtle or overt signs of bone marrow damage (leukopenia and thrombocytopenia) are frequent. The extent of red cell destruction in PNH depends on the number of PNH red cells in blood, the level of GPI on the red cell membrane (PNH III cells are devoid of GPI), and the surface activation of complement. The anemia is often aggravated by iron deficiency caused by chronic urinary iron loss in the form of hemosiderinuria. A DAT-negative hemolytic anemia associated with iron deficiency should raise the suspicion of PNH.12 The basis of the tendency to develop venous thrombosis is unclear. Hypercoagulability caused by the release of prothrombotic materials in red cell and platelet membranes (which are also abnormal in PNH) and impaired fibrinolysis have been implicated. Nitric oxide scavenging by free plasma hemoglobin may also damage endothelial cells and cause aggregation of platelets.  

DIAGNOSIS

The diagnosis of PNH is suspected in case of an acquired and Coombs-negative intravascular hemolytic anemia and can be established by demonstrating, by flow cytometry, a deficiency of CD59 on RBCs.13,14 Another reagent with utility in flow cytometry is aerolysin, a bacterial protein that binds to the GPI anchor. Data strongly suggest that the fluorescent aerolysin variant (FLAER) is the best and most reliable reagent to study GPI-linked antigens on leukocytes, helpful for diagnosing PNH.15 The clone size in PNH is usually correlated with the degree of intravascular hemolysis.

TREATMENT  The treatment of classical PNH has long been only supportive and based on regular transfusions and anticoagulation therapy in case of thrombosis. Eculizumab, a humanized monoclonal antibody against C5, reduces the signs of intravascular hemolysis, the requirement for transfusions, and the incidence of thrombosis in PNH. A5  In PNH, it is usually given at a dose of 600 mg every

1045

week for 4 weeks, followed 1 week later by a 900-mg dose, and then at 900 mg every other week. Complement blockade is not always complete. Before being treated with eculizumab, every patient should be vaccinated against Meningococcus A, B, and C and, in many countries, a long-term prophylaxis with oral penicillin is also strongly recommended. Beyond transfusion-dependent hemolytic anemia, the occurrence of disease-related venous thrombosis is also a strong indication for eculizumab in the setting of PNH in combination with anticoagulation. Ravulizumab, which is a newer complement inhibitor that also binds to C5, has a much longer terminal half-life compared with eculizumab, and can be given every 8 weeks rather than every 2 weeks. A6 

  HEMOLYTIC TRANSFUSION REACTIONS

The cause of hemolytic transfusion reactions (Chapter 167) is intravascular lysis of the donor’s red cells by antibodies (alloantibodies or isoantibodies) in the recipient that bind to one or more blood group antigens on the transfused cells. The recipient’s isoantibodies can be natural anti-A or anti-B antibodies, or they can be induced by previous transfusions or pregnancy. Whether IgM or IgG, the isoantibodies trigger the assembly of lytic complement components on the surface of the donor’s red cells. The rapid formation of large amounts of C3a and C5a fragments causes hypotension and bronchospasm. Renal failure is a consequence of severe, prolonged hypotension; the main renal lesion is renal cortical ischemia secondary to shunting of blood away from the kidneys. Hemoglobin itself is not nephrotoxic. The signs and symptoms of a hemolytic transfusion reaction are nonspecific and include fever, back pain, urticaria, dyspnea, hypotension, and evidence of disseminated intravascular coagulation. These nonspecific signs appear and worsen during administration of the transfusion. Immediate steps must be taken to stop the transfusion, submit the transfused blood and a sample of the patient’s blood to the blood bank, and order tests of plasma and urine for free hemoglobin. Hydration is necessary to prevent renal failure.

  OTHER CAUSES OF INTRAVASCULAR

HEMOLYSIS

Conditions in which vascular abnormalities (e.g., valve hemolysis), toxins (e.g., venomous snake bites), infections (e.g., malaria, babesiosis, clostridial sepsis), or drugs damage RBCs, cause them to lose pieces of membrane, and ultimately fragment into free hemoglobin-releasing particles should be considered in the differential diagnosis of intravascular hemolysis (see Table 151-1). Most of these conditions are readily apparent from the history and physical examination. Treatment focuses on the underlying cause of the hemolysis.

  Grade A References A1. Berentsen S, Randen U, Vagan AM, et al. High response rate and durable remissions following fludarabine and rituximab combination therapy for chronic cold agglutinin disease. Blood. 2010;116: 3180-3184. A2. Berentsen S, Randen U, Oksman M, et al. Bendamustine plus rituximab for chronic cold agglutinin disease: results of a Nordic prospective multicenter trial. Blood. 2017;130:537-554. A3. Birgens H, Frederiksen H, Hasselbalch HC, et al. A phase III randomized trial comparing glucocorticoid monotherapy versus glucocorticoid and rituximab in patients with autoimmune haemolytic anaemia. Br J Haematol. 2013;163:393-399. A4. Michel M, Terriou L, Roudot-Thoraval F, et al. A randomized and double-blind controlled trial evaluating the safety and efficacy of Rituximab for warm auto-immune hemolytic anemia in adults (the RAIHA study). Am J Hematol. 2017;92:23-27. A5. Martí-Carvajal AJ, Anand V, Cardona AF, et al. Eculizumab for treating patients with paroxysmal nocturnal hemoglobinuria. Cochrane Database Syst Rev. 2014;CD010340. A6. Lee JW, Sicre de Fontbrune F, Wong Lee Lee L, et al. Ravulizumab (ALXN1210) vs eculizumab in adult patients with PNH naive to complement inhibitors: the 301 study. Blood. 2019;133:530-539.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 151  Autoimmune and Intravascular Hemolytic Anemias  

GENERAL REFERENCES 1. Liebman HA, Weitz IC. Autoimmune hemolytic anemia. Med Clin North Am. 2017;101:351-359. 2. Berentsen S. How I manage patients with cold agglutinin disease. Br J Haematol. 2018;181: 320-330. 3. Parker V, Tormey CA. The direct antiglobulin test: indications, interpretation, and pitfalls. Arch Pathol Lab Med. 2017;141:305-310. 4. Hill A, Hill QA. Autoimmune hemolytic anemia. Hematology Am Soc Hematol Educ Program. 2018;2018:382-389. 5. Leaf RK, Ferreri C, Rangachari D, et al. Clinical and laboratory features of autoimmune hemolytic anemia associated with immune checkpoint inhibitors. Am J Hematol. 2019. [Epub ahead of print.] 6. Hill QA, Stamps R, Massey E, et al. Guidelines on the management of drug-induced immune and secondary autoimmune, haemolytic anaemia. Br J Haematol. 2017;177:208-220. 7. Go RS, Winters JL, Kay NE. How I treat autoimmune hemolytic anemia. Blood. 2017;129:2971-2979. 8. Serris A, Amoura Z, Canoui-Poitrine F, et al. Efficacy and safety of rituximab for systemic lupus erythematosus-associated immune cytopenias: a multicenter retrospective cohort study of 71 adults. Am J Hematol. 2018;93:424-429.

1045.e1

9. Hill A, DeZern AE, Kinoshita T, et al. Paroxysmal nocturnal haemoglobinuria. Nat Rev Dis Primers. 2017;3:1-14. 10. Wong EKS, Kavanagh D. Diseases of complement dysregulation—an overview. Semin Immunopathol. 2018;40:49-64. 11. Gnanaraj J, Parnes A, Francis CW, et al. Approach to pancytopenia: diagnostic algorithm for clinical hematologists. Blood Rev. 2018;32:361-367. 12. England JT, Dalal B, Leitch HA. Frequency of and reasons for paroxysmal nocturnal haemoglobinuria screening in patients with unexplained anaemia. J Clin Pathol. 2018;71:364-367. 13. Keeney M, Illingworth A, Sutherland DR. Paroxsymal nocturnal hemoglobinuria assessment by flow cytometric analysis. Clin Lab Med. 2017;37:855-867. 14. Röth A, Maciejewski J, Nishimura JI, et al. Screening and diagnostic clinical algorithm for paroxysmal nocturnal hemoglobinuria: expert consensus. Eur J Haematol. 2018;101:3-11. 15. Oldaker T, Whitby L, Saber M, et al. ICCS/ESCCA consensus guidelines to detect GPI-deficient calls in paroxysmal nocturnal hemoglobinuria (PNH) and related disorders part 4—assay validation and quality assurance. Cytometry B Clin Cytom. 2018;94:67-81.

1046

CHAPTER 152  HEMOLYTIC ANEMIAS  

152 

circulatory cycle throughout the body, an erythrocyte is subjected to high shear stress in the arterial system, dramatic size and shape changes in the microcirculation with capillary diameters as small as 7.5 µm, and marked fluctuations in tonicity, pH, and Po2.1

HEMOLYTIC ANEMIAS: RED BLOOD CELL MEMBRANE AND METABOLIC DEFECTS



PATRICK G. GALLAGHER

The mature erythrocyte differs from all other cells in the body. Lacking a nucleus, DNA, RNA, and ribosomes, it cannot synthesize RNA, DNA, or protein. It does not divide, it has no mitochondria, it cannot perform the Krebs cycle, and it lacks an electron transport system for oxidative phosphorylation. After enucleation, the reticulocyte, the precursor of the mature erythrocyte, leaves the marrow and enters the circulation equipped with a full complement of enzymes, transporters, signaling molecules, and all other proteins necessary to perform the essential functions of the red blood cell (RBC) during its lifespan. The erythrocyte membrane accounts for only about 1% of the total weight of an RBC, yet it plays a critical role in the maintenance of normal RBC homeostasis through a number of mechanisms. These include retention of vital compounds and removal of metabolic waste, regulation of erythrocyte metabolism and pH, and import of iron required for hemoglobin (Hb) synthesis during erythropoiesis. The membrane maintains a slippery exterior so that erythrocytes do not aggregate or adhere to endothelial cells. The membrane skeleton, a network of proteins on the inner surface of the RBC, provides the strength and flexibility needed to maintain the normal shape and deformability of the erythrocyte. The principal functions of erythrocyte metabolism in the mature erythrocyte include maintenance of adequate supplies of adenosine triphosphate (ATP), production of reducing substances to act as antioxidants, and control of oxygen affinity of Hb by production of adequate amounts of 2,3-diphosphoglycerate (2,3-DPG). Because the mature erythrocyte has lost its ability to perform oxidative phosphorylation, its energy is supplied by anaerobic glycolysis though the Embden-Meyerhof pathway, by oxidative glycolysis through the hexose monophosphate shunt, and through nucleotide salvage pathways.



Membrane Proteins

Membrane proteins are classified as integral, penetrating or crossing the lipid bilayer and interacting with the hydrophobic lipid core, or peripheral, interacting with integral proteins or lipids at the membrane surface but not penetrating into the bilayer core. Integral membrane proteins include the glycophorins, the Rh proteins, Kell and Duffy antigens, and transport proteins such as band 3 (AE1, anion exchanger 1, SLC4A1), Na+,K+-ATPase, Ca2+-ATPase, and Mg2+ATPase. Numerous membrane receptors and antigens are present on integral membrane proteins. Peripheral membrane proteins are on the cytoplasmic membrane face and include enzymes such as glyceraldehyde-3-phosphate dehydrogenase and the structural proteins of the spectrin-actin–based membrane skeleton.

Integral Membrane Proteins

  THE ERYTHROCYTE MEMBRANE

Band 3, the major integral protein of the RBC, has two primary functions, ion transport and maintenance of protein–protein interactions. Band 3 mediates chloride–bicarbonate exchange and provides a binding site for glycolytic enzymes, Hb, and the skeletal proteins ankyrin, protein 4.1, and protein 4.2.

Composed of a lipid bilayer and an underlying cortical membrane skeleton (Fig. 152-1), the membrane provides the erythrocyte the deformability and stability required to withstand its travels through the circulation. In one

Glycophorin A

Membrane Lipids

Red blood cell membrane lipids are asymmetrically distributed across the bilayer membrane, reflecting a steady state involving a constant exchange of phospholipids between the two bilayer hemileaflets. Glycolipids and cholesterol are intercalated between the phospholipids in the bilayer with their long axes perpendicular to the bilayer plane. Glycolipids, located in the external half of the bilayer with their carbohydrate moieties extending into the aqueous phase, carry several important RBC antigens and serve other important functions. Phospholipids are asymmetrically organized, with the choline phospholipids, phosphatidylcholine and sphingomyelin, primarily in the outer half of the bilayer, and the amino phospholipids, phosphatidylethanolamine and phosphatidylserine (PS), in the inner half of the bilayer. In pathologic states, such as thalassemia, sickle cell disease, and diabetes, loss of phospholipid asymmetry with externalization of PS leads to activation of blood clotting through conversion of prothrombin to thrombin and facilitates macrophage attachment to erythrocytes, marking them for destruction. Mature erythrocytes are unable to synthesize fatty acids, phospholipids, or cholesterol de novo and depend on lipid exchange and fatty acid acylation as mechanisms for phospholipid repair and renewal.

Glycophorin B

Glycophorin A

Band 3 Band 3

Band 3 4.2

Glycophorin C/D

Band 3

Rh RhAG LW

p55

CD47

Dematin

Ankyrin-1

4.1R Actin β Spectrin

Adducin 4.1R

α Spectrin

Tropomyosin Tropomodulin FIGURE 152-1.  The erythrocyte membrane. A model of the major proteins of the erythrocyte membrane is shown: α and β spectrin, ankyrin, band 3 (the anion exchanger), 4.1 (protein 4.1) and 4.2 (protein 4.2), actin, and glycophorin. (From Perrotta S, Gallagher PG, Mohandas N. Hereditary spherocytosis. Lancet. 2008;372:1411-1426.)

CHAPTER 152  HEMOLYTIC ANEMIAS  

ABSTRACT

Disorders of the erythrocyte membrane and erythrocyte metabolism comprise an important group of inherited hemolytic anemias. Abnormalities of proteins of the membrane and its underlying cytoskeleton compromise the strength and flexibility of the erythrocyte needed to maintain the normal shape and deformability needed to survive its travels through the circulation. Qualitative and quantitative membrane protein defects lead to various abnormalities of erythrocyte shape associated with varying degrees of hemolysis. In many cases, red blood cell shape abnormalities provide a clue to the pathobiology and diagnosis of the underlying disorder. Disorders of erythrocyte volume homeostasis associated with abnormalities in membrane proteins are now beginning to be understood. Abnormalities of metabolism are particularly crippling to the erythrocyte, because without a nucleus or mitochondria, the cell is left to depend on the array of proteins available after enucleation to maintain its metabolic functions. These include maintaining adequate supplies of energy via production of ATP, producing reducing substances to allow antioxidant production and activity, and controlling oxygen affinity of hemoglobin. Clinical, laboratory, and genetic manifestations of metabolic disorders of the erythrocyte are heterogeneous.

KEYWORDS

membrane metabolism spherocytosis elliptocytosis pyropoikilocytosis stomatocytosis

1046.e1

CHAPTER 152  HEMOLYTIC ANEMIAS  

A single N-glycan chain attached to an asparagine in the membrane spanning domain of band 3 is composed of N-acetyl-d-lactosamine units arranged in an unbranched, linear fashion in fetal erythrocytes (i antigen) and in a branched fashion in adult cells (I antigen). The glycophorins are the next most abundant family of integral membrane proteins. They provide most of the negative surface charge required by RBCs to avoid sticking to each other and to the vascular wall. They are involved in transmembrane signaling and carry receptors for Plasmodium falciparum, a number of viruses and bacteria, and several blood group antigens.

Peripheral Membrane Proteins

Spectrin is the major component of the membrane skeleton. It is composed of two subunits, α and β spectrin, that are structurally related but functionally distinct. Spectrin is highly flexible and assumes a variety of conformations, an unusual property that may be critical for normal membrane pliancy. The spectrin-based membrane skeleton is linked to the plasma membrane through the actin–protein 4.1 junctional complex; through spectrin–ankyrin interactions; and through binding of a multiprotein complex containing Rh proteins, Rh-associated glycoproteins, CD47, LW, glycophorin B, and protein 4.2 to ankyrin. Protein 4.1, a protein necessary for normal membrane stability, interacts with spectrin, actin, and other proteins of the RBC membrane. Ankyrin serves as the primary linkage protein for the high-affinity binding of spectrin to the inner membrane through interactions with the cytoplasmic domain of band 3. Protein 4.2 is a peripheral membrane protein that helps link the skeleton to the lipid bilayer through interactions with ankyrin and band 3. Erythrocyte membrane disorders result from alterations in the quantity or quality (or both) of individual proteins and their dynamic interactions with each other. Disruption of the vertical protein–protein interactions of the membrane, that is, the spectrin-ankyrin–band 3 linkage or the band 3–protein 4.2 interaction, leads to uncoupling of the membrane skeleton from the lipid bilayer. This leads to membrane instability with loss of lipids and some integral membrane proteins, resulting in loss of membrane surface area and the phenotype of spherocytosis. Disruption of the horizontal interactions of membrane skeleton proteins, including perturbation of spectrin self-association or junctional complex protein–protein interactions, leads to membrane instability, altered membrane deformability and mechanical properties, and the phenotype of elliptocytosis. Erythrocyte membrane disorders also result from abnormal cation perme­ ability or disturbed erythrocyte hydration (see later section on “Hereditary Stomatocytosis Syndromes.”)

  DISORDERS OF THE ERYTHROCYTE MEMBRANE

Hemolytic anemias caused by defects in the erythrocyte membrane comprise an important group of hereditary anemias. Hereditary spherocytosis, hereditary elliptocytosis, and hereditary pyropoikilocytosis are the most common disorders among this group.2 Detailed clinical studies carried out years ago have now been complemented by biochemical and genetic studies, providing both a better understanding of the pathogenesis of these disorders and a better understanding of the normal biology of the erythrocyte membrane.  

Hereditary Spherocytosis  

DEFINITION

Hereditary spherocytosis is a group of disorders characterized by spherical erythrocytes on the peripheral blood smear (Fig. 152-2A). Clinical, laboratory, and genetic heterogeneity characterize this group of disorders.  

EPIDEMIOLOGY

Hereditary spherocytosis affects approximately one in 2000 to 3000 individuals of northern European ancestry. Found worldwide, it is much more common in whites than individuals of African ancestry.  

PATHOBIOLOGY

The primary defect in hereditary spherocytosis is the loss of erythrocyte membrane surface area caused by defects in erythrocyte membrane proteins, including α spectrin, β spectrin, ankyrin, band 3, and protein 4.2. Qualitative or quantitative defects of one or more of these membrane proteins lead to membrane instability, which, in turn, leads to membrane loss. In approximately two thirds of hereditary spherocytosis patients, inheritance is autosomal dominant. In the remaining patients, inheritance is nondominant owing to a de novo mutation or autosomal recessive inheritance. Cases with autosomal recessive inheritance are caused by defects in either α spectrin or protein 4.2. Rare cases of homozygous hereditary spherocytosis have been reported, resulting in fetal

1047

death or severe hemolytic anemia. In most cases, hereditary spherocytosis mutations are “private,” that is, each individual has a unique mutation, implying that there is no selective evolutionary advantage to hereditary spherocytosis. The spleen plays a critical, albeit secondary, role in the pathophysiology of hereditary spherocytosis. Splenic destruction of poorly deformable spherocytes is the primary cause of hemolysis experienced by hereditary spherocytosis patients. Abnormal erythrocytes are trapped in the splenic microcirculation and ingested by phagocytes. Moreover, the splenic environment is hostile to erythrocytes, with low pH, low glucose, and low ATP concentrations and high local concentrations of toxic free radicals produced by adjacent phagocytes, all contributing to membrane damage.  

CLINICAL MANIFESTATIONS

The clinical manifestations of the spherocytosis syndromes vary widely.3,4 The classic triad of hereditary spherocytosis is anemia, jaundice, and splenomegaly. Rarely, patients may have severe hemolytic anemia presenting in utero or shortly after birth and continuing through the first year of life. These patients may require multiple blood transfusions, and in some cases, splenectomy in infancy. Many patients with hereditary spherocytosis escape detection throughout childhood. In these patients, the diagnosis of hereditary spherocytosis is not made until they are being evaluated for unrelated disorders later in life or when complications related to anemia or chronic hemolysis occur. Although the lifespan of an erythrocyte in these patients may be shortened to only 20 to 30 days, they adequately compensate for their hemolysis with increased bone marrow erythropoiesis. Chronic hemolysis leads to the formation of bilirubinate gallstones (Chapter 146), the most frequently reported complication in patients with hereditary spherocytosis. Although gallstones have been observed in early childhood, most appear in adolescents and young adults. Routine interval ultrasonography to detect gallstones should be performed even if patients are asymptomatic. Other complications of hereditary spherocytosis include aplastic, hemolytic, and megaloblastic crises. Aplastic crises occur after virally induced bone marrow suppression and present with anemia, jaundice, fever, and vomiting. The most common etiologic agent in these cases is parvovirus B19 (Chapter 347). Hemolytic crises, usually associated with viral illnesses and occurring before 6 years of age, are generally mild and present with jaundice, increased spleen size, and a decrease in hematocrit. Megaloblastic crises occur in hereditary spherocytosis patients with increased folate demands, such as the pregnant patient, growing children, or patients recovering from an aplastic crisis. Uncommon manifestations of hereditary spherocytosis include skin ulceration, gout, chronic leg dermatitis, cardiomyopathy, spinal cord dysfunction, movement disorders, and extramedullary erythropoiesis. In patients with untreated severe hereditary spherocytosis, poor growth and findings attributable to extramedullary hematopoiesis, such as hand and skull deformities, may be observed.  

DIAGNOSIS

Patients with hereditary spherocytosis may present at any age, usually with anemia, hyperbilirubinemia, or an abnormal blood smear. In evaluating a patient with suspected hereditary spherocytosis, particular attention should be paid to the family history, including questions about anemia, jaundice, gallstones, and splenectomy. The initial laboratory investigation should include a complete blood count with a peripheral smear, reticulocyte count, direct antiglobulin test (Coombs test), and serum bilirubin. When the peripheral smear or family history is suggestive of hereditary spherocytosis, flow cytometric analysis of eosin-5-maleimide–labeled erythrocytes (EMA binding) or an incubated osmotic fragility test (discussed later) should be obtained. Rarely, additional, specialized testing is required to confirm the diagnosis. Overall, laboratory findings in hereditary spherocytosis are heterogeneous. Erythrocyte morphology (Chapter 148) is distinctive but not diagnostic (Fig. 152-2A). Typical hereditary spherocytosis patients have blood smears with easily identifiable spherocytes lacking central pallor. Some patients present with only a few spherocytes on peripheral smear, but others present with numerous small, dense spherocytes and bizarre erythrocyte morphology. Specific morphologic findings have been identified in patients with certain membrane protein defects such as pincered erythrocytes (band 3) or spherocytic acanthocytes (β spectrin). When examining a smear in a case of suspected spherocytosis, it is important to have a high-quality smear with the erythrocytes well separated and some cells with central pallor in the field of examination because spherocytes are a common artifact on peripheral blood smears. Other disorders with spherocytes on peripheral blood smear are listed in Table 152-1.

1048

CHAPTER 152  HEMOLYTIC ANEMIAS  

A

B

C

D

FIGURE 152-2.  Peripheral blood smears in disorders of erythrocyte shape. A, Hereditary spherocytosis. Characteristic small spherocytes lacking central pallor are seen. B, Hereditary elliptocytosis. Smooth, cigar-shaped elliptocytes are seen. C, Hereditary pyropoikilocytosis. Pronounced microcytosis, poikilocytosis, fragmentation of erythrocytes, and elliptocytes are seen. D, Hereditary stomatocytosis.

TABLE 152-1 DISORDERS WITH SPHEROCYTES ON PERIPHERAL BLOOD FILM Hereditary spherocytosis Autoimmune hemolytic anemia Thermal injuries Microangiopathic and macroangiopathic hemolytic anemias Hepatic disease Clostridial septicemia Transfusion reactions with hemolysis Poisoning with certain snake, spider, and Hymenoptera venoms Severe hypophosphatemia Heinz body anemias ABO incompatibility (neonates)

The mean corpuscular hemoglobin concentration (MCHC) is increased (between 34.5 and 38) owing to relative cellular dehydration. The mean corpuscular volume (MCV) is usually normal or slightly decreased. Many cell counters provide a histogram of MCHCs claimed to be accurate enough to identify nearly all patients with hereditary spherocytosis. In a normal erythrocyte, a redundancy of cell membrane gives the cell its characteristic discoid shape and provides it abundant surface area. In spherocytes, there is a decrease in surface area relative to cell volume, resulting in their abnormal shape. This change is reflected in the increased osmotic fragility found in these cells. Osmotic fragility is tested by adding increasingly hypotonic concentrations of saline to RBCs. Normal erythrocytes are able to increase their volume by swelling, but spherocytes, which are already at maximal volume for surface area, burst at higher saline concentrations than normal. Approximately one fourth of hereditary spherocytosis individuals will have a normal osmotic fragility on freshly drawn RBCs, with the osmotic fragility curve approximating the number of spherocytes seen on peripheral smear. However, after incubation at 37° C for 24 hours, hereditary spherocytosis RBCs lose membrane surface area more readily than normal because their membranes have become leaky and unstable. Thus, incubation accentuates the defect in hereditary spherocytosis erythrocytes and brings out the defect on osmotic fragility, making incubated osmotic fragility the standard test in diagnosing hereditary spherocytosis (Fig. 152-3, bottom panel). When the spleen is present, a subpopulation of very fragile erythrocytes that have been conditioned by the spleen form the tail of the osmotic fragility curve. This tail disappears after splenectomy. The osmotic fragility test suffers from poor sensitivity, with as

many as 20% of mild cases of hereditary spherocytosis missed after incubation. It is unreliable in patients who have small numbers of spherocytes and in patients who have been recently transfused. It is abnormal in other conditions in which spherocytes are present. Eosin-5-maleimide binding is a flow cytometry–based test used in the diagnosis of hereditary spherocytosis.5,6 EMA is a fluorescent dye that binds to band 3 and Rh-related proteins in the erythrocyte membrane. In hereditary spherocytosis, the mean fluorescence of EMA-stained erythrocytes is lower compared with control because of the reduction of band 3 and related proteins, typically decreased to approximately 65% of normal (Fig. 152-3, top panel). Although primary defects of band 3 protein are seen in only about 25% of hereditary spherocytosis patients, decreased fluorescence intensity is also observed in the erythrocyte membranes of hereditary spherocytosis patients with defects in other membrane proteins such as ankyrin and spectrin. This is thought to be attributable to transmission of long-range effects of mutant protein defects across the membrane lattice, ultimately influencing the amount of EMA binding to band 3. EMA binding has good sensitivity and specificity and is simple and rapidly performed. Specialized testing is available for studying difficult cases or cases in which additional information is desired. Useful tests for these purposes include structural and functional studies of erythrocyte membrane proteins, such as protein quantitation, limited tryptic digestion of spectrin, spectrin, and ion transport. Membrane deformability may be examined using an ektacytometer. Genetic diagnosis of hereditary spherocytosis has become widely available and the associated costs continue to decrease. The precise role of molecular diagnosis is evolving. Next-generation sequencing panels could become a costeffective approach to molecular diagnosis of hereditary hemolytic anemia, especially when the family history is uninformative or when routine laboratory testing fails to identify the causative hemolytic process.7 In difficult cases, such as when there is transfusion dependence, genetic testing may be diagnostic. Some experts recommend genetic testing to verify the diagnosis prior to splenectomy. Other laboratory manifestations in hereditary spherocytosis are manifestations of ongoing hemolysis. Increased serum bilirubin, increased lactate dehydrogenase, increased urinary and fecal urobilinogen, and decreased serum haptoglobin reflect increased erythrocyte destruction. After diagnosing a patient with hereditary spherocytosis, family members should be examined for presence of the disease. This can be of great epidemiologic importance, particularly for very old and very young patients. Prenatal diagnosis of hereditary spherocytosis has been made in a few cases, but this is rarely necessary.

CHAPTER 152  HEMOLYTIC ANEMIAS  

who have hereditary spherocytosis and also have vascular compromise of vital organs. Whether patients with moderate hereditary spherocytosis and compensated, asymptomatic anemia should undergo splenectomy is controversial.8 When splenectomy is indicated, laparoscopic splenectomy has become the method of choice. This technique results in less postoperative discomfort, a quicker return to preoperative diet and activities, shorter hospitalization, decreased costs, and smaller scars. Even massive spleens can be removed laparoscopically because the spleen is placed in a large bag, diced intraoperatively, and eliminated through suction catheters. Partial splenectomy, initially advocated for infants and young children with significant anemia associated with erythrocyte membrane disorders to allow for palliation of hemolysis and anemia while maintaining some residual splenic immune function, is now being suggested by some for most hereditary spherocytosis patients.9 Guidelines, which reflect changes in current opinion about surgical management, include (1) preference for a laparoscopic approach, (2) performance of splenectomy ideally after the age of 6 years, (3) no indication for extended thrombosis prophylaxis after splenectomy for hereditary spherocytosis, and (4) avoidance of splenectomy in patients with some forms of hereditary stomatocytosis because of an increased risk of venous thromboembolism. Before splenectomy (Chapter 159), patients should be immunized with vaccines against pneumococcus, Haemophilus influenzae type B, and meningococcus. Postsplenectomy care includes counseling of patients or parents to seek prompt medical care in case of febrile illness, prophylaxis if traveling to malaria endemic areas, and importance of seeking medical care after dog or cat bites. Use of routine antibiotics after splenectomy for prevention of pneumococcal sepsis is controversial and data are lacking to indicate or refute their prescription. Many prescribe daily penicillin prophylaxis for 2 years postsplenectomy. Before splenectomy and postsplenectomy if hemolysis does not abate, hereditary spherocytosis patients should take folic acid (1 mg/day orally) to prevent folate deficiency.

Counts

282 HS

Control

188

94

8 100

101

EMA

102

100 Severe HS Tail Typical HS

Percent Lysis

80

1049

60

40

Control

20



Hereditary Elliptocytosis and Related Disorders  

0 0.8

DEFINITION

Hereditary elliptocytosis is characterized by the presence of elliptical or oval cigar-shaped erythrocytes on peripheral blood smears of affected individuals (see Fig. 152-2B). 0.7

0.5 0.6 NaCI concentration (%)

0.4

0.3

FIGURE 152-3.  Testing in hereditary spherocytosis. Top panel, Eosin-5-maleimide (EMA) binding. Histogram of fluorescence of EMA-labeled erythrocytes from a normal control and a patient with typical hereditary spherocytosis. Decreased fluorescence is observed from hereditary spherocytosis (HS) erythrocytes. Bottom panel, Osmotic fragility curves in hereditary spherocytosis. The shaded region is the normal range. Results representative of both typical, and severe spherocytosis are shown. A tail, representing fragile erythrocytes conditioned by the spleen, is common in spherocytosis patients prior to splenectomy. (From Gallagher PG. Abnormalities of the erythrocyte membrane. Pediatr Clin North Am 2013;60:1349-1352.)





TREATMENT AND PROGNOSIS  Splenic sequestration and destruction is the primary determinant of erythrocyte survival in hereditary spherocytosis patients. Splenectomy cures or alleviates anemia in most patients, reducing or eliminating the need for transfusions. The risk for cholelithiasis is also decreased to nearly background levels. After splenectomy, spherocytes remain in the peripheral blood, but their lifespan becomes near normal. In the past, splenectomy was routinely performed in all patients with hereditary spherocytosis. However, the risk of overwhelming postsplenectomy infection; the emergence of penicillin-resistant pneumococci; and the growing recognition of the increased risk of postsplenectomy cardiovascular disease, particularly thrombosis, pulmonary hypertension, and increased risk of atherosclerosis have led to reevaluation of the role of splenectomy in the treatment of hereditary spherocytosis (Chapter 159). In addition, with growing globalization, the important role of the spleen in protection of individuals living in or traveling to geographic regions where parasitic diseases such as malaria or babesiosis occur has reemerged. When splenectomy is considered, health care providers, the patient, and family members must review and weigh the benefits of splenectomy against the immediate and long-term risks of the procedure. Considering the risks and benefits, a reasonable approach is to splenectomize all patients with severe spherocytosis and all patients who have significant signs or symptoms of anemia, including growth failure, skeletal changes, leg ulcers, and extramedullary hematopoietic tumors. Other candidates for splenectomy are older patients

EPIDEMIOLOGY

Hereditary elliptocytosis has been estimated to occur in approximately one in 2000 to 4000 individuals. The true incidence of hereditary elliptocytosis is unknown because its clinical severity is heterogeneous, and many patients are asymptomatic. It is common in African Americans and people of Mediterranean ancestry, presumably because elliptocytes confer some resistance to malaria. In parts of Africa, the incidence of hereditary elliptocytosis approaches one in 100.

PATHOBIOLOGY

The principal defect in hereditary elliptocytosis is mechanical weakness or fragility of the erythrocyte membrane skeleton. Qualitative and quantitative defects in a number of RBC membrane proteins have been described in hereditary elliptocytosis, including α spectrin, β spectrin, protein 4.1, and glycophorin C. Most defects occur in spectrin, the principal structural protein of the erythrocyte membrane skeleton. αβ Spectrin heterodimers self-associate into tetramers and higher order oligomers that are critical for erythrocyte membrane stability as well as erythrocyte shape and function. Most spectrin defects in hereditary elliptocytosis impair the ability of spectrin dimers to self-associate into tetramers and oligomers, thereby disrupting the membrane skeleton. Structural and functional defects of protein 4.1 appear to disrupt the spectrin– actin contact in the membrane skeleton. Glycophorin C variants are also deficient in protein 4.1. The precise pathobiology of how elliptocytes are formed in these syndromes is unclear. Genetically, hereditary elliptocytosis is heterogeneous with multiple genetic loci. A wide variety of mutations have been described in the α spectrin, β spectrin, protein 4.1, and glycophorin C genes, including point mutations, gene deletions and insertions, and messenger RNA processing defects. Several mutations have been identified in a number of individuals of the same genetic background, suggesting a “founder effect” for these mutants, which supports the hypothesis that there has been genetic selection for elliptocytosis because these RBCs confer some resistance to malaria. Most cases of hereditary elliptocytosis are inherited in an autosomal dominant pattern, with rare cases of de novo mutations.

1050



CHAPTER 152  HEMOLYTIC ANEMIAS  



Hereditary Pyropoikilocytosis

Hereditary pyropoikilocytosis is a rare cause of anemia with distinctive erythrocyte morphology on peripheral blood smear (see Fig. 152-2C) and has a picture similar to that seen in patients with severe burns. Patients typically present in infancy with severe anemia and peripheral blood smear findings of elliptocytosis, poikilocytosis, pyknocytosis, and fragmentation. Microspherocytosis is common, and the MCV is usually very low (50-70 fL). Most patients are of African ancestry. At least one third of hereditary pyropoikilocytosis patients have a parent or sibling with typical hereditary elliptocytosis, as they share common mutations in the self-association site of spectrin. Patients with hereditary pyropoikilocytosis tend to experience severe hemolysis and anemia in infancy that gradually improves, evolving toward typical hereditary elliptocytosis later in life.  

Patients should be followed for signs of decompensation during acute illnesses. Interval ultrasonography to detect gallstones should be performed. In patients with significant hemolysis, folate should be administered daily.

CLINICAL MANIFESTATIONS

The clinical presentation of hereditary elliptocytosis is heterogeneous, ranging from asymptomatic carriers to patients with severe, life-threatening anemia. Most patients with hereditary elliptocytosis are asymptomatic and are diagnosed incidentally during testing for unrelated conditions. Asymptomatic carriers have been identified who possess the same molecular defect as an affected hereditary elliptocytosis relative but who have normal peripheral blood smears. The erythrocyte lifespan, normal in most patients, is decreased in only about 10% of patients. This subset of patients with hereditary elliptocytosis and decreased erythrocyte lifespan experience hemolysis, anemia, splenomegaly, and intermittent jaundice. Many of these patients have parents with typical hereditary elliptocytosis and thus are homozygotes or compound heterozygotes for defects inherited from each of the parents. Symptoms may vary among members of the same family; indeed, they may vary in the same individual at different times.

DIAGNOSIS

Cigar-shaped elliptocytes on peripheral blood smear are the hallmark of hereditary elliptocytosis (see Fig. 152-2B). These normochromic, normocytic elliptocytes vary in number from a few to 100%, with the likelihood of hemolysis not correlating with the number of elliptocytes present. Ovalocytes, spherocytes, stomatocytes, and fragmented cells may also be seen. In some cases, pyknocytes may be prominent. Elliptocytes may be seen in association with other disorders, including megaloblastic anemias, hypochromic microcytic anemias (iron deficiency anemia and thalassemia), myelodysplastic syndromes, and myelofibrosis; however, elliptocytes generally make up fewer than one third of RBCs in these conditions. History and additional laboratory testing usually clarify the diagnosis of these disorders. In typical cases, the incubated osmotic fragility is normal, but in severe hereditary elliptocytosis and hereditary pyropoikilocytosis, incubated osmotic fragility is increased, and EMA binding is decreased. Other laboratory findings in hereditary elliptocytosis are similar to those found in other hemolytic anemias and are nonspecific markers of increased erythrocyte production and destruction. The reticulocyte count generally is less than 5% but may be higher when hemolysis is severe. Similar to hereditary spherocytosis, specialized laboratory procedures are available to study the erythrocyte membranes of hereditary elliptocytosis and hereditary pyropoikilocytosis patients. These studies are not routinely required to make the diagnosis of hereditary elliptocytosis or hereditary pyropoikilocytosis, but they may be helpful in studying problematic cases and in elucidating the underlying molecular defects.

TREATMENT  Therapy is rarely needed in patients with hereditary elliptocytosis. In rare cases, occasional RBC transfusions may be required. In cases of severe hereditary elliptocytosis and hereditary pyropoikilocytosis, splenectomy has been palliative because the spleen is the site of erythrocyte sequestration and destruction. Many practitioners think that the same indications for splenectomy in hereditary spherocytosis should be applied to patients with symptomatic hereditary elliptocytosis or hereditary pyropoikilocytosis. Postsplenectomy patients with hereditary elliptocytosis or hereditary pyropoikilocytosis experience increased hematocrits, decreased reticulocyte counts, and improvement in clinical symptoms. Similar postsplenectomy guidelines as outlined for hereditary spherocytosis should be followed.



Hereditary Stomatocytosis Syndromes

Red blood cell hydration is primarily determined by the intracellular concentration of monovalent cations. A net increase in sodium and potassium ions causes water to enter, forming stomatocytes (see Fig. 152-2D) or hydrocytes, but a net loss of sodium and potassium produces dehydrated RBCs, or xerocytes. Numerous descriptions of congenital or familial hemolytic anemias associated with abnormal cation permeability and, in some cases, disturbed RBC hydration have been reported.10 These span the range from severe hydrocytosis to severe xerocytosis. In many cases, the molecular bases of this group of disorders are unknown. An unusual characteristic of the stomatocytosis syndromes is a predisposition to thrombosis after splenectomy. Acquired stomatocytosis has been associated with acute alcoholism and hepatobiliary disease, vinca alkaloid administration, neoplasms, and cardiovascular disease. Stomatocytosis is also sometimes observed as a processing artifact.  

OVERHYDRATED HEREDITARY STOMATOCYTOSIS (HYDROCYTOSIS)

This group of disorders is characterized by stomatocytes, erythrocytes with a mouth-shaped (stoma) area of central pallor on peripheral blood smear (see Fig. 152-2D), severe hemolysis, macrocytosis (110-150 f L), elevated erythrocyte sodium concentration, reduced potassium concentration, and increased total Na+ and K+ content. The excess cations expand cell water, producing large, osmotically fragile cells with low MCHCs (24%-30%). The clinical severity of overhydrated hereditary stomatocytosis is variable; some patients experience hemolysis and anemia, but others are asymptomatic. Missense mutations in the Rh-associated glycoprotein (RHAG) and in band 3 (SLC4A1) have been identified in discrete subsets of hydrocytosis patients.  

DEHYDRATED HEREDITARY STOMATOCYTOSIS (XEROCYTOSIS)

Blood smears from patients with dehydrated hereditary stomatocytosis exhibit contracted and spiculated RBCs, dessicytes, a variable number of stomatocytes, and target cells. Most patients have nearly normal erythrocyte morphology, with only a few target cells and an occasional echinocyte or stomatocyte. The MCV (95-115 f L) and MCHC are increased, and the osmotic fragility is reduced (i.e., resistance to osmotic lysis). The characteristic biochemical abnormality is a decreased potassium concentration and total monovalent cation content. Dominantly inherited mutations in PIEZO1 are found in most xerocytosis patients. PIEZO proteins are the recently identified pore-forming subunits of channels that mediate mechanotransduction in mammalian cells. Association of PIEZO variants with changes in erythrocyte hydration suggest that these proteins play an important role in erythrocyte volume homeostasis. A few hereditary xerocytosis patients have been found to have mutations in the Gardos channel, encoded by KCNN4, important in the pathogenesis of erythrocyte dehydration in sickle cell disease.  

INTERMEDIATE SYNDROMES AND HEREDITARY STOMATOCYTOSIS VARIANTS

Hydrocytosis and xerocytosis represent the extremes of a broad spectrum of RBC permeability defects. In fact, families with features of both conditions have been reported. Some patients with severe permeability defects have little or no hemolysis. The proportion of stomatocytes and the degree of sodium influx do not correlate with each other, and neither correlates with the amount of hemolysis or anemia.

  ERYTHROCYTE METABOLISM

The primary functions of the erythrocyte, gas transport and exchange, are maintained without a net change in energy state. However, several critical functions of the erythrocyte depend on the production and expenditure of energy. As erythrocytes age, glucose utilization and ATP levels fall, leading to decreased membrane deformability and, ultimately, a shortened lifespan. Lower potassium levels, higher sodium levels, and decreased membrane lipids are also seen in ATP-deficient, aging erythrocytes. Erythrocytes do not undergo oxidative phosphorylation and do not store glycogen; thus, they must constantly catabolize glucose from the blood stream through the Embden-Meyerhof pathway and the hexose monophosphate shunt as a source of energy (Fig. 152-4). Erythrocytes incorporate glucose from the

CHAPTER 152  HEMOLYTIC ANEMIAS  

Glucose

NADP

G-6-P

Glucose phosphate isomerase

Glucose-6-phosphate dehydrogenase

F-6-P ATP PhosphoR-5-P fructokinase ADP FDP Fructose diphosphate aldolase DHAP Triose phosphate Pi isomerase

Antioxidant activity

ATP ADP

Hexokinase

6-PG

Glutathione reductase

NADPH

Enolase PEP Pyruvate kinase

H2O 2H+

Glutathione (GSH)

GSH Methemoglobin reduction

Pi 2-PG

Glutathione peroxidase

GSSG

NAD DiphosphoglyceroNADH G-3-P mutase Dehydrogenase 1,3-DPG ADP 2,3-DPG Phosphoglycerate kinase Diphosphoglycerate ATP phosphatase 3-PG Phosphoglyceromutase

H2O2

GSH

CO2

G-3-P

1051

ADP ATP Glycine

Glucose catabolism Membrane structure Cation pump Glutathione synthesis Nucleotide salvage

Glutathione synthetase λ-Glutamylcysteine ADP

ATP Cysteine

λ-Glutamylcysteine synthetase Glutamic acid

ADP ATP

Pyruvate Lactate dehydrogenase

Adenylate kinase

NADH NAD

AMP 2 ADP

Lactate

FIGURE 152-4.  Pathways of energy metabolism in the erythrocyte. Glucose-6-phosphate may be degraded anaerobically to lactate through the Embden-Meyerhof pathway or oxidatively through the hexose monophosphate shunt. Pentose phosphates (R-5-P) can reenter anaerobic glycolysis as fructose-6-phosphate (F-6-P) and glyceraldehyde-3-phosphate (G-3-P) after conversion by enzymes of the terminal pentose phosphate pathway or as a product of adenosine or inosine degradation. 2,3-Diphosphoglycerate (2,3-DPG) may be generated instead of adenosine triphosphate (ATP) through diversion of triose through the Rapoport-Luebering shunt. Glutathione may be synthesized directly from constituent amino acids; its cycling from oxidized (GSSG) to reduced forms (GSH) depends on reduced pyridine cofactor (NADPH) generation. ADP = adenosine diphosphate; DHAP = dihydroxyacetone phosphate; FDP = fructose-1,6-diphosphate; NAD = nicotinamide adenine dinucleotide; NADP = nicotinamide adenine dinucleotide phosphate; NADPH = nicotinamide adenine dinucleotide phosphate, reduced form; PEP = polyestradiol phosphate.

plasma through facilitated transfer, with erythrocyte glucose levels rapidly equilibrating with changes in blood glucose levels. Glucose is the preferred carbohydrate of the RBC, but fructose and mannose are metabolized almost as readily. Inside the erythrocyte, glucose is converted to glucose-6-phosphate or to fructose by sorbitol. Glucose-6-phosphate follows one of three pathways: (1) most (≈90%) enters the Embden-Meyerhof pathway, where it is converted into lactate, pyruvate, and ATP; (2) some (≈5%-10%) enters the hexose monophosphate shunt to produce reduced intermediates and ribulose 5-phosphate, the latter of which eventually enters the Embden-Meyerhof pathway; and (3) a tiny fraction ( 3.5

10–50 (may be up to 100%) 40 years), hepatocellular carcinoma (Chapter 186) due to long-term untreated iron accumulation has been detected with resultant cirrhosis in the liver as found in hereditary hemochromatosis (Chapter 201). Hypogonadism, hypothyroidism, and diabetes mellitus are rare. Although patients with thalassemia intermedia generally experience puberty late, they have normal sexual development and are usually fertile. Women with thalassemia intermedia may have successful pregnancies, although complications during pregnancy may occur.11 Thalassemia minor is the heterozygous state of β-thalassemia. Subjects with thalassemia minor are “carriers” of a single β-globin gene defect, and they are usually asymptomatic except for a mild anemia of pregnancy. The carriers are usually identified in the course of a family study, incidentally12 during an intercurrent illness, or as part of a population survey. The anemia is mild, microcytic, and hypochromic and is associated with an elevated level of HbA2. The blood smear shows characteristic microcytosis and hypochromia, with some variation in size and shape of the red cells. The presence of target cells is variable. The hematologic features are remarkably similar among different ethnic groups. Carriers of β-thalassemia with normal HbA2 have been observed in settings in which an individual with mild thalassemia intermedia was found to have one parent with typical β-thalassemia minor with elevated HbA2, whereas the other showed either minimal or no hematologic abnormalities and normal HbA2. Subjects with normal HbA2 are usually carriers of silent β++ mutations (β++). Carriers of β-thalassemia may have a triplicated α-gene allele: its coinheritance may cause a mild thalassemia intermedia phenotype

CHAPTER 153  The Thalassemias  

Transferrin saturation due to frequent blood transfusions, or ineffective erythropoiesis leading to increasd iron absorption

Subsequent formation of NTBI in plasma

30%

Transferrin saturation %

Normal: no NTBI Iron produced overload 100%

Uncontrolled iron loading of organs Pituitary Parathyroid Thyroid Heart

Fe Fe Fe Fe

1059

Liver

Fe

Pancreas

Fe Fe

Gonads

A 85–07

Survival probability

1.00

75–79 70–74

0.75 65–69 0.50

60–64

0.25 P < 0.00005

0

B

80–84

0

5

10

15 Age (yrs)

20

25

30

FIGURE 153-3.  A, Transfusion iron overload and specific organ loading. B, Survival of Italian thalassemia major patients by cohort of birth. NTBI = non–transferrin-bound iron.

due to the excess of α chains production. Coincidental iron deficiency can lower elevated HbA2 levels to the normal range.

δβ-Thalassemia and Hereditary Persistence of Fetal Hemoglobin

The clinical manifestations of (δβ)0-thalassemias are similar to those of thalassemia intermedia, whereas the heterozygotes are distinguished from β-thalassemia heterozygotes by normal levels of HbA2 together with an increased HbF level of 5 to 20%. Homozygotes for HbLepore may have phenotypes similar to either thalassemia major or thalassemia intermedia. Subjects affected by deletion or nondeletion forms of HPFH are usually asymptomatic. Any of the β-thalassemia defects may be coinherited with β-chain variants (HbS, HbC, HbE) and cause a clinically relevant β-thalassemia phenotype of different severity. These variants further illustrate that β-thalassemia syndromes have a wide clinical spectrum and that specific therapeutic approaches may completely change the clinical course and the natural history of these disorders (Fig. 153-3).

Hemoglobin E and Hemoglobin E Thalassemias

Hemoglobin E (HbE) is caused by a substitution of glutamic acid by lysine at codon 26 of the β-globin gene. It is a common mutation, especially in the Indian subcontinent and Southeast Asia and among people who have emigrated from those regions. The frequency of HbE approaches 60% in many regions of Thailand, Laos, and Cambodia. HbE results in reduced synthesis of the β-E chain and therefore has the phenotype of a mild form of β-thalassemia. HbE trait has little clinical significance; it may be associated with slight microcytosis without anemia, but it could be mistaken for iron deficiency without appropriate laboratory testing. Homozygous HbEE individuals have mild anemia and microcytosis, occasional splenomegaly, and peripheral blood smears showing red cell changes similar to those seen in thalassemia trait, including target cells. In this condition, hepcidin is suppressed, thereby leading to iron accumulation—even in the absence of blood transfusions and certainly worsened by them.13 HbE interacts with different forms of α-thalassemia to produce a wide variety of clinical disorders. Its coinheritance with β-thalassemia, a condition called HbE/β-thalassemia, is the most common severe form of β-thalassemia in Asia and globally represents about 50% of the clinically severe β-thalassemias.14



DIAGNOSIS

The diagnosis of thalassemia may be required in a patient with an appropriate, suggestive clinical picture or for the identification of a heterozygote subject as part of a family study or population screening program. The general approach is common to any form of thalassemia, regardless of presentation. The primary evaluation is based on hematologic changes; the red cell indices by electronic cell counter and the red cell morphology examined on a well-stained blood film are sufficient to direct further investigations. Individuals with mean corpuscular volume below 80 f L and mean corpuscular hemoglobin below 27 pg with normal iron parameters need to be further investigated. The red blood cell number is usually higher than normal. In the presence of anemia with thalassemic red cell changes, the next step is the evaluation of hemoglobin fractions (HbA, HbA2, HbF, or hemoglobin variants) by electrophoresis on cellulose acetate at alkaline pH or, even better, by high-performance liquid chromatography that enables the precise measurement of HbA2, HbF, and HbA and the provisional identification of a large number of hemoglobin variants, including HbE. An HbA2 level above 3.5% associated with hypochromic microcytic red cells is diagnostic of β-thalassemia minor. HbA2 values between 3.2 and 3.5% (borderline) should be interpreted with care because they could be due to interaction of more than one thalassemic defect (α and β), a silent β mutation, or concomitant iron deficiency. If iron deficiency is present, it should be corrected and the HbA2 estimation repeated. The majority of individuals with thalassemic red cell indices with normal or low HbA2 and normal HbF will be α0-thalassemia carriers or α+-thalassemia homozygotes. Carriers of α0-thalassemia may have a few red cells with HbH inclusions. Microcytosis with low or normal HbA2 levels with elevated HbF (2 to 20%) indicates heterozygosity for δβ-thalassemia. Patients with HPFH usually have normal red blood cell indices but increased levels of HbF with different intercellular distribution (homogeneous or pancellular with the exception of heterocellular HPFH) compared with δβ-thalassemia (uneven or heterocellular). A radioactive method for measuring the α/β-globin synthesis ratio was introduced in the mid to late 1960s, and it was largely directed at prenatal diagnosis in the pre-DNA era. Although it gives a quantitative assessment of globin production, today its use is limited to difficult cases because of interaction of different globin chain defects. The definitive diagnosis of the thalassemia syndromes involves the identification of the underlying mutations through DNA analysis. There are several methods available for the diagnosis of any

1060

CHAPTER 153  The Thalassemias  

particular mutation, such as polymerase chain reaction (PCR) restriction enzyme analysis, PCR allele-specific oligonucleotides, gap PCR, and direct sequencing that at present is probably the easiest and most reliable method. For deletion forms of α-thalassemia, the multiplex ligation-dependent probe amplification is a recently introduced, useful method. During their clinical course, patients affected by different forms of thalassemia develop several complications mainly due to iron overload, which requires monitoring to direct iron chelation therapy. The principal methods of determining body iron levels (Chapter 201) are measurements of the serum ferritin level and assessment of liver iron concentration from biopsy tissue or, as an alternative noninvasive method, by R2 MRI. The reciprocals of T2 and T2*, known as R2 and R2*, are directly proportional to iron content and demonstrate the most promising results. High serum ferritin levels (>2500 µg/L) and high liver iron concentration (>15 g/dry weight) indicate high risk for significant morbidity and mortality. A cutoff of 800 µg/L ferritin and more than 7g/dry weight liver iron concentration have been associated with high risk for morbidity in thalassemia intermedia and other forms of non–transfusiondependent thalassemias. Cardiac iron can be measured by a T2* MRI procedure that allows estimation of the cardiac iron load. MRI T2* values below 10 milliseconds are always associated with severe iron load and high risk for heart failure within 1 year. MRI T2* values above 20 milliseconds are considered normal, meaning no iron in the heart. Echocardiography may also be useful to evaluate functional changes. For other complications, including endocrinopathies, liver disease, lung disease, thrombophilia, and bone disease, the diagnostic approaches are similar to those used in clinical practice; these are performed with consideration of test cost, performance characteristics, and preferences of the patients, as described in the corresponding chapters.

TREATMENT  Conventional Treatment

No specific treatment is required for α- or β-thalassemia heterozygotes (carriers, thalassemia minor), but they should receive appropriate genetic counseling. During pregnancy, thalassemia-carrying women may become more anemic, so they should be observed carefully, mainly during the second and third trimesters, and supported with folic acid. When real iron deficiency is associated with thalassemia traits, iron supplementation should be provided, monitoring transferrin saturation and ferritin. A few cases of in utero blood transfusions have been reported with hemoglobin Bart hydrops fetalis syndrome; most of the infants have been delivered prematurely by cesarean section, subsequent development has been abnormal, and survivors required regular blood transfusions after birth. HbH patients in general have higher hemoglobin levels (8 to 9 g/dL) and do not need regular blood transfusion. Supplementation with folic acid (2 to 5 mg/day) is generally recommended, especially in pediatric patients. The major complications in HbH disease are hemolytic crises that may occur during or after acute infections; in such cases, immediate intervention, including blood transfusions and treatment for infections, should be promptly administered. The clinical management of thalassemia major and thalassemia intermedia remains the major issue. The quality and duration of life of thalassemia major and thalassemia intermedia patients have been transformed in this century, with life expectancy increasing well into the third and fourth decades. Nevertheless, prolongation of life is accompanied by several complications, partly due to the underlying disorder and partly as a consequence of the treatment with blood transfusions and iron overload. Moreover, we are starting to deal with aging-related complications in the context of a multiorgan disease that requires management by a team of clinicians who have specific knowledge of thalassemias in adults, working together with different specialists and welltrained nurses. The conventional treatment of thalassemia major patients includes regular transfusion therapy and iron chelation. The definition of the optimal transfusion and iron chelation regimen has been the most important advance in the management of thalassemia major patients, with the primary objective being to control the ineffective erythropoiesis, its consequences, and the body iron burden. The optimal transfusion regimen involves regular blood transfusions, usually administered every 2 to 5 weeks, to maintain the pretransfusion hemoglobin levels above 9 to 10.5 g/dL. The decision to initiate lifelong transfusion therapy should be based on a definitive diagnosis of severe thalassemia, taking into account the molecular defects, the severity of anemia on repeated measurement, the level of ineffective erythropoiesis, and the clinical criteria (such as failure to thrive or bone changes). It is advisable that thalassemia major patients receive leukoreduced packed red cells to minimize transfusion reactions and pathogen transmission. Adverse reactions to red blood cell transfusions may occur during or after transfusion and can be hemolytic and nonhemolytic. Transfusion-related acute lung injury is rare but severe and must be immediately managed (Chapter 167).

Many patients with thalassemia major require splenectomy because of hypersplenism. However, optimal clinical management may delay or even obviate the need for splenectomy that was common in the past. Splenectomy should be considered only for patients whose annual blood consumption increases progressively and is responsible for significant increases in iron stores despite good chelation therapy or in the presence of symptoms due to spleen enlargement. Clinical problems related to leukopenia or thrombocytopenia due to hypersplenism could also be the reasons for considering splenectomy.15 The major complication of splenectomy is severe and sometimes overwhelming infection. Because removal of the spleen may reduce the primary immune response to encapsulated organisms, it is advisable to delay splenectomy until patients are at least 5 years old. The mortality rate for postsplenectomy overwhelming infection in thalassemia patients is approximately 50% despite intensive supportive care. Therefore, it is mandatory to adopt preventive measures including immunoprophylaxis (vaccination against Streptococcus pneumoniae, pneumococcus, and meningococcus) (Chapter 159), chemoprophylaxis, and education of the parent and patient to recognize and to report febrile illnesses. Increase of thrombotic risk has been well documented in thalassemia patients after splenectomy; thus this procedure must be avoided as much as possible. Iron overload is an inevitable and serious complication of long-term blood transfusion therapy and hyperabsorption of dietary iron that requires adequate treatment to prevent early death, mainly from iron-induced cardiac disease. Optimal chelation therapy extends complication-free survival (see Fig. 1533). The standard chelation therapy for more than 40 years was deferoxamine, given for 10 to 24 hours daily as a continuous subcutaneous infusion 5 to 7 days per week. A1  The long-term efficacy of deferoxamine has been extensively documented in large cohorts of patients in Italy and elsewhere. Unfortunately, compliance with the rigorous regimen of daily subcutaneous infusions is a serious limiting factor, and life expectancy in noncompliant patients is not different from that in the pre-deferoxamine era. This has been the rationale behind the intensive effort to identify alternative, orally effective iron chelators. At present, two oral iron chelators are on the market: deferiprone and deferasirox. Deferiprone is registered in Europe and more recently in the United States. By the guidelines of the EMEA countries (Europe, Middle East, and Africa), treatment with deferiprone at doses of 75 to 100 mg/kg/day is restricted to patients unable to use deferoxamine or patients with an unsatisfactory response to deferoxamine as judged by serum ferritin levels and by liver iron concentrations. Studies indicate that deferiprone may be more effective than deferoxamine in protecting the heart from the accumulation of iron. A2  A potential benefit of combined deferoxamine and deferiprone therapy has been observed, and according to Thalassemia International Federation guidelines, a combination treatment (deferoxamine and deferiprone) should be considered for patients with high levels of heart iron or cardiac dysfunction. The new orally effective iron chelator deferasirox has been shown to be effective and safe in removing excess iron from different organs, including the heart. A3  A4  Deferasirox is now available in most countries throughout the world as first-line treatment. Its use has clearly demonstrated that iron chelation is not a standard treatment, but it should be individualized according to age, history of compliance with previous chelation, and other factors. Monitoring and adjustment of iron chelation by repeated measurements of ferritin, calculation of iron intake by transfusions, and, whenever possible, measurement of cardiac and liver iron at least once by MRI are mandatory. A prospective phase 2 study evaluating combination of deferasirox with deferoxamine in patients with severe transfusional myocardial siderosis showed clinically meaningful improvements in cardiac MRI results. In patients with thalassemia major and cardiac siderosis, amlodipine added to chelation therapy reduced cardiac iron more effectively than chelation therapy alone, but larger studies will be required to inform the evaluations. A5  The use of proton pump inhibitors may further reduce serum ferritin levels. A6  The management of thalassemia intermedia patients is more complicated because of the wide heterogeneity of thalassemia intermedia phenotypes. A number of options are currently adopted for treatment of thalassemia intermedia patients, including transfusion therapy, splenectomy, modulation of HbF production, and hematopoietic stem cell transplantation (HSCT). However, increasing evidence is documenting the benefit of transfusion therapy in decreasing the incidence of complications. Thus, although the common practice has been to initiate transfusion when complications ensue, it may be worthwhile to start transfusion therapy earlier as a preventive approach, which will also help alleviate the increased risk for alloimmunization with delayed initiation of transfusion. The initiation of iron chelation therapy in patients with thalassemia intermedia depends not only on the amount of excess iron but also on the rate of iron accumulation, the duration of exposure to excess iron, and various other factors in individual patients.16 ,

Bone Marrow Transplantation and Experimental Therapies

Allogeneic HSCT (Chapter 168) in thalassemia syndromes has been increasingly successful during the past two decades, mainly in β-thalassemia major.17 Predictors of poor transplant outcome are hepatomegaly, history of irregular

chelation, and hepatic fibrosis. Patients are categorized into three risk classes. Class 1 patients have none of these adverse risk factors, class 2 patients have one or two adverse risk factors, and class 3 patients have all three. In the most recent update of the Pesaro group’s experience, the probabilities of thalassemiafree survival for patients younger than 17 years at the time of HSCT receiving the allograft from an HLA-identical relative were 87% and 85%, respectively, in patients belonging to class 1 and class 2 and much lower in young patients in class 3. The progressive adjustment of conditioning regimens in class 3 patients and in adults (>17 years old) has significantly reduced the incidence of transplantrelated mortality in patients in class 3. Only 25 to 30% of patients with diseases potentially curable by HSCT have a suitable HLA-compatible sibling. Bone marrow transplantation from unrelated donors increases significantly the incidence of acute and chronic graft-versus-host disease, particularly in thalassemia. A study from the Eurocord cooperative group reported the outcome of 33 patients with thalassemia belonging to class 1 and class 2 (Pesaro classification) who received cord blood hematopoietic stem cells from an HLA-identical sibling; no patient died of transplant-related complications, suggesting that related cord blood HSCT is a safe procedure for thalassemia patients. An alternative treatment of β-thalassemia consists of the pharmacologic stimulation of HbF synthesis. In humans, hemoglobin switch from HbF to HbA occurs in the period around birth as a result of γ- to β-globin gene switching. A number of pharmacologic agents able to reactivate HbF synthesis have been identified, including hypomethylating agents, histone deacetylase inhibitors, and hydroxyurea. Whereas the effect of these pharmacologic treatments (particularly hydroxyurea) in sickle cell disease is clear (Chapter 154), their benefit on the clinical course of β-thalassemia is presently limited. The discrepancy between these two conditions in the response to HbF inducers may be mainly related to the higher level of HbF required in β-thalassemia to achieve clinical results compared with those observed in sickle cell disease. The limited clinical response to γ-globin inducers observed in the majority of β-thalassemic patients may be also a reflection of the unfavorable effects of these agents on the other globin genes (i.e., increased α-globin synthesis). Although much attention has been paid to pathways that increase γ-globin expression, and hence the production of fetal hemoglobin, the reduction of α-globin expression may provide an equally plausible approach to ameliorating clinically severe forms of β-thalassemia, especially in patients with hemoglobin E β-thalassemia, who comprise about 50% of all patients born each year with severe β-thalassemia. Two new molecules, sotatercept18 and luspatercept,19 activin-type IIA and IIB receptor fusion proteins, have recently been shown to increase the release of mature erythrocytes into the circulation by acting mainly on late-stage erythropoiesis. Clinical data in healthy volunteers have shown that treatment with sotatercept results in increased red blood cell parameters. A phase 2a, multicenter, open-label, dose-finding study to determine a safe and active dose level of sotatercept and luspatercept in adult patients with β-thalassemia intermedia and major has been completed. The positive results prompted a phase 3 study with luspatercept in transfusion-dependent thalassemia patients. The phase 3 study has been recently completed and the results are very promising, showing a significant reduction of blood transfusions in transfusion-dependent patients treated with luspatercept compared to placebo group. Gene therapy (Chapter 38) is an attractive approach for thalassemia syndromes; however, this strategy poses major challenges in terms of controlling transgene expression, which should be erythroid specific and sustained over time. Treatment of β-thalassemia, sickle cell disease, and other disorders through lentivirus-mediated gene transfer has been reported in murine and primate models.20 At present, several phase 1/2 clinical studies are ongoing with encouraging results. Genome editing through CRISPR/Cas9 technology increases fetal hemoglobin production. It is in preclinical phase and will possibly enter the clinical phase in 2019. Moreover, new molecules with a potential ability to correct ineffective erythropoiesis or to generate iron-restricted erythropoiesis are in phase 1 clinical trials.

  UNSTABLE HEMOGLOBINOPATHIES

More than 80 rare mutant hemoglobins have been reported to cause hemolytic anemia by either amino acid replacements or deletions that significantly lower the solubility of hemoglobin. These mutant hemoglobins thereby form intracellular precipitates that can be detected as so-called Heinz bodies when the blood smear is exposed to a supravital stain. Structural abnormalities include mutations that weaken the linkage between heme and globin, disrupt secondary (α-helical) structure, or introduce a charged or polar side group into the hydrophobic interior of the globin subunit. This disorder, sometimes called congenital Heinz body hemolytic anemia, is inherited in an autosomal dominant manner. Severely affected individuals have jaundice, splenomegaly, and, on occasion, dark brown urine due to the release of heme and aberrant conversion to dipyroles. The instability of a few of these globin mutants is so extreme that they cannot be detected by routine laboratory methods. This results in a thalassemia phenotype with microcytosis

and ineffective erythropoiesis. Like individuals with glucose-6-phosphate dehydrogenase deficiency (Chapter 152), those with unstable hemoglobin mutants often lack clinical symptoms and signs of hemolysis until they develop an infection or are exposed to an oxidant drug. The diagnosis can be established by a combination of a positive Heinz body preparation and either abnormal hemoglobin electrophoresis or demonstration of a precipitate after exposure of a hemolysate to heat or isopropanol. Some clinics have access to a reference laboratory that can identify the specific mutation by α- and β-globin DNA sequencing. Most individuals with this disorder do not require treatment; some are symptomatic from severe anemia. Splenectomy generally results in a significant increase in red cell mass. However, the fraction of Heinz body–positive red cells increases markedly after splenectomy, and these patients are now at significant risk for development of pulmonary hypertension and cor pulmonale.

  Grade A References A1. Fisher SA, Brunskill SJ, Doree C, et al. Desferrioxamine mesylate for managing transfusional iron overload in people with transfusion-dependent thalassaemia. Cochrane Database Syst Rev. 2013;8:CD004450. A2. Fisher SA, Brunskill SJ, Doree C, et al. Oral deferiprone for iron chelation in people with thalassaemia. Cochrane Database Syst Rev. 2013;8:CD004839. A3. Taher AT, Porter J, Viprakasit V, et al. Deferasirox reduces iron overload significantly in nontransfusiondependent thalassemia: 1-year results from a prospective, randomized, double-blind, placebo-controlled study. Blood. 2012;120:970-977. A4. Dou H, Qin Y, Chen G, et al. Effectiveness and safety of deferasirox in thalassemia with iron overload: a meta-analysis. Acta Haematol. 2019;141:32-42. A5. Sadaf A, Hasan B, Das JK, et al. Calcium channel blockers for preventing cardiomyopathy due to iron overload in people with transfusion-dependent beta thalassaemia. Cochrane Database Syst Rev. 2018;7:CD011626. A6. Eghbali A, Khalilpour A, Taherahmadi H, et al. Pantoprazole reduces serum ferritin in patients with thalassemia major and intermedia: a randomized, controlled study. Therapie. 2019. [Epub ahead of print.]

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 153  The Thalassemias  

GENERAL REFERENCES 1. Taher AT, Weatherall DJ, Cappellini MD. Thalassaemia. Lancet. 2018;391:155-167. 2. Mettananda S, Higgs DR. Molecular basis and genetic modifiers of thalassemia. Hematol Oncol Clin North Am. 2018;32:177-191. 3. Antoniani C, Meneghini V, Lattanzi A, et al. Induction of fetal hemoglobin synthesis by CRISPR/ Cas9-mediated editing of the human β-globin locus. Blood. 2018;131:1960-1973. 4. Taher AT, Weatherall DJ, Cappellini MD. Thalassaemia. Lancet. 2018;391:155-167. 5. Sleiman J, Tarhini A, Bou-Fakhredin R, et al. Non-transfusion-dependent thalassemia: an update on complications and management. Int J Mol Sci. 2018;19:1-6. 6. Origa R. β-Thalassemia. Genet Med. 2017;19:609-619. 7. Wong P, Fuller PJ, Gillespie MT, Milat F. Bone disease in thalassemia: a molecular and clinical overview. Endocr Rev. 2016;37:320-346. 8. Pepe A, Gamberini MR, Missere M, et al. Gender differences in the development of cardiac complications: a multicenter study of a large cohort of thalassaemia major patients to optimize the timing of cardiac follow-up. Br J Haematol. 2018;180:879-888. 9. Ansari S, Rashid N, Hanifa A, et al. Laboratory diagnosis for thalassemia intermedia: are we there yet? J Clin Lab Anal. 2019;33:1-8. 10. Trinchero A, Marchetti M, Giaccherini C, et al. Platelet haemostatic properties in β-thalassaemia: the effect of blood transfusion. Blood Transfus. 2017;15:413-421. 11. Cassinerio E, Baldini IM, Alameddine RS, et al. Pregnancy in patients with thalassemia major: a cohort study and conclusions for an adequate management approach. Ann Hematol. 2017;96: 1015-1021.

1061.e1

12. Viprakasit V, Ekwattanakit S. Clinical classification, screening and diagnosis for thalassemia. Hematol Oncol Clin North Am. 2018;32:193-211. 13. Jones E, Pasricha SR, Allen A, et al. Hepcidin is suppressed by erythropoiesis in hemoglobin E β-thalassemia and β-thalassemia trait. Blood. 2015;125:873-880. 14. Vichinsky E. Non-transfusion-dependent thalassemia and thalassemia intermedia: epidemiology, complications, and management. Curr Med Res Opin. 2016;32:191-204. 15. Iolascon A, Andolfo I, Barcellini W, et al. Recommendations regarding splenectomy in hereditary hemolytic anemias. Haematologica. 2017;102:1304-1313. 16. Taher A, Musallam K, Cappellini MD. Guidelines for the management of non transfusion dependent thalassaemia (NTDT) 2nd edition. 2017. http://www.thalassemia.org/boduw/wp-content/ uploads/2011/09/Guidelines-for-Mgmt-of-NTDT-TIF-2017. Accessed March 7, 2019. 17. Strocchio L, Locatelli F. Hematopoietic stem cell transplantation in thalassemia. Hematol Oncol Clin North Am. 2018;32:317-328. 18. Cappellini MD, Porter J, Origa R, et al. Sotatercept, a novel transforming growth factor β ligand trap, improves anemia in β-thalassemia: a phase II, open-label, dose-finding study. Haematologica. 2019;104:477-484. 19. Piga A, Perrotta S, Gamberini MR, et al. Luspatercept improves hemoglobin levels and blood transfusion requirements in a study of patients with beta-thalassemia. Blood. 2019. [Epub ahead of print.] 20. Ghiaccio V, Chappell M, Rivella S, et al. Gene therapy for beta-hemoglobinopathies: milestones, new therapies and challenges. Mol Diagn Ther. 2019. [Epub ahead of print.]

1061.e2

CHAPTER 153  The Thalassemias  

REVIEW QUESTIONS 1. An 18-year-old woman from Pakistan has been observed for a lifelong severe microcytic and hypochromic anemia. Her hemoglobin levels have generally been in the range of 7 to 10 g/dL, but she recalls needing blood transfusions only twice in her life, on both occasions after traumatic bone fractures. Throughout childhood, her parents had kept her from playing sports or participating in activities requiring significant physical exertion because they had been told by pediatricians that she had a “weak heart.” She had markedly delayed puberty but now had regular menses. Physical examination shows marked hepatosplenomegaly; laboratory test abnormalities include indirect hyperbilirubinemia and hyperglycemia. What is the most likely diagnosis? A . β-Thalassemia major B. Homozygosity for hemoglobin E (HbEE) C. Hemoglobin H (HbH) disease D. Hydrops fetalis E. Hereditary juvenile hemochromatosis Answer: C  The clinical picture is most consistent with thalassemia intermedia. In the case of α-thalassemia, this would be most like hemoglobin H disease (most frequently resulting from the interaction of α+- and α0-thalassemia), with the β4 tetramers of hemoglobin H forming red cell inclusion bodies leading to chronic hemolysis. Patients with α- or β-thalassemia intermedia become severely iron overloaded from lifelong hyperabsorption of dietary iron, even without needing blood transfusions. A similar gut hyperabsorption of iron occurs in familial hemochromatosis states, but those individuals are not typically anemic. This patient’s hyperglycemia, growth retardation, possibly her bone fragility, and even the history of a “weak heart” (perhaps due to cardiomyopathy) are all manifestations of systemic iron overload. Thalassemias are especially prevalent not only in people of Mediterranean and Middle Eastern origin but also in those from the Indian subcontinent. In contrast to thalassemia intermedia, a patient with β-thalassemia major would be expected to be transfusion dependent throughout life. Hydrops fetalis, in which all four α-globin genes are deleted, is incompatible with life. Hemoglobin E, even in its homozygous state, causes little if any anemia and microcytosis.

2. Which of the following statements is correct regarding hematopoietic stem cell transplantation for thalassemia? A . It can lead to thalassemia-free survival of at least 85% in low-risk, younger patients. B. It should not be administered with cord blood for transplantation. C. It has been mostly effective in the α-thalassemias. D. It is mostly reserved for thalassemia patients who are poorly iron chelated and have liver involvement. E. It is associated with a reduced risk for graft-versus-host disease compared with comparable unrelated transplants for other hematologic diseases. Answer: A  The highest risk indicators in hematopoietic stem cell transplantation for thalassemia are hepatomegaly, hepatic fibrosis, and poor iron chelation history. On the other hand, the Pesaro group’s experience has demonstrated 87% disease-free survival in thalassemia patients who are younger than 17 years at the time of transplantation, have none of the risk factors noted, and are allografted from an HLA-identical related donor. Early experience with cord blood transplantation has shown it to be a safe procedure for thalassemia patients. Hematopoietic stem cell transplantation has been mainly successful in β-thalassemia. The risk for graft-versus-host disease (acute and chronic) after bone marrow transplantation from unrelated donors is at least as high in thalassemia as it is in other indications for bone marrow transplantation. 3. Heterozygosity for hemoglobin E: A . Is a rare form of unstable hemoglobinopathy B. Is associated with lifelong, low-grade hemolysis C. Causes iron overload from hyperabsorption of dietary iron D. Is frequently associated with iron deficiency E. Is of little if any clinical significance Answer: E  Hemoglobin E is one of the most common mutations in the world, especially in the Indian subcontinent and Southeast Asia. Its phenotype is that of a mild thalassemia, not an unstable hemoglobinopathy. It is of little if any clinical importance in its heterozygous form (HbE trait); even individuals who are homozygous for hemoglobin E (HbEE) usually have only mild anemia and microcytosis, without any clinical sequelae. Hemoglobin E can be mistaken for iron deficiency in the absence of further laboratory testing, but it is not known to be associated with iron deficiency.

1061

CHAPTER 154  Sickle Cell Disease and Other Hemoglobinopathies  

154  SICKLE CELL DISEASE AND OTHER HEMOGLOBINOPATHIES JO HOWARD

  SICKLE CELL DISEASE  

DEFINITION

Sickle cell disease is caused by a mutation in the β-globin gene (HBB). The resulting clinical syndromes are characterized by chronic hemolytic anemias and can be complicated by vaso-occlusion and vasculopathy that can cause widespread acute and chronic organ damage as well as premature mortality.  

EPIDEMIOLOGY

The sickle mutation became prominent several thousands of years ago in subSaharan African and Arab-Indian populations. Its persistence was due to a selective advantage in areas where falciparum malaria was endemic, with sickle trait carriers estimated to have approximately 90% protection against severe malaria infection. The carrier frequency is now highest in West and Central Africa (up to 50%), in Arab populations in the Middle East, and in tribal populations in India (Fig. 154-1). Forced movement of African populations during the slave trade and, more recently, migration due to war and economic pressures have led to increasing numbers of carriers and of those affected in the Americas, the Caribbean, Europe, and worldwide. The prevalence of sickle cell disease and sickle trait varies across the world, but more than 75% of the estimated 300,000 to 400,000 births affected annually with sickle cell anemia (HbSS) occur in sub-Saharan Africa, with these numbers predicted to increase.1

CHAPTER 154  Sickle Cell Disease and Other Hemoglobinopathies  

ABSTRACT

Sickle cell disease (SCD) is caused by a homozygous GAG-to-GTG mutation in codon 6 of the β-globin chain of human hemoglobin that results in a single amino acid substitution, Glu6Val. This point mutation is sufficient to cause increased polymerization of hemoglobin in red blood cells that are deoxygenated during transit through the microcirculation of tissues and organs, leading to distortion of red cell shape from round, pliable biconcave discs to rigid sickle cells that occlude the microvasculature. The clinical sequelae of this pathophysiology include episodic vaso-occlusive events that are associated with pain crises, hemolytic anemia, stroke, acute chest syndrome, pulmonary hypertension, urologic and renal complications that can lead to chronic kidney disease, and other ischemic organ involvement. Individuals who are heterozygous for the mutation have sickle trait that does not cause disease but can be infrequently associated with specific complications. Curative options include hematopoietic stem cell transplantation and gene therapy but both are associated with significant morbidity and mortality. In addition to well-coordinated palliative care, the use of hydroxyurea and transfusion therapy, including exchange transfusions, has been found to favorably affect the natural history of the disease in certain situations.

KEYWORDS

sickle cell vaso-occlusive pain crisis stroke acute chest syndrome priapism hydroxyurea exchange transfusion hematopoietic stem cell transplantation

1061.e3

1062

CHAPTER 154  Sickle Cell Disease and Other Hemoglobinopathies  



0–20% 2% 25%

8% 2%

3–35%

5%

5–30% 5–50% 0–15% 85

2-3

2-15

4

HbSS-α-thalassemia

Homozygous HbS α+-thalassemia

30% of HbSS patients

25-33

70-85

>85

4-6

2-15

4

HbSC disease (HbSC)

Compound heterozygous HbS, HbC

1 : 800

28-40

70-85

50

2-3

1-8

2

HbS-β0-thalassemia (HbS-β0-Thal)

Compound heterozygous HbS, β0-thalassemia

1 : 1600

20-30

65-75

>85

4-6

5-15

4

HbS-β+-thalassemia (HbS-β+-Thal)

Compound heterozygous HbS, β+-thalassemia

1 : 1600

30-40

60-70

70-95

4-6

2-10

1-3

HbSE disease (HbSE)

Compound heterozygous HbS, HbE

Rare§

30-45

70-80

60

2-3

1

1-2

HbS-HPFH

Compound heterozygous HbS and gene deletion HPFH

Rare

38-45

70-80

70

2

20-30

Sickle cell trait (HbAS)¶

Heterozygous HbS

1 : 12

38-50

80-90

35-40

2-3

2g/dL) may be associated with worsening symptoms of fatigue and breathlessness, and intervention with transfusion may be required. Causes of acute anemia include worsening hemolysis, parvovirus (erythrovirus) B19 infection (potentially causing pure red cell aplasia or hypoplasia), splenic sequestration, folate deficiency, and any cause of anemia seen in non–sickle cell disease patients. Parvovirus B19 causes slapped-cheek syndrome in children. It usually manifests as an acute viral illness with profound anemia and reticulocytopenia. Splenic sequestration is most common in young children before splenic infarction and is characterized by anemia and splenic enlargement. Both should be treated with simple transfusion, aiming for a return to the patients’ baseline hemoglobin.

Neurologic Complications Acute Ischemic Stroke

Acute ischemic stroke (AIS) was previously a common complication in children and adults with sickle cell disease and an important cause of morbidity and mortality, with rates of approximately 10% by the age of 20 and 25% by 40 years. Historically, the highest incidence was in children between 2 and 9 years of age, with high rates also seen in those older than 40 years. Cerebral

1066

CHAPTER 154  Sickle Cell Disease and Other Hemoglobinopathies  

vasculopathy is the most common cause in children, but in adults, comorbidities include diabetes, hypertension, and atrial fibrillation. AIS can also occur in the context of acute pain crisis or acute anemia. The typical features of AIS (Chapter 379) are usually seen, including hemiparesis, facial droop, or dysphasia, but patients may present with loss of consciousness or behavioral change. Urgent neuroimaging should be organized, and magnetic resonance imaging (MRI) or magnetic resonance angiography examinations will confirm and localize the stroke, although computed tomography (CT) may be used in the emergency situation to rule out acute hemorrhagic stroke (see later). After initial management with hydration and oxygenation, urgent exchange transfusion should be performed aiming to minimize ischemic damage and improve perfusion of the brain. If there is marked anemia (hemoglobin 50 mg/mmol). Progressive renal dysfunction is seen with increasing age in some patients and can result in end-stage kidney disease and a requirement for dialysis or

1067

CHAPTER 154  Sickle Cell Disease and Other Hemoglobinopathies  

renal transplantation.12 Renal function should be monitored regularly to allow optimization of treatment of hypertension and proteinuria and thereby slow progression. Hematuria is also common in sickle cell disease, as well as in sickle cell trait (see earlier), and can be due to renal papillary necrosis. This can be painful and profuse, requiring treatment with blood transfusion and occasionally with interventional radiology or surgery. Hematuria in patients older than 40 years, particularly if painless, should always be referred to the urologist. Renal medullary carcinoma is a highly aggressive malignancy that predominantly afflicts young adults and adolescents with sickle cell hemoglobinopathies. Acute kidney injury occurs in 2 to 10% of those admitted with acute painful crisis and can be precipitated by dehydration, sepsis, and drugs. Renal function and urine output should be monitored in those admitted with pain crises.

Priapism

Priapism is a persistent, prolonged, and painful penile erection unrelated to sexual stimulation. It is common in men and boys with sickle cell disease and may occur as short, intermittent, self-limited episodes (stuttering priapism) or as severe acute episodes lasting for more than an hour (fulminant priapism). Acute priapism may be preceded by stuttering priapism and can result in long-term erectile dysfunction. Young men with sickle cell disease should be educated about priapism and encouraged to come to the emergency department for episodes lasting more than 1 hour. Initial conservative management includes pain relief, fluids, gentle exercise and urination, and oral α-adrenergic agents (etilefrine). If the priapism fails to resolve, a urologist should be contacted to perform penile aspiration; this can be combined with the injection of α-adrenergic agents into the corpus cavernosa. The procedure will often cause detumescence, but if not, further surgical intervention will be needed. Transfusion therapy may be helpful to support surgery and possibly to relieve persistent priapism. Treatment of stuttering priapism should be considered for symptomatic relief and with the aim of preventing future episodes of fulminant priapism. Oral α-adrenergic agents and antiandrogens should be considered under guidance of an expert urologist, and phosphodiesterase type 5 (PDE5) inhibitors may also be considered.  

DIAGNOSIS

Routine Blood Tests

The usual steady-state hemoglobin level is 5 to 10 g/dL in HbSS and is higher in the milder phenotypes. The anemia in HbSS is typically normochromic and normocytic, but microcytosis is often seen in HbSC disease, and marked microcytosis and hypochromia are seen in HbSβ-thalassemia. There is usually reticulocytosis and evidence of hemolysis with a raised bilirubin and lactate dehydrogenase (LDH). The blood film in HbSS will show numerous sickle cells, polychromasia, and features of hyposplenism (in adults). In addition, hypochromia is also prominent in HbSβ-thalassemia. In HbSC disease, target cells, densely staining contracted boat-shaped red cells, and HbC crystals may be seen with less sickled cells and polychromasia than in HbSS (Figs. 154-3Aii and 154-3Aiii).

Analysis of Hemoglobin Type

High-pressure liquid chromatography (HPLC) is now the most common method of hemoglobin analysis. Variant hemoglobins are identified by their elution time from an ion exchange column and can be quantified. HPLC is quick, is reliable, and allows high throughput of samples (Fig. 154-3B). Confirmatory testing should be performed with alternative techniques. These include the following: • The sickle solubility test will detect sickling of hemoglobin in solution by treatment with a reducing agent. It is specific for sickle hemoglobin but does not differentiate between HbAS, HbSS, and other compound heterozygotes of sickle cell hemoglobin. • Hemoglobin electrophoresis was the first method used to demonstrate hemoglobin variants but is time-consuming and labor-intensive. • DNA sequence analysis identifies the specific globin gene mutation (Fig. 154-3C). • Tandem mass spectrometry identifies the specific amino acid change in the globin chain.  

SCREENING

Antenatal and Neonatal Screening

Antenatal screening aims to identify couples at high risk for an affected pregnancy to enable counseling and prenatal diagnosis (via chorionic villus sampling

or amniocentesis) early enough to allow reproductive choice. Preconceptual screening provides couples with information about the risks of an affected birth before conception and allows them to consider the option of preimplantation diagnosis. The aim of neonatal screening is to identify babies with sickle cell disease before their first clinical presentation in order to improve clinical outcomes and reduce early mortality by prompt infection prevention and engagement with comprehensive care.  

TREATMENT

Infection and Infection Prophylaxis

Sickle cell patients experience increased morbidity and mortality from infection, in part due to functional asplenia or hyposplenia and also due to increased rates of gram-negative sepsis. Hyposplenia develops from infancy and leads to a deficiency in response to encapsulated bacteria and hence increased risk for invasive infection with organisms such as Streptococcus pneumonia, Neisseria meningitides, and Haemophilus influenzae type b. Early treatment with antibiotic prophylaxis and pneumococcal vaccination has decreased the risk for early mortality and morbidity from invasive pneumococcal disease, although vaccine-resistant serotypes can still cause serious infections.13 Infants identified by neonatal screening should be started promptly on antibiotic prophylaxis and vaccination programs.

General Measures

Sickle cell disease is a chronic disease for which good nutrition and adequate fluid intake should be encouraged. Patients and their families should be educated about acute complications and should be taught how to palpate the spleen, how to manage simple pain crises, and when to go to the emergency department. Regular outpatient clinic visits will ensure appropriate screening for chronic disease complications (including screening for stroke risk) and early intervention. The increased production of red blood cells, combined with inadequate nutrition, leads to accelerated consumption of folic acid; supplementary folic acid is typically recommended, although patients with a good dietary intake may not require it. There are significant psychosocial implications of chronic illness, and patients and their families should be offered psychological and social support if needed. The transition from pediatric to adult care is a time of heightened vulnerability and should be managed in a structured way.

Management of Pregnancy

While most women with sickle cell disease should expect to have successful pregnancies and healthy babies, these pregnancies are associated with an increased rate of maternal and fetal morbidity and mortality. These include increased sickle complications (pain, chest complications), pregnancy complications (infection, raised blood pressure), and high rates of premature birth and fetal growth restriction. Women should be counseled about pregnancy risk during their routine outpatient appointments, and preconceptual care should include partner screening, review of medications, and evaluation of chronic complications that may affect the pregnancy. When pregnant, women should be enrolled into multidisciplinary antenatal care from an obstetrician with experience in sickle cell disease and a hematologist. Caregivers should consider prescribing daily low-dose aspirin (75 mg) for prevention of preeclampsia. Patients should be followed regularly throughout pregnancy with blood pressure, urinalysis, and ultrasound monitoring. Prophylactic transfusion is not recommended for all pregnancies but may reduce the risk for pain crisis, chest complications, and fetal complications.14 If there are no obstetric complications, the pregnancy should be allowed to continue to term, anticipating spontaneous labor and normal delivery. Women with sickle cell disease are at increased risk for venous thrombosis, and their risk should be assessed using appropriate tools and the need for thromboprophylaxis considered.

Management of Surgery

General anesthesia is associated with increased risk for complications, in particular, increased pain crises and acute chest crises. These complications can be reduced with careful preoperative care (hydration, oxygenation) and consideration of preoperative blood transfusion. Simple transfusion aiming for a hemoglobin concentration of more than 9g/dL is as effective as exchange transfusion for low- and moderate-risk surgery, but high-risk surgery (cardiac or neurosurgery) will need preoperative exchange transfusion. A4  A5  In low- to moderate-risk surgery in patients with non-HbSS genotypes, transfusion may not be needed. ,

1068

CHAPTER 154  Sickle Cell Disease and Other Hemoglobinopathies  

Hydroxyurea

Until recently, hydroxyurea has been the only drug licensed by the U.S. Food and Drug Administration for treatment of sickle cell disease, and it works via an increase in HbF levels and reduction of cellular adhesion and hemolysis. Randomized controlled trials have shown that hydroxyurea reduces acute pain episodes, chest crises, and rates of hospitalization in adults and children with HbSS. Although initial studies focused the impact of hydroxyurea on those with severe phenotype, a subsequent randomized trial included young children with HbSS irrespective of disease severity. This trial showed similar improvements in outcome with few side effects. Furthermore, studies in cohorts around the world have shown improved survival in patients on hydroxyurea.15 In addition, hydroxyurea reduces stroke risk in those with elevated transcranial Doppler studies. National guidelines recommend that hydroxyurea be offered to all infants, children, and adolescents with sickle cell anemia, irrespective of severity, and to adults with recurrent pain episodes, severe or recurrent acute chest syndrome, and symptomatic anemia.16 Hydroxyurea is well tolerated; the main side effect is myelosuppression, and regular blood monitoring is required while on the drug. There is evidence of teratogenicity and effects on spermatogenesis, and although this is lowquality evidence, it should be avoided in those trying to conceive. Despite the evidence of efficacy and national guidelines advocating its use, hydroxyurea is underutilized.

Blood Transfusion

Red cell transfusion is used both to treat the acute complications of sickle cell disease and as long-term disease-modifying therapy. It can be administered as a simple or “top-up” transfusion to increase hemoglobin levels and improve oxygen-carrying capacity or as an exchange transfusion (manual or automated). Exchange transfusion allows a more marked reduction of HbS compared with HbA and modification of vaso-occlusion.17 Methodology of transfusion will depend on baseline hemoglobin, reason for transfusion, and availability. In view of the increased risk for alloimmunization and delayed hemolytic transfusion reactions in patients with sickle cell disease, blood should be ABO Rh (CcDEe) and Kell compatible and blood should be antigen-negative for clinically significant antibodies that are currently or have previously been detected. Acute indications for transfusion include acute anemia, acute chest syndrome, stroke, and multiorgan failure. Transfusion is not indicated for the treatment of simple acute pain crises. Long-term transfusion is indicated for primary and secondary stroke prevention as well for the prevention of recurrent pain episodes or recurrent acute chest syndrome if there has not been a response to hydroxyurea therapy. The use of transfusion for the management of sickle cell disease has increased over recent years, but there is considerable variability in its use in different centers. Consistency and equity of transfusion use should improve with national guidance.18 Long-term or frequent intermittent transfusion can result in iron overload and iron deposition in the liver and other organs. The patient should be monitored with ferritin and liver iron quantification by MRI19 and iron chelation therapy instigated if the patient is iron-loaded. Both deferasirox and desferrioxamine are licensed for use in sickle cell disease and are effective. Deferiprone is not currently licensed for use in sickle cell disease but is also effective.

Stem Cell Therapy

Successful hematopoietic stem cell transplantation (HSCT) is a potentially curative option for sickle cell disease but is associated with significant morbidity and mortality. The majority of HSCTs have been performed in children with HLA-matched siblings. In a study of 1000 patients with sickle cell disease transplanted from HLA-identical siblings, of the 846 who were younger than 17 years, 5-year overall survival (OS) was 95%, but it was 81% in the 154 older patients.20 The utility of this approach is limited because only about 10 to 20% of sickle cell disease patients will have HLA-identical siblings available as donors. Alternative donor approaches (particularly haplo-identical transplants) have shown good outcomes in small studies but are associated with increased morbidity and mortality and should be offered as part of clinical trials. There is increasing evidence of the efficacy of reduced-intensity conditioning approaches in adults, associated with reduced adverse effects.21

Gene Therapy

Gene therapy is another potentially curative option for sickle cell disease and involves manipulation of hematopoietic stem cells to correct or circumvent the sickle mutation followed by autologous transplantation of the manipulated

cells. Clinical trials are ongoing using addition of modified antisickling β-globin variants by lentiviral vectors. Data on the first patient, 15 months after treatment, confirmed that the therapeutic antisickling β-globin was more than 50%, with no recurrence of sickle crises and with improvement in hallmarks of disease.22 Gene-editing techniques using CRISPR/Cas9 technology to correct the causative β-globin gene mutation or to encourage production of fetal hemoglobin are also under investigation.

New Therapies

There are a number of promising agents currently in development for the acute or long-term treatment of sickle cell disease acting through different mechanisms. Rivipansel is a pan-selectin inhibitor that reduced the requirement for parenteral opiates in acute crisis23 in a phase II trial and now is being investigated in a phase III trial. Crizanlizumab, an anti–P selectin agent, showed significant decreases in median number of crises per year and in median time to first and second crises. A6  Other antiadhesive agents are currently under investigation, although trials of vepoloxamer and prasugrel did not show efficacy. A7  Voxelotor, a hemoglobin modifier, is currently being investigated in a phase 3 trial, and drugs that modify nitric oxide signaling, inhibit PDE9, and reduce oxidative stress are also undergoing investigation. A phase III trial with L-glutamine, an antioxidant, has shown a reduction in sickle cell crisis (three vs. four events) and in median incidence of hospitalization in the treated group and has just been licensed for use in the United States. A8   

PROGNOSIS

In resource-rich countries with universal neonatal screening, antibiotic prophylaxis, improved supportive care, and successful research trials, most children with sickle cell disease reach adulthood.24 In the United States, nearly 95% of persons with sickle cell disease reach age 18 years, although adults with the most severe forms of sickle cell disease still have a life span that is 20 to 30 years shorter than those without the disease. In contrast, in sub-Saharan Africa, which in 2010 had almost 80% of the projected 306,000 newborns with sickle cell disease worldwide, mortality in children ranged from 50 to 90% before age 5 years.

  OTHER HEMOGLOBINOPATHIES

Hemoglobin C trait is a benign condition, but homozygous HbC disease is associated with a mild hemolytic anemia and splenomegaly. As noted in Figure 154-3Aiii and in Chapter 148, HbC disease is characterized by a peripheral smear with a large population of target cells, with some red cells containing crystals. Individuals with HbC disease generally do not have clinical sickle crises or other complications. Hemoglobin E and the HbE thalassemias are covered in Chapter 153. Other than HbS, HbE, and HbC, most hemoglobinopathies do not cause clinical abnormalities. Among those that may have some clinical complications are some of the unstable hemoglobin variants, in which the mutant hemoglobin tends to precipitate intracellularly and causes a so-called Heinz body hemolytic anemia. These are covered in the section on “Unstable Hemoglobinopathies” in Chapter 153. Another group of hemoglobinopathies are globin chain mutations that cause increased binding of oxygen by the abnormal hemoglobin, that is, the high-oxygen-affinity hemoglobin variants. These individuals generally have a secondary erythrocytosis (see Chapter 157) as a compensatory response to their relative ineffectiveness in releasing O2 to tissues.25 High-oxygen-affinity hemoglobinopathies are best diagnosed by formal study of the patient’s oxyhemoglobin dissociation curve and P-50 values, which can be obtained with a commercially available instrument such as the Hemox-Analyzer, which measures in vitro deoxygenation of the sample using nitrogen gas and simultaneously determines oxygen tension with a Clarke electrode. Estimations of the P-50 by blood gas analyzers on venous blood are more convenient but have not been well validated in the screening of hemoglobin variants with high oxygen affinity. Despite their polycythemia, these patients do not have the other diagnostic characteristics of polycythemia vera (Chapter 157).

  Grade A References A1. Ware RE, Helms RW, the SWiTCH Investigators. Stroke with transfusions changing to hydroxyurea (SWiTCH). Blood. 2012;119:3925-3932. A2. DeBaun MR, Gordon M, McKinstry RC, et al. Controlled trial of transfusions for silent cerebral infarcts in sickle cell anemia. N Engl J Med. 2014;371:699-710. A3. Ware RE, Davis BR, Schultz WH, et al. Hydroxycarbamide versus chronic transfusion for maintenance of transcranial Doppler flow velocities in children with sickle cell anaemia—TCD with transfusions

changing to hydroxyurea (TWiTCH): a multicentre, open-label, phase 3, non-inferiority trial. Lancet. 2016;387:661-670. A4. Howard J, Malfroy M, Llewelyn C, et al. The Transfusion Alternatives Preoperatively in Sickle Cell Disease (TAPS) study: a randomised, controlled, multicentre clinical trial. Lancet. 2013;381:930-938. A5. Estcourt LJ, Fortin PM, Trivella M, et al. Preoperative blood transfusions for sickle cell disease. Cochrane Database Syst Rev. 2016;4:CD003149. A6. Ataga KI, Kutlar A, Kanter J, et al. Crizanlizumab for the prevention of pain crises in sickle cell disease. N Engl J Med. 2017;376:429-439. A7. Heeney MM, Hoppe CC, Abboud MR, et al. A multinational trial of prasugrel for sickle cell vasoocclusive events. N Engl J Med. 2016;374:625-635. A8. Niihara Y, Miller ST, Kanter J, et al. A phase 3 trial of l-glutamine in sickle cell disease. N Engl J Med. 2018;379:226-235.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 154  Sickle Cell Disease and Other Hemoglobinopathies  

GENERAL REFERENCES 1. Kato GJ, Piel FB, Reid CD, et al. Sickle cell disease. Nat Rev Dis Primers. 2018;4:1-22. 2. Sundd P, Gladwin MT, Novelli EM. Pathophysiology of sickle cell disease. Annu Rev Pathol. 2019;14:263-292. 3. Kotila TR. Sickle cell trait: a benign state? Acta Haematol. 2016;136:147-151. 4. Singer DE, Byrne C, Chen L, et al. Risk of exertional health illnesses associated with sickle cell trait in U.S. military. Mil Med. 2018;183:e310-e317. 5. Lampert R, Zipes DP. Updated recommendations for athletes with heart disease. Annu Rev Med. 2018;69:177-189. 6. Naik RP, Smith-Whitley K, Hassell KL, et al. Clinical outcomes associated with sickle cell trait: a systematic review. Ann Intern Med. 2018;169:619-627. 7. Thein MS, Igbineweka NE, Thein SL. Sickle cell disease in the older adult. Pathology. 2017; 49:1-9. 8. Kozanoglu I, Ozdogu H. Use of red blood cell exchange for treating acute complications of sickle cell disease. Tranfus Apher Sci. 2018;57:23-26. 9. Adams RJ, Cox M, Ozark SD, et al. Coexistent sickle cell disease has no impact on the safety or outcome of lytic therapy in acute ischemic stroke. Stroke. 2017;48:686-691. 10. Kassim AA, Pruthi S, Day M, et al. Silent cerebral infarcts and cerebral aneurysms are prevalent in adults with sickle cell anemia. Blood. 2016;127:2038-2040. 11. Howard J, Hart N, Roberts-Harewood M, et al. Guideline on the management of acute chest syndrome in sickle cell disease. Br J Haematol. 2015;169:492-505. 12. Airy M, Eknoyan G. The kidney in sickle hemoglobinopathies. Clin Nephrol. 2017;87:55-68. 13. Oligbu G, Collins S, Sheppard C, et al. Risk of invasive pneumococcal disease in children with sickle cell disease in England: a National Observational Cohort Study, 2010-2015. Arch Dis Child. 2017; 103:643-647. 14. Malinowski AK, Shehata N, D’Souza R, et al. Prophylactic transfusion for pregnant women with sickle cell disease: a systematic review and meta-analysis. Blood. 2015;126:2424-2435.

1069.e1

15. Lê PQ, Gulbis B, Dedeken L, et al. Survival among children and adults with sickle cell disease in Belgium: benefit from hydroxyurea treatment. Pediatr Blood Cancer. 2015;62:1956-1961. 16. Qureshi A, Kaya B, Pancham S, et al; On behalf of the British Society of Haematology. Guidelines for the use of hydroxycarbamide in children and adults with sickle cell disease: a British Society for Haematology Guideline. Br J Haematol. 2018;181:460-475. 17. Davis BA, Allard S, Qureshi A, et al; the British Committee for Standards in Haematology. Guidelines on red cell transfusion in sickle cell disease. Part I: principles and laboratory aspects. Br J Haematol. 2017;176:179-191. 18. Davis BA, Allard S, Qureshi A, et al; on behalf of the British Society for Haematology. Guidelines on red cell transfusion in sickle cell disease part II: indications for transfusion. Br J Haematol. 2017;176:192-209. 19. Labranche R, Gilbert G, Cerny M, et al. Liver iron quantification with MR imaging: a primer for radiologists. Radiographics. 2018;38:392-412. 20. Gluckman E, Cappelli B, Bernaudin F, et al. Sickle cell disease: an international survey of results of HLA-identical sibling hematopoietic stem cell transplantation. Blood. 2017;129:1548-1556. 21. Fitzhugh CD, Abraham A, Hsieh MM. Alternative donor/unrelated donor transplants for the β-thalassemia and sickle cell disease. Adv Exp Med Biol. 2017;1013:123-153. 22. Ribeil JA, Hacein-Bey-Abina S, Payen E, et al. Gene therapy in a patient with sickle cell disease. N Engl J Med. 2017;376:848-855. 23. Sins JWR, Mager DJ, Davis SCAT, et al. Pharmacotherapeutical strategies in the prevention of acute, vaso-occlusive pain in sickle cell disease: a systematic review. Blood Adv. 2017;1:1598-1616. 24. Minniti CP, Vichinsky E. Lifespan care in SCD: whom to transition, the patients or the health care system? Am J Hematol. 2017;92:487-489. 25. Orvain C, Joly P, Pissard S, et al. Diagnostic approach to hemoglobins with high oxygen affinity: experience from France and Belgium and review of the literature. Ann Biol Clin (Paris). 2017;75: 39-51.

CHAPTER 155  Megaloblastic Anemias  

1069

infants can have severe developmental and growth impairment. There is an efficient enterohepatic circulation of cobalamin which prevents loss of the vitamin so that those on deficient or marginal diets will take years to become deficient if the gastrointestinal tract is functioning. Cobalamin is a trace component of food-bound proteins in the diet and most causes of severe deficiency relate to the inability to absorb the vitamin. The failure of secretion of gastric intrinsic factor due to an autoimmune disease of the gastric mucosa is known as pernicious anemia and results in lifethreatening anemia, pancytopenia, or myeloneuropathy unless treated. Surgery or disease of the stomach, pancreas, or ileum also will cause deficiency. The normal processes of cobalamin absorption are shown in Figure 155-1.

TABLE 155-2 CAUSES OF MEGALOBLASTIC ANEMIA COBALAMIN (VITAMIN B12) DEFICIENCY

155  MEGALOBLASTIC ANEMIAS SALLY P. STABLER



DEFINITION

Megaloblastic anemia is defined as an anemia or pancytopenia resulting from failure of DNA synthesis and imbalance of maturation between the nucleus and the cellular cytoplasm resulting in large cells subject to maturation arrest and early destruction. Large, macrocytic, red blood cells are also seen in conditions other than megaloblastic anemia, as shown in Table 155-1. In practical terms, however, many of the causes of macrocytic red blood cells overlap with those of full-blown megaloblastic anemias. The most frequent causes of severe megaloblastic anemia are deficiencies of cobalamin (vitamin B12) and folate, two vitamins with interrelated roles in the synthesis of thymidine and purines for DNA synthesis1 (Table 155-2).  

EPIDEMIOLOGY

Cobalamin Deficiency

Cobalamin is synthesized only by microorganisms, and humans obtain the vitamin through ingestion of animal-source foods because plants do not utilize cobalamin.2,3 There are many populations in the world that do not consume adequate cobalamin-containing food due to scarcity and expense of animalsource food or belief systems against the use of such foods. Recommended daily allowances in the United States have been set at 2.4 µg/day, which is readily achieved with an omnivorous diet containing meat, dairy, eggs, fish, or shellfish. Foods targeted to vegetarian populations in high-resource countries usually are supplemented with cobalamin. The fetus and breast-feeding infant obtain cobalamin from the mother, and thus requirements are increased to 2.6 µg for pregnancy and 2.8 µg for lactating women. The breast milk of cobalamin-deficient women has low concentrations of the vitamin and their

TABLE 155-1 CAUSES OF MACROCYTOSIS NONMEGALOBLASTIC DISORDERS Reticulocytosis Alcohol abuse/heavy use Chronic liver disease Thyroid disorders (hypothyroidism) Chronic lung disease with hypoxemia Cold agglutinins and paraproteinemias Hyperglycemia Primary bone marrow disorders (e.g., myelodysplasia) MEGALOBLASTIC ANEMIAS (SEE TABLE 155-2) Cobalamin (vitamin B12) deficiency Folate deficiency Other megaloblastic anemias

Lack of intrinsic factor Pernicious anemia (type A gastritis) Gastrectomy, gastric bypass Congenital defect in gastric intrinsic factor Food cobalamin malabsorption Atrophic gastritis (type B gastritis) Gastric hypo- or achlorhydria Exocrine pancreatic dysfunction Disorders of the distal ileum Surgical resection of distal 1-2 ft. Inflammatory bowel disease Surgical reconstruction—ileal conduit or cystoplasty Congenital defects of cubam receptor (Imerslund-Gräsbeck syndrome) Usurption of intestinal cobalamin Tropical sprue, bacterial overgrowth, blind loops Parasitic infection—Diphyllobothrium latum and Giardia lamblia Nutritional cobalamin deficiency Breast-fed infant of a deficient mother Diet lacking animal source food, vegan, lacto-ovo vegetarian, or impoverished low-meat diet Drug-induced cobalamin deficiency Nitrous oxide abuse or anesthesia in a cobalamin-deficient patient Metformin, histamine blocking or proton pump inhibitor drugs Inborn errors of cobalamin metabolism Transport—transcobalamin deficiency Cellular metabolism—mutations that cause combined methylmalonic acidemia and homocystinuria (CblC-J) FOLATE DEFICIENCY Nutritional folate deficiency Lack of fresh fruit, green leafy vegetables, legumes, organ meats Anorexia nervosa and other eating disorders Alcohol abuse combined with poor nutrition Folate malabsorption Celiac or tropical sprue Crohn disease and other small intestinal diseases Increased demand or loss Pregnancy Hemolytic anemias Proliferative skin diseases Leukemias Hemodialysis Drug-caused folate deficiency Antifolates—methotrexate, pemetrexed, pralatrexate, pyrimethamine, sulfasalazine, trimethoprim Anticonvulsants—phenytoin, carbamazepine, valproic acid Methylation-requiring drugs—L-Dopa, high-dose niacin, α-lipoic acid Inborn errors of folate metabolism MTHFD1 deficiency PCFT (SLC46A1) mutation MEGALOBLASTIC ANEMIA NOT DUE TO COBALAMIN OR FOLATE DEFICIENCY Inborn errors Thiamine responsive megaloblastic anemia (SLC19A2 mutation) Hereditary orotic aciduria Lesch-Nyhan syndrome Drugs affecting DNA synthesis Cancer chemotherapy—hydroxyurea, azathioprine, others Antiretrovirals—zidovudine, stavudine Tyrosine kinase inhibitors—imatinib, sunitinib, others CblC-J = cobalamin C, D, E, F, G, and J (defects).

1069.e2

ABSTRACT

CHAPTER 155  Megaloblastic Anemias  

The most common causes of megaloblastic anemia are cobalamin (B12) or folate deficiencies. Methylmalonic acid is the most sensitive and specific marker of cobalamin deficiency. Folate deficiency is rare in food folate fortified countries and is usually associated with alcohol abuse. Cobalamin deficiency can be treated with injections or high-dose oral preparations. The severity of myeloneuropathy due to cobalamin deficiency is inversely related to the severity of megaloblastic anemia but is reversible if treated early.

KEYWORDS

vitamin B12 cobalamin folate methylmalonic acid homocysteine

1070

CHAPTER 155  Megaloblastic Anemias  

Cobalamin absorption and effects Food • Vegetarian diet • Deficient breast milk

Food Cbl Saliva Enterohepatic circulation HC Liver

IF HC + Cbl Acid

Bile HC-Cbl

Pepsin HC-Cbl HC-Cbl

HC Cbl IF

Jejunum Proteases • Bacterial overgrowth IF-Cbl • Parasites IF-Cbl • Celiac disease Ileum • Ileal resection • Crohn disease Lysosome LMBD1 ABCD4

Stomach • Pernicious anemia (type A gastritis) • Partial or total gastrectomy • Gastric bypass, other bariatric surgery • Atrophic gastritis • Acid blocking drugs Pancreas • Exocrine pancreatic insufficiency

IF-Cbl Cubilin Amnion-less

Ileal cell • Imerslund-Gräsbeck syndrome • Cbl-F disease

IF-Cbl ? TC-Cbl

Blood vessel • Transcobalamin deficiency

FIGURE 155-1.  The normal processes of cobalamin (Cbl) absorption and associated defects are shown. ABCD4 = ATP binding cassette subfamily D member 4 (defect); Cbl = cobalamin; Cbl-F = cobalamin F (defect); HC = haptocorrin (also called R-protein or R-factor); IF = intrinsic factor; LMBD1 = probable lysosomal cobalamin transporter (defect); TC = transcobalamin. (Adapted from Stabler SP, Megaloblastic anemias: pernicious anemia and folate deficiency. In: Young NS, Green SL and High KA [Eds.] Clinical Hematology, Philadelphia, PA: Mosby Elsevier; 2006:242-251.)

The incidence of pernicious anemia is reported to range from 1 to 50 per 100,000 persons per year, depending on the criteria used for diagnosis. The prevalence has been described as 50 to 4000 cases per 100,000 persons, again depending on diagnostic criteria. Pernicious anemia becomes more common with age, and the highest prevalence occurs in persons of African or European ancestry with lower incidence and prevalence in East Asians. Mild atrophic gastritis with protein-bound cobalamin malabsorption occurs in up to 20% of older adults, particularly Caucasians. Dietary cobalamin deficiency is population dependent, but can range from 30 to 50% in areas of South America, Africa, and India.

Folate Deficiency

In contrast to cobalamin, folates are abundant in foods of plant origin and are also found in dairy and organ meats. Folates are susceptible to loss in cooking, particularly if the cooking water is discarded and not incorporated into the dish. Populations with limited access to or distaste for vegetables, legumes, or fruits may have folate deficiency.4 Alcohol and some other drugs interfere with folate absorption or promote excretion and are major causes of folate deficiency seen clinically. Folates are absorbed mainly in the upper small intestine; thus, diseases such as celiac or tropical sprue, bacterial overgrowth or inflammatory bowel diseases are complicated by clinical folate deficiency. Many countries including the United States, Canada, Chile, Costa Rica, and South Africa have fortified grain products with folic acid for the last 20 years. This has virtually eliminated folate deficiency except in those persons with impaired absorption or high alcohol intake. Countries without fortification may have a significant folate-deficient population depending on customary dietary food patterns which in resource-poor countries provide less than 50% of recommended folate intake.4 A recommended daily allowance for folate in the United States is 400 µg for adult men and nonpregnant women. The requirement has been increased to 600 µg for pregnant women and 500 µg for lactating women. Food folates are polyglutamated and are less bioavailable than the synthetic folic acid used

to fortify food and supplements, which has made estimation of total folate intake more difficult in fortified populations.  

PATHOBIOLOGY

Cobalamin

Cobalamin is necessary for only two enzymatic reactions in higher animals. In one reaction methyl-cobalamin is a cofactor for the enzyme methionine synthase, which uses methyl-tetrahydrofolate (methyl-THF) to methylate homocysteine to form methionine. Methionine can be activated to S-adenosylmethionine, which is the methyl donor for many reactions including the important synthesis of creatine, lipids, neurotransmitters, DNA methylation, and many other compounds. The demethylated folate product, THF, is the precursor for the other folate coenzymes. One coenzyme, 5, 10-methyleneTHF, participates in the formation of thymidine from dUMP by thymidylate synthase. Thymidine is critical for DNA synthesis and the accumulation of uracil is harmful; if incorporated into DNA it can cause prolongation of the cell cycle, chromosomal damage, and maturation arrest. Apoptosis may occur due to the impaired DNA repair with cell fragmentation and resulting cell death. The buildup of methyl-THF, which cannot be converted to THF due to the decrease in methionine synthase activity, is referred to as the “methylfolate trap.” This causes a secondary state of folate deficiency in the cobalamindeficient patient. Methyl-THF is not a good substrate for polyglutamation and therefore it diffuses out of cells and will be lost in the urine. The impairment in methionine synthase activity is the primary cause for the indistinguishable megaloblastic anemia seen with either cobalamin or folate deficiency. When methionine synthase activity is impaired serum homocysteine levels increase (Chapter 198). The second cobalamin-dependent reaction uses adenosyl-cobalamin to convert a product of propionic acid metabolism, L-methylmalonyl-CoA, to succinyl-CoA by the enzyme L-methylmalonyl-CoA mutase. When the mutase is blocked by deficiency of adenosyl-cobalamin, the methylmalonyl-CoA is

CHAPTER 155  Megaloblastic Anemias  

hydrolyzed to methylmalonic acid, a compound with unknown metabolic fate. Elevated methylmalonic acid has been shown to be a sensitive indicator of impaired mutase activity and thus of cobalamin deficiency in many experimental cellular or animal models as well as in human patients. There are no roles for folate metabolism in this pathway.

Cobalamin Absorption

The cobalamin in unfortified food consists of the coenzyme forms bound to the two enzymes. Because there are only trace amounts of cobalamin (average 4 to 6 µg in omnivorous diets) there is a multistep process involving binding proteins and receptors to provide for absorption of the vitamin. The normal processes of absorption are shown in Figure 155-1 and reviewed elsewhere in detail.5 Food is mixed with saliva and digested in the stomach where pepsin and gastric acid release cobalamin from the food protein allowing it to bind to haptocorrin (HC), also called R-protein or R-factor, a glycoprotein present in saliva. The HC-cobalamin complex moves to the duodenum where it is joined by HC-cobalamin recycled from bile. Pancreatic proteases with bicarbonate digest the HC releasing cobalamin to be bound by intrinsic factor (IF), which had been secreted by the gastric parietal cells. IF binding is specific for true cobalamin and does not bind the cobalamin analogues found widely in the environment. The IF-cobalamin complex travels to the distal ileum where it is bound by the cubam receptor, a complex of cubulin and amnionless. The IF-cobalamin complex is internalized in lysosomes and two specific exporters (products of the LMBD1 and ABCD4 genes) are required to export it into the cytoplasm. Eventually cobalamin is bound to transcobalamin (TC), which is the physiologic delivery protein. A specific transcobalamin receptor (CD 320) facilitates cellular uptake wherein the TC is degraded by lysosomes and the cobalamin becomes available for intracellular processes. There are many intracellular transporters and enzymes responsible for the eventual appearance of the coenzyme forms of methyl-cobalamin on methionine synthase and adenosyl-cobalamin on L-methylmalonyl-Co-A mutase. The study of congenital defects in these enzymes, which are designated Cbl A-Cbl J, has elucidated much of this complicated metabolism. Some of these disorders are included in Table 155-2.

Cobalamin Transport

About 80% of total cobalamin in blood is bound to HC and only 10 to 20% to the delivery protein, TC. TC is known to be the most important carrier because inborn errors in TC lead to infantile megaloblastic anemia and severe disability. In contrast, deficiencies of HC are moderately common and do not appear to result in disease. HC binds analogues of cobalamin that are abundant in the human colon, foods, and the environment and may have a role in sequestering the analogues to facilitate their excretion. HC is present in white blood cells such as granulocytes. HC-cobalamin is taken up by the asialoglycoprotein receptors in the liver, which may facilitate the removal of analogues by secretion into the bile. There are problems in interpreting total cobalamin values because the cobalamin bound to TC is a small part of the circulating total, as discussed in the section on Diagnosis.

Malabsorption of Cobalamin as a Cause of Deficiency Pernicious Anemia

Pernicious anemia was recognized many decades before cobalamin was identified because of the histamine-fast achlorhydria found in some patients with fatal progressive macrocytic pancytopenia. It is misnamed because it is a gastric disease, and many patients may not even have anemia. Pernicious anemia is due to a chronic inflammatory attack on the mucosa of the fundus and body of the stomach in which there are inflammatory infiltrates and an immune response to gastric H/K ATPase in the parietal cells.6 With atrophy of the gastric mucosa there is loss of the gastric parietal cells which produce IF in addition to stomach acid.7 This type A gastritis takes many years to develop and initially may cause iron deficiency anemia before the lesion is severe enough to result in the loss of IF. There may be production of anti-intrinsic factor antibodies, which can also block absorption of cobalamin. The cobalamin secreted into bile is lost in pernicious anemia because there is no IF in the duodenum to bind the cobalamin. Thus, instead of the usual normal loss of about 1 µg of cobalamin per day there may be 5 to 9 µg lost, greatly shortening the time to develop deficiency. Helicobacter pylori is another cause of chronic gastritis, although it mostly affects the antrum of the stomach (Type B gastritis). Whether H. pylori plays a role in the progression of pernicious anemia and cobalamin deficiency is still undefined.

1071

Pernicious anemia is found in association with other autoimmune disorders such as endocrinopathies like autoimmune thyroid disease, Type I diabetes mellitus, and vitiligo. Relatives of patients with pernicious anemia may have the full spectrum of autoimmune diseases. The relationship between Graves disease and pernicious anemia is especially strong with about a quarter of the patients having both illnesses. It is important that women with autoimmune thyroid disease have adequate cobalamin status prior to pregnancy and throughout lactation. Protein-Bound Cobalamin Malabsorption

Less severe forms of gastric atrophy appear in older adults, most frequently in those of European ancestry, wherein a loss of pepsin and gastric acid prevents the total release of cobalamin from protein binding in food. Stomach acid–blocking medications may also impair the absorption of protein-bound cobalamin. It is difficult to know what the actual risk is since these medications are used frequently in the aging population that also has age-related gastric atrophy. Patients with gastric atrophy may be identified by serum markers as discussed under Diagnosis. Cobalamin Malabsorption Due to Surgery

Total gastrectomy invariably results in severe cobalamin deficiency due to the loss of IF after 1 to 5 years depending on the underlying status of cobalamin stores. Bariatric surgery such as the Roux-en-Y gastric bypass and sleeve gastrectomy cause the food-cobalamin stream to be separated from the cells secreting IF because the stomach is bypassed. The cubam receptor is present in the distal 1 to 2 feet of ileum, so its loss due to inflammatory bowel disease or the surgical use of distal ileum to create an ileal conduit diversion or ileal cystoplasty causes cobalamin deficiency. Other Causes of Malabsorption

Exocrine pancreatic dysfunction may cause failure of digestion of HC and release of cobalamin for binding by IF. Bacterial overgrowth as seen in tropical sprue has been shown to cause both cobalamin and folate deficiency. Infestation with a freshwater fish tapeworm, Diphyllobothrium latum, is an uncommon cause of usurpation of cobalamin in the duodenum. Giardia lamblia is a more common protozoal intestinal infection that may cause decreased cobalamin absorption. Congenital Malabsorption of Cobalamin

Abnormal gastric IF (GIF) manifests as megaloblastic anemia in infancy or childhood, depending on the underlying cobalamin status at birth. Much more common disorders include mutations in the cubam receptor which have been found in either the cubulin or amnion-less protein. This was initially described in Scandinavia and was given the name Imerslund-Gräsbeck disease, but it is particularly common in the Middle East. It is not known whether heterozygotes for these mutations might be more predisposed to nutritionally or atrophic gastritis-induced cobalamin deficiency. Cobalamin Deficiency Due to Drugs

Many studies have reported on the effects of metformin on cobalamin status. Although it does appear to decrease serum cobalamin levels and may cause malabsorption in the ileum, it is not known whether it can induce full-blown megaloblastic anemia. The anesthetic agent nitrous oxide is a definite cause of cobalamin deficiency, usually manifesting as myeloneuropathy. The gas inactivates methionine-synthase and increases excretion of cobalamin, therefore repeated use induces a severe deficiency. Prolonged continuous use will cause bone marrow aplasia in about 10 days. Persons at risk are often employed in dental or medical offices where nitrous oxide is available or abuse nitrous oxide–containing whipping cream dispensers. There are also reports of a subacute neurologic progression of cobalamin deficiency occurring in patients with unrecognized pernicious anemia after only one episode of nitrous oxide anesthesia, symptoms occurring over days to weeks. Often neurologic damage is permanent despite adequate replacement with cobalamin.

Folate Folate Metabolism

The role of methyl-THF as a cofactor for methionine synthase was described above. However, there are other folate coenzymes that function as cofactors for many enzymes involved in amino acid metabolism, purine synthesis, methylation reactions, one carbon metabolism, and regulatory functions.5 Inhibition of these reactions due to folate deficiency can cause decreased cellular growth and proliferation, which is exploited in the use of antifolates for cancer

1072

CHAPTER 155  Megaloblastic Anemias  

chemotherapy. One cellular form, 5,10-methyleneTHF, is at a branch point to be used by thymidylate synthase to produce the thymidine necessary for DNA synthesis, or is reduced irreversibly by 5,10-methylenetetrahydrofolate reductase (MTHFR) to methyl-THF. As discussed above, methyl-THF must be converted by cobalamin-dependent methionine synthase before the folate can be used in other reactions. A common polymorphism in MTHFR (C677T), known as the thermolabile mutation, renders the enzyme somewhat less efficient and results in higher homocysteine and lower folate levels when folate intake is low. The prevalence of the polymorphism is quite variable, ranging from less than 1% in Africans to 30% in Mediterranean, Mexican, and Northern Chinese populations. Its association with risk for various human disorders has been extensively studied but not clearly defined. The polymorphism changes the proportions of folate coenzyme forms, decreasing methyl-THF, and thus circulating folate levels. The increase in homocysteine levels and decrease in folate levels is most apparent in populations on low or marginal folate intakes.

Folate Absorption and Transport

Natural food folates are polyglutamated and must be hydrolyzed by an upper small intestinal brush border enzyme, folylpolyglutamate hydrolase. Folate monoglutamate is transported into the cells by a surface protein–coupled folate transporter (PCFT). Folic acid (nonreduced folate), a stable form, is used in supplements and fortified grains and has a bioavailability of about 90%. Folic acid is reduced to THF and then methylated and released into the plasma. The reduced folate carrier (RFC, solute carrier family, SLC19A1) transports folates into and out of the cell. Absorbed folates in the portal circulation are taken up in the liver, polyglutamated, and retained or released into the circulation or bile. Circulating folates are bound to proteins, particularly albumin. Red blood cells accumulate folates during erythropoiesis, and thus the red cell folate level may more constantly reflect long-term folate status than serum folate. Concentrations of folates in CSF are three-fold higher than in blood, and the choroid plexus has abundant receptors for uptake and transport.

Causes of Nutritional Folate Deficiency

Nutritional folate deficiency occurs in the areas of the world that do not currently have folate food fortification. The foods containing the highest amount of naturally occurring folate are dark green leafy vegetables, beans and other legumes, nuts, orange juice, and liver or other organ meats. Potatoes, many fruits, and eggs have moderate amounts of folate but dairy items and muscle meat are much lower in folate. In high-resource countries, cereals, snack bars, sport drinks, and other foods promoted for health and exercise are often fortified with folic acid. Natural food folate intake has been estimated to be roughly 200 µg/day with an extra 200 µg in nonsupplement users from the added folic acid in fortified grains in persons in the United States. Total folate in supplement users was much higher, between 700 to 1000 µg per day. Low serum folate was found in approximately 15% in surveys of the United States population prior to the food folate fortification program. Nutritional folate deficiency is common in areas of the world with limited opportunity for consumption of fresh vegetable foods and is particularly a problem for women of childbearing age who require increased quantities to prevent folate deficiency–associated pregnancy complications. Folate-deficient diets are often deficient in other nutrients such as vitamin C, iron, and other vitamins and minerals. Malabsorption of Folate

The serum folate level decreases within 3 weeks of severe folate malnutrition or malabsorption and megaloblastic bone marrow changes occur after 15 to 20 weeks. Folate malabsorption occurs in diseases of the upper small intestine and is rarely seen after gastrectomy or with Roux-en-Y gastric bypass. Celiac disease, which causes villous atrophy of the small intestine, causes folate deficiency in up to half of the patients. Tropical sprue, a cause of chronic diarrhea and progressive small intestinal villous atrophy, also causes folate deficiency with or without cobalamin deficiency. Inflammatory bowel disease, particularly Crohn disease, causes folate deficiency in addition to iron and vitamin B12 deficiency. Conditions with Increased Folate Requirements

Diseases in which there is increased cellular proliferation may result in folate depletion. This includes skin diseases such as psoriasis and chronic hemolytic anemias, particularly sickle cell disease. Folates and other water-soluble vitamins are lost during renal hemodialysis.

Alcohol and Folate Deficiency

Alcohol abuse is the most common cause of folate deficiency manifesting megaloblastic anemia.8 Even well-nourished alcoholics may have macrocytosis and elevated homocysteine values, often when serum folate and hemoglobin are still normal. Alcohol-induced impairment of folate metabolism is exacerbated by poor nutrition. Megaloblastic anemia due to alcohol-associated folate deficiency still occurs despite the food folate fortification program in the United States. Folate-deficient alcoholics have been shown to have impaired intestinal absorption of folate, oxidative destruction of folates, impaired hepatic uptake, reduced storage, and enhanced renal excretion. Effects on the enzymes of methionine metabolism, particularly methionine synthase, are likely due to the toxic metabolites of ethanol. It may be difficult to recognize megaloblastic anemia in alcoholics because their frequently coexisting iron deficiency due to gastrointestinal bleeding may mask macrocytosis. In addition, alcohol induces sideroblastic anemia, an iron utilization defect, which can also mask macrocytosis. Nonalcohol related causes of cirrhosis can cause macrocytosis due to membrane abnormalities in the red cells, even with adequate folates. Drugs That Cause Folate Deficiency

There are many drugs that interfere with folate metabolism, as shown in Table 155-2. The anticonvulsants, particularly carbamazepine, phenytoin, valproate, and phenobarbital have all been shown to cause macrocytosis, folate depletion, and hyperhomocysteinemia. Folic acid supplements improve folate status without preventing the antiseizure efficacy. There are many antifolates in common use, being exploited for their inhibitory effect on cellular proliferation such as methotrexate, pemetrexed, and pralatrexate used in cancer chemotherapy and in rheumatic diseases. It is interesting that the beneficial effect of methotrexate in rheumatoid arthritis is not counteracted by providing supplemental folic acid. Some antibiotics such as trimethoprim, pyrimethamine, and some sulfa derivatives are antifolates. Sulfasalazine used for inflammatory bowel disease will also cause folate depletion. Functional folate deficiency can be induced by drugs that require methylation for metabolism by depleting the normal stores of methyl groups. Drugs such as L-dopa, high-dose niacin, and an over-the-counter agent, alpha-lipoic acid, all raise homocysteine. Inborn Errors of Folate Metabolism

There are two inborn errors of folate that present as megaloblastic anemia: methylenetetrahydrofolate dehydrogenase deficiency (MTHFD1) and hereditary folate malabsorption due to mutations in the protein-coupled folate transporter (PCFT).9 MTHFD1 deficiency has been found in six infants or children with megaloblastic anemia, immunodeficiency, or inflammatory conditions and hyperhomocysteinemia. Treatment with high doses of folic acid or folinic acid were beneficial in several of the patients. It is not yet known whether this disorder will be found in adult medicine. Severe deficiencies of MTHFR (not the common thermolabile polymorphism) cause hyperhomocysteinemia with low methionine, but not megaloblastic anemia. These patients may present with thrombotic complications of hyperhomocysteinemia (Chapter 73) and neurologic deficits (Chapter 388). Treatment with betaine and folic acid greatly improves the prognosis.  

CLINICAL MANIFESTATIONS

Definition of Macrocytosis

The best definition of macrocytosis is red blood cell mean cell volume (MCV) that is higher than the expected value for any individual patient. Clinicians who rely on laboratory reported ranges such as 80 to 100 f L will make errors in diagnosis because the MCV is affected by ethnicity and age. For instance, 30% of persons with African ancestry in the United States have one deleted α hemoglobin subunit and their median MCV is about 84 f L. It would take much longer for these individuals to increase the MCV to greater than 100 f L than someone whose normal MCV was 92 f L. Persons with β-thalassemia trait, hemoglobin E, and iron deficiency will also not usually have an increase in MCV to more than 100 f L despite the coexistence of severe megaloblastic anemia. Individuals with a combination of microcytic red cell indices and megaloblastosis may have oval macrocytes visible on the peripheral blood smear and a high red cell distribution width (RDW). The causes of macrocytosis are shown in Table 155-1. As indicated, macrocytosis is not equivalent to megaloblastosis: there are many causes of macrocytic peripheral red blood cell indices that are not associated with a megaloblastic state. Reticulocytes, newly released red cells that are less than 24 hours old, are at their largest size and decrease as they age over 120 days

CHAPTER 155  Megaloblastic Anemias  

in the circulation. Thus, reticulocytosis will increase the MCV, in many cases to over 100 f L. A rise in MCV may be the first sign of autoimmune hemolytic anemia and is also commonly found in patients with fragmentation anemia, sickle cell anemia, or with hemolysis due to infectious agents. Macrocytosis can be seen in patients who are post-splenectomy or have liver disease of any type, especially with associated alcoholism. Severe portal hypertension causes macrocytosis with thrombocytopenia and leukopenia mimicking a true megaloblastic anemia when examining only the peripheral blood smear. Chronic lung disease, particularly with associated hypoxia-induced polycythemia, is frequently macrocytic. Hypothyroidism and severe hyperglycemia also cause macrocytosis. The primary bone marrow disorders such as aplastic anemia, pure red cell aplasia, myelodysplastic syndrome, idiopathic myelofibrosis, and multiple myeloma may cause striking macrocytosis, with or without associated cytopenias. Cold agglutinins and paraproteinemias cause red cell clumping, and respectively, rouleau and may be associated with an erroneously high MCV. Drugs with effects on DNA synthesis such as antiretrovirals, chemotherapy agents, and antifolates cause macrocytosis.

Pathologic Abnormalities in Megaloblastic Anemia

Examples of peripheral blood smears and bone marrow aspirate and biopsy in megaloblastic anemia are shown in Figure 155-2A-D (also see Chapter 148). The hallmark of megaloblastic pathology in the marrow is that cells are large (“megalo”) and their nuclei have not matured (“blastic”); thus there are oval macrocytes and occasionally even a megaloblastic nucleated red blood cell detected on the peripheral blood smear. Other red blood cells are misshapen

1073

and fragmented and can be mistaken for schistocytes as seen in fragmentation anemias. There may be thrombocytopenia and leukopenia when anemia becomes severe. Hypersegmented granulocytes are present. The presence of one 6-lobed or five 5-lobed granulocytes per 100 cells is considered abnormal. Some red blood cells may show basophilic stippling. A megaloblastic blood smear with oval macrocytes and fragmented red blood cells with thrombocytopenia has frequently been mistaken for microangiopathic hemolytic anemia such as thrombotic thrombocytopenia, as documented by case reports.10 Reticulocytopenia in megaloblastic anemias differentiates these microangiopathic-appearing red cells in the peripheral smear from hemolytic anemias.11 The bone marrow biopsy and aspirate smear is hypercellular with a decreased myeloid:erythroid ratio. There is delayed maturation of erythroblasts, which to the inexperienced observer may even suggest a diagnosis of acute leukemia because they are large cells with open chromatin and blue cytoplasm. The normal process of shrinking and condensing of the nucleus during erythropoiesis is impaired so that the chromatin pattern remains open and lacy (i.e., “megaloblastic”). Although nuclear maturation is delayed, cytoplasmic maturation (e.g., hemoglobinization of the cytoplasm) proceeds unimpaired, leading to the morphologic phenomenon of “nuclear cytoplasmic dissociation.” These red cell precursors have poor survival within the bone marrow and pathologic abnormalities suggestive of cellular dysplasia with nuclear fragments in the red cells are present. The cellular apoptosis causes intramedullary hemolysis or “ineffective erythropoiesis.” Maturation of white cell precursors is delayed, producing giant metamyelocytes and bands. Flow cytometry of the bone marrow demonstrates left shifts in erythroblasts and white cell precursors which again could lead

A

B

C

D

FIGURE 155-2.  Photomicrographs of peripheral smear, bone marrow aspirate, and biopsy are shown. A, Oval macrocytes and fragmented red blood cells are present in a smear from a patient with severe cobalamin deficiency and thalassemia trait, demonstrating an extreme range in size and shape of the red blood cells. B, Hypersegmented neutrophil. C, A group of megaloblastic red cell precursors all with an open lacy chromatin pattern is shown from a bone marrow aspirate. Red cell dysplasia and giant metamyelocytes are also present. D, The bone marrow biopsy shows hypercellularity with a predominance of large erythroblasts and many mitotic figures, all potentially confused with acute leukemia. The erythroid hyperplasia in the bone marrow of a patient with megaloblastic anemia is due to “ineffective erythropoiesis.” (Photomicrographs: 2B courtesy of John W. Ryder, MD, Department of Pathology, University of Colorado School of Medicine; 2A, C, and D courtesy of Zenggang Pan, MD, PhD, Department of Pathology, University of Colorado School of Medicine.)

1074

CHAPTER 155  Megaloblastic Anemias  

the unwary to diagnose acute leukemia. Cytogenetic studies may show chromosomal fragmentation and in some cases even clonal abnormalities which correct with vitamin replacement and resolution of megaloblastic anemia. There is indirect hyperbilirubinemia, low serum haptoglobin, and sometimes extreme values for lactate dehydrogenase because of the intramedullary hemolysis. The presence of hemolysis markers and fragmented cells present on peripheral blood smear with normal instead of low reticulocyte counts will suggest a hemolytic anemia to the unwary clinician, as the literature well documents. To complicate the situation further, there are reports of previously undiagnosed defects of cobalamin metabolism such as cobalamin-C mutation presenting with hemolytic-uremic syndrome. Other tissues that have rapid cellular turnover may become likewise megaloblastic, which is particularly common in the gastrointestinal tract where they sometimes cause malabsorption. Cells of the uterine cervix have been noted to be megaloblastic and appear dysplastic. Impairment of spermatogenesis can cause infertility. Atrophy and glossitis of the tongue mucosa may occur. Some individuals may have areas of skin hyperpigmentation.

Neurologic Abnormalities Due to Cobalamin Deficiency

Cobalamin is necessary for the development and initial myelination and maintenance of the spinal cord and brain. Demyelination can be demonstrated on imaging in both brain and spinal cord as depicted schematically in Figure 155-3. Cobalamin-deficient infants may have delay or failure in normal myelination

Cobalamin deficiency of the nervous system

which occurs during the first year of life. The demyelinating lesion, often termed “subacute combined degeneration,” begins with symptoms of symmetric paresthesia, usually in the lower extremities but sometimes involving the hands first. There is early impairment of proprioception ending with loss of vibration and position sense manifesting as ataxic gait. Other common signs and symptoms are shown in Figure 155-3. As the severity of the spinal lesions increases weakness develops, with spasticity and hyperreflexia, and eventually patients may develop a segmental cutaneous sensory level and even paraplegia. Altered mental status occasionally occurs particularly with irritability, delusions, paranoia, lability, even mania called “megaloblastic madness” in the past. Infants present with unique abnormalities, especially hypotonia, irritability, lethargy, failure to thrive, and occasionally movement disorders. The underlying biochemical lesions causing the neurologic disease are still unknown although changes in lipids, growth factors, cytokines, or methylation status of proteins have all been studied. There is a strong inverse relationship between the severity of the neurologic abnormalities and the hematologic abnormalities, which has been demonstrated in a large series of patients with pernicious anemia and recognized for at least 100 years in the literature. Therefore, anemia is only helpful in a patient with a compatible neurologic disorder in suggesting cobalamin deficiency but is not required. Figure 155-4 shows that patients with normal hemoglobin and a neurologic presentation had similar metabolic abnormalities as those with anemia. There is also an inverse relationship between the duration of neurologic abnormalities prior to diagnosis and the completeness of correction after adequate cobalamin replacement, so early diagnosis is imperative. Further discussion of the neurologic abnormalities of megaloblastic anemia and also alcoholism appears in Chapter 388.  

DIAGNOSIS

Cobalamin and Folate Biochemical Assays

Brain • Delirium • Cognitive defects • Coma (infants) • Optic atrophy • Anosmia

Psychiatric symptoms • “Megaloblastic madness” • Mania • Paranoia • Irritability • Delusions • Emotional liability • Depression Spongy degeneration

Measurement of the serum cobalamin or folate level has been a time-honored method of screening. Original tests have been replaced today with automated assays of cobalamin levels performed on instruments with platforms for multiple analytes. These methods rely on purified intrinsic factor (IF) as the cobalamin binder, often using chemiluminescent microparticle immunoassays for detection. It has been reported that denaturation of anti-IF antibodies present in samples from patients with pernicious anemia can be

300,000

Peripheral nerve • Cutaneous sensory loss • Hyporeflexia • Symmetric weakness • Paresthesias

Other • Autonomic nervous system -Postural hypotension -Incontinence -Impotence • Movement disorders in infants and children

Spinal cord • Paresthesias • Proprioceptic loss -Vibration -Position -Ataxic gait -Positive Romberg sign • Spasticity -Hyperreflexia • Lhermitte sign • Limb weakness • Segmental cutaneous sensory level • Paraplegia

FIGURE 155-3.  The myriad symptoms and signs of cobalamin deficiency of the nervous

system are shown. These range from paresthesias without objective signs to severe motor loss, even quadriplegia. The spongy degeneration shown in the cross-section depiction of the spinal cord may be seen on MR imaging as increased signal intensity on T2 weighted images often described as an inverted “V” sign. Also see Chapter 388. (Adapted from Stabler SP, Megaloblastic anemias: pernicious anemia and folate deficiency. In: Young NS, Green SL and High KA [Eds.] Clinical Hematology, Philadelphia, PA: Mosby Elsevier; 2006:242-251.)

Methylmalonic acid nmol/L

100,000 50,000

10,000 5,000

1,000 500

100

10

50

100

150

200

250

300

350

400

450

Homocysteine µmol/L FIGURE 155-4.  Serum methylmalonic acid plotted against serum homocysteine in patients with clinically ascertained and documented cobalamin deficiency. The patients with hematocrit greater than 38% are shown as open circles. Hatched lines represent methylmalonic acid 376 nmol/L and homocysteine 21.3 µmol/L. (Previously published in Stabler S. Vitamin B12. In Erdman, Jr. JW, McDonald IA, Zeisel SH, [Eds.] Present Knowledge in Nutrition. 10th ed. Ames, IA: Wiley-Blackwell; 2012.)

CHAPTER 155  Megaloblastic Anemias  

incomplete, and thus normal to even high values of cobalamin are reported falsely. In general, the values reported from these automated methods are higher than with the historical assays, which complicates comparison of reference ranges. Extremely low values of serum cobalamin (60 years, history of thrombosis, cardiovascular risk factors, and presence of the JAK2 V617F mutation each retained significance in a multivariate analysis of thrombosis risk. A prognostic model (IPSET-Thrombosis) could stratify patients into low-, intermediate-, and high-risk groups with significantly different rates of annual thrombosis risk: 1.03, 2.35, and 3.56%, respectively. Further analysis of IPSET-thrombosis revealed that among low-risk patients, the risk of thrombosis was significantly lower in patients with JAK2 negative/unmutated ET in the absence of cardiovascular risk factors (0.44%) compared to the risk of thrombosis in patients with JAK2 unmutated ET in the presence of cardiovascular risk factors (1.05%). In addition, the risk of thrombosis in the presence of JAK2 mutation without cardiovascular risk factors and in the presence of both JAK2 mutation and cardiovascular risk factors were 1.59% and 2.57%, respectively. These findings led to creation of the revised IPSET-thrombosis classification that stratifies patients into 4 risk groups: very low risk (age ≤60 years, no prior history of thrombosis and no JAK2 mutation); low risk (age ≤60 years, no prior history of thrombosis and JAK2 mutation); intermediate risk (age >60 years, no prior history of thrombosis and no JAK2 mutation), and high risk (prior history of thrombosis and/or age >60 years with JAK2 mutation). CALR mutation status was not found to have a significant impact on the IPSET-thrombosis prognostic score for predicting thrombosis risk. 3. Which of the following mutated genes does not lead to activation of the JAK-STAT signaling pathway? A . JAK2 V617F B. TET2 C. CALR D. MPL E. LNK (SH2B3)

Answer: B  Dysregulated JAK-STAT signaling is a principal feature among the classic Philadelphia chromosome-negative MPNs. Activation of the JAKSTAT signaling can occur via mutations in several genes in this axis. First, the V617F point mutation in the JAK2 gene (≈95% of PV patients and 50 to 60% of ET and PMF patients) relieves the pseudokinase domain inhibition of the protein’s kinase activity. Mutation of codon 515 in the gene encoding the thrombopoietin receptor (MPL) (5% of ET and 5 to 10% of PMF patients) stabilizes the receptor in an active dimeric configuration, leading to downstream JAK-STAT activation. Mutations in exon 9 of CALR (20 to 25% of ET and 25 to 30% of PMF patients) result in a +1 frameshift in the coding sequence. Although some 50 CALR mutations have been identified, the two most common mutations are 52 base pair deletions (type 1 mutations), which are more common in MF patients, and 5 base pair insertions (type 2 mutations), which are more common in ET. The mutant CALR C-terminus also acquires a greater positive charge that leads to loss of multiple calcium binding sites. Mutant CALR binds to the extracellular domain of MPL, resulting in activation of the JAK STAT signaling. This is a neomorphic function that allows the mutated ER chaperone protein to activate a cytokine receptor, leading to cellular transformation. Mutant CALR has also been found to be secreted from cells, circulates in the plasma, and can stimulate cells in an autocrine and paracrine fashion by binding to MPL. LNK (SH2B3) is negative regulator of JAK2. Mutations in LNK (found in 1 to 2% of ET/MF patients, and rarely in patients with idiopathic erythrocytosis/PV) interfere with its inability to inhibit downstream STAT activation. Somatic mutations in TET2 are a mixture of deletions, frameshifts, stop codons, or conserved amino acid substitutions. TET proteins are alpha-ketoglutarate-dependent enzymes that catalyze the conversion of 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC). DNA methylation is an important modulator of gene expression; therefore, the conversion of 5mC to 5hmC may alter chromatin structure and restrict access to DNA methyltransferases, which mediate repression of gene transcription. In various TET2 knockout murine models, loss of TET2 results in a myeloproliferative phenotype (e.g., splenomegaly, extramedullary hematopoiesis) with expansion of the hematopoietic stem cell compartment. Mutations in TET2 do not appear to activate the JAK-STAT signaling pathway. 4. Which of the following is true about the impact of therapy on the clinicopathologic features or natural history of PV, ET, or MF? A . In PV and ET, complete molecular remission of the JAK2 V617F mutation with PEG-interferon-alfa-2a results in delay of transformation to MF and increased overall survival. B. Ruxolitinib reduces the rate of transformation to acute myeloid leukemia in intermediate and high-risk MF. C. Ruxolitinib induces resolution of bone marrow fibrosis and molecular remission of JAK2 V617F in the majority of patients with myelofibrosis. D. Allogeneic hematopoietic stem cell transplantation (HSCT) is the only curative modality in MF. E. With 5-year follow-up, there is a 70% relative reduction in the risk of death in intermediate- and high-risk MF patients. Answer: D  PEG-interferon-alfa-2a can elicit complete molecular remissions of JAK2 V617F in 15 to 20% of patients with PV and ET; however, there are no long-term data to indicate how such remissions translate into modification of the natural history of these diseases, including delay in transformation to MF and overall survival. In patients with intermediate- and high-risk MF, ruxolitinib has not been found decrease the frequency of leukemic transformation (B), nor produce significant decreases in bone marrow fibrosis or JAK2 V617F allele burden in the majority of patients (C). With 5-year follow-up, an approximate 30% (not 70%) reduction in risk of death was observed in ruxolitinib-treated intermediate- and high-risk patients compared to patients treated with placebo or best available therapy, respectively, in the phase III COMFORT-I and COMFORT-II studies. The correct response is D—allogeneic HSCT is the only modality that can cure patients with MF. Please refer to the treatment section for a discussion regarding risk stratification and decision making regarding patient selection for HSCT, and outcome data based on patient clinical and laboratory variables.

CHAPTER 157  POLYCYTHEMIA, THROMBOCYTHEMIA, MYELOFIBROSIS  

5. A 17-year-old Indian woman presents with increasing headaches over the last 6 months. The patient has no smoking history. She is not taking any medications. Oxygen saturation is 97%. A CBC reveals a WBC count of 7.8 × 109/L, hemoglobin 21.3 g/dL, and platelet count of 178 × 109/L. Labs reveal iron deficiency and serum erythropoietin level is elevated at 52 IU/L (upper limit of normal 20 IU/L). What is the most likely cause of this patient’s erythrocytosis? A . JAK2 V617F mutation B. Exon 12 JAK2 mutation C. VHL mutation D. CALR exon 9 mutation E. MPL W515K mutation Answer: C  In patients with polycythemia vera, the JAK2 V617F mutation is identified in approximately 95% of patients, and exon 12 JAK2 mutations are found in ≈3% of patients. These mutations lead to autonomous red blood cell production which results in negative feedback on endogenous erythropoietin (EPO) production by the kidneys, thereby resulting in low serum EPO levels, in contrast to the elevated serum EPO level identified in this patient. The somatic CALR and MPL mutations are not found in patients with erythrocytosis, but instead are identified in patients with ET and MF with nonmutated JAK2.

1094.e3

The correct answer is C, Von Hippel-Lindau (VHL) mutation. The Arg200Trp mutation (C598T), leading to a R200W amino acid substitution in the VHL protein, causes the Chuvash form of polycythemia, which is inherited in an autosomal recessive manner. Alternatively, compound heterozygosity for different germline mutations within the VHL gene can lead to this form of congenital erythrocytosis. Besides the original cluster described in the Chuvash region, diverse ethnic groups have now been identified with VHL. The von Hippel-Lindau protein plays an important role in hypoxia sensing. It binds to the hydroxylated hypoxia-inducible factor 1α (HIF-1α) and serves as a recognition component of an E3-ubiquitin ligase complex. In hypoxia or secondary to a mutated VHL gene, the non-degraded HIF-1α forms a heterodimer with HIF-β and leads to increased transcription of hypoxia-inducible genes, including EPO.

1094

CHAPTER 158  Leukocytosis and Leukopenia  

158  LEUKOCYTOSIS AND LEUKOPENIA NANCY BERLINER The normal peripheral white blood cell count (WBC) ranges between 4500/µL and 10,000/µL, with a mean of 7500/µL, and is composed of neutrophils, lymphocytes, monocytes, basophils, and eosinophils. Because neutrophils usually represent about 60% of the peripheral WBC, derangement in the WBC usually reflects elevation or reduction in the absolute neutrophil count. Leukocytosis (an elevated WBC) and leukopenia (a depressed WBC) may be secondary to an underlying disease or exposure, or they may be manifestations of a primary hematologic disorder. This chapter outlines both the primary and secondary causes of leukocytosis and leukopenia, focusing particularly on neutrophilia and neutropenia.

  NORMAL NEUTROPHIL DYNAMICS

Neutrophils arise from multipotent progenitor cells in the bone marrow that also give rise to erythrocytes, megakaryocytes, eosinophils, basophils, and monocytes. Neutrophil precursors in the marrow mature over 6 to 10 days to form a storage pool of mature neutrophils (Fig. 158-1). Acetylation of C/ EBPε, a member of the CCAAT/enhancer binding protein family of transcription factors, is required for secondary and tertiary granule formation and for terminal neutrophil differentiation.1 Together, the marrow populations make up about 95% of the body’s total granulocyte mass (20% neutrophil precursors, 75% mature bands and neutrophils). The circulating neutrophil pool thus represents only the remaining approximately 5% of the body’s total neutrophils, just over half of which at any given time are adherent to the vascular endothelium and pooled in the spleen, a phenomenon termed margination. Marginated neutrophils are poised for immediate release into the circulation during stress. The lifespan of neutrophils in the peripheral blood was thought to be very short, only 6 to 12 hours; however, newer in vivo studies suggest that they circulate for up to 3 to 4 days. They subsequently migrate into tissues,

1094.e4

ABSTRACT

CHAPTER 158  Leukocytosis and Leukopenia  

Disruption of the normal peripheral white blood cell count may reflect primary hematologic disease or may be secondary to a host of other diseases. As the primary component of the white blood cell compartment, leukocytosis and leukopenia usually reflect disorders of neutrophils. The focus of this chapter is on the differential diagnosis, evaluation, and management of leukocytosis and leukopenia, with a strong focus on disorders of neutrophils. Leukocytosis is most commonly reactive and secondary to nonhematologic disorders, reflecting the normal response of a healthy bone marrow to infectious or inflammatory stress. Evaluation should be focused upon identifying the underlying disorder and bone marrow examination is rarely necessary. Primary myeloproliferative neoplasms leading to leukocytosis are much less common, but can largely be identified by molecular testing for pathognomonic mutations on peripheral blood. By contrast, neutropenia is often a primary disorder, either congenital or acquired, and requires more intensive hematologic evaluation. Outside of chronic neutropenic disorders, leukopenia may also occur because of immune-mediated neutrophil destruction, either as a reaction to a drug or toxin, or in the setting of autoimmune disease. Most patients with unexplained neutropenia require bone marrow examination, although repeated examination is not indicated in patients with chronic idiopathic neutropenia. Age of onset and chronicity can help guide the evaluation of patients with leukopenia.

KEYWORDS

leukocytosis leukopenia neutrophilia neutropenia immune neutropenia congenital neutropenia

1095

CHAPTER 158  Leukocytosis and Leukopenia  

Myeloblast

Promyelocyte

Nucleolus: should be prominent in myeloblast and much less in promyelocyte, and absent thereafter

Myelocyte

Metamyelocyte

Primary Granules: should be present mostly in promyelocytes and myelocytes, and fewer in the later stages

Band

Segmented neutrophil

Secondary Granules: should be present at myelocyte stage and beyond. Lighter pink than primary granules

FIGURE 158-1.  Myeloid maturation in the bone marrow. Nucleoli are prominent in myeloblasts, much less frequent in promyelocytes, and absent in more mature forms. Primary granules are present in the cytoplasm of promyelocytes and myelocytes, and secondary granules predominate beyond the myelocyte stage.

where they can survive for 1 to 4 days. Changes in neutrophil number reflect these dynamics. Neutrophilia can occur as the result of increased marrow production, increased release of neutrophils from the storage pool, or mobilization of neutrophils from the marginated pool. Neutropenia may be due to decreased marrow production, increased margination with or without sequestration by the spleen, or increased destruction of peripheral cells.

  NEUTROPHILIA

Neutrophilia is usually reactive or secondary to an underlying inflammatory process.2 This includes neutrophilia due to infection, chronic inflammation, stress, drugs, nonhematologic malignancy, marrow stimulation (as in hemolysis or idiopathic thrombocytopenic purpura), or splenectomy. Primary causes of neutrophilia may be congenital, including hereditary neutrophilia, Down syndrome, and leukocyte adhesion deficiency or acquired as in the case of chronic myelogenous leukemia (CML) and other myeloproliferative neoplasms (Chapter 157) (Table 158-1).  

EPIDEMIOLOGY AND PATHOBIOLOGY

Secondary Causes of Neutrophilia Infection

Many acute bacterial infections can present with leukocytosis with a “left shift,” referring to the circulation of more immature myeloid cells. This left shift most commonly is restricted to release of an increased number of band forms; however, in severe stress, one may also see circulating metamyelocytes and even earlier cells (“leukemoid reaction”) (see Fig. 158-1) in the peripheral blood. Leukocytosis occurs within minutes to hours of infection owing to release of neutrophils from both the marrow and marginated pools. Examination of these neutrophils on peripheral smear may reveal evidence of toxic granulation (see Fig. 148-18), Döhle bodies (see Fig. 148-19), and cytoplasmic vacuoles. Certain infections (e.g., Clostridium difficile or tuberculosis in particular) are known to cause elevations in the WBC to greater than 30,000/µL in about one fourth of infected patients and may result in a leukemoid reaction, defined as a WBC of greater than 50,000/µL with a pronounced left shift (Fig. 158-2).

Chronic Inflammation

Leukocytosis due to chronic inflammation results from increased leukocyte (specifically neutrophil and monocyte) production as opposed to altered neutrophil distribution. Mature neutrophil pools become depleted with ongoing inflammation, and the marrow myeloid compartment expands to increase neutrophil production. Myriad cytokines, including tumor necrosis factor-α (TNF-α), granulocyte colony-stimulating factor (G-CSF), granulocytemacrophage colony-stimulating factor (GM-CSF), macrophage inflammatory protein-1 (MIP-1), interleukin-1 (IL-1), IL-6, and IL-8, have been implicated in this marrow stimulation. Chronic inflammatory conditions that are commonly associated with leukocytosis and neutrophilia include juvenile rheumatoid arthritis, rheumatoid arthritis, Still disease, Crohn disease, ulcerative colitis, granulomatous infection, and chronic hepatitis. The WBC and neutrophil elevation in these cases is typically more modest than that seen in acute infection or inflammation.

TABLE 158-1 DIFFERENTIAL DIAGNOSIS OF NEUTROPHILIA Primary hematologic etiologies Congenital neutrophilia Hereditary neutrophilia Chronic idiopathic neutrophilia Down syndrome Leukocyte adhesion deficiency (LAD) LAD-1 LAD-2 Acquired hematologic neoplasms Myeloproliferative neoplasm Chronic myelogenous leukemia Polycythemia vera Secondary to other disease entities Infection Acute via release from marginated and storage pools Chronic via increased myelopoiesis (e.g., tuberculosis, fungal infection, chronic abscess, other chronic infections) Chronic inflammation Rheumatic disease: juvenile rheumatoid arthritis, rheumatoid arthritis, Still disease, and others Inflammatory bowel disease Granulomatous disease Chronic hepatitis Cigarette smoking Stress Drug induced Corticosteroids β-Agonists Lithium Recombinant cytokine administration Nonhematologic malignancy Cytokine-secreting tumors (lung, tongue, kidney, urothelial tumors) Marrow metastasis (myelophthisis) Marrow stimulation Hemolytic anemia, immune thrombocytopenia Recovery from marrow suppression Recombinant cytokine administration Post-splenectomy

Cigarette Smoking

Cigarette smoking can cause a neutrophilia in about 25 to 50% of chronic smokers that can persist for up to 5 years after smoking cessation. The mechanism of neutrophila is unknown, although there is recent evidence that cigarette smoke slows neutrophil apoptosis.

Stress

Within minutes of exercise, surgery, or stress, one can see an increased number of circulating neutrophils. This is presumed to be due to the effects of catecholamines on marginated neutrophils (causing demargination). Some cases of stress-induced neutrophilia can be prevented by pretreatment with β-adrenergic antagonists (e.g., propranolol), supporting the role of catecholamines in the

1096

CHAPTER 158  Leukocytosis and Leukopenia  

FIGURE 158-2.  Peripheral blood from a patient with leukemoid reaction. From this smear, it would be impossible to distinguish this from chronic phase chronic myelogenous leukemia (Chapter 175). Distinction would depend on determination of presence or absence of BCR-ABL fusion.

FIGURE 158-3.  Myelophthisic changes in erythrocyte morphology. Note prominent teardrop forms. (From Rose M, Berliner N. Disorders of red blood cells. In: Andreoli TE, Benjamin IJ, Griggs RC, et al, eds. Andreoli and Carpenter’s Cecil Essentials of Medicine, 8th ed. Philadelphia: Saunders; 2010:522, Fig. 49-2.)

Pelger-Huët Anomaly

Drug Induced

The most well-known and widely used drugs associated with leukocytosis are corticosteroids. Other drugs that elevate the neutrophil count include β-agonists and lithium. Lithium causes neutrophilia by increasing the production of endogenous colony-stimulating factors (CSFs). G-CSF or GM-CSF treatment (Chapter 147) likewise cause neutrophilia, and although this is the desired effect, the neutrophilia can be quite pronounced.

Patients with the Pelger-Huët anomaly are often misdiagnosed as having a left-shifted WBC because many of their mature neutrophils are misinterpreted as band forms.3 Although these patients do not have leukocytosis, the PelgerHuët anomaly often raises suspicion for an acute infection because of this apparent left shift. Pelger-Huët anomaly is due to a mutation in the lamin B receptor gene and manifests with mature neutrophils and condensed, clumped chromatin within a bilobed nucleus (see Fig. 148-20). A number of drugs can reversibly induce pseudo-Pelger-Huët anomaly, including colchicine, sulfonamides, ibuprofen, taxoids, and valproate. Pseudo-Pelger-Huët anomaly is also seen in some patients with myelodysplasia (Chapter 172). Vitamin B12 or folate deficiency can cause increased nuclear lobation of neutrophils in patients with Pelger-Huët anomaly, perhaps leading to a missed diagnosis. With correction of the vitamin deficiency, however, the aberrant neutrophil nuclear morphology returns.

Nonhematologic Malignancy

Down Syndrome

demargination process. Exercise-induced neutrophilia, however, is not blocked by propranolol, suggesting that it may instead be due to flow and mechanical perturbation of neutrophils in the lungs. An elevated WBC has also been noted in the setting of acute myocardial infarction.

Leukocytosis can be seen in a number of nonhematologic malignancies. Some tumors (lung, tongue, kidney, bladder) may secrete G-CSF as an ectopic hematopoietic growth factor. In cancer patients with hyperleukocytosis, genomic testing can discriminate myeloproliferative neoplasms from paraneoplastic leukemoid reaction. Other tumors (lung, stomach, breast), when metastatic to the bone marrow, can cause a leukoerythroblastic reaction, characterized by left-shifted leukocytosis, thrombocytosis, and red cell abnormalities including nucleated and teardrop-shaped red blood cells (Fig. 158-3). The presence of nonhematopoietic entities invading the bone marrow (metastatic cancer, fibrosis, granulomatous disease) is termed myelophthisis.

Marrow Stimulation

Peripheral destruction of red cells and platelets, as seen with hemolytic anemia and idiopathic thrombocytopenic purpura, can result in stimulation of the bone marrow and a “spillover” leukocytosis. Recovery of cell counts following marrow suppression, as in the case of chemotherapy, can result in a rebound leukocytosis that may last several weeks.

Primary Causes of Neutrophilia Hereditary Neutrophilia

Hereditary neutrophilia is an autosomal dominant genetic disease that is characterized by an elevated WBC in the 20,000 to 100,000/µL range with splenomegaly and widened diploe of the skull. The neutrophils function normally, and patients have no increased risk for bacterial infection or other sequelae. Hereditary neutrophilia caused by an autosomal-dominant GCSF3 gene mutation has been reported, causing constitutive activation of the G-CSF receptor.

Chronic Idiopathic Neutrophilia

Chronic idiopathic neutrophilia is a condition marked by leukocytosis in the 11,000 to 40,000/µL range with a normal bone marrow biopsy. In one series with a 20-year follow-up, patients with this condition had no medical sequelae from the elevated WBC.

Up to 10% of patients with Down syndrome develop a transient myeloproliferative disorder manifesting as peripheral blood leukocytosis with blasts in association with an accumulation of megakaryoblasts in the blood, liver, and marrow. Similar reactions have also been reported in phenotypically normal individuals with trisomy 21 mosaicism. Transient myeloproliferative disorder usually resolves spontaneously but can progress to acute megakaryoblastic leukemia in 23 to 30% of affected patients. This disorder is attributable to acquisition of mutations in the GATA1 gene, which encodes a key transcription factor for hematopoietic regulation, leading to loss of normal GATA-1 expression and expression of a truncated GATA-1 protein. Evidence supports that these mutations are acquired during fetal life, and patients present in early infancy with transient myeloproliferative disorder. The pathogenesis of progression to acute megakaryoblastic leukemia presumably involves additional genetic and epigenetic events and is the focus of intense study.

Leukocyte Adhesion Deficiency

Patients with leukocyte adhesion deficiency also have persistent leukocytosis, defects in stimulus-dependent activation of neutrophils, recurrent infections, and delayed separation of the umbilical cord. Leukocyte adhesion deficiency is an abnormality of leukocyte adhesion reflecting the loss of surface adhesion molecules.4 Leukocyte adhesion deficiency-1 is due to absence or marked reduction in the common β chain of β2 integrins, resulting in loss of expression of leukocyte function–associated antigen 1 (LFA-1), the C3bi receptor, and GP150;95. This results in a failure to ingest and kill microbes opsonized by C3bi. In leukocyte adhesion deficiency-2, neutrophils lack sialyl Lewis X, the ligand for L-selectin expressed on endothelial cells. Neutrophils appear morphologically normal but are defective in chemotaxis, adherence, and phagocytosis.

Familial Cold Urticaria

Familial cold urticaria is marked by episodic fevers, leukocytosis, urticaria, rash, conjunctivitis, and muscle and skin tenderness with cold exposure. The

CHAPTER 158  Leukocytosis and Leukopenia  

1097

Neutrophilia

Repeat counts

Normalized

No further evaluation

Persistent Examine smear: Leukoerythroblastic? Yes

No

Bone marrow examination Morphology, cytogenetics, culture

Fever, signs of infection?

Yes

Evaluate and treat

No Serologies for AID Tumor Granulomatous dx

JAK2 mutation

Ph1 or BCR-ABL

Peripheral blood for BCR-ABL and JAK2 Non-CML MPN (P. vera, myelofibrosis)

CML

Potential drugs (lithium, steroids)

ddx: Autoimmune dx Myeloproliferative dx Drug induced Smoking related Idiopathic

Smoker? FIGURE 158-4.  Diagnostic approach to neutrophilia. AID = autoimmune disease; CML = chronic myelogenous leukemia; ddx = differential diagnosis; dx = disease; MPN = myeloproliferative neoplasm; Ph1 = Philadelphia chromosome; PV = P. vera, ET = essential thrombocytopenia.

rash is composed of infiltrating neutrophils. The syndrome appears to be related to decreased levels of C1-esterase inhibitor and is associated with mutations in the CIAS1 gene on chromosome 1q.

Chronic Myelogenous Leukemia and Other Myeloproliferative Neoplasms

Chronic myelogenous leukemia (CML), chronic neutrophilic leukemia, and the other myeloproliferative neoplasms (namely polycythemia vera and essential thrombocythemia) are discussed in detail in Chapters 157 and 175. They are the principal acquired primary hematologic disorders associated with neutrophilia. CML on presentation often has to be distinguished from a leukemoid reaction. In contrast to a leukemoid reaction, CML is characterized by the presence of abnormalities of other blood cell lines (“panmyelosis”) and by the presence of specific abnormalities. Therefore, the peripheral blood smear in CML (but not leukemoid reaction) demonstrates increased numbers in all cells of the neutrophilic series, classically with a greater proportion of myelocytes to metamyelocytes, and may display concomitant basophilia, eosinophilia, anemia, and thrombocytosis. Chronic neutrophilic leukemia, a rare myeloproliferative neoplasm, is characterized by hepatosplenomegaly and leukocytosis of at least 25,000/µL, with more than 80% of leukocytes being segmented neutrophil/band forms and less than 10% being immature granulocytes.5 At least 50% of patients with chronic neutrophilic leukemia harbor mutations in the receptor for CSF-3 (CSF3R; GCSFR). Polycythemia vera and essential thrombocythemia, on the other hand, are notable for also having a marked increase in red cell mass and a marked thrombocytosis, respectively, which is often accompanied by leukocytosis.

Post-splenectomy

Patients develop leukocytosis following splenectomy, often long-standing, reflecting the loss of a major site of neutrophil margination. This is of no clinical importance, except insofar as it leads to unnecessary evaluation for other pathology.  

CLINICAL MANIFESTATIONS AND DIAGNOSIS

As outlined previously, acquired leukocytosis is most commonly the result of acute or chronic infection or inflammation. When it occurs in the absence of fever, aberrations in acute phase reactants, effusions and edema, or other signs and symptoms of inflammation, it still may be secondary to drugs or an underlying nonhematologic malignancy. As such, it should be seen as the

sign of a healthy hematopoietic system responding to an outside stress. Bone marrow evaluation is therefore rarely indicated. However, persistence of leukocytosis in the absence of signs and symptoms of inflammation or infection, nonhematologic malignancy, and offending drugs should prompt an evaluation for a primary myeloproliferative disease or clonal hematologic neoplasm, particularly when there is evidence of a leukoerythroblastic reaction. CML and other myeloproliferative neoplasms can be ruled out by molecular diagnosis on the peripheral blood, as described in Chapter 175. In this setting, bone marrow examination is indicated to evaluate for marrow infiltration by infection, tumor, or fibrosis and should include cultures for tuberculosis and fungal infection as well as cytogenetics and flow cytometry (Fig. 158-4).

Leukocytosis Due to Expansion of Other Cell Lines

Monocytosis and lymphocytosis can also elevate the WBC. Monocytosis, defined by an absolute monocyte count of greater than 500/µL, usually occurs in chronic inflammation resulting from infections like tuberculosis, syphilis, or subacute bacterial endocarditis, autoimmune or granulomatous disease, and sarcoidosis.6 It can also be seen in malignancies, such as preleukemic states, nonlymphocytic leukemia including acute myelomonocytic and monocytic leukemia, histiocytosis, Hodgkin disease, non-Hodgkin lymphoma, and various carcinomas. Finally, it can be seen in the setting of chronic neutropenia, after splenectomy, and in the setting of recovery from marrow suppression (Table 158-2). Lymphocytosis is defined by an absolute lymphocyte count of more than 5000/µL. The most common causes of lymphocytosis are viral infections such as Epstein-Barr virus (Chapter 346) and the hepatitis viruses (Chapter 139 and 140). Although most bacterial infections cause neutrophilia, pertussis and cat-scratch disease due to Bartonella henselae can cause lymphocytosis. Other infections that may cause a secondary lymphocytosis include toxoplasmosis and babesiosis. Hypersensitivity reactions due to drugs or serum sickness may also be associated with lymphocytosis. Primary disorders causing lymphocytosis include chronic lymphocytic leukemia and monoclonal B-cell lymphocytosis (Table 158-3; see also Chapter 174). Low-count monoclonal B lymphocytosis can be detected in about 5% of adults over age 40 years and rarely progresses to chronic lymphocytic leukemia. By comparison, high-count monoclonal B lymphocytosis progresses to chronic lymphocytic leukemia at a rate of 1 to 2% per year.7 Eosinophilia (Chapter 161) is defined by an absolute eosinophil count of more than 400/µL.8 Eosinophils proliferate under the influence of IL-5 and

1098

CHAPTER 158  Leukocytosis and Leukopenia  

TABLE 158-2 DIFFERENTIAL DIAGNOSIS OF MONOCYTOSIS

TABLE 158-4 DIFFERENTIAL DIAGNOSIS OF NEUTROPENIA

Infection Granulomatous disease (tuberculosis, fungal disease) Endocarditis Syphilis Autoimmune diseases Lupus, rheumatoid arthritis Giant cell arteritis Vasculitis Inflammatory bowel disease Sarcoid Malignancy Primary hematologic malignancy Chronic myelomonocytic leukemia Acute myelomonocytic leukemia Lymphoma Solid tumors Neutropenia Associated with chronic neutropenia Recovery form marrow suppression Post-splenectomy

Congenital neutropenia Ethnic and benign familial (constitutional) neutropenia Severe congenital neutropenia Autosomal dominant (ELANE mutation) Autosomal recessive (Kostmann syndrome; HAX2 mutation) X-linked (WASP mutation) Other rare defects (G-CSFR mutation, unknown) Cyclic neutropenia Shwachman-Diamond syndrome Fanconi anemia Dyskeratosis congenita Glycogen storage disease type Ib Myelokathexis Chédiak-Higashi syndrome Griscelli syndrome type II Hermansky-Pudlak syndrome II Barth syndrome Acquired neutropenia Infection Postinfection Drug induced Immune neutropenia Primary immune neutropenia Secondary to autoimmune disease Rheumatoid arthritis Felty syndrome Large granular lymphocyte disease Systemic lupus erythematosus Granulomatosis with polyangiitis (formerly Wegener granulomatosis) Hyperthyroidism Pure white cell aplasia associated with thymoma Large granular lymphocyte disease (independent of rheumatoid arthritis) Primary bone marrow failure Aplastic anemia Myelodysplastic syndrome Acute leukemia Margination and hypersplenism Vitamin and mineral deficiencies (including B12, folate, copper) Chronic idiopathic neutropenia in adults (CINA)

TABLE 158-3 DIFFERENTIAL DIAGNOSIS OF LYMPHOCYTOSIS Infection Viral infection Epstein-Barr virus Cytomegalovirus Hepatitis Bacterial infection Pertussis Bartonella Tuberculosis Syphilis Rickettsia Babesia Hypersensitivity reactions Serum sickness Drug hypersensitivity Primary hematologic disease Chronic lymphocytic leukemia Monoclonal B-cell lymphocytosis Non-Hodgkin lymphoma

play a role in phagocytosis and modulating toxicity due to mast cell degranulation in hypersensitivity reactions. Eosinophilia is therefore most often seen in the setting of drug reactions, allergy, atopy, and asthma. A variety of infections, particularly parasitic infections and, to a lesser degree, fungal infections, can cause eosinophilia. Eosinophilia can also be the result of autoimmune and inflammatory conditions, as in eosinophilic granulomatosis with polyangiitis (formerly Churg-Strauss vasculitis). Atheroembolic disease and adrenal insufficiency may also cause eosinophilia. A number of cancers have been associated with polytypic expansion of eosinophils, including lymphomas and solid tumors. There are also a number of clonal disorders of eosinophils that occur in the setting of some leukemias. Finally, there is a heterogeneous group of disorders termed hypereosinophilic syndromes (Chapter 161). A FIP1L1PDGFRA fusion gene has confirmed that some of these are primary clonal disorders of eosinophils; the clonality of other hypereosinophilic syndromes can be difficult to establish (see Table 161-1).

  NEUTROPENIA

The risk for infection with neutropenia is highly dependent on the size of the neutrophil storage pool. Although neutropenia is defined by an absolute neutrophil count of less than 1500/µL, patients with neutrophil counts below this number have different risks and rates of infection depending on the cause of neutropenia. Patients who are neutropenic from chemotherapy, marrow failure, or marrow exhaustion experience infection at much higher rates than those with chronic neutropenic syndromes and immune-mediated neutropenia. Neutropenia may be congenital or acquired. The following sections first discuss congenital neutropenic disorders, the study of which has provided critical

ELANE = neutrophil elastase; G-CSFR = granulocyte colony-stimulating factor receptor; WASP = Wiskott-Aldrich syndrome protein.

insights into normal myelopoiesis, and then the secondary causes of neutropenia (Table 158-4).  

ETIOLOGY

Primary Causes of Neutropenia Ethnic and Benign Familial (Constitutional) Neutropenia

The normal range of the neutrophil count is genetically determined and is variable. Several racial and ethnic groups have a relatively large proportion of individuals who are neutropenic by comparison to the published normal range, usually based on young, largely white individuals. This is termed constitutional neutropenia and is seen among a variety of ethnic groups, including African Americans, Yemenite Jews, Falasha Jews, and African Bedouins.9 Singlenucleotide polymorphisms in the gene for the Duffy antigen receptor for chemokine are postulated to explain racial and ethnic differences in neutrophil counts. There is also an autosomal dominantly inherited condition called benign familial neutropenia that is characterized by neutrophil counts in the 800 to 1400/µL range. Neither ethnic nor benign familial neutropenia are associated with any increased risk for infection.

Severe Congenital Neutropenia

First described by Rolf Kostmann in 1956, severe congenital neutropenia presents in the neonatal period with recurrent bacterial infections and neutrophil counts of less than 500/µL.10 Infections usually occur in the first months of life and often include omphalitis and perirectal abscesses. There is often an increase in monocytes and eosinophils. Bone marrow biopsy reveals a “maturation arrest” with an absence of mature neutrophil elements. Severe congenital neutropenia can follow autosomal dominant and recessive and X-linked patterns of inheritance and is associated with mutations in a variety of genes, as summarized in Table 158-5.

CHAPTER 158  Leukocytosis and Leukopenia  

1099

Shwachman-Diamond syndrome

X-Linked

SBDS

Fanconi anemia

Autosomal recessive

FANCA-FANCO

Dyskeratosis congenita

Variable

Telomerase and related genes

Glycogen storage disease type 1b

Autosomal recessive

G-6-PT

and their contribution to malignant transformation in patients with severe congenital neutropenia (see below under Treatment). Two classes of G-CSF receptor mutations have been associated with severe congenital neutropenia and either hyper-responsiveness or hyporesponsiveness of the receptor. Initial studies aimed at demonstrating that severe congenital neutropenia was caused by mutation of the G-CSF or G-CSF receptor gene did not implicate such mutations in its pathogenesis. However, a small number of patients have been shown to have mutations in the G-CSF receptor gene that block ligand binding and produce a G-CSF-resistant form of severe congenital neutropenia. More important, however, the studies identified an acquired missense mutation that introduces a stop codon that leads to the deletion of the distal intracellular domain of the receptor known to be responsible for differentiation signaling. This has been shown to cause hypersensitivity to G-CSF, suggesting that the mutation may play an important role in the development of myelodysplasia/AML. However, whether this mutation in fact influences the development of myelodysplasia/AML and whether G-CSF increases the risk for developing the mutation remain subjects of significant controversy. Even in the presence of the G-CSF mutation, development of AML is associated with the acquisition of other myeloid leukemia-associated genes.

Myelokathexis

Autosomal dominant

CXCR4

Cyclic Neutropenia

Chédiak-Higashi syndrome

Autosomal recessive

LYST

Griscelli syndrome type 2

Autosomal recessive

RAB27A

Hermansky-Pudlak syndrome type 2

Autosomal recessive

AP3B1

TABLE 158-5 CONGENITAL NEUTROPENIA SYNDROMES SYNDROME

INHERITANCE PATTERN

GENE

SCN

Autosomal dominant Autosomal recessive X-linked X-linked Autosomal recessive Autosomal dominant Autosomal dominant

ELANE (~60%) HAX1 (~5%) WASP (~5%) TAZ (rare) G6PC3 (~2%) Gfi1 (rare) G-CSFR (rare)

CYCLIC NEUTROPENIA

Autosomal dominant

ELANE

OTHER CONGENITAL SYNDROMES

ELANE = neutrophil elastase; G6PC3 = glucose-6-phosphatase catalytic subunit 3; G-6-PT = glucose-6-phosphatase translocase; G-CSFR = granulocyte colony-stimulating factor receptor; LYST = lysosomal trafficking regulatory gene; SBDS = Shwachman-Bodian-Diamond syndrome; SCN = severe congenital neutropenia; WASP = Wiskott-Aldrich syndrome protein.

Neutrophil elastase (ELANE) is a serine protease that is synthesized at high levels in the promyelocyte stage of neutrophil maturation and is packaged in primary granules. Though originally hypothesized to cause defective cellular trafficking and cytoplasmic accumulation leading to neutrophil apoptosis, newer evidence supports that ELANE mutations lead to accumulation of mutant elastase in the endoplasmic reticulum, with apoptosis resulting from activation of the unfolded protein response.11 Severe congenital neutropenia due to ELANE mutations is inherited in an autosomal dominant fashion. Hax-1 is a mitochondrial protein with weak homology to bcl-2, and its absence results in mitochondrial-dependent apoptosis. HAX-1 mutations cause the autosomal recessive severe congenital neutropenia originally described by Kostmann. Although all severe congenital neutropenia cases were originally referred to as Kostmann syndrome, that term is now reserved for this subgroup of autosomal recessive cases. Wiskott-Aldrich syndrome protein regulates actin polymerization in hematopoietic cells, and deficiency of this protein results in the Wiskott-Aldrich syndrome, characterized by small platelets in low number, sinopulmonary infections, and eczema. Another phenotype of mutated Wiskott-Aldrich syndrome protein, however, is X-linked thrombocytopenia and neutropenia. Mutations in the glucose-6-phosphatase catalytic subunit 3 (G6PC3) are a more recently discovered cause of severe congenital neutropenia; homozygous loss of this metabolic enzyme also appears to lead to activation of the unfolded protein response and increased apoptosis of neutrophil precursors. Severe congenital neutropenia was formerly a disease of infancy and early childhood because few, if any, patients survived to adulthood. However, the observation that G-CSF can raise neutrophil counts and prevent infection in most patients has allowed these children to survive. Some patients require very high doses of G-CSF, but responses are seen in 80 to 90% of individuals. Increased survival led to the realization that severe congenital neutropenia predisposes to the development of myelodysplasia and acute leukemia (myelodysplasia/AML), with the development of myelodysplasia/AML at a rate of approximately 2% per year, and a cumulative risk of about 20 to 30% over 10 years. Patients refractory to, or requiring very high doses of, G-CSF have a higher risk for leukemic transformation. As discussed below, development of myelodysplasia/AML is often associated with the acquisition of a G-CSF receptor gene mutation. Considerable controversy exists concerning the role of lifelong G-CSF administration in the acquisition of other mutations

Cyclic neutropenia is defined by periods of neutropenia (≤200/µL) lasting 3 to 5 days and occurring at approximately 21-day intervals. Neutropenic periods may be marked by recurrent fevers, mouth sores, and infections of the skin, upper respiratory tract, and ears. The disorder can be dominantly inherited or sporadic. Cyclic neutropenia is also associated with ELANE mutations in virtually all cases. The diagnosis formerly required demonstration of transient neutropenia through frequent blood counts over a course of 6 weeks but now can be established by sequencing of the elastase gene. It is successfully treated with G-CSF and is not associated with an increased risk for leukemic transformation. Acquired cyclic neutropenia has been rarely associated with systemic diseases such as large granular lymphocytosis or T-cell lymphoma.

Other Congenital Syndromes with Associated Neutropenia

Several congenital disorders are associated with neutropenia.12 These include Shwachman-Diamond syndrome, Fanconi anemia, dyskeratosis congenita, glycogen storage disease Ib, myelokathexis, Chédiak-Higashi syndrome, Griscelli syndrome II, and Hermansky-Pudlak syndrome II. Shwachman-Diamond syndrome usually begins as an isolated neutropenia but progresses to marrow failure and is also associated with pancreatic dysfunction and skeletal abnormalities. The responsible gene, the ShwachmanBodian-Diamond syndrome gene (SBDS), is involved in the regulation of ribosomal RNA. These patients carry an increased risk for leukemic transformation. Fanconi anemia (Chapter 156) is due to mutations in genes involved in DNA repair, and as such, marrow failure develops later in these patients (median age, 7 years). Patients with Fanconi anemia often have, in addition to marrow failure, short stature with upper limb anomalies and hyperpigmented caféau-lait spots, although about one third have no physical abnormalities. Screening is done by chromosomal fragility testing following exposure to diepoxybutane or mitomycin C and by direct assessment for known Fanconi gene mutations. Stem cell transplantation is curative but carries a high risk for morbidity and mortality owing to the toxicity of preparative regimens. Dyskeratosis congenita (Chapter 156) is a syndrome of nail dystrophy, leukoplakia, and skin pigmentation abnormalities with associated neutropenia or aplastic anemia. It is inherited in an autosomal dominant or recessive or X-linked fashion and is associated with mutations in genes involved in telomere maintenance. It typically does not present until the second decade of life. Dyskeratosis congenita is one of several diseases associated with telomere abnormalities, with wide-ranging manifestations including pulmonary fibrosis and hepatic cirrhosis as well as bone marrow failure. Glycogen storage disease Ib (Chapter 196) is inherited as an autosomal recessive disorder characterized by intermittent neutropenia due to defects in the neutrophil respiratory burst with subsequent apoptosis of circulating neutrophils. Hepatomegaly and metabolic crises are also the hallmarks of this disease and are due to mutations in the gene for the glucose-6-phosphatase translocase enzyme. The neutropenia of myelokathexis is due to retention of mature neutrophils in the bone marrow. During infection, however, patients with myelokathexis typically have a sudden rise in their neutrophils, making their clinical course relatively more benign. There is an association between this condition and

1100

CHAPTER 158  Leukocytosis and Leukopenia  

hypogammaglobulinemia and warts, the WHIM syndrome (warts, hypogammaglobulinemia, immunodeficiency, and myelokathexis). It is caused by heterozygous mutations in the gene encoding chemokine receptor CXCR4. Chédiak-Higashi syndrome, Griscelli syndrome II, and Hermansky-Pudlak syndrome II are all syndromes of albinism and neutropenia due to defects in vesicular trafficking. Chédiak-Higashi syndrome is due to mutations in a lysosomal trafficking regulatory gene (LYST) and is characterized by oculocutaneous albinism, bleeding, progressive neurologic disease, and increased susceptibility to hemophagocytic syndrome. Griscelli syndrome II, with nearly identical clinical manifestations as Chédiak-Higashi syndrome, is caused by mutations in the gene encoding the small guanosine triphosphatase RAB27A, which is involved in the release of myeloperoxidase from the primary granules of neutrophils. Hermansky-Pudlak syndrome II is due to mutations in the AP3B1 gene, which encodes a part of a protein transport complex involved in vesicular trafficking and appears to be involved in the trafficking of neutrophil elastase. It is also marked by albinism, platelet abnormalities, and pulmonary fibrosis. Barth syndrome is an X-linked autosomal recessive disorder characterized by neutropenia, cardiomyopathy, and growth retardation, with a high early childhood mortality from the associated heart disease. The causative mutation is in the TAZ gene, which encodes tafazzin protein that is critical to remodeling cardiolipin in the mitochondrial membrane.

Secondary Causes of Neutropenia Infection-Related Neutropenia

4

Rituximab

11

Cephapirin

12

Oxacillin

13

Drug-induced neutropenia is the most common cause of neutropenia. Multiple drugs have been implicated in neutropenia and agranulocytosis, in both predictable and idiosyncratic patterns. Drug-induced neutropenias may reflect either suppression of marrow granulopoiesis or increased destruction of peripheral neutrophils. Many drugs cause direct dose-dependent marrow suppression, which is predictable and often mild. Others incite an idiosyncratic immunemediated destruction that can present with profound agranulocytosis. Druginduced neutropenia typically presents with a marked decline in neutrophils occurring after 1 to 2 weeks of exposure to the drug with recovery within 3 to 10 days of stopping the drug. However, atypical cases can present long after drug initiation, and others may be associated with a markedly delayed recovery (Fig. 158-5). Patients with drug-induced agranulocytosis may present with acute sepsis, leading to significant acute mortality. Recovery is often preceded by the appearance of monocytes and immature neutrophil forms. The more hypercellular the marrow is at diagnosis, the earlier marrow recovery may occur. Some common drugs known to cause neutropenia, in addition to antineoplastic, antiviral, and immunosuppressive agents, include clozapine, the antithyroidal thioamides including carbimazole, methimazole and propylthiouracil, quinidine, procainamide, sulfasalazine, and Levamisole, which is widely used as a cocaine adulterant, causing cocaine-associated neutropenia. Neutrophil recovery is speeded by G-CSF, although there are no definitive data that this improves survival. Radiation can induce marrow injury leading to acute or chronic marrow failure; in high doses, it is also a risk factor for the development of myelodysplasia and leukemia. These malignant hematopoietic diseases can themselves cause marrow failure because the malignant cells proliferate within the marrow and can also cause marrow fibrosis, both of which lead to cytopenias. These diseases are discussed in greater detail in Chapters 156 and 172, respectively.

19 22

4

Nafcillin

22

8

Penicillin G

Duration of drug exposure before onset of acute agranulocytosis Time between onset of acute agranulocytosis and normalization of neutrophil count

10

25

4

Aprindine

29

12

Captopril

32

10

Ranitidine

33

9

Quinidine

35

4

Mianserin

35

12

Propylthiouracil

36

10

Ticlopidine

24

Dapsone

39 39

10

Carbimazole

Several viral infections cause a transient neutropenia that typically resolves as the viremia abates. These include varicella, measles, rubella, hepatitis A and B, Epstein-Barr virus, influenza, parvovirus, and cytomegalovirus. The mechanisms are diverse and involve redistribution, decreased production, and immune destruction of neutrophils. Human immunodeficiency virus and acquired immunodeficiency syndrome also cause multifactorial leukopenia and neutropenia (Chapter 366). Patients often have splenomegaly and sequestration, but more commonly the neutropenia reflects immune-mediated destruction. A myriad of atypical infections like Mycobacterium tuberculosis, ehrlichiosis, rickettsia, tularemia, brucellosis, and some staphylococcal infections can cause moderate neutropenia. Any infection leading to overwhelming sepsis can cause neutropenia, but this usually reflects consumption of marrow neutrophil reserves and is typically seen in newborns and elderly patients with limited marrow reserves. There is also increased margination of neutrophils during sepsis due to systemic activation of complement, exacerbating the neutropenia.

Drug-Induced Neutropenia and Neutropenia Due to Marrow Injury

2

Dipyrone

41

8

Terbinafine

42

4

Sulfasalazine

42

11

Methimazole

42

10

Chlorpromazine

45

11

Amodiaquine

46

20

Procainamide

47

8

Carbamazepine

49

6

Clozapine

56

12

Levamisole

60

10

0

10

20

30 40 Time, days

50

60

70

FIGURE 158-5.  Median duration of treatment and neutropenia. Only drugs with more than three definite or probable reports of the time between onset of acute agranulocytosis and normalization of neutrophil count and the duration of treatment before onset of acute agranulocytosis were considered. (From Andersohn F, Konzen C, Garbe E. Systematic review: agranulocytosis induced by nonchemotherapy drugs. Ann Intern Med. 2007;146: 657-665.)

Likewise, metastatic carcinoma to the marrow can also cause marrow failure as the marrow becomes replaced by the metastatic cells.

Immune Neutropenia

Immune neutropenia can occur as an isolated phenomenon (primary immune neutropenia) or as a manifestation of an underlying systemic autoimmune disease (secondary immune neutropenia). Primary autoimmune neutropenia is primarily a disease of children, with a median age of onset of 6 to 12 months. Although infectious risk is increased, treatment is restricted to prophylactic antibiotics, with G-CSF reserved for acute infectious episodes. Ninety-five percent of patients undergo spontaneous remissions within 2 years. Nearly all of these patients have antineutrophil antibodies directed against antigens derived from the FcγIIIb receptor; these antibodies mediate neutrophil destruction by either sequestration in the spleen or complement-mediated neutrophil lysis. Secondary autoimmune neutropenia is a disease of adults and can be seen in association with hyperthyroidism, Wegener granulomatosis, rheumatoid arthritis, and systemic lupus erythematosus (SLE). The role of antineutrophil antibodies in these patients is less clear. More than 50% of patients with SLE, for example, have antineutrophil antibodies, but many have normal neutrophil counts, and there is a poor correlation between the presence of the antibodies and neutrophil number.

CHAPTER 158  Leukocytosis and Leukopenia  

Felty syndrome and large granular lymphocyte syndrome deserve separate mention. Felty syndrome occurs in association with long-standing rheumatoid arthritis (Chapter 248) and is characterized by splenomegaly and profound neutropenia. Large granular lymphocyte syndrome can occur either in the setting of rheumatoid arthritis or as an isolated phenomenon. Both Felty syndrome and large granular lymphocyte syndrome are associated with a proliferation of large granular lymphocytes, with a characteristic surface phenotype (CD3+, CD8+, CD16+, and CD57+). In rheumatoid arthritis, these two syndromes were originally thought to be separate diseases, with Felty syndrome being polyclonal and large granular lymphocyte syndrome representing a monoclonal proliferation. However, with increasing sensitivity of detection of monoclonal populations of lymphocytes, this distinction has become blurred. More than 90% of rheumatoid arthritis patients with either syndrome are human leukocyte antigen (HLA)-DR4-positive, leading to the postulate that the two entities represent the extremes of a single disease spectrum. This HLA restriction is not found among non-rheumatoid arthritis patients with large granular lymphocyte syndrome. Both syndromes cause immune-mediated neutrophil destruction by multiple mechanisms, including antineutrophil antibodies and cell-mediated destruction. Some patients may also have G-CSF resistance mediated by inhibitory G-CSF antibodies. The pathophysiology of LGL is incompletely understood. However, up to 50% of patients have activating mutations in Stat3, and all patients show constitutive activation of the Stat3 signaling pathway. There are other rare forms of immune neutropenia. Isoimmune neonatal neutropenia is a moderate to severe neutropenia of the newborn due to transplacental passage of maternal antibodies against alleles inherited from the father, resulting in neutropenia in a manner similar to the development of anemia in Rh hemolytic disease. Pure white cell aplasia is a rare disease associated with severe pyogenic infections, and also with thymoma in more than two thirds of cases. It has also occurred following ibuprofen therapy. There is a complete absence of myeloid precursors on bone marrow examination. It is immune mediated, but removal of the thymoma in thymoma-associated cases may not be sufficient for remission. Adjuvant therapy with cyclophosphamide, corticosteroids, cyclosporine, and intravenous immunoglobulin may be needed.

Neutropenia Due to Increased Margination and Hypersplenism

Complement activation can result in both acute and chronic neutropenia as a result of increased margination of the circulating neutrophil pool. This is attributed to the fact that C5a renders neutrophils more adherent and prone to aggregation within the pulmonary vasculature. This has been seen in patients suffering from burns and transfusion reactions. Complement activation may also lead to neutrophil destruction, as in paroxysmal nocturnal hemoglobinuria. Neutropenia can also reflect hypersplenism, although this is typically less common and less pronounced than the anemia and thrombocytopenia seen in patients with splenomegaly.

Neutropenia Due to Nutritional Deficiency

B12, folate (Chapter 155), and copper deficiency are associated with neutropenia. Copper deficiency is an under-recognized cause of neutropenia (usually seen with anemia), found especially in situations like total parenteral nutrition without copper supplementation, protein-losing enteropathies, celiac disease, gastric bypass surgery and postgastrectomy, malabsorption syndromes, and zinc toxicity. In addition to low serum copper and ceruloplasmin levels, distinct morphologic findings include hypogranularity and hypolobation (PHA) on peripheral smear, and cytoplasmic vacuolization of myeloid as well as erythroid precursors and ringed sideroblasts. Deficiencies in these vitamins and minerals result in ineffective myelopoiesis, maturation arrest, and megaloblastic changes with nuclear-cytoplasmic dyssynchrony. The characteristic finding in the setting of megaloblastic anemia is hypersegmentation of the neutrophil (Fig. 158-6).

Chronic Idiopathic Neutropenia

Chronic idiopathic neutropenia in adults is an acquired disorder characterized by prolonged neutropenia in the absence of any apparent underlying etiology.14 Chronic neutropenia is often an incidental finding on routine blood tests, making it impossible to know the duration of the neutropenia. Patients have no evidence of autoimmune disease, nutritional deficiency, or myelodysplasia. The syndrome is heterogeneous, with a wide range of neutrophil counts. A group of adults from Greece with chronic idiopathic neutropenia were found to have increased production of transforming growth factor-β (TGF-β) and consequent suppression of granulopoiesis by the bone marrow. These patients

1101

FIGURE 158-6.  Peripheral blood with macrocytosis and hypersegmented neutrophils in megaloblastic anemia.

tend to have very mild neutropenia, with an absolute neutrophil count that is rarely less than 800. They may have an ethnic predisposition to neutropenia; indeed, the neutropenia has been demonstrated to be linked to a genetic polymorphism in the TGF-β locus. However, these patients should be distinguished from other adults with chronic idiopathic neutropenia, many of whom have an absolute neutrophil count below 200. The etiology of neutropenia in these patients is completely unknown, but the natural history of chronic idiopathic neutropenia in adults, even in the face of very low neutrophil counts, is generally benign. Most patients require no therapy, although those with very low counts may require G-CSF if they manifest severe infections.15 Some patients with recurrent infections or troublesome aphthous ulcers require chronic G-CSF treatment. These patients typically respond to fairly low doses of G-CSF, and there is no increase in the development of myelodysplasia/AML.  

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Neutropenia is not usually associated with clinical signs and symptoms other than those of the underlying condition causing the low WBC. It becomes clinically evident, however, when it results in infection. Although patients with an absolute neutrophil count below 1000/µL have a slightly increased risk for infection, the risk increases substantially when the neutrophil count falls below 500/µL. With few neutrophils, the signs and symptoms of infection may be attenuated; pneumonia may be present with minimal infiltrate on chest radiograph, or a urinary tract infection may yield only a very mild pyuria. Therefore, fever in any neutropenic patient must be considered an emergency with prompt acquisition of cultures and administration of empirical antibiotic therapy. Because drug-induced neutropenia is the most common cause of acquired neutropenia, a careful inventory for drugs and toxins is warranted.16 Likewise, patients should be evaluated for potential precipitating malignant and inflammatory conditions. The time course of the neutropenia and infections can provide insight into the etiology (acute versus chronic, persistent versus cyclic, neonatal versus childhood versus adult onset). Attention should be paid to the skin, bones, appendages, and nails because abnormalities in these may point toward one of the congenital neutropenia syndromes. Evaluation of the complete blood count, peripheral blood smear, and vitamin B12 and folate levels should also be performed. When the neutropenia is mild and associated with anemia and thrombocytopenia, one should consider the possibility of hypersplenism. Diagnosis can often be made by the finding of palpable splenomegaly. However, especially in obese patients, abdominal imaging should be used to evaluate spleen size. Abdominal ultrasound also allows the assessment of portal venous flow with Doppler studies. If splenic enlargement is confirmed, the etiology of the splenomegaly (Chapter 159) should be determined. In patients with chronic neutropenia without a history of infection, drug or toxin exposure, or B12 or folate deficiency, bone marrow examination should be performed to rule out myelodysplasia, with assessment of morphology, flow cytometry for large granular lymphocyte syndrome, and

1102

CHAPTER 158  Leukocytosis and Leukopenia  

Neutropenia

Stop potential drugs Look for toxins

Fever, localizing signs?

Documented to be new?

Yes

No

+FH?

Yes

ANC < 500/mL

Single lineage or multilineage cytopenia

Yes

Yes

No Cycle?

Yes

Admit for IV antibiotics Consider G-CSF

ELANE mutation? Yes Cyclic neutropenia

Isolated neutropenia

Pancytopenia

Bone marrow examination Morphology, cytogenetics, flow

Low B12, folate?

Yes

Yes No Abnormal cytogenetics

LGL by flow

All normal

Megaloblastic anemia Autoimmune disease?

No

Appropriate ethnic group Yes Constitutional neutropenia

Familial neutropenia

Yes Immune cytopenia Myelodysplasia

LGL

Idiopathic neutropenia

FIGURE 158-7.  Diagnostic approach to neutropenia. ANC = absolute neutrophil count; ELANE = neutrophil elastase gene; FH = family history; G-CSF = granulocyte colony-stimulating factor; IV = intravenous; LGL = large granular lymphocyte syndrome.

cytogenetics. Once a normal marrow has been obtained, further bone marrow examination is not indicated in adults with chronic idiopathic neutropenia (Fig. 158-7).

TREATMENT  Management of the neutropenia patient is nuanced and depends on the etiology of the depressed count, its severity, and the presence or absence of fever.17

Neutropenia after Chemotherapy

In afebrile adults with acquired neutropenia following chemotherapy, antibiotic prophylaxis with a quinolone (ciproflaxin 500 mg orally twice a day, switched to 200 mg intravenously twice a day if the patient develops severe mucositis or intolerance to the oral formulation) or oral sulfamethoxazole/trimethoprim (1 double-strength tablet daily) can reduce all-cause mortality A1  and is recommended along with antifungal prophylaxis (e.g., oral voriconazole 200 mg twice daily) in patients who are expected to have profound, protracted neutropenia (7 days).18 Trimethoprim-sulfamethoxazole is specifically recommended in patients at risk for Pneumocystis jirovecii pneumonia (Chapter 321), based on neutropenia in the setting of the equivalent of 20 mg or more of prednisone/day for 1 month or longer. In afebrile neutropenia, the routine prophylactic administration of colonystimulating factors does not appear to be beneficial, A2  but many oncologic professional society guidelines recommend the use of prophylactic granulocytecolony stimulating factor (G-CSF) in patients who are receiving chemotherapy and who have a 20% or greater risk for developing febrile neutropenia based on their age, comorbid illnesses, disease characteristics, and the myelotoxicity of the chemotherapy regimen. G-CSF is also recommended during hematopoietic stem cell transplantation (Chapter 168), for patients receiving chemotherapy for non-Hodgkin lymphoma (Chapter 176) or dose-dense chemotherapy, and for patients with a history of febrile neutropenia receiving further chemotherapy. In afebrile neutropenia, granulocyte transfusions may reduce the risk of infection but do not reduce mortality. A3  The approach to the patient with fever and neutropenia is discussed in detail in Chapter 265. Neutropenia with fever is a clinical emergency because patients

are at risk for hemodynamic collapse and septic shock. Therefore, such patients should be evaluated thoroughly and cultured promptly, and empiric antibiotics should be administered within 30 to 60 minutes of presentation while awaiting culture results. The timely administration of a combination of antipseudomonal antibiotics in neutropenic patients at the onset of fever confers significant clinical benefit, with more rapid response and improved survival. With the advent of newer-generation cephalosporins, monotherapy with a third- or fourthgeneration cephalosporin may be sufficient. Randomized trials also demonstrate the survival superiority of β-lactam monotherapy compared with β-lactamaminoglycoside combination therapy. A4  The addition of a second antipseudomonal agent, vancomycin, or antifungal agent is warranted in patients at risk for resistant pseudomonal infection, resistant gram-positive infection, or fungal infections, respectively, or if the patient fails to defervesce within 3 to 5 days of antibiotic administration. These considerations are discussed in more detail in Chapter 265.

Other Neutropenias

Patients with immune neutropenia are typically treated with immunosuppressive therapy, including steroids, antithymocyte globulin, or cyclosporine, aimed primarily at treatment of the underlying autoimmune disease. Patients usually respond to G-CSF, although this treatment may induce a flare of joint symptoms in the setting of rheumatoid arthritis. Patients with large granular lymphocyte syndrome often respond to low-dose methotrexate (10 mg/m2 orally once a week), cyclosporine (100 to 600 mg or 2 to 10 mg/kg orally daily), or low-dose cyclophosphamide (50 to 100 mg orally daily). Patients with congenital neutropenia, including idiopathic, severe congenital, or cyclic neutropenia, are usually successfully managed with G-CSF for years. Before the use of G-CSF, the mean age of death for patients with severe congenital neutropenia was 2 to 3 years. With G-CSF, however, life expectancy has been extended into adulthood. Therapy is daily and chronic, given by subcutaneous injection, with doses varying by the type of neutropenia and the individual responsiveness to therapy. It is usually well tolerated, although accelerated bone loss has been observed. Growth and development appear unaffected. Patients with severe congenital neutropenia typically require the highest doses, whereas those with idiopathic neutropenia require the lowest, and patients with cyclic neutropenia fall somewhere in between.

In adults with chronic idiopathic neutropenia, G-CSF should be reserved for acute febrile episodes unless the patient has recurrent infections. Adults with chronic idiopathic neutropenia treated with G-CSF may experience significant side effects, including fever, gastrointestinal symptoms, and splenomegaly. Consequently, when adults with chronic idiopathic neutropenia require chronic G-CSF, it should be administered at the lowest dose necessary to prevent infections; it is usually sufficient to treat to maintain the absolute neutrophil count in the range of 300 to 500. For patients with an inflammatory, infectious, or drug-induced neutropenia, the recommendation is to treat the underlying condition or stop the offending agent.19 The use of antibiotics to prevent infection in other patients with chronic neutropenias has not been extensively studied but is typically not recommended and should be based on clinical context. There is no evidence that a low bacterial diet can prevent infection or mortality in neutropenic cancer patients. A5  Stem cell transplantation (Chapter 168) can be curative for a number of the congenital neutropenia and bone marrow failure syndromes. It is, however, not without risk and should therefore be reserved for patients with severe neutropenia complicated by recurrent infection definitively shown to be due to marrow failure.



Leukopenia Due to Deficiency of Other Cell Lines

Lymphocyte production takes place in a variety of anatomic sites, and lymphocyte trafficking from those sites is bidirectional, making it difficult to understand lymphocyte dynamics in the same way that we do for neutrophils. Despite this, the peripheral lymphocyte count seems to be maintained in a narrow range at 2000 to 4000/µL, 20% of which are B cells and 70% of which are T cells. Lymphocytopenia is a total lymphocyte count of less than 1500/µL. It can reflect decreased production, defective trafficking, or increased loss or destruction. Decreased production can result from protein and calorie malnutrition; lymphocyte progenitor pool injury secondary to radiation, chemotherapy, or immunosuppressive agents; and congenital immunodeficiency states. Endogenous or exogenous glucocorticoid excess can cause lymphocytopenia by altering lymphocyte trafficking. This can also occur as the result of acute bacterial or fungal infections, certain viral infections, and granulomatous disease. Finally, many viruses cause direct destruction of lymphocytes, as do antilymphocyte antibodies seen in patients with underlying autoimmune diseases. Lymphocytes can also be lost from intestinal lymphatics in cases of protein-losing enteropathy, primary disease of the gut or intestinal lymphatics, or gut edema secondary to severe heart failure. Lymphocytopenia should prompt comprehensive assessment of the immune system, including lymphocyte subtyping, quantitative immunoglobulins, and skin testing to detect deficiencies of cell-mediated immunity. Treatment is aimed at the underlying disease, but intravenous immunoglobulin can be administered to patients with hypogammaglobulinemia, and transplantation can be performed in patients with severe deficiencies of cell-mediated immunity due to impaired lymphocyte production and function. Monocytopenia, eosinopenia, and basophilopenia can accompany bone marrow failure syndromes or acute infection, malignancy, or severe injury. This is thought to reflect elevations in glucocorticoids, prostaglandins, and epinephrine. These humoral factors have the greatest impact on eosinophils, and absence of eosinopenia in these settings should prompt suspicion for adrenal insufficiency, a primary myeloproliferative syndrome, parasitic infection, or primary hypereosinophilic syndrome. Monocytopenia is less frequent, probably owing to the diverse roles monocytes play in normal human physiology; prolonged and extreme monocytopenia may not be compatible with life.

  Grade A References A1. Gafter-Gvili A, Fraser A, Paul M, et al. Antibiotic prophylaxis for bacterial infections in afebrile neutropenic patients following chemotherapy. Cochrane Database Syst Rev. 2012;1:CD004386. A2. Skoetz N, Bohlius J, Engert A, et al. Prophylactic antibiotics or G(M)-CSF for the prevention of infections and improvement of survival in cancer patients receiving myelotoxic chemotherapy. Cochrane Database Syst Rev. 2015;12:CD007107. A3. Estcourt LJ, Stanworth S, Doree C, et al. Granulocyte transfusions for preventing infections in people with neutropenia or neutrophil dysfunction. Cochrane Database Syst Rev. 2015;6:CD005341. A4. Paul M, Dickstein Y, Schlesinger A, et al. Beta-lactam versus beta-lactam-aminoglycoside combination therapy in cancer patients with neutropenia. Cochrane Database Syst Rev. 2013;6:CD003038. A5. van Dalen EC, Mank A, Leclercq E, et al. Low bacterial diet versus control diet to prevent infection in cancer patients treated with chemotherapy causing episodes of neutropenia. Cochrane Database Syst Rev. 2016;4:CD006247.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 158  Leukocytosis and Leukopenia  

GENERAL REFERENCES 1. Bartels M, Govers AM, Fleskens V, et al. Acetylation of C/EBPepsilon is a prerequisite for terminal neutrophil differentiation. Blood. 2015;125:1782-1792. 2. Widick P, Winer ES. Leukocytosis and leukemia. Prim Care. 2016;43:575-587. 3. Shah SS, Parikh RS, Vaswani LP, et al. Familial Pelger-Huet anomaly. Indian J Hematol Blood Transfus. 2016;32:347-350. 4. Wolach B, Gavrieli R, Wolach O, et al. Leucocyte adhesion deficiency—a multicentre national experience. Eur J Clin Invest. 2018;1-17. 5. Venugopal S, Mascarenhas J. Chronic neutrophilic leukemia: current and future perspectives. Clin Lymphoma Myeloma Leuk. 2018;19:129-134. 6. Lynch DT, Hall J, Foucar K. How I investigate monocytosis. Int J Lab Hematol. 2018;40: 107-114. 7. Vander Meeren S, Heyrman B, Renmans W, et al. Lymphoma-like monoclonal B cell lymphocytosis in a patient population: biology, natural evolution, and differences from CLL-like clones. Ann Hematol. 2018;97:1219-1227. 8. Larsen RL, Savage NM. How I investigate eosinophilia. Int J Lab Hematol. 2019;41:153-161. 9. Palmblad J, Höglund P. Ethnic benign neutropenia: a phenomenon finds an explanation. Pediatr Blood Cancer. 2018;65:1-7. 10. Donadieu J, Beaupain B, Fenneteau O, et al. Congenital neutropenia in the era of genomics: classification, diagnosis, and natural history. Br J Haematol. 2017;179:557-574.

1103.e1

11. Nayak RC, Trump LR, Aronow BJ, et al. Pathogenesis of ELANE-mutant severe neutropenia revealed by induced pluripotent stem cells. J Clin Invest. 2015;125:3103-3116. 12. Furutani E, Newburger PE, Shimamura A. Neutropenia in the age of genetic testing: advances and challenges. Am J Hematol. 2019;94:384-393. 13. Andrés E, Mourot-Cottet R. Non-chemotherapy drug-induced neutropenia—an update. Expert Opin Drug Saf. 2017;16:1235-1242. 14. Dale DC, Bolyard AA. An update on the diagnosis and treatment of chronic idiopathic neutropenia. Curr Opin Hematol. 2017;24:46-53. 15. Sicre de Fontbrune F, Moignet A, Beaupain B, et al. Severe chronic primary neutropenia in adults: report on a series of 108 patients. Blood. 2015;126:1643-1650. 16. Dale DC. How I diagnose and treat neutropenia. Curr Opin Hematol. 2016;23:1-4. 17. Palmblad J, Nilsson CC, Hoglund P, et al. How we diagnose and treat neutropenia in adults. Expert Rev Hematol. 2016;9:479-487. 18. Taplitz RA, Kennedy EB, Bow EJ, et al. Antimicrobial prophylaxis for adult patients with cancerrelated immunosuppression: ASCO and IDSA clinical practice guideline update. J Clin Oncol. 2018;3043-3054. 19. Curtis BR. Non-chemotherapy drug-induced neutropenia: key points to manage the challenges. Hematology Am Soc Hematol Educ Program. 2017;2017:187-193.

1103.e2

CHAPTER 158  Leukocytosis and Leukopenia  

REVIEW QUESTIONS 1. A 20-year-old African American college student receives a scholarship to spend her junior year abroad in France. She needs to have a health form completed before her departure. A complete blood count (CBC) reveals neutropenia, and her physician advises her to postpone the scholarship and to see you for evaluation. She is healthy with no history of excessive infections. Her physical examination is normal. CBC shows: white blood cell count 3400 w/ 30% neutrophils, hematocrit 41, platelets 200,000. What should you advise her? A . She should have a bone marrow biopsy to rule out a bone marrow process causing neutropenia. B. She should have serial blood counts to rule out cyclic neutropenia. C. She should have chromosome analysis of her peripheral blood. D. She should be evaluated for an elastase mutation. E. She should go to France without any further evaluation. Answer: E  The normal neutrophil count is partially determined by ethnic background. African American males frequently have a neutrophil count of 1000 to 1500, and African American females may have counts that are even lower. Many may actually have counts below 1000. With a total neutrophil count of 1120, a normal hematocrit and platelet count, and an unremarkable history and physical examination, the patient should be diagnosed with constitutional neutropenia. This requires no further evaluation, although if Duffy antigen receptor for chemokine testing is available, confirming she is Duffy negative will confirm the diagnosis. The patient should be reassured that these counts are normal for her and present no undue risk. She should leave for France as planned. (See Ethnic and Benign Familial [Constitutional] Neutropenia.) 2. A 1-month-old baby boy presents with fever and cough and is found to have pneumonia associated with severe neutropenia. His parents have normal blood counts but had lost a baby girl at age 3 months under similar circumstances 2 years before. Which of the following statements is not true? A . The patient likely has a mutation in neutrophil elastase. B. The child should be treated with granulocyte colony-stimulating factor (G-CSF). C. The child is at risk for developing myelodysplastic syndrome and acute myeloid leukemia (myelodysplasia/AML). D. If the child survives, his offspring will probably have normal neutrophil counts. E. The child’s white blood cell count likely has a 21-day cycle in and out of the normal range. Answer: A  This child has a disease that is apparent in neither parent but has a sibling that died of the same disease. This is consistent with an autosomal recessive form of severe congenital neutropenia. Severe congenital neutropenia related to neutrophil elastase mutation is an autosomal dominant disorder. Cyclic neutropenia is a milder autosomal dominant disorder, also due to mutations in ELANE; it is clinical mild and rarely presents in infancy. Although it can appear sporadically, the occurrence in a sibling makes that virtually impossible. This case is most consistent with Kostmann syndrome, which is due to mutations in the HAX1 gene. Because it is a very rare recessive disorder, if the child lives to adulthood and has offspring, they will be obligate heterozygotes and, like the patient’s parents, will have normal counts. Like patients with dominantly inherited severe congenital neutropenia associated with neutrophil elastase mutations, these patients respond to G-CSF and have an increased risk for developing myelodysplasia/AML. (See Severe Congenital Neutropenia.) 3. A 45-year-old man presents to the emergency department after a motor vehicle crash. He is afebrile with a blood pressure of 160/95 mm Hg and a pulse of 110 beats per minute. He has several scrapes and bruises, but no major injuries. Admission complete blood count reveals: white blood cell count 16,000, hematocrit 45, and platelets 280,000. What is the most likely explanation for his leukocytosis? A . Acute bacterial infection B. Undiagnosed chronic myelogenous leukemia C. Stress-induced demargination of neutrophils D. Cytokine release stimulating increased marrow production and release of neutrophils E. Unsuspected splenic rupture

Answer: C  Acute stress is commonly associated with leukocytosis. Acute leukocytosis in almost any setting occurs by demargination, in this case in response to the release of catecholamines during the stress of the accident, also causing his tachycardia. Cytokine-induced marrow proliferation and release of neutrophils reflect more chronic inflammation and could not have occurred in less than about a week. Unsuspected splenic rupture would likely lead to hypotension and shock. There is no reason to consider a diagnosis of chronic myeloproliferative disease or sepsis in the absence of other evidence for a secondary condition. The patient should have his counts rechecked in several days or weeks to rule out any other pathology. (See Stress.) 4. A 3-year-old child presents with recurrent infections and a diagnosis of Chédiak-Higashi syndrome is made. Which of the following is not true of his disease? A . He is likely to have associated nystagmus. B. His sibling has a 50% chance of having the disease. C. He is likely to have abnormal neutrophil granules. D. He is likely to develop hemophagocytic syndrome. E. He is likely to have a defect in skin pigmentation. Answer: A  Chédiak-Higashi syndrome is a generalized disorder of granule formation caused by abnormalities in the LYST gene, which is required for the correct trafficking of proteins into granules. Consequently, patients with Chédiak-Higashi syndrome have other defects associated with abnormal granule formation, including oculocutaneous albinism, with decreased pigmentation and nystagmus, and abnormal-appearing neutrophil granules. Because the disease is an autosomal recessive disorder, the heterozygous parents are usually totally unaffected, and siblings have a 25% chance of inheriting the disease. Patients with Chédiak-Higashi syndrome often have a terminal phase of the disease associated with the development of hemophagocytic syndrome, manifesting as fever, splenomegaly, and pancytopenia, often as a result of EpsteinBarr virus infection. (See Other Congenital Syndromes with Associated Neutropenia.) 5. A 1-month-old infant presents with a temperature of 103° F and a white blood cell count (WBC) of 90,000. He has a history of delayed umbilical cord separation, poorly healing skin lesions, recurrent otitis media, and failure to thrive. He is admitted to the hospital, where he is documented to have methicillin-resistant Staphylococcus aureus (MRSA). He is placed on appropriate antibiotics, but he remains febrile, and his WBC is persistently elevated in the 80,000 to 100,000 range. This patient is likely to have a mutation in which of the following? A . ELANE B. HAX1 C. Integrin receptor β chain D. LYST E. G-CSFR Answer: C  The child has a congenital disorder of neutrophils that results in leukocytosis. The diagnosis is leukocyte adhesion deficiency. This disease arises from defects in leukocyte adhesion to endothelium. It can arise from defects in the integrin receptor common β chain (leukocyte adhesion deficiency-1), with loss of expression of leukocyte function–associated antigen 1 (LFA-1), C3bi receptor, and gp150;95, leading to a failure to ingest and kill microbes opsonized by C3bi. It can also arise from an abnormality of selectin glycosylation (leukocyte adhesion deficiency-2). ELANE and HAX1 mutations are causes of severe congenital neutropenia, and affected patients present with profound neutropenia. LYST mutations are the cause of Chédiak-Higashi syndrome, which is also associated with neutropenia. Congenital G-CSFR mutations are very rare, although two or three patients in the literature have been described to have such mutations as the basis for G-CSF–resistant severe congenital neutropenia. More commonly, G-CSFR mutations are somatic mutations that arise in patients with severe congenital neutropenia and may be a harbinger of the development of myelodysplastic syndrome or acute myeloid leukemia. (See Leukocyte Adhesion Deficiency.)

1103

CHAPTER 159  THE PATIENT WITH LYMPHADENOPATHY AND SPLENOMEGALY  

159  APPROACH TO THE PATIENT WITH LYMPHADENOPATHY AND SPLENOMEGALY JANE N. WINTER

  LYMPHADENOPATHY  

PATHOBIOLOGY AND CLINICAL MANIFESTATIONS

Lymph nodes are secondary lymphoid tissues located along the course of the lymphatic vessels. These round or bean-shaped structures typically measure less than 1 cm in diameter and are strategically situated to filter lymphatic fluid and to intercept microorganisms and other antigens. The hilus is the point of indentation in these otherwise round tissues where blood vessels enter and leave the lymph node. Lymphatic fluid enters the lymph node in afferent lymphatic vessels that empty into the subcapsular sinus, traverses the lymph node, and then leaves through a single efferent lymphatic at the hilus. Lymph nodes consist of B and T lymphocytes, macrophages, and dendritic cells that interact to generate an immune response to the antigens transported in the lymphatic fluid. These cells function together to generate the normal immune response by means of antigen processing, presentation, and recognition and the resulting proliferation of effector B and T lymphocytes. Lymph node enlargement generally results from cellular proliferation and expansion of lymph node components associated with the normal immune response to foreign antigens. Alternatively, it can be caused by infiltration of the lymph node by a neoplastic process or deposition of a foreign substance (e.g., in storage disease). Whether (lymph) adenopathy is localized or generalized, its location, and the age of the patient, help to focus the investigation into its cause. Diffuse lymphadenopathy may occur in the setting of generalized infections, systemic disorders of the immune system, or widespread neoplasia (Table 159-1). Localized lymphadenopathy usually but not always results from a regional problem. The particular site may be informative. For example, epitrochlear or supraclavicular adenopathy is problematic in any adult or child, and it signifies likely disease, whereas inguinal adenopathy is common and of little concern in the absence of other findings, commonly related to infection or trauma to the lower extremities. Thoracic adenopathy may be caused by a variety of nonmalignant conditions including sarcoidosis and Castleman disease in addition to infection.1 In young children who are continually exposed to new antigens, palpable lymphadenopathy is typical. The location of the enlarged lymph nodes reflects the site of infection and differs with the age of the child. For example, recurrent sore throats and upper respiratory infections are regularly associated with tender submandibular and/or cervical adenopathy.2

Differential Diagnosis

Lymph nodes that are larger than 1 cm are generally considered to be abnormal, although adenopathy in some locations is considered abnormal and worthy of investigation even when subcentimeter in size (supraclavicular, epitrochlear, or popliteal lymph nodes). The differential diagnosis of lymphadenopathy is vast.3 A careful history and physical examination, with consideration of the age of the patient, associated symptoms, and distribution of the adenopathy, will often but not always guide the investigation to the underlying cause (Table 159-2). Often, no cause is identified, and careful follow-up is required. The general categories of infection, autoimmune disorders, and malignancy cover the majority of cases. Worldwide, infection by bacteria, mycobacteria, fungi, chlamydia, parasites, and viruses are the major causes of lymph node enlargement. The location of the lymph node enlargement and whether the lymphadenopathy is localized

CHAPTER 159  THE PATIENT WITH LYMPHADENOPATHY AND SPLENOMEGALY  

ABSTRACT

The patient with lymphadenopathy and/or splenomegaly is one of the most challenging for the general internist. The differential diagnosis is broad, with infection, autoimmune disorders, and malignancy representing the vast majority of cases. A careful and complete history is the first and most important step in the assessment of the patient with lymphadenopathy and/or splenomegaly, including travel history and possible exposures to infectious agents and sexual and family history. Similarly, the physical examination must be complete, with attention to the distribution of the adenopathy, size, and associated erythema or tenderness. In the patient with splenomegaly, its overall size, the presence of associated lymphadenopathy, tonsillar enlargement, and hepatomegaly are of particular importance. A complete blood count with review of the peripheral smear may be especially helpful in identifying the underlying cause. Both lymphadenopathy and splenomegaly are often first detected incidentally when imaging is performed for an unrelated issue. In many cases, close observation rather than biopsy or splenectomy is the best course of action in the asymptomatic patient.

KEYWORDS

lymphadenopathy lymphoproliferative disorder splenomegaly hypersplenism splenectomy

1103.e3

1104

CHAPTER 159  THE PATIENT WITH LYMPHADENOPATHY AND SPLENOMEGALY  

TABLE 159-1 DIFFERENTIAL DIAGNOSIS OF LYMPHADENOPATHY CAUSES Infectious  Bacteria

 Fungi  Parasites  Viruses Immune-mediated   Autoimmune disorders   Drug reactions Malignant  Hematologic   Solid tumors Unusual   Storage disease  Sarcoidosis   Castleman disease   Kawasaki vasculitis   Kikuchi lymphadenitis   Kimura disease  Endocrinopathies   Sinus histiocytosis with massive lymphadenopathy (Rosai-Dorfman disease)

EXAMPLES Group A streptococcus, staphylococcal skin infections, syphilis, cat-scratch disease, Chlamydia trachomatis, typhoid fever, tuberculosis Histoplasmosis, blastomycosis, coccidioidomycosis Malaria, toxoplasmosis Infectious mononucleosis (Epstein-Barr virus), cytomegalovirus, HIV Rheumatoid arthritis, systemic lupus erythematosus

A

Lymphoma, leukemia, myeloproliferative neoplasms Metastatic to lymph nodes Gaucher disease, Niemann-Pick disease

Addison disease

HIV = human immunodeficiency virus.

B FIGURE 159-1.  Axillary (A) and retroperitoneal lymph nodes (B) in a patient with follicular lymphoma. Arrows indicate examples of enlarged lymph nodes.

TABLE 159-2 CLUES TO THE CAUSE OF LYMPHADENOPATHY Age Symptoms (associated or regional) Localized versus generalized lymphadenopathy Tenderness, warmth, overlying erythema, mobility, pulsatile Size Location/distribution

or generalized may provide a clue to the cause. Lymph node enlargement in the drainage region of any pyogenic infection is expected. Dramatic regional lymph node enlargement with fluctuant lymph nodes (i.e., buboes) is a hallmark of infection with the coccobacillus Yersinia pestis, causing bubonic plague. In cat-scratch fever, enlarged nodes close to the site of the flea bite carrying Bartonella henselae are common presenting signs. Mediastinal lymphadenopathy may occur in inhalational anthrax. Cervical adenopathy is commonly a manifestation of tuberculosis, and in parts of the world with limited resources, it may lead to empirical antituberculous therapy rather than biopsy. In contrast, diffuse adenopathy may represent a systemic infection with Epstein-Barr virus (i.e., infectious mononucleosis), cytomegalovirus, or human immunodeficiency virus (HIV) or with the parasite Toxoplasma gondii. Autoimmune disorders may be associated with either localized or generalized lymphadenopathy. Lymph node enlargement in this setting is especially challenging given the increased frequency of malignant lymphoproliferative disorders in these patients. Biopsy and expert hematopathology consultation are required to exclude the possibility of lymphoma. Lymphadenopathy occurring as a result of a drug reaction is uncommon, but it may be exceptionally difficult to distinguish histopathologically from malignant lymphoma. A wide range of drugs, including antiepileptics and antibiotics, have been reported to cause lymphadenopathy, but associated rash, eosinophilia, and liver function abnormalities are helpful in distinguishing a drug reaction from other causes. Malignancy commonly manifests as lymphadenopathy. Enlarged lymph nodes may be involved or replaced by a hematologic malignancy such as Hodgkin or non-Hodgkin lymphoma (Fig. 159-1), or they may represent a

site of metastasis from a solid tumor, most often occurring in the drainage area of the primary tumor; for example, breast cancer is often accompanied by ipsilateral axillary nodal involvement and prostate cancer by iliac node involvement. Distant spread to the lymph nodes may also occur. In addition to infectious, autoimmune, and malignant causes of lymphadenopathy, there are some less common causes such as storage disease (as in Gaucher or Niemann-Pick disease) or amyloidosis (associated with a hematologic malignancy such as multiple myeloma or non-Hodgkin lymphoma, an inherited form of the disease, or with chronic inflammation). Sarcoidosis may manifest as enlarged lymph nodes, as can endocrinopathies such as thyroid disease or adrenal insufficiency.

Evaluation of the Patient with Lymphadenopathy History and Physical Examination

A careful and complete history is the first and most important step in the evaluation of the patient with lymphadenopathy (Table 159-3). The age of the patient, recent travel history, possible infectious or toxic exposures through employment or hobbies, recent immunizations, current medications, and social, sexual, and family history may provide important clues to the cause. A complete review of systems is required. Not only can the presence of symptoms be informative, but also their pattern, character, and duration (e.g., the high spiking fevers characteristic of Hodgkin lymphoma, known as Pel-Ebstein fevers) may suggest a particular cause. Similarly, particular symptoms such as pruritus suggest specific diagnoses (e.g., also Hodgkin lymphoma). The physical examination must also be meticulous with attention not only to the distribution of the adenopathy but also to the presence of other physical findings (e.g., tonsillar enlargement in the patient with cervical adenopathy and Burkitt lymphoma, parotid enlargement in Sjögren syndrome, rash in T-cell lymphomas, joint findings in autoimmune disorders, or a breast mass in the patient with axillary lymphadenopathy) (Table 159-4). Malignant lymph nodes are generally painless, whereas those that represent infection may be tender. Lymphadenopathy resulting from metastatic solid tumor is

CHAPTER 159  THE PATIENT WITH LYMPHADENOPATHY AND SPLENOMEGALY  

TABLE 159-3 TAKING A COMPLETE HISTORY IN THE ADULT PATIENT WITH LYMPHADENOPATHY History of present illness   Duration, recent changes   Associated symptoms: fevers, drenching night sweats, shaking chills, weight loss, sore throat, arthritis or arthralgias, rash or skin lesions, shortness of breath, chest pain   Ill contacts, recent and remote travel, tick or insect bites Past medical history   Autoimmune disease, history of immunosuppressive therapy, history of infectious disease Family history   Malignancy, especially hematologic malignancy, or infection Social history   Sexual activity—risk for HIV infection, other sexually transmitted diseases   Occupation—exposure to toxins, livestock, pets   Smoking or drinking Medications: immunosuppressive agents, phenytoin, allopurinol, antibiotics HIV = human immunodeficiency virus.

TABLE 159-4 EVALUATING THE PATIENT WITH LYMPHADENOPATHY Complete physical examination with particular attention to:   Tonsillar enlargement, evidence of superior vena cava syndrome   Attention to all lymph node groups including epitrochlear, femoral  Hepatosplenomegaly  Rash   Joints, deformities Preliminary laboratory examination   Complete blood count with differential, platelet count   Sedimentation rate, testing for mononucleosis in the right setting Imaging   Chest radiograph (assess for mediastinal mass, hilar enlargement)   Ultrasound for children up to age 14 years   CT with intravenous contrast or contrast-enhanced MRI  PET/CT not recommended unless diagnosis of lymphoma or other malignancy Pathologic diagnosis   Needle aspiration for investigation of infectious causes   Core biopsy if excisional biopsy unsafe   Excisional biopsy preferred especially if malignancy suspected CT = computed tomography; MRI = magnetic resonance imaging; PET = positron emission tomography.

characteristically hard and may be fixed or matted, whereas that representing hematologic malignancies is typically soft and may be mobile, but exceptions occur. Whereas borderline and slightly enlarged lymph nodes are commonly related to infectious causes, sizeable nodes are worrisome and warrant prompt investigation.

Imaging

Imaging with computed tomography (CT) or magnetic resonance imaging (MRI) (the latter in certain instances where radiation exposure is to be avoided as in pregnancy), but not combined positron emission tomography (PET) and CT (PET/CT), is the first step in the evaluation of worrisome palpable adenopathy in adults.4 Ultrasonography is recommended as the preferred approach for children with cervical adenopathy. Clearly, borderline adenopathy or lymphadenopathy associated with symptoms or history suggestive of a benign origin (e.g., mononucleosis in a college student) does not merit immediate imaging and may be watched.5 PET/CT is reserved for staging and evaluation during and after treatment of patients with an established diagnosis of malignant lymphoma or a variety of solid tumors. Although most patients with lymphadenopathy come to medical attention because they detect a lump on self-examination, an increasing number of patients with lymphadenopathy are identified incidentally when they undergo imaging for unrelated issues (e.g., chest radiography for infectious symptoms; MRI of the spine for back pain related to a disc herniation). These patients pose a special challenge.6 Evaluation of the patient with incidentally noted intra-abdominal or pelvic adenopathy, like that of the patient with a new palpable lump, requires that the clinician integrate the clinical history and radiographic findings. This is

1105

best done together with the radiologist reading the imaging study. A comparison of the radiologic findings to prior imaging studies can often eliminate concern if the findings are long-standing. Imaging findings considered worrisome by the American College of Radiology include: (1) short axis diameter 1 cm or larger in the retroperitoneum; (2) abnormal architecture (round, indistinct hilum); (3) presence of enhancement (necrosis; hypervascularity); and (4) increased number of nodes defined as a cluster of three or more in a single nodal station or two or more in at least two regions. Biopsy should be considered if the nodes fall into any of these categories and there is clinical or laboratory evidence supporting the presence of a lymphoproliferative disorder. In many cases, the patient may be followed with repeat imaging in 3 to 6 months. If stable for 1 year, an abnormal lymph node or group of lymph nodes is most likely benign. A trial of antibiotics is often indicated, especially in children. Steroids should be avoided because they may mask the symptoms and cause shrinkage of the nodes in hematologic malignancies.

Biopsy

The decision to perform a biopsy of a lymph node depends on the clinical history and presentation and the ease with which a biopsy can be performed. For patients with symptoms, an early biopsy of a palpable lymph node may provide an answer to a puzzling case but may add to the confusion if the biopsy result is not definitive. An excisional rather than a core biopsy is most likely to provide a definitive diagnosis, but it is not practical in all cases. When there are no easily accessible peripheral nodes for biopsy, a core biopsy may be obtained by the interventional radiologists with the assistance of ultrasound or CT guidance. Fine-needle aspiration may be adequate to diagnose a solid tumor or infection but is insufficient for distinguishing malignant lymphoma from a reactive lymph node, and it is discouraged if lymphoma is in the differential diagnosis. An excisional biopsy remains the gold standard.7 Studies for infectious agents including cultures and special stains for infectious agents, immunophenotypic analysis by flow cytometry, and molecular studies complement morphology in making a diagnosis.8 In contrast, if head and neck cancer is in the differential diagnosis, the patient should have a complete ear, nose, and throat examination at the outset to avoid complicating subsequent surgery. Face-to-face discussion between the clinician and the pathologist is the most effective approach to determining a diagnosis, especially for the most challenging cases.9

  SPLENOMEGALY  

PATHOBIOLOGY

The spleen is the largest lymphatic organ in the body, and some liken it to a very large lymph node. Similar to lymph nodes, the spleen has multiple and diverse functions, including its role as a filter removing defective red blood cells and infectious agents, its important role in immune response to bloodborne antigens, and a role in regulating blood volume. Consistent with its multiple functions and its role as a massive filter, the spleen has an open circulation with three kinds of compartments: the white pulp, marginal zone, and red pulp. Blood enters the spleen through the splenic artery and its branches, filters through the splenic cords and periarterial lymphatic sheath, and is exposed in the white pulp to immunologically active cells including macrophages, B lymphocytes, and T lymphocytes. There, microorganisms and foreign proteins are recognized, and an immune response to blood-borne antigens is initiated. In splenectomized individuals, the absence of this splenic function makes them especially susceptible to certain infections, especially those with encapsulated organisms such as Streptococcus pneumoniae. Separating white pulp from red, the marginal zone acts as a transition zone. More than half the volume of the spleen is composed of splenic red pulp and consists primarily of erythrocytes along with macrophages and dendritic cells and other white blood cells. Senescent or defective red blood cells are identified and destroyed in the red pulp. The process of hemolysis may be mechanical and/or immunologic as in autoimmune hemolytic anemia, where the antibody-coated red blood cells are phagocytosed by macrophages. In the absence of a functioning spleen, basophilic inclusions representing nuclear remnants, known as Howell-Jolly bodies, are seen in circulating red blood cells. The presence of Howell-Jolly bodies (Fig. 159-2) in peripheral blood indicates that the patient has either undergone splenectomy or has a nonfunctioning spleen, as in patients with sickle cell disease who have experienced repeated splenic infarcts. Last, and perhaps less well appreciated, the spleen plays a role in regulating blood volume by means of a variety of mechanisms. These include release of

CHAPTER 159  THE PATIENT WITH LYMPHADENOPATHY AND SPLENOMEGALY  

How to Approach the Patient with Lymphadenopathy

The patient with lymphadenopathy may come to medical attention in a number of ways. An adult may identify an enlarged lymph node through self-examination, whereas a child may only come to attention when there are associated symptoms or the node has become easily visible to a parent. As noted earlier, it is more and more common for asymptomatic lymphadenopathy to be identified on imaging performed for unrelated purposes. The location of the enlarged node(s), size, consistency, presence of generalized lymphadenopathy or hepatosplenomegaly, or associated symptoms are all helpful in developing an individualized approach. If nodes are large, 2 to 3 cm in size, and firm with associated symptoms, the risk for malignancy is high, and a diagnostic procedure should be performed expeditiously. A complete blood count with differential and examination of the peripheral smear along with a plain chest radiography may be helpful in sorting out which patient is best served by observation for what may be infectious mononucleosis or another infectious and uncomplicated process and who should proceed quickly to biopsy. A short course of antibiotics with close observation is often the preferred strategy with little risk if the patient is reliable and will return if the lump grows rather than regresses.

1105.e1

1106

CHAPTER 159  THE PATIENT WITH LYMPHADENOPATHY AND SPLENOMEGALY  

FIGURE 159-2.  Howell-Jolly body in an erythrocyte. This is evidence of splenectomy or a nonfunctional spleen.

TABLE 159-5 DIFFERENTIAL DIAGNOSIS OF SPLENOMEGALY CAUSES Infectious  Bacteria  Fungi  Parasites  Viruses Immune-mediated   Autoimmune disorders

EXAMPLES Endocarditis, brucellosis, syphilis, tuberculosis, Rocky Mountain spotted fever Histoplasmosis, toxoplasmosis Malaria, leishmaniasis Infectious mononucleosis (Epstein-Barr virus), HIV, cytomegalovirus

  Drug reaction

Rheumatoid arthritis, systemic lupus erythematosus, autoimmune hemolytic anemia Phenytoin

Malignant  Hematologic   Solid tumor

Lymphoma, leukemia, myeloproliferative neoplasm Metastatic to spleen

Congestive

Liver disease, portal vein thrombosis, congestive heart failure

Nonmalignant hematologic disorders

Hereditary spherocytosis, congenital or acquired hemolytic anemias, hemoglobinopathies, extramedullary hematopoiesis

Unusual   Storage disease  Sarcoidosis   Cysts, pseudocysts  Amyloidosis

Gaucher disease, Niemann-Pick disease

HIV = human immunodeficiency virus.

blood through splenic contraction and the splenorenal reflex promoting fluid and sodium retention by the kidneys during septic shock.  

DIAGNOSIS

As with lymphadenopathy, the differential diagnosis of splenomegaly is wide ranging (Table 159-5).10 The same principles hold. First, a careful history and physical examination, with attention to the age of the patient, the size of the spleen, and associated symptoms will often but not always guide the investigation to the underlying cause. Often, no cause is identified, and careful follow-up is required. Again, the general categories of infection, autoimmune disorders, and malignancy cover the majority of cases. As with lymphadenopathy, the infectious causes of splenomegaly are myriad. Geography and travel history are often important clues to an underlying infection. The most likely causes of splenomegaly in equatorial Africa or in a recent African immigrant are different from those of a North American urbanite. Bacterial infections commonly associated with splenomegaly include endocarditis, brucellosis, typhoid fever, and rickettsial disorders such as Rocky Mountain spotted fever. Mycobacterial infections including typical and atypical mycobacteria may manifest with splenomegaly, as can fungal infections such

FIGURE 159-3.  Splenomegaly (indicated by arrow) in a patient with hemophagocytic lymphohistiocytosis.

as histoplasmosis. Among the parasitic disorders, malaria is nearly always accompanied by an enlarged spleen and should be high on the differential diagnosis list in the right setting. Epstein-Barr virus, the cause of infectious mononucleosis, is a common cause of splenomegaly that is usually self-limited. HIV may also be associated with splenomegaly, although often the causative agent is a coexisting infection with hepatitis C virus. Autoimmune disorders such as rheumatoid arthritis (RA) and systemic lupus erythematosus are commonly associated with mild to moderate splenomegaly. Felty syndrome is characterized by splenomegaly, usually in the setting of long-standing RA, with associated neutropenia. Autoimmune hemolytic anemia is characterized by splenomegaly. Reactions to a variety of commonly prescribed drugs such as antibiotics may result in splenomegaly, through direct effects on the spleen, serum sickness, or indirectly through effects on the liver. Hematologic malignancies including myeloproliferative neoplasms frequently manifest with splenomegaly. Patients with lymphoproliferative disorders such as chronic lymphocytic leukemia or mantle cell lymphoma will in many cases have easily appreciated lymphadenopathy in addition to a palpable spleen. Isolated splenomegaly without associated lymphadenopathy is characteristic of hairy cell leukemia and myeloproliferative neoplasms such as idiopathic myelofibrosis. Splenomegaly with weight loss related to compression of the stomach and early satiety may be the presenting complaint. Solid tumors may sometimes metastasize to the spleen and result in its enlargement, but this is an unusual occurrence. Among solid tumors, malignant melanoma is the one most likely to be associated with splenic involvement, but even so, palpable splenomegaly is rare. Congestive splenomegaly may occur as the result of chronic liver disease with portal hypertension or in the case of splenic or portal vein thrombosis. Disorders that result in excessive destruction of abnormal red blood cells such as hereditary spherocytosis or hemoglobinopathies are also associated with splenomegaly. Storage diseases such as Gaucher disease, sarcoidosis, amyloidosis, thyrotoxicosis, and hemophagocytic lymphohistiocytosis are some of the other less common causes of splenomegaly (Fig. 159-3).

Evaluation of the Patient with Splenomegaly History

As with lymphadenopathy, a careful and complete history is the first and most important step in the evaluation of the patient with splenomegaly. The age of the patient, travel history, possible infectious or toxic exposures through employment or hobbies, recent immunizations, current medications, and

CHAPTER 159  THE PATIENT WITH LYMPHADENOPATHY AND SPLENOMEGALY  

social, sexual, and family history all may contribute important clues to the cause. A complete review of systems is required. A review of old records including imaging may provide evidence that splenomegaly is long-standing.

Physical Examination

Today, splenomegaly is commonly first detected incidentally on imaging performed for what at first appears to be an unrelated issue such as diverticulitis or nephrolithiasis. The borderline or slightly enlarged spleen may be difficult to appreciate on examination. There are many approaches to palpating the spleen. Positioning the patient in the right lateral decubitus position with knees slight flexed is an approach favored by some. For most right-handed examiners, the patient should be relaxed and in the right lateral decubitus position. It is best to begin very low in the pelvis and work up to the left upper quadrant so as not to miss the edge of a massively enlarged spleen. Inspection and percussion combine with palpation to provide the best chance of identifying the enlarged spleen. It is best to have the patient take deep breaths while checking for the tip of the spleen as it descends with deep inspiration. Splenomegaly is typically recorded as the number of centimeters the spleen descends below the left costal margin in the midclavicular line on deep inspiration. The physical examination in the patient with splenomegaly should be meticulous, including attention to the presence of lymphadenopathy, jaundice, tonsillar enlargement, joint effusions or swelling, and hepatomegaly.

Laboratory Evaluation

Laboratory studies, especially the complete blood count and peripheral blood smear, are important tools in the investigation of the patient with splenomegaly. An enlarged spleen is frequently hyperfunctional (“hypersplenism”) sequestering red cells and platelets, leading to cytopenias. Leukopenia may also occur. In the presence of hypersplenism, the bone marrow is hypercellular, and there may be nucleated red cells and an increased number of reticulocytes in the peripheral blood. Examination of the peripheral smear (Chapter 148) in the patient with splenomegaly may show red cell abnormalities characteristic of chronic liver disease. As noted earlier, the presence of Howell-Jolly bodies in red cells is a marker of a hypofunctioning or absent spleen. The presence of abnormal lymphocytes (e.g., cells with projections as in hairy cell leukemia or with villous projections in splenic marginal zone lymphoma with villous lymphocytes) suggests the presence of a malignant leukemia or lymphoma. The presence of lymphoblasts points to acute lymphoblastic leukemia. Flow cytometric analysis is the appropriate next step in identifying the abnormal cell population in many cases. A bone marrow aspirate and biopsy with immunohistochemistry, special stains, flow cytometric analysis, fluorescent in situ hybridization, cytogenetics, and additional molecular testing for particular mutations will be required to evaluate fully the patient with splenomegaly and a hematologic malignancy. Elevated lactate dehydrogenase and uric acid levels are seen in lymphoid malignancies such as Burkitt lymphoma or other aggressive non-Hodgkin lymphomas, but also in myeloid malignancies. Examination of thick smears in patients suspected of malaria may show the offending parasite. Today, genetic testing may confirm the suspected diagnosis (e.g., in children and in adults with suspected hemophagocytic lymphohistiocytosis).

Imaging

Not all palpable spleens are abnormal, especially in very thin individuals. The least invasive and most inexpensive approach to imaging the spleen is ultrasound, an approach that allows for sequential and accurate measurements. By ultrasound criteria, a spleen that is 13 cm or greater in length and more than 5 cm in “thickness” qualifies as enlarged. Today, splenomegaly is most often identified incidentally in the course of CT imaging for an often unrelated process (e.g., diverticulitis). Chest CT scans often capture the upper abdomen so that individuals scanned for chest pain or respiratory complaints may be found to have unrelated splenomegaly. In comparison with ultrasound, CT scans generally focus on the entire abdomen, not just the spleen, and provide important information that often provides an explanation for the splenomegaly. CT scans also will show whether the spleen is diffusely enlarged or enlarged because of multiple nodules. Whereas fluorodeoxyglucose-PET does not distinguish among a variety of malignancies, infections, and other causes of splenomegaly, it is generally used later to evaluate patients with lymphoid malignancies to assess response and not in the diagnostic evaluation of the patient with splenomegaly. The technetium-labeled liver-spleen scan has been replaced by CT, although it does have the potential to identify patients with liver disease as the cause of splenomegaly.

1107

TABLE 159-6 RECOMMENDATIONS FOR PNEUMOCOCCAL, MENINGOCOCCAL AND HAEMOPHILUS INFLUENZAE VACCINATION FOR ASPLENIC INDIVIDUALS The following vaccines are used: Meningococcus

Quadrivalent meningococcal conjugate vaccine (Menveo or Menactra) Meningococcal serogroup B vaccine (Trumenba or Bexsero)

Pneumococcus

23-valent pneumococcal vaccine (PPSV23, Pneumovax) 13-valent conjugate vaccine (PCV13, Prevnar)

Haemophilus influenzae

H. influenza type b (Hib) conjugate vaccine

GENERAL GUIDELINES • Adults with functional asplenia (e.g., those with sickle cell disease) or those who require surgical splenectomy should be immunized with pneumococcal, meningococcal, and H. influenza vaccines. • If undergoing elective splenectomy, meningococcal, pneumococcal, and H. influenzae vaccines should be given at least 14 days before surgery. If not possible, they should be given after no less than 14 days after splenectomy. • Some recommend that the same vaccine combination be given again 8 weeks after the initial doses. • Multiple vaccines can be given at the same visit but injected at separate sites. • There is some variation in the specific recommendations listed below between guidelines from the United States Centers for Disease Control (CDC), the United States Advisory Committee on Immunization Practices, various institutions, and some European agencies. MENINGOCOCCAL VACCINES • Quadrivalent meningococcal conjugate vaccines: two doses of Menveo or Menactra, given 2 months apart plus • Serogroup B meningococcal vaccines: either Trumenba (three doses at 0, 1-2, and 6 months apart) or Bexsero (two doses at least 1 month apart) • Boost every 5 years with Menveo or Menactra • The two serogroup B vaccines are not interchangeable; the same vaccine product must be used for all doses in a series. PNEUMOCOCCAL VACCINES • For before splenectomy: Pneumovax and Prevnar • For functional asplenia: Pneumovax and Prevnar given at least 8 weeks apart • Revaccination: At least 5 years after most recent Pneumovax dose to be followed at least 8 weeks later by Prevnar HAEMOPHILUS INFLUENZAE VACCINE • Most adults are not at risk because they have already acquired specific antibodies against it through routine childhood vaccination or previous infection. However, • One dose of Hib vaccine can be given before splenectomy • One dose of Hib vaccine can be given to functionally asplenic individuals

Most incidentally identified splenic masses in asymptomatic patients will prove to be benign. Follow-up imaging in short intervals (3 to 6 months) may establish a lesion to be unchanged and provide reassurance of its benign nature. Additional imaging modalities (e.g., MRI, PET/CT, and ultrasound) may be helpful in selecting appropriate patients for further investigation, including biopsy.

Biopsy

Because of the spleen’s location and propensity to bleed, needle aspiration or core biopsy of the spleen is performed only at centers skilled and experienced in the procedure. If an extensive evaluation has not provided a likely explanation for splenomegaly and observation is judged to be imprudent, splenectomy may be performed laparoscopically or by laparotomy.11,12 Because of the risk of surgical complications and long-term susceptibility to infection in splenectomized individuals, the relative risks and benefits of the procedure must be given serious consideration.13 Once splenectomy is a possibility, immunization should proceed promptly in an attempt to reduce long-term risks. Guidelines for vaccination before splenectomy and in asplenic individuals are presented in Table 159-6. Partial splenectomy is performed in some children with hereditary spherocytosis, splenic tumors, or splenic trauma in an attempt to reduce the risk of sepsis long term.14 GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 159  THE PATIENT WITH LYMPHADENOPATHY AND SPLENOMEGALY  

How to Approach the Patient with Splenomegaly

Similar to the patient with lymphadenopathy, the child or adult with splenomegaly may come to medical attention in many different ways. Whereas splenomegaly is often associated with lymphadenopathy, an enlarged spleen may be detected on physical examination or by imaging in the course of an evaluation for an enlarged lymph node. Pain in the left upper quadrant sometimes radiating to the left shoulder may be the first indication of a splenic infarct and an enlarged spleen. Rarely, the catastrophic signs and symptoms of splenic rupture are the first indications of splenomegaly. Unexplained weight loss related to early satiety caused by compression of the stomach by the enlarged spleen or unexplained cytopenias caused by hypersplenism may be the first indication of splenomegaly. Today, it is uncommon for splenomegaly to be detected first on physical examination given the obesity epidemic and the increasing reliance on testing rather than careful physical examination. In contrast to lymphadenopathy, an enlarged spleen discovered on physical examination is nearly always abnormal and requires explanation. Like lymphadenopathy, splenomegaly is often first detected on imaging studies performed for unrelated complaints. When borderline in size, the significance of the incidentally noted spleen, like marginally enlarged lymph nodes, is unclear and requires follow-up.

1107.e1

A careful history should focus on excluding a systemic illness that could explain the splenomegaly. Prior history or symptoms of infection, autoimmune disorders, or neoplasia should be reviewed. If a likely cause, such as infectious mononucleosis, is identified and confirmed by testing, appropriate management should follow. The associated splenomegaly should resolve over time. The patient without an obvious explanation for splenomegaly presents a difficult diagnostic challenge. Occult liver disease or an undiagnosed autoimmune process may emerge over time as an explanation for persistent splenomegaly in some of these patients. Although the otherwise asymptomatic patient may be monitored, the symptomatic patient requires a diagnosis so that appropriate treatment may be initiated. The presence of fevers, weight loss, and other symptoms requires persistence and consultation to exclude active infection or neoplasia. Bone marrow examination with culture and flow cytometric analysis may be helpful in identifying a cause in these patients. Splenectomy or biopsy of focal splenic lesions may be the only diagnostic option in patients with persistent symptoms or progressive cytopenias. Splenectomy may have unintended consequences in patients with occult liver disease and portal hypertension and should be avoided in that setting. In the asymptomatic patient, close observation is often preferable to splenectomy.

1107.e2

CHAPTER 159  THE PATIENT WITH LYMPHADENOPATHY AND SPLENOMEGALY  

GENERAL REFERENCES 1. Nin CS, de Souza VVS, do Amaral RH, et al. Thoracic lymphadenopathy in benign diseases: a state of the art review. Respir Med. 2016;112:10-17. 2. Indolfi P, Perrotta S, Rossi F, et al. Childhood head and neck lymphadenopathy: a report by a single institution (2003-2017). J Pediatr Hematol Oncol. 2019;41:17-20. 3. Gaddey HL, Riegel AM. Unexplained lymphadenopathy: evaluation and differential diagnosis. Am Fam Physician. 2016;94:896-903. 4. American College of Radiology. ACR Appropriateness Criteria: neck mass/adenopathy. https:// acsearch.acr.org/docs/69504/Narrative. Accessed March 12, 2019. 5. Ebell MH, Call M, Shinholser J, Gardner J. Does this patient have infectious mononucleosis? The rational clinical examination systematic review. JAMA. 2016;315:1502-1509. 6. Abou Youssef HA, Elzorkany MA, Hussein SA, et al. Evaluation of mediastinal lymphadenopathy by diffusion weighted MRI; correlation with histopathological results. Adv Respir Med. 2019;87:175-183. 7. Farndon S, Behjati S, Jonas N, Messahel B. How to use…lymph node biopsy in paediatrics. Arch Dis Child Educ Pract Ed. 2017;102:244-248.

8. Tzankov A, Dirnhofer S. A pattern-based approach to reactive lymphadenopathies. Semin Diagn Pathol. 2018;35:4-19. 9. Winter JN, Peterson LC. Lymphocytosis, lymphadenopathy: benign or malignant? Hematology Am Soc Hematol Educ Program. 2015;2015:106-110. 10. Curovic Rotbain E, Lund Hansen D, Schaffalitzky de Muckadell O, et al. Splenomegaly—diagnostic validity, work-up, and underlying causes. PLoS ONE. 2017;12:1-11. 11. Hangge PT, Sheaffer WW, Neville M, et al. The diagnostic value of splenectomy in patients without a definitive preoperative diagnosis. Am J Surg. 2019;217:78-82. 12. Misiakos EP, Bagias G, Liakakos T, Machairas A. Laparoscopic splenectomy: current concepts. World J Gastrointest Endosc. 2017;9:428-437. 13. Dionne B, Dehority W, Brett M, et al. The asplenic patient: post-insult immunocompetence, infection, and vaccination. Surg Infect (Larchmt). 2017;18:536-544. 14. Costi R, Castro Ruiz C, Romboli A, et al. Partial splenectomy: who, when and how. A systematic review of the 2130 published cases. J Pediatr Surg. 2018. [Epub ahead of print.]

1108

CHAPTER 160 Histiocytoses  

160  HISTIOCYTOSES BARRETT J. ROLLINS AND NANCY BERLINER

The histiocytoses comprise a family of diseases characterized by an accumulation of abnormal cells with features reminiscent of macrophages or dendritic cells. The term histiocyte refers to tissue macrophages, and although the cell of origin in some histiocytoses is indeed a macrophage, many of the other histiocytoses arise from nonmacrophage lineages or from hematopoietic stem cells. The diagnostic category of histiocytosis embraces a large number of heterogeneous but distinct clinical entities. The widely divergent nature of these diseases has produced a confusing literature that may not help a medical provider confronted with a patient who has one of these diseases. For example, a long-standing categorical division between Langerhans cell histiocytosis (LCH) and non-Langerhans cell histiocytoses (non-LCHs) has been undermined by a new appreciation for the frequent co-occurrence of these diseases in a single patient. This and other considerations have led to an updated understanding of the histiocytoses. In particular, recent advances in molecular genetics provide the basis for a revised nosology that places the histiocytic diseases into five groups (Table 160-1)1: • L group. These diseases include LCH and Erdheim-Chester disease (ECD) because the occasional appearance of both diseases in the same patient suggests a common origin and because both diseases are characterized by somatic mutations in genes of the mitogen-activated protein kinase (MAPK) pathway. Indeterminate cell histiocytosis (ICH) is also placed in the L group because it shares some characteristics of LCH. • C group. These are non-LCHs that involve the skin and/or mucosal surfaces. As in the L group, patients may present with mixtures of C group histiocytoses. They are subdivided into the xanthogranuloma (XG) and non-XG diseases. Some C group diseases may also have a major systemic component. • M group. These are the malignant histiocytoses, which include primary malignancies such as malignant histiocytosis (MH) and secondary malignant

TABLE 160-1 CLASSIFICATION OF HISTIOCYTOSES AND MACROPHAGE/DENDRITIC CELL NEOPLASMS L group

Langerhans cell histiocytosis (LCH) Indeterminate cell histiocytosis (ICH) Erdheim-Chester disease (ECD)

C group

Cutaneous non-LCH Xanthogranuloma (XG) family Juvenile xanthogranuloma ( JXG) Adult xanthogranuloma (AXG) Solitary reticulohistiocytoma (SRH) Benign cephalic histiocytosis (BCH) Generalized disruptive histiocytosis (GEH) Progressive nodular histiocytosis (PNH) Xanthoma disseminatum (XD) Non-xanthogranuloma (non-XG) family Cutaneous Rosai-Dorfman disease (RDD) Necrobiotic xanthogranuloma (NXG) Multicentric reticulohistiocytosis (MRH) Cutaneous non-LCH histiocytosis with major systemic involvement

R group

Familial Rosai-Dorfman Disease (RDD), also known as sinus histiocytosis with massive lymphadenopathy (SHML) Sporadic RDD Classical RDD Extranodal RDD RDD with neoplasia or immune disease Not otherwise classified

M group

Primary malignant histiocytoses Secondary malignant histiocytoses (associated with hematologic neoplasia) Subtypes: histiocytic, interdigitating, Langerhans, indeterminate

H group

Primary hemophagocytic lymphohistiocytosis (HLH) Secondary HLH HLH of unknown or uncertain origin

histiocytoses that occur in the setting of leukemias, lymphomas, or other hematologic neoplasms. • R group. The paradigmatic member of this group is Rosai-Dorfman disease (RDD), also known as sinus histiocytosis with massive lymphadenopathy (SHML). However, the R group also includes all of the other miscellaneous noncutaneous, non-LCH histiocytoses. • H group. This group contains the varieties of hemophagocytic lymphohistiocytosis (HLH) and macrophage activation syndromes (MASs).

  L GROUP

Langerhans Cell Histiocytosis  

EPIDEMIOLOGY

LCH occurs predominantly in children (median age, 3.8 years) but is also seen in adults.2 Its overall incidence is 2.6 cases per million child years for all children younger than 14 years of age but with a strong trend toward a higher incidence in younger children: 9.0 cases per million child years for infants less than 1 year old, and 0.7 cases per million child years for children 10 to 14 years old. There is a slight excess of cases in male patients (1.2:1). The incidence of high-risk LCH (defined later) is lower in African Americans than it is in whites and higher in Hispanics than it is in non-Hispanics. The incidence in adults has been estimated at 1 to 2 cases per million. Pulmonary LCH, which occurs primarily in adults, is strongly associated with a smoking history.  

PATHOBIOLOGY

The cardinal feature of LCH is an accumulation of abnormal histiocytes having abundant pale cytoplasm and a single kidney- or coffee bean–shaped nucleus (Fig. 160-1A). Pathologic LCH cells all express the surface marker CD1a (Fig. 160-1B), and most also express CD207, a mannose-binding lectin also known as Langerin. A majority of cells contain intracellular organelles called Birbeck granules, which can be detected by electron microscopy and are likely composed of CD207 molecules. LCH lesions are commonly accompanied by a marked inflammatory infiltrate. The quality of the infiltrate may vary depending on the clinical subtype (e.g., abundant eosinophils in eosinophilic granuloma of bone) (see Clinical Manifestations). LCH is a neoplastic disease: LCH cells are clonal, and most patients carry somatic activating mutations in genes that encode members of the MAPK pathway.3 Specifically, 50 to 60% have mutated BRAF (predominantly the mutation that encodes the oncogenic variant BRAF V600E), and 25 to 30% have activating mutations of MAP2K1. All cases show evidence for activation of extracellular signal-regulated kinase (ERK), the target of this kinase pathway, and the remaining 10 to 25% of cases without BRAF or MAP2K1 mutations have mutations in other genes that activate the pathway, including ARAF and RAS family members, or translocations resulting in fusion proteins that activate BRAF.4 Historically, LCH was acknowledged to be a disease of histiocytes, but because the origin was unknown, the disease was called histiocytosis X. The discovery that LCH cells express CD1a and CD207 and have Birbeck granules, all of which are characteristic of normal Langerhans cells, led to the suggestion that LCH cells are derived from Langerhans cells. However, the profile of genes expressed by LCH cells is closer to that of myeloid precursor cells than mature Langerhans cells. In addition, BRAF mutations occur in hematopoietic stem cells of patients whose LCH cells have BRAF mutations, a finding suggesting that LCH is more properly viewed as a myeloid neoplasm in which myeloid precursors may partially differentiate in a direction that allows them to acquire attributes that are also characteristic of Langerhans cells.  

CLINICAL MANIFESTATIONS

LCH may occur at any age and involve any organ. However, the disease is more common in children, and the organs most frequently affected include the following: • Bone (80% of cases). Skeletal lesions may be asymptomatic or may be associated with tenderness and swelling. The most common locations (in descending order of frequency) are skull, pelvis, femur, ribs, humerus, mandible, and spine. Radiographically these are punched-out lytic lesions without sclerosis (Fig. 160-1C). Resorption of alveolar bone can lead to the appearance of so-called “floating teeth.” • Skin (33% of cases). The rash may be intensely petechial and include scaly and encrusted papules (Fig. 160-1D and E). When skin involvement is the presenting manifestation, the LCH rash may be confused with seborrheic dermatitis and “cradle cap” in infants.

CHAPTER 160 Histiocytoses  

ABSTRACT

The histiocytoses are characterized by the accumulation in various organs of abnormal cells having features reminiscent of macrophages or dendritic cells. These diseases are heterogeneous in their clinical behaviors and outcomes, but recent advances in molecular genetics have led to a new nosology that divides histiocytic diseases into five categories: L Group: Langerhans cell histiocytosis (LCH), Erdheim-Chester disease (ECD), and indeterminate cell histiocytosis; C Group: non-LCH histiocytoses involving skin or mucosa and comprising both xanthogranuloma and non-xanthogranuloma subtypes; M Group: primary and secondary malignant histiocytoses; R Group: RosaiDorfman disease and other noncutaneous, non-LCH histiocytoses; H Group: hemophagocytic lymphohistiocytosis (HLH) and macrophage activation syndrome (MAS). The discovery of constitutive activation of the mitogenactivated protein kinase (MAPK) pathway, due largely to mutations in BRAF and MAP2K1, in LCH and ECD has led to case reports describing major clinical responses to BRAF and MEK1 inhibitors. Targeted therapies based on gene mutations in other histiocytoses have also been effective. However, the role of these interventions in the standard treatment of these diseases awaits the results of ongoing clinical trials. In its familial form, HLH is associated with null germline mutations in genes encoding immune proteins such as perforin. Acquired HLH, triggered by malignancy, infection, or autoimmune disease, may be associated with hypomorphic alleles of these genes. Both forms are treated by immune suppression; hematopoietic stem cell transplantation may also be considered.

KEYWORDS

histiocytosis dendritic cell mitogen-activated protein kinase BRAF perforin

1108.e1

CHAPTER 160 Histiocytoses  

A

B

C

D

1109

G

E

F

FIGURE 160-1.  Histologic, radiographic, and cutaneous features of Langerhans cell histiocytosis (LCH) and Erdheim-Chester disease (ECD). A, Hematoxylin and eosin stain of an LCH bone lesion demonstrating the characteristic kidney- or coffee bean–shaped nuclei and abundant pink cytoplasm of LCH histiocytes (1000× magnification). B, CD1a immunostain of an LCH skin lesion showing uniformly positively staining LCH histiocytes infiltrating the upper dermis and focally involving the epidermis (200× magnification). C, Radiographic image showing lytic skull bone lesions in LCH. D, Cutaneous lesions in a child with multisystem LCH. E, Intertriginous lesion in an adult with cutaneous LCH. F, Periorbital xanthelasma in a patient with ECD. G, Fluorine-18–radiolabeled fluorodeoxyglucose positron emission tomographic (PET) scan of a patient with ECD showing bilateral and symmetrical abnormal signals in femora, tibiae, and humeri. (A and B, From Degar, BA, Fleming, MD, Rollins, BJ, Rodriguez-Galindo C. Histiocytoses. In: Orkin SH, Fisher, DE, Ginsburg D, et al. eds. Nathan and Oski’s Hematology and Oncology of Infancy and Childhood, 8th ed. Philadelphia: Elsevier Saunders; 2015:2104; C to G, From Emile J-F, Abla O, Fraitag S, et al. Revised classification of histiocytoses and neoplasms of the macrophage-dendritic cell lineages. Blood. 2016;127:2672-2781.)

• Hypothalamus and pituitary stalk (25% of cases). Pituitary involvement can lead to diabetes insipidus, which was initially observed to occur in 25 to 50% of patients with LCH. With earlier diagnosis and effective chemotherapy, the incidence of diabetes insipidus has been reduced to 7 to 20%. Involvement of the liver, spleen, bone marrow, or lungs occurs in 15% of cases; involvement of lymph nodes in 5 to 10%, and central nervous system (CNS; excluding pituitary) in 2 to 4%. Lung involvement is more frequent in adults than in children. The clinical manifestations of LCH are extraordinarily diverse. However, some presentations are sufficiently stereotypic that they had been recognized as separate entities before the understanding that they are manifestations of the same disease. The eponyms for these entities are no longer in use but are listed here to aid in understanding older literature: • Hand-Schüller-Christian disease: triad of exophthalmos, diabetes insipidus, and bone lesions (especially cranial) in children; 15 to 40% of LCH cases • Letterer-Siwe disease (or Abt-Letterer-Siwe disease): multisystem disease involving skin, bone, lymph nodes, liver, and spleen in infants; poor prognosis • Eosinophilic granuloma: bone only, often solitary; 60 to 80% of LCH cases • Hashimoto-Pritzker disease (or congenital self-healing reticulohistiocytosis): self-limited skin involvement in newborns Contemporary classification of LCH is based on clinical attributes that correlate with risk and therefore guide therapy. The two dimensions of risk

are the number of organs involved (single system vs. multisystem disease) and the presence of disease in “risk organs” (defined as bone marrow, liver, or spleen). The Histiocyte Society identified three clinical groups using these characteristics for its international LCH-III trial (Table 160-2). A late manifestation of LCH is neurodegeneration, which occurs in about 10% of surviving patients. Manifestations include ataxia, dysarthria, and dysmetria, along with neurocognitive decline and psychological difficulties. Magnetic resonance imaging (MRI) shows signal intensity changes in the cerebellar peduncles, basal ganglia, and pons. The clinical manifestations may wax and wane, but the disease is relentlessly progressive.  

DIAGNOSIS

Because the clinical manifestations of LCH are nonspecific, one must have a high index of suspicion for LCH in adult patients who present with a nonhealing seborrheic rash or solitary osteolytic lesion.5 In patients with diabetes insipidus, characteristic infiltration of the pituitary stalk may be seen on MRI. Pulmonary LCH should be considered in young smokers with dyspnea or a nonproductive cough and whose plain chest radiograph shows diffuse bilateral reticulonodular disease in the middle and upper lung zones. Computed tomography (CT) may show nodules or cysts. Tissue analysis is mandatory. Histopathologic examination shows an infiltrate of histiocytic cells with pale cytoplasm and a single kidney- or coffee

1110

CHAPTER 160 Histiocytoses  

TABLE 160-2 CLINICAL CLASSIFICATION OF LANGERHANS CELL HISTIOCYTOSIS* CLINICAL GROUP

INVOLVED SYSTEM

INVOLVED ORGANS

1

Multisystem

Any risk organ†

2

Multisystem

≥2 organs not including risk

3

Single system • Multifocal‡ or special site§

≥2 lesions in one organ or



Single system • Unifocal or localized

1 lesion in 1 organ

organs

in a special site§

*Developed for the Langerhans Cell Histiocytosis-III (LCH-III) trial. A1  † Risk organs: bone marrow, liver, or spleen. (Lung is no longer classified as a risk organ.) ‡ Multiple sites of involvement in a single organ system. § Special sites confer increased risk of specific complications and include intracranial soft tissue extension or vertebral lesions with intraspinal soft tissue extension.

bean–shaped nucleus accompanied by varying numbers and types of inflammatory cells (see Fig. 160-1A). Eosinophils may be abundant. Definitive diagnosis requires a demonstration that the histiocytes stain for CD1a (see Fig. 160-1B). Positive staining for CD207 may further support the diagnosis of LCH. Electron microscopic examination for Birbeck granules has been supplanted by CD207 staining.

TREATMENT 

show remarkable responses to inhibitors of RAF, such as vemurafenib and dabrafenib, or MEK1, such as trametinib or cobimetinib. Optimal dosing has not yet been explored in a clinical trial, and treatment has generally been devised based on efficacy in other diseases. Most reports indicate that patients continue to respond to these inhibitors for as long as they are treated and have relapses when treatment is discontinued. Therefore, despite the remarkable efficacy of these agents, the roles that these inhibitors may play in curative or first-line treatment of LCH are currently unclear. For adults with pulmonary LCH, smoking cessation is usually sufficient treatment. If not, these patients may be treated with corticosteroids. Patients with refractory cases may require lung transplantation.



Erdheim-Chester Disease  

Treatment strategies for LCH are evolving rapidly based on new molecular characterization of the disease, but historically, therapy and its intensity have been matched to risk stratification. For example, patients with single-system disease without organ dysfunction have an excellent prognosis and can be treated with local therapy (e.g., surgical curettage or radiation) or simply monitored and treated when symptomatic. More extensive disease requires systemic therapy, which has been refined through a series of international trials. The first such trial, LCH-I, tested the efficacy of adding etoposide to the historical standard of care, which was vinblastine and prednisone. Patients with multisystem LCH were randomized to receive weekly vinblastine or etoposide every 3 weeks for 24 weeks. Overall survival was excellent at 80%, and there were no differences between treatment arms of the trial. LCH-II tested the effects of early intensification by randomizing patients with multisystem LCH to receive the following: vinblastine and prednisone, followed by continuation therapy with 6-mercaptopurine and pulses of vinblastine/ prednisone for 24 weeks; or addition of etoposide to the induction therapy. Survival and relapse rates were the same in both arms. However, the subset of patients with risk organ involvement (see Table 160-2) had better survival with the addition of etoposide. LCH-III carefully stratified patients by risk and asked whether the addition of methotrexate would further improve outcomes in patients with risk organ involvement. In addition, because an earlier nonrandomized trial (DAL-HX83) suggested that prolonged treatment could be beneficial, LCH-III also tested in a randomized fashion whether prolonging treatment in patients with multisystem disease would reduce the risk of relapse. A1  Although the addition of methotrexate had no effect on outcomes, prolonging therapy in patients with multisystem disease decreased the risk of relapse from 54 to 37%. Thus current standard of care for multisystem disease is vinblastine and prednisone induction followed by 12 months of continued therapy. Various approaches are available for relapsed or recurrent LCH, referred to historically as “reactivation,” which occurs in 20 to 50% of patients. These patients can be placed into low-risk and high-risk categories. Low-risk patients are those who initially had low-risk multisystem LCH (i.e., no risk organ involvement or multifocal bone disease). Responses to a variety of second-line therapies have been documented, including 6-mercaptopurine and methotrexate, indomethacin, bisphosphonates, and nucleoside analogues such as cladribine, cytarabine, and clofarabine. High-risk patients are those who had risk organ involvement at presentation or showed a poor response to first-line treatment. Cladribine and high-dose cytarabine may be effective in this setting, as is hematopoietic stem cell transplantation, but mortality is still significant. The discovery of activating mutations in genes encoding MAPK pathway components has led to the use of rapidly advancing fibrosarcoma (RAF) and MAP kinase/ERK kinase 1 (MEK1) inhibitors primarily in the relapsed setting. Most reports describe small series of treated patients, although one so-called basket trial using the RAF inhibitor vemurafenib in all patients with BRAF V600E included a small number of patients with LCH.6 In general, patients carrying sensitive mutations in BRAF or MAP2K1, including heavily pretreated patients,

PROGNOSIS

Outcomes in single-system and low-risk disease are quite good. Overall survival for single system disease is more than 99%, with a relapse rate of 18%. In the LCH-III trial, 5-year overall survival in low-risk multisystem disease was 99% despite a 40 to 50% cumulative rate of relapse. Nonetheless, morbidities associated with the initial site of disease, especially orthopedic problems, may be long-standing and significant. Patients with multisystem high-risk disease had an 84% 5-year overall survival with a 27% relapse rate in LCH-III. Failure to respond to initial therapy carries a dismal prognosis: children younger than 2 years of age with risk organ involvement who do not respond to initial therapy have a 3-year survival of only 17%. Neurodegeneration may appear in up to 10% of patients several years after otherwise successful treatment and in the absence of active histiocytic disease.

EPIDEMIOLOGY

ECD is extremely rare. Fewer than 1000 cases have been reported, and its precise incidence and prevalence are unknown. In contrast to LCH, ECD occurs primarily in adults between 40 and 70 years of age (mean age at diagnosis of 53 years),7 although it may occur in children and infants. Male patients outnumber female patients by a ratio of nearly 3:1.  

PATHOBIOLOGY

Histopathologic examination of lesions shows infiltration by foamy mononuclear histiocytes often accompanied by fibrosis and inflammation involving lymphocytes, neutrophils, or plasma cells. Occasional multinucleated giant cells may be observed. Because the histiocytes stain for CD68 and CD163, and not for CD1a, ECD histiocytes are thought to derive from the monocytemacrophage lineage. Although these characteristics distinguish ECD from LCH, 20% of patients with ECD have concurrent LCH lesions. As in LCH, about 50% of patients with ECD have somatic mutations encoding BRAF V600E. Essentially all remaining patients also have genetic variants that activate the MAPK pathway, including mutations in MAP2K1, ARAF, KRAS, and NRAS, as well as gene fusions that activate BRAF, ALK, and NTRK.8 Mutations in PIK3CA have been observed and may co-occur with BRAF mutations.  

CLINICAL MANIFESTATIONS

Histiocytic infiltration of bone occurs in nearly all patients, manifesting as symmetrical cortical sclerosis of diaphyseal and metaphyseal areas of long bones, often in the lower extremities (Fig. 160-1G). Patients may complain of mild juxta-articular pain but may also be asymptomatic in the presence of radiographic disease. Osteosclerosis of facial bones or other bones of the skull may also occur. CT or MRI may be necessary to detect lesions not seen on plain radiographs. Cardiovascular involvement by fibrosis occurs in the majority of patients and may manifest as valvular abnormalities, conduction defects, or periaortic coating.9 This is one of the main causes of morbidity and mortality in ECD. Similar fibrosis may occur in the retroperitoneum, encircling the kidneys and leading to a progressive obstructive nephropathy. About 50% of patients with ECD have pulmonary disease involving the pleura, parenchyma, or both. About 40% of patients have CNS disease, including pituitary involvement leading to diabetes insipidus, as in LCH, and mass lesions in various parts of the brain. CNS involvement portends a worse outcome. Cognitive impairment and behavioral changes may also occur in the absence of MRI-documented infiltration or masses, although they may be associated with gray matter volume loss. Cutaneous disease occurs in about 25% of patients and usually manifests as xanthelasma, most frequently involving the eyelids (Fig. 160-1F).

CHAPTER 160 Histiocytoses  



DIAGNOSIS

Diagnosis must be made by histologic examination of involved sites, preferably osteosclerotic bone. Because genetic analysis for mutations in MAPK pathway genes may inform clinical care, and because decalcification for histologic examination may preclude such analysis, sufficient material must be collected and processed separately for both histologic and genetic testing. Absence of CD1a or CD207 staining will distinguish ECD from LCH. However, as noted earlier, 20% of patients with ECD have concurrent LCH. Clinical manifestations of ECD may be confused with those of Paget disease or POEMS (polyneuropathy, organomegaly, endocrinopathy, myeloma protein, and skin changes) syndrome (Chapter 178), but the histopathologic picture of these disorders is distinct. Juvenile xanthogranuloma ( JXG) is a histiocytic disease of the C group (see later), and its histopathologic features are very similar to those of ECD. However, JXG typically occurs in young children and only rarely involves extracutaneous sites. However, when JXG is disseminated, lesional cells may harbor activating mutations in genes of the MAPK pathway, which is reminiscent of ECD and places this form of JXG in the L group. When the tissue diagnosis of ECD is confirmed, the extent of disease should be determined. A full endocrine evaluation should be performed to assess pituitary involvement, and imaging studies should include long bones, skull, and pelvis, as well as CT or MRI evaluation of the brain, full length of the aorta, and retroperitoneum. Fluorine-18–radiolabeled fluorodeoxyglucose uptake demonstrated by positron emission tomography has been reported to provide high specificity in the setting of osteosclerosis (Fig. 160-1G). The presence of conduction abnormalities should be evaluated by electrocardiography. Macrophage-like cells that stain for S100 and show emperipolesis may be seen histologically and could suggest a diagnosis of Rosai-Dorfman disease (RDD), but the clinical presentation of ECD is distinct from that of RDD (see later).

TREATMENT  Because most cases of ECD are indolent, asymptomatic patients who have no CNS disease may be followed without treatment. If patients have CNS disease or develop systemic symptoms, treatment with interferon (IFN) alfa (conventional or pegylated) is recommended because of the suggestion of a survival benefit in a retrospective analysis. For patients in whom IFN treatment fails and are symptomatic, cladribine or cyclophosphamide may produce responses. Glucocorticoids and interleukin-1 receptor (IL-1R) antagonists may also have activity. Radiotherapy may provide relief of local symptoms. The nearly universal presence of mutations in genes encoding components of the MAPK pathway suggests that therapies targeted against some of these activated variants could be effective. A basket trial of the RAF inhibitor vemurafenib in patients with a variety of diseases carrying a somatic mutation encoding BRAF V600E included several cases of ECD and LCH. Prolonged responses were seen in six of 18 patients. There are also reports of responses to the MEK1 inhibitors trametinib and cobimetinib.



PROGNOSIS

The 1- and 5-year overall survival rates of patients with ECD who were treated with IFN alfa are 96 and 68%, respectively, versus 86 and 17% in control patients in this small study. The impact of therapy targeted against BRAF V600E or MEK1 on survival is currently unknown.

Indeterminate Cell Histiocytosis ICH is an exceedingly rare disease that may manifest as a cutaneous eruption mimicking LCH, although some cases may involve lymph nodes only. It occurs in adults and may predominantly affect women, although no rigorous epidemiologic analyses have been performed. Histologic analysis of lesions shows a nonepidermotropic infiltration of histiocytes that may stain for CD1a, similar to LCH. In contrast to LCH, however, lesional cells do not express CD207. One report describes a case with a somatic mutation encoding BRAF V600E that may support the notion that ICH is related to LCH. However, another report collected four cases, three of which showed the same ETV3-NCOA2 gene fusion.10 Although the pathogenic relevance of this translocation is unknown, its recurrent appearance in ICH and its absence from LCH indicate that these are distinct entities.

1111

  C GROUP

Cutaneous and Mucosal non-LCH Histiocytoses These non-LCH histiocytoses are divided into those that are predominantly cutaneous versus those that have a significant systemic component. Each group is further divided into the xanthogranuloma (XG) and non-XG categories.

XG Family  

EPIDEMIOLOGY

This group comprises several diseases that are distinguished primarily on the basis of clinical presentation: patient age, anatomic site of involvement, and whether lesions are solitary, multiple, or disseminated. Juvenile XG ( JXG) is the most commonly occurring non-LCH histiocytosis. Although its published prevalence is 0.5% of pediatric tumors, the benign nature of JXG likely leads to underreporting. Forty to 70% of cases occur in the first year of life, and JXG may be appear in neonates. A slight excess of cases in female patients has been observed (1.4:1).  

PATHOBIOLOGY

Lesions contain a dermal infiltrate that includes histiocytes, occasional multinucleated giant cells which may have the features of Touton cells, and nonspecific inflammatory cells. Regressing lesions show replacement by fibrous tissue. Histiocytes stain for CD68, CD14, CD163, factor XIIIa, and fascin; S100 staining may be absent or variable, and the cells do not stain for CD1a or CD207.  

CLINICAL MANIFESTATIONS

JXG is a benign disease in which a variable number of small (500 ng/mL Soluble CD25/sIL2r >2400 U/mL Hemophagocytosis in bone marrow, spleen, or lymph nodes Low or absent NK cell activity HLH = hemophagocytic lymphohistiocytosis; NK = natural killer; sIL2r = soluble interleukin-2 receptor.

in adults. Other diagnostic algorithms such as the H-score have been designed with the intent of improving diagnostic accuracy, especially in adults, but they probably add little to the ability to diagnose HLH in adult patients, except perhaps at initial presentation. Diagnosis of HLH requires demonstration of homozygous or compound heterozygous null mutations in familial HLH genes (seen in children) or fulfillment of five of eight criteria (see Table 160-3): fever, cytopenias, hypofibrinogenemia or hypertriglyceridemia, elevated ferritin, elevated soluble IL-2 receptor (sIL2r or sCD25), splenomegaly, hemophagocytosis in the bone marrow, and impaired NK cell function. Because NK cell function is highly variable outside of f HLH, it is generally not a helpful test in adult patients; given that NK cell function is unreliable and rarely tested, diagnosis in adults is therefore usually made by satisfying five of the other seven criteria in Table 160-3. Fever and cytopenias are universal findings in patients presenting with HLH. These signs are reflective of the exuberant secretion of inflammatory cytokines released by activated T and NK cells, including IFN-γ, tumor necrosis factor-α, IL-2, IL-4, IL-6, and IL-10. Splenomegaly has been reported to occur in about half of adults with HLH, and its presence may reflect the duration of the illness before diagnosis. Hypertriglyceridemia and/or hypofibrinogenemia occur in at least 50% of adults with HLH. Elevated liver function test results, although not a criterion for the diagnosis, are seen in almost all patients. Hyperferritinemia has often been considered the hallmark of HLH and MAS. In children, a serum ferritin level higher than 10,000 mg/dL has been demonstrated to be more than 90% sensitive and specific for HLH. However, in adults, ferritin is less powerful for establishing the diagnosis. Survey of a database of patients for elevated ferritin revealed that even with ferritin levels higher than 50,000 mg/dL, fewer than 20% of patients will have HLH; conversely, among adults diagnosed with HLH, the ferritin level ranges from 600 mg/dL to more than 50,000 mg/dL. Therefore, the negative predictive value of a normal ferritin result is high, but it is not a discerning marker for HLH in adults. Because hyperferritinemia is a nonspecific consequence of macrophage activation, it is also elevated in other chronic inflammatory conditions, hemolytic anemia, and renal failure, all of which are more common in adults than in children.17 The soluble form of the IL-2 receptor (sIL2r, sCD25) is also elevated in HLH. The IL-2 receptor is widely expressed by lymphocytes, monocytes, eosinophils, and NK cells. It is cleaved by macrophage gelatinase (MMP9) to release the soluble form of the receptor. It is a marker of immune activation in many inflammatory disorders. Although it is elevated in nearly all, if not all, patients with HLH, it is not a readily available test. However, it can be an extremely valuable diagnostic marker. In particular, in adults, it is usually extremely elevated in lymphoma-associated HLH because the malignant lymphoma cells themselves express high levels of IL2r. The ratio of sIL2r to ferritin can help distinguish lymphoma-associated HLH from other forms of HLH in adults because the ratio is usually less than 1 in the absence of malignancy and greater than 2 in the setting of lymphoma-associated HLH. Hemophagocytosis is often a prominent feature of HLH, but it is neither sensitive nor specific for the diagnosis. Even in children with f HLH, it may be difficult to detect hemophagocytosis in the marrow without multiple samplings. In adults, it appears to be somewhat more prominent. However, it must be remembered that many other diagnoses are associated

with hemophagocytosis. Many acute viral infections are associated with hemophagocytosis, and hemolytic anemias are associated with the same finding. The first critical step in the diagnosis of HLH is to consider the diagnosis because until recently it was rarely entertained in adult patients. In considering the diagnosis, a thorough history should include details regarding recent infections, history of autoimmune disease, underlying malignancy, and underlying immune disorders. Testing should include complete blood count, full metabolic panel, triglycerides, fibrinogen, ferritin, and sIL2r where available. Thorough evaluation for bacterial, parasitic, mycobacterial, and fungal infection should be performed, as well as evaluation for viral infection to include both serology and viral titers for EBV, cytomegalovirus, human immunodeficiency virus, influenza, hepatitis B and C, herpes simplex virus, varicella-zoster virus, parvovirus, adenovirus, and human herpesvirus-8. Patients should be evaluated for evidence of systemic autoimmune disease. Importantly, because lymphoma is a trigger of HLH in about half of adult patients, every effort should be made to find evidence for lymphoma, including performance of flow cytometry and positron emission tomography scan. Patients should undergo marrow examination, and in addition to routine studies, marrow samples should be sent for flow cytometry and T-cell gene rearrangement studies. When possible, full evaluation for lymphoma should be performed before initiation of therapy because HLH therapy is likely to mask accompanying lymphoma and impede timely diagnosis. Evaluation for genetic mutations in familial HLH genes may be helpful in later management, but it is not urgent for the initial evaluation.

TREATMENT  Therapy of HLH is aimed at suppressing the abnormal immune response, usually through a combination of immunosuppressive and targeted anti–T-cell therapy (Fig. 160-2). Because there have been no large trials of therapy for adult HLH, most adults are treated with the HLH-94 protocol, which was studied in a large prospective trial for pediatric HLH (Table 160-4). This therapy entails dexamethasone and etoposide with an 8-week induction course. Children with neurologic symptoms receive intrathecal methotrexate, but this is rarely necessary in adults. Following completion of the 8-week course, patients can be put on maintenance therapy with the addition of calcineurin inhibitors to intermittent dexamethasone and etoposide, usually as a bridge to stem cell transplantation. The HLH-94 was modified with the HLH-2004 protocol. Both protocols are shown in Table 160-4. Patients with significant liver dysfunction can have initial dose reduction of etoposide with dose increases as function improves. Renal failure may also require dose reduction. Etoposide is bound to albumin, and patients with HLH frequently present with very low serum albumin. This reduced protein binding may increase the potency of etoposide, but dosing is not usually adjusted. Treatment should not usually be held for cytopenias because these often reflect disease activity rather than drug toxicity. Therapy is tailored to the underlying trigger for the syndrome. In patients with malignancy-associated HLH, therapy for the malignancy should be initiated immediately. Because most malignancy-associated HLH accompanies a lymphoma, therapy with an etoposide-containing regimen is recommended, and steroid therapy should be continued until the HLH has resolved. EBV-related HLH is driven by EBV infection of T lymphocytes, so treatment with the HLH-94 protocol should be administered without delay. In addition, rituximab is recommended to clear the B-cell reservoir of virus. Patients with no clear precipitating trigger (“idiopathic HLH”) should receive the HLH-94 protocol with continued evaluation for underlying undiagnosed lymphoma or infection. The one exception to the recommendation for cytotoxic therapy for adult HLH is for patients with MAS. This subtype of HLH is often difficult to distinguish from an acute flare of the patient’s underlying systemic autoimmune disease. MAS usually responds to intensification of immunosuppressive therapy for the primary disease and does not usually require cytotoxic therapy. Patients who fail to respond to the HLH-94 protocol may receive salvage therapy with alemtuzumab, a monoclonal anti-CD52 antibody that destroys mature T cells. This should be viewed as a bridge to stem cell transplant in eligible patients. In this setting, alemtuzumab has also been demonstrated to reduce the incidence of graft versus host disease (GVHD) after stem cell transplantation and is now incorporated into the conditioning regimen before stem cell transplantation for HLH. The role of hematopoietic stem cell transplantation (HSCT) (Chapter 168) in HLH is evolving. fHLH is uniformly fatal without HSCT, but the necessity for other patients with HLH to undergo HSCT is less clear. In many children with infection-associated HLH, HLH resolves completely with HLH-94 therapy. The

CHAPTER 160 Histiocytoses  

1115

History Fever x7 days, immune disorder, autoimmune disease, malignancy, IBD Physical Fever, splenomegaly, lymphadenopathy Laboratory studies CBC, BMP, LFTs, albumin, ferritin, fibrinogen, triglycerides, sCD25, LDH, viral studies, cultures Imaging and pathologic studies Bone marrow aspirate, biopsy, flow cytometry, TCR studies, CAP CT, PET scan

Rheumatologic disorder

Malignancy

Immunosuppression and disease-specific therapy

Chemotherapy including etoposide as appropriate

Infection Antimicrobials as appropriate; HLH-94 protocol. Add rituximab for EBV

Idiopathic HLH-94 protocol; no IT therapy

Relapsed/refractory disease Allogeneic stem cell transplant FIGURE 160-2.  Approach to diagnosis and treatment of adult hemophagocytic lymphohistiocytosis (HLH). BMP = basic metabolic panel; CAP CT = chest, abdomen, pelvis computed tomography; CBC = complete blood count; EBV = Epstein-Barr virus; HLH-94 protocol (see text and grade A reference A2); IBD = inflammatory bowel disease; IT = immunotherapy; LDH = lactate dehydrogenase; LFTs = liver function tests; PET = positron emission tomography; sCD25 = soluble CD25; TCR = T-cell receptor.

TABLE 160-4 HLH-94 AND HLH-2004 PROTOCOLS FOR TREATMENT OF HEMOPHAGOCYTIC LYMPHOHISTIOCYTOSIS (HLH) HLH-94 PROTOCOL Initial Therapy for the First 8 Weeks: Dexamethasone at 10 mg/m2 per day weeks 1-2; 5 mg/m2 per day weeks 3-4; 2.5 mg/m2 per day weeks 5-6; 1.25 mg/m2 per day week 7; taper during week 8 Etoposide at 150 mg/m2 twice weekly for weeks 1-2; then once weekly Intrathecal MTX (for children with neurologic symptoms only) at 12 mg per dose, given at weeks 3 and 4, with two additional doses at weeks 5 and 6 if progressive neurologic symptoms and/or abnormal CSF not improved after 2 weeks Continuation Therapy/HSCT: Dexamethasone at 10 mg/m2 per day for 3 days, given every 2nd week Cyclosporine A beginning at week 9, dosing aimed at trough level of 200 µg/L HLH-2004 PROTOCOL Initial Therapy for the First 8 Weeks: Same as HLH-94 Protocol, except (1) cyclosporine A beginning at day 1 and continuing into the Continuation phase. Recommend starting cyclosporine A at a daily dose of 6 mg/kg, with dosing aimed at trough level of 200  µg/L; and (2) adding prednisone to MTX for intrathecal injections (for children with neurologic symptoms only) Continuation Therapy/HSCT as soon as an Accepted Donor is Identified CSF = cerebrospinal fluid; MTX = methotrexate. Modified from Bergsten E, Horne A, Aricó M, et al: Confirmed efficacy of etoposide and dexamethasone in HLH treatment: long-term results of the cooperative HLH-2004 study. Blood. 2017;130:2728-2738.

outcomes in adults are less clear. However, adult patients with refractory or relapsed disease, and those who have had life-threatening disease at presentation, should be evaluated for transplantation. Patients must be in remission at the time of HSCT because patients with active HLH at the time of transplantation fare poorly and usually die of overwhelming GVHD. Outcomes are superior with reduced intensity conditioning. Successful HSCT is curative. Novel therapies that may have an increasing role in the treatment of HLH include ruxolitinib and emapalumab. Because most inflammatory cytokines signal through the JAK-STAT pathway, the Janus kinase 1/2 inhibitor ruxolitinib has been studied in murine models and in early clinical trials. Although it is not approved for use in the setting of HLH, it is a possible adjunct to therapy, especially in patients with intractable HLH symptoms. Emapalumab is an investigative antibody targeting IFN-γ that has been studied for the treatment of refractory HLH in children; preliminary results suggest that it is extremely effective. Trials of the agent in adult HLH should follow.

Grade A Reference A1. Gadner H, Minkov M, Grois N, et al. Therapy prolongation improves outcome in multisystem Langerhans cell histiocytosis. Blood. 2013;121:5006-5014.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 160 Histiocytoses  

GENERAL REFERENCES 1. Emile JF, Abla O, Fraitag S, et al. Revised classification of histiocytoses and neoplasms of the macrophage-dendritic cell lineages. Blood. 2016;127:2672-2681. 2. Kobayashi M, Tojo A. Langerhans cell histiocytosis in adults: advances in pathophysiology and treatment. Cancer Sci. 2018;109:3707-3713. 3. Rollins BJ. Genomic alterations in Langerhans cell histiocytosis. Hematol Oncol Clin North Am. 2015;29:839-851. 4. Chakraborty R, Burke TM, Hampton OA, et al. Alternative genetic mechanisms of BRAF activation in Langerhans cell histiocytosis. Blood. 2016;128:2533-2537. 5. Allen CE, Merad M, McClain KL. Langerhans-cell histiocytosis. N Engl J Med. 2018;379: 856-868. 6. Hyman DM, Puzanov I, Subbiah V, et al. Vemurafenib in multiple nonmelanoma cancers with BRAF V600 mutations. N Engl J Med. 2015;373:726-736. 7. Campochiaro C, Tomelleri A, Cavalli G, et al. Erdheim-Chester disease. Eur J Intern Med. 2015;26: 223-229. 8. Diamond EL, Durham BH, Haroche J, et al. Diverse and targetable kinase alterations drive histiocytic neoplasms. Cancer Discov. 2016;6:154-165.

1115.e1

9. Ghotra AS, Thompson K, Lopez-Mattei J, et al. Cardiovascular manifestations of Erdheim-Chester disease. Echocardiography. 2019;36:229-236. 10. Brown RA, Kwong BY, McCalmont TH, et al. ETV3-NCOA2 in indeterminate cell histiocytosis: clonal translocation supports sui generis. Blood. 2015;126:2344-2345. 11. Janssen D, Harms D. Juvenile xanthogranuloma in childhood and adolescence: a clinicopathologic study of 129 patients from the Kiel Pediatric Tumor Registry. Am J Surg Pathol. 2005;29:21-28. 12. Hilal T, DiCaudo DJ, Connolly SM, et al. Necrobiotic xanthogranuloma: a 30-year single-center experience. Ann Hematol. 2018;97:1471-1479. 13. Kumar B, Singh N, Rahnama-Moghadam S, et al. Multicentric reticulohistiocytosis: a multicenter case series and review of literature. J Clin Rheumatol. 2018;24:45-49. 14. Piris MA, Aguirregoicoa E, Montes-Moreno S, et al. Castleman disease and Rosai-Dorfman disease. Semin Diagn Pathol. 2018;35:44-53. 15. Campo M, Berliner N. Hemophagocytic lymphohistiocytosis in adults. Hematol Oncol Clin North Am. 2015;29:915-925. 16. Schram AM, Berliner N. How I treat hemophagocytic lymphohistiocytosis in the adult patient. Blood. 2015;125:2908-2914. 17. Schram AM, Campigotto F, Mullally A, et al. Marked hyperferritinemia does not predict for HLH in the adult population. Blood. 2015;125:1548-1552.

1115.e2

CHAPTER 160 Histiocytoses  

REVIEW QUESTIONS 1. A 45-year-old man with a 25 pack-year smoking history presents to the emergency room with chest pain and mild dyspnea. His chest radiograph shows a small pneumothorax along with reticulonodular opacities in the middle to upper lung zones. Pulmonary function tests show a restrictive pattern. Bronchoalveolar lavage shows 20% CD1a-positive cells. What is the first intervention you should make? A . Transbronchial biopsy to confirm the diagnosis of pulmonary LCH B. Systemic steroid therapy to improve lung function C. Smoking cessation D. PET scan to establish baseline disease activity E. Test for circulating DNA encoding BRAF V600E Answer: C  Smoking cessation. The finding of greater than 5% CD1a-positive cells on bronchial lavage in a middle-aged smoker with this chest radiographic finding establishes the diagnosis of pulmonary LCH. Smoking cessation is a mandatory first step and may lead to disease resolution. 2. A 76-year-old man complains of abdominal pain. Work-up includes a CT scan that showed gallstones but also revealed circumferential periaortic fibrosis and a similar fibrotic accumulation around both kidneys. His general physical examination is unremarkable except for yellowish plaques in his periorbital skin. You consider the diagnosis of Erdheim-Chester disease and confirm it by histopathologic examination of a skin biopsy. Molecular assessment of the biopsy reveals the presence of the mutation encoding BRAF V600E. Brain MRI shows no involvement, and ECG is unremarkable. PET scan shows increased activity in the diaphyses of his long bones. Appropriate treatment for this patient is the following: A . Vemurafenib 960 mg bid B. Dabrafenib 100 mg bid plus trametinib 3 mg qd C. Pegylated interferon-α starting at 135 µg per week D. Weekly vinblastine 6 mg/m2 plus daily prednisone 40 mg/m2 for 6 weeks, followed by the same doses of vinblastine on day 1 and prednisone on days 1 through 5 every 21 days for 46 weeks E. Observation alone Answer: E  Observation alone. Asymptomatic Erdheim-Chester disease without evidence for CNS involvement requires no treatment. Patients should be followed and treatment instituted for symptom management or if CNS disease appears. 3. Rosai-Dorfman disease is characterized by all of the following except: A . Prominent lymphadenopathy often involving cervical lymph nodes B. Fever and leukocytosis C. Diabetes insipidus D. Accumulation of IgG4-positive plasma cells along with histiocytes in extranodal lesions E. A familial form associated with “H syndrome” involving mutation of SLC29A3

Answer: C  Diabetes insipidus. Diabetes insipidus due to histiocytic involvement of the pituitary may occur in up to 20% of patients with Langerhans cell histiocytosis and some patients with Erdheim-Chester disease, but it is not a complication of Rosai-Dorfman disease. 4. A 58-year-old woman presents with pain in multiple joints and a new rash consisting of nonpruritic yellow-to-brown papules and nodules on her face, ears, and forearms. She also has some periungual papules. Several joints are red and warm to the touch. She tells you that she has lost about 10 pounds, unintentionally, over the past 4 months. In addition to histologic and immunohistologic analysis of a skin biopsy, the most useful diagnostic strategy will be: A . Serum protein electrophoresis B. Assessment of mutational status of BRAF, and MAP2K1 in skin biopsy C. MRI of the abdomen, chest, and brain to look for malignancy D. Aspiration of fluid from inflamed joints to identify cellular composition E. PET scan Answer: C  MRI of the abdomen, chest, and brain to look for malignancy. The skin lesions, especially the periungual “coral bead” papules, in the setting of inflammatory arthritis are characteristic of multicentric reticulohistiocytosis (MRH). Approximately 25% of patients with this disease, the majority of whom are women, have an underlying malignancy. Treatment of MRH is directed toward the malignancy, and identifying the underlying disease provides the best chance for effective intervention. 5. A 43-year-old man with no past medical history presents to the hospital with a 2-month history of fevers and night sweats. Repeated cultures by his PCP have been negative, and despite three admissions to the hospital, his illness has evaded diagnosis. He has lost 20 pounds, and he has been unable to work for 6 weeks. Physical examination reveals a cachecticappearing young man with a palpable liver and spleen. He has no adenopathy. Laboratory studies reveal pancytopenia with an ANC of 726 and a platelet count of 15,000, elevated transaminases, and diffuse infiltrates on chest CT. He is begun on empirical antibiotics. Which of the following tests is LEAST likely to be helpful in making a diagnosis: A . Bone marrow aspirate and biopsy B. Liver biopsy C. Serum soluble CD25 D. Serum ferritin E. Fibrinogen Answer: B  Liver biopsy. This patient almost certainly has hemophagocytic lymphohistiocytosis (HLH). He has pancytopenia, splenomegaly, and fever, with no obvious infectious source. Other diagnostic features that would support the diagnosis would be the finding of hemophagocytosis on bone marrow, an elevated serum ferritin, an elevated soluble CD25, elevated triglycerides, and low fibrinogen. A liver biopsy, although it might show hemophagocytosis, is likely to be dangerous in someone with this low a platelet count and is not considered a particularly helpful diagnostic procedure.

1116

CHAPTER 161  Eosinophilic Syndromes  

161  EOSINOPHILIC SYNDROMES AMY D. KLION



DEFINITIONS



Eosinophils are terminally differentiated cells of the myeloid lineage that circulate in low numbers in the peripheral blood. Although normal ranges vary depending on the laboratory, blood eosinophilia is typically defined as an absolute eosinophil count (AEC) of greater than 450/µL and hypereosinophilia (HE) as an AEC of 1500/µL or greater. Percentages of eosinophils in the total white blood count (WBC) can be misleading because they are affected by the relative numbers of neutrophils, which show significant racial differences. Tissue HE is loosely defined as extensive eosinophil infiltration in the opinion of the pathologist and may or may not be accompanied by blood eosinophilia. Consensus thresholds for tissue HE have not been defined, except for bone marrow eosinophilia and eosinophilic esophagitis. The term hypereosinophilic syndrome (HES) was first used to describe a heterogeneous group of patients with eosinophil-related clinical manifestations of unknown etiology. Since that time, a number of definitions and classification schemes have been proposed by consensus groups to attempt to distinguish between HES of known and unknown etiologies.1,2 For the purposes of this chapter, HES will be defined by an AEC of 1500/µL or greater on at least two occasions with evidence of clinical manifestations attributable to the eosinophilia, regardless of the underlying etiology. This definition encompasses a number of clinical subtypes that differ in etiology and/or clinical manifestations and have prognostic and therapeutic implications (Table 161-1). HE or HES can be due to neoplastic or clonal proliferation of eosinophils (myeloid HE or HES), aberrant lymphocyte production of cytokines that promote eosinophilia (lymphoid HE or HES), inherited genetic abnormalities (familial HE or HES), or unknown causes (idiopathic HE or HES). Overlap HES refers to eosinophilic disorders involving a single organ system, such as eosinophilic gastrointestinal disorders and eosinophilic fasciitis, or defined clinical entities, such as eosinophilic granulomatosis with polyangiitis (EGPA), or episodic angioedema and eosinophilia (Gleich syndrome), that overlap in clinical presentation with other subtypes of HES. Finally, HE or HES can complicate a wide variety of diseases and conditions, including helminth infection, neoplasia, and drug hypersensitivity. When this occurs, the eosinophilic manifestations are collectively referred to as secondary or associated HE or HES.  

of mild to moderate eosinophilia worldwide is helminth infection, although atopic disease is the most likely culprit in industrialized settings. In contrast, HES (with the exception of some of the single-organ variants) is extremely rare, with an overall estimated prevalence of 0.3 to 6.3 per 100,000 in North America. The prevalence of eosinophilic esophagitis appears to be increasing, in keeping with an overall increase in the incidence of atopic disease, and has been estimated at 2.5 to 5 per 10,000 in the United States and Europe. Racial and ethnic differences have also been described.

EPIDEMIOLOGY

Eosinophilia (AEC >450 eosinophils/µL) is common, occurring in 5 to 15% of the general population depending on the setting. The most frequent cause

PATHOBIOLOGY

Eosinophil Biology and Mediators of Tissue Damage

Eosinophils develop from pluripotent hematopoietic stem cells in the bone marrow (Chapter 147) under the influence of a complex array of transcription factors and cytokines,3 the most important of which is interleukin-5 (IL-5). Mature eosinophils are characterized by the presence of cytoplasmic granules that stain red with the acidic dye, eosin. These granules contain four cationic proteins (major basic protein [MBP1], eosinophilic cationic protein [ECP], eosinophil-derived neurotoxin [EDN], and eosinophil peroxidase [EPO]) and a panoply of pre-formed cytokines and chemokines within a complex membranovesicular network. Release of granule proteins and mediators can occur in an indiscriminate way (through cytolysis or classic exocytosis) or selectively through a process called piecemeal degranulation.4 Under homeostatic conditions, mature eosinophils are released into the blood, where they remain in the circulation for up to 36 hours, before trafficking to the tissues. Data suggest that these “resident” eosinophils may play important regulatory roles in the maintenance of tissue integrity and cell-cell interactions.5 When eosinophils become activated, irrespective of the cause, they can cause tissue damage and other clinical manifestations through direct and indirect mechanisms. The latter include tissue deposition of cytotoxic eosinophil granule proteins and reactive oxygen intermediates, recruitment of inflammatory cells through the selective secretion of cytokines and chemokines, formation of lipid bodies with production of leukotrienes and other inflammatory mediators, and activation of mast cells, fibroblasts, and endothelial cells to promote fibrogenesis and thrombosis. Modulation of surface receptor expression and increased responsiveness to chemokines in response to activation signals also play important roles in eosinophil recruitment to the tissues and, ultimately, in disease pathogenesis.

Mutations

A number of recurrent genetic mutations have been described in clonal (neoplastic) eosinophilia.6 The most common of these, an interstitial deletion in chromosome 4 leading to formation of the fusion gene FIP1L1-PDGFRA, likely causes eosinophil activation through a network of molecules including the IL-5 receptor, JAK2, and Lyn.7 The mechanisms of eosinophil activation in the setting of other mutations associated with clonal eosinophilia remain largely unexplored. Disease risk loci have been identified in a number of allergic disorders characterized by eosinophil accumulation and activation in tissues, including eosinophilic esophagitis.8 Some of these, such as 5q22, encode genes known to be involved in eosinophil activation.

TABLE 161-1 CLINICAL SUBTYPES OF HYPEREOSINOPHILIA/HYPEREOSINOPHILIC SYNDROMES CLINICAL SUBTYPE

DEFINITION

EXAMPLES

Myeloid (clonal, neoplastic)

Presumed or proven primary neoplastic or clonal eosinophilia

PDGFR-associated myeloid neoplasms Chronic eosinophilic leukemia–not otherwise specified (reviewed in reference 6)

Lymphocytic

Eosinophilia secondary to production of interleukin-5 or other eosinophilpromoting cytokines by aberrant and/or clonal lymphocytes



Familial

Hypereosinophilia in multiple generations

Autosomal dominant familial eosinophilia

Idiopathic

Unknown cause and exclusion of other clinical subtypes



Overlap

Clinically-defined eosinophilic disorders, including those with single organ involvement, that overlap in clinical presentation with idiopathic or lymphocytic variant hypereosinophilic syndrome

Eosinophilic gastrointestinal disease Eosinophilic fasciitis Eosinophilic granulomatosis with polyangiitis Wells syndrome Episodic angioedema and eosinophilia

Secondary (associated)

Eosinophilia in the setting of a distinct diagnosis in which eosinophilia has been described in a subset of patients

See Table 161-2

Hypereosinophilia of unknown significance

Asymptomatic hypereosinophilia without evidence of end-organ manifestations or a primary myeloid etiology



CHAPTER 161  Eosinophilic Syndromes  

ABSTRACT

Eosinophils are terminally differentiated cells of the myeloid lineage that are primarily tissue based and circulate in low numbers in the blood of healthy individuals. Increased levels of blood eosinophils are associated with a wide variety of conditions, ranging from atopic disorders and helminth infections to rare hypereosinophilic syndromes. Since the clinical manifestations of eosinophilia are protean and can be very similar regardless of the underlying etiology, a careful history is essential to narrow the differential diagnosis. Treatment is guided by the acuity and severity of clinical manifestations and the most likely cause of the eosinophilia. Whereas glucocorticoids have been the mainstay of treatment for many eosinophilic disorders, the availability of therapies that directly target the eosinophils is rapidly changing the therapeutic paradigm.

KEYWORDS

eosinophil hypereosinophilic syndrome myeloid neoplasm drug hypersensitivity helminth infection

1116.e1

1117

CHAPTER 161  Eosinophilic Syndromes  



CLINICAL MANIFESTATIONS

The clinical manifestations of eosinophilia are very varied and include asymptomatic eosinophilia detected on a routine blood count, nonspecific symptoms such as fatigue and myalgia, and signs and symptoms related to eosinophilic tissue infiltration and/or complications of eosinophil-associated hypercoagulability. Although skin, pulmonary, and gastrointestinal manifestations are most frequent, any organ can be affected.9 Moreover, there is considerable heterogeneity in the clinical manifestations within any given organ system. For instance, common presentations of eosinophilic skin involvement include intractable pruritus without rash, eczematous rash, urticaria, bullous lesions, and mucosal ulcerations. Although some clinical manifestations are more common in specific subtypes of HES (e.g., skin manifestations in lymphoid HES), there is considerable overlap in presentation among clinical subtypes. Moreover, the most serious clinical manifestations, such as endomyocardial fibrosis and thromboembolism, can occur in any of the varied HES subtypes, including secondary or associated HES caused by untreated helminth infection or drug hypersensitivity.  

DIAGNOSIS

Differential Diagnosis

Although the differential diagnosis of eosinophilia is very broad, ranging from common allergic disorders to rare idiopathic HESs, a careful history can substantially narrow the possibilities. Particular attention should be paid to medications and supplements, travel and exposures, risk factors for neoplasia, and prior medical history (especially asthma, atopic disease, and/or recurrent infections) because these may provide clues to secondary causes of eosinophilia, such as helminth infection, neoplasia, and immunodeficiency syndromes, that require therapy directed at the underlying cause rather than the eosinophilia itself (Table 161-2).10,11 Whereas common allergic disorders, including asthma and atopic dermatitis, can be associated with HE, especially in children, extremely high AECs (≥5000/µL) should prompt

TABLE 161-2 SELECTED SECONDARY CAUSES OF EOSINOPHILIA REQUIRING TREATMENT DIRECTED AT THE UNDERLYING CAUSE CATEGORY Infection Helminthic Protozoan Ectoparasitic Viral Bacterial Fungal Neoplasia

ETIOLOGIES Varied Isospora belli infection Sarcocystis infection Scabies Myiasis HIV infection Chronic tuberculosis Coccidioidomycosis Cryptococcosis Lymphoma Leukemia Carcinoma

COMMENTS Reviewed in reference 10 Most pronounced early in infection and with tissue invasive species Other protozoan infections are not associated with eosinophilia Hypersensitivity reactions

Immunodeficiency

Hyper-IgE syndromes Omenn syndrome Others (reviewed in reference 11)

Other

The first step in the diagnostic evaluation of the patient with eosinophilia is confirmation of the eosinophil count, while keeping in mind that fever, bacterial infection, and therapies, including glucocorticoids, that may be prescribed for other reasons, can all suppress eosinophilia.12 Once eosinophilia is confirmed, the next step is a search for secondary causes that require specific therapy directed at the underlying disorder (associated HES). If such a cause is identified, the patient should be treated and reassessed for resolution of the eosinophilia. Patients with persistent eosinophilia and those without evidence of a clear secondary cause should undergo an initial evaluation aimed at the identification of end-organ involvement and classification by clinical subtype (Fig. 161-1). When possible, this should occur before institution of therapy because many diagnostic tests are affected by glucocorticoids and other therapies. That said, treatment should not be delayed if life-threatening or disabling complications are present or imminent. The diagnostic evaluation should be performed in a step-wise manner beginning with a careful history and physical examination, complete blood count, and differential and routine chemistries (including liver function tests and serum immunoglobulins). An initial screen for evidence of myeloid or lymphocytic variant HES should include serum vitamin B12 and tryptase levels, as well as immunophenotyping for T- and B-cell clonality by flow cytometry. Abnormal results, a peripheral eosinophil count of 5000/µL or greater, or other evidence suggestive of a hematologic malignancy (including lymphadenopathy, splenomegaly, or dysplastic or precursor cells on peripheral blood smear) should prompt a bone marrow aspirate and biopsy with standard cytogenetic analysis. Specific testing for recurrent molecular abnormalities associated with HES should be guided by the clinical features and, if suspected, should be

Confirmation of eosinophilia Y

Hodgkin, non-Hodgkin Especially T-cell leukemias and pre-B cell ALL Most commonly adenocarcinoma Prescription and non-prescription drugs, dietary supplements

Autoimmune and Varied other immunologic disorders

Diagnostic Evaluation

Advanced disease Rare Eosinophilia can complicate a wide variety of fungal infections

Drug hypersensitivity Varied Varied

evaluation for another cause. A detailed review of systems should be obtained in all patients because the pattern of clinical manifestations can be suggestive of a particular diagnosis. For example, a history of aspirin sensitivity in an eosinophilic patient with worsening asthma and sinusitis is highly suggestive of aspirin-exacerbated respiratory disease (AERD), whereas symptoms of peripheral neuropathy in a similar patient should point to a diagnosis of EGPA. It is extremely important to identify patients with primary myeloid disorders presenting as HES because these patients often have aggressive disease, and the approach to therapy is quite different from that of other eosinophilic disorders (see the later Treatment section). Clinical and laboratory features suggestive of a primary myeloid disorder include dysplastic eosinophils and eosinophil precursors in the blood, anemia or erythrocytosis, thrombocytopenia, splenomegaly, and elevated serum levels of vitamin B12 and/or tryptase. Male gender should raise suspicion for PDGFR-associated myeloid neoplasms in this setting.

Cholesterol embolism — Hypoadrenalism Radiation Post-transplant status Cytokine therapy: IL-2, GM-CSF

ALL = acute lymphoblastic leukemia; GM-CSF = granulocyte-macrophage colony-stimulating factor; HIV = human immunodeficiency virus; IgE = immunoglobulin E; IgG4 = immunoglobulin G4; IL-2 = interleukin-2.

Secondary treatable cause identified?

Reassess in 1-3 months. Eosinophilia?

Y

Y

N

Treat and reassess. Persistent eosinophilia?

Sarcoidosis Inflammatory bowel disease IgG4 disease Others

Y Evaluation for end-organ manifestations and clinical phenotype

N

N

Evaluation for end-organ manifestations and clinical phenotype

N STOP

STOP

FIGURE 161-1.  Initial approach to the clinically stable patient with unexplained

eosinophilia.

included with the bone marrow examination, where their yield may be higher than in peripheral blood.13 Although the recommended testing for end-organ damage depends on the clinical signs and symptoms, serum troponin levels, echocardiography, and pulmonary function tests should be performed in all patients to assess clinically evident or occult cardiopulmonary involvement such as endomyocardial fibrosis that can be suggested by echocardiographic abnormalities. Radiologic imaging of the chest, abdomen, and pelvis should also be considered in all patients to identify splenomegaly, lymphadenopathy, and clinically silent neoplastic or inflammatory disease.

TREATMENT  Initial Therapy

The most appropriate initial treatment for a patient with eosinophilia depends on the acuity of the clinical presentation, the severity of the clinical signs and symptoms, and the most likely etiology of the eosinophilia. At the present time, glucocorticoids remain the first-line agent for most hypereosinophilic disorders, with three important exceptions: HES secondary to a treatable cause (associated HES), primary clonal or neoplastic HES with an identified mutation for which targeted therapy is available (i.e., imatinib-sensitive, PDGFR mutation–positive HES), and patients with HE of unknown significance (who may not require therapy at all). The most appropriate dose of glucocorticoids for an individual patient has not been adequately studied, although clinical subtype appears to be a useful predictor of systemic glucocorticoid response in HES.14 Some eosinophilic disorders, including eosinophilic esophagitis and eosinophilic dermatitis, can be successfully managed with topical steroid therapy alone. If life-threatening or disabling complications, such as endomyocardial fibrosis, respiratory failure, or neurologic impairment, are present or imminent, high-dose glucocorticoid therapy should be initiated without delay. This is true even in the setting of associated HES unless a treatable secondary cause is identified and is likely to respond rapidly to specific therapy. Ivermectin (200 µg/kg/day for 1 to 2 days) should be administered if there is any potential history of exposure to strongyloidiasis to prevent hyperinfection syndrome. If there is no response to glucocorticoid therapy, a second agent should be added based on the clinical subtype and most likely etiology of the eosinophilia, as discussed later. Initial treatment of myeloid neoplasms presenting as HES should be guided by the molecular findings. Imatinib is clearly the therapy of choice in PDGFR mutation–associated disease, with near 100% response rates reported in most series and the possibility of cure in some patients with long-term therapy.15 Results with other targeted therapies, including JAK inhibitors, have been variable in patients with myeloid forms of HES to date, even in the presence of mutations in the corresponding gene. Consequently, glucocorticoids are often used initially with rapid addition of hydroxyurea or other second-line agents as needed. This approach is likely to change, however, as more experience is gained with targeted therapies.

Second-Line Agents

The selection of a second-line agent depends on a variety of factors, including the clinical subtype, physicians’ and patients’ preference, and cost. Conventional agents used to treat glucocorticoid-resistant, PDGFR mutation–negative patients include hydroxyurea, interferon-α, methotrexate, and imatinib. Patients who are PDGFR mutation positive and who are intolerant of or resistant to imatinib should be treated with a second- or third-generation tyrosine kinase inhibitor with activity against PDGFR, with early consideration of hematopoietic stem cell transplant in nonresponders. The recent availability of therapeutic agents that specifically target eosinophils is rapidly changing the approach to the treatment of eosinophilic disorders. Three biologics that target eosinophils through the IL-5 axis, mepolizumab, reslizumab, and benralizumab, have been approved for the treatment of eosinophilic asthma on the basis of phase 3 trials documenting reduction in asthma exacerbations with little or no toxicity. A1-A3  A placebo-controlled, double-blind phase 3 trial in treatment-refractory or relapsing EGPA demonstrated improved remission rates and length of remission in patients receiving mepolizumab (300 mg subcutaneously every 4 weeks), A4  resulting in approval of mepolizumab for this indication. All three agents have been used to treat PDGFRA-negative HESs other than EGPA in phase 2 trials with promising results, although only mepolizumab is currently available on a compassionate use protocol for patients with life-threatening, treatment-refractory HES.16 Other novel agents currently in development for eosinophilic disorders, including biologics targeting eotaxin, Siglec-8 (an inhibitory receptor whose expression is restricted to eosinophils, basophils, and mast cells), and the IL-4/IL-13 axis, are likely to provide additional therapeutic options for the heterogeneous group of eosinophilic disorders.



PROGNOSIS

Although the prognosis of an individual patient with HES depends substantially on the clinical subtype, the severity of clinical manifestations at presentation,

and treatment responsiveness, the overall prognosis has improved dramatically (>90% 5-year survival in most series) with advances in diagnostic tools and therapeutic approaches. The most clear-cut example of this has been the dramatic decrease in mortality in patients with clinical features consistent with an eosinophilic myeloid neoplasm (the majority of whom have FIP1L1-PDGFRA– associated disease) following the discovery of imatinib. In fact, mortality and morbidity in this group of patients (which were 30 to 50% at 5 years before the introduction of imatinib) are now almost entirely restricted to those patients who present with irreversible cardiac or other life-threatening manifestations. Progression to lymphoma remains an issue in patients with lymphocytic variant, particularly in those patients with CD3-CD4+ T-cell clones.

  Grade A References A1. Ortega HG, Liu MC, Pavord ID, et al. Mepolizumab treatment in patients with severe eosinophilic asthma. N Engl J Med. 2014;371:1198-1207. A2. Castro M, Zangrillia J, Wechsler ME, et al. Reslizumab for inadequately controlled asthma with elevated blood eosinophil counts: results from two multicenter, parallel, double-blind, randomised, placebo-controlled, phase 3 trials. Lancet Respir Med. 2015;3:355-366. A3. FitzGerald JM, Bleeker ER, Nair P, et al. Benralizumab, an anti-interleukin-5 receptor α monoclonal antibody, as add-on treatment for patients with severe, uncontrolled, eosinophilic asthma (CALIMA): a randomised, double-blind, placebo-controlled phase 3 trial. Lancet. 2016;388:2128-2141. A4. Wechsler ME, Akuthota P, Jayne D. Mepolizumab or placebo for eosinophilic granulomatosis with polyangiitis. N Engl J Med. 2017;376:1921-1932.

  GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 161  Eosinophilic Syndromes  

GENERAL REFERENCES 1. Gotlib J. World Health Organization-defined eosinophilic disorders: 2017 update on diagnosis, risk stratification, and management. Am J Hematol. 2017;92:1243-1259. 2. Klion AD. How I treat hypereosinophilic syndromes. Blood. 2015;126:1069-1077. 3. Willebrand R, Voehringer D. Regulation of eosinophil development and survival. Curr Opin Hematol. 2017;24:9-15. 4. Weller PF, Spencer LA. Functions of tissue-resident eosinophils. Nat Rev Immunol. 2017;17: 746-760. 5. Wen T, Rothenberg ME. The regulatory function of eosinophils. Microbiol Spectr. 2016;1-12. 6. Reiter A, Gotlib J. Myeloid neoplasms with eosinophilia. Blood. 2017;129:704-714. 7. Li B, Zhang G, Li C, et al. Lyn mediates FIP1L1-PDGFRA signal pathway facilitating IL-5RA intracellular signal through FIP1L1-PDGFRA/JAK2/Lyn/Akt network complex in CEL. Oncotarget. 2016;5:64984-64998. 8. Kottyan LC, Rothenberg ME. Genetics of eosinophilic esophagitis. Mucosal Immunol. 2017;10: 580-588.

1118.e1

9. Khoury P, Bochner BS. Consultation for elevated blood eosinophils: clinical presentations, high value diagnostic tests, and treatment options. J Allergy Clin Immunol Pract. 2018;6:1446-1453. 10. O’Connell EM, Nutman TB. Eosinophilia in infectious diseases. Immunol Allergy Clin North Am. 2015;35:493-522. 11. Navabi B, Upton JE. Primary immunodeficiencies associated with eosinophilia. Allergy Asthma Clin Immunol. 2016;12:27. 12. Wang SA. The diagnostic work-up of hypereosinophilia. Pathobiology. 2019;86:39-52. 13. Maric I, Sun X. Advances in diagnosis of mastocytosis and hypereosinophilic syndrome. Semin Hematol. 2019;56:22-29. 14. Khoury P, Abiodun AO, Holland-Thomas N, et al. Hypereosinophilic syndrome subtype predicts responsiveness to glucocorticoids. J Allergy Clin Immunol Pract. 2018;6:190-195. 15. Khoury P, Desmond R, Pabon A, et al. Clinical features predict responsiveness to imatinib in platelet-derived growth factor receptor-alpha-negative hypereosinophilic syndrome. Allergy. 2016;71: 803-810. 16. Kuang FL, Klion AD. Biologic agents for the treatment of hypereosinophilic syndromes. J Allergy Clin Immunol Pract. 2017;5:1502-1509.

1118.e2

CHAPTER 161  Eosinophilic Syndromes  

REVIEW QUESTIONS 1. A 32-year-old immigrant from Cameroon presents for evaluation of eosinophilia (absolute eosinophil count: 1600/µL) detected on routine bloodwork. She has a history of seasonal allergic rhinitis for which she takes intermittent cetirizine but is otherwise healthy. Physical examination is unremarkable, and laboratory testing is notable for a normal complete blood count with the exception of the eosinophilia and a mildly elevated serum IgE level (230 IU/mL). Three stool samples for ova and parasites are negative. Which of the following is the most likely diagnosis? A . Atopic disease B. Drug hypersensitivity C. Helminth infection D. Hypereosinophilia of unknown significance E. Idiopathic hypereosinophilic syndrome Answer: C  The most common cause of eosinophilia worldwide is helminth infection, and this patient is an immigrant from a high-prevalence area. The finding of negative stools does not exclude a helminth infection. Stool examination is insensitive for detection of Strongyloides and Schistosoma species, both of which are endemic in Cameroon, and does not detect infection with filarial parasites, such as Loa loa, which commonly cause asymptomatic eosinophilia in Cameroon. Although atopic disease or drug hypersensitivity could cause asymptomatic eosinophilia in this patient, they are less likely given the epidemiology. Hypereosinophilia of unknown significance and idiopathic hypereosinophilic syndrome are not only uncommon, but also are diagnoses of exclusion. 2. A 55-year-old man presents with chest pain, shortness of breath, and wheezing. Physical examination reveals a papular erythematous rash on the upper arms bilaterally, a blowing systolic murmur at the apex, inspiratory and expiratory wheezes, and a palpable spleen tip. Complete blood count is notable for an absolute eosinophil count of 14,640/µL and hemoglobin 10.6 g/dL. Peripheral smear shows promyelocytes and dysplastic eosinophils. Serum tryptase is elevated at 45 ng/dL. Serum IgE is within normal limits. Which of the following tests is most likely to provide a definitive diagnosis? A . FIP1L1-PDGFRA testing in peripheral blood B. Echocardiography C. Lymphocyte subset analysis D. D816V KIT testing in bone marrow E. Karyotype analysis Answer: A  The clinical features in this case are classic for FIP1L1-PDGFRA– associated myeloid neoplasm, including male gender, eosinophilic endomyocardial fibrosis, dysplastic eosinophils, anemia, splenomegaly, and elevated serum tryptase levels. The interstitial deletion that produces the fusion gene is detectable by polymerase chain reaction or fluorescence in situ hybridization in peripheral blood. Cytogenetic abnormalities detectable on karyotype analysis are extremely uncommon. Echocardiography would likely confirm endomyocardial fibrosis, but this can be seen in any of the clinical subtypes of HES. The history is inconsistent with lymphocytic variant HES, which is characterized by skin manifestations and elevated serum IgE. Finally, although elevated tryptase, anemia, and splenomegaly are characteristic of D816V KIT mutation–positive systemic mastocytosis, and eosinophilia can be present, the clinical manifestations of systemic mastocytosis (e.g., anaphylaxis, urticaria pigmentosa, bone pain) differ from those of HES. 3. A 36-year-old obese man presents to the emergency room with presyncope. He has no prior cardiac history, and his only travel outside of the United States was to rural China more than 10 years earlier. Evaluation is notable for a mildly elevated serum troponin (2.8 ng/mL), a normal electrocardiogram, and an absolute eosinophil count of 12,456/µL. Echocardiogram and exercise stress test are normal. Bone marrow biopsy is hypercellular with dysplastic eosinophils, but no increase in blasts. What is the most appropriate course of action? A . Withhold therapy until diagnostic testing is complete B. Prednisone burst taper C. Prednisone (1 mg/kg daily) D. Prednisone (1 mg/kg daily) and ivermectin (200 µg/kg) E. Imatinib mesylate (400 mg daily)

Answer: D  This patient meets criteria for hypereosinophilic syndrome with potentially life-threatening clinical manifestations and should be treated expeditiously. In the absence of data pointing to a likely secondary cause or convincing evidence for PDGFR-associated myeloid neoplasm, high-dose glucocorticoids are indicated.* Because Strongyloides stercoralis is endemic in rural China and infection can persist for decades, ivermectin should be administered concomitantly to prevent hyperinfection syndrome. (*Klion AD. How I treat hypereosinophilic syndromes. Blood. 2015;126:1069-1077.) 4. A 55-year-old man with a history of seasonal allergic rhinitis and hypertension is referred for eosinophilia (5200/µL) noted on routine bloodwork. He lives in Boston, where he works as an accountant. He denies foreign travel, recent medication changes, or unusual exposures. There is no family history of eosinophilia, and his most recent bloodwork 1 year ago was notable for an eosinophil count of 1200/µL. He is completely asymptomatic, with a normal physical examination. Evaluation, including routine bloodwork, echocardiography, pulmonary function tests, bone marrow biopsy, testing for T- and B-cell clonality, and FIP1L1-PDGFRA, is unrevealing except for an absolute eosinophil count of 7600/µL. Which of the following is the most appropriate management? A . Initiate prednisone therapy. B. Initiate imatinib therapy. C. Recheck eosinophil count in 1 week. D. Recheck eosinophil count in 1 month. E. Recheck eosinophil count in 1 year. Answer: C  Based on the available information, the most likely diagnosis in this patient would be hypereosinophilia of unknown significance. However, this is a diagnosis of exclusion, and the eosinophil count appears to be rising. Published data indicate that progression is unlikely if there are no signs of clinical manifestations after 5 years of hypereosinophilia in the absence of specific treatment.* Conversely, progression has been reported when less stringent criteria are used.† The safest and most appropriate option for this patient in the short term is to monitor his eosinophil count and clinical symptoms closely. (*Chen YY, Khoury P, Ware JM, et al. Marked and persistent eosinophilia in the absence of clinical manifestations. J Allergy Clin Immunol. 2014;133:1195-1202. †Helbig G, Hus M, Francuz T, et al. Characteristics and clinical outcome of patients with hypereosinophilia of undetermined significance. Med Oncol. 2014;31:815.) 5. A 77-year-old Indian woman with a history of chronic cough/asthma and peripheral eosinophilia is referred for hand and foot pain of several weeks’ duration. Past medical history is significant for eosinophilic pneumonia 10 years ago and eosinophilic myocarditis with pleural and pericardial effusions earlier this year that responded to glucocorticoid therapy. Physical examination is unremarkable. Laboratory testing is notable for an eosinophil count of 5810/µL, an elevated erythrocyte sedimentation rate at 88 mm/ hour, and microscopic hematuria. Chest CT shows a nodular infiltrate versus mass lesion in the left upper lobe, patchy densities bilaterally, mediastinal lymphadenopathy, and a left pleural effusion. ANCA antibody testing is negative. What is the most likely diagnosis? A . Tropical pulmonary eosinophilia B. Chronic eosinophilic pneumonia C. Sarcoidosis D. Drug hypersensitivity E. Eosinophilic granulomatosis with polyangiitis Answer: E  Whereas the differential diagnosis of eosinophilia and pulmonary infiltrates includes all of the choices listed, this patient’s history of asthma, prior episodes of eosinophilic pneumonia, and myocarditis and symptoms suggestive of peripheral neuropathy are most consistent with eosinophilic granulomatosis with polyangiitis (EGPA).* ANCA antibody testing is positive in less than 50% of patients with EGPA. (*Groh M, Pagnoux C, Baldini C, et al. Eosinophilic granulomatosis with polyangiitis (Churg-Strauss) (EGPA) Consensus Task Force recommendations for evaluation and management. Eur J Intern Med. 2015;26:545-553.)

1118

CHAPTER 162  Approach to the Patient with Bleeding and Thrombosis  

162  APPROACH TO THE PATIENT WITH BLEEDING AND THROMBOSIS ANDREW I. SCHAFER

MECHANISMS OF HEMOSTASIS AND THROMBOSIS

Normal Hemostasis

The coagulation system is normally quiescent, and blood fluidity within blood vessels is maintained by the actions of a continuous monolayer of endothelial cells that line the intimal surface of the vasculature throughout the circulatory tree. At a site of vascular damage, the antithrombotic properties of endothelium are lost, and thrombogenic constituents of the subendothelial vessel wall become exposed to circulating blood. The result is rapid formation of a hemostatic clot that consists of platelets and fibrin and is localized to the area of vascular injury. Activation of platelets and formation of fibrin occur essentially simultaneously and interdependently to effect hemostasis. Subsequently, vascular repair is accomplished by thrombolysis and recanalization of the occluded site.1 Platelet activation at a site of vascular injury begins with the adhesion of platelets to the locally de-endothelialized intimal surface (platelet–vessel wall interaction).2 Platelet adhesion is mediated by von Willebrand factor, which sticks circulating platelets to the area of damaged vessel wall by binding to its receptors located in platelet membrane glycoprotein Ib. The adherent platelets then undergo a “release reaction,” during which they discharge constituents of their storage granules, including adenosine diphosphate (ADP), and simultaneously elaborate thromboxane A2 from arachidonic acid through the aspirin-inhibitable cyclooxygenase reaction. ADP, thromboxane A2, and other components of the release reaction act in concert to recruit and activate additional platelets from the circulation to the site of vascular injury. These activated platelets expose binding sites for fibrinogen by forming the surface membrane glycoprotein IIb/IIIa complex. In the process of platelet aggregation (platelet-platelet interactions), fibrinogen (or von Willebrand factor under conditions of high shear stress) mediates the final formation of an occlusive platelet plug.

CHAPTER 162  Approach to the Patient with Bleeding and Thrombosis  

ABSTRACT

Physiological coagulation in response to injury is referred to as hemostasis. Blood platelets and coagulation factors participate together in forming a hemostatic plug. Pathological coagulation, however, is referred to as thrombosis, which occurs when coagulation is abnormally hyperactive and is unneeded. Systemic bleeding disorders are caused by quantitative deficiencies or functional abnormalities in the coagulation cascade and in platelets. Thrombosis, particularly venous thrombosis, can be the result of deficiencies or functional abnormalities in physiological antithrombotic proteins (like antithrombin III, protein C, and protein S). Screening tests for systemic bleeding disorders (platelet count, prothrombin time [PT], and activated partial thromboplastin time [aPTT]) can rule out most, but not all, clinically significant coagulation abnormalities. For example factor XIII deficiency or only mildly low coagulation factor levels will not be picked up by a prolonged PY or aPTT. Algorithms for more specialized tests are provided, and newer global coagulation and point-of-care tests are discussed. Evaluation of hypercoagulable states requires a panel of blood tests, since no well validated screening tests are available.

KEYWORDS

hemostasis thrombosis bleeding disorders coagulation factors platelets inhibitors

1118.e3

CHAPTER 162  Approach to the Patient with Bleeding and Thrombosis  

TF XI VII

VIIa

HMWK

XII PK HMWK XIIa

IX Ca2+

VIIa/TF Ca2+ PL VIIIa

XIa

Ca2+ PL PT

History and Physical Examination

VIII

Va

V

Th

Fibrinogen

FIBRIN monomer FIBRIN polymer XIIIa

XIII

Cross-linked FIBRIN polymer FIGURE 162-1.  Coagulation cascade. This scheme emphasizes an understanding of (1) the importance of the tissue factor pathway in initiating clotting in vivo, (2) the interactions among pathways, and (3) the pivotal role of thrombin in sustaining the cascade by feedback activation of coagulation factors. HMWK = high-molecular-weight kininogen; PK = prekallikrein; PL = phospholipid; PT = prothrombin; TF = tissue factor; Th = thrombin. (From Schafer AI. Coagulation cascade: an overview. In: Loscalzo J, Schafer AI, eds. Thrombosis and Hemorrhage. Cambridge, MA: Blackwell Scientific Publications; 1994:3-12.)

Fibrin, which anchors the hemostatic platelet plug, is formed from soluble plasma fibrinogen by the action of the potent protease enzyme thrombin (Fig. 162-1). The fibrin mesh is stabilized by covalent cross-linking mediated by factor XIII. Thrombin is formed from its inactive (zymogen) plasma precursor, prothrombin, by the action of activated factor X (Xa) and its cofactor, factor Va. This sequence of reactions has classically been referred to as the common pathway of coagulation. Factor X can be activated by either the tissue factor (extrinsic) pathway or the contact activation (intrinsic) pathway of coagulation. The tissue factor pathway is now considered to be the major physiologic initiator of coagulation activation. It is triggered by the formation of the complex of tissue factor, which is exposed on the surfaces of activated vascular and blood cells, with activated factor VII (VIIa). The contact activation pathway involves a series (or cascade) of zymogen-protease reactions that are initiated by factor XII, high-molecular-weight kininogen, and prekallikrein. Activated factor XII (XIIa) converts factor XI to XIa, which in turn activates factor IX to IXa. Factor IXa is an enzyme that converts factor X to Xa, a reaction that requires factor VIIIa as a cofactor.

Physiologic Antithrombotic Mechanisms

binds to the glycosaminoglycan thrombomodulin. TFPI is a plasma protease inhibitor that specifically quenches tissue factor–induced coagulation. Finally, what little fibrin can be produced, despite these potent physiologic antithrombotic mechanisms, is digested rapidly by the endogenous fibrinolytic system. Fibrinolysis is mediated by the protease plasmin, which is generated from plasminogen in plasma by the action of endothelium-derived plasminogen activators.

EVALUATION OF THE PATIENT WITH A POSSIBLE BLEEDING DISORDER

IXa

Xa

X

1119

Intact, normal endothelium promotes blood fluidity by inhibiting platelet activation. It likewise plays a crucial role in preventing fibrin accumulation. Among the physiologic antithrombotic systems that produce this latter effect are (1) antithrombin III, (2) protein C and protein S, (3) tissue factor pathway inhibitor (TFPI), and (4) the fibrinolytic system. Antithrombin is the major protease inhibitor of the coagulation system: it inactivates thrombin and other activated coagulation factors. Heparin functions as an anticoagulant by binding to antithrombin and greatly accelerating these reactions. Heparin and heparin sulfate proteoglycans are naturally present on endothelial cells, so antithrombin inactivation of thrombin and other coagulation proteases most likely occurs physiologically on vascular surfaces rather than in fluid plasma. Activated protein C, with its cofactor protein S, functions as a natural anticoagulant by destroying factors Va and VIIIa, two essential cofactors of the coagulation cascade. Thrombin itself is the activator of protein C, and this reaction occurs rapidly only on the surfaces of intact vascular endothelial cells, where thrombin

A thorough history is paramount in evaluating a patient for a possible systemic bleeding disorder. In addition to asking the patient about spontaneous bleeding episodes in the past, the responses to specific hemostatic challenges should be recorded. A bleeding tendency may be suspected if a patient previously experienced excessive hemorrhage after surgery or trauma, including common events such as circumcision, tonsillectomy, labor and delivery, menses, dental procedures, vaccinations, and injections. Conversely, a history of normal blood clotting after such specific challenges in the recent past is just as important to note. It may be a better test of the integrity of systemic hemostasis than any laboratory measurement can provide. In a patient with a history of excessive or unexplained bleeding, the initial goal is to determine whether the cause is a systemic coagulopathy or localized anatomic or mechanical problem with a blood vessel. This situation is encountered most frequently in patients with excessive postoperative bleeding, which could be due to either local surgical trauma or a coagulation abnormality, or both. A history of prior bleeding suggests a coagulopathy, as does the finding of bleeding from multiple sites. However, this is not always the case. Even diffuse bleeding may arise from anatomic rather than hemostatic abnormalities. An example of this is recurrent mucosal hemorrhage in patients with hereditary hemorrhagic telangiectasia (Chapter 164). Conversely, a single episode of bleeding from an isolated site may be the initial manifestation of a coagulopathy. The history must include a survey of coexisting systemic diseases and drug ingestions that could affect hemostasis. Renal failure and the myeloproliferative neoplasms are associated with impaired platelet–vessel wall interactions and qualitative platelet abnormalities, connective tissue diseases and lymphomas are associated with thrombocytopenia, and liver disease causes a complex coagulopathy (Chapter 166). Ingestion of aspirin and other nonsteroidal antiinflammatory drugs (NSAIDs) that cause nonselective inhibition of cyclooxygenase leads to platelet dysfunction; these drugs are often contained in over-the-counter preparations that patients may neglect to report without specific questioning. Other drugs, such as antibiotics, also may be associated with a bleeding tendency by causing abnormal platelet function or thrombocytopenia.3 Finally, it is important to elicit a family history of bleeding problems. Although a positive history provides an important clue to a possible inherited coagulopathy, a negative history does not exclude a familial cause; for example, 20% of patients with classic hemophilia have a completely negative family history of bleeding. Mild bleeding events are commonly reported by patients with and without subsequently laboratory-documented bleeding disorders, sometimes making it difficult for hematologists to define a “significant bleeding history.” Using a web-based questionnaire, 25% of subjects in a healthy population reported epistaxis, 18% easy bruising (more commonly in women), 18% prolonged bleeding after dental extraction, and 47% of women heavy menstrual bleeding. More precise quantification of bleeding symptoms is being attempted by using “bleeding score” instruments such as the Vicenza bleeding score to help discriminate, in conjunction with laboratory testing, between healthy subjects and those with mild bleeding disorders. Patterns of clinical bleeding, as revealed by the history and physical examination, may be characteristic of certain types of coagulopathy (Table 162-1).4 In general, patients with thrombocytopenia or qualitative platelet or vascular disorders present with bleeding from superficial sites in the skin and mucous membranes. These may involve petechiae, which are pinpoint cutaneous hemorrhages that appear particularly over dependent extremities (characteristic of severe thrombocytopenia), ecchymoses (common bruises), purpura, gastrointestinal and genitourinary tract bleeding, epistaxis, and hemoptysis. In these types of disorders, bleeding tends to occur spontaneously or immediately after trauma. In contrast, patients with inherited or acquired coagulation factor deficiencies, such as hemophilia, or those on anticoagulant therapy tend to bleed from deeper tissue sites (e.g., hemarthroses, deep hematomas, retroperitoneal hemorrhage) and in a delayed manner after trauma.

1120

CHAPTER 162  Approach to the Patient with Bleeding and Thrombosis  

TABLE 162-1 CHARACTERISTIC PATTERNS OF BLEEDING IN SYSTEMIC DISORDERS OF HEMOSTASIS TYPE OF DISORDER Platelet-vascular disorders

SITES OF BLEEDING GENERAL

SKIN

MUCOUS MEMBRANES

Superficial surfaces Petechiae, ecchymoses Common: oral, nasal, gastrointestinal, genitourinary

Coagulation factor Deep tissues deficiency

Hematomas

Rare

OTHER Rare

ONSET OF BLEEDING

CLINICAL EXAMPLES

Spontaneous or immediately after trauma

Common: joint, muscle, Delayed after trauma Inherited coagulation factor retroperitoneal deficiency, acquired inhibitor, anticoagulation, disseminated intravascular coagulation, liver disease

Laboratory Testing Screening Tests of Hemostasis

A few simple screening tests have traditionally been used in the initial evaluation of patients with a suspected coagulopathy: platelet count, bleeding time, prothrombin time (PT) (also reported as the INR), activated partial thromboplastin time (aPTT), and thrombin time (TT).5 The North American Specialized Coagulation Laboratory Association (NASCOLA) has reported that most coagulation laboratories currently perform these tests as their “bleeding disorder panels.” Thrombocytopenia, reported by electronic particle counting, should be verified by examination of the peripheral smear. Pseudothrombocytopenia, a laboratory artifact of ex vivo platelet clumping, may be caused by the ethylenediaminetetraacetic acid (EDTA) anticoagulant used in tubes for blood cell counts, by other anticoagulants, or by nonphysiologic cold agglutinins acting at room temperature. It should be suspected whenever a very low platelet count is unexpectedly reported in a patient who does not exhibit any clinical bleeding. Pseudothrombocytopenia is indicated by the finding of platelet clumps on the peripheral smear, and the diagnosis is supported by the finding of simultaneously normal platelet counts in blood samples obtained by finger stick, in tubes containing other anticoagulants, or in a tube maintained at 37° C before platelet counting. Examination of the blood smear can also reveal clues to the cause of real thrombocytopenia, such as fragmented red blood cells in thrombotic thrombocytopenic purpura. The bleeding time was a widely used clinical screening test for disorders of platelet–vessel wall interactions. It measures the time to cessation of bleeding after a standardized incision over the volar aspect of the forearm. However, the test is prone to problems related to quality control, reproducibility, sensitivity, and specificity. Therefore, because von Willebrand disease is the most common genetic cause of abnormal platelet–vessel wall interactions, most experts now recommend replacing the bleeding time with specific tests for von Willebrand disease (VWD screen) in the initial evaluation of patients with a suspected coagulopathy (Chapter 164). As an additional replacement for the bleeding time, especially when a functional (qualitative) abnormality of platelets is suspected by characteristic mucocutaneous bleeding or bruising, a global assay of platelet function can be appended to the panel of screening tests. This is most commonly a platelet function analyzer (PFA) closure time. The PFA and other platelet function studies (see later) must be performed after discontinuation of drugs that interfere with platelet function (e.g., aspirin and other NSAIDs). The PT measures the integrity of the extrinsic and common pathways of coagulation (factors VII, X, and V; prothrombin; and fibrinogen) (Fig. 162-2). The aPTT measures the integrity of the intrinsic and common pathways of coagulation (high-molecular-weight kininogen; prekallikrein; factors XII, XI, IX, VIII, X, and V; prothrombin; and fibrinogen). The sensitivity of the PT and aPTT in detecting coagulation factor deficiencies may vary with the reagents used to perform these tests, and each laboratory must determine its own reference standards. The TT is a screen for quantitative deficiencies and qualitative defects of plasma fibrinogen. With a few notable exceptions, normal results for all these screening tests of hemostasis essentially exclude any clinically significant systemic coagulopathy. However, patients with factor XIII deficiency6 may have a serious bleeding diathesis but normal screening tests; specific tests for factor XIII deficiency should be performed if this disease is suspected (Chapter 165). The PT and aPTT detect only the more severe deficiencies of coagulation factors, usually

Thrombocytopenia, functional platelet disorder, vascular fragility, disseminated intravascular coagulation, liver disease

HMWK PK F XII F XI F VII F IX

F VIII

aPTT

PT FX FV Prothrombin Fibrinogen

F XIII

Cross-linked fibrin

FIGURE 162-2.  Classic coagulation cascade. The prothrombin time (PT) measures the integrity of the extrinsic and common pathways, whereas the activated partial thromboplastin time (aPTT) measures the integrity of the intrinsic and common pathways. Factor (F) XIII deficiency is not detected by PT or aPTT. HMWK = high-molecular-weight kininogen; PK = prekallikrein.

at levels less than 30% of normal; specific factor levels should be determined if a mild coagulation factor deficiency is suspected (Table 162-2).7 Rare disorders of fibrinolysis also may be associated with normal screening tests, necessitating more specialized tests when indicated (Chapter 165). Abnormalities in these screening tests of hemostasis may be pursued by more specialized tests to establish a specific diagnosis (Fig. 162-3); other similar algorithms have been published. An abnormal VWD screen should be followed up with more specialized tests, including von Willebrand factor (VWF) multimer analysis, to identify the type of von Willebrand disease involved. Abnormalities in the PFA closure time should be followed by more specialized light transmission aggregometry (LTA) of platelets using a panel of agonists (ADP, epinephrine, collagen, arachidonic acid, ristocetin) that induce characteristic changes in light transmission (or optical density) in stirred suspensions of freshly isolated platelets in plasma (PRP). LTA is still considered the “gold standard” for platelet function testing8 (Chapter 164). The finding of a prolonged PT and/or aPTT indicates either a deficiency of one or more coagulation factors or the presence of an inhibitor (Fig. 1624), usually an antibody, directed at one or more components of the coagulation system. These two possibilities can be distinguished by performing a simple inhibitor screen, which involves a 1 : 1 mix of the patient’s plasma and normal plasma. The premise of the test is that even if the patient’s plasma is completely deficient (0% level) in a certain factor, mixing it 1 : 1 with normal plasma (100% level) should bring the concentration of that factor to 50% in the

CHAPTER 162  Approach to the Patient with Bleeding and Thrombosis  

1121

History and physical exam suggest possible systemic bleeding disorder Discontinue all antithrombotics and other drugs that may interfere with tests Screening laboratory tests:

Prothrombin time (INR)

A

Partial thromboplastin time

Thrombin time

Platelet count

VWD screen

Platelet function (PFA)

Either or both prolonged

Prolonged

Measure fibrinogen (clottable and antigen)

Inhibitor screen: 1:1 mix

Prolonged time does not correct

Test for lupus anticoagulant and specific coagulation factor inhibitor(s)

Prolonged time corrects

Measure individual coagulation factor levels

History and physical exam suggest possible systemic bleeding disorder Discontinue all antithrombotics and other drugs that may interfere with tests Screening laboratory tests:

Prothrombin time (INR)

Partial thromboplastin time

Thrombin time

See Chapter 163 for evaluation of thrombocytopenia

B

Platelet count

VWD screen

Platelet function (PFA)

VWD screen

Platelet function (PFA)

Low

History and physical exam suggest possible systemic bleeding disorder Discontinue all antithrombotics and other drugs that may interfere with tests Screening laboratory tests:

C

Prothrombin time (INR)

Partial thromboplastin time

Thrombin time

Platelet count

Low VIII:C, VWF:Ag, and/or VWF:RCo

Abnormal

Perform VWF multimer analysis

Platelet aggregation in PRP

Perform specialized platelet studies

Abnormal

FIGURE 162-3.  A to C, Algorithms for clinical and laboratory approach to the diagnosis of a patient with a suspected systemic bleeding disorder (coagulopathy). The critical importance of a thorough personal and family history and physical examination is emphasized before initiating a laboratory work-up. INR = international normalized ratio; PFA = platelet function analyzer; PRP = platelet-rich plasma; VIII:C = factor VIII coagulant activity; VWF = von Willebrand factor; VWF:Ag = von Willebrand factor antigen; VWF:RCo = von Willebrand factor: ristocetin cofactor activity.

1122

CHAPTER 162  Approach to the Patient with Bleeding and Thrombosis  

TABLE 162-2 FEATURES OF COAGULATION FACTOR DEFICIENCIES DEFICIENCY

TYPICAL LABORATORY FINDINGS

APPROXIMATE PREVALENCE

Fibrinogen

Afibrinogenemia: TT ↑↑, aPTT ↑↑, PT ↑↑ Dysfibrinogenemia and hypofibrinogenemia: TT ↑, aPTT ↑, PT ↑↑

1 in 1 million

Prothrombin

TT normal, aPTT ↑, PT ↑

1 in 2 million

Factor V

TT normal, aPTT ↑, PT ↑

1 in 1 million

Combined factor V and VIII

TT normal, aPTT ↑, PT ↑

1 in 1 million

Factor VII

TT normal, aPTT normal, PT ↑

1 in 500,000

Factor X

TT normal, aPTT ↑, PT ↑

1 in 1 million

Factor XI

TT normal, aPTT ↑, PT normal

1 in 1 million

Factor XIII

TT, aPTT, and PT normal; specific assays required

1 in 2 million

Vitamin K–dependent coagulation factors

TT normal, aPTT ↑, PT ↑↑

90% of cases) Deep vein thrombosis and/or pulmonary embolism most common Mesenteric, cerebral vein thrombosis rare but characteristic Frequent family history of thrombosis Typically autosomal dominant First episode of thrombosis typically in young adulthood (25%), so mild hemophiliacs (with either factor VIII or IX deficiency) may have a normal PTT and yet still have serious bleeding complications when provoked with surgery. Therefore, factor VIII and IX levels must be obtained here to rule out the mild hemophilia, which would require prophylactic preoperative (as well as intraoperative and postoperative) replacement therapy. The bleeding time, a method to test the integrity of platelet-vascular interactions, is now largely an obsolete test; and the kinds of bleeding problems this patient describes (as well as the notable absence of pathologic bruising with trauma) argue strongly against a qualitative platelet abnormality. Liver function tests could be done but are unlikely to shed additional light on this diagnostic problem; furthermore, this patient appears to have an inherited, not acquired, coagulopathy. The possibility of von Willebrand disease is not unreasonable (even though it is usually autosomal dominant), although, again patients with von Willebrand disease tend to bleed superficially like patients with platelet problems. In any case, multimer analysis is not the first step to diagnosing von Willebrand disease: it should be preceded by a von Willebrand panel to assay levels of functional VWF, antigenic VWF, and factor VIII activity. 2. A 68-year-old man with coronary artery disease, diabetes mellitus, hypertension, and a history of stroke followed by seizures is now 3 days after an uncomplicated coronary artery bypass graft (CABG) surgery, and you are asked to consult for severe thrombocytopenia. The patient is on multiple medications for his chronic conditions and is now on postoperative subcutaneous heparin for thromboprophylaxis. There have been no bleeding problems postoperatively, and physical examination shows no signs of bleeding from external orifices and no ecchymoses or petechiae. The platelet count today is 5000, which triggered the STAT consult; it had been previously normal, including a count of 280,000 yesterday. Which of the following should be the first step? A . Examine the peripheral blood smear for platelet clumping. B. Stop heparin. C. Stop heparin and all other medications. D. Order STAT platelet transfusions. E. Take a thorough history and do a comprehensive physical examination.

Answer: A  The first step is to promptly determine whether the apparently sudden drop in platelet count is real or artifactual. The abrupt drop (when counts were normal as recently as 1 day ago), accompanied by the striking absence of symptoms or signs of bleeding in this patient raises a high level of suspicion for “pseudothrombocytopenia.” Seeing large clumps of platelets on the peripheral smear, which should be examined by the hematologist, would confirm the suspicion, which should be then promptly documented by ordering repeat platelet counts in both the customary purple-top tube containing ethylenediaminetetraacetic acid (EDTA) as the anticoagulant along with tubes containing other anticoagulants (heparin or citrate). (The platelet count in the EDTA tube should be again very low, but not in blood collected into test tubes with the other anticoagulants.) Occasionally, pseudothrombocytopenia occurs as a result of the presence of a nonphysiologic “cold agglutinin,” which clumps platelets in the test tube while sitting at room temperature before counting. This can be ruled out by having the laboratory do simultaneous platelet counts on a tube kept at 37° C versus one that has been intentionally left standing at room temperature: the former should have a normal platelet count if this is the case. These findings represent pseudothrombocytopenias in which the platelet counts are artifactually low and of no clinical consequence. These tests can be all done very quickly without stopping important medications, including heparin, and before undertaking further work-up and treatment of the thrombocytopenia if they remain necessary. 3. Which of the following statements is correct regarding the relationship between cancer and thrombosis? A . In a patient being treated for gastric cancer, secondary prevention following a first episode of venous thromboembolism is best provided by warfarin for 3 to 6 months at doses to target an international normalized ratio of 2.0 to 3.0. B. A first episode of apparently unprovoked deep vein thrombosis in a previously healthy individual requires comprehensive search for an occult malignancy, including computed tomography of the chest, abdomen, and pelvis. C. Pulmonary embolism in a patient being actively treated for cancer requires the placement of an inferior vena cava (IVC) filter to prevent recurrence. D. A patient with breast cancer who completed treatment 6 months ago and is now in complete remission, with no evidence of disease, is not at increased risk for venous thromboembolism. E. In a previously healthy individual, the spontaneous development of extensive IVC thrombosis is highly suggestive of an occult renal cell carcinoma. Answer: E  IVC thrombosis, a rare and dramatic form of venous thrombosis that can extend into the right atrium, is associated with renal cell carcinoma. Other characteristic (but not diagnostic) cancer-associated thrombosis syndromes include Budd-Chiari syndrome (hepatic vein thrombosis) or portal vein thrombosis in patients with myeloproliferative neoplasms and superficial, migratory thrombophlebitis (Trousseau syndrome) with occult cancers of the gastrointestinal tract and pancreatic cancer. Subcutaneous low-molecularweight heparin is superior to dose-adjusted warfarin for secondary prophylaxis after an episode of venous thromboembolism in a cancer patient, even when in remission, and it should be continued indefinitely. Although the incidence of occult malignancy in a previously healthy person who presents with unprovoked venous thromboembolism is definitely increased, most experts agree that a comprehensive history and physical examination, stool for occult blood, chest radiograph, blood counts and basic chemistries, and a mammogram in women is a sufficient initial screen for occult cancer, as long as any abnormalities picked up on that screen are more aggressively pursued. Increased risk for thrombosis in a patient with cancer continues for months or even years after complete remission has been achieved. An IVC filter is inadequate to prevent recurrent pulmonary embolism in patients with or without cancer.

CHAPTER 163 Thrombocytopenia  

1123

163  THROMBOCYTOPENIA CHARLES S. ABRAMS



DEFINITION

Thrombocytopenia is defined as a platelet count lower than the normal range, typically below 140,000/µL.  

EPIDEMIOLOGY

With the widespread use of the “routine CBC,” asymptomatic individuals with incidentally discovered mild thrombocytopenia are being increasingly

CHAPTER 163 Thrombocytopenia  

ABSTRACT

Thrombocytopenia can be caused by three general mechanisms: (1) increased destruction of circulating platelets; (2) impaired production of platelets by the bone marrow; or (3) increased splenic sequestration of circulating platelets, a passive rather than actively destructive process caused by expansion of the splenic platelet pool as a result of splenomegaly due to congestive (e.g., portal hypertension) or infiltrative (e.g., lymphoma or storage disease) etiologies. Destructive thrombocytopenias can be immune- or nonimmune-mediated. The former includes ITP and drug-induced thrombocytopenia; the latter can be caused by TTP, HUS, DIC, and ITP and intravascular devices. Before initiating an extensive workup for the cause of the thrombocytopenia, the peripheral blood smear should be examined for artifactual platelet clumping (“pseudothrombocytopenia”), an idiosyncratic in vitro phenomenon that should be especially suspected in a patient reported to have an unexpectedly low platelet count in the absence of any clinical bleeding. Mild thrombocytopenias can occur physiologically with the progression of a normal pregnancy (gestational thrombocytopenia) or congenitally.

KEYWORDS

idiopathic (immune) thrombocytopenic purpura (ITP) thrombotic thrombocytopenic purpura (TTP) hemolytic-uremic syndrome (HUS) disseminated intravascular coagulation (DIC) pseudothrombocytopenia

1123.e3

1124

CHAPTER 163 Thrombocytopenia  

recognized. The clinical significance of mild thrombocytopenia (platelet counts between 100,000/µL and 150,000/µL) in the absence of other CBC abnormalities can be difficult to ascertain. Many, and perhaps most of these individuals with mild chronic thrombocytopenia, represent outliers of the normal distribution of platelet counts. However, others have thrombocytopenia that represents the early manifestation of an unrecognized disease. In the absence of other signs, many of these individuals are diagnosed with idiopathic (immune) thrombocytopenic purpura (ITP)—a disease that will be discussed later in this chapter. After a decade, approximately 10% of patients with mild idiopathic thrombocytopenia will develop ITP or another autoimmune disorder. However, approximately 90% of the subjects will continue to manifest borderline thrombocytopenia without developing another disorder during this period of time. It can be concluded that most individuals with isolated and asymptomatic chronic mild thrombocytopenia simply represent the lower end of the normal platelet count distribution. Consequently, international consensus groups have set the threshold for the diagnosis of ITP at a platelet count of less than 100,000/µL.  

PATHOBIOLOGY

The causes of thrombocytopenia can be divided into three major categories: (1) increased destruction of platelets, such as seen in immune-mediated causes; (2) decreased production of platelets, usually due to an underlying bone marrow disorder; or (3) sequestration of platelets within the spleen, as occurs in conditions that cause splenomegaly (hypersplenism).

Increased Platelet Destruction

In the absence of splenomegaly, the presence of increased megakaryocytes in an otherwise normal bone marrow implies that thrombocytopenia is due to increased platelet destruction. An acute drop in a patient’s platelet count also implies that a peripheral destructive process is the likely cause. For example, a patient who develops abrupt thrombocytopenia during hospitalization most likely has a low platelet count caused by an infection or by the introduction of a new medication. Both nonimmune and immune processes can lead to a shortened platelet lifespan. Nonimmune reasons for the accelerated destruction of platelets include sepsis, disseminated intravascular coagulation (DIC), thrombotic thrombocytopenic purpura (TTP), hemolytic-uremic syndrome (HUS), preeclampsia or eclampsia, cardiopulmonary bypass, and giant cavernous hemangioma. Thrombocytopenia that occurs in these circumstances usually resolves with treatment of the underlying disorder, and platelet transfusions are rarely necessary. Thrombocytopenia due to TTP, HUS, or heparininduced thrombocytopenia (HIT) is more characteristically associated with microvascular occlusion or thrombosis than with bleeding (Table 163-1), so platelet transfusions are also rarely necessary for the management of these conditions. In addition, reports have noted the clinical deterioration of patients with TTP or HIT following platelet transfusion, leading to the controversial suggestion that platelets should not be given to most patients with these two particular disorders. Immune-mediated platelet destruction can result from drugs, alloimmune sensitization, or autoimmunity. Medications should always be considered as a possible cause of acute thrombocytopenia. The list of potential offending agents is long, but drugs with strong evidence of antibody-mediated platelet destruction include quinine, quinidine, sulfonamides (such as trimethoprimsulfamethoxazole), β-lactam antibiotics (both penicillins and cephalosporins), vancomycin, procainamide, and gold salts. In addition to stopping the offending medication, platelet transfusions might be required for severe thrombocytopenia.

Decreased Platelet Production

Decreased platelet production occurs in primary diseases of the bone marrow, such as acute leukemia and aplastic anemia; myelophthisic processes in which bone marrow is affected by metastatic carcinoma, fibrosis, or other clonal

TABLE 163-1 DISORDERS THAT CAUSE BOTH THROMBOCYTOPENIA AND THROMBOSIS Thrombotic thrombocytopenic purpura (TTP) Hemolytic-uremic syndrome (HUS) Heparin-induced thrombocytopenia (HIT) Disseminated intravascular coagulation (DIC) Paroxysmal nocturnal hemoglobinuria (PNH) Vasculitis (such as systemic lupus erythematosus) Antiphospholipid antibody syndrome (APS)

hematopoietic disorders; following toxin-induced (such as chemotherapy or ethanol exposure) or radiation therapy-induced myelosuppression; and during infections with viruses such as HIV, cytomegalovirus (CMV), Epstein-Barr virus (EBV), and varicella. Thrombocytopenia also occurs when normal megakaryocyte proliferation is impaired by myelodysplasia. Thrombocytopenia due to decreased platelet production is often accompanied by abnormalities of the white blood cells or red blood cells. This should be readily apparent by inspection of the CBC and review of the peripheral blood smear. When thrombocytopenia in this setting is accompanied by bleeding, the bleeding usually requires treatment by platelet transfusion. Additionally, when decreased platelet production is profound (less than 10,000/µL) and cannot be readily reversed, prophylactic platelet transfusion to prevent bleeding should be considered. However, as will be discussed in the following section, prophylactic platelet transfusion is problematic because of the short lifespan of platelets and the potential to develop alloantibodies that will limit the effectiveness of future platelet transfusions.

Platelet Sequestration

Approximately 30 to 40% of the circulating platelet mass is normally sequestered within the spleen. Enlargement of the spleen due to portal hypertension or infiltration is accompanied by expansion of the splenic platelet pool. Because hypersplenism by itself rarely causes a platelet count of less than 40,000/µL to 50,000/µL, bleeding due to thrombocytopenia from hypersplenism alone is unusual. In addition, if a patient with a large spleen has a platelet count of less than 40,000/µL, an additional etiology for the thrombocytopenia needs to be considered.  

CLINICAL MANIFESTATIONS

In the absence of qualitative platelet defects (Chapter 164), excessive bleeding does not occur in thrombocytopenic patients following trauma or surgery unless the platelet count is lower than 75,000/µL. In otherwise hemostatically normal patients, spontaneous hemorrhage typically does not occur with platelet counts above 30,000/µL. Patients with platelet counts of less than 5000/µL to 10,000/µL are at a high risk for spontaneous, life-threatening hemorrhage. However, there is no absolute threshold for spontaneous bleeding due to thrombocytopenia. Bleeding may occur at higher counts when fever, sepsis, severe anemia, and other hemostatic defects are present, or when platelet function is impaired by medication. It is sometimes said that patients with wet purpura (mucocutaneous bleeding with blood blisters) are at particularly high risk for sudden catastrophic bleeding events. Therefore, it is critical that the physician consider the gamut of possibilities when diagnosing and managing a patient with a low platelet count.  

DIAGNOSIS

Typically, the first step in diagnosing a patient with a low platelet count is to determine whether the thrombocytopenia is attributable to increased platelet destruction, decreased platelet production, or hypersplenism. See Figure 163-1 for an approach to evaluating patients with chronic thrombocytopenia. Because there is no easy test to differentiate among the three possibilities, clinical evaluation is critical. Therefore, a thorough history and physical examination, with attention to possible alternative explanations for thrombocytopenia, are mandatory. Particular attention should be paid to the duration of symptoms, which helps to determine whether the patient has acute or chronic thrombocytopenia. The clinician should also focus on the patient’s recent exposure to new medications that might induce thrombocytopenia as a side effect. Preliminary laboratory tests include a complete blood count (CBC) with a differential, along with an examination of the peripheral blood smear (Chapter 148). Abnormalities in the number or morphology of the leukocytes or erythrocytes may indicate a systemic inflammatory disorder or a pathology within the bone marrow; fragmentation of red cells may suggest a microangiopathy such as thrombotic thrombocytopenic purpura (TTP) or hemolyticuremic syndrome (HUS). Clumping of the platelets on the peripheral blood smear may indicate that the patient has pseudothrombocytopenia (Chapter 162). This anomaly is usually due to an antibody that binds to and thereby agglutinates platelets only in the presence of hypocalcemia induced by ethylenediaminetetraacetic acid (EDTA) in the blood collection tube, or when the blood sample is allowed to cool to room temperature. Repeating the platelet count in a blood sample that has been anticoagulated with heparin and kept warm until it is analyzed can help exclude this spurious diagnosis. Any additional testing to determine the cause of thrombocytopenia is based solely on the available clinical information derived from the history and examination of the patient. For example, testing for human immunodeficiency virus (HIV)

CHAPTER 163 Thrombocytopenia  

1125

Chronic thrombocytopenia Abnormal blood smear or abnormality of WBC or RBC suggests bone marrow disorder

Exclude pseudothrombocytopenia on peripheral blood smear Platelet count less than 100,000/μL

Large spleen and platelets > 40,000/μL suggests hypersplenism Immune mediated; primary or secondary ITP; drug induced thrombocytopenia Fragmented red cells suggest microangiopathy

Exclude pseudothrombocytopenia on peripheral blood smear Platelet count greater than 100,000/μL

Abnormal smear or abnormal WBCs and/or RBCs suggests bone marrow disorder Fragmented red cells suggest microangiopathy: proceed with differential diagnosis as in above algorithm

Stable; continue to follow CBC Follow CBC Platelet count less than 100,000/μL; see above algorithm

FIGURE 163-1.  Systematic approach to the evaluation of chronic thrombocytopenia. CBC = complete blood count; ITP = idiopathic (immune) thrombocytopenic purpura; RBC = red blood cells; WBC = white blood cells.

would be prudent in a patient with risk factors. A bone marrow aspirate and biopsy are typically not required for a patient with thrombocytopenia who has an otherwise normal CBC and peripheral blood smear. However, a bone marrow aspirate and biopsy should be considered in older patients and in patients whose thrombocytopenia is progressive or in whom standard therapy has not been effective.

TREATMENT  For patients with decreased platelet production, the cornerstone of treatment is platelet transfusion (Chapter 167). For patients with other specific causes, however, platelet transfusions may or may not be a beneficial therapeutic option.

Platelet Transfusions

Clinicians usually consider “wet” bleeding to be much more ominous than “dry” bleeding. Signs of wet bleeding include epistaxis, gingival bleeding, gastrointestinal bleeding, genitourinary bleeding, and bleeding around intravenous sites. Dry bleeding is defined as ecchymosis or petechiae (Fig. 163-2). Overt wet bleeding that is clearly due to thrombocytopenia is usually treated with platelet transfusion (Chapter 167). Prophylactic platelet transfusion for patients who are not bleeding is controversial. When making the decision whether to treat a nonbleeding patient with thrombocytopenia, the practitioner must consider the short lifespan of platelets (10 days), the 5- to 7-day shelf life of stored platelets, and the potential for the patient to develop transfusion-induced platelet alloantibodies. In patients undergoing treatment for acute leukemia, clinical trials have supported the practice of prophylactic platelet transfusions when the platelet count is below 5000/µL or 10,000/µL, rather than waiting until bleeding occurs in the severely thrombocytopenic patient. A1  The indications for prophylactic platelet transfusions before interventional procedures are predominantly based on clinical experience and expert opinions, rather than on clinical trials. Current guidelines recommend platelet transfusion in patients with therapy-induced hypoproliferative thrombocytopenia and a platelet count of less than 10,000/µL. Prophylactic platelet transfusion is also indicated for patients having elective central venous catheter placement with a platelet count less than 20,000/µL, and for patients having elective diagnostic lumbar puncture or major elective nonneuraxial surgery with a platelet count less than 50,000/µL (Table 163-2).1,2 Platelet transfusions should continue for

FIGURE 163-2.  Petechiae. Multiple pinpoint, nonblanching, erythematous macules found predominantly on skin of dependent extremities in patients with severe thrombocytopenia.

several days following procedures where the risk for bleeding or the complications resulting from bleeding are high. For extremely high-risk procedures that involve the nervous and ocular systems, the prophylactic platelet transfusions should be continued for at least 7 to 10 days. A donated unit of blood contains approximately 50 billion platelets. Infusing this number of platelets into a patient should increase his or her platelet count by 20,000/µL divided by the patient’s body surface area in square meters. Therefore, transfusion of one unit into an average-sized adult patient should increase the individual’s platelet count by approximately 10,000/µL to 12,000/µL. Bags of platelets used for transfusions are typically obtained by pooling platelets derived from four to six blood donors. Therefore, a “unit” at one hospital may be derived from more donors than a “unit” obtained at another hospital. Large quantities of platelets can be derived from a single donor by using apheresis technology (plateletpheresis) (Chapter 167). Because there is minimal

1126

CHAPTER 163 Thrombocytopenia  

TABLE 163-2 SUGGESTED MINIMUM PLATELET COUNTS BEFORE INVASIVE PROCEDURES VERY-HIGH-RISK PROCEDURES—75,000/µL TO 100,000/µL Neurosurgery Ocular surgery (except cataract extraction) Thyroid surgery Prostatectomy MODERATE-RISK PROCEDURES—50,000/µL Liver biopsy Dental extraction Most surgical procedures LOW-RISK PROCEDURES—30,000/µL Endoscopy with no biopsy Bronchoscopy with no biopsy Lumbar puncture (with scrupulous technique and no intrathecal injection) VERY-LOW-RISK PROCEDURES—NO PLATELET TRANSFUSIONS NECESSARY Bone marrow biopsy Cataract extraction

loss of red blood cells with this technique, plateletpheresis allows one individual to donate very large numbers of platelets (equivalent to the number of platelets obtained from 6 units of donated blood). As discussed in the following section, the risk for platelet alloimmunization is partially dependent on the number of individuals donating platelets to a patient. Therefore, single-donor plateletpheresis can minimize the risk of alloimmunization. Several trials have addressed how many platelets should be transfused into thrombocytopenic patients to maintain hemostasis and to minimize exposure to blood products (also see Chapter 167). The largest of these trials was the multicenter Platelet Dose (PLADO) Trial. A2  In this study, patients with thrombocytopenia were randomized to receive transfusions of low-dose (110 billion/ m2), medium-dose (220 billion/m2), or high-dose (440 billion/m2) platelets while recovering from hematopoietic stem cell transplantation. Patients who received the low-dose therapy required more frequent platelet transfusions. However, overall they received fewer transfused platelets over the course of the study. Importantly, bleeding complications were identical with all platelet transfusion strategies. This study demonstrates that low doses of platelets (110 billion/m2 of body surface area) are as safe as larger doses in patients with hypoproliferative thrombocytopenia who are receiving prophylactic platelet transfusion therapy. Several complications of platelet transfusion therapy merit mention (also see Chapter 167). First, some patients can become alloimmunized against platelet antigens and can become refractory to future platelet transfusion therapy (discussed in detail later). Second, bacterial contamination of stored platelets is a much more common complication than the infectious risk associated with red blood cell transfusions. Unlike red blood cells, which are stored frozen after being harvested from donated blood, platelets are sensitive to cool temperatures and therefore need to be stored at room temperature. This method of storage results in more bacterial overgrowth in the transfusion bags. Transfusion-associated bacterial sepsis is one of the most frequently reported causes of transfusion-induced mortality in the United States, and platelet transfusions are the most common blood products associated with sepsis. It should also be noted that room temperature storage of platelets contributes to their short shelf life of approximately 5 to 7 days. In contrast to the risk for bacterial infection, the risk for viral infection from platelet transfusions is no higher than it is for red blood cell transfusions. From a single unit of platelets, the risk for being exposed to HIV is less than 1 in 1.5 to 2 million, the risk for exposure to hepatitis B virus is less than 1 in 200,000, and the risk for exposure to hepatitis C virus is less than 1 in 1 to 2 million. Platelet transfusions have been considered contraindicated for thrombocytopenia caused by TTP or HIT. However, this recommendation is driven entirely by retrospective reports of thrombotic events that have occurred soon after platelet transfusions had been given to such patients.3 This recommendation is not based on randomized trials designed to address this issue. Because thrombotic events are a known complication of both TTP and HIT, it is difficult to determine whether the thrombi that occurred in these patients were due to the platelet transfusions or merely to the intrinsic prothrombotic risk for TTP and HIT. In any case, there is usually no indication for platelet transfusions in patients with TTP and HIT because thrombosis is a much greater risk than hemorrhage in these disorders.

The Transfusion-Refractory Patient

Many patients do not have an optimal increase in their platelet count following transfusion. Although there are various definitions of platelet refractoriness, it is simplest to base this definition on the timing of platelet loss following

transfusion. Patients who do not have the predicted rise in their platelet count (between 10,000/µL and 12,000/µL for every unit of donated platelets) should have their platelet count analyzed before their next platelet transfusion, 1 hour after that transfusion, and again 24 hours later. If their platelet count rises appropriately 1 hour after the transfusion but falls substantially 24 hours later, the patient has ongoing platelet consumption. This is often seen in patients with sepsis, DIC, severe active hemorrhage, or sometimes drug-mediated immune destruction of platelets. In these situations, the best therapy is to treat the underlying cause and to continue to support the patient with platelet transfusions as clinically indicated. Alternatively, some patients fail to have a significant increase in platelets even 1 hour after a transfusion. These patients may have (1) hypersplenism, (2) an autoantibody that eliminates not only endogenous platelets but also allogeneic donor platelets (as seen in patients with ITP or some patients with drug-mediated immune destruction of platelets), or (3) alloantibodies that react with antigens on the transfused platelets. Alloantibodies develop in approximately 20% of patients who are repeatedly exposed to platelet transfusions, and these patients present some of the most challenging management issues in transfusion medicine (Chapter 167). The alloantibodies may be directed against human leukocyte antigen (HLA) class I antigens (HLA-A and HLA-B) or against platelet-specific antigens present on the surface of the platelets. Because it appears that HLA class II antigens present on the surface of leukocytes are essential for the development of antibodies directed against HLA class I antigens, efforts to remove the contamination of leukocytes in platelet transfusion preparations can minimize alloantibody formation. Therefore, filters that trap leukocytes during transfusions are useful to prevent patients from becoming refractory to future platelet transfusions. After a patient develops alloantibodies against transfused platelets, the clinician and the blood bank should attempt to identify whether the alloantibody is directed against an HLA class I antigen or against a platelet-specific antigen. If the alloantibody can be identified, platelets derived from donors matched for an HLA class I antigen or the platelet-specific antigen can be used. Corticosteroids, intravenous immunoglobulin, splenectomy, and recombinant factor VIIa are probably of no value in maintaining hemostasis in platelet-refractory thrombocytopenic patients. Antifibrinolytic agents (e.g., ε-aminocaproic acid) might be helpful for patients with bleeding that is predominantly in the oral cavity or in the genitourinary tract. However, antifibrinolytic agents are contraindicated in patients with DIC (Chapter 166) because these agents can precipitate thrombotic events.

  SPECIFIC CAUSES OF THROMBOCYTOPENIA

Drug-Induced Thrombocytopenia  

PATHOBIOLOGY

Drug-induced thrombocytopenia is one of the most frequent causes of cytopenias evaluated by physicians.4,5 Typically, drug-induced thrombocytopenia is the result of an immune reaction elicited by either the drug or by one of its metabolites. Sometimes the drug is simply deposited on the platelet surface, and antidrug antibodies lead to accelerated platelet clearance, primarily in the spleen. More frequently, the drug binds to a protein on the platelet surface and induces expression of a neoantigen that is ultimately recognized by the immune system. In the absence of the drug or drug metabolites, this platelet neoantigen disappears, and the thrombocytopenia slowly resolves as new platelets are released from the bone marrow. Immune-mediated platelet destruction explains most drug-induced thrombocytopenias, but there are some exceptions. For example, chemotherapy and other bone marrow toxins can decrease platelet production and thereby induce thrombocytopenia by a mechanism independent of the immune system.  

CLINICAL MANIFESTATIONS

The thrombocytopenia induced by medications can have a mild, moderate, or even severe course. Almost all types of drug-induced thrombocytopenia predispose the patient to hemorrhagic but not thrombotic complications. The notable exception to this rule is heparin-induced thrombocytopenia (discussed separately in the next section). A common clinical scenario is an infected or otherwise critically ill patient who develops abrupt thrombocytopenia. In that situation, it can be difficult to discern whether the thrombocytopenia is the result of the recent illness (such as a bacterial or viral infection) or its treatment (such as an antibiotic or another unrelated medication). Another difficult diagnostic dilemma is to discern drug-induced thrombocytopenia from ITP. For this distinction, the patient’s medical history may be helpful. A gradually progressive thrombocytopenia is more consistent with ITP than with one induced by a medication. Conversely, the rapid onset of thrombocytopenia following the initiation of a new medication indicates that the drug is the most likely culprit. The time between the initiation of the offending

1127

CHAPTER 163 Thrombocytopenia  

drug and the development of thrombocytopenia has not been well documented for most medications. However, thrombocytopenia typically begins days to a few weeks after the administration of most medications with this type of toxicity. Presumably, this lag time represents the period required for the patient to develop an immune response against the drug-platelet complex. Medications that have been taken safely for years before the onset of thrombocytopenia are unlikely to be the offending agents. In addition to an assessment of the patient’s prescribed medications, a careful and comprehensive evaluation of the patient’s over-the-counter medications is necessary. Agents that contain quinine lead the list of often nonprescribed drugs that can induce life-threatening thrombocytopenia. Quinine is frequently contained in over-the-counter pills for leg cramps. Even the quinine in tonic water can induce severe thrombocytopenia (“gin and tonic purpura”) when ingested by certain patients, since most drug-induced, immune-mediated thrombocytopenias are not related to the dose of the offending drug.  

DIAGNOSIS

Some medications are much more likely to induce thrombocytopenia than others. Table 163-3 contains a partial list of drugs that frequently induce this type of toxicity. One particularly useful online database (www.ouhsc.edu/ platelets/ditp.html) lists and periodically updates the level of evidence for specific drugs that may induce thrombocytopenia. With the exception of tests for HIT, laboratory tests for drug-induced thrombocytopenia do not have widespread applicability. Usually, drug-induced thrombocytopenia is proved only in retrospect, when the platelet count improves after the discontinuation of the suspected medication. Even then, the diagnosis is not conclusive unless thrombocytopenia recurs after rechallenging the patient with the offending drug (a practice that is almost never recommended).

TREATMENT  The most efficacious method of treating drug-induced thrombocytopenia is to stop all suspected offending medications. Usually, no additional therapy is indicated. The thrombocytopenia typically begins to resolve without further intervention within days to a week of stopping the drug. The notable exceptions involve drugs with particularly long half-lives. Sulfonamides, quinine, and quinidine are notorious for inducing severe or even life-threatening thrombocytopenia. In patients with profound thrombocytopenia (95%

99%

30-70%

ELISA = enzyme-linked immunosorbent assay; PF4 = platelet factor 4.

exposure emphasize the importance of conducting a careful and critical history in establishing the diagnosis. HIT should be suspected in any patient who develops thrombocytopenia while on heparin therapy. It is important to note that the normal platelet count varies widely (140,000/µL to 450,000/µL), and some patients may have a substantial decrease in their platelet count but still remain within the normal range. Therefore, a greater than 50% decrease in the platelet count of a patient on heparin should raise the suspicion of this syndrome. HIT should also be suspected in any patient who develops a thrombotic event while on heparin therapy. Some have used the 4T scoring system to clinically assess the likelihood of HIT, as opposed to a different cause of thrombocytopenia. The 4T score uses a combination of platelet count fall, timing of platelet count fall, coexisting occurrence of thrombosis, and other potential causes of thrombocytopenia to arrive at clinical likelihood of HIT that is low probability, intermediate probability, or high probability. Although antibodies against the heparin-PF4 complex are almost always present in patients with HIT, these antibodies are also frequently present in heparin-treated patients who do not have this disorder (i.e., those patients with neither thrombocytopenia nor thrombosis). There are two general categories of laboratory assays for the diagnosis of HIT: functional assays and immunologic assays (Table 163-4). Functional assays analyze whether the combination of heparin and the patient’s plasma can induce normal platelets to aggregate or to secrete serotonin; these assays have very high specificity but relatively low sensitivity. Immunologic assays test the patient’s plasma for antibodies that bind to the heparin-PF4 complex; these assays have very high sensitivity but lack specificity. Some studies show a good correlation between the amount of anti-PF4 antibodies present (such as determined by the optical density of an ELISA-based immunoassay) and the chance that the patient has clinically meaningful HIT. A negative immunologic assay is extremely useful in excluding this diagnosis, and a positive functional assay is useful in confirming the diagnosis of HIT. Several days may elapse before the test results are available. Therefore, in practice, these laboratory assays provide only confirmatory information. Urgent clinical decisions should not be deferred until such test results are available. The timing of the thrombocytopenia with respect to the heparin exposure and the degree of the platelet drop are the most important pieces of information needed by the physician when evaluating a patient suspected of having HIT. Because platelets are consumed as they become activated, the clinical presentation of HIT is thrombocytopenia. However, it is unusual for the thrombocytopenia to actually cause bleeding in patients with HIT. Instead, HIT is a highly prothrombotic disorder. Both venous and arterial thromboses are common in HIT. Patients with HIT who do not have thrombosis on initial presentation still have a 20 to 30% chance of developing a thrombus within the next month. Nevertheless, the incidence of HIT-related clinical events is greatest immediately after diagnosis. More importantly, clinical studies report that the cessation of heparin alone frequently fails to prevent the development of new thrombotic events.

TREATMENT AND PREVENTION  If a patient is suspected of having HIT, all heparin should be stopped immediately. This includes subcutaneous injections of “minidose” heparin, heparin flushes of intravenous lines, and low-molecular-weight heparin; even heparin-coated intravenous catheters should be withdrawn. Alternative anticoagulation, such as the direct thrombin inhibitors argatroban or lepirudin (depending on renal and liver function), the low-molecular-weight heparinoid danaproid, or more recently, the direct thrombin inhibitor dabigatran,9 should be administered, at least until the platelet count normalizes. Although not as well studied (nor FDA approved), both bivalirudin and fondaparinux appear to be viable alternatives. There are reports of extremely small cohorts of HIT patients treated with

direct oral anticoagulants. However, this has not been studied well enough to safely recommend this practice. In contrast, low-molecular-weight heparin should not be used because this drug can cross-react with the pathologic HIT antibodies. Warfarin should also not be used initially in cases of acute HIT because of its delayed therapeutic effect and its association with venous limb gangrene. When the platelet count has returned to a normal level after an acute episode of HIT, however, warfarin can be slowly introduced at a dose of 5 mg or less daily and gradually increased to achieve an international normalized ratio (INR) of 2 to 3 for a duration of at least 4 to 6 weeks because HIT is a risk factor for subsequent venous thromboembolism. Because patients rarely become profoundly thrombocytopenic as a result of HIT alone, platelet transfusions are typically not required. In fact, some reports suggest that platelet transfusions can actually precipitate thrombotic complications, although this premise remains controversial. If low-molecular-weight heparin is used for prophylaxis instead of unfractionated heparin, the risk of HIT is much lower. In one interventional program, for example, the rate of HIT declined by more than 50%.10

Sepsis Along with the exposure to certain drugs, bacterial and viral infections are among the most common causes of acute thrombocytopenia in hospitalized patients.11 Acute thrombocytopenia can be caused by the deposition of antibodyantigen complexes on the platelet surface through an “innocent bystander” phenomenon. These antibody-coated platelets are then cleared from the circulatory system by Fc receptor–expressing macrophages in the spleen. Thrombocytopenia associated with infections can also be due to DIC (as discussed in Chapter 166). The treatment of sepsis-induced thrombocytopenia is directed at treating the underlying cause, along with the administration of platelet transfusions as clinically indicated.

Idiopathic (Immune) Thrombocytopenic Purpura  

DEFINITION

ITP is an autoimmune disorder caused by circulating antiplatelet autoantibodies. An ITP-like picture can also be found in patients with autoimmune diseases, such as in systemic lupus erythematosus (Chapter 250); in low-grade lymphoproliferative disorders, such as chronic lymphocytic leukemia (Chapter 174); and in HIV infection (Chapter 366).  

CLINICAL MANIFESTATIONS

ITP was originally thought to be a disease of young women. Although this description is appropriate for many individuals with this disorder, more recent data indicate that ITP can occur in patients of either sex and at any age. ITP is a chronic, recurring disorder in most adults with this disease. This is in stark contrast to pediatric patients, who usually suffer from acute ITP and rarely have the chronic variant of this disorder. In contrast to patients with coagulation factor deficiencies who present with bleeding deep within their tissues, individuals with ITP (or other platelet disorders) typically have excessive mucocutaneous bleeding. Consequently, the clinician should inquire whether the patient has noticed epistaxis, gingival bleeding, easy bruising, hematuria, melena, or hematochezia. Female patients should also be asked about inappropriate or excessive vaginal bleeding. The physical examination should pay particular attention to signs of mucocutaneous bleeding. The patient should be thoroughly examined for petechiae (see Fig. 163-2) and ecchymosis as well as for evidence of hemorrhage within the conjunctiva, retina, and central nervous system.  

DIAGNOSIS

In patients with ITP, the CBC is usually normal except for the thrombocytopenia, and their peripheral blood smear is remarkable only for a decreased number of platelets, some of which may be larger than normal. Splenomegaly is absent unless the ITP is due to an underlying disorder, such as lymphoma that is associated with splenomegaly. A bone marrow examination is usually not necessary in the absence of abnormalities other than thrombocytopenia on the CBC and peripheral blood smear. If performed, the bone marrow aspirate and biopsy of ITP patients typically show normal or increased numbers of megakaryocytes. However, the rest of their bone marrow is otherwise normal. These bone marrow findings are similar to what is observed in the bone marrows of patients who have other forms of destructive thrombocytopenia such as drug-induced thrombocytopenia. Antiplatelet antibody assays are insufficiently sensitive or specific and therefore should not be done at this time.

CHAPTER 163 Thrombocytopenia  

TREATMENT  Because platelet production is assumed to be increased in patients with ITP as a result of the immune-mediated accelerated platelet destruction, traditional therapy has focused on moderating this immune response. For most of the latter half of the 20th century, splenectomy and corticosteroids were the sole therapies for ITP. Corticosteroids in the form of either prednisone (1 mg/kg orally daily) or high-dose dexamethasone (4-day pulses of 40 mg intravenously per day every 28 days for several cycles) are both effective, with a few trials showing better results with dexamethasone, and others (including the largest trial) showing no difference in long-term response rates. A3  Although corticosteroids remain the first-line therapeutic modalities, other immunomodulating agents, such as intravenous immunoglobulin (IVIG), anti-Rh immunoglobulin (anti-D), cyclophosphamide, azathioprine, cyclosporine, mycophenolate mofetil, dapsone, interferon, and etanercept, can be useful as well. The basic mechanism of and rationale for ITP treatment has revolved around altering the immune system by corticosteroids, splenectomy, or immune modulators, such as IVIG or anti-D. More recently, rituximab, a “humanized” murine monoclonal antibody against CD20 (a B-cell antigen), has been added as an effective therapeutic option. Although rituximab has not been approved by the U.S. Food and Drug Administration (FDA) for the treatment of ITP, it is currently widely used off label in patients who are unresponsive to first-line therapy. Response rates vary significantly between studies, ranging from 28 to 44% in larger trials. In a study of 133 newly diagnosed adult ITP patients with followup of up to 4 years, the combination of rituximab (375 mg/m2 intravenously weekly for 4 weeks) plus dexamethasone (40 mg per day orally for 4 days) induced higher response rates and longer time to relapse than dexamethasone alone. A4  However, grade 3 to 4 adverse events were increased when rituximab was combined with dexamethasone. Although most ITP patients have a compensatory increase in megakaryopoiesis in response to their rapid platelet destruction, plasma derived from some ITP patients was unexpectedly found to inhibit platelet production. This prompted a re-evaluation of whether impaired megakaryopoiesis contributes to the development of thrombocytopenia in this disease. Thrombopoietin (TPO) is a potent megakaryocyte colony-stimulating factor, and along with other cytokines, it increases the size and number of megakaryocytes (Chapter 147). TPO levels are not markedly elevated in ITP, suggesting that supplemental TPO could help to increase platelet production and correct the thrombocytopenia. Peptides have been developed that bind to and activate the TPO receptor; these agents are called TPO receptor agonists or thrombopoiesis stimulating agents (TSAs). Because these recombinant drugs bear little structural similarity to native TPO, they do not trigger autoimmune anti-TPO antibodies. The FDA has approved two of these drugs, and several more are being used in other countries. One of these, called romiplostim (N-plate), is composed of several copies of a TPO receptor–binding peptide spliced into a recombinant antibody. This peptide agonist competes with TPO for binding to the TPO receptor and activates the receptor in an identical fashion as would endogenous TPO. Another FDA-approved TPO receptor agonist is eltrombopag (Promacta). It is an oral drug that activates the TPO receptor by binding to the receptor’s transmembrane region. Both subcutaneously administered romiplostim and orally administered eltrombopag are capable of increasing platelet counts in approximately 70% of patients with ITP. Remarkably, these drugs can also increase the platelet counts of patients with ITP that are refractory to other treatment modalities, including splenectomy. However, adverse events, including bone marrow fibrosis and thromboembolism, have been reported. It should be noted that the drugs’ effects disappear soon after their discontinuation.

General Principles of ITP Therapy

Initial management of ITP is guided by both symptoms and platelet count. Asymptomatic patients with platelet counts of greater than 30,000/µL can be followed without treatment. If the patient is bleeding or has a platelet count of less than 30,000/µL, initial treatment with corticosteroids is recommended (Table 163-5). Refractory patients may require splenectomy, other immunosuppressive medications, or one of the new thrombopoiesis-stimulating agents (Table 163-6). Splenectomy has a long history of success in this disorder, and durable complete response rates are approximately 66 to 70%. Approximately half of the remaining patients who do not have normal platelet counts following splenectomy achieve a partial response that is clinically meaningful. Unfortunately, 10 to 15% of patients derive no benefit from splenectomy. There is no test to predict whether a particular patient will respond to this treatment, although initial steroid responders tend to have a better outcome. It should be noted that splenectomy is associated with small long-term risks of both infections and thrombosis. ITP patients with severe thrombocytopenia (3.0 plus low-dose ASA (81 mg), as long as benefits outweigh risks E. Stop Coumadin after 6 months, with careful monitoring for VTE recurrence Answer: A  The occurrence of β2-GPI antibody and LAC with deep venous thrombosis is consistent with the antiphospholipid syndrome (APS). Individuals with APS are at increased risk for recurrent venous thromboembolism, and thus anticoagulation should be continued as long as the benefits of anticoagulation outweigh bleeding risk. Several studies have demonstrated that standard Coumadin to an INR 2.0-3.0 is adequate VTE prophylaxis for individuals with APS, with no added benefit to maintain INR above 3 for patients with antiphospholipid syndrome. Aspirin, other than in the setting of pregnancy, appears to provide no added benefit in clot protection in individuals with APS.

CHAPTER 166  HEMORRHAGIC DISORDERS  

166  HEMORRHAGIC DISORDERS: DISSEMINATED INTRAVASCULAR COAGULATION, LIVER FAILURE, AND VITAMIN K DEFICIENCY ANDREW I. SCHAFER

  DISSEMINATED INTRAVASCULAR

COAGULATION



DEFINITION

Disseminated intravascular coagulation (DIC), also referred to as consumptive coagulopathy or defibrination, is caused by a wide variety of serious disorders (Table 166-1). In most patients, the underlying process dominates the clinical picture, but in some cases (e.g., occult malignant neoplasm, envenomation), DIC may be the initial or predominant manifestation of the disorder. DIC never occurs in isolation, without an inciting cause.  

PATHOBIOLOGY

DIC is pathophysiologically a primarily thrombotic process. However, its clinical manifestation may be widespread hemorrhage in acute cases. The basic pathophysiologic mechanism (Fig. 166-1), regardless of cause, is entry into the circulation of procoagulant substances that trigger systemic activation of

TABLE 166-1 MAJOR CAUSES OF DISSEMINATED INTRAVASCULAR COAGULATION INFECTIONS Gram-negative bacterial sepsis Other bacteria, fungi, viruses, Rocky Mountain spotted fever, malaria IMMUNOLOGIC REACTIONS Transfusion reactions (ABO incompatibility) Transplant rejection OBSTETRIC COMPLICATIONS Amniotic fluid embolism Retained dead fetus Abruptio placentae Toxemia, preeclampsia Septic abortion MALIGNANT NEOPLASMS Pancreatic carcinoma Adenocarcinomas Acute promyelocytic leukemia Other neoplasms LIVER FAILURE ACUTE PANCREATITIS ENVENOMATION RESPIRATORY DISTRESS SYNDROME TRAUMA, SHOCK Brain injury Crush injury Burns Hypothermia or hyperthermia Fat embolism Hypoxia, ischemia Surgery VASCULAR DISORDERS Giant hemangioma (Kasabach-Merritt syndrome) Aortic aneurysm Vascular tumors

1147

the coagulation system and platelets. This results in disseminated deposition of fibrin-platelet thrombi within the microvasculature.1 In most cases, the procoagulant stimulus is tissue factor, a lipoprotein that is not normally exposed to blood. In DIC, tissue factor gains access to blood by tissue injury, its elaboration by malignant cells, or its expression on the surface of monocytes and endothelial cells by inflammatory mediators. Components of the inflammatory response and the coagulation system are reciprocally activated in some forms of DIC, such as sepsis. Tissue factor triggers generation of the coagulation protease thrombin, which induces fibrin formation and platelet activation. In some specific cases of DIC, procoagulants other than tissue factor (e.g., a cysteine protease or mucin in certain malignant neoplasms) and proteases other than thrombin (e.g., trypsin in pancreatitis, exogenous enzymes in envenomation) provide the procoagulant stimulus. In acute, uncompensated DIC, coagulation factors are consumed at a rate in excess of the capacity of the liver to synthesize them, and platelets are consumed in excess of the capacity of bone marrow megakaryocytes to release them. The resulting laboratory manifestations under these circumstances are a prolonged prothrombin time (PT), activated partial thromboplastin time (aPTT), and thrombocytopenia. Increased fibrin formation in DIC stimulates a heightened process of secondary fibrinolysis, in which plasminogen activators generate plasmin to digest fibrin (and fibrinogen) into fibrin(ogen) degradation products (FDPs). FDPs are potent circulating anticoagulants that further contribute to the bleeding manifestations of DIC. Intravascular fibrin deposition can cause fragmentation of red blood cells and lead to the appearance of schistocytes in blood smears; however, frank microangiopathic hemolytic anemia is unusual in DIC. Occlusive microvascular thrombosis in DIC can compromise the blood supply to some organs and lead to multiorgan failure, particularly when it is accompanied by systemic hemodynamic, inflammatory, and metabolic derangements.

Underlying Causes

DIC always has an underlying cause that generally must be identified and eliminated if the coagulopathy is to be managed successfully. The development of DIC in many of these disorders is associated with an unfavorable outcome. Infection is the most common cause of DIC. The syndrome is particularly associated with gram-negative or gram-positive sepsis (Chapter 100), although it can be triggered by a variety of other bacterial, fungal, viral, rickettsial, and protozoal microorganisms. The placenta and uterine contents are rich sources of tissue factor and other procoagulants that are normally excluded from the maternal circulation; a spectrum of clinical manifestations of DIC may accompany obstetric complications when this barrier is breached, especially in the third trimester. These syndromes range from acute, fulminant, and often fatal DIC in amniotic fluid embolism to chronic or subacute DIC with a retained dead fetus. Other obstetric problems associated with DIC include abruptio placentae, toxemia, and septic abortion. Chronic forms of DIC are caused by a variety of malignant neoplasms, particularly pancreatic cancer (Chapter 185) and mucin-secreting adenocarcinomas of the gastrointestinal tract (Chapter 184), in which thrombotic rather than bleeding manifestations predominate. Treatment with all-trans-retinoic acid has greatly reduced the incidence of severe DIC in patients with acute promyelocytic leukemia (Chapter 173). It is not known whether liver failure (see later) can cause DIC or whether its coexistence merely exacerbates intravascular coagulation because of impaired clearance of activated clotting factors, plasmin, and FDPs. Snake venom contains a variety of substances that can affect coagulation and endothelial permeability. Bites from rattlesnakes and other vipers can induce profound DIC by introduction of these exogenous toxins and release of endogenous tissue factor through tissue necrosis. The likelihood and degree of DIC caused by trauma, surgery, and shock (Chapter 98) are related to the extent of tissue damage and the organs involved. The brain is a particularly rich source of tissue factor, so traumatic brain injury (Chapter 371) can precipitate acute DIC. Large aortic aneurysms (Chapter 69), giant hemangiomas, and other vascular malformations can cause subclinical or clinical, chronic DIC that is initiated locally within the abnormal vasculature but can “spill” into the systemic circulation.  

CLINICAL MANIFESTATIONS

The clinical manifestations of DIC are determined by the nature, intensity, and duration of the underlying stimulus. The coexistence of liver disease exacerbates DIC of any cause. Low-grade DIC is often asymptomatic and diagnosed only by laboratory abnormalities. Thrombotic complications of DIC occur most often with chronic underlying diseases, as exemplified by Trousseau

CHAPTER 166  HEMORRHAGIC DISORDERS  

ABSTRACT

Mixed hemorrhagic disorders that involve deficiencies in multiple coagulation factors as well as platelets include “disseminated intravascular coagulation (DIC)” and the coagulopathy of liver failure. Vitamin K deficiency causes deficiencies of specific vitamin K–dependent coagulation factors II (prothrombin), VII, IX, and X, as well as specific vitamin K–dependent prothrombotic factors protein C and protein S, but does not involve platelets. DIC, also called consumptive coagulopathy, is a syndrome caused by a wide variety of disorders that all have in common their activation of the clotting system at different points in the coagulation cascade, leading to increased fibrin deposition in the microvasculature. So DIC is primarily a thrombotic disorder and may in fact present in its chronic form as thrombosis or occlusive organ dysfunction. The bleeding that more typically accompanies DIC, especially in acute cases, results from consumption of all clotting factors, consumption of platelets, and activation of the fibrinolytic system. The coagulopathy of liver failure is likewise a complex disorder that has both prothrombotic and antithrombotic components. Therefore, while most patients with liver failure have a bleeding diathesis, thrombosis can also occur depending on the balance of these opposing forces. Complete resolution of the coagulopathy of DIC and liver failure, which often coexist, depends on the ability to control the underlying disease.

KEYWORDS

disseminated intravascular coagulation consumptive coagulopathy coagulopathy of liver failure vitamin K deficiency thrombocytopenia coagulation factor deficiencies

1147.e1

1148

CHAPTER 166  HEMORRHAGIC DISORDERS  

Stimulus

Elaboration/release of tissue factor Generation of thrombin

Thrombocytopenia

Bleeding

Platelet activation

Secondary fibrinolysis

Intravascular fibrin

Fibrin degradation products

Depletion of coagulation factors

Thrombosis

Bleeding

Tissue necrosis FIGURE 166-1.  Pathophysiologic process of bleeding, thrombosis, and ischemic manifestations in patients with disseminated intravascular coagulation.

FIGURE 166-2.  Disseminated intravascular coagulation resulting from staphylococcal septicemia. Note the characteristic skin hemorrhage ranging from small purpuric lesions to larger ecchymoses. (From Forbes CD, Jackson WF. Color Atlas and Text of Clinical Medicine. 3rd ed. London: Mosby; 2003.)

syndrome in cancer (Chapter 73). The thrombotic phenotype of DIC, which occurs prototypically in association with solid tumor malignancies, can manifest as large vessel venous or arterial thrombosis, or as ischemic organ dysfunction due to microvascular thrombosis, or even as a thrombotic microangiopathy.2 DIC can present as acrocyanosis and gangrene of the digits in critically ill, hemodynamically compromised patients receiving vasopressors. Hemorrhagic necrosis of the skin (Fig. 166-2) and purpura fulminans may also be manifestations of DIC.3 Bleeding is the most common clinical finding in acute, uncompensated DIC. Bleeding can be limited to sites of intervention or anatomic abnormalities, but it tends to be generalized in more severe cases, including widespread ecchymoses and diffuse oozing from mucosal surfaces and orifices. In these cases, causes of which are exemplified by major trauma, abruptio placentae, or acute promyelocytic leukemia, the precipitating trigger is also a prothrombotic activation of the coagulation system but a fulminant, uncontrolled fibrinolytic response creates a bleeding or so-called fibrinolytic phenotype.4  

DIAGNOSIS

The laboratory diagnosis of severe, acute DIC is not usually difficult.5 Consumption and inhibition of the function of clotting factors cause prolongation of the PT, aPTT, and thrombin time. Consumption of platelets causes thrombocytopenia. Secondary fibrinolysis generates increased titers of FDPs, which can be measured by latex agglutination or D-dimer assays. Some schistocytes may be seen in the peripheral blood smear, but this finding is neither sensitive nor specific for DIC. Chronic or compensated forms of DIC are more difficult to diagnose, with highly variable patterns of abnormalities in “DIC screen” coagulation tests. Increased D-dimers and a prolonged PT are generally more sensitive measures than are abnormalities of the aPTT and platelet count. Overcompensated synthesis of consumed clotting factors and platelets in some

chronic forms of DIC may actually cause shortening of the PT and aPTT or thrombocytosis (or both), even though elevated levels of D-dimers indicate secondary fibrinolysis in such cases. The most difficult differential diagnosis of DIC occurs in patients who have coexisting liver disease. The coagulopathy of liver failure (see later section on liver failure) is often indistinguishable from that of DIC, partly because advanced hepatic dysfunction is in fact accompanied by a state of DIC. In liver failure, the combination of decreased synthesis of clotting factors, impaired clearance of activated clotting factors, secondary fibrinolysis, and thrombocytopenia from portal hypertension and hypersplenism may make the coagulopathy virtually impossible to differentiate from DIC. The coagulopathy associated with postpartum hemorrhage is rarely caused by DIC alone; there may be also varying proportions of dilutional coagulopathy, localized consumption, and increased fibrinolysis.6 Thrombotic microangiopathies including thrombotic thrombocytopenic purpura and hemolytic-uremic syndrome, the syndrome of “hemolysis, elevated liver enzymes, and low platelet count” (HELLP) in obstetric patients, and other forms of platelet consumption and thrombocytopenia (Chapter 163) are not accompanied by activation of clotting factors or secondary fibrinolysis; therefore, the PT, aPTT, thrombin time, and D-dimers are generally normal in these disorders. Schistocytes, often with frank hemolysis, are much more prominent in the peripheral smear in thrombotic thrombocytopenic purpura and hemolytic-uremic syndrome (Chapter 163) than in DIC. One situation where DIC and thrombotic microangiopathy can overlap and even coexist is in patients with cancer-associated thrombotic microangiopathy. These patients typically have bone marrow involvement and systemic microvascular metastases, either in the setting of known metastatic cancer or unexpectedly on presentation.

TREATMENT  Successful treatment of DIC (Table 166-2) requires that the underlying cause be identified and eliminated. All other therapies, including hemodynamic support, replacement of coagulation factors and platelets, and pharmacologic inhibitors of coagulation and fibrinolysis, are just temporizing measures. Because of the difficulty of testing this complex syndrome with multiple causes by controlled, randomized clinical trials, current guidelines for treatment are generally not based on high-grade evidence.7,8 In many patients with asymptomatic, self-limited DIC who have only laboratory manifestations of the coagulopathy, no treatment may be necessary. In patients with DIC who are actively bleeding or who are at high risk for bleeding, the blood component treatments of choice are transfusions of platelets to improve the thrombocytopenia and fresh-frozen plasma to replace all consumed coagulation factors and to correct the prolonged PT and aPTT. In the absence of robust evidence-based guidelines for the treatment of DIC, an international consensus group recommended VTE prophylaxis in critically ill DIC patients, switching from pharmacologic to mechanical prophylaxis in case of bleeding, and maintaining platelets at greater than 20,000. Some patients who have particularly profound hypofibrinogenemia, the additional transfusion of cryoprecipitate, a plasma concentrate that is enriched in fibrinogen, may be useful.

CHAPTER 166  HEMORRHAGIC DISORDERS  

TABLE 166-2 TREATMENT OF DISSEMINATED INTRAVASCULAR COAGULATION Identify and eliminate the underlying cause No treatment if mild, asymptomatic, and self-limited Hemodynamic support, as indicated, in severe cases Blood component therapy Indications: active bleeding or high risk for bleeding Fresh-frozen plasma Platelets In some cases, consider cryoprecipitate, antithrombin III Drug therapy Indications: heparin for DIC manifested by thrombosis or acrocyanosis and without active bleeding; antifibrinolytic agents are generally contraindicated except with life-threatening bleeding and failure of blood component therapy DIC = disseminated intravascular coagulation.

The theoretical concern that these blood products could “fuel the fire” and exacerbate the DIC has not been supported by clinical experience. The use of pharmacologic inhibitors of coagulation and fibrinolysis in DIC is controversial. Heparin is of theoretical benefit because it blocks thrombin activity and quenches intravascular coagulation and the resultant secondary fibrinolysis. In practice, heparin might exacerbate the bleeding tendency in acute DIC. Heparin is usually reserved for special forms of DIC, including those manifested by thrombosis or acrocyanosis and forms that accompany cancer, vascular malformations, retained dead fetus, and possibly acute promyelocytic leukemia, in which active bleeding is not present. A meta-analysis of randomized controlled trials of anticoagulant therapy specifically in patients with sepsis and sepsis-induced DIC showed an increased risk of bleeding but significant reduction in mortaility. (In the study, anticoagulation had no benficial effects on mortality in the absence of DIC). A1  In cases of DIC in which thrombosis or acral ischemia predominates, unfractionated heparin should be used by continuous infusion because of its short half-life and reversibility in the event of increased bleeding. Monitoring of the aPTT in the presence of DIC may be problematic, so heparin infusion in this setting should be followed mainly by clinical response and improvement in results of other tests of coagulation (e.g., thrombocytopenia). Antifibrinolytic agents, including ε-aminocaproic acid and tranexamic acid, are generally contraindicated in DIC. By blocking the secondary fibrinolytic response to DIC, these drugs cause unopposed fibrin deposition and may precipitate thrombosis. Antifibrinolytic agents may be effective in decreasing life-threatening bleeding in DIC, however, particularly in extreme cases in which aggressive blood component replacement fails to control the hemorrhage; in such situations, simultaneous infusion of low doses of heparin is needed to reduce the risk for thrombosis. Recombinant human activated protein C (rhAPC) has not been confirmed to reduce mortality in patients with septic shock. A2  There is insufficient evidence to support the use of antithrombin III in patients with sepsis and DIC.9 Recombinant human soluble thrombomodulin is under active investigation for management of DIC in different settings.10 Therefore, rhAPC was withdrawn by its manufacturer in 2011. Although recombinant tissue factor pathway inhibitor, antithrombin III concentrates, and human soluble thrombomodulin may have some efficacy in improving laboratory parameters of DIC, the overall survival benefit with these agents in patients with DIC has not been demonstrated.

  LIVER FAILURE

Bleeding complications in patients with advanced liver disease (Chapters 144 and 145) can be severe and even fatal and directly account for about 20% of the deaths associated with hepatic failure. The extent of the bleeding tendency depends on the severity and type of liver disease involved. About one third of deaths in patients undergoing liver transplantation are attributable to perioperative hemorrhage.  

PATHOBIOLOGY

The pathophysiologic mechanism of bleeding in liver failure is multifactorial.11,12 The complexity of the systemic coagulopathy of hepatic disease is not surprising inasmuch as the liver is the principal organ site for the synthesis of almost all the coagulation and fibrinolytic factors as well as their inhibitors. (Hepatocytes synthesize all the clotting factors except for von Willebrand factor (vWF), which serves as a physiologic “carrier” and stabilizer of Factor VIII; so the elevated levels of vWF that are seen in liver disease can enable near-normal circulating levels of F VIII despite diminished synthesis of the

1149

latter.) Thus, the abnormal hemostasis of liver failure is accompanied by opposing antithrombotic and prothrombotic alterations. Unfortunately, currently available laboratory tests of coagulation cannot measure the net effect of the complex, incongruous hemostatic forces that are at work at any given time on clinical bleeding, or thrombotic risk in a patient with liver disease. For example, the prothrombin time (PT)/INR is still accepted as a valid prognostic marker of liver failure, but can no longer be considered a reliable indicator of bleeding tendency in these patients. Although the degree of prolongation of the PT/INR was commonly used in the past to guide treatment with freshfrozen plasma (FFP) or prothrombin complex concentrate (PCC), it is now recognized that it is not a useful predictor of bleeding after surgery or an invasive procedure.13 This explains why having certain prolongations of the PT/INR and activated partial thromboplastin time (aPTT) as a function of liver disease cannot be regarded as being functionally “auto-anticoagulated:” this notion is a seriously misleading fallacy. The opposing hemostatic forces at work due to liver failure include (1) decreased total levels of circulating clotting factors (caused by consumption of the factors), (2) hypo- or dysfibrinogenemia, (3) enhanced fibrinolysis, and (4) increased levels of fibrin(ogen) degradation products (FDPs) which act as circulating anticoagulants, all promoting bleeding, with simultaneously (1) increased levels of activated clotting factors (caused by impaired hepatic clearance), (2) decreased levels of the physiologic anticoagulant proteins, antithrombin III (AT III), protein C and protein S (caused by consumption of those proteins), (3) elevated levels of von Willebrand factor and Factor VIII, and (4) low plasminogen, all promoting thrombosis. Two factors that may tip the balance more toward a bleeding than a thrombotic phenotype in most of these patients are: (1) thrombocytopenia, and (2) anatomically vulnerable sites of bleeding in the GI tract in liver failure patients, such as esophageal and gastric varices, gastritis (congestive gastropathy), and hemorrhoids. There is an increased incidence of deep vein thrombosis and/or pulmonary embolism in patients with chronic liver disease (ranging from 0.5 to 6% in different studies). There is an even higher risk of portal vein thrombosis (ranging from 7 to 18% in different series).14 Quantitative and qualitative abnormalities of platelets also contribute to the bleeding diathesis of liver failure. Congestive splenomegaly secondary to portal hypertension causes increased pooling of platelets in the spleen (hypersplenism). The resultant thrombocytopenia, the degree of which generally correlates with spleen size, rarely causes a reduction in the platelet count to less than 30,000/µL. In alcoholic patients, suppression of bone marrow thrombopoiesis by the acute toxic effects of alcohol or folate deficiency may contribute to the thrombocytopenia. Qualitative platelet abnormalities have also been described in patients with liver disease. Liver transplantation (Chapter 145) poses special problems to the coagulation system. During the anhepatic stage of surgery, which lasts about 2 hours, the complete cessation of synthesis of coagulation factors causes further prolongation of the PT and aPTT. Release of tissue plasminogen activator from the newly grafted liver leads to increased fibrinolysis and transient exacerbation of bleeding risk in the postoperative period.  

CLINICAL MANIFESTATIONS

The most common hemorrhagic complication of liver disease is gastrointestinal bleeding, which is usually caused by anatomic abnormalities and exacerbated by the systemic coagulopathy of liver failure. Bleeding from other mucosal sites, extensive ecchymoses, or more serious hemorrhage into the retroperitoneum or central nervous system generally indicates more significant derangements of the coagulation system. The severe coagulopathy in patients with liver disease makes liver biopsy a potentially hazardous procedure. The PT and platelet count may be the best guides to bleeding risk, but they too lack reliability as predictors of risk of bleeding with liver biopsy. In general, liver biopsy can be performed safely if the platelet count is higher than 50,000/µL. The American Association for the Study of Liver Diseases has concluded that the bleeding risk-benefit ratio of liver biopsy must be carefully considered on a case-by-case basis.  

DIAGNOSIS

Although the PT and aPTT are often prolonged in patients with advanced liver disease, the PT tends to be a more sensitive assay early in the course; a disproportionate prolongation of the aPTT should raise suspicion of a coexisting coagulation abnormality, such as a lupus anticoagulant or clotting factor inhibitor. A prolonged PT is also a useful prognostic indicator of poor outcome in patients with cirrhosis, acute acetaminophen hepatotoxicity, and acute viral hepatitis; in acute viral hepatitis, it is a better index of prognosis than are

1150

CHAPTER 166  HEMORRHAGIC DISORDERS  

serum albumin and transaminases. A disproportionate prolongation of the thrombin time should suggest the presence of dysfibrinogenemia. Hypersplenism (Chapter 159), possibly associated with nutritional folate deficiency or the acute toxic effects of alcohol on bone marrow, often causes mild to moderate thrombocytopenia in patients with liver disease. Thrombocytopenia in liver disease can also be due to impaired synthesis of thrombopoietin, the primary megakaryocytopoietic growth factor (Chapter 147). However, consideration should be given to other coexisting causes of thrombocytopenia if the platelet count is significantly less than 30,000/µL. The coagulopathy of liver failure is often indistinguishable from that of DIC, in part because some degree of DIC is in fact a necessary accompaniment of advanced liver disease. In general, patients with DIC have more marked decreases in levels of factor VIII and increases in D-dimer than do patients with liver failure. Global coagulation tests, such as thrombelastography (ROTEM), thrombin generation assay, and clot waveform analysis (Chapter 162), would seem to lend themselves ideally to assessing bleeding or thrombotic risk and monitoring the overall hemostatic state of a complex coagulopathy like the one encountered in patients with liver disease. While ROTEM is increasingly used perioperatively in liver transplantation, the clinical utility of these global coagulation assays in chronic liver disease has not yet been validated.

TREATMENT  Given more recognition that simple clinical coagulation tests like the PT/ INR are of little value in predicting bleeding problems in patients with liver disease, and the continued dearth of evidence that blood component transfusions can predictably avert bleeding complications based on certain cutoff values of PT/INR or platelet count, there has been a significant movement toward more restrictive transfusion strategies in these patients. This shift in practice has been driven not only by the lack of proven benefit in preventing bleeding, but also the increasing recognition that attempting to correct abnormal laboratory values with therapeutic transfusions may paradoxically exacerbate bleeding in some cases by further expanding intravascular volumes and increasing portal pressures. This may, for example, lead to increased risk of variceal bleeding. Restrictive transfusion policies are now widely used in liver transplantation, where they have led to significant reductions in intraoperative blood loss and red cell transfusion requirements. By reducing central venous pressures, and consequently portal pressures, even performing intraoperative phlebotomy has decreased blood loss with liver transplantation.15 For these reasons, routine transfusion of FFP for prophylaxis of bleeding in coagulopathic liver disease patients is not recommended because patients are unlikely to benefit. Response to FFP is unpredictable and usually does not fully normalize the PT/INR even when multiple units are transfused. This is in part due to the fact that FFP and especially stored thawed plasma itself often does not have an INR of less than 1.3. Some institutional guidelines have recommended empirically giving an intravenous dose of vitamin K, which may slightly reduce the PT/INR in liver failure, as well as transfusing cryoprecipitate if the plasma fibrinogen is less than 100 mg/dL (as a concentrated source of fibrinogen) in patients with or without active bleeding who are undergoing invasive procedures. There is little experience with the efficacy of four-factor prothrombin complex concentrate (PCC, Kcentra), which is currently approved for urgent reversal of vitamin K antagonists. Transfusing plasma (10 mL/Kg) is usually reserved for actively bleeding patients who are not responding to these treatments and platelet transfusions if indicated. In addition to the uncertainty regarding platelet count thresholds for platelet transfusion in patients with liver disease, there are other considerations in those who have portal hypertension with congestive splenomegaly and hypersplenism. Normally, about one third of the total peripheral platelet pool is sequestered in the spleen at any given time. In a patient with hypersplenism, platelet sequestration in the spleen can expand to as much as 90%, depending on the size of the spleen. Much of this splenic platelet pool can be mobilized into the circulation with catecholamines. It is unknown whether this phenomenon also occurs in response to endogenous catecholamines, but it is plausible that the stress of a hemostatic challenge could induce a transient increase in platelet counts, making the baseline level of thrombocytopenia in these patients a significant underestimation of what actually occurs in vivo. Related to the potentially marked expansion of the splenic platelet pool with hypersplenism, transfusion of a given amount of platelets often does not produce the expected “bump” in platelet count, and more units of platelets might be needed to achieve a calculated recovery in peripheral blood. The use of a thrombopoietin-receptor agonist to increase the platelet count is contraindicated at this time. A randomized trial using eltrombopag 75 mg daily versus placebo to increase the platelet count of thrombocytopenic patients (platelets 100,000 per µL) to allow the full-dose initiation and maintenance of these antiviral drugs, leading to improved sustained virologic response. A4  However, improved treatment today for HCV with direct-acting antivirals may obviate the need for raising the platelet count. Because of the increased risk of thrombosis, as well as bleeding, in patients with chronic liver disease, prophylactic anticoagulation, especially when hospitalized or immobilized, should not be withheld. Prophylactic doses of lowmolecular-weight heparin are safe to use. If there is coexisting kidney disease that precludes its use, prophylactic doses of unfractionated heparin or warfarin can be substituted, although monitoring of the latter may be complicated by a preexisting prolonged INR. In patients who are actively bleeding or have a very low platelet count, mechanical thromboprophylaxis can be substituted for pharmacologic thromboprophylaxis.

  VITAMIN K DEFICIENCY

Vitamin K is required for γ-carboxylation of glutamic acid residues of the procoagulant factors II (prothrombin), VII, IX, and X and the anticoagulant factors protein C and protein S via the enzyme, γ-glutamyl carboxylase (GGCX). This post-translational modification normally renders these proteins functionally active in coagulation. The PT is more sensitive than the aPTT in detecting vitamin K deficiency states because factor VII, the only vitamin K–dependent factor that is in the extrinsic pathway of coagulation, is the most labile of these proteins. The two major sources of vitamin K are dietary intake and synthesis by the bacterial flora of the intestine. In the absence of malabsorption, nutritional deficiency alone rarely causes clinically significant vitamin K deficiency. The condition can arise, however, when eradication of gut flora is combined with inadequate dietary intake. This situation typically occurs in critically ill patients in intensive care units who have no oral intake and are receiving broad-spectrum antibiotics for prolonged periods. Vitamin K deficiency can also develop in patients receiving total parenteral nutrition unless the infusions are supplemented with vitamin K. Certain mutations of the gene encoding GGCX can also cause a bleeding diathesis.16 Vitamin K is absorbed predominantly in the ileum and requires the presence of bile salts. Clinically significant vitamin K deficiency occurs with malabsorption of fat-soluble vitamins secondary to obstructive jaundice (Chapter 146) or with malabsorption caused by intrinsic small bowel diseases, including celiac sprue, short-bowel syndrome, and inflammatory bowel disease (Chapters 131 and 132). Warfarin acts as an anticoagulant by competitive antagonism of vitamin K. Rare cases of hereditary deficiency of the vitamin K–dependent coagulation factors may cause a lifelong bleeding tendency. Correction of vitamin K deficiency, when it is clinically significant, can be achieved with oral supplementation, unless malabsorption is present, in which case parenteral vitamin K (10 mg subcutaneously daily) should be administered. Emergency treatment of bleeding caused by vitamin K deficiency is transfusion of fresh-frozen plasma or four-factor prothrombin complex concentrate (PCC, Kcentra).

  Grade A References A1. Umemura Y, Yamakawa K, Ogura H, et al. Efficacy and safety of anticoagulant therapy in three specific populations with sepsis: a meta-analysis of randomized controlled trials. J Thromb Haemost. 2016;14:518-530. A2. Ranieri VM, Thompson BT, Barie PS, et al. Drotrecogin alfa (activated) in adults with septic shock. N Engl J Med. 2012;366:2055-2064. A3. Afdhal NH, Giannini EG, Tayyab G, et al. Eltrombopag before procedures in patients with cirrhosis and thrombocytopenia. N Engl J Med. 2012;367:716-724. A4. Afdhal NH, Dusheiko GM, Giannini EG, et al. Eltrombopag increases platelet numbers in thrombocytopenic patients with HCV infection and cirrhosis, allowing for effective antiviral therapy. Gastroenterology. 2014;146:442-452.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 166  HEMORRHAGIC DISORDERS  

GENERAL REFERENCES 1. Boral BM, Williams DJ, Boral LI. Disseminated intravascular coagulation. Am J Clin Pathol. 2016;146:670-680. 2. Levi M. Management of cancer-associated disseminated intravascular coagulation. Thromb Res. 2016;140(suppl 1):S66-S70. 3. Warkentin TE. Ischemic limb gangrene with pulses. N Engl J Med. 2015;373:642-655. 4. Squizzato A, Rancan E, Thachil J, et al. Diagnosis of overt and non-overt disseminated intravascular coagulation: a survey among experts and a call for action from the ISTH. Thromb Res. 2017;152:74-76. 5. Levi M, Sivapalaratnam S. Disseminated intravascular coagulation: an update on pathogenesis and diagnosis. Expert Rev Hematol. 2018;11:663-672. 6. Collins P, Abdul-Kadir R, Thachil J. Management of coagulopathy associated with postpartum hemorrhage: guidance from the SSC of the ISTH. J Thromb Haemost. 2016;14:205-210. 7. Scully M, Levi M. How we manage haemostasis during sepsis. Br J Haematol. 2019. [Epub ahead of print.] 8. Squizzato A, Hunt BJ, Kinasewitz GT, et al. Supportive management strategies for disseminated intravascular coagulation. An international consensus. Thromb Haemost. 2016;115:896-904.

1150.e1

9. Allingstrup M, Wetterslev J, Ravn FB, et al. Antithrombin III for critically ill patients. Cochrane Database Syst Rev. 2016;2:CD005370. 10. Hayakawa M, Kushimoto S, Watanabe E, et al. Pharmacokinetics of recombinant human soluble thrombomodulin in disseminated intravascular coagulation patients with acute renal dysfunction. Thromb Haemost. 2017;117:851-859. 11. Allison MG, Shanholtz CB, Sachdeva A. Hematological issues in liver disease. Crit Care Clin. 2016;32:385-396. 12. Tripodi A, Primignani M, Mannucci PM, et al. Changing concepts of cirrhotic coagulopathy. Am J Gastroenterol. 2017;112:274-281. 13. Barton CA. Treatment of coagulopathy related to hepatic insufficiency. Crit Care Med. 2016;44: 1927-1933. 14. Leonardi F, De Maria N, Villa E. Anticoagulation in cirrhosis: a new paradigm? Clin Mol Hepatol. 2017;23:13-21. 15. Yates SG, Gavva C, Agrawal D, et al. How do we transfuse blood components in cirrhotic patients undergoing gastrointestinal procedures? Transfusion. 2016;56:791-798. 16. Tie J-K, Carneiro JDA, Jin CD, et al. Characterization of vitamin K-dependent carboxylase mutations that cause bleeding and nonbleeding disorders. Blood. 2016;127:1847-1855.

1150.e2

CHAPTER 166  HEMORRHAGIC DISORDERS  

REVIEW QUESTIONS 1. Which of the following statements is correct regarding disseminated intravascular coagulation (DIC)? A . There is always an underlying cause for DIC. B. Laboratory screening test results showing normal to increased platelet counts or normal to shortened (not prolonged) prothrombin time (PT) and activated partial thromboplastin time (aPTT) essentially rule out DIC. C. The primary pathophysiologic mechanism of DIC is bleeding; thrombosis can sometimes occur as a secondary event. D. Large numbers of schistocytes (fragmented red cells) are characteristic blood smear findings in DIC. E. The prolongation of PT and aPTT in DIC is a function of depletion of coagulation factors. Answer: A  DIC is a syndrome with many possible causes and never occurs in isolation as a primary event without an underlying cause. The results of laboratory screening tests of DIC (PT, aPTT, platelet count, D-dimers) can be highly variable; in fact, in chronic forms of so-called overcompensated DIC, most commonly due to malignant disease, the PT and aPTT can be actually shortened, and the platelet count can even be elevated. The primary pathophysiologic mechanism of DIC, regardless of cause, is activation of the coagulation cascade leading to increased fibrin generation and microvascular fibrin deposition; therefore, DIC is a primarily thrombotic process. The systemic bleeding that is more often seen in acute DIC is due to a combination of depleted (“consumed”) coagulation factors and platelets as well as the circulating anticoagulant effect generated by increased levels of D-dimers. Schistocytes (fragmented red cells) are nowhere nearly as prominent in the peripheral smear of a patient with DIC as they are in thrombotic thrombocytopenic purpura. As noted before, the prolongation of PT and aPTT in DIC is caused not just by the depletion of clotting factors (“consumption”) but also by the powerful circulating anticoagulant effects of fibrin(ogen) degradation products. 2. Which of the following statements is correct regarding the coagulopathy of liver failure? A . In considering anticoagulation in a patient with liver failure, it should be taken into account that the already prolonged PT (and sometimes also aPTT) intrinsic to liver failure has already made the patient at least partially “auto-anticoagulated.” B. The degree of decrease in serum albumin is a better prognostic marker than the PT (or international normalized ratio [INR]) in a patient with acute viral hepatitis who has fulminant liver failure. C. Patients with the coagulopathy of liver failure (prolonged PT and sometimes aPTT, with thrombocytopenia) have a serious bleeding risk but not thromboembolic risk. D. In general, liver biopsy can be safely performed in a patient with liver failure when the PT and aPTT do not exceed 1.5 times control values and the platelet count is above 50,000. E. In case of bleeding in a liver failure patient, prothrombin complex concentrates and recombinant human activated factor VII (rFVIIa) are first-line hemostatic agents. Answer: D  The American Association for the Study of Liver Diseases has stated that the bleeding risk-benefit ratio of liver biopsy must be carefully considered on a case-by-case basis. Nevertheless, most experts think that in general, biopsy can be performed without excessive risk of bleeding with the PT and aPTT both being less than 1.5 times control and the platelet count above 50,000. The notion that a patient in liver failure who already has a

prolonged PT and thrombocytopenia is auto-anticoagulated is misleading. It ignores the complex pathophysiologic process of the coagulopathy of liver failure, which also involves important prothrombotic abnormalities. Patients with liver failure have a two-fold increase in the risk of venous thromboembolism compared with control individuals, even when the PT (INR) is prolonged. Prothrombin complex concentrates and rVIIa are effective “general” hemostatic agents but carry a risk of thrombosis, especially in liver failure patients, who may have an impaired ability to clear those activated clotting factors. A prolonged PT is a useful prognostic indicator in various forms of liver disease and is actually a better index of prognosis than the serum albumin in acute viral hepatitis. 3. A 62-year-old man with known cirrhosis with portal hypertension and splenomegaly as well as alcoholic cardiomyopathy with poorly controlled heart failure is admitted to the intensive care unit with septic shock secondary to gram-negative sepsis. He is still intubated on hospital day 6. A hematology consultation is requested now because it is noted that his INR (PT) has been progressively rising since admission, from a baseline of 1.2 to 1.9. Which of the following is not likely to be contributing to this patient’s rising INR? A . Cirrhosis B. Splenomegaly with hypersplenism C. Sepsis D. Heart failure E. Vitamin K deficiency Answer: B  Worsening liver function could be contributing to the rising INR in this setting but not the associated portal hypertension, splenomegaly, and hypersplenism. Sepsis could be causing DIC, and worsening right-sided heart failure could be causing passive hepatic congestion, both of which might be likewise contributing. The two major sources of vitamin K are its dietary intake and its synthesis by bacterial flora in the intestine. Both of these sources of vitamin K have been essentially eliminated in this patient because he has not been eating and his gut flora has been largely sterilized with the use of broad-spectrum antibiotics for sepsis; this is a prototypical setting for vitamin K deficiency. (Elimination of only one source of vitamin K, that is, poor nutrition or use of antibiotics, is usually not sufficient to cause clinically significant vitamin K deficiency.) 4. Which of the following is not a coagulation abnormality in advanced liver disease? A . Dysfibrinogenemia B. Disseminated intravascular coagulation C. Acquired factor VIII inhibitor D. Increased fibrinolytic activity E. Decreased synthesis of coagulation factors Answer: C  The pathophysiologic mechanism of bleeding in liver failure is complex and multifactorial. Impaired liver function reduces the synthesis of all coagulation factors because that organ is the major site of their synthesis. Because the liver is also the major site of clearance of trace amounts of circulating activated clotting factors, liver failure is associated with increased levels of these, leading to essentially a state of DIC. (It is why it is so difficult and sometimes impossible to distinguish the coagulopathy of liver failure from the coagulopathy of DIC; they may in fact often coexist in liver failure.) Enhanced fibrinolytic activity, partly secondary to DIC, is part of the picture in liver disease, as is the development of dysfibrinogenemia (functional abnormality of fibrinogen).

1151

CHAPTER 167  Transfusion Medicine  

167  TRANSFUSION MEDICINE

and platelets are stored at room temperature to preserve function and posttransfusion survival. As needed, components can be modified (e.g., irradiated) or selected (e.g., hemoglobin S negative) to meet specific patient requirements.  

BETH H. SHAZ AND CHRISTOPHER D. HILLYER

Blood transfusion can be lifesaving and is currently the safest it has ever been. However, blood transfusion is associated with multiple adverse outcomes, including both noninfectious and infectious complications. Thus, benefits and risks of transfusion must be considered for each patient for optimal outcome. In fact, the United States and other countries have implemented patient blood management programs and evidence-based medicine. Patient blood management, as defined by the American Association of Blood Banks (AABB), is an evidence-based, multidisciplinary approach to optimize care of patients who might need transfusion. Through patient blood management, hospital costs can be reduced, quality of care improved, and patient outcome bettered. As a result, blood utilization has plummeted in the United States and worldwide. In the United States, 11.3 million units of red blood cells (RBCs) were transfused in 2015, which is markedly decreased from the high of 14.8 million units in 2008.1 Platelet transfusions have remained unchanged at around 2.0 million units while plasma transfusions have decreased from 4.3 to 2.7 million units/year.

  BLOOD COMPONENTS

The majority of blood collections are whole blood, which can then be separated into plasma, platelets, and red blood cell (RBC) components. In the United States, 85% of RBC units are manufactured from whole blood and greater than 90% of platelet components are collected by apheresis. The majority of blood is transfused as component therapy, tailored to the need of a specific patient. Blood components such as RBCs, plasma, platelets, or granulocytes are obtained either from donation of whole blood or via apheresis (Table 167-1). Each component is able to be stored under their respective optimal conditions; RBCs are refrigerated with additive solutions to improve long-term storage, plasma is frozen to preserve clotting factors,

Whole Blood

Whole blood contains RBCs, platelets, and plasma. It is stored refrigerated unless used shortly after collection. The shelf-life depends on the storage conditions and preservation solution used. Whole blood provides oxygen-carrying capacity (RBCs) and coagulation support (plasma and platelets). Whole blood has largely been replaced by individual blood components (“component therapy”) because most patients only require one component (RBCs, platelets, or plasma). Recently, there has been an increase in the use of whole blood, particularly for trauma patients. The military uses a walking donor program for fresh whole blood, which is stored at 22° C for less than 24 hours. This product has addressed blood component shortages in austere environments, particularly platelet products. For civilian trauma some institutions are using cold stored (2 to 6° C) whole blood, which may or may not be leukoreduced (i.e., depleted of leukocytes). Additionally, whole blood is sometimes used for pump-priming in pediatric cardiac surgery, with the goal of decreasing donor exposure. The benefits for use of whole blood in either clinical situation is not supported by randomized controlled trials.  

Red Blood Cell Components

RBC components are usually prepared from whole blood by removing 200 to 250 mL of plasma after centrifugation. RBC components can also be manufactured by apheresis, which can collect two components from a single donor. RBC components are stored at 1 to 6° C, which allows for 42-day storage. General characteristics of RBC components include 130 to 240 mL of RBCs, 50 to 80 g of hemoglobin, and 150 to 250 mg of iron per unit. The total volume, 250 to 350 mL, and hematocrit, 55 to 80%, vary depending on the volume of blood collected (450 versus 500 mL) and the specific preservative used, which also determines its shelf life. RBCs stored: in citrate-phosphatedextrose (CPD) have a hematocrit of 65 to 80% and a shelf life of 21 days, in CPDA-1 have a hematocrit of 65 to 80% and a shelf life of 35 days, and in additive solution have a hematocrit of 55 to 65% and a shelf life of 42 days. In the United States, most RBC components are stored in additive solutions.

TABLE 167-1 BLOOD PRODUCT CHARACTERISTICS PRODUCT

VOLUME

COMPOSITION

STORAGE CONDITIONS

Red Blood Cells

300-350 mL

Platelets

Platelets suspended in 20-24° C with constant 50 mL per whole blood sufficient plasma or gentle agitation derived platelet platelet additive solution component; 200-400 mL/apheresis component

Plasma

200-250 mL

Cryoprecipitate 10-15 mL/unit

200 mL RBCs, 50-100 mL plasma, 50-100 mL anticoagulant and additive solution

1-6° C

One mL of plasma contains −18° C; 1-6° C after one unit of coagulation thawing factor activity

80-120 units Factor VIII, >150 mg Fibrinogen, Factor XIII, VWF

−18° C; room temperature

after thawing

STORAGE LENGTH CPD: 21 days CPDA-1: 35 days AS: 42 days

COMMENTS • Leukoreduced: 40) is one feature that would prompt consideration of a patient for bariatric surgery when medical treatments have failed. Patients with class II obesity (BMI of 35.0 to 39.9) may be considered for bariatric surgery if medical treatments have failed and if severe, life-threatening complications are present. Waist circumference measures are recommended as an office assessment tool to help with the treatment decision-making process. The guidelines agree that the waist circumference cut points of more than 102 cm (40 inches) for men and more than 88 cm (35 inches) for women are indicators of increased metabolic risk. The relationships between disease risk and waist circumference are continuous and progressive, with no obvious cut points. The recommendation is to measure waist circumference in overweight and class I obesity adults.

TABLE 207-1 CLASSIFICATION OF OVERWEIGHT AND OBESITY BY BODY MASS INDEX (BMI) OBESITY CLASS

BMI (kg/m2) 150 mg/dL), hypertension, glucose intolerance and diabetes, and hyperuricemia should be documented. A history suggestive of sleep apnea should prompt a referral for overnight oximetry or a sleep disorder evaluation. A review of the patient’s lifestyle, including an assessment of physical activity level and eating habits, may help provide information about why the patient is obese. A family history of obesity, or long-standing obesity, provides evidence against a secondary cause of obesity. A careful medication history, including over-the-counter medications, and social history may help the clinician identify precipitating factors that can be modified. By emphasizing the role of modifiable lifestyle factors that predispose to disease risk, as opposed to focusing solely on the patient’s weight, it may be possible to initiate a conversation about weight/disease management in a less threatening manner from the patient’s perspective. Before a patient enters a weight management program, it is helpful to ensure that the patient is interested and ready to make lifestyle changes and has realistic goals and expectations. Patients who expect to lose large amounts of weight in a short time are virtually doomed to disappointment. Medical treatment programs, even if they include pharmacotherapy, struggle to routinely achieve sustained weight loss of more than 10%. Although this amount of weight loss is sufficient to markedly reduce the medical complications of obesity, disappointment with “only” 10% weight loss may cause patients to abandon a medically successful program. Helping the patient to accept that lifestyle changes resulting in achievable (10%) weight loss is a reasonable, initial goal can be challenging for a physician. It is sometimes necessary to delay entry into any treatment program if a patient is not ready to make lifestyle changes. Every form of treatment currently available requires some degree of lifestyle accommodation, and those uninterested in doing so should be advised to reconsider their goals at a later date. A reasonable strategy is to periodically remind the patient of the health benefits of improved activity and eating habits and reconsider efforts once a willingness to make changes is apparent.

TREATMENT  Obesity represents an individual’s response to the environment based on genetics and learned behavior and is best viewed as a chronic disease. Therefore treatment must be considered a long-term issue, much like diabetes, hypertension, or dyslipidemia. Substantial weight loss can be induced through severe dietary calorie restriction, but without approaches to ensure behavioral changes, body fat is invariably regained. To the extent that environmental factors contribute to a patient’s overweight status, and to the extent that the macroenvironment is unlikely to change, patients must learn how to make permanent lifestyle changes (eating and activity behavior) to hope for permanent weight loss. Behavior modification approaches,16 which can help patients recognize and circumvent environmental cues for sedentary behavior and overeating, can increase the likelihood that patients will accomplish these lifestyle changes. A1  A randomized study has shown that intensive lifestyle intervention (compared with only support and education) is associated with fewer hospitalizations, fewer medications, and lower health care costs in overweight or obese adults with type 2 diabetes. A2  Reducing energy intake is the most efficient and effective means to lose weight.17 For example, creating a 500 kcal/day deficit by reduced food intake will initially theoretically result in the loss of 1 pound of fat per week. It is more difficult to increase energy expenditure by 500 kcal/day through exercise. Greater amounts of physical activity can prevent weight gain (or weight regain after weight loss). Some patients are able to change eating and activity habits on their own, given the proper information, whereas others require formal or informal behavior modification interventions (see later) to help make these changes. In some instances, pharmacotherapy, endoscopic obesity treatments or bariatric surgery may be needed for treatment of obesity. A flow diagram on how to evaluate and to manage patients with overweight and obesity is presented in Figure 207-1.

Diet

Changes in eating habits must be permanent if weight loss is to be maintained. An experienced registered dietitian can be helpful in the evaluation of a patient’s eating habits and will be able to provide the needed education. The diet history may identify eating behaviors that result in excess energy intake. Although it is important to address specific adverse eating behaviors, patients need to understand some general principles regarding diet. Reducing the energy density of food (most commonly accomplished by reducing dietary fat) can allow patients to feel satiated while consuming fewer calories. A consensus recommendation is that providers prescribe 1200 to 1500 kcal/day for women and 1500 to 1800 kcal/day for men.18 Alternatively, diets that produce an energy deficit of 500 to 750 kcal/day can be recommended. Because there appears to be no clear superiority of one diet over another (e.g., low-carbohydrate compared with low-fat) with regard to weight loss, A3  it is recommended that providers prescribe one of the evidence-based diets that restricts selected food types (e.g., high-carbohydrate foods, low-fiber foods, or high-fat foods) to create an energy deficit by reduced food intake as well as to address issues such as dyslipidemia, diabetes, and hypertension. Patients should be informed that consuming foods high in water and fiber (fruits, vegetables, legumes, and soups) can provide satiety without excess calories. Patients also should be counseled to reduce the intake of beverages containing substantial calories, most often sugar-sweetened beverages. Finally, a regular pattern of eating should be encouraged. Commercial weight-loss programs (Weight Watchers, Jenny Craig, Nutrisystem) generally result in 2 to 4% more weight loss than usual education and counseling. A4  Contrary to expectations, however, the opportunity to choose what diet to follow does not necessarily improve weight loss. New diets are continually being promoted with the promise of easy weight loss. A common feature of these diets is the claim that special properties of certain foods help people lose weight or are the cause of obesity. If followed, most of these diets result in weight loss because of a reduced energy intake. The reduced intake can often be explained by the monotony of the diet, and no diets have been identified that cause persons to lose weight out of accordance with physiologic principles. Although a number of dietary approaches can be successful in promoting weight loss, if there is no peer-reviewed evidence for safety and success of new diets, a review by a dietitian for nutritional safety is warranted. The NHLBI/AHA/ACC/TOS obesity guideline concluded that many types of diets are able to help patients achieve long-term, medically significant weight loss. Thus it may be less important what specific type of diet (DASH diet, Mediterranean diet, high-carbohydrate/low-fat or high-fat/low-carbohydrate diet) is recommended than for the patient to find dietary adherence to be relatively easy. A comprehensive lifestyle intervention that includes a high-intensity, on-site behavioral intervention provides the best results. Very low calorie diets (20  µg/dL. In most normal individuals, the basal ACTH increases two- to four-fold and reaches a peak (20-100 pg/mL). ACTH responses may be delayed in cases of hypothalamic dysfunction. Cortisol levels usually reach 20-25 µg/dL. A normal response is cortisol >18  µg/dL. In suspected hypothalamic-pituitary deficiency, a low-dose (1-µg) test may be more sensitive.

Thyroid-stimulating hormone (TSH)

Basal thyroid function tests: free T4, free T3, TSH

Low free thyroid hormone levels in the setting of TSH levels that are not appropriately increased

Luteinizing hormone (LH), follicle-stimulating hormone (FSH)

Basal levels of LH, FSH, testosterone, estrogen

Basal LH and FSH should be increased in postmenopausal women. Low testosterone levels in conjunction with low or low-normal LH and FSH are consistent with gonadotropin deficiency.

*Values are with polyclonal assays. CRH = corticotropin-releasing hormone; GHRH = growth hormone–releasing hormone; IM = intramuscularly; IV = intravenously; PO = orally; T3 = triiodothyronine; T4 = thyroxine.

TABLE 211-3 HORMONAL REPLACEMENT THERAPY IN HYPOPITUITARISM* PITUITARY AXIS

HORMONAL REPLACEMENTS

Growth hormone (GH)

In children, GH (0.25 mg/kg) SC daily. In adults, GH (0.3-1.2 mg) SC daily. Titrate dose to achieve IGF-I levels in middle to upper part of normal range. Women receiving oral estrogens require higher doses.

Prolactin

None

Adrenocorticotropic hormone–cortisol

Hydrocortisone (10-15 mg PO q am; 5-10 mg PO q pm) or prednisone (2.5 mg PO q am; 2.5 mg PO q pm). Dose adjusted on clinical basis. Stress dosing: 50-75 mg hydrocortisone IV q8h

Thyroid-stimulating hormone–thyroid

l-thyroxine (0.075-0.15 mg) PO daily

Gonadotropins–gonads

FSH and LH (or HCG) can be used to induce ovulation in women. HCG alone or with FSH can be used to induce spermatogenesis in men. In men, testosterone enanthate (100-300 mg) IM q1-3 wk or testosterone cyclopentylpropionate (100-300 mg) IM q1-3 wk. Testosterone transdermal patches can also be used (5 mg daily). Testosterone gel 5-10 g daily In women, conjugated estrogens (0.625-1.25 mg) PO days 1-25 each month, cycled with medroxyprogesterone acetate (5-10 mg) PO days 15-25 each month. Low-dose contraceptive pills may also be used. Estrogen-containing transdermal patches are also available.

Posterior pituitary

Desmopressin, 0.05-0.2 mL (5-20 µg) intranasally once or twice daily, or tablets (0.1-0.4 mg q8-12h) or 0.5 mL (2 µg) SC

*Replacement therapy is dictated by the types of hormone deficiencies and by the clinical circumstances. In each case, the recommended preparations and doses are representative but need to be adjusted for individual patients. Other hormonal preparations are also available. FSH = follicle-stimulating hormone; GnRH = gonadotropin-releasing hormone; HCG = human chorionic gonadotropin; IGF-I = insulin-like growth factor-I; IM = intramuscularly; LH = luteinizing hormone; PO = orally; SC = subcutaneously.

hormone receptor in the respective glands may be defective, thereby resulting in resistance to the action of the pituitary hormones. Finally, genetic defects in the gene coding for the hormone or the hormone’s receptor can result in profound hormone deficiency. Acquired defects such as a tumor, specifically a macroadenoma in the sella, is the most common cause of hypopituitarism. Tumor enlargement within the confines of the bony sella turcica can compress the normal pituitary cells, resulting in their failure to function. Irradiation of the pituitary, pituitary infarction, surgical intrusion, and infiltrative disease can compromise normal pituitary functioning. In addition to structural damage to the cells, dysfunction can occur in malnutrition, critical illness, and psychiatric disorders. Diagnosis of hypopituitarism in patients with secreting or non-secreting pituitary masses usually involves a combination of static and stimulatory blood tests (Table 211-2). While there are multiple tests available, each has advantages and disadvantages, and the choice of the test may depend on the age of the patient and the clinical setting. Sometimes multiple tests may be performed if results of one test are equivocal. Treatment is replacement of the deficient hormones (Table 211-3).5



Hypophysitis

Inflammation of the pituitary is known as hypophysitis and can be classified clinically as primary hypophysitis when isolated inflammation of the pituitary is not associated with other inflammatory conditions, infections, or medications. Histologically, lymphocytic hypophysitis is the most common form and occurs three times more frequently in women, typically in the fourth decade. It was previously thought that lymphocytic inflammatory hypophysitis was predominately related to the end of pregnancy or the first few months after delivery; however, rare cases have been described in men and children. The other histologic types, granulomatous and xanthomatous hypophysitis, are not associated with pregnancy although they are more common in females. The least common form, plasmacytic or IgG4-related hypophysitis, occurs in males in a 2 : 1 ratio and usually presents in the seventh decade. Furthermore, IgG4-related hypophysitis cases usually have other organ involvement. The etiology of any of the forms of hypophysitis is unknown, and an autoantigen in the pituitary has yet to be clearly identified.6

1444

CHAPTER 211  Anterior Pituitary  

TABLE 211-4 SIGNS AND SYMPTOMS OF PITUITARY APOPLEXY SYMPTOM

TABLE 211-5 PREDISPOSING CONDITIONS ASSOCIATED WITH PITUITARY APOPLEXY

INCIDENCE

Headache

95%

Vomiting

70%

Vision Defects: Visual field defect Decreased visual acuity Diplopia (CN III, IV, V and VI)

64% 52%

Hemiplegia

Rare

Meningismus

Rare

Hypotension (cardiovascular collapse)

95%

Pituitary Tumor Nonfunctioning pituitary macroadenoma Certain functional tumors Hypertension and/or hypotension Surgery Cardiac surgery (heart lung bypass; coronary artery grafts) Major orthopedic procedures Drugs Cabergoline Bromocriptine Endocrine stimulation tests (thyrotropin-releasing hormone stimulation; insulin tolerance test) Anticoagulants Estrogen Head Trauma Pregnancy and delivery (when significant hemorrhage and hypovolemia occurs, called Sheehan syndrome) Infections Dengue fever Hypophysitis Radiation therapy Diabetes, poorly controlled Sickle cell anemia

CN = cranial nerves.

Patients with hypophysitis usually present with headaches and multiple deficiencies of anterior pituitary hormones. Diabetes insipidus (Chapter 212) is reported in half the patients, and most have associated mild hyperprolactinemia. Diagnosis typically cannot be confirmed by imaging alone. An MRI cannot reliably differentiate hypophysitis from a pituitary adenoma, but hypophysitis usually causes diffuse pituitary enlargement, normal sellar size, and thickening of the midline stalk. Diagnosis can be made based on radiologic and clinical characteristics during or just after pregnancy or if there is a proximal history of use of anti-CTLA-4 agents several months prior to the onset of symptoms (headaches and weakness). Definitive diagnosis, however, is usually made by biopsy. Surgery may be indicated when a biopsy is needed for diagnosis, when the history is unclear or if there is a mass effect. Careful pituitary function testing is mandatory even in patients who do not undergo surgery because they might otherwise die from undiagnosed adrenocortical insufficiency. The prognosis is unclear, but both the structural and functional abnormalities can occasionally resolve spontaneously. Glucocorticoids in physiologic doses may have a direct effect on the lymphocytic invasion in lymphocytic hypophysitis, but there are no prospective treatment trials of this rather rare condition.  

Pituitary Apoplexy

Insult to the pituitary can be in the form of hemorrhage, infarction, or both. When abrupt, and sometimes catastrophic hemorrhagic infarction occurs in the pituitary, it is defined as apoplexy. The constellation of headache, vomiting, visual impairment, and altered consciousness with hemodynamic instability is not specific for pituitary apoplexy but raises suspicion for the diagnosis (Table 211-4). Often, this dramatic presentation is the first time the patient is aware of a potential pituitary tumor. Asymptomatic hemorrhage and infarction into a pituitary tumor can occur in 10 to 25% of patients, but true apoplexy occurs in only 2 to 10% of pituitary tumor patients. The most common presenting complaint, headache, can present variably from retro-orbital to unilateral to bilateral temporal headaches during the acute phase of apoplexy. As the hemorrhagic infarction resolves, the patient often is left with hypopituitarism. Certain conditions predispose a patient to pituitary apoplexy (Table 211-5). Although all large pituitary tumors are at risk for hemorrhagic infarction, certain functional pituitary tumors, such as those in Cushing disease or acromegaly, may be particularly prone. Nearly 25% of all patients with apoplexy have inadequately treated hypertension. The main symptoms and consequences of apoplexy are due to the increased pressure present within the bony walls of the sella turcica in which the pituitary resides. A sudden increase in the sellar contents, due to blood and edema, results in increased pressure. This increased pressure and meningeal irritation are responsible for the neurologic symptoms described in Table 211-4, including the increased pressure in the cavernous sinus and the cranial nerve palsies as well as bitemporal hemianopsia. Extravasation of blood into the subarachnoid space causes meningeal irritation.  

(hypofunctioning of thyroid, adrenal, or gonadal systems) may be helpful. The cornerstone for diagnosis is urgent radiologic assessment. MRI T2-weighted images are the test of choice and should be performed emergently in all patients with visual symptoms. A CT scan can be useful when an MRI is not available or possible. Urgent measurement of blood chemistries, including electrolytes, kidney function, liver function, complete blood count with platelets, and prothrombin time can be useful. Because more than 80% of patients will have endocrine dysfunction, urgent measurement of free T4, TSH, prolactin, and random cortisol can be helpful. Less rapidly available and helpful (and less important in the initial diagnosis and management) are other pituitary hormones such as LH, FSH, estradiol or testosterone, growth hormone, and IGF-I. Examination of cerebrospinal fluid (CSF) is usually not diagnostic and is unnecessary if the diagnosis of apoplexy is certain. However, if there is bleeding into the CSF as a result of the apoplexy, red blood cells as well as an elevated protein level and xanthochromia can be seen. The differential diagnosis of pituitary apoplexy should include other conditions that result in the symptoms of headache, vomiting, visual disturbances, and hemodynamic instability such as infection (meningitis [Chapter 384]), cavernous sinus thrombosis (Chapter 385), migraine (Chapter 370), Rathke cyst hemorrhage (Chapter 380), and hyperemesis gravidarum (Chapter 226). Each of these conditions is itself a medical emergency that requires specific treatment.

TREATMENT AND PROGNOSIS  The initial management is stabilization of the hemodynamic status with intravenous 0.9% NaCl boluses to maintain normal tissue perfusion, usually accompanied by high-dose parenteral glucocorticoids (100 mg hydrocortisone every 8 hours intravenously). Although 80% of patients have residual hypopituitarism following apoplexy (with or without surgical decompression), some patients do not display immediate evidence of hypopituitarism. In addition, recurrent apoplexy has been reported to occur. MRI of the pituitary should be obtained at 3- to 6-month intervals until the anatomy is stable and then yearly for 5 years. A month after discharge from the hospital and recovery from the acute event, patients should have repeat endocrine testing to determine if the endocrine defect persists. Repeat testing will confirm whether the patient needs to remain on life-long hormone replacement therapy.

DIAGNOSIS

Prompt recognition of patients presenting with the triad of headache, vomiting, and visual disturbances is required to prevent death or irreversible neurologic impairment. Clinical evaluation of the patient should begin with a thorough history from the patient, if sufficiently conscious to give one, or from family members. A history of a pituitary tumor should raise the suspicion of apoplexy. More subtle abnormalities associated with pituitary dysfunction



Empty Sella

Empty sella is a radiologic diagnosis that refers to the observation on imaging of the presence of CSF in the sella turcica accompanied by a flattened pituitary gland (Fig. 211-4). It is usually noted as an incidental finding when an MRI or CT scan of the skull is performed because of problems unrelated to the pituitary. It can be primary or acquired when a defect in the diaphragma sella

CHAPTER 211  Anterior Pituitary  

A

1445

B

FIGURE 211-4.  Magnetic resonance image of an empty sella. A, Sagittal postcontrast. Note the void in the sella (large arrow) with compression of the normal pituitary against the sella turcica (small, thin arrow). B, Coronal postcontrast. Sella filled with cerebrospinal fluid and appears as empty.

allows the arachnoid membrane to herniate into the sella. If the herniation has been present for years, the sella enlarges, probably owing to persistent exposure to intracranial pressure. Primary empty sella is more frequent in women and may be accompanied by benign intracranial hypertension (Chapters 180 and 370). Pituitary function is usually normal, but about 10% of patients also have mild hyperprolactinemia, probably owing to stretching of the pituitary stalk. Patients with acquired empty sella may have a prior history of an infarcted pituitary, surgery, or radiation. After the diagnosis of empty sella is made and the patient is asymptomatic, there is no need to measure hormone levels unless there is suspicion of a clinical abnormality related to the pituitary. Because pituitary function is usually normal, no specific treatment is required. However, when there is concern about hypopituitarism, investigation as outlined above should be performed.  

Other Causes of Hypopituitarism

Functional causes of defective pituitary function can be seen in severe malnutrition (Chapter 203). Such nutritional deprivation can occur in critically ill patients who are in a state of severe calorie deprivation or in anorexia nervosa (Chapter 206). In both examples, there are likely central (hypothalamic and higher) mechanisms and molecular mediators (e.g., cachexins and tumor necrosis factor) that inhibit the release of pituitary hormone. This form of hypopituitarism is generally reversible when the caloric situation improves. Deposition and infiltration of amyloid protein in amyloidosis (Chapter 179) or iron in hemochromatosis (Chapter 201) can result in hypopituitarism.

  PITUITARY TUMORS  

PATHOBIOLOGY

Clonal proliferation of the different cell types in the pituitary result in tumor formation. Although the molecular basis for this clonal proliferation is not well understood, pituitary tumors are classified clinically as “functioning,” meaning they produce one or more hormones in excess, or “nonfunctioning,” meaning that they do not produce an excess hormone in blood.7 Because most functional tumors do not have appropriate physiologic feedback due to their oncologic nature, they usually secrete excess hormone(s) that cause a variety of clinical syndromes as described. Not uncommonly, confirmation of the secretory nature of the pituitary tumor is made at the time of removal when the surgical pathologist, based on immunochemistry, demonstrates the immunoreactivity of a hormone(s) in the tumor. However, not all immunohistochemically positive tumors result in excess blood levels of the hormone. The exact reason is unknown but may result when a defect in the intracellular processing of the hormone inhibits release of that hormone. Nevertheless, a functioning pituitary tumor is a diagnosis made clinically and based on provocative testing as discussed below. The five cell types result in different pituitary adenomas. The reported prevalences differ among large surgical series, which represent only patients who undergo surgery rather than being managed medically.

Prolactinomas, which may account for up to 50% of pituitary tumors, are likely caused when proliferation of the lactotroph results in excess prolactin and subsequent galactorrhea and hypogonadism. Somatotrophs producing growth hormone, corticotrophs making ACTH, and gonadotrophs making LH and FSH together account for 10 to 20% of adenomas. Nonfunctioning tumors, which mainly cause a mass effect or hypopituitarism, account for 10 to 25%. The size of the tumor usually is proportional to the amount of hormone it produces when functioning. Depending on the severity of the clinical perturbation caused, certain tumors are usually diagnosed sooner and therefore are usually smaller on diagnosis than others. For example, gonadotropinomas are usually diagnosed as macroadenomas because their symptoms are usually more subtle than in TSH-secreting tumors, where the symptoms present early and the tumors are usually smaller at the time of diagnosis. In addition to functioning and nonfunctioning, tumors are classified by their anatomy based on imaging studies: microadenomas are less than 10 mm in diameter versus macroadenomas which are greater than 10 mm in diameter (and typically with extrasellar extension). Pituitary adenomas are monoclonal and can be locally invasive, but they are rarely malignant. The somatic mutations causing most pituitary tumors remain unknown, and the hormonal environment may also affect the rate of tumor growth (e.g., the growth of ACTH-secreting tumors following bilateral adrenalectomy as seen in Nelson syndrome). At least six types of inherited predispositions to pituitary tumors are recognized (Table 211-6). When macroadenomas are found in children or when there is a family history of pituitary adenomas, genetic testing is warranted (Table 211-6) so that appropriate monitoring can occur before the patient is symptomatic.8  

CLINICAL MANIFESTATIONS

The clinical manifestations of pituitary tumors depend in large part on whether they are functional and whether their size compromises the normal function of the pituitary gland and thereby results in clinical symptoms of hypopituitarism.9 In general, the diagnosis of a tumor due to excess or diminishing hormone secretion presents the greater clinical challenge. Pituitary tumors have historically been diagnosed more frequently in women of reproductive age because even a slight perturbation in hormonal milieu will disrupt the multiple hormone interactions necessary for a normal menstrual period. In men or postmenopausal women, slight perturbations in hormone levels (either excess or deficiency) are usually not noticed for several years. More recently, the diagnosis of pituitary adenomas is usually an incidental finding on an imaging study obtained for unrelated reasons. Even careful endocrine evaluation of patients with incidental tumors reveals hyper- or hyposecretion of pituitary hormones in only a small percent, and such incidental tumors usually do not increase in size over time. Other than hormonal effects, reasons for suspicion of a pituitary mass are headaches and visual field abnormalities. A reported 33 to 72% of patients with a pituitary lesion have headaches, a percentage that is greater than the

1446

CHAPTER 211  Anterior Pituitary  

TABLE 211-6 INHERITED SYNDROMES OF PITUITARY ADENOMAS SYNDROME

PITUITARY TUMOR

MOLECULAR PATHOGENESIS

MODE OF INHERITANCE

McCune Albright

F/NF

Gsα SU

Somatic mutation

Ovary, bone, thyroid dysfunction

OTHER MANIFESTATIONS

Multiple endocrine neoplasias*

F (30% PRL)/NF

Menin (MEN1) and MEN4 (CDKN1B)

AD

Parathyroid and pancreas tumors

Familial Isolated Pituitary Adenoma (FIPA) Syndrome*

F/NF

Arylhydrocarbon receptor interacting protein (AIP)†

AD with variable penetrance

None

Carney Complex (with or without PPNAD)

1/3 with Cushing

Type 1A regulatory subunit of protein kinase A (PRKARIA)

AD

Atrial myxomas, spotty skin pigmentation, schwannomas

Primary Pigmented Nodular Adrenocortical Disease (PPNAD)

Cushing

Succinate dehydrogenase subunit (SDH)

AD

Pheochromocytomas, paragangliomas

Early childhood gigantism

GH

CD40LG, ARHGEF6, RBMX, GPR101

X-linked

gigantism

*Tumors are generally more aggressive and occur at a younger age. † About 15% of FIPA and 50% of familial acromegaly AD = autosomal dominant; F = functioning; GH = growth hormone; NF = nonfunctioning; PPNAD = primary pigmented nodular adrenocortical disease; PRL = prolactin.

TABLE 211-7 SELECTED TESTS OF EXCESS PITUITARY FUNCTION HORMONE

TEST

Growth hormone (GH) Basal IGF-I Oral glucose suppression test: after 75-g glucose load, GH is measured at −30, 0, 30, 60, 90, 120 min.

INTERPRETATION Elevated IGF-I levels are consistent with acromegaly when interpreted in the context of age and nutritional status. GH should be suppressed to 2 cm but ≤4 cm in greatest dimension limited to the thyroid

T3

Tumor >4 cm limited to the thyroid, or gross extrathyroidal extension invading only strap muscles Tumor >4 cm limited to the thyroid Gross extrathyroidal extension invading only strap muscles (sternohyoid, sternothyroid, or omohyoid muscles) from a tumor of any size

T3a T3b T4 T4a T4b

Includes gross extrathyroidal extension beyond strap muscles Gross extrathyroidal extension invading subcutaneous soft tissues, larynx, trachea, esophagus, or recurrent laryngeal nerve from a tumor of any size Gross extrathyroidal extension invading prevertebral fascia or encasing carotid artery/mediastinal vessels, any size

N0

No evidence of regional lymph nodes metastasis

N1 N1a N1b

Metastasis to cervical/upper mediastinal lymph nodes Central neck only (Level VI, VII) Lateral neck (Level I, II, III, IV, V)

M0

No distant metastasis

M1

Distant metastasis

From Tarasova VD, Tuttle RM. Current management of low risk differentiated thyroid cancer and papillary microcarcinoma. Clin Oncol. 2017;29(5):290-29.

THEN THE STAGE GROUP IS …

From: Tarasova VD, Tuttle RM. Current management of low risk differentiated thyroid cancer and papillary microcarcinoma. Clin Oncol. 2017;29(5):290-297.

ten-year survival is 100% for stage I disease, compared with approximately 20% in stage IV.  

Anaplastic Thyroid Carcinoma

Anaplastic thyroid carcinoma is a rare, histologically undifferentiated, clinically aggressive malignant neoplasm that typically arises in older patients, about 25% of whom have had a prior differentiated thyroid cancer. This cancer typically presents with a rapidly enlarging anterior or lateral neck mass associated with pain, tenderness, and compressive symptoms, including dysphagia, dysphonia, and stridorous dyspnea. Fine-needle aspiration biopsy of the mass usually yields large, pleomorphic, undifferentiated cells, but open surgical biopsy is sometimes required to confirm the diagnosis. Most anaplastic thyroid cancers cannot be fully resected because they already have invaded other cervical structures when diagnosed. Surgery targets airway preservation. In the case of esophageal impingement, a percutaneous gastrostomy tube is often required to ensure adequate nutrition. Patients typically relapse within a few months and die within a median of 3 to 7 months.34 However, overall survival of 22 months has been reported with trimodal therapy including surgery, radiation, and chemotherapy (potentially including targeted therapy and immunotherapy) compared with 6.5 months with just radiation and chemotherapy.  

Primary Thyroid Lymphoma

Lymphoma (Chapter 176), which rarely arises in the thyroid gland, nevertheless accounts for 1 to 5% of all thyroid malignancies. It typically presents in older persons, with an average age of diagnosis of 66 years. Hashimoto thyroiditis is the strongest risk factor for primary thyroid lymphoma. Thyroid lymphoma typically presents as a rapidly enlarging, diffuse goiter with compressive symptoms. Fine-needle aspiration biopsy usually shows abundant lymphocytes, and flow cytometry can identify monoclonal lymphocytes, usually B cells. Surgical biopsy is sometimes required to establish the diagnosis. Most tumors respond to chemotherapy and external beam radiation therapy (Chapter 176). Surgery is rarely indicated, but tracheal compression occasionally requires elective tracheostomy. Disease-free survival rates vary, but early stage disease localized to the thyroid gland is associated with a 5-year survival rate of over 85%.

  Grade A References A1. Hoang TD, Olsen CH, Mai VQ, et al. Desiccated thyroid extract compared with levothyroxine in the treatment of hypothyroidism: a randomized, double-blind, crossover study. J Clin Endocrinol Metab. 2013;98:1982-1990. A2. Stott DJ, Rodondi N, Kearney PM, et al. Thyroid hormone therapy for older adults with subclinical hypothyroidism. N Engl J Med. 2017;376:2534-2544. A3. Feller M, Snel M, Moutzouri E, et al. Association of thyroid hormone therapy with quality of life and thyroid-related symptoms in patients with subclinical hypothyroidism: a systematic review and meta-analysis. JAMA. 2018;320:1349-1359. A4. Casey BM, Thom EA, Peaceman AM, et al. Treatment of subclinical hypothyroidism or hypothyroxinemia in pregnancy. N Engl J Med. 2017;376:815-825. A5. Rajendram R, Taylor PN, Wilson VJ, et al. Combined immunosuppression and radiotherapy in thyroid eye disease (CIRTED): a multicentre, 2 x 2 factorial, double-blind, randomised controlled trial. Lancet Diabetes Endocrinol. 2018;6:299-309. A6. Smith TJ, Kahaly GJ, Ezra DG, et al. Teprotumumab for thyroid-associated ophthalmopathy. N Engl J Med. 2017;376:1748-1761. A7. Brose MS, Nutting CM, Jarzab B, et al. Sorafenib in radioactive iodine-refractory, locally advanced or metastatic differentiated thyroid cancer: a randomised, double-blind, phase 3 trial. Lancet. 2014;384:319-328. A8. Schlumberger M, Tahara M, Wirth LJ, et al. Lenvatinib versus placebo in radioiodine-refractory thyroid cancer. N Engl J Med. 2015;372:621-630.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 213 Thyroid  

GENERAL REFERENCES 1. Mendoza A, Hollenberg AN. New insights into thyroid hormone action. Pharmacol Ther. 2017;173:135-145. 2. Anyetei-Anum CS, Roggero VR, Allison LA. Thyroid hormone receptor localization in target tissues. J Endocrinol. 2018;237:R19-R34. 3. Iwen KA, Oelkrug R, Brabant G. Effects of thyroid hormones on thermogenesis and energy partitioning. J Mol Endocrinol. 2018;60:R157-R170. 4. Soh SB, Aw TC. Laboratory testing in thyroid conditions—pitfalls and clinical utility. Ann Lab Med. 2019;39:3-14. 5. Dighe M, Barr R, Bojunga J, et al. Thyroid ultrasound: state of the art part 1—thyroid ultrasound reporting and diffuse thyroid diseases. Med Ultrason. 2017;19:79-93. 6. Byun DJ, Wolchok JD, Rosenberg LM, et al. Cancer immunotherapy—immune checkpoint blockade and associated endocrinopathies. Nat Rev Endocrinol. 2017;13:195-207. 7. Chaker L, Bianco AC, Jonklaas J, et al. Hypothyroidism. Lancet. 2017;390:1550-1562. 8. Sinha RA, Singh BK, Yen PM. Direct effects of thyroid hormones on hepatic lipid metabolism. Nat Rev Endocrinol. 2018;14:259-269. 9. Alexander EK, Pearce EN, Brent GA, et al. 2017 Guidelines of the American Thyroid Association for the diagnosis and management of thyroid disease during pregnancy and the postpartum. Thyroid. 2017;27:315-389. 10. Dhital R, Basnet S, Poudel DR. Impact of hypothyroidism on occurrence and outcome of acute coronary syndrome from the national inpatient sample. Am J Cardiol. 2017;120:2160-2163. 11. Peeters RP. Subclinical hypothyroidism. N Engl J Med. 2017;376:2556-2565. 12. de Vries EM, Fliers E, Boelen A. The molecular basis of the non-thyroidal illness syndrome. J Endocrinol. 2015;225:R67-R81. 13. Ross DS, Burch HB, Cooper DS, et al. 2016 American Thyroid Association guidelines for diagnosis and management of hyperthyroidism and other causes of thyrotoxicosis. Thyroid. 2016;26:1343-1421. 14. Smith TJ, Hegedus L. Graves’ disease. N Engl J Med. 2016;375:1552-1565. 15. Khong JJ, McNab AA, Ebeling PR, et al. Pathogenesis of thyroid eye disease: review and update on molecular mechanisms. Br J Ophthalmol. 2016;100:142-150. 16. De Leo S, Lee SY, Braverman LE. Hyperthyroidism. Lancet. 2016;388:906-918. 17. Burch HB, Cooper DS. Management of Graves disease: a review. JAMA. 2015;314:2544-2554. 18. Nguyen CT, Sasso EB, Barton L, et al. Graves’ hyperthyroidism in pregnancy: a clinical review. Clin Diabetes Endocrinol. 2018;4:4-13.

1476.e1

19. Biondi B, Cooper DS. Subclinical hyperthyroidism. N Engl J Med. 2018;378:2411-2419. 20. Chiha M, Samarasinghe S, Kabaker AS. Thyroid storm: an updated review. J Intensive Care Med. 2015;30:131-140. 21. Elnaggar MN, Jbeili K, Nik-Hussin N, et al. Amiodarone-induced thyroid dysfunction: a clinical update. Exp Clin Endocrinol Diabetes. 2018;126:333-341. 22. Knobel M. Etiopathology, clinical features, and treatment of diffuse and multinodular nontoxic goiters. J Endocrinol Invest. 2016;39:357-373. 23. Durante C, Grani G, Lamartina L, et al. The diagnosis and management of thyroid nodules: a review. JAMA. 2018;319:914-924. 24. Fisher SB, Perrier ND. The incidental thyroid nodule. CA Cancer J Clin. 2018;68:97-105. 25. Dighe M, Barr R, Bojunga J, et al. Thyroid ultrasound: state of the art. Part 2—focal thyroid lesions. Med Ultrason. 2017;19:195-210. 26. Roth MY, Witt RL, Steward DL. Molecular testing for thyroid nodules: review and current state. Cancer. 2018;124:888-898. 27. Eszlinger M, Lau L, Ghaznavi S, et al. Molecular profiling of thyroid nodule fine-needle aspiration cytology. Nat Rev Endocrinol. 2017;13:415-424. 28. Fagin JA, Wells SA Jr. Biologic and clinical perspectives on thyroid cancer. N Engl J Med. 2016;375:1054-1067. 29. Cabanillas ME, McFadden DG, Durante C. Thyroid cancer. Lancet. 2016;388:2783-2795. 30. Bibbins-Domingo K, Grossman DC, Curry SJ, et al. Screening for thyroid cancer: US Preventive Services Task Force recommendation statement. JAMA. 2017;317:1882-1887. 31. Sosa JA, Duh QY, Doherty G. Striving for clarity about the best approach to thyroid cancer screening and treatment: is the pendulum swinging too far? JAMA Surg. 2017;152:721-722. 32. Haugen BR, Alexander EK, Bible KC, et al. 2015 American Thyroid Association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association Guidelines Task force on Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid. 2016;26:1-133. 33. Naoum GE, Morkos M, Kim B, et al. Novel targeted therapies and immunotherapy for advanced thyroid cancers. Mol Cancer. 2018;17:1-15. 34. Tiedje V, Stuschke M, Weber F, et al. Anaplastic thyroid carcinoma: review of treatment protocols. Endocr Relat Cancer. 2018;25:R153-R161.

1476.e2

CHAPTER 213 Thyroid  

REVIEW QUESTIONS 1. An 84-year-old man is found to have an elevated TSH level of 6.5 mU/L (reference range 0.5 to 4.5 mU/L) on routine screening. The patient has been in good health previously and feels entirely well. His physical examination is normal, including the thyroid, which is not palpable. Repeat laboratory results 2 months later: Free T4 1.1 ng/dL (0.8-1.8 ng/mL), TSH 5.9 mU/L. Anti-thyroperoxidase antibodies are negative. Which of the following statements is correct regarding this patient? A . TSH in this range is associated with increased cardiovascular mortality. B. Treatment with levothyroxine will improve his quality of life. C. The abnormal serum TSH is likely an artifact due to assay interference. D. His TSH values are normal for an octogenarian. Answer: D  Cross-sectional and prospective studies show that serum TSH levels rise with age, and that this increase is not associated with underlying thyroid disease or adverse effects. In fact, some data suggest that elderly individuals with higher serum TSH levels have lower mortality rates and better functional status. The largest published cross-sectional study, using data from the U.S. National Health and Nutrition Epidemiologic Study (NHANES), showed that the 2.5 to 97.5% confidence intervals for serum TSH for healthy octogenarians was 1.9 to 7.5 mU/l. 2. A 59-year-old male was started on nivolumab (an immune checkpoint inhibitor) for treatment of metastatic melanoma. Five months after beginning therapy the patient developed nausea, vomiting, weakness, and fatigue. A viral gastroenteritis was initially suspected, but hypophysitis and secondary adrenal insufficiency were later recognized. The patient was treated with hydrocortisone for secondary adrenal insufficiency and had resolution of his nausea and vomiting. However, he continued to have weakness and fatigue. Hypothyroidism was suspected. Which of the following laboratory tests would be most useful in confirming hypothyroidism in this particular situation? A . An MRI of the pituitary B. A serum TSH level C. A free T4 level D. Thyroid peroxidase antibodies Answer: C  Immune checkpoint inhibitors are commonly used for cancer therapy. Although very effective, they have several side effects, including the development of endocrinopathies as a result of autoimmune damage to endocrine glands. This patient has sustained damage to his pituitary gland, and although an MRI of the pituitary gland would show evidence of inflammation, such imaging would not delineate which pituitary hormones were affected. If the patient had developed secondary hypothyroidism in addition to secondary adrenal insufficiency, a serum TSH would not be elevated (as it would be in primary hypothyroidism), and could be normal or low. Although thyroid peroxidase antibodies are associated with autoimmune damage to the thyroid gland itself, including in association with nivolumab-induced thyroiditis, they do not provide information about whether thyroid insufficiency has actually developed. The best test to confirm the suspicion of secondary hypothyroidism is a free T4, which would be low or in the lower part of the normal range. 3. A 30-year-old woman develops heat intolerance, palpitations, tremor, and a 10-lb weight loss despite an increase in food intake. There is a strong family history of both hypothyroidism and hyperthyroidism. On physical exam, her pulse is 100 beat/minute and there is a mild stare but no proptosis. Her thyroid is enlarged 2-fold, symmetrical, and rubbery. There is a fine tremor. Thyroid function tests show the following: free T4 2.8 ng/dL (0.8 to 1.8 ng/dL), T3 300 ng/dL (80 to 180 ng/dL) and TSH less than 0.02 mU/L. AntiTSH receptor antibody titers are positive. The patient is started on atenolol 50 mg/day with improvement in her adrenergic symptoms, and one week later wishes to discuss specific management strategies for her Graves disease. Which is the best specific treatment for this patient? A . Methimazole for 12 to 18 months to see if she has a remission B. Radioiodine therapy to induce permanent hypothyroidism C. Thyroidectomy with permanent hypothyroidism D. Either methimazole, radioiodine, or surgery E. No specific treatment as her Graves disease may spontaneously remit

Answer: D  There is no “best” treatment for Graves disease in a young otherwise healthy individual. The only prospective randomized controlled study of antithyroid drugs, radioiodine, and surgery found that the three treatments were equivalent in terms of efficacy, patient satisfaction, and quality of life (Abraham-Nording M, et al. Thyroid. 2005;15:1279-1285). The American Thyroid Association hyperthyroidism guidelines recommends all three therapies, recognizing that many factors play into decision making, including patient preferences, other medical problems and comorbidities, and desire for pregnancy, among many others. Although some patients do have spontaneous remissions, these are uncommon, and overt hyperthyroidism always requires treatment. 4. A 60-year-old woman is found to have a 1.5-cm right thyroid nodule after undergoing a cervical MRI. On physical exam, a vague rubbery thyroid nodule is noted, corresponding to the nodule seen on imaging. Thyroid function tests are normal and thyroid ultrasound reveals an isoechoic 1.6-cm nodule with a sonolucent rim and mainly peripheral flow. FNA of the nodule is benign (an adenomatoid nodule). What is the most likely scenario in the follow-up of this patient over the next 3 to 5 years? A . The nodule will grow slowly B. The nodule will remain stable C. The nodule will decrease in size D. The nodule will need to be rebiopsied because of a greater than 50% increase in volume Answer: B  Two prospective studies have shown that over a 5-year follow-up period, thyroid nodules remain stable. In one of the studies of 1567 nodules monitored annually, 15% of nodules grew, 18% decreased in size, and the rest remained stable. Nodule growth was associated with larger nodule size at baseline, the presence of multiple nodules, male sex, and younger age (1 cm) purple striae, and new irritability, decreased cognition, and decreased short-term memory.3 Testing is indicated when clinical features have progressed over time and in patients with adrenal masses that are incidentally detected on imaging obtained for unrelated reasons (so-called adrenal incidentalomas).4 For example, oligomenorrhea is more suggestive of Cushing syndrome if a woman previously had regular menses. Serial seven subtractions and recall of three cities (or objects) are useful bedside strategies to identify deficits in cognition and memory.

Laboratory Findings

Exogenous administration of glucocorticoid should be excluded before screening for endogenous Cushing syndrome. In the absence of pseudo-Cushing states (see later), at least two different screening test results should be abnormal to establish the diagnosis. Tests for the differential diagnosis of Cushing

1478

CHAPTER 214  Adrenal Cortex  

Cushing syndrome suspected (consider endocinologist consultation)

TABLE 214-2 ETIOLOGY OF CUSHING SYNDROME EXOGENOUS Most common cause of Cushing syndrome: Glucocorticoid or ACTH driven May be factitious or iatrogenic

Exclude exogenous glucocorticoid exposure Perform one of the following tests

24-h UFC (≥2 tests)

Overnight 1-mg DST

Late night salivary cortisol (≥2 tests)

Use 48-h, 2-mg DST in certain populations (see text)

ENDOGENOUS ACTH independent—autonomous adrenal activation (20% of all cases) Adrenal adenoma (40-50%) Adrenal carcinoma (40-50%) Primary pigmented nodular adrenal disease McCune-Albright syndrome Massive macronodular adrenal disease Gastric inhibitory polypeptide or food induced ACTH dependent—adrenal activation by excessive ACTH (80% of all cases) Corticotrope adenoma (80%) Ectopic ACTH secretion (20%) Ectopic CRH secretion (rare)

ACTH = adrenocorticotropic hormone; CRH = corticotropin-releasing hormone.

Any abnormal result

Normal (CS unlikely)

with certainty unless values reach this threshold. Conversely, patients with Cushing syndrome may have normal UFC excretion because of mild or intermittent hypercortisolism or altered renal metabolism of cortisol. If UFC is only mildly elevated and clinical features are minimal, it is best to treat any pseudo-Cushing state and to remeasure UFC excretion with the expectation that it will normalize. Alternatively, if UFC values are normal but clinical suspicion is high, repeated measurement might disclose intermittent hypercortisolism. Measurement of plasma cortisol at midnight distinguishes pseudo-Cushing states from Cushing syndrome with 95% diagnostic accuracy; a level greater than 7.5 µg/dL is required for the diagnosis of Cushing syndrome. Measurement of salivary cortisol at bedtime or at midnight works as well, is more convenient, and may be the best screening test in patients with mild or intermittent hypercortisolism.6 However, the criteria for its interpretation differ, so each assay must be validated before it is used for this purpose.

Exclude physiologic causes of hypercortisolism Consult endocrinologist Perform 1 or 2 other studies shown above Suggest consider repeating the abnormal study Suggest Dex-CRH or midnight serum cortisol in certain populations

Discrepant (Suggest additional evaluation)

Abnormal

Normal (CS unlikely)

Cushing syndrome

FIGURE 214-4.  Algorithm for testing of patients suspected of having Cushing syndrome (CS). All statements are recommendations except for those prefaced by “suggest.” Diagnostic criteria that suggest Cushing syndrome are urine free cortisol (UFC) greater than the normal range for the assay, serum cortisol greater than 1.8 µg/dL (50 nmol/liter) after 1 mg dexamethasone (1-mg DST), and late-night salivary cortisol greater than 145 ng/ dL (4 nmol/liter). Dex-CRH = dexamethasone–corticotropin-releasing hormone test; DST = dexamethasone suppression test. (Reprinted with permission from Nieman LK, Biller BM, Findling JW, et al. The diagnosis of Cushing syndrome: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2008;93:1526-1540.)

syndrome should not be used to make the diagnosis. Figure 214-4 is the Endocrine Society’s recommended algorithm for testing of patients suspected of having Cushing syndrome.

Urine, Saliva, and Serum Cortisol Measurements

Urine free cortisol (UFC) excretion during 24 hours is a good screening test. Specific, structurally based assay techniques, such as high-performance liquid chromatography and tandem mass spectrometry, are the “gold standard.” The upper-normal limit of these tests is much lower and more specific than that of antibody-based assays, in which other steroids may cross-react. This crossreactivity may be an advantage in screening for hypercortisolism. UFC excretion is elevated in about 20% of adrenal incidentalomas5 and also may be increased in the so-called pseudo-Cushing states, including psychiatric disorders (depression, anxiety disorder, obsessive-compulsive disorder), chronic pain, severe exercise, alcoholism, uncontrolled diabetes, and morbid obesity. Here, it is hypothesized that higher brain pathways stimulate CRH release and activation of the entire hypothalamic-pituitary-adrenal axis (see Fig. 214-3E). Cortisol negative feedback inhibition on CRH and pituitary ACTH release restrains the resulting hypercortisoluria to less than four-fold greater than normal. Thus Cushing syndrome cannot be diagnosed

Dexamethasone Suppression Tests

The dexamethasone suppression test is a simple screening test that takes advantage of the negative feedback effect of glucocorticoids to reduce ACTH (and hence serum cortisol). Dexamethasone 1 mg is given orally between 11:00 pm and midnight, and plasma cortisol is measured between 8:00 and 9:00 the next morning. The test has an 8% false-negative rate in patients with Cushing disease and a 30% false-positive rate in chronic illness, obesity, psychiatric disorders, and normal individuals. As a result, Cushing syndrome cannot be diagnosed by this test alone unless the result is extremely abnormal. The 2-day, 2-mg dexamethasone suppression test discriminates patients with a pseudo-Cushing state if plasma cortisol end points of less than 1.4 or 2.2 µg/ dL are used. Dexamethasone 500 µg is given orally every 6 hours for eight doses, and plasma cortisol is measured 2 hours after the last dose. The test has excellent sensitivity (90 to 100%) and specificity (97 to 100%) for discriminating Cushing syndrome, but it is costly and requires excellent compliance of the patient. The immediate subsequent administration of CRH (1 mg per kilogram of body weight intravenously) and the measurement of cortisol 15 minutes later increased the sensitivity and specificity to 100% in a small study of patients, with values above 1.4 µg/dL indicating Cushing syndrome. Although this combined dexamethasone-CRH test has high diagnostic accuracy, it has the same disadvantages as the 2-day dexamethasone suppression test and the added cost of CRH testing. Because of these drawbacks, these tests are usually reserved for patients with ambiguous or confusing results on other screening tests. CRH is available commercially (Acthrel), with U.S. Food and Drug Administration–approved labeling for the differential diagnosis of Cushing syndrome. Its use in the dexamethasone-CRH test is an off-label use. Any dexamethasone test may give false results in patients with abnormal metabolic clearance of the drug. Agents that induce the cytochrome P-450 CYP3A4 enzymes (alcohol, rifampin, phenytoin, phenobarbital) increase dexamethasone clearance, whereas renal or hepatic failure decreases it. Measurement of a dexamethasone level can determine whether its clearance has been altered.

Differential Diagnosis

The causes of endogenous Cushing syndrome can be divided broadly into ACTH-dependent (80%) and ACTH-independent (20%) forms (Table 214-2).

CHAPTER 214  Adrenal Cortex  

Hypercortisolism from autonomously functioning adrenal tumors suppresses ACTH, whereas in primary disorders of ACTH excess, the adrenal glands respond to tumor-derived ACTH. Plasma ACTH concentration distinguishes between these causes. ACTH is usually less than 10 pg/mL in primary adrenal disorders but is also suppressed by exogenous steroids, whether they are prescribed intentionally (iatrogenic Cushing syndrome) or taken factitiously. Patients in the latter group often have had multiple surgical procedures and do not reveal that they are self-administering steroids. As a result, patients must be queried closely about exogenous steroid administration, recognizing that parenteral, inhaled, and topical steroids can all cause glucocorticoid excess. Patients with endogenous Cushing syndrome and low ACTH concentrations should undergo adrenal imaging to identify the site of adrenal abnormality. Nonautonomous adrenal tissue atrophies when ACTH support is subnormal. Because of this, the common ACTH-independent forms of Cushing syndrome— adrenal adenoma and carcinoma—are manifested as a unilateral adrenal mass, with atrophy of the adjacent and contralateral tissue on magnetic resonance imaging or computed tomography. Bilateral forms of primary adrenal disease are rare and may be manifested with small or large adrenal nodules. Primary pigmented nodular adrenal disease occurs primarily in children and young adults and is characterized by small to normal-sized adrenal glands containing small (15 pg/mL; 3.3 pmol/L) is consistent with an ACTH-producing tumor. Intermediate ACTH concentrations between 5 and 15 pg/mL (1.1 to 3.3 pmol/L) in a two-site sandwich assay are not diagnostic. In these patients, suboptimal cortisol responses to CRH stimulation may identify the minority of cases of ACTH-independent Cushing syndrome with borderline basal ACTH values. In addition, a suppressed plasma DHEA-S value supports the diagnosis of an ACTH-independent disorder. Cushing disease,8 an ACTH-secreting pituitary adenoma, is the most common cause of Cushing syndrome (Chapter 211). It is more common in women than in men (6 : 1 ratio), with a mean age at onset in the fourth decade. ACTH also may be secreted ectopically by a variety of neuroendocrine tumors, as shown in Table 214-3. Pituitary magnetic resonance imaging shows a tumor in only about 40 to 50% of patients with Cushing disease, but it is obtained routinely in patients

TABLE 214-3 THE INCIDENCE AND TYPES OF TUMORS CAUSING THE SYNDROME OF ECTOPIC ACTH SECRETION TUMOR TYPE Carcinoma of lung (small cell or oat cell)

PERCENTAGE 19-50

Carcinoid of bronchus

2-37

Carcinoid of thymus

8-12

Pancreatic tumors, carcinoid and islet cell

4-12

Pheochromocytoma, neuroblastoma, ganglioma, paraganglioma

5-12

Medullary carcinoma of the thyroid

0-5

Miscellaneous*

20) of morning aldosterone to plasma renin activity (Fig. 214-5). Except as noted in the figure, one of four tests (usually salt loading) is used to confirm primary hyperaldosteronism by demonstrating a lack of aldosterone suppression.14

Differential Diagnosis  

Mineralocorticoid Excess  

DIAGNOSIS

Patients with mineralocorticoid excess often have few clinical symptoms apart from fatigue and muscle weakness or cramps related to hypokalemia. Most often the condition is suspected because of hypertension, especially if it occurs at an early age in association with spontaneous hypokalemia or is difficult to control.11 The prevalence of primary aldosteronism among patients with hypertension is uncertain but is generally felt to be no higher than about 5%.12 Mineralocorticoid excess can result from primary adrenal disease, in which aldosterone (or another mineralocorticoid) is produced autonomously (and renin levels are low),13 or it may be due to nonadrenal causes as a result of elevated renin values, which stimulate aldosterone secretion. The latter situations include states of contracted arterial intravascular volume, such as congestive heart failure or cirrhosis with ascites, decreased renal arterial blood flow, and tumor production of renin (Table 214-4).

TABLE 214-4 CAUSES OF MINERALOCORTICOID EXCESS PRIMARY HYPERALDOSTERONISM: HIGH ALDOSTERONE, LOW RENIN Aldosterone-producing adenomas (30-50%) Bilateral zona glomerulosa hyperplasia Familial hyperaldosteronism Type 1: glucocorticoid-remediable hyperaldosteronism—this results from formation of a chimeric gene containing the regulator portion of 11β-hydroxylase (normally regulated by ACTH) and the synthetic region of aldosterone synthase; as a result, ACTH stimulates aldosterone synthase and hence aldosterone production Type 2: adrenal adenomas or hyperplasia expressed in a familial pattern Type 3: caused by mutant KCNJ5, often younger and more severe than type 2 Aldosterone-producing adrenal carcinoma Ectopic aldosterone secretion (rare): kidney, ovary

Having made the diagnosis of aldosterone-dependent mineralocorticoid excess, one must differentiate between the two most common adrenal causes—hyperplasia and adenoma—after excluding potential rare causes of hyperaldosteronism. Two rare autosomal dominant forms of familial hyperaldosteronism are type 1, a glucocorticoid-suppressible hyperaldosteronism, and type 2. Familial hyperaldosteronism type 1 is caused by a genetic swap of the promoter for CYP11B1 (11β-hydroxylase) with that of CYP11B2 (aldosterone synthase), forming a chimeric gene in which ACTH stimulates aldosterone synthase. It should be suspected in the setting of familial disease, particularly if there is a history of early-onset cardiovascular events, and is confirmed by gene testing (see http://www.brighamandwomens.org/Departments_and_ Services/medicine/services/endocrine/Services/gra/default.aspx). Analysis of a multiplex family with familial hyperaldosteronism type 2 and 80 additional probands with unsolved early-onset primary aldosteronism revealed that eight had novel heterozygous variants in the CLCN2 gene that encodes a voltagegated chloride channel that is expressed in adrenal glomerulosa cells.15 For the more common conditions, adrenal computed tomography scans may show nonfunctioning nodules and falsely suggest an adenoma. However, further testing is not recommended in a patient less than 35 years old with marked aldosteronism and an adrenal mass.14 The responses to physiologic maneuvers, such as upright posture, and salt loading with oral or intravenous sodium tend to be preserved in patients with hyperplasia, but there is significant overlap among groups of patients. The best diagnostic test involves the measurement of cortisol and aldosterone in bilateral adrenal venous effluent and a peripheral vein before and during an ACTH infusion. Cortisol is used to evaluate catheter placement in the adrenal veins, as levels from the two sides should be similar. When an adenoma is present, the aldosterone-tocortisol ratio on one side is usually at least five-fold greater than the other, which may be similar to the periphery, indicating suppression. Bilateral hyperplasia tends to produce similar values on each side.

TREATMENT 

SECONDARY HYPERALDOSTERONISM: HIGH ALDOSTERONE, HIGH RENIN

Treatment of primary hyperaldosteronism includes laparoscopic or robotic resection for adenomas.16 Afterward, hypokalemia generally resolves, but hypertension persists in up to 65% of patients. A mineralocorticoid antagonist, spironolactone or eplerenone, is used to treat patients unable to undergo surgery or those with hyperplasia. Eplerenone is a more selective mineralocorticoid antagonist (with fewer side effects of sexual dysfunction and gynecomastia compared with spironolactone). A sodium channel blocker (e.g., amiloride) may be helpful, and antihypertensive agents are continued as needed.

Renovascular hypertension and aortic stenosis Diuretic use Renin-secreting tumors Severe cardiac failure APPARENT MINERALOCORTICOID EXCESS: LOW ALDOSTERONE, LOW RENIN Licorice ingestion: licorice (candy or flavored tobacco) containing glycyrrhetinic acid (or similar compounds such as carbenoxolone) inhibits renal 11β-hydroxysteroid dehydrogenase type 2, reducing cortisol conversion to cortisone and enabling cortisol to act as an endogenous mineralocorticoid Severe hypercortisolism: similar in mechanism to licorice ingestion; very high cortisol levels are thought to overwhelm the ability of 11β-hydroxysteroid dehydrogenase type 2 to convert cortisol to cortisone in the kidney; cortisol itself then acts as a potent mineralocorticoid Liddle’s syndrome: mutation of the gene for the β or γ subunit of the collecting tubule sodium channel leads to a constitutive increase in sodium reabsorption and potassium excretion 11β-Hydroxylase deficiency form of congenital adrenal hyperplasia: 11-deoxycortisol accumulates because of an inability to convert it to cortisol 17-Hydroxylase deficiency form of congenital adrenal hyperplasia: deoxycorticosterone and corticosterone are increased



Androgen Excess  

DEFINITION

Women with excess circulating androgens or increased sensitivity to androgens present with complaints of hirsutism, acne, and anovulation or infertility. When testosterone is secreted in great excess, women may virilize and exhibit a deepened voice, clitorimegaly, masculinized body habitus, and alopecia.  

DIAGNOSIS

The adrenal causes of hyperandrogenism—congenital adrenal hyperplasia, Cushing disease, adrenal cancer, and androgen-producing adrenal adenoma—are

CHAPTER 214  Adrenal Cortex  

1481

Patients with hypertension that are at increased risk for PA

PA unlikely



ARR to detect cases (1|⊕⊕⊕⊕)

K+, renin PAC >20 ng/dL

+

PA unlikely



+

Patient unwilling/ unable to proceed

Confirmatory testing (1|⊕⊕⊕⊕) No need for confirmatory testing (2|⊕⊕⊕⊕)b

+

Adrenal CT (1|⊕⊕⊕⊕)

Treat with MR antagonist (2|⊕⊕⊕⊕)a

If surgery desired If surgery not desired

Subtype testing

Marked PA, young age, and + CT (2|⊕⊕⊕⊕)c

AVS (1|⊕⊕⊕⊕) Bilateral

Unilateral

Treat with MR antagonist (1|⊕⊕⊕⊕)

Treat with laparoscopic adrenalectomy (1|⊕⊕⊕⊕)

FIGURE 214-5.  Algorithm for the diagnosis and treatment of primary aldosteronism (hyperaldosteronism). Cross-filled circles indicate the quality of evidence, such that ⊕○○○ denotes very low quality evidence; ⊕⊕○○, low quality; ⊕⊕⊕○, moderate quality; and ⊕⊕⊕⊕, high quality. ARR = aldosterone-to-renin ratio; AVS = adrenal venous sampling; CT = computed tomography; MR = mineralcorticoid receptor; PA = primary aldosteronism. (Reprinted, with slight modification of text, from Funder JW, Carey RM, Mantero F, et al. The management of primary aldosteronism: case detection, diagnosis, and treatment: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2016;101:1889-916.)

uncommon. Most women have no clear-cut cause (idiopathic hirsutism) or polycystic ovary syndrome. Rarely, androgen-secreting ovarian tumors, hyperprolactinemia, glucocorticoid resistance, or exogenous drugs cause hyperandrogenism. Patients with an adrenal source of hyperandrogenism usually have increased serum levels of DHEA, DHEA-S, or androstenedione, in contrast to the testosterone excess that is more typical of an ovarian source. DHEA and DHEA-S are weak androgens that can be converted locally to testosterone in the hair follicles. Because DHEA and DHEA-S levels decline throughout adult life, these values must be interpreted within age-specific normal ranges. Although a tumor is more likely if DHEA-S is greater than 500 µg/dL or testosterone is greater than 200 ng/mL, it is not excluded at lower levels. Imaging identifies nearly all adrenal tumors but may miss a small intraovarian one. UFC may be elevated in patients with virilizing adrenal carcinoma or Cushing disease (see earlier) and in those with glucocorticoid resistance. By contrast, androgen-secreting adrenal adenomas do not have glucocorticoid excess. In women suspected of having nonclassic forms of congenital adrenal hyperplasia, precursor and product hormones should be measured before and after ACTH to confirm the diagnosis.

TREATMENT  Treatment of adrenal causes of hyperandrogenism varies according to the disorder. Classic congenital adrenal hyperplasia is treated by glucocorticoids to normalize ACTH and hence androgen levels (typically dexamethasone 0.125 to 0.375 mg at bedtime). The nonclassic forms respond well to oral contraceptive or antiandrogen treatment, with dexamethasone reserved for ovulation induction. Surgery with adjunctive medical treatment may be used in adrenal carcinoma (see earlier).

tend to be manifested either in childhood (type 1), in association with hypoparathyroidism and mucocutaneous candidiasis, or in adulthood (type 2), in association with insulin-dependent diabetes mellitus, autoimmune thyroid disease, alopecia areata, or vitiligo. The glands are small on imaging.

Adrenoleukodystrophy

Adrenoleukodystrophy, a rare (1 in 25,000) X-linked condition, is characterized by a deficiency of peroxisomal membrane adrenoleukodystrophy protein, which transports activated acyl-coenzyme A derivates into the peroxisomes, where they are shortened by β-oxidation. This deficiency results in the accumulation of very long chain fatty acids in the central nervous system and other tissues and increased plasma C26:0 fatty acids. Incomplete penetrance of the genetic defect and variable accumulation of very long chain fatty acids in the adrenal gland, brain, testis, and liver account for the clinical phenotypes, which differ by age and presentation.17

Replacement of Adrenal Tissue

Infections cause about 15% of primary adrenal insufficiency. Typical infections include tuberculosis and systemic fungal diseases (histoplasmosis, coccidioidomycosis, blastomycosis), in which the adrenal tissue is replaced by caseating granulomas. End-stage AIDS-associated opportunistic infections, such as cytomegalovirus or Mycobacterium avium-intracellulare, may reduce adrenal function. Adrenal tissue may be replaced by bilateral metastases (most commonly primary carcinoma of the lung, breast, kidney, or gut) or primary lymphoma, although adrenal insufficiency is uncommon. Intra-adrenal hemorrhage may also lead to insufficient steroidogenesis. Hemorrhage typically occurs in a stressed, hospitalized patient receiving long-term prophylactic anticoagulation and is often accompanied by back pain. The adrenal glands tend to be large on imaging.

Congenital Adrenal Hyperplasias Mixed Mineralocorticoid and Glucocorticoid Deficiency: Adrenal Insufficiency  



PATHOBIOLOGY

Primary Adrenal Insufficiency Autoimmune Destruction

Autoimmune destruction is the most common cause of primary adrenal insufficiency in industrialized countries and may occur alone or, rarely, in association with autoimmune polyglandular syndromes. These syndromes

The congenital adrenal hyperplasias18 are a disparate group of diseases caused by a genetic deficiency of one of the enzymes needed for adrenal steroidogenesis. Patients with nearly complete deficiency of an enzyme required for cortisol synthesis present in infancy with adrenal insufficiency and salt-wasting crisis. This is most problematic in patients with mutation of the 21-hydroxylase (CYP21A2) or 11β-hydroxylase (CYP11B1) gene. The increase in ACTH levels caused by cortisol deficiency drives the intact steroidogenic pathways so that there is excessive production of the steroids just proximal to the enzymatic block—17-hydroxyprogesterone and 11-deoxycortisol, respectively,

1482

CHAPTER 214  Adrenal Cortex  

TABLE 214-5 CAUSES OF ADRENAL INSUFFICIENCY AND ANCILLARY TESTS SPECIFIC CAUSES Primary adrenal insufficiency Idiopathic autoimmune destruction Polyglandular failure type 1 Polyglandular failure type 2 Infections: tuberculosis, systemic fungal diseases, AIDS-associated opportunistic infections (e.g., cytomegalovirus) Space-occupying adrenal lesions Bilateral adrenalectomy or treatment with steroidogenesis inhibitors Adrenoleukodystrophy

SUGGESTIVE CLINICAL FEATURES

USEFUL ANCILLARY TESTS

Hyperpigmentation, orthostatic hypotension Most common cause (80%) in developed countries; with or without other endocrinopathies, as below Hypoparathyroidism, mucocutaneous candidiasis, vitiligo; age 40 years 15% of patients in U.S. series

Hyperkalemia, elevated ACTH Antibodies to 21-hydroxylase are present; on imaging, adrenal glands are small

Metastases from carcinoma of lung, breast, kidney, gut; lymphoma or hemorrhage (heparin use)

Abnormal shape of adrenal glands on CT; evidence of hemorrhage Ketoconazole, mitotane, aminoglutethimide, trilostane, and metyrapone reduce cortisol levels Deficiency of peroxisomal very long chain acyl-coenzyme A synthetase leads to elevated plasma C26:0 fatty acid levels

X-linked—screen males; in childhood, cognitive and gait disturbances; in adults, spastic paraparesis

Secondary adrenal insufficiency Suppression of the adrenal axis by exogenous or Medication history; history of Cushing syndrome endogenous glucocorticoids Structural lesions of the hypothalamus or pituitary Other pituitary deficiencies gland (tumors, destruction by infiltrating disorders, x-irradiation, and lymphocytic hypophysitis) Isolated ACTH deficiency Head trauma

On imaging, adrenal glands are small Adrenal glands tend to be large on CT and may be calcified

Adrenal glands are small on imaging Adrenal glands are normal or small on imaging; MRI or CT may show pituitary or hypothalamic lesion

ACTH = adrenocorticotropic hormone; AIDS = acquired immunodeficiency syndrome; CT = computed tomography; MRI = magnetic resonance imaging.

in 21-hydroxylase and 11β-hydroxylase deficiency. The increased levels of precursor steroids enable increased adrenal androgen synthesis, so that severely affected girls may be virilized in utero. Girls and women with nonclassic congenital adrenal hyperplasia present later. They have greater enzyme activity, so that cortisol production is adequate, but increased ACTH levels cause hyperandrogenism.

Rare Causes

Other rare causes of primary adrenal insufficiency include ACTH resistance, congenital adrenal hypoplasia, Smith-Lemli-Opitz syndrome, and amyloidosis. Patients with primary adrenal insufficiency should undergo further evaluation to determine its cause (Table 214-5). Detection of antibodies to 21-hydroxylase identifies nearly all patients with idiopathic disease. In a male with negative results, measurement of plasma C26:0 fatty acids will detect adrenoleukodystrophy. Taken together, this strategy identifies the cause in nearly all adult patients with idiopathic adrenal insufficiency. Patients with autoimmune disease should be tested for other endocrine deficiencies, and those with adrenoleukodystrophy require neurologic evaluation.

Secondary Adrenal Insufficiency Suppression of the Pituitary Axis

Suppression of the hypothalamic-pituitary-adrenal axis by exogenous or endogenous glucocorticoids is the most common cause of secondary adrenal insufficiency. This phenomenon depends on the dose, duration, and schedule of glucocorticoid administration. Thus adrenal suppression is unusual with “replacement” doses of glucocorticoid that are roughly equivalent to daily production (e.g., total daily doses of 20 mg hydrocortisone, 5 mg prednisone, or 0.3 to 0.5 mg dexamethasone). At higher doses, adrenal suppression is usually not seen until after 3 weeks of administration, and a single morning administration is less suppressive than are divided doses given during the day. When potentially suppressive doses of glucocorticoids are stopped, symptoms of adrenal insufficiency may occur within 48 hours, and the entire axis may not recover for up to 18 months. During this time, the patient should receive replacement glucocorticoid treatment or supplemental steroids at times of physiologic stress, depending on the degree of impairment (see later). Chronic opiate use may suppress pituitary ACTH secretion, but the frequency of frank adrenal insufficiency appears uncommon.19

Lesions of the Hypothalamus or Pituitary

Secondary adrenal insufficiency also may result from structural lesions of the hypothalamus or pituitary gland that interfere with CRH production or transport or with corticotrope function. These causes include tumors,

trauma, destruction by infiltrating disorders, x-irradiation, and lymphocytic hypophysitis. In general, these are not reversible conditions. Patients with secondary adrenal insufficiency not ascribed to glucocorticoid use should undergo imaging of the pituitary and hypothalamus to exclude a structural or infiltrating lesion as well as tests of other pituitary function to exclude additional deficiencies.  

CLINICAL MANIFESTATIONS

The clinical presentation of adrenal insufficiency reflects the cause and duration of this uncommon condition. Primary adrenal insufficiency eventually destroys the entire adrenal cortex, with loss of both glucocorticoid and mineralocorticoid activity. By contrast, secondary adrenal insufficiency reflects an inability of the hypothalamic-pituitary unit to deliver CRH or ACTH, thus reducing trophic support to otherwise normal glands. As a result, only cortisol production decreases because mineralocorticoid production is not very ACTH dependent (Fig. 214-6). The characteristic clinical presentation of acute primary adrenal insufficiency includes orthostatic hypotension, agitation, confusion, circulatory collapse, abdominal pain, and fever.20 These features are most likely to be caused by hemorrhage, metastasis, or acute infection and can lead to death if not treated. In contrast, the typical history and clinical findings of chronic primary adrenal insufficiency include a longer history of malaise, fatigue, anorexia, weight loss, joint and back pain, and darkening of the skin (especially in the creases of the hands, extensor surfaces, recent scars, buccal and vaginal mucosa, and nipples). Patients may crave salt and may develop unusual food preferences, such as drinking the brine from pickles. Associated biochemical features for both acute and chronic presentations include hyponatremia, hypoglycemia, hyperkalemia, unexplained eosinophilia, and mild prerenal azotemia. Chronic secondary adrenal insufficiency is manifested in a similar way, but without hyperpigmentation or mineralocorticoid abnormalities.  

DIAGNOSIS

Biochemical testing confirms the diagnosis of adrenal insufficiency. A morning serum cortisol measurement is an inexpensive but relatively insensitive screening test for adrenal insufficiency in patients who are not acutely ill. The diagnosis is virtually excluded by values greater than 19 µg/dL (524 nmol/L) and is likely if the value is less than 3 µg/dL (83 nmol/L). However, both healthy individuals and patients with adrenal insufficiency may have indeterminate results (3 to 19 µg/dL) that require additional evaluation. Patients with acute adrenal insufficiency should be evaluated for sepsis, adrenal metastases, and hemorrhage. Imaging of the glands and other testing

CHAPTER 214  Adrenal Cortex  

Primary adrenal insufficiency

Normal

Secondary adrenal insufficiency

Hypothalamus

CRH +

CRH

CRH

ACTH

ACTH



Cortisol

Cortisol

ALDO

Renin

ALDO

Cortisol

Renin

ALDO

FIGURE 214-6.  Physiology of the adrenal axis in health, primary adrenal insufficiency, and secondary adrenal insufficiency. In healthy individuals (Normal), cortisol production is stimulated by the increased hypothalamic release of corticotropin-releasing hormone (CRH), which then travels down the pituitary stalk to stimulate adrenocorticotropic hormone (ACTH) secretion and release by corticotropes. Circulating ACTH stimulates adrenal gland production and secretion of cortisol. Cortisol then functions in a negative feedback mechanism to inhibit both CRH and ACTH. In patients with primary adrenal insufficiency, destruction or replacement of the entire adrenal cortex results in decreased cortisol, aldosterone, and dehydroepiandrosterone (DHEA, not shown) secretion by the adrenal glands. As a result of decreased cortisol negative feedback, the normal hypothalamus and pituitary gland increase CRH and ACTH secretion. The decreased aldosterone levels lead to an increase in renin levels. In patients with secondary adrenal insufficiency, ACTH or CRH secretion is reduced because of destruction or replacement of the hypothalamus or pituitary gland or because of disruption of the pituitary stalk. The decreased ACTH stimulation results in decreased cortisol (and, not shown, DHEA) secretion by the adrenal glands. Aldosterone production is only slightly affected by ACTH stimulation, and levels remain normal. The abnormal hypothalamus and pituitary gland do not increase CRH and ACTH secretion in response to the decreased cortisol negative feedback.

may reveal an infectious cause. In acute adrenal insufficiency, a serum cortisol value is generally inappropriately normal or subnormal in the setting of hypotension, in which cortisol values are usually well above 18 µg/dL. There is controversy about the best test to diagnose chronic adrenal insufficiency. Many use the cortisol response to exogenous ACTH as a gold standard test of adrenal steroidogenic ability. In the classic test, 250 µg of ACTH (1-24, cosyntropin) is given intravenously at any time of day. This dose of ACTH is a maximal stimulus to the adrenal gland, so that the serum cortisol level measured 30 to 60 minutes later is greater than 18 µg/dL. A 1-µg dose is equally good for establishing the diagnosis, but both doses are unfortunately associated with a substantial risk of a false-negative result.21 Lower values indicate adrenal insufficiency. Insulin-induced hypoglycemia and metyrapone stimulation have been proposed as better tests for patients with mild or recent secondary adrenal insufficiency, who may respond to pharmacologic doses of ACTH. None of these is ideal. Also, because there is no commercial formulation of ACTH for lower-dose tests, the product must be diluted on site, leading to concerns about the accuracy of the administered dose and the validity of results. Metyrapone has limited availability in the United States. Cerebral adrenoleukodystrophy, presenting in childhood, is characterized by cognitive and gait disturbances; the adult form, adrenomyeloneuropathy, is characterized by spinal cord and peripheral nerve demyelination. In both forms, the accumulation of very long chain fatty acids in the adrenal cortex alters membrane function and inhibits signal transduction by ACTH. Because a substantial minority of patients in both groups present first with adrenal insufficiency, boys and young men with adrenal insufficiency should be screened for adrenoleukodystrophy.

Differential Diagnosis

Acute Adrenal Insufficiency

Chronic Adrenal Insufficiency

Adrenal

Renin

TREATMENT  In suspected acute adrenal insufficiency, hydrocortisone is the treatment of choice because it has both glucocorticoid and mineralocorticoid activity. Treatment with intravenous saline for volume expansion, glucose for hypoglycemia, and intravenous hydrocortisone (100 mg) is started immediately after placement of an intravenous line and withdrawal of blood for documentation of the cortisol value.

Pituitary

ACTH

1483

Primary and secondary adrenal insufficiency can be distinguished by measurement of plasma ACTH. In primary adrenal insufficiency, ACTH levels are generally above the normal range and may exceed the normal range before the cortisol response to exogenous ACTH stimulation is subnormal. In addition, hyperkalemia and elevated renin values are characteristic of primary but not of secondary adrenal insufficiency, which is identified by a suppressed or inappropriately normal ACTH level.

Therapy for chronic adrenal insufficiency22 aims to provide the physiologic replacement of steroids. Glucocorticoid replacement is achieved by administration of 10 to 12 mg/m2 of hydrocortisone daily in one to three oral doses, attempting to mimic the physiologic diurnal variation of cortisol concentrations. Hydrocortisone offers the advantage of multiple-dose tablets, which allows fine adjustment and splitting of the daily dose. Ideally, the morning dose is given as soon after waking as possible; for individuals who feel extremely fatigued in the morning before the agent is absorbed, a strategy of taking the medication 30 minutes before arising may be helpful. Some data suggest that once-daily dosing is preferable in terms of immune function, weight gain, and quality of life. A3  Although many patients do well with a single dose, others complain of pronounced fatigue in the afternoon and evening. For them, a split-dose regimen, in which about one third of the daily dose is given around 4:00 PM or two afternoon doses are given, may be useful. Other glucocorticoids may be used for daily replacement therapy. Prednisone, 5 to 7.5 mg daily, has the advantage of a long half-life and may be particularly helpful in patients with afternoon or evening fatigue. Dexamethasone may be used, but because of variable interindividual metabolism, it is difficult to recommend a specific replacement dose; in addition, few options for fixed doses are available, so it is difficult to adjust the dose. Patients with primary adrenal insufficiency should be encouraged to salt their food and not to limit salt intake. Nearly all patients also require a mineralocorticoid, such as fludrocortisone 50 to 300 µg/day. The dose is adjusted until plasma renin activity is normal. If a mineralocorticoid is not given, the dose of hydrocortisone or other steroid with mineralocorticoid activity is often mistakenly increased to reduce an “unwell” feeling or hyperkalemia or salt craving. However, if a supraphysiologic dose is given, the patient becomes cushingoid. Patients with primary adrenal insufficiency also have decreased serum DHEA levels. Controversy exists about its replacement. A meta-analysis concluded that there is insufficient evidence to support its routine use in these patients.

Ensuring Proper Dosing Education of the Patient

All patients receiving chronic glucocorticoid replacement therapy should be instructed that they must take the glucocorticoids as prescribed and that failure to take or to absorb the medication will lead to adrenal crisis and possibly death. They should wear medical information bracelets or necklaces that identify this requirement. It is important to educate patients and their families about glucocorticoid adjustment during physiologic stress conditions, including the emergency administration of intramuscular glucocorticoid by means of a kit containing prefilled syringes with injectable steroid.

Dosing for Stress

The daily oral glucocorticoid dose is usually doubled for “stressful” physiologic conditions, such as fever, nausea, and diarrhea, although there are few data to support this strategy. In addition, this practice may lead to chronic overmedication by the patient because of a liberal interpretation of what constitutes physical stress. Thus education about when and how to change the dose of steroid should be reinforced periodically, preferably with written material, and the dangers of excessive steroid use should be emphasized. If the patient is vomiting, has severe diarrhea, or has collapsed, intramuscular glucocorticoids should be given before transport to a medical facility. The glucocorticoid dose is increased in proportion to the amount of stress. Thus during maximally stressful situations (e.g., adrenal crisis, major surgery, trauma, labor and delivery), the daily hydrocortisone dose is 100 to 300 mg. Few data support the need for this supraphysiologic dose, but the safety of not following this practice has not been established. The dose may be tapered by 50% per day if the patient is clinically stable. For more moderate stress, such as that of cholecystectomy, 75 to 100 mg of hydrocortisone is given on the day of surgery, and the dose is tapered more rapidly. Patients undergoing minimal stress, such as tooth extraction or short operative orthopedic procedures, may not require any additional supplementation.

Assessment to Ensure Proper Dosing

Clinical assessment is the best way to judge whether the glucocorticoid dose is correct. Symptoms of adrenal insufficiency improve with adequate therapy. The development of cushingoid features or osteopenia suggests frank or subtle overreplacement, respectively, and the presence of adrenal insufficiency symptoms (fatigue, anorexia, weight loss) suggests underreplacement.

In women, DHEA replacement increases testosterone levels, so that hirsutism, acne, or other signs of androgen excess may suggest overreplacement. In primary adrenal insufficiency, adequate hormone replacement results in plasma ACTH levels that decrease but remain elevated, in the range of 100 to 200 pg/mL. Renin values, however, normalize completely and may be used to judge the adequacy of mineralocorticoid replacement. Although hydrocortisone is metabolized to cortisol, plasma cortisol values should not be used to monitor therapy because clearance from the bloodstream is rapid, and circulating values are low for most of the day. UFC does not reflect adequate replacement; the increase in plasma cortisol levels after a single daily dose may exceed corticosteroidbinding globulin capacity, resulting in excessive urine levels and overestimation of integrated cortisol levels.



PROGNOSIS

Even in educated patients with chronic adrenal insufficiency, adrenal crisis develops at a rate of 8.3 crises per 100 patient-years. Common precipitating causes include gastrointestinal infection, fever, and emotional stress (about 20% of cases each), but major pain, surgery, strenuous physical activity, heat, and pregnancy are other precipitants. About 6% of crises are fatal.23  

Mineralocorticoid Deficiency

Hypoaldosteronism may be classified as a low-normal or a high renin state on the basis of plasma renin activity after 4 hours of upright posture. Renin deficiency is the most common cause of hypoaldosteronism, occurring most often in older patients with mild, nonoliguric renal disease who often have insulin-dependent diabetes and potentially diabetic nephropathy. Indomethacin and other prostaglandin synthesis inhibitors as well as autonomic dysfunction associated with prolonged bed rest can also result in hyporeninemic hypoaldosteronism.  

CLINICAL MANIFESTATIONS

There are few clinical features associated with mineralocorticoid deficiency; as a result, it is usually suspected when laboratory results reveal hyperkalemia, hyponatremia, and a mild metabolic alkalosis. If glucocorticoid deficiency is excluded, isolated hypoaldosteronism is established if the circulating level of aldosterone is inappropriately low. High renin states of hypoaldosteronism include congenital adrenal hyperplasias with mineralocorticoid deficiency and primary adrenal insufficiency when it is treated with pure glucocorticoid replacement. Treatment of these conditions involves sodium replacement with at least 10 mEq/kg/day, roughly equivalent to the 4 g of sodium chloride found in a typical diet in the United States. For individuals who do not maintain such a diet (often the elderly or the young), fludrocortisone can be given at the same doses used in primary adrenal insufficiency.

  Grade A References A1. Lacroix A, Gu F, Gallardo W, et al. Efficacy and safety of once-monthly pasireotide in Cushing’s disease: a 12 month clinical trial. Lancet Diabetes Endocrinol. 2018;6:17-26. A2. Fassnacht M, Terzolo M, Allolio B, et al. Combination chemotherapy in advanced adrenocortical carcinoma. N Engl J Med. 2012;366:2189-2197. A3. Isidori AM, Venneri MA, Graziadio C, et al. Effect of once-daily, modified-release hydrocortisone versus standard glucocorticoid therapy on metabolism and innate immunity in patients with adrenal insufficiency (DREAM): a single-blind, randomised controlled trial. Lancet Diabetes Endocrinol. 2018;6:173-185.

  GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 214  Adrenal Cortex  

GENERAL REFERENCES 1. Fuller PJ, Yang J, Young MJ. 30 years of the mineralocorticoid receptor: coregulators as mediators of mineralocorticoid receptor signaling diversity. J Endocrinol. 2017;234:T23-T34. 2. Loriaux DL. Diagnosis and differential diagnosis of Cushing’s syndrome. N Engl J Med. 2017;376: 1451-1459. 3. Yorke E, Atiase Y, Akpalu J, et al. Screening for Cushing syndrome at the primary care level: what every general practitioner must know. Int J Endocrinol. 2017;2017:1-6. 4. Vaidya A, Hamrahian A, Bancos I, et al. The evaluation of incidentally discovered adrenal masses. Endocr Pract. 2019;25:178-192. 5. Araujo-Castro M, Sampedro Núñez MA, Marazuela M. Autonomous cortisol secretion in adrenal incidentalomas. Endocrine. 2019. [Epub ahead of print.] 6. Blair J, Adaway J, Keevil B, Ross R. Salivary cortisol and cortisone in the clinical setting. Curr Opin Endocrinol Diabetes Obes. 2017;24:161-168. 7. Bonnet-Serrano F, Bertherat J. Genetics of tumors of the adrenal cortex. Endocr Relat Cancer. 2018;25:R131-R152. 8. Molitch ME. Diagnosis and treatment of pituitary adenomas: a review. JAMA. 2017;317:516-524. 9. Sharma ST; AACE Adrenal Scientific Committee. An individualized approach to the evaluation of Cushing syndrome. Endocr Pract. 2017;23:726-737. 10. Beuschlein F, Weigel J, Saeger W, et al. Major prognostic role of Ki67 in localized adrenocortical carcinoma after complete resection. J Clin Endocrinol Metab. 2015;100:841-849. 11. Mete O, Duan K. The many faces of primary aldosteronism and Cushing syndrome: a reflection of adrenocortical tumor heterogeneity. Front Med (Lausanne). 2018;5:1-12.

1484.e1

12. Kayser SC, Dekkers T, Groenewoud HJ, et al. Study heterogeneity and estimation of prevalence of primary aldosteronism: a systematic review and meta-regression analysis. J Clin Endocrinol Metab. 2016;101:2826-2835. 13. Byrd JB, Turcu AF, Auchus RJ. Primary aldosteronism. Circulation. 2018;138:823-835. 14. William TA, Reincke M. Management of Endocrine Disease: diagnosis and management of primary aldosteronism: the Endocrine Society guideline 2016 revisited. Eur J Endocrinol. 2018;179:R19-R29. 15. Scholl UI, Stölting G, Schewe J, et al. CLCN2 chloriode channel mutations in familial hyperaldosteronism type II. Nat Genet. 2018;50:349-354. 16. Colvin J, Krishnamurthy V, Jin J, et al. A comparison of robotic versus laparoscopic adrenalectomy in patients with primary hyperaldosteronism. Surg Laparosc Endosc Percutan Tech. 2017;27:391-393. 17. Köhler W, Curiel J, Vanderver A. Adulthood leukodystrophies. Nat Rev Neurol. 2018;14:94-105. 18. El-Maouche D, Arit W, Merke DP. Congenital adrenal hyperplasia. Lancet. 2017;390:2194-2210. 19. Gibb FW, Stewart A, Walker BR, Strachnan MW. Adrenal insufficiency in patients on long-term opioid analgesia. Clin Endocrinol (Oxf). 2016;85:831-835. 20. Pazderska A, Pearce SH. Adrenal insufficiency—recognition and management. Clin Med (Lond). 2017;17:258-262. 21. Ospina NS, Al Nofal A, Bancos I, et al. ACTH stimulation tests for the diagnosis of adrenal insufficiency: systematic review and meta-analysis. J Clin Endocrinol Metab. 2016;101:427-434. 22. Esposito D, Pasquali D, Johannsson G. Primary adrenal insufficiency: managing mineralocorticoid replacement therapy. J Clin Endocrinol Metab. 2018;103:376-387. 23. Hahner S, Spinnler C, Fassnacht M, et al. High incidence of adrenal crisis in educated patients with chronic adrenal insufficiency: a prospective study. J Clin Endocrinol Metab. 2015;100:407-416.

1484.e2

CHAPTER 214  Adrenal Cortex  

REVIEW QUESTIONS 1. A 35-year-old woman has had weight gain, insomnia, decreased short-term memory, and depression for the last year. She states that her symptoms are worsening. Two recent urine free cortisol (UFC) test results are threefold increased. She takes lisinopril for recent-onset hypertension. What is the next best diagnostic test? A . A 1-mg dexamethasone suppression test B. A pituitary magnetic resonance imaging study C. An adrenocorticotropic hormone (ACTH) measurement D. Another UFC determination E. A dexamethasone–corticotropin-releasing hormone test Answer: A  Recent guidelines and expert opinion suggest that at least two different screening test results should be abnormal to confirm the clinical suspicion of Cushing syndrome (see Fig. 214-4). This patient has had two abnormal UFC results and should undergo another screening test, either measurement of late-night salivary cortisol or a 1-mg or 2-day 2-mg dexamethasone suppression test. Another UFC determination will not obviate the need for a different screening test. The dexamethasone–corticotropinreleasing hormone stimulation test is more cumbersome and expensive than the dexamethasone suppression test and would not be a better choice. Tests that are used for the differential diagnosis, such as ACTH level or magnetic resonance imaging, should not be used to establish the diagnosis of Cushing syndrome. 2. A 32-year-old white woman with known idiopathic primary adrenal insufficiency complains of recent fatigue and a feeling of being unwell. This has been worse lately when she has been more active with gardening and is worse in the afternoon and evening. She denies dizziness, anorexia, increase in pigmentation, or change in weight. On physical examination, she weighs 125 pounds, with a height of 63 inches. Blood pressure is 110/75, and pulse is 86 without orthostasis. She has pigmented palmar creases and some buccal pigmentation, unchanged from the last visit. There are no other skin lesions. She does not appear cushingoid, there is no abdominal tenderness, and the joint examination findings are normal. Her only medications are hydrocortisone, 15 mg in the morning and 5 mg in the afternoon, and fludrocortisone, 50 mg every other day. What is the next best step? A . Measure a plasma renin level. B. Increase the afternoon hydrocortisone dose to 10 mg. C. Measure the urine free cortisol excretion. D. Add a late afternoon hydrocortisone dose of 5 mg. E. Measure plasma ACTH before the morning hydrocortisone dose. Answer: A  This woman has somewhat nonspecific symptoms that are exacerbated by gardening (perhaps by heat). She is receiving adequate hydrocortisone replacement but taking a low fludrocortisone dose. Such nonspecific symptoms may occur in the setting of relative mineralocorticoid deficiency, which can be evaluated through renin measurement. If the renin were elevated, an increased fludrocortisone dose might eliminate her vague symptoms of being unwell. The history and the amount of prescribed hydrocortisone do not suggest that she needs additional glucocorticoid. If she is mineralocorticoid deficient, she might feel better with a higher hydrocortisone dose because of its mineralocorticoid activity; however, a slightly supraphysiologic dose would put her at risk for osteopenia and other features of glucocorticoid excess. (See the section on adrenal insufficiency, assessment to ensure proper dosing.) Measurement of ACTH and urine free cortisol usually is not helpful in assessing adequacy of dosage.

3. A 30-year-old woman presents to the emergency department with the acute onset of nausea, vomiting, diarrhea, and dizziness. Her only medication is levothyroxine, 88 µg daily. On examination, she is hypotensive, tachycardic, and orthostatic. She has dark skin, and no hyperpigmentation is seen on cursory examination. She has abdominal pain and increased bowel sounds. She is extremely weak but able to move all extremities and to respond to questions. The sodium concentration is 129 mEq/L, potassium concentration is 5.3 mEq/L, and blood urea nitrogen level is 40 mg/dL. Apart from initiating fluid resuscitation and evaluation for sepsis, what treatment and testing should be ordered immediately? A . Draw blood for cortisol measurement before giving hydrocortisone 100 mg IV. B. Give dexamethasone 8 mg and perform an ACTH stimulation test. C. Measure an ACTH level and give hydrocortisone 100 mg IV. D. Give dexamethasone 8 mg and defer further testing. E. Give hydrocortisone 100 mg IV and obtain an adrenal computed tomography scan to evaluate for hemorrhage. Answer: A  This woman presents with the signs and symptoms of acute primary adrenal insufficiency. The knowledge that she takes thyroid hormone suggests polyglandular failure. However, it is possible that this constellation of features is caused by sepsis, renal failure, or hypothyroidism. Thus it is prudent to measure cortisol in blood obtained during intravenous line insertion to confirm the diagnosis of adrenal insufficiency. Other biochemical and stimulation testing can be deferred. She has no apparent reason to have adrenal hemorrhage. Dexamethasone 8 mg is more than what is needed for treatment of adrenal insufficiency; hydrocortisone 100 mg is the recommended initial dose. (See the section on adrenal insufficiency, replacement of adrenal tissue, and treatment.) 4. A 50-year-old man is evaluated for hyperaldosteronism because of difficultto-control hypertension. He is otherwise healthy. Evaluation was done during treatment with agents that do not affect aldosterone or renin levels and include an aldosterone-to-renin ratio of more than 40 and lack of aldosterone suppression to oral salt loading. A computed tomography (CT) scan of the adrenal glands shows a 1-cm mass on the right with a density of 10 HU. What is the next best step? A . Adrenal venous sampling B. Right laparoscopic adrenalectomy C. Treatment with eplerenone D. Review of CT scan to exclude hyperplasia of the left adrenal gland E. Ultrasound ablation of the adrenal gland Answer: A  Adrenal venous sampling should be performed to identify the source of excess aldosterone. It is possible that the visualized mass is a nonfunctioning tumor. This man is a good surgical candidate, and laparoscopic adrenalectomy would be recommended if venous sampling is consistent with unilateral disease. The CT scan is not helpful in identifying bilateral disease. Treatment with eplerenone and ultrasound ablation would be possibilities if he were not a surgical candidate; eplerenone would be an appropriate treatment for bilateral disease. (See section on mineralocorticoid excess, diagnosis and treatment.)

CHAPTER 214  Adrenal Cortex  

5. A 35-year-old woman presents with a 6-month history of increasing body hair growth. Before that time, she states that she was mildly “hairy” but now needs to shave her face daily. (She has stopped shaving for 3 days so that her hair growth would be visible.) She has had only two episodes of vaginal spotting during this time. Her menses started at the age of 13 years and are usually irregular, occurring every 30 to 45 days. The extra hair growth started in her teens and until 6 months ago was similar to the hair pattern of her mother and sister. She admits to some recent irritability and perhaps has gained 5 pounds. She has always had a low voice and does not notice any recent change. She has a 3-year-old daughter. She has been able to continue her 35 minutes of daily cardiovascular exercise. On physical examination, body mass index is 30, blood pressure 125/85 mm Hg, pulse 82. The Ferriman-Galway score is 17. Fat distribution is normal with some increased abdominal fat, and she appears “well developed” but not especially muscular. There is no temporal hair recession, but her scalp hair appears thin. Acne is present on the chest and back. The clitoral index is 100 mm. Laboratory results include low luteinizing hormone and follicle-stimulating hormone, estradiol 50 pg/mL, dehydroepiandrosterone sulfate (DHEA-S) 653 ng/dL (reference range, 4 to 332), and testosterone 235 ng/dL. These abnormalities are confirmed on a second set of tests. What is the next best diagnostic step? A . Obtain a CT scan of the adrenal glands. B. Obtain an ovarian ultrasound. C. Perform a 1-mg overnight dexamethasone suppression test. D. Measure a morning17-hydroxyprogesterone level. E. Measure an androstenedione level.

1484.e3

Answer: A  This woman has recent-onset hirsutism and virilization. Although worsening polycystic ovary syndrome is a possibility, the degree of DHEA-S and testosterone elevation suggests the possibility of an adrenal tumor. The slight weight gain and irritability might also point to possible cortisol excess. Although androstenedione would confirm an adrenal source, it is unlikely to point to another cause, and so proceeding to adrenal CT scan is reasonable, particularly because adrenal cancer is part of this differential diagnosis. An ovarian ultrasound is less likely to be abnormal (i.e., an ovarian tumor is less likely). Congenital adrenal hyperplasia is consistent with these signs and symptoms but not with the time course of rapid worsening. Cushing syndrome is a possibility, but it is less likely to be Cushing disease with these levels of hormones; adrenal cancer is a possibility that would be identified by the CT scan, making a 1-mg dexamethasone suppression test less urgent. (See the section on androgen excess.)

1484

CHAPTER 215  Adrenal Medulla, Catecholamines, and Pheochromocytoma  

215  ADRENAL MEDULLA, CATECHOLAMINES, AND PHEOCHROMOCYTOMA WILLIAM F. YOUNG, JR.

  ADRENAL MEDULLA AND CATECHOLAMINES

The adrenal medulla occupies the central portion of the adrenal gland. Adrenomedullary cells are called chromaffin cells (they stain brown with chromium salts). Chromaffin cells differentiate in the center of the adrenal gland in response to cortisol; some chromaffin cells also migrate to form paraganglia. The largest cluster of chromaffin cells outside the adrenal medulla is near the level of the inferior mesenteric artery and is referred to as the organ of Zuckerkandl. Catecholamines are substances that contain catechol (o-dihydroxybenzene) and a side chain with an amino group—the catechol nucleus (Fig. 215-1). Epinephrine is synthesized and stored in the adrenal medulla and released into the systemic circulation. Norepinephrine is synthesized and stored not only in the adrenal medulla but also in the peripheral sympathetic nerves. Dopamine, the precursor of norepinephrine, is found in the adrenal medulla and peripheral sympathetic nerves. Catecholamines have many cardiovascular and metabolic actions, including increasing the heart rate, blood pressure, myocardial contractility, and cardiac conduction velocity. Three types of specific adrenergic receptors mediate their biologic actions: α, β, and DA. Their receptor subtypes are α1, α2, β1, β2, β3, DA1, and DA2. The α1 subtype is a postsynaptic receptor that mediates vascular and smooth muscle contraction; stimulation causes vasoconstriction and increased blood pressure. The α2-receptors are located on presynaptic sympathetic nerve endings and, when activated, inhibit the release of norepinephrine; stimulation causes suppression in central sympathetic outflow and decreased blood pressure. Stimulation of the β1-receptor causes positive inotropic and chronotropic effects on the heart, increased renin secretion in the kidney, and lipolysis in adipocytes as well as bronchodilation, vasodilation in skeletal muscle, glycogenolysis, and increased release of norepinephrine from sympathetic nerve terminals. The β3-receptor regulates energy expenditure and lipolysis. DA1 receptors are localized to the cerebral, renal, mesenteric, and coronary vasculature; stimulation causes vasodilation in these vascular beds. DA2 receptors are presynaptic and localized to sympathetic nerve endings, sympathetic ganglia, and brain; their stimulation inhibits the release of norepinephrine, inhibits ganglionic transmission, and inhibits prolactin release. Catecholamines are synthesized from tyrosine by a process of hydroxylation and decarboxylation (see Fig. 215-1). Tyrosine is derived from the diet or synthesized from phenylalanine in the liver, and it enters neurons and chromaffin cells by active transport. Tyrosine is converted to 3,4-dihydroxyphenylalanine (dopa) by tyrosine hydroxylase, the rate-limiting step in catecholamine synthesis. α-Methyl-p-tyrosine (metyrosine) is a tyrosine hydroxylase inhibitor that may be used therapeutically in patients with catecholamine-secreting tumors. Aromatic l-amino acid decarboxylase catalyzes the decarboxylation of dopa to dopamine. Dopamine is actively transported into granulated vesicles to be hydroxylated to norepinephrine by the dopamine β-hydroxylase. These reactions occur in the synaptic vesicle of adrenergic neurons and the chromaffin cells of the adrenal medulla. In the adrenal medulla, norepinephrine is released from the granule into the cytoplasm, where phenylethanolamine N-methyltransferase converts it to epinephrine. Expression of phenylethanolamine N-methyltransferase is positively regulated by glucocorticoids. Thus catecholamine-secreting tumors that secrete primarily epinephrine are localized to the adrenal medulla. In normal adrenal medullary tissue, approximately 80% of the catecholamine released is epinephrine. The biologic half-life of circulating catecholamines is between 10 and 100 seconds. Thus plasma concentrations of catecholamines fluctuate widely. Catecholamines are removed from the circulation by either reuptake by sympathetic nerve terminals or metabolism through two enzyme pathways (see Fig. 215-1), followed by sulfate conjugation and renal excretion. Almost 90% of catecholamines released at sympathetic synapses are taken up locally by the nerve endings (uptake 1). Uptake 1 can be blocked by cocaine, tricyclic antidepressants, and phenothiazine. Extraneuronal tissues also take up

1484.e4

ABSTRACT

CHAPTER 215  Adrenal Medulla, Catecholamines, and Pheochromocytoma  

Catecholamines affect many cardiovascular and metabolic processes and form the basis for many cardiovascular, pulmonary, and neuroendocrine pharmacotherapeutics. Catecholamine-secreting tumors that arise from chromaffin cells of the adrenal medulla and the sympathetic ganglia are referred to as pheochromocytomas and extra-adrenal catecholamine-secreting paragangliomas, respectively. Although catecholamine secreting tumors are rare, it is important to suspect, confirm, localize, and resect these tumors, because (1) the associated hypertension is curable with surgical removal of the tumor; (2) a risk of lethal paroxysm exists; (3) 10% of the tumors are malignant; and (4) 40% of these tumors are familial and their detection in the proband may result in early diagnosis in other family members.

KEYWORDS

pheochromocytoma paraganglioma catecholamines metanephrines adrenal medulla

1485

CHAPTER 215  Adrenal Medulla, Catecholamines, and Pheochromocytoma  

OH

OH

OH HO

OH

TH Catechol

HCH

HCH

HC

HC

COOH

NH2

NH2 Tyrosine

Dopa AADC

OH HO HO

CH CH2

OH

PNMT HO HO

DBH

CH CH2

NHCH3

Dopamine

MAO

MAO

OH HO HO

CH C

HO HO

COMT

Dihydroxymandelic acid OH CH CH2 NHCH3

COMT

O

CH2 CH2 NH2

Dihydroxyphenylacetic acid

3-Methoxytyramine

COMT

CH

MAO CH3O

CH2

HO

NH2

CH2 C

O

OH

Normetanephrine

Metanephrine MAO

C

CH3O HO

OH

OH CH3O HO

COMT CH2

O

OH

CH3O HO

CH2 NH2

Norepinephrine

MAO

CH2

NH2

Epinephrine

COMT

HO HO

Homovanillic acid (HVA)

MAO OH

CH3O HO

CH C

O

OH Vanillylmandelic acid (VMA) FIGURE 215-1.  Biosynthetic and metabolic pathways for catecholamines. The term catecholamine comes from the catechol (o-dihydroxybenzene) structure and a side chain with an amino group—the catechol nucleus (top left). Tyrosine is converted to 3,4-dihydroxyphenylalanine (dopa) in the rate-limiting step by tyrosine hydroxylase (TH). Aromatic L-amino acid decarboxylase (AADC) converts dopa to dopamine. Dopamine is hydroxylated to norepinephrine by dopamine β-hydroxylase (DBH). Norepinephrine is converted to epinephrine by phenylethanolamine N-methyltransferase (PNMT); cortisol serves as a cofactor for PNMT, and this is why epinephrine-secreting pheochromocytomas are almost exclusively localized to the adrenal medulla. Metabolism of catecholamines occurs through two enzymatic pathways. Catechol-O-methyltransferase (COMT) converts epinephrine to metanephrine and norepinephrine to normetanephrine by meta-O-methylation. Metanephrine and normetanephrine are oxidized by monoamine oxidase (MAO) to vanillylmandelic acid by oxidative deamination. Monoamine oxidase also may oxidize epinephrine and norepinephrine to dihydroxymandelic acid, which is then converted by catechol-O-methyltransferase to vanillylmandelic acid. Dopamine is also metabolized by monoamine oxidase and catechol-O-methyltransferase, with the final metabolite homovanillic acid.

catecholamines (uptake 2). Most of these catecholamines are metabolized by catechol-O-methyltransferase. Metanephrine and normetanephrine are oxidized by monoamine oxidase to vanillylmandelic acid by oxidative deamination. Monoamine oxidase may also oxidize epinephrine and norepinephrine to 3,4-dihydroxymandelic acid, which is then converted by catechol-O-methyltransferase to vanillylmandelic acid. In the storage vesicle, norepinephrine is protected from metabolism by monoamine oxidase. Monoamine oxidase and catechol-O-methyltransferase metabolize dopamine to homovanillic acid (see Fig. 215-1).

  PHEOCHROMOCYTOMA AND PARAGANGLIOMA  

DEFINITION

Catecholamine-secreting tumors that arise from chromaffin cells of the adrenal medulla and the sympathetic ganglia are referred to as pheochromocytomas

and extra-adrenal catecholamine-secreting paragangliomas, respectively. Because the tumors have similar clinical presentations and are treated with similar approaches, many clinicians use the term pheochromocytoma to refer to both entities. However, the distinction between pheochromocytoma and paraganglioma is an important one because there are differences in the risk for associated neoplasms, risk for malignant transformation, and type of genetic testing that should be considered.1,2  

EPIDEMIOLOGY

Catecholamine-secreting tumors are rare; the annual incidence is 2 to 8 cases per million people. Nevertheless, it is important to suspect, confirm, localize, and resect these tumors. The associated hypertension is curable with surgical removal of the tumor, a risk of lethal paroxysm exists, and at least 10% of the tumors are malignant. Approximately 40% of cases are familial, so detection of this tumor in the proband may result in early diagnosis in other family members.

1486



CHAPTER 215  Adrenal Medulla, Catecholamines, and Pheochromocytoma  

PATHOBIOLOGY

Genetics

Approximately 40% of patients with catecholamine-secreting tumors have germline mutations (inherited mutations present in all cells of the body) in genes associated with the genetic disease.3 Hereditary catecholamine-secreting tumors typically are manifested at a younger age than sporadic neoplasms. Sporadic pheochromocytoma is typically diagnosed on the basis of symptoms or incidental discovery on computed imaging, whereas syndromic pheochromocytoma and paraganglioma are frequently diagnosed earlier in the course of disease because of biochemical surveillance or genetic testing.

Multiple Endocrine Neoplasia

Multiple endocrine neoplasia (MEN) type 2A is an autosomal dominant disorder (Chapter 218). The phenotype includes adrenal pheochromocytoma in 50% (usually bilateral and may be asynchronous), medullary carcinoma of the thyroid in 100%, hyperparathyroidism in 20 to 30%, and cutaneous lichen amyloidosis in 5%.4 Medullary carcinoma of the thyroid is usually detected before pheochromocytoma. Numerous activating mutations in the RET protooncogene have been documented in persons with MEN type 2A. MEN type 2B is also an autosomal dominant disorder, and it represents approximately 5% of all MEN type 2 cases. The phenotype includes pheochromocytoma in 50% (usually bilateral), aggressive medullary carcinoma of the thyroid in 100%, mucosal neuromas (typically involving the tongue, lips, and eyelids) in most patients, thickened corneal nerves, intestinal ganglioneuromatosis, and marfanoid body habitus. MEN 2B–associated tumors are caused by mutations in the RET protein’s intracellular domain, as described in detail in Chapter 218.

Genetic Testing

Since 1990, 15 different pheochromocytoma and paraganglioma susceptibility genes have been reported: NF1, RET, VHL, SDHD, SDHC, SDHB, EGLN1/ PHD2, KIF1B, SDHAF2, IDH1, TMEM127, SDHA, MAX, HIF2A, and FH gene encoding fumarate hydratase.8 Genetic testing should be considered if a patient has one or more of the following: paraganglioma, bilateral adrenal pheochromocytomas, unilateral adrenal pheochromocytoma and a family history of pheochromocytoma or paraganglioma, onset of unilateral adrenal pheochromocytoma at a young age (before 45 years), or other clinical findings suggestive of one of the previously discussed syndromic disorders. Genetic testing can be complex; testing of one family member has implications for related individuals.9,10 Genetic counseling is recommended to help families understand the implications of genetic test results; to coordinate the testing of at-risk individuals; and to help families work through the psychosocial issues that may arise before, during, or after the testing process. A list of clinically approved molecular genetic diagnostic laboratories is available at www.genetests.org.  

CLINICAL MANIFESTATIONS

Catecholamine-secreting tumors occur with equal frequency in men and women, primarily in the third, fourth, and fifth decades of life. These tumors are rare in children; when discovered, they may be multifocal, extra-adrenal, metastatic, and associated with a hereditary syndrome.11 When symptoms are present, they are due to the pharmacologic effects of excess concentrations of circulating catecholamines (Table 215-1). The resulting hypertension may be

Von Hippel–Lindau Disease

Von Hippel–Lindau (VHL) disease is an autosomal dominant disorder characterized by pheochromocytoma (frequently bilateral), paraganglioma (skull base, mediastinal, abdominal, pelvic), hemangioblastoma (involving the cerebellum, spinal cord, or brain stem), retinal angioma, clear cell renal cell carcinoma, pancreatic neuroendocrine tumor, endolymphatic sac tumor of the middle ear, serous cystadenoma of the pancreas, and papillary cystadenoma of the epididymis and broad ligament.5 Pheochromocytoma occurs in about 10 to 20% of patients with VHL disease. Nearly 100% of patients have an identifiable gene mutation (VHL tumor suppressor gene).6 Certain missense mutations appear to be associated with a “pheochromocytoma only” presentation of VHL disease.

Neurofibromatosis

Neurofibromatosis type 1 (NF1) is an autosomal dominant disorder characterized by neurofibromas, multiple café au lait spots, axillary and inguinal freckling, iris hamartomas (Lisch nodules), bone abnormalities, central nervous system gliomas, pheochromocytoma and paraganglioma, macrocephaly, and cognitive deficits. The expression of these features is variable. Approximately 3% of patients with NF1 develop catecholamine-secreting tumors; in these patients, the tumor is usually a solitary benign adrenal pheochromocytoma, occasionally bilateral adrenal pheochromocytomas, and rarely an abdominal paraganglioma.7 Inactivating NF1 mutations cause the disorder (NF1 tumor suppressor gene).

Familial Paraganglioma

Familial paraganglioma is an autosomal dominant disorder characterized by paragangliomas that are located in the skull base and neck, thorax, abdomen, and pelvis. Most cases of familial paraganglioma are caused by mutations in the succinate dehydrogenase (SDH; succinate:ubiquinone oxidoreductase) subunit genes (SDHB, SDHC, SDHD, SDHAF2, SDHA), which compose portions of mitochondrial complex II. In patients with SDHD mutations, penetrance depends on the mutation’s parent of origin. Hence, the disease does not manifest when the mutation is inherited from the mother but is highly penetrant when it is inherited from the father. This phenomenon is known as maternal imprinting. Multiple cofactors are required for normal activity of the SDH complex, including flavin adenine dinucleotide (FAD) in the SDH1 subunit. FAD is covalently attached to Sdh1, and deletion of SDHAF2 causes a complete loss of FAD cofactor attachment (flavination) of Sdh1. Like families with mutations in SDHD, those with mutations in SDHAF2 also exhibit maternal imprinting and parasympathetic paragangliomas that typically occur in the skull base and neck. Patients with SDHB mutations have an increased risk for malignant paraganglioma.

TABLE 215-1 SIGNS AND SYMPTOMS ASSOCIATED WITH CATECHOLAMINE-SECRETING TUMORS SPELL RELATED Anxiety and fear of impending death Diaphoresis Dyspnea Epigastric and chest pain Headache Hypertension Nausea and vomiting Pallor Palpitation (forceful heartbeat) Tremor CHRONIC Cold hands and feet Congestive heart failure—dilated or hypertrophic cardiomyopathy Constipation (megacolon) Diaphoresis Dyspnea Ectopic hormone secretion–dependent symptoms (e.g., CRH/ACTH, GHRH, PTH-RP, VIP) Epigastric and chest pain Erythrocytosis Fatigue Fever Grade II to IV retinopathy Headache Hyperglycemia Hypertension Livedo reticularis Orthostatic hypotension Painless hematuria (associated with urinary bladder paraganglioma) Palpitation (forceful heartbeat) Raynaud phenomenon Tremor Weight loss NOT TYPICAL OF PHEOCHROMOCYTOMA Flushing ACTH = adrenocorticotropic hormone; CRH = corticotropin-releasing hormone; GHRH = growth hormone–releasing hormone; PTH-RP = parathyroid hormone–related peptide; VIP = vasoactive intestinal polypeptide. Modified from Young WF Jr. Pheochromocytoma: 1926-1993. Trends Endocrinol Metab. 1993;4:122-127.

CHAPTER 215  Adrenal Medulla, Catecholamines, and Pheochromocytoma  

sustained (in approximately half of patients) or paroxysmal (in approximately a third of patients). The remaining patients have normal blood pressure. Episodic symptoms may occur in spells, or paroxysms, that can be extremely variable in presentation but typically include forceful heartbeat, pallor, tremor, headache, and diaphoresis. The spell may start with the sensation of a “rush” in the chest and a sense of shortness of breath, followed by a “pounding” heartbeat in the chest that typically progresses to a throbbing headache. Peripheral vasoconstriction with a spell results in cool or cold hands and feet and facial pallor. Increased sense of body heat and sweating are common symptoms that occur toward the end of the spell. Spells may be spontaneous or precipitated by postural change, anxiety, medications (e.g., metoclopramide, β-adrenergic inhibitors, anesthetic agents, corticosteroids), exercise, or maneuvers that increase intraabdominal pressure (e.g., change in position, lifting, defecation, exercise, colonoscopy, pregnancy, trauma). Although the types of spells experienced by patients are highly variable, spells tend to be stereotypical for each patient. Spells may occur multiple times a day or as infrequently as once a month. The typical duration of a pheochromocytoma spell is 15 to 20 minutes, but it may be much shorter or last several hours. The clinician must recognize that most patients with spells do not have a pheochromocytoma. Fasting hyperglycemia and diabetes mellitus are caused in part by the α-adrenergic inhibition of insulin release. Some of the cosecreted hormones that may dominate the clinical presentation include corticotropin (Cushing syndrome), parathyroid hormone–related peptide (hypercalcemia), vasopressin (syndrome of inappropriate antidiuretic hormone secretion), vasoactive intestinal peptide (watery diarrhea), and growth hormone–releasing hormone (acromegaly) (see Table 215-1). Cardiomyopathy and congestive heart failure are the symptomatic presentations of pheochromocytoma that are most frequently unrecognized by clinicians. Cardiomyopathy, whether dilated or hypertrophic, may be totally reversible with tumor resection. Some patients with pheochromocytoma may be asymptomatic despite high circulating levels of catecholamines, probably reflecting adrenergic receptor desensitization related to chronic stimulation. Symptomatic pheochromocytomas are localized to the adrenal glands, with an average diameter of 4.5 cm (Fig. 215-2). Paragangliomas are found where there is chromaffin tissue: along the para-aortic sympathetic chain, within the organs of Zuckerkandl (at the origin of the inferior mesenteric artery), in the wall of the urinary bladder, and along the sympathetic chain in the neck or mediastinum. Paragangliomas in the skull base and neck region (e.g., carotid body tumors, glomus tumors, chemodectomas) usually arise from parasympathetic tissue and typically do not hypersecrete catecholamines and metanephrines, whereas paragangliomas in the mediastinum, abdomen, and pelvis usually arise from sympathetic chromaffin tissue and usually do hypersecrete catecholamines and metanephrines.



DIAGNOSIS

Differential Diagnosis

Numerous disorders can cause signs and symptoms that may lead the clinician to test for pheochromocytoma. These disorders span much of medicine and include endocrine disorders (e.g., primary hypogonadism), cardiovascular disorders (e.g., idiopathic orthostatic hypotension), psychological disorders (e.g., panic disorder), pharmacologic causes (e.g., withdrawal from an adrenergic inhibitor), neurologic disorders (e.g., postural orthostatic tachycardia syndrome), and miscellaneous disorders (e.g., mast cell disease). Indeed, most patients tested for pheochromocytoma do not have it. In addition, fractionated catecholamines and metanephrines may be elevated in several clinical scenarios: withdrawal from medications or drugs (e.g., clonidine, alcohol), any acute illness (e.g., subarachnoid hemorrhage, migraine headache, preeclampsia), and administration of many drugs and medications (e.g., tricyclic antidepressants, levodopa, cocaine, phencyclidine, lysergic acid diethylamide, amphetamines, ephedrine, pseudoephedrine, phenylpropanolamine, isoproterenol) (Table 215-2). Pheochromocytoma should be suspected in patients who have one or more of the following: hyperadrenergic spells (e.g., self-limited episodes of nonexertional palpitations, diaphoresis, headache, tremor, or pallor); resistant hypertension; a familial syndrome that predisposes to catecholamine-secreting tumors (e.g., MEN type 2, NF1, VHL disease); a family history of pheochromocytoma or a history of a resected pheochromocytoma and a present history of recurrent hypertension or spells; an incidentally discovered adrenal mass with increased precontrast CT attenuation of more than 20 Hounsfield units (HU); hypertension and diabetes; pressor response during anesthesia, surgery, or angiography; onset of hypertension at a young age (before 20 years); and idiopathic dilated cardiomyopathy.

Laboratory Findings

The diagnosis must be confirmed biochemically by increased concentrations of fractionated metanephrines in the plasma or fractionated catecholamines and metanephrines in a 24-hour urine collection (Fig. 215-3). Most laboratories now measure fractionated catecholamines (dopamine, norepinephrine, and epinephrine) and metanephrines (metanephrine and normetanephrine) by high-performance liquid chromatography with electrochemical detection or tandem mass spectroscopy.12,13 These techniques have overcome the problems with fluorometric analysis (e.g., false-positive results caused by α-methyldopa, labetalol, sotalol, and imaging contrast agents). One of the most reliable methods of identifying catecholamine-secreting tumors is measurement of fractionated metanephrines and catecholamines in a 24-hour urine collection (sensitivity, 98%; specificity, 98%). If clinical suspicion is high, plasma fractionated metanephrines should also be measured. Some groups have advocated the measurement of plasma fractionated metanephrines as a first-line test for pheochromocytoma; the predictive value of a negative test result is extremely high, and the finding of normal plasma fractionated metanephrines excludes pheochromocytoma, except in patients with early preclinical disease and those with strictly dopamine-secreting neoplasms. A plasma test is also attractive because of its simplicity. Although measurement of plasma fractionated metanephrines has a sensitivity of 96 to 100%, the specificity is poor at only 85 to 89%; the specificity falls to 77% in patients older than 60

TABLE 215-2 MEDICATIONS THAT MAY INCREASE MEASURED LEVELS OF CATECHOLAMINES AND METANEPHRINES Tricyclic antidepressants (including cyclobenzaprine) Levodopa Drugs containing adrenergic receptor agonists (e.g., decongestants) Amphetamines FIGURE 215-2.  Contrast-enhanced computed tomography of the abdomen in a 32-year-old second-year medical student with the peripartum discovery of a pheochromocytoma. The plasma fractionated metanephrines were abnormal: metanephrine, 0.19 nmol/L (normal, 95%; specificity, >65%). Approximately 85% of these tumors are found in the adrenal glands, and 95% are found in the abdomen. If the results of abdominal imaging are normal, localization with gallium 68 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA)-octreotate positron emission tomography (68-Ga DOTATATE PET) is indicated (see Fig. 215-3). Ga-68 DOTATATE PET CT has a lesionbased detection rate of 96-98% and is superior to 123I metaiodobenzyl guanidine (123I-MIBG) scintigraphy.14–16 If a typical (20 Hounsfield units) E. Rapid contrast washout with more than 50% washout at 10 minutes Answer: D  See section on diagnosis. Imaging characteristics of an incidentally discovered adrenal mass consistent with pheochromocytoma include increased precontrast CT attenuation (e.g., >20 Hounsfield units); marked enhancement with intravenous contrast medium on CT; high signal intensity on T2-weighted MRI; cystic and hemorrhagic changes; bilaterality; and large size (>4 cm). In addition, pheochromocytomas are characterized by slow contrast washout (e.g., 200 mg/dL) and typical symptoms of polyuria, polydipsia, and weight loss. Because plasma glucose levels range on a continuum, the selection of a specific diagnostic threshold is in some respects arbitrary. Current criteria are based on the plasma glucose or HbA1c level above which the risk of diabetesspecific microvascular complications (e.g., retinopathy) is perceptibly increased. In situations of altered red blood cell turnover or certain hemoglobinopathies, HbA1c may not accurately reflect mean plasma glucose levels (see later section on glycosylated hemoglobin), and direct glucose measurement should be used. Separate glucose criteria exist for the diagnosis of gestational diabetes (see section on gestational diabetes under clinical manifestations of type 2 diabetes). States of impaired glucose regulation, not meeting the criteria for diabetes, have also been defined (fasting glucose concentration of 100 to 125 mg/dL, 2-hour glucose concentration of 140 to 199 mg/dL, or HbA1c level of 5.7 to 6.4%). Individuals in these categories are at increased risk for diabetes, although not all will progress and some may revert to normal glucose regulation. Impaired glucose tolerance (oral glucose tolerance test 2-hour glucose concentration of 140 to 199 mg/dL) has also been associated with increased risk of atherosclerotic cardiovascular disease (CVD), which may be independent of future development of diabetes.  

Glycosylated Hemoglobin

Measurements of glycosylated hemoglobin have been in clinical use since the 1980s as a means of assessing glucose control in patients with diabetes and more recently for the diagnosis of diabetes and pre-diabetic states. Hemoglobin A1c (HbA1c) is formed by the nonenzymatic glycosylation of hemoglobin, and its percentage reflects the exposure of the hemoglobin A molecule to glucose during the lifespan of circulating red blood cells (about 120 days). Thus HbA1c has a predictable (but nonlinear) relationship with mean plasma glucose levels during the preceding 3 to 4 months, although more recent glucose exposure

TABLE 216-2 DIAGNOSTIC CRITERIA FOR DIABETES NORMAL

IMPAIRED (PRE-DIABETES)

DIABETES

15 years) are reported in some patients with treatment with somatostatin analogs and other tumor-directed therapies.  

VIPomas  

DEFINITION

The VIPoma syndrome, also called the Verner-Morrison syndrome, pancreatic cholera, and the WDHA syndrome (for watery diarrhea, hypokalemia, and achlorhydria), results from a neuroendocrine tumor, usually in the pancreas, that ectopically secretes vasoactive intestinal polypeptide (VIP).13

1524



CHAPTER 219  Neuroendocrine Tumors  

EPIDEMIOLOGY AND PATHOBIOLOGY

VIPomas in adults are found in the pancreas in 80 to 90% of cases; rare cases result from intestinal carcinoids, ganglioneuromas, ganglioneuroblastomas, and pheochromocytomas. In children less than 10 years and rarely in adults, the VIPoma syndrome is caused by ganglioneuromas or ganglioneuroblastomas at extrapancreatic sites. VIPomas are usually large and solitary; 50 to 75% occur in the pancreatic tail, and 40 to 70% have metastasized at the time of diagnosis. VIPomas frequently secrete both VIP and peptide histidine methionine, but VIP is responsible for the symptoms. VIP is a potent stimulant of secretion in the small and large intestine, which causes the cardinal clinical features. VIP also causes relaxation of GI smooth muscle, and this may contribute to the dilated loops of bowel that are common in this syndrome, as well as a dilated atonic gallbladder. Hypochlorhydria is thought to result from the inhibitory effect of VIP on gastric acid secretion, the flushing is related to the vasodilatory effects of VIP, and the hyperglycemia is caused by the glycogenolytic effect of VIP. The mechanism of the hypercalcemia remains unclear.  

CLINICAL MANIFESTATIONS

The cardinal clinical feature is severe, large-volume, watery diarrhea (>1 L/ day) in all cases, which is secretory and occurs during fasting. Hypokalemia (67 to 100%) and dehydration (83%) commonly occur because of the volume of the diarrhea. Achlorhydria is only occasionally noted, but hypochlorhydria is usually found (34 to 72% of cases). Flushing occurs in 20% of patients, hyperglycemia in 25 to 50%, and hypercalcemia in 41 to 50%. Steatorrhea is uncommon (16%) despite the volume of diarrhea.  

DIAGNOSIS/DIFFERENTIAL DIAGNOSIS

The diarrhea of VIPomas characteristically persists during fasting and is large in volume (>3 L/day in 70 to 80%); the diagnosis is excluded when fasting stool volume is less than 700 mL/day. To differentiate VIPomas from other causes of large-volume fasting diarrhea, fasting plasma VIP levels should be determined. The normal value in most laboratories is less than 190 pg/mL, and elevated levels are present in 90 to 100% of patients. The differential diagnosis of large-volume fasting diarrhea (>700 mL/day) includes ZES, diffuse islet-cell hyperplasia, surreptitious use of laxatives, the pseudopancreatic cholera syndrome, and, rarely, HIV infection (Chapter 366). Serum gastrin levels identify patients with ZES, and plasma VIP levels are normal in most patients who abuse laxatives, in 82% of patients with pancreatic islet cell hyperplasia, and in patients with HIV-induced secretory diarrhea.

TREATMENT  The symptoms caused by the VIP are controlled initially in greater than 85% of patients by daily doses of octreotide (50 to 400 µg 1 to 3 times/daily) or by monthly injections of a depot form (octreotide [LAR]/lanreotide [Autogel]), but increased doses may be needed over time. Before the availability of octreotide, small numbers of patients were reported to respond to a variety of agents, including high-dose prednisone (60 to 100 mg/day; 40 to 50%), clonidine, lithium carbonate, indomethacin, loperamide, metoclopramide, and phenothiazines. After tumor localization studies, surgical resection should be attempted if it is possible to remove all visible tumor; however, greater than 50% of patients have generalized liver metastases at diagnosis, so complete resection may not be possible. For patients with unresectable advanced disease, treatment is similar to that for other advanced nonresectable pNETs (outlined later).

cases; 70 to 92% will have metastasized at the time of diagnosis, and they are usually large (mean, 5 cm) and solitary. In contrast, duodenal somatostatinomas are smaller (mean, 2.4 cm), frequently associated with psammoma bodies on histologic examination (11%), and less frequently have metastases at diagnosis (30 to 40%).  





PROGNOSIS



Somatostatinomas  

DEFINITION AND PATHOBIOLOGY

Somatostatinomas are neuroendocrine tumors that occur primarily in the pancreas or upper small intestine and ectopically secrete somatostatin. In the GI tract, somatostatin inhibits basal and stimulated gastric acid secretion, pancreatic secretion, intestinal absorption of amino acids, gallbladder contractility, and release of numerous hormones, including cholecystokinin and gastrin. Pancreatic somatostatinomas occur in the pancreatic head in 60 to 80% of

DIAGNOSIS/DIFFERENTIAL DIAGNOSIS

Somatostatinomas are usually found incidentally, particularly during exploratory laparotomy or cholecystectomy, during endoscopy, or on imaging studies. The presence of psammoma bodies on histologic examination of a duodenal neuroendocrine tumor or any duodenal lesions in patients with von Recklinghausen disease should raise the suspicion of a duodenal somatostatinoma. The diagnosis of the somatostatinoma syndrome requires the demonstration of increased concentrations of somatostatin-like immunoreactivity in the plasma and the resected tumor. However, other tumors outside the pancreas or intestine, such as small cell lung cancer, medullary thyroid carcinoma, pheochromocytomas, and paragangliomas, may also have elevated concentrations of somatostatinlike immunoreactivity. Somatostatinomas can be imaged using somatostatin receptor scintigraphy or, if needed, other conventional imaging studies to assess the tumor’s location.

TREATMENT  Treatment with octreotide or lanreotide can improve symptoms. Surgery, if possible, should be performed. For patients with unresectable advanced disease, treatment is similar to that for other advanced nonresectable pNETs (outlined later).



PROGNOSIS

Patients with intestinal somatostatinomas, which uncommonly cause the somatostatinoma syndrome, and are less malignant, have an excellent prognosis (5-year survival rate >80%), whereas those with pancreatic somatostatinomas, which frequently cause the somatostatinoma syndrome and present with metastatic disease (>70%), have a much reduced 5-year survival rate (5 years) after a diagnosis of metastatic pancreatic adenocarcinoma should be suspected of having an NF-pNET. In older series of symptomatic patients, primary tumors were large (70% were >5 cm) and liver metastases were frequent (38 to 62%) at presentation, whereas in recent series of asymptomatic patients, the NF-pNETs detected incidentally have been frequently less than 2 cm, without metastases. An elevated plasma chromogranin A or pancreatic polypeptide level or a positive somatostatin receptor scintigraphic scan is strong evidence that a pancreatic mass is a pNET. Malignancy correlates with vascular/perineural invasion, a proliferative index of greater than 2%, a mitotic rate of greater than 2, a size of at least 4 cm, capsular penetration, nuclear atypia, lack of progesterone receptors, and the presence of calcitonin immunoreactivity in the tumor.

TREATMENT  Survival is better in patients with smaller tumors, patients who are asymptomatic at presentation, patients with no metastases, and patients in whom surgical resection can be performed. Surgical resection should be performed, whenever possible, in patients with NF-pNETs greater than 2 to 3 cm, if duct compression is found, or if a higher grade tumor is present. The role of surgery in NF-pNETs that are less than 2 to 2.5 cm is controversial, with some recommending observation with reevaluation and others surgery.15 For patients with unresectable advanced disease, treatment is similar to that outlined later for other advanced nonresectable pNETs.



PROGNOSIS

The overall 5-year survival rate varies in different series from 30% to greater than 90%, but it is heavily dependent on the extent of the disease at diagnosis, with survival rates of 96% in patients without metastases at presentation, decreasing to 30 to 50% with metastatic disease.  

ACTHomas and Other Uncommon Tumors

pNETs that ectopically secrete ACTH cause 4 to 16% of the cases of ectopic Cushing syndrome.16 Cushing syndrome (Chapter 214) occurs in 5% of all cases of ZES, but in patients with sporadic ZES it is a late feature, occurring

1525

with metastatic liver disease. Its development is associated with a poor prognosis, and the response to chemotherapy is generally poor; however, some patients benefit from the use of long-acting somatostatin analogs (octreotide, lanreotide). Paraneoplastic hypercalcemia (Chapter 169) can result from a pNET that releases parathormone-related peptide or an unknown hypercalcemic substance. Tumors are generally large, with metastatic liver disease at diagnosis. Somatostatin analogs may help control the hypercalcemia, but surgery, chemotherapy, hepatic embolization, and chemoembolization are the mainstays of treatment. pNETs causing the carcinoid syndrome (see later) are usually large, and 68 to 88% are malignant. Octreotide may control the symptoms and the newly released tryptophan hydrolase 1 inhibitor, telotristat, may be helpful to control the diarrhea. Surgery, chemotherapy, hepatic embolization, chemoembolization, or molecular targeted therapy (everolimus, sunitinib) may be helpful. A single case of a pNET that secreted renin manifested with severe hypertension; the tumor was localized with somatostatin receptor scintigraphy, and the patient’s symptoms improved significantly after tumor resection. A single case of an erythropoietin-secreting pNET resulting in polycythemia and a single case of a pNET secreting IGF-II or GLP-1 causing hypoglycemia have been described. Two symptomatic cases of pNETs that secreted luteinizing hormone have been described; virilization occurred in the female patient, whereas the male patient had increased acne and a rash. In both cases, the tumors were resectable, and symptoms improved postoperatively. A single case of a pNET secreting cholecystokinin (CCKoma) has been described, with the patient demonstrating PUD, gallbladder disease, diarrhea, and weight loss. A case of a pNET secreting enteroglucagon causing small intestinal hypertrophy has also been described.

Neuroendocrine Tumors of the Gastrointestinal Tract and Thorax (Carcinoids, non-pNETs)  

These well-differentiated NETs, also called carcinoid, occur most often in the GI tract (about 25% in the small intestine, 15% in the rectum, 10% in the appendix, and 5% in the stomach) and the lung (30%), but they also can occur in many other organs.17  

Stomach Neuroendocrine Tumors

Gastric NETs occur in several subtypes with differing pathophysiology, natural history, and treatment.18,19 Type 1 gastric NET (carcinoid) arises in the body and fundus of the stomach in the setting of autoimmune atrophic gastritis. This is the most common type, comprising 80% of cases. In type 1 lesions gastrin is markedly elevated due to the absence of negative feedback from gastric acid, not because the tumors are secreting gastrin. The high gastrin stimulates proliferation of gastric enterochromaffin-like (ECL) neuroendocrine cells in the body and fundus of the stomach, resulting in neuroendocrine cell hyperplasia and small multifocal neuroendocrine lesions, which occasionally progress to type 1 gastric carcinoids. These NETs are generally of low grade and have a low metastatic potential when less than 2 cm in size and are usually removed endoscopically. Multifocal NETs with multiple recurrences can be treated with laparoscopic antrectomy to normalize gastrin level, and cause regression of the NET and neuroendocrine hyperplasia. Type 2 NETs, which are associated with Zollinger-Ellison syndrome with MEN1, comprise only about 5% of gastric NETs, thereby making them the least common type. These tumors are almost invariably multifocal. Unlike type 1, which is also associated with elevated gastrin, the gastric mucosa is hypertrophic rather than atrophic. The type 2 lesions are usually small and of low to intermediate aggressiveness, best treated with endoscopic excision. Type 3 NET is sporadic gastric NET. In comparison to types 1 and 2 NET, type 3 NETs are not associated with hypergastrinemia and are usually of higher stage/grade, giving a worse prognosis. They can produce an atypical carcinoid syndrome. In most cases, patients require a partial gastrectomy, as in adenocarcinoma of the stomach.  

Small Bowel and Cecum Neuroendocrine Tumors

Patients with small bowel and cecal NET may present with symptoms that are related to the tumor mass or with symptoms caused by biologically active peptides that are released into the circulatory system (e.g., carcinoid syndrome).20,21 Abdominal primary tumors or metastases can cause abdominal pain and even intestinal or bowel obstruction owing to growing tumor and associated desmoplasia. A primary tumor may infrequently cause GI bleeding. Massive liver metastases can cause palpable hepatomegaly. 5-HIAA, serotonin, chromogranin A, pancreastatin, and neurokinin A are frequently measured biomarkers.

1526

CHAPTER 219  Neuroendocrine Tumors  

Most small bowel NETs arise within 60 cm of the ileocecal valve, approximately 25% having multiple primary small bowel foci. Localized tumors should be treated with right hemicolectomy or resection of the involved segment of small bowel depending on location. Metastatic potential increases with size, from less than 5% in tumors less than 1 cm to approximately 50% in tumors greater than 2 cm. When metastatic, 5-year overall survival ranges from 40 to 85% and 10-year survival ranges from 40 to 60%. Duodenal carcinoids can produce several types of hormonal syndromes. Duodenal gastrinomas can be sporadic or can occur in a background of MEN-1, where they are often associated with ZES. In MEN-1–associated cases, hyperplasia of background G cells can be observed. Duodenal somatostatinomas may be associated with psammoma bodies and type-1 neurofibromatosis. The nonfunctional duodenal NET can still produce calcitonin and serotonin. Ampullary NETs have metastatic potential even when small, and are also associated with neurofibromatosis. These tumors are treated with endoscopic resection when possible, although larger, more aggressive tumors, particularly those of the ampulla, may require a modified Whipple resection.  

Appendix Neuroendocrine Tumors

Appendix NETs are frequently found incidentally at the time of simple appendectomy. If less than 2 cm in diameter and there are no unfavorable pathologic features, prognosis is outstanding and no additional surgery is needed. However, if these tumors are greater than 2 cm, a right hemicolectomy should be done. Goblet cell carcinoid (GCC) contains both carcinoid elements and adenocarcinoma elements. These tumors arise most commonly in the appendix and because of their aggressive nature should be treated with a right hemicolectomy, even when less than 2 cm in diameter. These tumors can present as solid masses of GCC, as mucinous tumors, which can lead to peudomyxoma peritoneii, or as an adenocarcinoma, which overgrows the GCC. GCCs should be distinguished from “tubular carcinoids,” which often have focal mucin present but have an excellent prognosis and are treated as usual appendiceal NETs rather than as GCCs.  

Rectum Neuroendocrine Tumors

Hindgut NETs arise in the distal colon and rectum and are almost never associated with functional syndromes.22 These are frequently detected at screening colonoscopy and if less than 2 cm can be treated with endoscopic mucosal resection or transrectal excision with low risk of recurrence. However, if greater than 2 cm or of higher histologic grade, a low anterior resection of the rectum or anterior posterior resection of the rectum is necessary, and metastatic risk is greater than 50%. Mixed adeno-neuroendocrine carcinomas may also rarely occur in the colon, containing high-grade neuroendocrine and adenocarcinoma elements, which should be differentiated from high-grade adenocarcinoma of the rectum with neuroendocrine differentiation (latter not a neuroendocrine tumor).  

Lung Neuroendocrine Tumors

Lung NETs (bronchial carcinoids) comprise 1 to 2% of lung malignancies (Chapter 182) in adults and 20 to 30% NET cases.23 The annual incidence is 1.35 per 100,000 population/year. There is no definite relationship to cigarette smoking. Among the lung NETs, 90% are sporadic, 5% are associated with MEN-1, and occasional cases arise from diffuse idiopathic pulmonary neuroendocrine cell hyperplasia (DIPNECH). There is a 5:1 female preponderance. Typical carcinoid (low-grade NE) has a mitotic count less than 2 per 10 high-powered fields and no necrosis. Atypical carcinoid (intermediate grade) has a mitotic count of 2 to 20 per 10 high-powered fields and/ or focal necrosis. Staging is exactly the same as lung cancer (Chapter 182). Carcinoid syndrome (see later) with serotonin production occurs in less than 10% of cases, but can be severe even in the absence of metastatic disease. Histamine-mediated atypical flushing, which can last over 30 minutes, is bright red and can involve limbs and upper trunk, and is associated with sweating, tearing, wheezing, and burning, and sometimes rosacea. Cushing syndrome from ACTH or CRH secretion occurs in less than 5% of patients with lung carcinoid tumors. This can be cured by resection of a localized tumor. Laparoscopic bilateral adrenalectomy is the treatment of choice, if caused by unresectable or metastatic NET, because medical management is usually inadequate. Acromegaly from ectopic production of GHRH can occur with bronchial NET, and often responds to somatostatin analogs or surgical debulking. DIPNECH is characterized by infiltration of lung parenchyma with neuroendocrine cells and tumorlets (1 cm in diameter; macroadenomas) by the time they are detected. Male patients with prolactin-secreting macroadenomas usually present with hypogonadism, erectile dysfunction, and visual manifestations from suprasellar extension. In small tumors, hypogonadotropic hypogonadism may be due to suppressive effects on GnRH described earlier, but in large tumors, it also may be due to a mass effect damaging the nonneoplastic gonadotrophs. Large non–prolactin-secreting pituitary tumors (growth hormone, adrenocorticotropic hormone, glycopeptide, and null cell) also may produce gonadotropin insufficiency from damage to the adjacent normal pituitary gland (Chapter 211), resulting in decreased serum LH and testosterone levels.  

CLINICAL MANIFESTATIONS AND DIAGNOSIS OF HYPOGONADISM

The diagnosis is based on clinical symptoms and signs and a reduced serum testosterone level. The reference range of serum total testosterone in a young adult male population varies across different laboratories and populations. The harmonized normal range in a healthy nonobese population of European and American men from 4 large cohort studies, 19 to 39 years, is 264 to 916 ng/dL.

History

The medical history should focus on testicular descent, pubertal development, shaving frequency, changes in body hair, and present and past systemic illnesses. A complete sexual history includes changes in libido, erectile and ejaculatory functions, frequency of masturbation, coital activity, and fertility (including that of present and previous partners). Information should be obtained on previous orchitis, sinopulmonary complaints, sexually transmitted diseases, human immunodeficiency virus (HIV) status, genitourinary infections, and previous surgical procedures that might affect the reproductive tract (e.g., vasectomy, hernia repair, prostatectomy, varicocele ligation). Social history includes tobacco and alcohol intake. Medication and self-prescribed drug history includes recreational drugs; opioids; anabolic steroids; glucocorticoid use; 5α-reductase inhibitors; and psychiatric, antihypertensive, antiandrogenic, cytotoxic, and alternative medicine therapies; environmental toxins; and exposure to heat (including saunas and Jacuzzis) and irradiation.

Physical Examination

The general physical examination is supplemented by height and span measurements; characterization of facial, pubic, and body hair distribution; presence of acne and facial wrinkling; breast examination for gynecomastia; assessment of muscle mass and adiposity; measurement of penile length and urethral meatus localization; digital rectal prostate examination; and visual field assessment if secondary hypogonadism is suspected. The scrotal examination should include an assessment of midline fusion (e.g., bifid scrotum, hypospadias); testicular size and consistency; presence of intratesticular masses; abnormalities of the epididymis; bilateral presence of vas deferens; and varicoceles, hydroceles, or hernias. Normal testicular size ranges from 3.6 to 5.5 cm in length, 2.1 to 3.2 cm in width, and 15 to 35 mL in volume in white and black men. Asian men have a slightly smaller mean testicular size. A decrease in testicular volume usually implies decreased spermatogenic cells because the seminiferous tubules account for more than 80% of testicular volume.

Laboratory Studies

Because there is a strong diurnal rhythm in testosterone secretion in young men (highest in the morning), testosterone, LH, and FSH are routinely determined from morning blood samples. There is a broad range of reference values of these hormones due partly to measurement variability but also influenced by the selection criteria for the reference population. Most hospital laboratories have used immunoassay methods to measure serum testosterone, which may lack precision at low serum testosterone levels. Recent studies suggest that methods using liquid or gas chromatography and mass spectrometry give more accurate results even at very low serum testosterone levels. The U.S.

1542

CHAPTER 221  THE TESTIS AND MALE HYPOGONADISM, INFERTILITY  

Centers for Disease Control (CDC) has developed an external system to harmonize laboratory measurements to a validated standard method. Total testosterone measurements may be misleading indicators of Leydig cell secretory status in conditions in which SHBG levels are abnormal (see earlier section). In these circumstances, a measurement of free testosterone (by an equilibrium dialysis method), bioavailable testosterone (consisting of free plus albumin bound), or calculated free testosterone (by total testosterone and SHBG measurements) may be useful. Elevated LH and FSH levels distinguish primary from secondary hypogonadism (both have low serum testosterone levels), but many older men with low serum testosterone levels may have a normal LH concentration. Serum prolactin levels should be measured in all low testosterone, low LH cases (hypogonadotropic hypogonadism) and in men with known pituitary mass lesions, or galactorrhea. DHT is measured in cases of abnormal differentiation of the genitalia and when 5α-reductase deficiency is suspected. Serum estradiol should be measured in cases of gynecomastia. Assessment of other testosterone precursors and products may be required in special circumstances, including suspected congenital enzyme defects. The semen analysis is the “cornerstone” of the laboratory examination for male infertility. The following rules on measurement of serum testosterone apply to most young and middle-aged men thought to have hypogonadism. If a morning serum total testosterone level is repeatedly below 230 ng/dL (8 nmol/L), and he has symptoms or signs compatible with low testosterone state, the patient is probably hypogonadal, and testosterone replacement is indicated. If the serum testosterone level is between 230 and 320 ng/dL with normal serum LH levels, the patient may or may not be clinically hypogonadal and androgen replacement may not improve the symptoms (e.g., sexual dysfunction). Thus, when serum total testosterone is borderline and LH is not increased, one of the measurements of bioactive testosterone is indicated (e.g., free testosterone). The guidelines for men older than 60 years are less certain. Because SHBG levels are often increased, total testosterone levels may overestimate the biologically active forms. A serum total testosterone level above 350 to 400 ng/dL indicates that hypogonadism is very unlikely to be the cause of the symptoms, and the clinician should look for other etiologies for the symptoms.

TREATMENT  Indications

The main medical indication for androgen replacement therapy is male hypogonadism (Table 221-4). In approximately 10% of men with idiopathic hypogonadism reversed by testosterone therapy, the reversal is sustained after therapy is stopped. This suggests that some patients with low serum testosterone may have a transient cause of the deficiency. Beneficial effects of testosterone replacement have been demonstrated in elderly men with relatively low serum testosterone levels. Testosterone replacement therapy (up to 3 years) decreases fat mass, increases lean body mass, improves strength, and increases bone mineral density in older men. A1  Treatment with testosterone gel sufficient to raise serum testosterone to the mid-normal young male range improves sexual activity libido, erectile function, and mood, A2  A3  corrects unexplained anemia, A4  and increases bone mineral density and strength. A5  The effect on physical activity is modest at best, and no improvement is seen in cognition. A6  Testosterone-treated men have significantly increased noncalcified plaque volume by coronary computed tomographic angiography, A7  but cardiovascular adverse events were not increased over the less than 3 year or so duration of the various randomized trials. A8  In cohort studies with longer-duration follow-up, however, the data on cardiovascular risk are less clear.7-9 Testosterone replacement therapy does not appear to affect lower urinary tract symptoms. A9  At present, testosterone treatment is not recommended for older men with or suspected of having prostate cancer, moderate-to-severe heart failure, severe ,

uncorrected sleep apnea, or high red blood cell mass.10-12 A digital rectal examination should be performed, prostate-specific antigen level determined, and symptoms of severe urinary tract obstruction evaluated before testosterone treatment is instituted. In older men with DHEA deficiency, oral administration of DHEA raises serum DHEA and DHEA-S concentrations to the levels found in young men. However, serum testosterone levels do not change, and patients have not reported beneficial effects on quality of life, sexual function, mood, body composition, or exercise capacity. A10  Absolute contraindications to androgen replacement therapy include carcinoma of the prostate and the male breast. Androgens should be used with caution in older men with an enlarged prostate and urinary symptoms, elevated hematocrit, and sleep-related breathing disorders.

Testosterone Delivery Systems

The various methods of delivering testosterone treatment are shown in Table 221-5. Testosterone esters, such as testosterone enanthate (or cypionate) injections, are widely used in the United States and throughout the world. The recommended dose is 150 to 200 mg administered intramuscularly once every 2 to 3 weeks. More recent studies showed that these medium long–acting testosterone esters can be administered by subcutaneous (SC) injections weekly, producing more stable testosterone levels compared to intramuscular injections. Testosterone undecanoate injections, 750 mg administered every 10 weeks or 1000 mg every 12 weeks, are the preferred testosterone replacement therapy in many parts of the world but not in the United States because of the regulatory requirement of an observation period for about 30 minutes after injection for cough episodes. Modified 17α-alkylated androgens (methyl testosterone and many anabolic steroids), which are available in oral preparations, are not recommended as androgen replacement. These agents may lead to abnormalities in liver function, marked decreases in high-density lipoprotein cholesterol, and increases in total cholesterol levels compared with the testosterone esters. Oral testosterone undecanoate capsules have recently been approved in the United States. The capsules, containing from 158 to 396 mg testosterone undecanoate in a selfemulsifying drug delivery system, have to be taken twice a day with food. Transbuccal delivery of testosterone by mucoadhesive tablets (30 mg applied twice daily) results in physiologic-range testosterone levels through direct absorption into the systemic circulation, thus avoiding first-pass effects on the liver. The tablets may be dislodged from the buccal mucous membrane. Implants (pellets) of crystalline testosterone are available for chronic treatment of hypogonadism. Serum testosterone levels are maintained in the physiologic range for 4 to 6 months depending on the number of pellets that are inserted SC. Implants are not usually used but are gaining some popularity with urologists in the United States; they are widely used in Australia and the United Kingdom.

TABLE 221-5 ANDROGEN PREPARATIONS ROUTE

DOSE AND FREQUENCY OF ADMINISTRATION

Oral*

Testosterone undecanoate

158 to 396 mg PO two or three times daily with food

Buccal

Transbuccal testosterone, mucoadhesive tablets

30 mg two times daily

Injection

Testosterone enanthate and cypionate

100 mg/wk IM or 150-200 mg IM every 2 wk 50 to 100 mg/week SC 750-1000 mg IM every 10-12 wk

Testosterone undecanoate Implant

Testosterone implants

75-mg pellets (in United States), 6-10 inserted once every 4-6 months

Transdermal

Nonscrotal patch Androderm

Two patches, each delivering testosterone 2.5 mg/day; or one patch delivering testosterone 5 mg/day One patch delivering testosterone 5 mg/day

TABLE 221-4 INDICATIONS FOR ANDROGEN THERAPY Androgen deficiency (hypogonadism) (symptoms, signs and low testosterone level) Microphallus (neonatal) Delayed puberty in boys Angioneurotic edema Other possible uses or under investigation Hormonal male contraception Sarcopenia associated with cancer, human immunodeficiency virus infection, chronic infection, frailty in older men and women Hypoactive sexual disorder in postmenopausal women

PREPARATION

Testoderm TTS Transdermal gels

AndroGel or Testogel; Testim; Axiron; Fortesta

1 to 2% gel applied once daily delivering 50-100 mg testosterone on skin and 5 to 10 mg to body

*Oral modified 17α-alkylated androgens such as methyltestosterone, fluoxymesterone, oxymetholone, stanozolol, and oxandrolone are not recommended for the treatment of androgen-deficient states because of potential hepatotoxicity and adverse effects on serum lipids. IM = intramuscularly; PO = orally; SC = subcutaneously.

CHAPTER 221  THE TESTIS AND MALE HYPOGONADISM, INFERTILITY  

TABLE 221-6 ANDROGEN THERAPY: RISKS VERSUS BENEFITS BENEFITS Development or maintenance of secondary sex characteristics Improved libido and sexual function Increased muscle mass and strength Increased bone mineral density Improves anemia Decreased body and visceral fat Improved mood Effect on cognition (?) Effect on vitality and quality of life (?) Decreased cardiovascular disease risk (epidemiologic studies); clinical study no benefits/risk

TABLE 221-7 MALE INFERTILITY: BASIC LABORATORY TESTS

RISKS Fluid retention Gynecomastia Acne, oily skin Increased hematocrit, erythrocytosis Decreased high-density lipoprotein cholesterol (oral agents produce more effect) Sleep apnea Aggressive behavior (?) Prostate disease Benign prostatic hyperplasia (?) Carcinoma of prostate (aggravate existing cancer) Increased cardiovascular adverse events in one study in frail elderly men with multiple comorbid conditions

SEMEN ANALYSES Volume, pH Microscopy: Agglutination, debris Sperm: Concentration, motility, morphology, vitality Leukocytes Immature germ cells Sperm autoantibodies (sperm and semen biochemistry, sperm function tests)

HORMONE ANALYSES (IN PATIENTS WITH ABNORMAL SEMEN ANALYSES) Serum luteinizing hormone and follicle-stimulating hormone Serum testosterone If luteinizing hormone and testosterone levels are low, serum prolactin

TABLE 221-8 SEMEN ANALYSIS: REFERENCE RANGE FROM FERTILE MEN* PARAMETER Semen volume

Transdermal testosterone delivery through skin patches and gels have been available in the United States for over 15 years. The nonscrotal patches deliver 5 mg/day of testosterone, which is the physiologic production rate. These patches deliver levels of testosterone within the normal range but have a high incidence of skin irritability (redness, swelling, and blisters). Hydroalcoholic and nonalcoholic testosterone gels/lotion have been developed for transdermal application and have become the most widely used testosterone formulations in the United States. The usual dosage is 50 to 100 mg of 1, 1.62, and 2% testosterone gel applied daily to the skin, delivering 5 to 10 mg of testosterone to the body. Transdermal delivery results in a more consistent serum concentration and causes little skin irritation. Transfer from the user to other people is possible during routine use and may be a concern if there is close skin contact with women and children. Protective clothing or a shower is necessary to avoid transferring testosterone through skin-to-skin contact. Table 221-6 shows the benefits and potential side effects of androgen treatment. In hypogonadal men, androgen replacement leads to the development and maintenance of secondary sexual characteristics. Testosterone has important anabolic effects on muscle and bone and improves libido and sexual dysfunction. It has less effect on erectile dysfunction. It has no major short-term effects on the prostate gland but increases prostatic volume by an average of about 4.5 mL after 12 months of testosterone gel treatment.13

Sperm Concentration Total count Motility Morphology Vitality (live) Leukocytes



DEFINITION

Infertility is defined as the failure of a couple to achieve pregnancy after at least 1 year of frequent unprotected intercourse. If a pregnancy has not occurred after 3 years, infertility will most likely persist without medical treatment.  

EPIDEMIOLOGY

Studies in the United States and Europe showed a 1-year prevalence of infertility in 15% of couples. The prevalence in developing countries is likely to be higher because of the higher prevalence of genital tract infection. Of subfertility cases, 30 to 35% can be attributed to predominantly female factors, 25 to 30% to male factors, and 25 to 30% to problems in both partners.  

PATHOBIOLOGY

Hypothalamic-pituitary disorders are infrequent causes of male infertility and are discussed in the section on hypogonadism and androgen deficiency. Testicular disorders are the most frequent identifiable cause of infertility (see Table 221-2). Y chromosome microdeletions are increasingly recognized as a genetic cause of azoospermia and severe oligozoospermia. Up to 25% of infertile men have microdeletions in the long arm of the Y chromosome, many of which map to the Yq11 region of the chromosome, which is called the azoospermic factor (AZF). Mutations in the AZFa and b regions are associated with azoospermia, whereas mutations of AFZc region may be associated with oligozoospermia. The gr/gr deletion removes a large segment of the AFZc gene and represents a significant risk factor for oligozoospermia in some but not all populations. Mutations in the testis-expressed 11 gene (TEX11) and other genes such as NR5A1, DMRT1, have been described and are a common

REFERENCE RANGE >1.5 mL >15 million/mL >39 million/ejaculate >40% motile >32% progressively motile >4% normal† >58% 30-40 mIU/mL) Gonadal dysgenesis With stigmata of Turner syndrome Pure (46,XX or 46,XY) Mixed Ovarian failure with normal ovarian development Genetic disorders Autoimmune disorders Gonadotropin receptor or postreceptor defects (resistant ovary or Savage syndrome?) Enzymatic defects (17α-hydroxylase deficiency, galactosemia) Physical causes: irradiation, chemotherapeutic agents, viral agents Idiopathic

Hypogonadotropic or normogonadotropic hypogonadism (LH and FSH 10 mIU/mL) Isolated gonadotropin deficiency In association with midline defects (Kallmann syndrome) Independent of associated disorders Neoplasms of the hypothalamic-pituitary axis Craniopharyngiomas Pituitary tumors Others Infiltrative processes (Langerhans-type histiocytosis) Idiopathic hypopituitarism “Hypothalamic” forms of amenorrhea Psychogenic Exercise associated Associated with malnutrition Anorexia nervosa Miscellaneous disorders Prader-Labhart-Willi syndrome Laurence-Moon-Bardet-Biedl syndrome Primary hypothyroidism Constitutional delayed puberty ASYNCHRONOUS PUBERTAL DEVELOPMENT Incomplete forms of androgen insensitivity Complete forms of androgen insensitivity HETEROSEXUAL PUBERTAL DEVELOPMENT Polycystic ovary syndrome Congenital adrenal hyperplasia (female pseudohermaphroditism) 21-Hydroxylase deficiency 11β-Hydroxylase deficiency 3β-ol-Hydroxysteroid dehydrogenase deficiency Male pseudohermaphroditism due to 5α-reductase deficiency Male pseudohermaphroditism due to partial androgen insensitivity Mixed gonadal dysgenesis Androgen-producing neoplasms Ovarian Adrenal Cushing syndrome

*No development by age 13 yr, absence of menarche by age 15 yr, or passage of ≥5 yr from breast budding without menarche. FSH = follicle-stimulating hormone; LH = luteinizing hormone.

peripheral, in which case it is considered gonadotropin-releasing hormone (GnRH)–independent precocious puberty or precocious pseudopuberty. Careful questioning of the patient and her parents may indicate the inadvertent ingestion or absorption of sex steroids (iatrogenic or factitious). As many as 20% of individuals with true precocious puberty have one of several organic brain diseases, including any of several neoplasms, tuberous sclerosis, neurofibromatosis, encephalitis, meningitis, vascular malformations, and hydrocephalus. Because of the seriousness of intracranial lesions, girls with precocious puberty must have radiographic evaluation of the CNS, most effectively by MRI. In at least 75% of girls with true precocious puberty, however, no cause is identified (idiopathic or constitutional). The physical examination may also provide critical information about the cause of the precocious development. Cutaneous café au lait spots, facial asymmetry, polyostotic fibrous dysplasia and other skeletal abnormalities, cranial nerve deficits, and multiple ovarian follicular cysts suggest McCune-Albright syndrome (Chapters 211 and 234) in a girl with precocious development. It is now known that various clones of cells in the endocrine glands of girls with this disorder function autonomously with respect to cyclic adenosine monophosphate production as a consequence of a mutation within exon 8 of the G protein α–subunit. This same mutation probably accounts for the bone lesions and café au lait hyperpigmentation. Other endocrine cells may be similarly affected and lead to pituitary adenomas (usually secreting growth hormone), hyperthyroidism, and, rarely, adrenal hyperplasia. Studies on the etiologic causes for precocious puberty are in their infancy, but one molecule, kisspeptin, is known to be involved in pubertal

development. Kisspeptin and its receptor GPR54 are essential regulators of GnRH-induced gonadotropin secretion and pubertal onset. Activation results in stimulation of the HPO axis, and elevated kisspeptin levels are associated with precocious puberty. Additionally, kisspeptin loss-of-function mutations result in normosmic idiopathic hypogonadotropic hypogonadism. Abdominal and rectal examination may reveal a mass, suggesting an adrenal or ovarian tumor. Because palpable ovarian cysts may rarely develop before ovulation in true precocious puberty, the presence of a mass does not confirm the diagnosis of precocious pseudopuberty. When vaginal bleeding is the only sign of development, the diagnosis of sexual precocity should be suspect. Common causes of bleeding in this age group include irritation from a vaginal infection or foreign body, sexual assault, prolapse of the urethral meatus, and ingestion of estrogen-containing medications (most commonly, oral contraceptive preparations). A vaginal or cervical neoplasm is also a rare possibility. Thus vaginal bleeding requires a vaginal examination, which is often best performed with the patient under anesthesia, before further evaluation is undertaken. Heterosexual precocity in an apparent prepubertal female is almost always due to congenital adrenal hyperplasia or to an androgen-secreting adrenal or ovarian neoplasm. Only rarely must another disorder of sexual differentiation be considered (Chapter 220). It is important to examine the external genitalia carefully, because congenital adrenal hyperplasia is usually associated with some degree of sexual ambiguity. Excessive androgens produced endogenously by abnormal fetal adrenal glands in utero or diffusing across the placenta to the fetus from the mother

1550

CHAPTER 222  Ovaries and Pubertal Development  

can virilize the external genitalia and result in female pseudohermaphroditism. The extent of virilization varies from only an enlarged clitoris to sexual ambiguity sufficient to make gender assignment difficult. Excessive maternal androgen secretion, typically from an ovarian or adrenal neoplasm, can lead to virilization of a female fetus. This occurs very rarely because of the great capacity of the placenta to aromatize naturally occurring androgens to estrogens. Virilization of a female fetus is much more likely to occur if a pregnant woman has ingested a synthetic steroid preparation with androgenic properties, because synthetic compounds generally cannot be aromatized. Excessive androgen secretion beginning in utero is usually associated with defective cortisol synthesis. As a consequence, pituitary corticotropin secretion is increased, resulting in congenital adrenal hyperplasia and excessive androgen secretion. The three different enzyme defects in the steroidogenic pathway that can lead to virilization of the female fetus are described in Chapter 220. The most common form of congenital adrenal hyperplasia is 21-hydroxylase deficiency, accounting for the disorder in more than 90% of affected individuals. The defect may vary from partial to complete deficiency of the enzyme.

Diagnostic Tests Measurement of Peptide and Steroid Hormones

Increased levels of immunoreactive human chorionic gonadotropin (HCG) may suggest an HCG-secreting neoplasm, most commonly an ovarian teratoma or dysgerminoma.8 In such cases, the HCG, which is antigenically and biologically similar to LH, stimulates ovarian steroid secretion and pseudopubertal development. Because even specific LH immunoassays show some cross-reactivity with HCG, values for serum LH may be elevated in individuals with HCGsecreting tumors. Immunoreactive HCG is always elevated in the presence of such tumors. Levels and ratios of FSH and LH typical of pubertal as opposed to prepubertal girls help in the diagnosis of true precocious puberty. Timed urine collections rather than blood samples can be used to measure gonadotropin secretion if necessary. The use of exogenous GnRH to stimulate endogenous LH and FSH secretion can help differentiate gonadotropin-dependent from gonadotropin-independent precocious puberty and is regarded as the “gold standard” in the diagnosis of central precocious puberty. If GnRH is not available, a GnRH analog can be substituted. Excessively high circulating levels of estrogen (>100 pg estradiol) suggest an estrogen-producing neoplasm or a functioning ovarian cyst. High levels of serum testosterone suggest an ovarian source of excess androgen in girls with heterosexual development, whereas increased levels of dehydroepiandrosterone or its sulfate (the principal precursors of 17-ketosteroids) suggest an adrenal source. High levels of serum l7-hydroxyprogesterone imply congenital adrenal hyperplasia secondary to 21-hydroxylase deficiency, whereas high levels of serum 11-deoxycortisol imply an 11β-hydroxylase deficiency (Chapter 220). In congenital adrenal hyperplasia, these hormone levels should decrease promptly after the oral administration of suppressive doses of dexamethasone. Suppression in response to exogenous corticoids occurs much less consistently in individuals with adrenal cortical adenomas and carcinomas (Chapter 214) and rarely in those with ovarian androgen-secreting neoplasms.

Additional Studies

Imaging of the CNS is the most important test if true precocious puberty is present or if there are any neurologic deficits. Ultrasonography of the adrenals and ovaries or computed tomography (CT) of the adrenals may be indicated to confirm clinical suspicions. In girls with ovarian or adrenal neoplasms, the tumor can almost always be localized radiographically. Catheterization of the ovarian and adrenal veins and measurements of the effluent steroids from each gland should be pursued only when CT, ultrasonography, or MRI fails to identify a suspected neoplasm. Radiographic estimation of bone age is also indicated and serves as a useful tool to follow the results of treatment.

TREATMENT  Treatment of precocious puberty should be initiated promptly so that the patient’s ultimate height is not compromised as a result of sex steroid–induced premature epiphyseal closure and to prevent or attenuate emotional disturbances in the patient and her parents.9-11 GnRH analogs are now the preferred therapy for suppressing gonadotropin secretion, and they also may prevent early bone maturation. No randomized trials have been conducted, but there is universal acknowledgment that GnRH

analogs increase ultimate adult height in girls presenting before 6 years of age. Two unresolved issues are whether to initiate treatment with GnRH analogs in girls between 6 and 8 years of age and at what age to stop treatment. In addition, some data suggest that oral contraceptive pills or metformin may improve hirsutism and oligomenorrhea in the teenage years.12 The analogs are not effective in children with McCune-Albright syndrome, and ketoconazole and testolactone have been only marginally successful. Aqueous depot medroxyprogesterone acetate (100 to 200 mg intramuscularly every 2 to 4 weeks) also may be used to suppress gonadotropin secretion; however, it does not always prevent premature epiphyseal closure and the resultant short stature. Individuals with CNS or steroid-secreting neoplasms must undergo therapy appropriate for the particular lesion. Girls with congenital adrenal hyperplasia are appropriately managed with glucocorticoids (plus mineralocorticoids when indicated), as outlined in Chapter 220.



Delayed Puberty

Girls who have no evidence of thelarche by age 13 or who fail to undergo menarche by age 15 have delayed puberty and should be evaluated.13 Ovarian failure, congenital absence of the uterus and vagina, and constitutional delay constitute about two thirds of cases in large series. Because of the anxiety generated by delayed puberty, some evaluation is always indicated regardless of the age of the patient. When pubertal development progresses normally but menstruation does not begin, an abnormality in the genital tract should be considered. Congenital malformations of the müllerian ducts are uncommon, occurring in 0.02% of all women. Most do not cause amenorrhea, and many do not impair reproduction. The anomalies associated with amenorrhea vary in severity from an imperforate hymen to complete aplasia of all müllerian duct derivatives, with vaginal atresia. Although aplasia generally involves all the müllerian duct derivatives, defects may involve only a single part of the distal genital tract. Family aggregates of the most common disorders of müllerian differentiation in females—müllerian aplasia and incomplete müllerian fusion—do occur and are best explained by polygenic or multifactorial inheritance. The HOX genes, a family of regulatory genes that encode transcription factors, are essential for proper development of the müllerian tract. A müllerian duct anomaly is suggested by (1) normal levels of serum gonadotropins and steroids, (2) an abnormal outflow tract, (3) a history of cyclic abdominal pain with or without a palpable mass, and (4) normal development of secondary sex characteristics. Normal ovarian function still induces endometrial growth and shedding after menarche if the uterus is normal. In the absence of a normal outflow tract, however, the menstrual effluent is retained and may or may not escape into the abdominal cavity. Free in the abdominal cavity, the effluent may cause endometriosis. Constrained to the uterine cavity, the effluent causes hematometra and a large abdominal mass. In the absence of a mass or cyclic pain, karyotyping is indicated in girls with evidence of an abnormal genital tract to rule out disorders of sexual differentiation (Chapter 220). Such disorders, however, almost never occur together with completely normal pubertal development. In girls with a normal karyotype and a genital tract anomaly, examination under anesthesia and diagnostic laparoscopy should be undertaken to delineate the extent of the defect. When the abnormality consists of an imperforate hymen or transverse vaginal septum only, surgical restoration can be accomplished relatively simply. Attempts to provide an outflow tract for the uterus should not be undertaken if there is no cervix because of the high risk of recurrent pelvic infection. Even with a functional cervix, the construction of an outflow tract that permits successful pregnancy is unlikely. A functional vagina can be constructed surgically or by the daily use of ever-larger dilators. To prevent shrinkage and scarring, surgery should be deferred until the patient is willing to use dilators on a daily basis or she is about to become sexually active. Other causes of delayed puberty and primary amenorrhea are the same as those that cause amenorrhea in older women (Chapter 223).14 When no apparent cause of delayed development is found, constitutional delayed puberty must be entertained as a diagnosis of exclusion.15 A strong family history of delayed maturation supports this presumption. Small doses of estrogen can be administered to induce some pubertal development, but this may obscure a pathologic cause of the delay and may compromise linear growth and ultimate height.  

Asynchronous Pubertal Development

Asynchronous pubertal development is characteristic of male pseudohermaphroditism due to androgen insensitivity, especially complete testicular

feminization. This syndrome of androgen insensitivity is inherited as either an X-linked recessive trait or a sex-limited autosomal dominant trait. Despite the presence of intra-abdominal or inguinal testes, there is complete failure of virilization. Affected individuals develop breasts (but only to Tanner stage 3) and a typical female habitus with unambiguous female external genitalia, but with the absence of internal female structures and generally only a foreshortened, blind-ending vagina. Little or no pubic and axillary hair develops. The karyotype is 46,XY in these individuals. Circulating testosterone levels are equivalent to or higher than those found in normal men, LH levels are elevated, and FSH levels are normal compared with those in menstruating women. For a more detailed description, see Chapter 220.  



Heterosexual Pubertal Development

POLYCYSTIC OVARY SYNDROME

Polycystic ovary syndrome (PCOS), by far the most common cause of heterosexual pubertal development, is associated with the development of some secondary sex features characteristic of males at the normal age of puberty.16 Feminization occurs in affected girls, and they develop normal breasts and a typical female habitus, but masculinization also occurs (in contrast, girls with congenital adrenal hyperplasia generally show little if any female development at puberty). A heterogeneous syndrome, PCOS typically begins at or near puberty with hirsutism and irregular menses from the time of menarche. Many girls who develop PCOS are overweight in childhood, and obesity is clearly a risk factor. It now appears that many girls who develop PCOS have alterations in insulin signaling.17 Menarche may be delayed in a few cases, so young women may present with primary amenorrhea. Basal LH levels tend to be somewhat elevated in perhaps two thirds of cases, and circulating levels of all androgens are moderately elevated. Some degree of insulin resistance is commonly present as well, and hypercholesterolemia may predispose to cardiovascular disease later in life. This is discussed more completely in Chapter 223.  

CONGENITAL ADRENAL HYPERPLASIA

Congenital adrenal hyperplasia is generally diagnosed before puberty, and heterosexual precocious pseudopuberty is typical.18 However, if the defect is mild and changes to the external genitalia are minimal, masculinization may occur at the expected age of puberty. This attenuated or nonclassic form of 21-hydroxylase deficiency seems to occur in families with a strong history of hirsutism. Affected girls generally have some defeminization, with flattening of the breasts, severe hirsutism, relatively short stature, and obesity. For a more detailed description, see Chapter 220.  

MIXED GONADAL DYSGENESIS

Mixed gonadal dysgenesis designates asymmetrical gonadal development, with a germ cell tumor or a testis on one side and an undifferentiated streak, rudimentary gonad, or no gonad on the other.19 The extent of genital virilization before puberty is variable in this rare disorder. Most individuals are reared as girls, in whom virilization occurs at puberty; some may note breast development as well. Affected individuals generally have a mosaic karyotype, with 45,X/46,XY being most common. Short stature and other stigmata associated with a 45,X karyotype in Turner syndrome are less common in patients with tumors than in patients with testes. Gonadectomy is indicated in all individuals with a Y chromosome to eliminate the increased neoplastic potential of such dysgenetic gonads and in all patients in whom virilization occurs at puberty to remove the source of androgen. Estrogen replacement therapy is warranted after gonadectomy. Other causes of male pseudohermaphroditism associated with heterosexual pubertal development are described in Chapter 220.  

OTHER CAUSES

An androgen-producing adrenal neoplasm or Cushing syndrome may occur rarely during the pubertal years and lead to heterosexual development (Chapter 214). GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 222  Ovaries and Pubertal Development  

GENERAL REFERENCES 1. Manfredi-Lozano M, Roa J, Tena-Sempere M. Connecting metabolism and gonadal function: novel central neuropeptide pathways involved in the metabolic control of puberty and fertility. Front Neuroendocrinol. 2018;48:37-49. 2. Castellano JM, Tena-Sempere M. Metabolic control of female puberty: potential therapeutic targets. Expert Opin Ther Targets. 2016;20:1181-1193. 3. Almeide M, Laurent MR, Dubois V, et al. Estrogens and androgens in skeletal physiology and pathophysiology. Physiol Rev. 2017;97:135-187. 4. Biggs WS, Marks ST. Diagnosis and management of adnexal masses. Am Fam Physician. 2016;93: 676-681. 5. Masch WR, Daye D, Lee SI. MR imaging for incidental adnexal mass characterization. Magn Reson Imaging Clin N Am. 2017;25:521-543. 6. Klein DA, Emerick JE, Sylvester JE, et al. Disorders of puberty: an approach to diagnosis and management. Am Fam Physician. 2017;96:590-599. 7. Leka-Emiri S, Chrousos GP, Kanaka-Gantenbein C. The mystery of puberty initiation: genetics and epigenetics of idiopathic central precocious puberty (ICPP). J Endocrinol Invest. 2017;90:789-802. 8. Guaraldi F, Beccuti G, Gori D, et al. Management of endocrine disease: long-term outcomes of the treatment of central precocious puberty. Eur J Endocrinol. 2016;174:R79-R87. 9. Schoelwer M, Eugster EA. Treatment of peripheral precocious puberty. Endocr Dev. 2016;29: 230-239. 10. Chen M, Eugster EA. Central precocious puberty: update on diagnosis and treatment. Paediatr Drugs. 2015;17:273-281.

1551.e1

11. Durá-Travé T, Ortega Pérez M, Ahmed-Mohamed L, et al. Central precocious puberty in girls: diagnostic study and auxological response to triptorelin treatment. Endocrinol Diabetes Nutr. 2019. [Epub ahead of print.] 12. Al Khalifah RA, Florez ID, Dennis B, et al. Metformin or oral contraceptives for adolescents with polycystic ovarian syndrome: a meta-analysis. Pediatrics. 2016;137:1-14. 13. Wei C, Davis N, Honour J, Crowne E. The investigation of children and adolescents with abnormalities of pubertal timing. Ann Clin Biochem. 2017;54:20-32. 14. Matalliotakis M, Koliarakis I, Matalliotaki C, et al. Clinical manifestations, evaluation and management of hyperprolactinemia in adolescent and young girls: a brief review. Acta Biomed. 2019;90:149-157. 15. Varimo T, Miettinen PJ, Känsäkoski J, et al. Congenital hypogonadotropic hypogonadism, functional hypogonadotropism or constitutional delay of growth and puberty? An analysis of a large patient series from a single tertiary center. Hum Reprod. 2017;32:147-153. 16. Rothenberg SS, Beverley R, Barnard E, et al. Polycystic ovary syndrome in adolescents. Best Pract Res Clin Obstet Gynaecol. 2018;48:103-114. 17. Tola EN, Yalcin SE, Dugan N. The predictive effect of inflammatory markers and lipid accumulation product index on clinical symptoms associated with polycystic ovary syndrome in nonobese adolescents and younger aged women. Eur J Obstet Gynecol Reprod Biol. 2017;214:168-172. 18. Witchel SF. Congenital adrenal hyperplasia. J Pediatr Adolesc Gynecol. 2017;30:520-534. 19. Makhija D, Shah H, Tiwari C, et al. Mixed gonadal dysgenesis with an unusual “inverted” Y chromosome. Dev Period Med. 2016;20:178-180.

1551.e2

CHAPTER 222  Ovaries and Pubertal Development  

REVIEW QUESTIONS 1. Which statement is true regarding pubertal development? A . It is characterized by resetting the positive gonadal steroid feedback loop. B. Development of breast and pubic hair results from increasing ovarian estrogen production. C. Testosterone is the primary hormone mediating pubertal bone growth in boys. D. Prepubertal ovaries produce sex steroids. E. Ovaries produce low levels of deoxycorticosterone and androstenedione. Answer: D  Though produced at low levels, the prepubertal ovaries regulate pituitary production of FSH and LH. Option A is incorrect; pubertal development is characterized by resetting negative feedback loops. Option B is incorrect; pubic hair results from androgen production. Option C is incorrect; estradiol is the primary hormone for both girls and boys. Option E is incorrect; ovaries do not produce measurable deoxycorticosterone, whereas a primary ovarian product is androstenedione. See Ovarian Function in Childhood and Puberty. 2. Which statement is true regarding ovarian cysts? A . Identification of ovarian cyst is a concerning finding requiring workup. B. Palpable masses in a child or postmenopausal woman are not follicular growths. C. Ovarian cysts identified radiologically require gynecologic consultation. D. Ectopic pregnancy can be definitively diagnosed by ultrasound. E. CA-125 is used to diagnose ovarian cancer in all women with pelvic masses. Answer: B  Women outside of reproductive years do not produce follicles (that would ultimately result in an ovulated oocyte). Option A is incorrect; in the reproductive years, follicular development occurs in all women with menstrual cycles. Option C is incorrect; ovarian cysts, particularly in cycling women, are common and are not pathologic. Option D is incorrect; a pregnancy test is also required. Option E is incorrect; CA-125 may be helpful in postmenopausal women, but is nonspecific in reproductive-age women. See Ovarian Cysts and Adnexal Masses. 3. Which statement is true regarding precocious puberty? A . Diagnostic investigation is required if puberty initiates before 8 years of age. B. It is more common in boys than girls. C. Premature adrenarche requires diagnosis and treatment. D. Organic brain disease is a common cause of true precocious puberty. E. In the majority of girls, the cause of true precocious puberty can be identified.

Answer: D  Organic brain disease occurs in as many as 20% of patients with true precocious puberty. Option A is incorrect; requirement for workup is controversial, and it is often focused on girls younger than 6 years old. Option B is incorrect; it is more common in girls. Option C is incorrect; it is a normal variant and should be managed based on patient concern. Option E is incorrect; in at least 75%, no cause is identified. See Precocious Puberty. 4. Which statement is true regarding delayed puberty? A . It is defined as absence of thelarche by 13 years, or menarche by 15 years. B. Constitutional delay is diagnosed in less than 5% of cases. C. Evaluation is not indicated until 15 years of age. D. Most congenital anomalies of the müllerian ducts cause amenorrhea. E. HOX gene mutations are sufficient to disrupt müllerian tract development. Answer: A  The definition is correct as presented. Option B is incorrect; constitutional delay can be diagnosed in about two thirds of cases. Option C is incorrect; evaluation is concerned at any age if there is patient concern. Option D is incorrect; most have little impact on amenorrhea. Option E is incorrect; HOX gene mutations, though essential, are not sufficient. See Delayed Puberty. 5. Which statement is true regarding polycystic ovary syndrome? A . It is a rare cause of heterosexual pubertal development. B. It commonly begins in the third decade. C. It is commonly associated with ovulation of more than one oocyte per cycle. D. Obesity is a diagnostic criterion for polycystic ovary syndrome (PCOS). E. Abnormalities in insulin signaling are common. Answer: E  Insulin resistance is common, but not required. Option A is incorrect; it is a common cause of heterosexual pubertal development. Option B is incorrect; it commonly begins at or shortly after the initiation of puberty. Option C is incorrect; it is associated with anovulation or oligo-ovulation of one oocyte. Option D is incorrect; obesity can worsen PCOS, but it is not a diagnostic criterion. See Polycystic Ovary Syndrome.

1551

CHAPTER 223  Reproductive Endocrinology and Infertility  

223  REPRODUCTIVE ENDOCRINOLOGY AND INFERTILITY WILLIAM H. CATHERINO



  THE NORMAL MENSTRUAL CYCLE Endometrium

The endometrium undergoes histologic and cytologic changes that culminate with menstrual bleeding when the corpus luteum ceases to secrete progesterone (Fig. 223-1). The basal layer of the endometrium then regenerates the superficial layer of compact epithelial cells lining the uterine cavity and an intermediate layer of spongiosa. Both superficial layers are shed at menstruation. Endometrial glands proliferate under the influence of estrogen, and the mucosa thickens. In the luteal phase, the glands become coiled and secretory, with increased vascularity and edema of the stroma. When estradiol and progesterone decline, the stroma becomes edematous, endometrial and blood vessel necrosis occurs, and bleeding ensues. Local release of prostaglandins may initiate vasospasm with ischemic necrosis and uterine contractions that accompany menstrual flow. Prostaglandin synthetase inhibitors can relieve menstrual cramping. The histologic changes are characteristic; therefore endometrial biopsies can be used to characterize the stage of the cycle and to assess the tissue response to gonadal steroids.  

Cervix and Cervical Mucus

During the follicular phase, cervical vascularity, congestion, and edema increase as a result of estrogen. Cervical mucus increases in quantity (10- to 30-fold)

Pituitary cycle

LH

FSH

Ovarian cycle

Maturation of follicle

E2

Corpus luteum

Ovulation

P

Endometrial cycle

M –12

–8

–4

0

Follicular phase

+4

+8

+12

Luteal phase Ovulatory phase

FIGURE 223-1.  The idealized cyclic changes observed in gonadotropins, estradiol (E2), progesterone (P), and uterine endometrium during the normal menstrual cycle. The data are centered on the day of the luteinizing hormone (LH) surge (day 0). Days of menstrual bleeding are indicated by M. FSH = follicle-stimulating hormone; LH = luteinizing hormone. (From Endocrine and Metabolism Continuing Education Quality Control Program, 1982. Copyright American Association for Clinical Chemistry, Inc.)

CHAPTER 223  Reproductive Endocrinology and Infertility  

ABSTRACT

With an understanding of the biological basis of the normal menstrual cycle, outlined in the first section of the chapter, the diagnosis and management of abnormalities during the reproductive years can be addressed. Dysmenorrhea and endometriosis, premenstrual syndrome, abnormal uterine bleeding, and amenorrhea are discussed. Amenorrhea can be due to anatomic factors (disorders of sexual differentiation, intrauterine adhesions, gestational trophoblastic disease), chronic anovulation (CNS-hypothalamic-pituitary dysfunction, inappropriate steroid feedback as in polycystic ovary syndrome, thyroid or adrenal disorders), or ovarian “failure” (menopause, genetic abnormalities, physical and environmental factors, autoimmune disease). Multiple male factors and female factors can cause infertility. The approach to sexual function and dysfunction is also discussed.

KEYWORDS

infertility endometrium gonadotropin embryo ovulation menstrual cycle dysmenorrhea endometriosis premenstrual syndrome amenorrhea polycystic ovary syndrome sexual dysfunction

1551.e3

1552

CHAPTER 223  Reproductive Endocrinology and Infertility  

and in elasticity. So-called ferning becomes prominent. Progesterone stimulates cervical mucus thickening and loss of elasticity and ability to fern. These characteristics are useful in evaluating the stage of the cycle and the amount of estrogen present.  

Brain

Vagina

Low estrogen is associated with pale, thin vaginal epithelium. As estrogens rise, the number of cornified epithelial cells increases. Subsequently, progesterone decreases the percentage of cornified cells and increases the number of precornified intermediate cells. There is also increased cellular debris and clumping of shed desquamated cells. Histologic changes in the vaginal epithelium are sensitive indicators of estrogen status.  

Ovary

The ovaries produce a single dominant graafian follicle that grows and develops to the preovulatory stage during the follicular phase. This process is brought about by combined action of follicle-stimulating hormone (FSH) and luteinizing hormone (LH) on the follicle wall to increase estradiol biosynthesis. The LH surge acts on the preovulatory follicle to cause the secretion of the mature fertilizable oocyte. After ovulation, the follicle wall transforms into the corpus luteum, which produces progesterone and estradiol. If implantation does not occur, the corpus luteum undergoes luteolysis and stops hormone production. In the late luteal phase, another dominant follicle develops, and a new menstrual cycle begins.  

GnRH

Hypothalamus

Pituitary

LH

Chronology of Folliculogenesis

Theca

The preovulatory follicle begins its development when a primordial follicle is recruited into the pool of growing follicles. There are two major phases of folliculogenesis: the preantral (gonadotropin-independent) and the antral (gonadotropin-dependent) periods (see Fig. 223-1). The first phase is characterized by growth of the oocyte and granulosa proliferation. Preantral folliculogenesis proceeds slowly, requiring at least 300 days. During the second phase, granulosa and theca cells proliferate, and the antrum enlarges. The graafian follicle increases relatively rapidly as it develops. The mature graafian follicle that will ovulate requires 40 to 50 days to complete the antral phase.

Ovulation

At midpoint in the menstrual cycle, the preovulatory surges of LH and FSH act on the preovulatory follicle to initiate the events leading to ovulation (see Fig. 223-1). The LH surge induces meiotic maturation, a process that converts the oocyte into a fertilizable egg arrested at the second meiotic metaphase. During meiotic maturation, the granulosa cells next to the oocyte are stimulated by FSH to undergo cumulus expansion (Fig. 223-2). This is a prerequisite for the oocyte’s pickup and transport by the oviduct. The LH surge also stimulates production of proteolytic enzymes in the vicinity of the presumptive stigma. This process requires the LH stimulation of progesterone and prostaglandins, which are obligatory for stigma formation. After 36 hours, the fertilizable egg and surrounding cumulus cells are secreted through the stigma (see Fig. 223-1). A serum progesterone level higher than 3 ng/mL 1 week before menses is probably diagnostic of ovulation.1

Luteogenesis

Ovulation leads to changes in the granulosa and theca cells of the ovulated follicle that result in increased production of progesterone and estradiol during the first week of the luteal phase. This event, termed luteinization, is important for the formation and development of a secretory endometrium. Three major

Inhibin

A Aromatase

Ovary

Granulosa E2

E2 FIGURE 223-2.  The hypothalamic-pituitary-ovarian axis in the regulation of follicular maturation and steroidogenesis. A = androgens; E2 = estradiol; FSH = follicle-stimulating hormone; GnRH = gonadotropin-releasing hormone; LH = luteinizing hormone. (Modified from Endocrine and Metabolism Continuing Education Quality Control Program, 1982. Copyright American Association for Clinical Chemistry, Inc.)

Selection

The dominant follicle is selected from a cohort at the end of the luteal phase of the previous menstrual cycle. The selected follicle requires 20 days to develop to the ovulatory stage. Shortly after the midluteal phase of the cycle, the granulosa cells show a sharp increase in the rate of mitosis. The first indication of selection is that the granulosa cells continue dividing at a high rate. As a consequence of the high sustained mitotic rate and the progressive accumulation of follicular fluid, the dominant follicle undergoes remarkable growth. The increase in plasma FSH levels that begins at the end of luteal phase and continues through the early follicular phase evokes follicle selection. The concentration of FSH in the follicular fluid of the healthy (dominant) follicle increases but does not increase in the nondominant atretic follicles. The manner in which this selective increase in FSH is controlled is unknown. More than 99.9% of all follicles are not selected and undergo atresia.

FSH

physiologic mechanisms are responsible for luteinization: removal of luteinization inhibitors, secretion of LH by the pituitary, and delivery of high levels of cholesterol. The induction of steroidogenic acute regulatory protein StAR, P450c22, and 3β-hydroxysteroid dehydrogenase in the granulosa lutein cells leads to progesterone production by the corpus luteum. The two-cell, twogonadotropin mechanism is responsible for estradiol production. If implantation does not occur, the corpus luteum initiates luteolysis, leading to decreases in progesterone, estradiol, and apoptosis. When luteolysis occurs, another dominant follicle is selected, and a new menstrual cycle begins.

ABNORMALITIES OF THE REPRODUCTIVE YEARS



Dysmenorrhea and Endometriosis  

DEFINITION

Dysmenorrhea, defined as painful menstruation, affects about 50% of postpubertal women and can be classified as primary or secondary. Endometriosis, which may result in dysmenorrhea, infertility, and dyspareunia (i.e., painful intercourse), is the ectopic occurrence of endometrial tissue, most commonly within the abdominal cavity but sometimes in surgical scars, on the vulva, in the umbilicus, and elsewhere.  

PATHOBIOLOGY

Primary dysmenorrhea occurs only in ovulatory cycles. Prostaglandins produce dysmenorrhea by initiating painful, exaggerated uterine contractions and myometrial ischemia. Associated systemic symptoms include nausea, diarrhea, headache, and emotional changes. In secondary dysmenorrhea, there is a pathologic cause, with endometriosis being the most common. Other causes include pelvic inflammatory disease; congenital abnormalities, such as atresia

CHAPTER 223  Reproductive Endocrinology and Infertility  

of a portion of the distal genital tract and cystic duplication of the paramesonephric ducts; and cervical stenosis. Studies have suggested the possibility that the pain of endometriosis is caused by the presence of nerve fibers within ectopic endometrium.

TREATMENT  Prostaglandin synthetase inhibitors, such as naproxen, ibuprofen, mefenamic acid, and indomethacin, are used to treat primary dysmenorrhea. A1  If the dysmenorrhea persists, addition of an oral contraceptive to inhibit ovulation and limit prostaglandin release is generally effective. In cases in which the pelvic pain remains intractable, additional evaluation is warranted.2 If thorough evaluation of the gastrointestinal and urinary tracts fails to reveal a definitive cause, examination under anesthesia and diagnostic laparoscopy may be indicated. If endometriosis is diagnosed at laparoscopy, treatment varies according to the severity of the disease and the goals of the patient regarding fertility.3 It may be possible to fulgurate implants or to lyse adhesions. Available data do not demonstrate a benefit of excision over ablation at the time of initial diagnosis as the preferred approach to controlling symptoms. From this point onward, efforts should be directed toward treating endometriosis medically, with additional surgery deferred until infertility (if present) becomes manifest. Medical therapy can consist of continuous suppression with oral contraceptives, A2  progestins (oral, injectable, or implantable), A3  or gonadotropin-releasing hormone (GnRH) analogs or danazol for 3 to 6 months. GnRH analogs A4  A5  are currently the most frequently used form of medical suppressive therapy.4 After a course of treatment, use of oral contraceptive agents should probably be continued until fertility is desired. Conservative surgical resection of endometriotic tissue should almost always be deferred until it is established as the cause of infertility. Surgery may be required, however, for continuing severe pain, severe endometriosis, or large ovarian cysts containing endometriosis (endometriomas). If symptoms continue despite adequate treatment or if psychological overlay is suspected, psychiatric evaluation may be indicated. Medical causes of dysmenorrhea, however, should be eliminated first.

1553

difficulties for 3 to 5% of women of reproductive age. The diagnosis of both PMS and PMDD is best established by requiring affected individuals to keep prospective daily records of symptoms during a 2- to 3-month period. Less than 50% of women complaining of PMS are found to have the syndrome when such records are examined. Most women seek help for PMS in their 30s after 10 or more years of symptoms. Many report that their symptoms began at menarche; approximately half state that symptoms followed childbirth. Severity and duration of symptoms are often reported to increase after each successive pregnancy and to become more severe with advancing age. Women with severe long-standing PMS almost always describe associated psychological reactions, including social difficulties, such as marital discord, difficulty relating to their children, difficulty maintaining friendships, and withdrawal from social activities.

TREATMENT  General Measures

The cause of PMS is unknown, and patients should be informed that no one therapy has been effective in all women. Women with mild premenstrual symptoms often benefit from simple changes in lifestyle, including daily mild aerobic exercise; reduction in intake of caffeine-containing beverages, salt, and refined sugar, particularly in the luteal phase; stress reduction; and adequate rest.

,



Medical Therapy

Women with more severe PMS may benefit from symptomatic treatment.6 Continuous oral contraceptives have inconsistent but generally positive therapeutic benefit,7 although intermittent drospirenone/ethinyl estradiol dosed on a 21-7 schedule has been found to have comparable efficacy to the continuous combined oral contraceptive. A6  Bromocriptine (generally 2.5 mg twice a day) or danazol (100 to 400 mg/day in two divided doses) may be given continuously for relief of mastalgia (breast pain), although this use is not listed in the manufacturer’s directive or approved by the U.S. Food and Drug Administration (FDA) and efficacy has not been documented by rigorous randomized trials. Prostaglandin synthetase inhibitors may help reduce dysmenorrhea and may alleviate headaches. Mild sedatives and tranquilizers may help reduce insomnia and anxiety. Low doses of fluoxetine (10 to 20 mg) and other selective serotonin reuptake inhibitors, administered either daily or for the last 2 weeks of each menstrual cycle, are highly effective in reducing the emotional symptoms associated with PMS. A7  Mild diuretics (especially spironolactone at doses up to 100 mg each morning) may benefit cyclic edema. Natural progesterone, given in the form of vaginal suppositories, has been used, but results of double-blind placebo-controlled trials show no efficacy.8 Likewise, the value of large quantities of multiple vitamins or of oil of evening primrose, containing the essential fatty acid γ-linolenic acid, a precursor of prostaglandins, is unsubstantiated.

Premenstrual Syndrome  

DEFINITION

Premenstrual syndrome (PMS), also known as premenstrual tension, is a complex of physical and emotional symptoms that occur repetitively in a cyclic fashion before menstruation and that diminish or disappear with menstruation.5  

DIAGNOSIS

The cyclic symptoms typically are sufficiently severe to interfere with some aspects of life. More than 150 different symptoms are now thought to vary with the menstrual cycle (Table 223-1). Estimates of the prevalence of PMS range from 25 to 100%. The Diagnostic and Statistical Manual of Mental Disorders classifies severe PMS as premenstrual dysphoric disorder (PMDD). For most women, PMS is merely annoying; severe PMS (or PMDD) causes serious

Surgical Therapy

Because PMS requires the occurrence of cyclic ovulation, oophorectomy is occasionally considered for patients with particularly intractable symptoms. However, oophorectomy may create new problems related to estrogen deficiency for women with PMS treated in this permanent fashion. Several trials employing a GnRH agonist together with exogenous steroids have been described as reducing PMS. Whether such therapy can be used in the long term remains to be determined.

TABLE 223-1 COMMON SYMPTOMS OF CYCLIC PREMENSTRUAL SYNDROME SOMATIC SYMPTOMS



Abdominal bloating

Constipation or diarrhea

Acne

Headache

Alcohol intolerance

Peripheral edema

Breast engorgement and tenderness

Weight gain

Clumsiness EMOTIONAL AND MENTAL SYMPTOMS Anxiety

Insomnia

Change in libido

Irritability

Depression

Lethargy

Fatigue

Mood swings

Food cravings (especially salt and sugar)

Panic attacks

Hostility

Paranoia

Inability to concentrate

Violence toward self and others

Increased appetite

Withdrawal from others

Abnormal Uterine Bleeding  

DIAGNOSIS

Differential Diagnosis

Because there is considerable confusion about terminology of abnormal uterine bleeding, it is important to determine just what is being included in any term used other than abnormal uterine bleeding.9 Postmenarchal bleeding in adolescents secondary to immaturity of the hypothalamic-pituitary-ovarian axis (resulting in anovulation) accounts for about 20% of cases, and perimenopausal bleeding consequent to incipient ovarian failure constitutes more than half. The causes of abnormal uterine bleeding in the reproductive years include complications from the use of oral contraceptives; complications of pregnancy, especially threatened, incomplete, or missed miscarriages and ectopic pregnancy; coagulation disorders, most commonly idiopathic thrombocytopenic purpura and von Willebrand disease; and pelvic disease, such as intrauterine polyps, leiomyomas, and tumors of the vagina and cervix. Clear cell adenocarcinoma of the vagina or cervix may occur in women exposed to diethylstilbestrol during fetal life. Affected women may also have congenital abnormalities of the upper vagina, cervix, and uterus. Women with a history

1554

CHAPTER 223  Reproductive Endocrinology and Infertility  

of diethylstilbestrol exposure should be reassured that the incidence of malignant change is extremely low. Trauma (coital or otherwise), foreign bodies, systemic illnesses including various endocrinopathies (such as diabetes mellitus, hypothyroidism and hyperthyroidism, Cushing syndrome, and Addison disease), leukemia, and renal disease may also be associated with abnormal bleeding as the presenting manifestation. Abnormal uterine bleeding with no demonstrable organic genital or extragenital cause (75% of cases) is most frequently associated with anovulation and is appropriately termed anovulatory (sometimes termed dysfunctional) bleeding. Most anovulatory bleeding is due to either estrogen withdrawal or estrogen breakthrough bleeding. In anovulatory women, estrogen stimulates the endometrium unopposed by progesterone. The endometrium proliferates, becomes thicker, and may shed irregularly. Anovulatory bleeding tends to occur at less frequent intervals, and organic lesions tend to cause bleeding more frequently than cyclic menses.

Clinical Evaluation

Endometrial ablation by any of several methods is being used increasingly to treat persistent bleeding. However, ablation is not 100% effective, and medical management remains the first line of therapy for most women. Hysterectomy may be an appropriate choice for a small number of women.



Amenorrhea  



All cases of abnormal bleeding should be evaluated, beginning with a thorough history emphasizing the amount and duration of blood loss. Prospective charting of the days on which bleeding occurs may be required to evaluate the bleeding pattern. Complications of pregnancy or a bleeding diathesis must always be ruled out. The findings on physical examination (including the Papanicolaou smear) are normal in anovulatory bleeding except for signs of anemia in the more severe cases. Laboratory tests should include a complete blood count, platelet count, coagulation studies, including von Willebrand disease screening, thyroid function tests, and fasting blood glucose concentration. Anovulatory bleeding must be a diagnosis of exclusion, with management depending on the age of the patient and the extent of the bleeding. A sample of the endometrium should be obtained by biopsy or by dilation and curettage from all women older than 35 years and from those at increased risk for endometrial carcinoma because of prolonged anovulatory bleeding.

TREATMENT  Even profuse bleeding in hemodynamically stable anovulatory women can almost always be successfully treated by the administration of one combination oral contraceptive pill every 6 hours for 5 to 7 days, although this use is not listed in the manufacturer’s directive or approved by the FDA. Bleeding should cease within 24 hours, but patients should be warned to expect heavy bleeding 2 to 4 days after therapy is stopped. If anemia is profound, blood transfusion may be necessary. If the bleeding continues despite therapy, curettage can be carried out. Recurrence can be prevented by giving the patient combination oral contraceptive agents cyclically if pregnancy is not desired. If pregnancy is desired, ovulation can be induced. Acute episodes of anovulatory bleeding can also be treated with conjugated estrogens administered intravenously (25 mg every 4 hours for up to three doses) until bleeding ceases, although this use is not listed in the manufacturer’s directive or approved by the FDA. Progestin therapy (medroxyprogesterone acetate, 5 to 10 mg orally for 10 days) should be started simultaneously. Withdrawal bleeding occurs after cessation of therapy, and the patient can then be treated with oral contraceptive agents for at least three cycles. For individuals with anovulatory bleeding without an episode of profuse bleeding, treatment with cyclic oral contraceptive agents or progestin can be provided unless pregnancy is desired, in which case ovulation must be induced.

DEFINITION

Amenorrhea is the absence of menstruation for 3 months or more in women with past menses (secondary amenorrhea) or the absence of menarche by the age of 15 years regardless of the absence or presence of secondary sex characteristics (primary amenorrhea).

PATHOBIOLOGY

If an intact genital outflow tract exists and there is no primary disease of the uterus, amenorrhea is a sign of failure of the hypothalamic-pituitary-ovarian axis to produce cyclically the hormones necessary for menses. Amenorrhea is physiologic in the prepubertal girl, during pregnancy and early in lactation, and after menopause. At any other time, it is pathologic and demands evaluation. Use of the term postpill amenorrhea to refer to failure to resume menses within 3 months of discontinuation of oral contraceptives is inappropriate. Women so affected should be evaluated in the same manner as any woman with amenorrhea. Similarly, individuals with menses occurring at infrequent intervals of more than 40 days or having fewer than nine menses per year, termed oligomenorrhea, should be evaluated identically to women with amenorrhea.  

DIAGNOSIS

Clinical Evaluation

In patients with amenorrhea, even subtle hormonal abnormalities may lead to signs and symptoms. Breast development indicates exposure to estrogens, and the presence of pubic and axillary hair indicates androgenic stimulation. Patients should be questioned especially closely for evidence of psychological disturbances, dietary and exercise habits,10 lifestyle, environmental stresses,11 family history of genetic anomalies, abnormal growth and development, and signs of hyperandrogenism, including hirsutism, temporal balding, deepening of the voice, increased muscle mass, clitorimegaly, and increased libido, and signs of defeminization, including decreasing breast size and vaginal atrophy. Any history of galactorrhea should be determined. A history of symptoms related to thyroid and adrenal dysfunction should also be sought (Chapters 213, 214, and 215). The physical examination should focus on the evaluation of body dimensions and habitus, extent and distribution of body hair, breast development and secretions, and genitalia. In normal adult women, the arm span is similar to the height; in hypogonadal women, the span is generally more than 5 cm greater than the height. The distribution and quantity of body hair should be considered in view of the family history. The extent of any hirsutism should be recorded, preferably by photographs. Other signs of virilization should be sought carefully. Breast development should be graded according to the method of Tanner (Table 223-2).12 Breast secretion should be sought by applying pressure to the breasts while the patient is seated. Any secretion should be examined microscopically for the presence of perfectly round fat globules of

TABLE 223-2 CRITERIA FOR DISTINGUISHING TANNER STAGES 1–5 DURING PUBERTAL MATURATION TANNER STAGE

BREAST

PUBIC HAIR

1 (Prepubertal)

No palpable glandular tissue or pigmentation of areola; elevation of areola only

No pubic hair; short, fine vellus hair only

2

Glandular tissue palpable with elevation of breast and areola together as a small mound; areolar diameter increased

Sparse, long, pigmented terminal hair chiefly along the labia majora

3

Further enlargement without separation of breast and areola; although more darkly pigmented, areola still pale and immature; nipple generally at or above midplane of breast tissue when individual is seated upright

Dark, coarse, curly hair, extending sparsely over mons

4

Secondary mound of areola and papilla above breast

Adult-type hair, abundant but limited to mons and labia

5 (Adult)

Recession of areola to contour of breast; development of the Montgomery glands and ducts on areola; further pigmentation of areola; nipple generally below midplane of breast tissue when individual is seated upright; maturation independent of breast size

Adult-type hair in quantity and distribution; spread to inner aspects of the thighs in most racial groups

Data from Ross GT. Disorders of the ovary and female reproductive tract. In: Wilson JD, Foster DW, eds. Textbook of Endocrinology, 7th ed. Philadelphia: WB Saunders; 1985:206; Speroff L, Glass RH, Kase N. Clinical Gynecologic Endocrinology and Infertility, 3rd ed. Baltimore: Williams & Wilkins; 1983:377; and Kustin J, Rebar RW. Menstrual disorders in the adolescent age group. Primary Care. 1987;14:139-166.

CHAPTER 223  Reproductive Endocrinology and Infertility  

TABLE 223-3 CAUSES OF AMENORRHEA ANATOMIC CAUSES Pregnancy Various disorders of sexual differentiation Distal genital tract obstruction (müllerian agenesis or dysgenesis) Gonadal dysgenesis* Ambiguity of external genitalia (male and female pseudohermaphroditism) Intrauterine adhesions (Asherman syndrome) Gestational trophoblastic disease CHRONIC ANOVULATION Due to CNS-hypothalamic-pituitary dysfunction With inappropriate steroid feedback (e.g., polycystic ovary syndrome) Due to thyroid or adrenal disorders OVARIAN “FAILURE” Menopause Genetic abnormalities Physical and environmental causes (e.g., chemotherapeutic agents, irradiation) Autoimmune disorders Idiopathic *Gonadal dysgenesis may be viewed as both a disorder of sexual differentiation and a form of gonadal “failure.” CNS = central nervous system.

varying size, which indicate galactorrhea. Finally, the female genitalia should be examined carefully because they are such sensitive indicators of the hormonal milieu. The Tanner stage of pubic hair development should be noted (see Table 223-2). Because the sensitivity of the genitalia to androgens decreases onward from early in fetal development, the extent of any virilization is important. Fusion of the labia and enlargement of the clitoris with or without formation of a penile urethra are observed in women exposed to androgens during the first 3 months of fetal development (Chapter 220). Significant clitorimegaly in the absence of other signs of sexual ambiguity and in the presence of other signs of virilization requires marked androgenic stimulation and strongly implicates an androgen-secreting neoplasm. The development of the labia minora in postpubertal women indicates the influence of estrogens. Overt anomalies of the distal genital tract and any evidence of obstruction to the escape of menstrual blood should be sought. Under the influence of estrogen, the vaginal mucosa changes during sexual maturation from a tissue with a shiny, bright red appearance with sparse, thin secretions to a dull, gray-pink rugated surface with copious, thick secretions. The history and physical examination quickly differentiate among several causes of amenorrhea (Table 223-3). The various disorders of sexual differentiation and other anatomic causes are often apparent on inspection. Distal genital tract obstruction should be identified at the time of pelvic examination even if the specific abnormality is not obvious. The physical stigmata of Turner syndrome, discussed subsequently, generally make the diagnosis simple. Any sexual ambiguity indicates the need for chromosomal analysis and the measurement of 17α-hydroxyprogesterone to rule out congenital adrenal hyperplasia. Pregnancy and gestational trophoblastic disease may be diagnosed by measurement of human chorionic gonadotropin (hCG). The possibility of intrauterine synechiae or adhesions (Asherman syndrome) must be considered in individuals in whom amenorrhea develops after curettage or endometritis. Tuberculous endometritis, especially in younger women, may also lead to this disorder. Without hormonal measurements, it may be impossible to distinguish between individuals with chronic anovulation, in whom hypothalamic-pituitaryovarian function is insufficiently coordinated to produce cyclic ovulation, and those with ovarian failure. However, it is generally possible to form a clinical impression about the cause of the amenorrhea. It can be noted whether the patient has absence of, incomplete, or complete development of secondary sex characteristics. The presence of excess body hair or galactorrhea may provide clinical evidence of the pathogenesis of the amenorrhea. Signs and symptoms of adrenal or thyroid dysfunction may be important as well. Administration of a progestin (typically medroxyprogesterone acetate, 5 to 10 mg given orally for 5 to 10 days, or progesterone in oil, 100 mg given intramuscularly) has been advocated to assess the level of endogenous estrogen. This test is of limited value, however, because almost half the young women with premature ovarian failure experience withdrawal bleeding in response to progestin.

1555

To ascertain whether the outflow tract is intact, an orally active estrogen, such as 2.5 mg of conjugated estrogen daily for 21 days, with 5 to 10 mg of oral medroxyprogesterone acetate for the last 5 to 10 days, is administered. Withdrawal bleeding should occur if the endometrium is normal. Still, hysterosalpingography and hysteroscopy may be required for the diagnosis of Asherman syndrome because some patients with a normal endometrium may not have a withdrawal bleed due to obstruction of the cervical os by scar tissue.

Laboratory Findings

Basal levels of FSH, prolactin, and thyroid-stimulating hormone (TSH) should be measured in all amenorrheic and oligomenorrheic women to confirm the clinical impression (Fig. 223-3). Increased TSH levels with or without increased levels of prolactin imply primary hypothyroidism, and further evaluation for this disorder is indicated (Chapter 213). Although hypothyroidism commonly results in anovulation, amenorrhea occurs in only some hypothyroid women. Menorrhagia and oligomenorrhea may occur as well. The sensitive immunoassays for TSH permit identification of women with hyperthyroidism as well because TSH levels are suppressed in those individuals. If the prolactin concentration is minimally increased and the TSH level is normal, measurement of the prolactin concentration should be repeated before more extensive evaluation is undertaken because prolactin levels are increased by nonspecific stressful stimuli, sleep, and food ingestion. Prolactin levels may be elevated in as many as one third of women with amenorrhea. Increased FSH levels (generally above 30 mIU/mL) imply ovarian failure and require further evaluation. Incipient ovarian failure should be considered in any woman with basal FSH levels of 15 mIU/mL or higher other than during the midcycle LH surge. Many clinicians believe that chromosomal evaluation is indicated in all individuals with elevated FSH levels before age 40 years, and it is certainly indicated if hypergonadotropic amenorrhea begins before age 30 years. If FSH levels are low or normal, the measurement of total testosterone levels may be helpful whether or not there is any evidence of hirsutism or virilization. Hyperandrogenic women need not be hirsute because some have relative insensitivity of the hair follicles to androgens. Mildly increased levels of testosterone (and perhaps dehydroepiandrosterone sulfate as well) suggest polycystic ovary syndrome (PCOS). However, total circulating androgen levels need not be elevated because of the alterations in metabolic clearance rate and sex hormone–binding globulin that are present in PCOS. Consequently, some clinicians prefer to measure circulating free testosterone levels. Circulating levels of LH and FSH may aid in differentiation of PCOS from hypothalamic-pituitary dysfunction. LH levels are often elevated in PCOS so that the ratio of LH to FSH is increased; however, LH levels may be identical to those observed in normal women in the follicular phase. In contrast, levels of LH and FSH are normal or slightly reduced in hypothalamic-pituitary dysfunction. There is some overlap between women with “polycystic ovarian-like” disorders and those with hypothalamic-pituitary dysfunction. Radiographic assessment of the sella turcica is indicated in all amenorrheic women in whom both LH and FSH levels are consistently low (both below 10 mIU/mL) to exclude a pituitary or parapituitary neoplasm (Chapter 211). Other pituitary functions should be evaluated in any individual with significantly impaired LH and FSH secretion. Both total testosterone and dehydroepiandrosterone sulfate levels should be measured in hirsute or virilized women. Testosterone levels greater than 200 ng/dL should lead to investigation for an androgen-producing neoplasm, most likely of ovarian origin. Dehydroepiandrosterone sulfate levels greater than 7.0 µg/mL should lead to evaluation for an adrenal neoplasm, and levels between 5.0 and 7.0 µg/mL should lead to evaluation for adult-onset congenital adrenal hyperplasia (Chapter 220).

Hypergonadotropic Amenorrhea (Presumptive Ovarian Failure, Primary Hypogonadism, Primary Ovarian Insufficiency)  



DIAGNOSIS

Differential Diagnosis

Gonadal failure may begin at any time during embryonic or postnatal development and may result from many causes. Normally, the ovaries fail at menopause, when virtually no functioning follicles remain. However, premature loss of oocytes before the age of 40 years may occur and lead to premature ovarian failure. Circulating gonadotropin levels increase whenever

1556

CHAPTER 223  Reproductive Endocrinology and Infertility  

Measure FSH, PRL, TSH

TSH

Primary hypothyroidism

PRL Normal TSH

FSH

Evaluate for hyperprolactinemia

Evaluate for gonadal “failure”

Normal TSH, PRL Normal or low FSH

Hirsutism present

No hirsutism or virilization

Measure T, DS

Measure T

T > 200 ng/dL or DS > 7.0 µg/mL

Normal or mildly T DS 5.0-7.0 µg/mL

Evaluate for neoplasm

Evaluate for “adult-onset” CAH

Normal or mildly T and/or DS (normal or LH)

Mildly T (normal or LH)

PCOlike

Normal T (normal or LH)

HCA

Evaluate for “adult-onset” CAH FIGURE 223-3.  Biochemical evaluation of amenorrhea. This schema must be considered an adjunct to the clinical evaluation of the patient. See text for details. CAH = congenital adrenal hyperplasia; DS = dehydroepiandrosterone sulfate; FSH = follicle-stimulating hormone; HCA = hypothalamic chronic anovulation; LH = luteinizing hormone; PCO-like = polycystic ovarian syndrome–like; PRL = prolactin; T = testosterone; TSH = thyroid-stimulating hormone.

ovarian failure occurs because of decreased negative estrogen feedback to the hypothalamic-pituitary unit.  

PATHOBIOLOGY

There are several causes for premature ovarian failure, including genetic causes (a growing list that includes karyotypic abnormalities, single gene mutations, and complex multifactorial polygenic inheritance), physical and environmental causes, and autoimmune disturbances. In addition, there may be families in which menopause begins earlier than the expected age without any further pathologic cause.

Genetic Abnormalities

Several pathologic conditions with dysgenetic gonads involve elevated gonadotropin levels and amenorrhea as well as abnormalities of the X chromosome. The term gonadal dysgenesis refers to individuals with undifferentiated streak gonads without any association with either extragonadal stigmata or sex chromosome aberrations. Because individuals with gonadal dysgenesis have the normal complement of oocytes at 20 weeks of fetal age but virtually none by birth, this disorder is a form of premature ovarian failure.

Turner Syndrome

Turner syndrome (also see Chapter 220) describes patients with streak gonads composed of fibrous stroma and four cardinal features: a female phenotype; sexual infantilism; short stature; and several physical abnormalities, sometimes including webbed neck, low-set ears, multiple pigmented nevi, double eyelashes, micrognathia, epicanthal folds, shieldlike chest with microthelia, short fourth metacarpals, increased carrying angle of the arms, and certain renal and cardiovascular defects (most commonly coarctation of the aorta and aortic stenosis). The diagnosis can sometimes be made at birth because of unexplained lymphedema of the hands and feet. The syndrome is associated with an abnormality of sex chromosome number, morphology, or both. Most commonly, the second sex chromosome is absent (45,X). Turner syndrome is the single most common chromosome disorder in humans, but more than 95% of such fetuses are aborted, and the incidence in newborns is approximately 1 in 3000 to 5000. Chromosome breakage and mosaicism occur as well. In mosaic individuals with a normal 46,XX cell line, sufficient follicles may persist postnatally

to initiate pubertal changes and to cause ovulation so that pregnancy is possible. Deletions of the X-chromosome-linked SHOX gene explain many of the dysmorphic skeletal features that are present, including the short stature. It is believed that the number of phenotypic findings may be related to the percentage of cells that are 45,X. There also may be an effect of imprinting with the variation in phenotype partly explained by the parental origin of the one remaining X chromosome.

Pure Gonadal Dysgenesis

Pure gonadal dysgenesis is the term given to phenotypically female individuals with streak gonads who are of normal stature and have none of the physical stigmata associated with Turner syndrome. Such individuals have either a 46,XX or 46,XY karyotype. The 46,XX defect may be inherited as an autosomal recessive, with 10% having associated nerve deafness. The 46,XY defect may be inherited as an X-linked recessive, with clitorimegaly occurring in 10 to 15% and gonadal tumors developing in 25% if the gonads are not removed.

Mutations in the X Chromosome Associated with Premature Ovarian Failure

Several regions of the X chromosome are now recognized to contain mutations in genes that may result in premature ovarian failure. Of particular note is the fragile X mental retardation (FMR1) gene. More than 5% of women with 46,XX spontaneous premature ovarian failure have mutations of the FMR1 gene. This risk is increased if there is a family history of premature ovarian failure. A family history of fragile X syndrome, unexplained mental retardation, dementia, developmental delay of a child, or tremor-ataxia syndrome is reason for genetic counseling. Mutations in the FMR1 gene are known to be associated with a neurodegenerative disorder. Women with mutations in the FMR1 gene are at risk of having a child with mental retardation, should they be one of the 6 to 8% of women with premature ovarian failure who conceive spontaneously. For FMR1, a CGG repeat sequence occurs, with up to 60 repeats being normal. Expansion to more than 200 repeats leads to the fragile X syndrome, with the high level of repeats causing hypermethylation of the gene promoter and silencing of the gene. Female carriers of the permutation have an unstable intermediate number of repeats (i.e., 60 to 199) and the predisposition for premature ovarian failure.

CHAPTER 223  Reproductive Endocrinology and Infertility  

Trisomy X

Trisomy X (46,XXX karyotype) is also associated with premature menopause, although many such individuals have normal reproductive lives. Premature menopause can also occur in mosaic individuals with cell lines with excess X chromosomes. When gonadal abnormalities occur in women with excess X chromosomes, they seem to occur after ovarian differentiation so that some ovarian function is possible. Only later in life do such women develop secondary amenorrhea and premature ovarian failure.

Known Genetic Alterations of Specific Genes

In girls with the rare syndrome of 17α-hydroxylase deficiency involving p450c17 who survive until the expected age of puberty, sexual infantilism and primary amenorrhea occur together with elevated levels of gonadotropins (also see Chapter 220). Defects in the 20,22-lyase (p450scc) or aromatase (p450arom) enzymes may also lead to ovarian failure. Women with galactosemia also experience ovarian failure early in life, even when a galactose-restricted diet is introduced early in infancy. Mutations of several autosomal genes result in premature ovarian failure. Included in this growing list are mutations involving FSHR (the FSH receptor gene), FOXL2 (a forkhead transcription factor associated with the blepharophimosis-ptosis-epicanthus inversus syndrome), INHA (the inhibin-α gene), E1F2B (a family of genes associated with central nervous system leukodystrophy and ovarian failure), PMM2 (the gene for phosphomannomutase), GALT (the gene for galactose-1-phosphate uridyltransferase), and AIRE (leading to the autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy syndrome). Myotonic dystrophy (Chapter 393) is caused by an autosomal triple repeat mutation, like the fragile X syndrome, that is similarly associated with premature loss of germ cells from the ovary. The list of mutations associated with early ovarian failure continues to increase as the function of more genes is determined.

1557

whom the diagnosis of premature ovarian failure is made. Ovarian biopsy is not indicated to document the existence of follicles because only a small portion of each ovary can be sampled and because pregnancies have resulted in patients who had biopsy samples devoid of follicles. Estrogen replacement is warranted to prevent the accelerated bone loss known to occur in affected women (Chapter 230). The estrogen should be given sequentially with a progestin to prevent endometrial hyperplasia. Young women with ovarian failure may require twice as much estrogen as postmenopausal women for relief of signs and symptoms of hypoestrogenism. Inexplicably, women with premature ovarian failure may conceive while taking exogenous estrogen, even in the form of oral contraceptive agents, at the same rate as those not taking estrogen, so barrier contraception should be discussed if pregnancy is not desired. Women with hypergonadotropic amenorrhea are rarely able to become pregnant. It is not clear why pregnancy may rarely occur in such women, but the pregnancy and delivery rate is 6 to 8%. Infertility treatment of young women with hypergonadotropic amenorrhea involves hormone replacement to mimic the normal menstrual cycle and embryo transfer by use of donor oocytes. Whether women with gonadal dysgenesis should be offered pregnancy by use of donor oocytes is now the subject of debate because a markedly increased incidence of aortic rupture during pregnancy secondary to medial necrosis has been documented. Women with Turner syndrome contemplating pregnancy should be counseled regarding the risks.

  CHRONIC ANOVULATION

Chronic anovulation, the most frequent form of amenorrhea encountered in women of reproductive age, implies that functional ovarian follicles remain and that cyclic ovulation can be induced with appropriate therapy (Table 223-4). The cause of the anovulation should be determined. The pathophysiologic

Mutations Involving Reproductive Hormones, Their Receptors, and Action

The resistant ovary (Savage) syndrome occurs in young amenorrheic women who have elevated peripheral gonadotropin concentrations, normal (although immature) follicles present on ovarian biopsy, 46,XX karyotype with no evidence of mosaicism, fully developed secondary sex characteristics, and ovarian resistance to stimulation with human menopausal or pituitary gonadotropins. At least some of these women have mutations in the FSH receptor. It is probably inappropriate to use the term “resistant ovary syndrome” because it is likely that this is a heterogeneous disorder due to various genetic mutations.

Other Causes Physical and Environmental

Irradiation and chemotherapeutic agents used to treat various malignant diseases may also cause premature ovarian failure. Ovulation and cyclic menses return in some of these patients even after prolonged intervals of hypergonadotropic amenorrhea associated with signs and symptoms of profound hypoestrogenism. In general, the younger the individual at the time of treatment, the less likely is she to have permanent ovarian failure after the completion of therapy. Rarely, mumps affects the ovaries and causes ovarian failure.

Autoimmune Disorders

Premature ovarian failure may occur in conjunction with a variety of autoimmune disorders. The most well known syndrome (autoimmune polyglandular syndrome type 1) involves hypoadrenalism, hypoparathyroidism, and mucocutaneous candidiasis together with ovarian failure (Chapter 218). Testing for adrenal antibodies by indirect immunofluorescence will identify the 4% of women with spontaneous premature ovarian failure who have steroidogenic cell autoimmunity and are at risk for adrenal insufficiency. Thyroiditis is the most commonly associated abnormality. Antibodies to the FSH receptor have been identified in a few cases. These associations make it mandatory to rule out other potentially life-threatening endocrinopathies in young women with hypergonadotropic amenorrhea.

TREATMENT  Women with hypergonadotropic amenorrhea and ovarian failure should be treated identically whether or not they have signs of hypoestrogenism or desire pregnancy. Counseling and psychological support are indicated in women in

TABLE 223-4 CAUSES OF CHRONIC ANOVULATION Chronic anovulation of hypothalamic-pituitary origin Hypothalamic chronic anovulation Psychogenic Exercise associated Associated with diet, weight loss, or malnutrition Anorexia nervosa and bulimia Pseudocyesis Forms of isolated (idiopathic) hypogonadotropic hypogonadism (including Kallmann syndrome) Due to hypothalamic-pituitary damage Pituitary and parapituitary tumors Empty sella syndrome Following surgery Following irradiation Following trauma Following infection Following infarction Idiopathic hypopituitarism Hypothalamic-pituitary dysfunction or failure with hyperprolactinemia (multiple causes) Due to systemic diseases Chronic anovulation due to inappropriate feedback (i.e., polycystic ovary syndrome) Excessive extraglandular estrogen production (i.e., obesity) Abnormal buffering involving sex hormone–binding globulin (including liver disease) Functional androgen excess (adrenal or ovarian) Neoplasms producing androgens or estrogens Neoplasms producing chorionic gonadotropin Chronic anovulation due to other endocrine and metabolic disorders Adrenal hyperfunction Cushing syndrome Congenital adrenal hyperplasia (female pseudohermaphroditism) Thyroid dysfunction Hyperthyroidism Hypothyroidism Prolactin or growth hormone excess Hypothalamic dysfunction Pituitary dysfunction (microadenomas and macroadenomas) Drug induced Malnutrition

1558

CHAPTER 223  Reproductive Endocrinology and Infertility  

bases for several forms of anovulation are unknown, but the anovulation can be interrupted transiently by nonspecific induction of ovulation in most affected women. Anovulation can result in either amenorrhea or irregular (generally less frequent) menses.  

TREATMENT  Psychological counseling or a change in lifestyle, especially for women engaged in strenuous exercise programs, may be effective in inducing cyclic ovulation and menses in women with functional HCA. Cognitive-behavioral therapy is effective in a proportion of women with functional HCA. For women desiring pregnancy, ovulation can also be induced with clomiphene citrate (50 to 100 mg/day for 5 days beginning on the third to fifth day of withdrawal bleeding). Treatment with exogenous gonadotropins to induce follicular maturation followed by hCG to induce follicular rupture may be effective in women who do not ovulate in response to clomiphene. Because women with HCA have low circulating levels of leptin, investigators have given recombinant leptin and documented that ovulation may resume in some affected women. Given the heterogeneous nature of the disorder, it is not surprising that exogenous leptin is not effective in all women. Most physicians advocate the use of exogenous gonadal steroids to prevent osteoporosis. A regimen can consist of daily oral conjugated or esterified estrogens (0.625 to 1.25 mg), ethinyl estradiol (20 µg), or micronized estradiol-17β (1 to 2 mg) or transdermal estradiol-17β (0.05 to 0.10 mg) daily, with oral medroxyprogesterone acetate (5 to 10 mg) added for the first 12 to 14 days of each month. Sexually active women can be given oral contraceptive agents as an alternative. If steroid therapy is administered, patients must be informed that the amenorrhea will probably be present when therapy is discontinued. Other physicians believe that only periodic observation is indicated, with barrier methods of contraception recommended for fertility control. Adequate ingestion of calcium should be ensured regardless of therapy. Contraception is needed for sexually active women with HCA because the functional defect is mild in these disorders and may resolve spontaneously at any time, with ovulation occurring before any episode of menstruation.

Hypothalamic Chronic Anovulation  

DEFINITION

Hypothalamic chronic anovulation (HCA) represents a heterogeneous group of disorders with similar manifestations. Emotional and physical stress, excessive exercise, nutritional deficiencies, weight loss, reduced body fat, and other unrecognized factors may contribute in varying proportions to the anovulation. Women with HCA have normal neuroanatomic findings.  

ANOREXIA NERVOSA

Individuals with amenorrhea and significant weight loss should be examined for the possibility of anorexia nervosa (Chapter 206).  

ISOLATED HYPOGONADOTROPIC HYPOGONADISM

Affected individuals have absence of spontaneous pubertal development. Most have functional GnRH deficiency, but some have abnormalities of gonadotropin deficiency localized to the pituitary gland. Kallmann syndrome is a familial disorder consisting of gonadotropin deficiency, anosmia or hyposmia, and color blindness in men or, more rarely, in women (Chapter 210). Partial or complete agenesis of the olfactory bulb is present on autopsy, accounting for use of the term olfactogenital dysplasia. Isolated gonadotropin deficiency in the absence of anosmia occurs as well. Sexual infantilism with a eunuchoid habitus is the clinical hallmark of this disorder, but moderate breast development may occur. Circulating LH and FSH levels are low but almost always detectable. Ovulation induction requires use of exogenous gonadotropins and hCG or pulsatile GnRH. Estrogen replacement therapy is indicated in these women until pregnancy is desired. It may not be possible to distinguish between partial isolated gonadotropin deficiency and functional HCA in all cases.  



HYPOPITUITARISM

Hypopituitarism may be obvious on cursory inspection or sufficiently subtle to require endocrine testing (Chapter 211). The clinical presentation depends on the age at onset, the cause, and the nutritional status of the individual. Failure of development of secondary sex characteristics must always raise the question of hypopituitarism. Ovulation can be induced successfully with exogenous gonadotropins when pregnancy is desired and after the hypopituitarism is treated appropriately. Replacement therapy with estrogen is indicated.  

HYPERPROLACTINEMIA

Galactorrhea associated with hyperprolactinemia, whatever the cause, almost always occurs together with amenorrhea caused by hypothalamic-pituitary dysfunction or failure. Many conditions can cause excess prolactin secretion (Chapter 211). A prolactinoma must be excluded. Hirsutism may be observed occasionally in association with amenorrhea-galactorrhea and hyperprolactinemia. Elevated levels of the adrenal androgens dehydroepiandrosterone and dehydroepiandrosterone sulfate may be observed and may account for the polycystic-type ovaries present in some hyperprolactinemic women.  

FAILURE OF THE HYPOTHALAMIC-PITUITARY UNIT

The hypothalamic-pituitary unit may also fail to function normally in a number of stressful, debilitating, systemic illnesses that interfere with somatic growth and development. Chronic renal failure, liver disease, and diabetes mellitus are the most prominent examples.  

DIAGNOSIS

Abrupt cessation of menses in women younger than 30 years who have no anatomic abnormalities of the hypothalamic-pituitary-ovarian axis and no other endocrine disturbances suggests a diagnosis of HCA. Affected individuals tend to be bright, educated, and engaged in intellectual occupations and may well give a history of psychosexual problems and socioenvironmental trauma. HCA is characterized by low to normal levels of gonadotropins and relative hypoestrogenism. Rarely, however, do affected women present with signs and symptoms of estrogen deficiency. It is important to rule out a central lesion as the cause of the hypogonadotropic hypogonadism in women who appear to have HCA.



Chronic Anovulation Related to Inappropriate Feedback

POLYCYSTIC OVARY SYNDROME DEFINITION  

Polycystic ovary syndrome (PCOS) is a heterogeneous disorder in which there is considerable clinical and biochemical variability among affected individuals.13 PCOS is currently considered to exist in women with any two of the following: (1) oligo-ovulation or anovulation, (2) hyperandrogenism, or (3) polycystic ovaries on ultrasound, and in whom other etiologies have been eliminated. PCOS is the classic disorder in which the amenorrhea or oligomenorrhea results from inappropriate feedback of gonadal steroids from the ovaries.14  

PATHOBIOLOGY

Current evidence suggests that the hypothalamic-pituitary unit is intact and that a functional derangement, perhaps involving insulin-like growth factors such as IGF-I within the ovary, results in abnormal gonadotropin secretion. PCOS is characterized by insulin resistance and compensatory hyperinsulinemia. The insulin resistance has been found in affected women of many racial and ethnic groups, implying that it is a universal characteristic and that a common defect may be present. There is increasing evidence of specific genetic abnormalities in some women with PCOS.  

CLINICAL MANIFESTATIONS

Although patients usually present with amenorrhea, hirsutism, and obesity, affected women may instead complain of irregular and profuse uterine bleeding, may not have hirsutism, and may be of normal weight. Excess androgen from any source or increased extraglandular conversion of androgens to estrogens can lead to the typical findings of PCOS. Included are such diverse disorders as Cushing syndrome, mild congenital adrenal hyperplasia, virilizing tumors of adrenal or ovarian origin, hyperthyroidism and hypothyroidism, obesity, and primary PCOS with no other recognizable cause. In the primary syndrome, the irregular menses, mild obesity, and hirsutism begin during puberty and typically become more severe with time, although there is increasing evidence of improvement in the years just before menopause. Obesity alone can lead to a polycystic ovarian-like syndrome, with the degree of obesity required to cause anovulation varying widely. The increase in the prevalence of obesity is leading to an increased prevalence of PCOS. All such patients are well estrogenized regardless of whether they present with primary or secondary amenorrhea or dysfunctional bleeding. LH concentrations tend

CHAPTER 223  Reproductive Endocrinology and Infertility  

to be elevated, with relatively low and constant FSH levels, but both may be in the normal range for the follicular phase of the menstrual cycle. Levels of most circulating androgens, especially testosterone, tend to be mildly elevated.  

pregnancy; there was no further improvement when the two agents were used concurrently.16

Surgical Treatment

Laparoscopic ovarian surgery can achieve unifollicular ovulation or make it easier for medical ovulation induction, but increases the risk for development of ovarian adhesions (themselves leading to infertility). It may be successful in a small subset of women with PCOS who are geographically removed from good medical care.

DIAGNOSIS

A consensus conference in Rotterdam in 2003 concluded that after exclusion of other etiologies, two of the following three are required for diagnosis of PCOS: (1) hyperandrogenism (clinical or biochemical), (2) oligo-ovulation or anovulation, (3) polycystic ovaries on ultrasound examination or at surgery. This definition is confusing to clinicians because it implies that hirsute women with polycystic ovaries on ultrasound examination who ovulate regularly should be considered to have PCOS. Moreover, it is clear that polycystic ovaries may be identified on ultrasound examination in normal women. In any case, the aim of the diagnostic evaluation is to rule out any causes (such as neoplasms) that require definitive therapy. Hirsutism should be evaluated as detailed in Chapter 413. A particularly severely affected subset of women present with marked obesity, anovulation, mild glucose intolerance with high levels of circulating insulin, acanthosis nigricans, hyperuricemia, severe hirsutism, and elevated circulating androgen levels. These women have hyperthecosis of the ovaries, in which the androgen-producing cells in the stromal, hilar, and thecal regions are increased greatly in number. Hyperthecosis should probably be viewed as a part of the spectrum of disorders constituting PCOS.

Chronic Anovulation Related to Other Endocrine and Metabolic Disorders  

Adrenal hyperfunction appears to cause chronic anovulation by inducing a polycystic ovarian-like syndrome secondary to increased adrenal androgen secretion. Both hyperthyroidism and hypothyroidism are associated with a variety of menstrual disturbances, including dysfunctional uterine bleeding and amenorrhea as a result of alterations in the metabolism of androgens and estrogens. These metabolic changes in turn result in inappropriate steroid feedback and chronic anovulation.

  INFERTILITY  

TREATMENT  Patients generally require therapy for hirsutism, for induction of ovulation if pregnancy is desired, and for prevention of estrogen-induced endometrial hyperplasia and cancer. No ideal therapy exists; the therapeutic approach must be individualized. The risks for metabolic syndrome, cardiovascular disease, and diabetes mellitus are increased in women with PCOS, at least in part because of the increased androgens and insulin resistance. Moreover, many women have elevated cholesterol levels.

Medical Therapy

In the anovulatory woman not desiring pregnancy who is not hirsute, therapy with intermittent progestin administration (such as medroxyprogesterone acetate, 5 to 10 mg orally for 10 to 14 days each month) or oral contraceptives can be provided to reduce the increased risk for endometrial carcinoma that is present in such a woman with unopposed estrogen. All women using intermittent progestin administration should be cautioned about the need for effective contraception if they are sexually active because these agents do not inhibit ovulation when they are administered intermittently. Improvements in insulin sensitivity in women with polycystic ovaries, either through lifestyle changes (e.g., exercise and diet) or through pharmacologic intervention, consistently result in improvements in the reproductive and metabolic abnormalities. Resumption of ovulation may occur in up to 60 to 70% of affected women.15 The longest and largest published experiences with any agent that improves insulin sensitivity in PCOS is with metformin, a biguanide that functions primarily by suppressing hepatic gluconeogenesis and also improves insulin sensitivity. Its use in PCOS leads to reductions in insulin and androgen levels and resumption of menses in some women. Divided doses of 1500 to 2000 mg/day have proved effective. Some clinicians advocate giving metformin to all women with polycystic ovaries, whereas others would administer such an agent only to those with documented insulin resistance. Some clinicians also advocate giving metformin first to women who desire pregnancy and then adding an agent to induce ovulation if the metformin proves ineffective. These agents are not approved for use in pregnant women or for the induction of ovulation.

Treatment Considering Pregnancy

Oral contraceptive agents are the first line of therapy for hirsute anovulatory woman not desiring pregnancy and offer protection from endometrial hyperplasia. In women with PCOS desiring pregnancy, clomiphene citrate or letrozole can be used to induce ovulation. A8  A9  Letrozole is not approved for this use by the FDA, but a large multicenter randomized trial has demonstrated its superiority to clomiphene in obese women with PCOS. About 75 to 80% conceive with such therapy. In addition to insulin-sensitizing agents, other possible methods of inducing ovulation include use of exogenous gonadotropins and hCG, and laparoscopic ovarian surgery with multiple punctures of the ovary by diathermy or laser. A large clinical trial documented that clomiphene citrate is more effective than metformin in inducing ovulation and resulting in ,

1559

DEFINITION

The World Health Organization (WHO) has defined infertility as “a disease of the reproductive system defined by the failure to achieve a clinical pregnancy after 12 months or more of regular unprotected sexual intercourse.” Sterility is total inability to reproduce. More than 10% of couples in the United States seek medical assistance for infertility. The requirements for pregnancy to occur are several: • The male must produce adequate numbers of normal, motile spermatozoa. • The male must be capable of ejaculating the sperm through a patent ductal system. • The sperm must be able to traverse an unobstructed female reproductive tract. • The female must ovulate and release an ovum. • The sperm must be able to fertilize the ovum. • The fertilized ovum must be capable of developing and implanting in appropriately prepared endometrium. In approximately 40% of cases, infertility is caused by the male (Table 223-5). In one third of couples, more than one cause contributes to the infertility. Peak age for fertility in the female is 25 years. For nulliparous women of this age, the average time during which unprotected intercourse occurs until conception is 5.3 months. For parous women, the average duration of intercourse until conception is 2.7 months. The reproductive performance of couples is influenced by the ages of the female and male partners, the frequency of intercourse, and the length of time the couple has been attempting to conceive. There is a decline in both female and male reproductive performance after the age of 25 years.  

DIAGNOSIS

Couples who complain of infertility merit evaluation regardless of the length of infertility. Evaluation is warranted in all women after 12 months and in women 35 years of age or older after 6 months of regular unprotected intercourse. The evaluation begins with a detailed history obtained from both partners and physical examinations of both individuals. If possible, the couple should be seen together. Each couple should be questioned together and separately because separate interviews may uncover information that would not be imparted in the presence of the partner. Initial evaluation for infertility includes assessment of semen; documentation of ovulation by basal body temperature, serum progesterone determination 6 to 8 days before menses, serum thyroid hormone, or (rarely) endometrial biopsy less than 3 days before onset of menses; and evaluation of the female genital tract by hysterosalpingography or sonohysterography. Diagnostic laparoscopy with tubal dye instillation may be performed if results of all previous tests are normal because 30 to 50% of women are found to have endometriosis or tubal disease on surgical evaluation; alternatively, patients with initial normal findings may be merely treated as having idiopathic infertility.

1560

CHAPTER 223  Reproductive Endocrinology and Infertility  

TABLE 223-5 CAUSES OF INFERTILITY AND THEIR APPROXIMATE INCIDENCE Male factors (40%) Decreased production of spermatozoa Varicocele Testicular failure Endocrine disorders Cryptorchidism Stress, smoking, caffeine, nicotine, recreational drugs Ductal obstruction Epididymal (after infection) Congenital absence of vas deferens Ejaculatory duct (after infection) After vasectomy Inability to deliver sperm into vagina Ejaculatory disturbances Hypospadias Sexual problems (i.e., impotence), medical or psychological Abnormal semen Infection Abnormal volume Abnormal viscosity Immunologic factors Sperm-immobilizing antibodies Sperm-agglutinating antibodies Female factors Fallopian tube disease (20-30%) Pelvic inflammatory disease or puerperal infection Congenital anomalies Endometriosis Secondary to past peritonitis of nongenital origin Amenorrhea and anovulation (15%) Minor ovulatory disturbances (400 mg/24 hr), and increased stone risk by biochemical stone risk analysis • Significant reduction in creatinine clearance (i.e., 95%) SLE (30%); Sjögren syndrome (70-80%) SLE (15%); Sjögren syndrome (50-70%) SLE (10%) PSS (40-70%); CREST (10-20%)

Nucleolar

Anti–PM-Scl Anti–Mi-2 Anti-RNA polymerase

Nucleolar protein complex Nuclear protein complex Subunits of RNA polymerase I

PSS (3%); PM (8%) DM (15-20%) PSS (4%)

Dividing cell

Anticentromere Antiproliferating cell nuclear antigen

Centromere/kinetochore protein Auxiliary protein of DNA polymerase δ

CREST (80%); PSS (30%) SLE (3%)

Cytoplasmic

Anti–Jo-1 Anti–PL-7 Anti–PL-12 Anti-SRP Anti–ribosomal P

Histidyl tRNA synthetase Threonyl tRNA synthetase Alanyl tRNA synthetase Signal recognition particle Large ribosomal subunit

ILD in PM/DM (18-25%) PM/DM (3%) PM (4%) PM/DM (3%) SLE (10%)

CREST = calcinosis, Raynaud phenomenon, esophageal dysmotility, sclerodactyly, and telangiectasia; DM = dermatomyositis; ILD = interstitial lung disease; MCTD = mixed connective tissue disease; PM = polymyositis; PSS = progressive systemic sclerosis (diffuse scleroderma); SLE = systemic lupus erythematosus; snRNP = small nuclear ribonucleoprotein; tRNA = transfer RNA.

CHAPTER 242  Laboratory Testing in the Rheumatic Diseases  

complaints such as arthralgias, fatigue, and fever, a positive test must be interpreted with caution and not used as proof of a CTD in the absence of correlative clinical or laboratory findings. Because the molecular target of many ANAs is now known, testing of individual antibodies can now be performed by ELISA or LINE-based approaches. Furthermore, multiplex assays provide simultaneous assessment of antibodies to a limited number of specificities and therefore may fail to detect certain ANAs. Although multiplex assays are operationally easier than conventional ANA testing by immunofluorescent staining, in clinical situations, the IF assay remains an important laboratory test for patient evaluation and can be used to verify results from a multiplex assay Among many ANA specificities now identified, only a few are performed routinely because of their value for diagnosis and prognosis. For certain CTDs, diagnosis can be readily determined from clinical findings or other laboratory tests. In these instances, the ANA determination provides confirmatory information as well as clues to the occurrence of certain clinical manifestations.

Antibodies to DNA

Antibodies to DNA (anti-DNA) are serologic markers of SLE and represent a criterion in the classification of patients with this disease (Chapter 250).10 These antibodies bind sites on both single-stranded (ss) and doublestranded (ds) DNA, although anti-dsDNA antibodies are more specific for SLE and therefore routinely measured. Although these antibodies can bind free DNA, DNA in the cell occurs in association with histones to form a structure called the nucleosome, with DNA wrapped around a histone core. Anti-DNA may therefore be considered a subset of antibodies to nucleosomes, with nucleosomes probably serving as the driving antigen for this response. Anti-DNA determinations, in addition to their value in diagnosis, can serve as an index of disease activity. The association with disease activity appears strongest with glomerulonephritis, most likely because of the role of DNA– anti-DNA immune complexes in immunopathogenesis. The association of anti-DNA antibodies with other disease manifestations is less certain, limiting the use of this marker as a measure of overall disease activity. The presence of anti-DNA may nevertheless be important in assessing likelihood of response to therapies such as belimumab (anti-BLyS or anti-BAFF), an agent indicated for treatment of patients with active disease as evidenced by the presence of either anti-DNA antibodies or a positive test for ANA.11 Several immunochemical approaches can be used to detect anti-DNA antibodies. The assays vary with regard to the spectrum of anti-DNA antibodies detected, and results between assays may not correlate. Nevertheless, for each assay, the dynamic range for testing is large. With treatment and disease quiescence, anti-DNA antibodies may essentially disappear; with flare, levels may increase dramatically. This property distinguishes anti-DNA antibodies from other ANAs in SLE, levels of which tend to be more consistent over time. As is the case for other ANAs, the appearance of anti-DNA antibodies in the serum may precede other manifestations of SLE, suggesting vigilance if these antibodies are present in patients who have symptoms that suggest a CTD but lack other evidence to establish a firm diagnosis.

Other Antinuclear Antibodies

Anti-Sm and anti-RNP antibodies are related specificities that commonly occur together in the sera of patients with SLE, a phenomenon called linkage. These antibodies bind proteins on subcellular particles called snRNPs (small nuclear ribonucleoproteins) that are composed of a set of proteins and uridine-rich RNAs. Anti-Sm and anti-RNP antibodies differ in protein specificity and in the ability to cause immunoprecipitation of the bound RNA molecules. Anti-Sm antibodies occur only in patients with SLE and represent a serologic marker for disease classification. In contrast, anti-RNP antibodies can appear in the sera of patients with other clinical presentations and, in the absence of anti-Sm, may characterize patients with overlapping CTD features, so-called mixed CTD or MCTD. In SLE, the frequencies of anti-Sm and anti-RNP antibodies vary among racial and ethnic groups, although a clear association with particular clinical manifestations has not been established. Anti-Ro and anti-La antibodies (or anti–SS-A and anti–SS-B), another set of linked ANAs, are directed to protein-RNA complexes that are involved in cellular metabolism of RNA. Two forms of Ro exist. Ro60 is an RNA-protein complex, whereas Ro52 is a protein known as TRIM22. Anti-Ro and anti-La antibodies are expressed more widely in patients with CTD and appear in the sera of patients with SLE, RA, and Sjögren syndrome, among others. Assessment of these antibodies is important because of their association with the neonatal lupus syndrome, which results from the transplacental passage of antibodies

1677

and causes congenital heart block as well as rash in the neonate. Although both Sm/RNP and Ro/La are complexes of proteins and RNA, these antibodies appear to be expressed by different patient subsets, suggesting distinct mechanisms of induction and clinical associations. Although ANAs are directed to ubiquitous antigens, they nevertheless are expressed in disease-specific patterns and may show association with particular organ-specific manifestations. These associations include anti–ribosomal P antibodies with central nervous systemic involvement in SLE, antibodies to DNA topoisomerase 1 (anti–SCL-70) with progressive systemic sclerosis (diffuse scleroderma), antibodies to centromeres with CREST syndrome (calcinosis, Raynaud phenomenon, esophageal dysmotility, sclerodactyly, and telangiectasia), and antibodies to histidyl transfer RNA synthetase (anti–Jo-1) with interstitial lung disease in scleroderma (Chapter 251). In inflammatory myopathies, the presence of certain autoantibodies may be associated with particular patterns of disease, with antibodies to the enzyme 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase present in a syndrome of necrotizing myositis; the syndrome can occur in patients treated with statins, which can inhibit the enzyme.12 In addition to their association with specific disease manifestations, antibodies to both DNA- and RNA-binding proteins such as Sm and RNP may contribute to overall immune dysregulation in patients with autoimmune disease because of their formation of immune complexes containing DNA or RNA. These complexes can stimulate the production of type 1 interferon by triggering both TLR and non-TLR nucleic acid sensors as well as other cellular receptors (e.g., Fc receptors). Because immunoassays of interferon with patients’ sera are limited, the presence of interferon is observed more clearly in the pattern of gene expression known as the interferon signature in peripheral blood cells. This signature can be assessed by both microarray assays and measurement of more limited sets of messenger RNA molecules. Because antibodies to RNA-binding proteins in particular may promote this pattern, the serologic assay of these ANAs may allow assessment of the likelihood of both nonspecific and specific immunologic disturbances.

Antibodies to Phospholipids

Originally defined by their effects on in vitro clotting tests, antibodies to phospholipids (APLs) are associated with in vivo thrombosis and have been termed lupus anticoagulants (LACs).13 Patients with these antibodies display a clinical condition, termed the antiphospholipid antibody syndrome,14 which is characterized by arterial or venous thrombosis, thrombocytopenia, and first trimester spontaneous abortions (Chapter 73). This syndrome may occur by itself or in the context of SLE, where it may contribute to the acceleration of atherosclerosis, premature stroke, and myocardial infarction. The laboratory evaluation of this condition involves specific assays of antibodies to phospholipids and related proteins as well as functional assays of clotting. Because expression of these antibodies may vary over time, testing must be performed on more than one occasion at least 6 weeks apart. Furthermore, the results of immunochemical and functional assays may not be congruent, likely related to the heterogeneity of antibodies. The serology of APLs is complicated because it is related to the nature of the antigenic targets as well as heterogeneity among patients. These antigens include phospholipids such as cardiolipin. Cardiolipin, however, can bind to the protein β2-glycoprotein 1, which is also a target for antibodies in this condition. Serologic evaluation thus involves assays with a complex of cardiolipin and β2-glycoprotein 1 as well as β2-glycoprotein in an ELISA format using reagents to measure IgG, IgA, and IgM, although the association of antibodies with thrombosis appears strongest with IgG antibodies. In interpreting these assays, it is important to know the cutoff values used to define positivity. Functional assays for LACs involve tests directed at inhibition of in vitro clotting (e.g., activated partial thromboplastin time, dilute Russell viper venom time), recognizing the discordance between in vivo thrombosis and in vitro anticoagulation. Functional assays to detect LACs involve a mixing step in which patient plasma is mixed with normal plasma to determine the presence of an inhibitor (i.e., an antibody) as opposed to a deficiency state. The mechanisms by which antibodies to phospholipids and related proteins may cause thrombosis in vivo are unknown, although these antibodies may interact with the surface of cells (e.g., endothelium) to promote a prothrombotic state. Assessing the likelihood of the syndrome is best accomplished by considering assay results in the context of the individual patient.

Complement

Assessment of the complement system can provide valuable information on the activity of diseases in which immune complex deposition may promote

inflammation and tissue injury (Chapter 44).15 This system involves a large number of proteins that function in enzyme cascades to generate degradation products that amplify immunologic reactions and promote the destruction or removal of foreign organisms as well as damaged cells. In the setting of SLE and in certain forms of vasculitis and glomerulonephritis, immune complexes activate complement, which can be measured in terms of the total complement level, levels of C3 and C4, and levels of complement fragments bound to red blood cells. Proteins of the complement system are acute phase reactants and can increase with inflammation, including active disease. Correspondingly, low levels may reflect inherited complement deficiency or copy number variations rather than consumption; genetic deficiency of C1q, for example, is highly associated with SLE.

Antineutrophil Cytoplasmic Antibodies

Antineutrophil cytoplasmic antibodies (ANCAs)16,17 are autoantibodies that react to determinants in the neutrophil and occur prominently in patients with certain forms of necrotizing vasculitis or rapidly progressive glomerulonephritis. Reflecting the serology, conditions have been called ANCA-associated vasculitis (AAV). Two main forms of ANCA have been distinguished on the basis of the target antigens and pattern of immunofluorescence staining of fixed neutrophils: PR3-ANCA (C-ANCA), which reacts with proteinase-3 (PR3), and MPO-ANCA (P-ANCA), which reacts with myeloperoxidase (MPO). By immunofluorescence, PR3-ANCA shows staining in the cytoplasm; staining by MPO-ANCA localizes in the perinuclear area. ANCAs can occur in other conditions including an immune syndrome resulting from the drug levamisole, which is used as an adulterant in cocaine. In the evaluation of severe, multisystem inflammatory disease, ANCA testing is important to evaluate diagnostic possibilities. ANCAs occur in association with varying clinical manifestations in patients with AAV and help define patterns of clinical involvement in terms of organ system involvement as well as histopathology (e.g., presence of granulomatous inflammation).18,19 PR3ANCA occurs commonly in patients with granulomatosis with polyangiitis (GPA, formerly called Wegener granulomatosis) as well eosinophilic granulomatosis with polyangiitis (EGPA, formerly called Churg-Strauss disease); MPO-ANCA marks the course of vasculitis caused by microscopic polyangiitis. Although there is overlap between serology and clinical features, PR3-ANCA occurs commonly in patients with upper airway disease, whereas MPO-ANA occurs commonly patients with patients with rapidly progressive renal disease (Chapter 254). In patients with ANCA-associated glomerulonephritis, the kidney lacks evidence of immune deposits, as indicated by the lack of staining for immunoglobulins or complement. Kidney disease of this kind is termed pauci-immune glomerulonephritis. Although ANCA testing is useful in initial diagnosis, its role for assessing disease activity is less certain. Occasionally, in patients who are desperately ill and cannot tolerate a lung or kidney biopsy, the presence of an ANCA can be used as preliminary evidence for diagnosis to allow the initiation of immunosuppressive therapy. ANCA testing is also useful for assessing the likelihood for relapse because patients who express PR3-ANCA appear at risk for recurrent disease.  

Cryoglobulins

Cryoglobulins are serum immunoglobulins that precipitate in the cold and can mediate disease through tissue deposition.20 The presence of a cryoglobulin is detected by allowing blood, collected warm, to remain cool at 2° to 4° C for 1 or more days. After centrifugation, the amount of cryoprecipitate is measured and expressed as a cryocrit. In the preanalytical phase, it is important that the blood remain at a temperature of 37° C during all steps.21 Subsequent analysis of the cryoprecipitate by immunochemical assays allows determination of its components. Cryoglobulins can be classified into three main types on the basis of their composition: (1) single, or type I; (2) mixed, type II; and (3) mixed, type III. A type I cryoglobulin consists of only a monoclonal immunoglobulin that precipitates in the cold. A mixed-type cryoglobulin contains RFs bound to polyclonal IgG to form an immune complex. In type II cryoglobulins, the IgM RF is monoclonal, and in type III, the IgM RF is polyclonal. Type I cryoglobulins occur in patients with lymphoproliferative disorders such as Waldenström macroglobulinemia, multiple myeloma, or chronic lymphocytic lymphoma (Chapters 174 and 178). In contrast, patients with mixed cryoglobulins can present with a wide range of signs and symptoms resulting from vasculitis. These manifestations include purpura (a sign of leukocytoclastic vasculitis), weakness, arthritis, and neuropathy, representing a syndrome known as essential mixed cryoglobulinemia. Most patients with this condition have

infection with hepatitis C virus, with viral components present in the complexes. These patients have serologic evidence of this infection as well as manifestations attributable to the underlying liver disease. As in the case of other CTDs and systemic inflammatory diseases, the evaluation of patients with essential mixed cryoglobulinemia demands attention to the entire patient and the impact of disease on multiple organs.   GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 242  Laboratory Testing in the Rheumatic Diseases  

GENERAL REFERENCES 1. George MD, Giles JT, Katz PP, et al. The impact of obesity and adiposity on inflammatory markers in patients with rheumatoid arthritis. Arthritis Care Res (Hoboken). 2017;69:1789-1798. 2. Derksen VF, Ajeganova S, Trouw LA, et al. Rheumatoid arthritis phenotype at presentation differs depending on the number of autoantibodies present. Ann Rheum Dis. 2017;76:716-720. 3. Miller A, Nightingale AL, Sammon CJ, et al. Estimating the diagnostic accuracy of rheumatoid factor in UK primary care: a study using the Clinical Practice Research Datalink. Rheumatology (Oxford). 2015;54:1882-1889. 4. Malmstrom V, Catrina AI, Klareskog L. The immunopathogenesis of seropositive rheumatoid arthritis: from triggering to targeting. Nat Rev Immunol. 2017;17:60-75. 5. van Zanten A, Arends S, Roozendaal C, et al. Presence of anticitrullinated protein antibodies in a large population-based cohort from the Netherlands. Ann Rheum Dis. 2017;76:1184-1190. 6. Heselden EL, Freemont AJ. Synovial fluid findings and demographic analysis of patients with coexistent intra-articular monosodium urate and calcium pyrophosphate crystals. J Clin Rheumatol. 2016;22:68-70. 7. Pisetsky DS. Antinuclear antibody testing: misunderstood or misbegotten? Nat Rev Rheumatol. 2017;13:495-502. 8. Stochmal A, Czuwara J, Trojanowska M, et al. Antinuclear antibodies in systemic sclerosis: an update. Clin Rev Allergy Immunol. 2019. [Epub ahead of print.] 9. Fatemi G, Kermani TA. Clinical significance of a positive antinuclear antibody test. JAMA. 2015; 314:827-828. 10. Qu C, Zhang J, Zhang X, et al. Value of combined detection of anti-nuclear antibody, anti-doublestranded DNA antibody and C3, C4 complements in the clinical diagnosis of systemic lupus erythematosus. Exp Ther Med. 2019;17:1390-1394.

1678.e1

11. Pisetsky DS, Rovin BH, Lipsky PE. New perspectives in rheumatology: biomarkers as entry criteria for clinical trials of new therapies for systemic lupus erythematosus: the example of antinuclear antibodies and anti-DNA. Arthritis Rheumatol. 2017;69:487-493. 12. Satoh M, Tanaka S, Ceribelli A, et al. A comprehensive overview on myositis-specific antibodies: new and old biomarkers in idiopathic inflammatory myopathy. Clin Rev Allergy Immunol. 2017;52:1-19. 13. Gupta A, Johnson DH, Nagalla S. Antiphospholipid antibodies. JAMA. 2017;318:959-960. 14. Garcia D, Erkan D. Diagnosis and management of the antiphospholipid syndrome. N Engl J Med. 2018;378:2010-2021. 15. Birmingham DJ, Hebert LA. The complement system in lupus nephritis. Semin Nephrol. 2015;35: 444-454. 16. Romero-Sanchez C, Benavides-Solarte M, Galindo-Ibanez I, et al. Frequency of positive ANCA test in a population with clinical symptoms suggestive of autoimmune disease and the interference of ANA in its interpretation. Rheumatol Clin. 2019. [Epub ahead of print.] 17. Csernok E, Mahrhold J, Hellmich B. Anti-neutrophil cytoplasm antibodies (ANCA): recent methodological advances-lead to new consensus recommendations for ANCA detection. J Imunol Methods. 2018;456:1-6. 18. Bui VL, Kermani TA. Clinical significance of a positive antineutrophil cytoplasmic antibody (ANCA) test. JAMA. 2016;316:984-985. 19. Cornec D, Cornec-Le Gall E, Fervenza FC, Specks U. ANCA-associated vasculitis: clinical utility of using ANCA specificity to classify patients. Nat Rev Rheumatol. 2016;12:570-579. 20. Kolopp-Sarda MN, Miossec P. Cryoglobulins: an update on detection, mechanisms and clinical contribution. Autoimmun Rev. 2018;17:457-464. 21. Muchtar E, Magen H, Gertz MA. How I treat cryoglobulinemia. Blood. 2017;129:289-298.

1678.e2

CHAPTER 242  Laboratory Testing in the Rheumatic Diseases  

REVIEW QUESTIONS 1. Which of the following antibodies is a classification criterion for systemic lupus erythematosus? A . Anti-La B. Anti-Ro C. Anti-RNP D. Anti-Sm E. Anti–ribosomal P Answer: D  Although the other antibodies are often found in patients with lupus, only anti-Sm is a classification criterion. Anti-Sm and anti-RNP commonly occur together, and in lupus, anti-RNP may be more frequent. AntiRNP, however, has been associated with an entity known as mixed connective tissue disease, which has overlapping features with other rheumatic diseases. As a result, anti-RNP has not been a classification criterion. 2. What is the antigenic target of rheumatoid factors? A . Lambda light chain B. Kappa light chain C. Antibody Fc D. Antibody Fab E. Antibody F(ab′)2. Answer: C  RFs are IgM antibodies directed to the Fc portion of the IgG molecule. These antibodies likely react with a conformational determinant that becomes available when an IgG molecule binds to its antigen. Denatured IgG preparations may also expose this determinant. Although the predominant antigenic target of RFs is the Fc portion, some antibodies to IgG may react to other parts of the IgG molecule. 3. Citrulline is a post-translational modification of which amino acid? A . Arginine B. Lysine C. Asparagine D. Aspartic acid E. Glutamic acid Answer: A  Citrulline is a post-translational modification of arginine that is mediated by an enzyme called peptidylarginine deiminase. The function of this modification is not well understood. Antibodies to citrullinated proteins (ACPAs) can be assayed using synthetic antigens known as cyclic citrullinated peptide (CCP). Antibodies reacting with the peptide are known as anti-CCP. Antibodies can also be assessed using specific citrullinated proteins as antigens.

4. Antibodies to centromeres occur most commonly in which of the following diseases? A . Progressive systemic sclerosis B. CREST syndrome C. Polymyositis D. Dermatomyositis E. Sjögren syndrome Answer: B  Antibodies to centromere proteins occur most commonly in CREST syndrome or limited scleroderma. This syndrome is characterized by calcinosis, Raynaud phenomenon, esophageal dysmotility, and telengiectasias. Antibodies to centromeres can be recognized by a characteristic staining pattern on the ANA as well as ELISA assay using specific proteins as antigens. In contrast to the CREST syndrome, the diffuse form of progressive systemic sclerosis demonstrates antibodies to a protein called SCL-70 or DNA topoisomerase. The basis for these patterns of expression is unclear. Both antigens are ubiquitous and intracellular. It is difficult to understand why these antibodies would be associated with distinct clinical patterns even if there is some overlap in the clinical features. 5. Which of the following antibodies produces rim staining by the indirect immunofluorescence assay for antinuclear antibodies? A . Anti-Sm B. Anti-RNP C. Anti-DNA D. Anti–Jo1 E. Anti-Ro Answer: C  The pattern of immunofluorescence reflects the localization of the target antigen. The difficulty with this approach relates to the complexity of sera from patients with rheumatic diseases, especially lupus. Thus, sera may contain many different specificities, and although each may produce a characteristic staining pattern, the ability to see that pattern depends on the concentration of the different antibodies. Sometimes, serum will show different patterns depending on the concentration. Thus, serum can show homogeneous staining at a titer of 1 : 640 and speckles at a titer of 1 : 2560. Homogenous staining often results from the presence of antibodies to chromatin. Because antibodies to chromatin and anti-DNA commonly coexist, the rim pattern may not be readily observed.

1678

CHAPTER 243  Imaging Studies in the Rheumatic Diseases  

243  IMAGING STUDIES IN THE RHEUMATIC DISEASES RONALD S. ADLER Historically, rheumatic disorders have been well characterized by conventional imaging. In as much as these disorders often manifest in characteristic distributions and present with specific alterations in the appendicular and/or axial skeleton and adjacent soft tissues, radiographic evaluation has been sufficient to characterize the abnormalities as well as provide a relatively small number of differential possibilities as to the specific disease. The most well-studied example is rheumatoid arthritis (RA), in which symmetric involvement of the metacarpophalangeal joints, uniform joint space narrowing, periarticular osteopenia, and juxta-articular erosions along the “bare areas” are pathognomonic. The development of new therapeutic alternatives for the inflammatory arthritides, so-called disease-modifying antirheumatic drugs (DMARDs), and chondroprotective strategies in the case of osteoarthrosis, require methods to diagnose these diseases at an earlier stage, characterize the degree of inflammation, and provide a useful metric to assess therapeutic response (Chapters 32 and 33). Indeed, it has become necessary to assess for possible joint and soft tissue abnormalities before irreversible tissue damage when the radiographic findings are not yet abnormal. Fortunately, the requirement to achieve earlier diagnosis has paralleled advances in imaging. Ultrasonography and magnetic resonance imaging (MRI) have largely supplanted conventional radiographic evaluation in the imaging work-up of patients with suspected rheumatologic disorders and negative radiographs. The term molecular imaging has been applied, particularly in the case of MRI and positron emission tomography (PET), in as much as these modalities reflect local tissue environment or metabolic activity.1,2

  RADIOGRAPHIC EVALUATION

Radiographic evaluation is among the first studies ordered in patients with a suspected rheumatologic disorder. In the current digital era, conventional analog-based radiographs have been largely replaced by computed radiography. Images are usually displayed on workstations with high-resolution monitors within the context of a picture archiving system (PACS). Digital radiographs are of high spatial resolution but relatively poor soft tissue contrast. These images are amenable to a variety of image processing schemes, resulting in enhanced definition of the cortical surfaces and cancellous bone, which may be of value in displaying subtle erosions. It is important to recognize that radiographs are projection images. To detect an abnormality, it may be necessary to view a joint or other structure at a specific angle. For instance, subtle erosions may be apparent only when viewed tangentially, as opposed to en face. It is therefore necessary to have specific image protocols to optimally display the joint, cortical surface, or soft tissue structure. Most radiographic evaluations contain at least two orthogonal projections. The addition of an oblique view or other specialized projection may be necessary to address a specific clinical question. The nature and distribution of joint space narrowing, presence of osteopenia, new bone formation, soft tissue swelling, soft tissue calcification, chondrocalcinosis, presence and nature of erosions, and assessment for joint malalignment may allow a specific diagnosis, as well as help determine the severity of disease (Fig. 243-1). For instance, the presence of a juxta-articular erosion

CHAPTER 243  Imaging Studies in the Rheumatic Diseases  

ABSTRACT

Historically, rheumatic disorders have been well characterized by conventional imaging. Radiographic evaluation has been sufficient both to characterize these conditions and to provide a relatively small number of differential diagnostic possibilities regarding the specific disease. The development of new therapeutic alternatives for the inflammatory arthritides, the so-called disease-modifying antirheumatic drugs, and chondroprotective strategies in the case of osteoarthrosis require methods to diagnose these diseases at an earlier stage, characterize the degree of inflammation, and provide a useful metric to assess therapeutic response. Fortunately, the requirement to achieve earlier diagnosis has paralleled advances in imaging. Ultrasound and magnetic resonance imaging have largely supplanted conventional radiographic evaluation in much of the imaging work-up of patients with suspected rheumatologic disorders. Radiographic evaluation, computed tomography, magnetic resonance imaging, ultrasound, and nuclear medicine collectively provide important information in evaluating patients with these disorders and are briefly reviewed.

KEYWORDS

imaging MRI ultrasound computed tomography nuclear medicine arthritis scintigraphy

1678.e3

CHAPTER 243  Imaging Studies in the Rheumatic Diseases  

A

1679

C

B

FIGURE 243-1.  Three hands with different diagnoses. A, Gout. Radiograph of the left hand showing multiple dense soft tissue nodules (n) with multiple small erosions affecting the ulnar styloid, triquetrum, and fifth ray. A large erosion (arrow) at the fifth distal interphalangeal (DIP) joint demonstrates bone formation extending circumferentially about the adjacent tophaceous deposit typical of an overhanging edge. Bone mineralization and joint spaces are preserved. B, Rheumatoid arthritis (RA). There is ulnar deviation of the second through fifth metacarpophalangeal (MCP) joints with uniform joint space loss involving the MCP joints and the carpus. There is mild ulnar translation of the carpus. The DIP joints are spared. Periarticular demineralization is present with small erosions along the radiovolar aspect of the second (arrow) MCP joint. L = left. C, Osteoarthrosis. Soft tissue swelling affecting the third digit with joint space narrowing and bone production affecting the DIP joints, third and fifth proximal interphalangeal (PIP) joints, basal joint of the thumb, and scaphotrapeziotrapezoid joint. There are subchondral cystic changes at the third PIP joint having an erosive character (arrow). Mineralization is preserved, and radiocarpal as well as MCP joint spaces are preserved.

TABLE 243-1 DISTINGUISHING RADIOGRAPHIC FEATURES OF SEVERAL COMMON RHEUMATIC DISEASES CONDITION Rheumatoid arthritis

COMMON SITES

DISTRIBUTION

Hands: MCP, PIP; wrists: intercarpal, Bilateral, symmetric, polyarticular DRUJ, ulnar styloid; feet: fifth MTP; cervical spine (atlantoaxial, apophyseal)

RADIOGRAPHIC FEATURES Juxta-articular osteopenia, periarticular swelling, subluxations (e.g., ulnar, volar), uniform joint space loss, erosions (bare areas)

Osteoarthritis (primary) Hands (DIP), wrists (basal joint, STT), feet (first MTP), hips (superolateral), knees (medial), spine (discs, facet, or apophyseal, uncovertebral)

Symmetric, weight-bearing joints

Normal or increased density, nonuniform joint space loss, subchondral sclerosis, cysts, bone formation (osteophytes) Spine: disc space narrowing, end plate sclerosis, and bone formation

Psoriatic arthritis

Hands (DIP, terminal tufts), feet (IP joints), entheses (calcaneus-plantar, posterior), spine, SI joints

Asymmetric (single ray), polyarticular, Normal or increased density, periosteal bone formation, soft segmental (intervertebral, apophyseal) tissue swelling, ankylosis (SI joints), thick hyperostosis spine (nonmarginal syndesmophytes), juxta-articular and periarticular erosions

Ankylosing spondylitis

Spine, SI joints, fibrous joints (pubic symphysis), entheses (adductor origin), rhizomelic joints (hips, shoulders)

Symmetric, continuous (may affect entire Normal or increased density, erosions (spine: squaring, spine: bamboo spine) shining corner) with superimposed bone formation (ankylosis: SI, thin [marginal] syndesmophytes)

Gout

Feet (first MTP), other damaged joints, elbow, knee, hindfoot

Asymmetric, extensor surfaces (elbow), abnormal joints (e.g., osteoarthritic joints)

Normal joint space, normal or increased density, dense soft tissue nodules (tophi), para-articular and subchondral erosions with bone formation along tophi (overhanging edge)

CPPD

Hands (second, third MCP), wrists (radiocarpal, TFC), knees (lateral compartment and patella-femoral, menisci)

Symmetric, fibrocartilaginous joints

Normal or increased density, hypertrophic bone formation, subchondral or periarticular cysts, chondrocalcinosis (hyaline, fibrocartilage), periarticular, peritendinous, periligamentous calcification

Infection

Any joint (pyogenic, TB)

Monoarticular (mostly), any joint

Pyogenic: osteopenia (days), joint space widened (early), joint space loss (rapid development), soft tissue swelling, erosions—both sides of joint, sequestra, periostitis TB: joint space and mineralization may be preserved, juxta-articular erosions Spine: disc space loss and end plate erosion

CPPD = calcium pyrophosphate deposition disease; DIP = distal interphalangeal; DRUJ = distal radial ulnar joint; IP = interphalangeal; MCP = metacarpophalangeal; MTP = metatarsophalangeal; PIP = proximal interphalangeal; SI = sacroiliac; STT = scaphotrapeziotrapezoid; TB = tuberculosis; TFC = triangular fibrocartilage.

extending over an adjacent area of slightly hyperdense soft tissue swelling in the setting of normal bone mineralization with maintenance of the adjacent joint space is diagnostic of gout, in contrast to RA noted earlier. The seronegative arthritides, such as psoriatic arthritis, have a characteristic appearance in the small joints of the hand and feet, with a predilection for distal joints, asymmetry, and appositional new bone formation.

Table 243-1 summarizes some of the features of several of the more common diseases that may be encountered in clinical practice. Finally, radiographs provide a direct means for needle localization during percutaneous procedures, predominantly joint injections, aspirations, and some biopsies. These are generally performed while imaging in real time (fluoroscopy) using short bursts of low-intensity x-rays enhanced through an image

1680

CHAPTER 243  Imaging Studies in the Rheumatic Diseases  

intensifier. Injection of joints under fluoroscopic guidance provides a convenient means to ensure intra-articular deposition of the therapeutic agent or for diagnostic aspiration. Intra-articular location is verified by injection of a small amount of a standard iodinated contrast material. Arthrography using fluoroscopic guidance can be used as a primary diagnostic tool, but this application has largely been replaced by intra-articular injection of contrast followed by computed tomography (CT) or MRI. For some procedures, CT may be preferable, depending on the location of the abnormality. The principal disadvantages of fluoroscopy relate to the use of ionizing radiation and poor soft tissue contrast. The latter becomes important with needle placement near neurovascular structures that may be potentially compromised by poor needle position. CT allows greater control over needle placement at the cost of greater levels of radiation exposure. Ultrasonography has replaced fluoroscopy and CT for a large number of percutaneous procedures. MRI provides another method to perform a variety of procedures without the necessity of ionizing radiation. These options will be discussed in greater detail below.

  COMPUTED TOMOGRAPHY

CT provides a two-dimensional map of tissue attenuation obtained from external x-ray source(s) located on a rotating gantry, whose radiation is detected by a series of detectors opposite the source. The current generation of CT scanners employs multiple detectors (16, 32, 64, and so on), allowing rapid image acquisition that can be displayed in a single plane in real time (CT-fluoroscopy) or as extremely thin section contiguous or overlapping acquisitions in the axial plane. The acquired images can be reconstructed in multiple planes with equivalent (isotropic) resolution elements (voxels) or as a three-dimensional rendering. Image data are generally obtained with the scanner operating in a helical mode (as the subject is advanced continuously while data are obtained), enabling rapid acquisitions. Image reconstruction has traditionally been performed using a technique known as filtered back-projection. Newer techniques involve iterative reconstruction, which is promising as a method to achieve significant decrease in image reconstruction times, as well as radiation dose reduction. Some scanners use dual energy sources, taking advantage of differences in the attenuation characteristics of various tissues at different energies. This has received greatest attention in the setting of gout, enabling a definitive diagnosis with greater sensitivity in depicting tophaceous deposits even in anatomic locations not conducive to radiographs or ultrasound.3 CT allows the best assessment of trabecular and cortical bone, providing an excellent means to assess fractures and erosions, the presence of new bone formation (e.g., fracture callus), and degenerative or inflammatory arthritis. Soft tissue mineralization can likewise be well characterized, providing important information as to its etiology. Joints that are difficult to assess on radiographs, including the sacroiliac, temporomandibular, wrist, and sternoclavicular joints, are well seen on CT (Fig. 243-2).

A

CT generally has poor soft tissue contrast. Nevertheless, it is still very useful in performing a number of guided procedures because of its tomographic nature and rapid image acquisition capability. Improved soft tissue contrast can be obtained with use of iodinated contrast material. A number of soft tissue tumors, inflammatory synovitis, and infectious processes display pathologic enhancement after contrast administration. CT can likewise be used to produce angiographic displays (CTA) when used in combination with contrast, providing exquisite detail of central and peripheral vascular disease, including in patients with suspected vasculitis. These agents are typically administered intravenously following well-defined enhancement characteristics. CTA has become the method of choice in evaluating patients with suspected pulmonary embolism. Likewise, contrast agents may be used to improve intra-articular contrast (CT arthrography), currently the method of choice in assessing internal derangement in the postoperative shoulder, knee, and so on and in patients who are unable to undergo MRI (e.g., those with claustrophobia, aneurysm clips, or cardiac pacemakers). Imaging of cartilage and soft tissue abnormalities usually depends on pathologic inhibition of contrast material, indicative of degeneration or tearing. A limitation of this approach resides in the fact that some abnormalities may remain occult. An example is the inability to detect a bursal-sided rotator cuff tear after shoulder CT arthrography. The radiation dose from CT can be high, especially when using the newer scanners. This is most significant when one is looking to minimize exposure, such as in children, requiring protocols specifically designed for the pediatric population. Newer iterative image reconstruction techniques as well as sparse sampling (compressed sensing) may allow for compensatory dose reduction. Intravenous (IV) use of iodinated contrast agents is contraindicated in patients with impaired renal function or a history of allergic reaction. Nonionic agents can diminish the associated risks but still should be used with caution.

  ULTRASONOGRAPHY

Ultrasound imaging takes advantage of the near uniform speed of sound and predictable attenuation characteristics of sound propagation in soft tissue. Diagnostic frequencies used in musculoskeletal ultrasound range from approximately 5 to 20 megahertz (MHz). Images are formed using a pulseecho technique, whereby a transducer produces a short duration series of pulses (lasting on the order of microseconds) and then goes into a receive mode from which image information is acquired and processed. In general, anatomic images derive from specular surfaces whose dimensions exceed the ultrasound wavelength; inherent noise (speckle) within the image derives from small scatterers, smaller than the resolution element of the transducer. Modern ultrasound equipment contains various methods to reduce speckle in the image, resulting in a more anatomic rendition of the soft tissues. Rapid image acquisition and processing enables ultrasound to be performed in real time (approximately 30 frames per second). Ultrasonography is also conducive to evaluation of blood flow from which estimates of flow velocity can

B

FIGURE 243-2.  Infectious sacroiliitis in a 12-year-old boy with a 2-week history of back and left hip pain. A, Axial computed tomography (CT) image of the pelvis at the level of the sacroiliac (SI) joints photographed using window settings optimized for bone detail. There is clear asymmetry in the two SI joints, with the left appearing more irregular. The cortical margins of the left sacral ala are less distinct, and there is an isolated bone fragment (arrow) surrounded by soft tissue suspicious for a sequestrum. B, CT-guided aspiration of the left SI joint confirmed an infectious origin.

CHAPTER 243  Imaging Studies in the Rheumatic Diseases  

be obtained through the Doppler equation. Doppler information is typically reported by either continuously estimating velocity at a specific depth (spectral Doppler) or through a color encoded two-dimensional map (color or power Doppler). There is great appeal for using ultrasonography in patients with rheumatic disorders. There is no ionizing radiation, and it is real-time, inexpensive, relatively portable, and well tolerated. Historically, however, ultrasonography has played only a limited role in the diagnostic assessment and treatment of patients with suspected musculoskeletal abnormalities, being used to differentiate fluid-filled from solid masses. The detection of a Baker cyst in the knee or the presence of a joint effusion constituted two major applications. There has also been limited application of ultrasonography to perform image-guided aspirations and biopsies. Within the United States in particular, the development of MRI further limited the musculoskeletal applications of ultrasonography. With the development of linear high-frequency small parts transducers, new imaging capabilities of ultrasound scanners, and the evolution of a new class of compact and portable (laptop) ultrasound units that have excellent image quality, the role of ultrasonography has dramatically changed in recent years.4,5 These new applications have paralleled the development of new classes of DMARDs for which diagnosis of inflammatory synovitis prior to joint destruction is a prerequisite. The current generation of ultrasound scanners enables examination of the small joints of the hands and feet, allowing early detection of synovitis (Fig. 243-3). Typically, a 10-MHz or higher-frequency linear transducer is used. The displacement of the joint capsule by hypoechoic (dark) soft tissue that displays vascularity on Doppler imaging or is incompressible with direct pressure by the transducer is characteristic, allowing differentiation of synovitis from an effusion. In addition to the detection of synovitis, ultrasonography has been shown to be more sensitive than conventional radiographs in the detection of erosions. Erosions appear as discrete irregular discontinuities in the normally smooth hyperechoic (bright), reflecting cortical surfaces, often seen in continuity with adjacent inflammatory soft tissue. There is some variation in the appearance of synovitis among various arthritides. The distribution, presence, or lack of symmetry and other concomitant findings may be necessary to obtain a specific diagnosis. The level of vascularity on color flow imaging can reflect active inflammation, correlating with clinical and biochemical parameters. A parametric image encoding either mean Doppler shift (color Doppler) or amplitude (power Doppler) is typically used as a standard Doppler map. Both maps can be used to detect abnormal levels of vascularity. Whereas power Doppler provides an indirect measure of the number of moving scatterers within the region being scanned, color Doppler provides a velocity map and therefore is more subject to artifact (angle dependence and sampling errors). When combined with color flow imaging, the activity of the synovitis can be estimated. Ultrasound contrast agents can depict capillary flow, resulting in significantly improved detection sensitivity of synovial inflammation and are used extensively in Europe. They constitute microbubble agents encased in a lipid or polysaccharide shell that can be instilled as either bolus or constant infusion, with the shell being metabolized in the liver and the gas exhaled in the lungs. These agents have biologic half-lives on the order of minutes and are best suited to examining target joints. Contrast agents have received U.S. Food and Drug Administration

A

1681

approval for only cardiovascular and abdominal applications at this time and therefore can be used only off label for the assessment of synovitis. Articular cartilage and fibrocartilage have characteristic appearances on ultrasonography. Whereas the former appears as a thin hypoechoic band paralleling the articular surface, fibrocartilage appears hyperechoic. Chondrocalcinosis appears as discrete hyperechoic foci within the substance of the cartilage, in which case its presence is suggestive of calcium pyrophosphate deposition disease. Calcification along the margin of the articular cartilage gives rise to the double-line sign seen in gout. Tendons and muscles have characteristic appearances on ultrasonography. The presence of tendinosis, tendon tears, muscle edema or inflammation, atrophy, and tears can be diagnosed. Ultrasonography is very sensitive, although not specific, for the detection of small amounts of calcification or ossification. It is an excellent method to assess for calcific peritendinitis or periarthritis and to provide guidance for treatment. Abnormal fluid distention of synovial lined structures can be assessed and treated under ultrasound guidance. Ultrasonography is an excellent modality to provide image guidance for therapeutic aspiration and injection of small and large joints, tendon sheaths, and cysts (e.g., bursae, ganglion, paralabral cysts, hematomas, abscesses) (Fig. 243-4). The real-time capability of ultrasonography is useful to demonstrate the presence of subluxations, to evaluate painful snapping, to document the distribution of injected material, and to assess adhesions. Ultrasonography is considered the method of choice to detect foreign bodies. Newer ultrasound techniques provide a map of soft tissue mechanical properties, also known as ultrasound elastography (USE). These techniques fall into two categories, compressionbased or shear wave elastography. The former employs speckle tracking to estimate local tissue strain following minimal external compression of the tissue being examined, thereby producing a color-encoded strain map or map of tissue hardness. The latter method employs a shear-wave produced either by an external source or by the transducer itself, the speed of the shear wave being an indirect measure of local Young modulus; a color-encoded map or quantitative estimate of local shear wave speed or elastic modulus can be derived. Both techniques provide information regarding alterations of mechanical properties of tendons and muscles in a variety of degenerative, traumatic, and reparative states. For example, fatty infiltration resulting from chronic myopathic disorders produces measurable alterations in local muscle elastic modulus. Nerves also have a characteristic appearance on ultrasonography. In crosssection, a nerve often has a “cluster of grapes” or “honeycomb” appearance, with nerve fascicles appearing hypoechoic and surrounded by hyperechoic internal and external epineural fat. In long axis, nerves display a characteristic “tram-track” appearance. Ultrasonography has been shown to be useful in the diagnosis and treatment of carpal tunnel syndrome and cubital tunnel syndrome. It is an excellent modality to assess for the presence of post-traumatic or post-surgical and interdigital neuromas and to provide image guidance for treatment, including therapeutic injections, nerve blocks, and ablative therapy. Although ultrasonography is well suited to the evaluation of superficial structures, it is less well suited to assess deep structures. Frequency and penetration are reciprocally related: the higher the frequency, the better the axial resolution but poorer the degree of penetration. A 15-MHz linear transducer

B

FIGURE 243-3.  Synovitis on ultrasound in a female patient with normal hand radiographs. A, Gray scale ultrasound image obtained along the dorsal aspect of the radioscaphoid joint shows hypoechoic soft tissue (arrows) distending the dorsal recess. The cortical margins of the scaphoid (S) and radius (R) appear as bright reflectors on ultrasound. B, Power Doppler image depicts the marked vascularity (red hues) of the soft tissue illustrating the level of disease activity.

1682

CHAPTER 243  Imaging Studies in the Rheumatic Diseases  

B

A

C

D

FIGURE 243-4.  Ultrasound-guided therapy in the first metatarsophalangeal (MTP) joint of a female patient with pain and swelling. A, Longitudinal gray scale image of the dorsal recess of the first MTP joint. Fluid and soft tissue distend the joint capsule (arrows). The metatarsal (m) and proximal phalanx (p) are labeled. Note that a thin hypoechoic (dark) band parallels the surface of the metatarsal head, corresponding to the overlying articular cartilage. B, Increased vascularity (red hues) demonstrated on power Doppler imaging within the dorsal recess reflects the level of disease activity. C, Transverse gray scale ultrasound image shows a needle (N) within the distended dorsal recess from which several drops of synovial fluid were aspirated, followed by therapeutic injection. D, Postinjection transverse sonogram depicts low-level echoes (small echogenic foci within dorsal recess) and microbubbles (arrows) within the distended joint capsule from injected material. Microbubbles will aggregate along the nondependent portion of the distended joint capsule, whereas injected material tends to settle to the deep potion of the joint.

would work well in the hand but not in the hip. Examination of a hip might require a 5-MHz transducer and curved transducer geometry with reduced image quality. Excessive abdominal fat can further limit acoustic penetration and distort the ultrasound beam, limiting image quality. Diagnostic ultrasonography does not penetrate bone, resulting in limited acoustic access to joint structures. In some instances, soft tissue contrast can be poor. An inexperienced scanner may find ligaments and tendinous insertions difficult to differentiate from adjacent fibrofatty structures.

  MAGNETIC RESONANCE IMAGING

The natural abundance of hydrogen in biologic systems and an inherent property of hydrogen, called spin, form the basis of conventional MRI. When placed in a strong magnetic field, protons tend to align themselves along the direction of the field. Magnetic field strengths are specified as Tesla and can be variable between clinical scanners. The majority of scanners in clinical usage vary between 1 and 3 Tesla. Application of a radiofrequency (RF) pulse to a system of protons induces the spins to rotate away from the direction of the field, during which time they precess about the direction of the magnetic field at a characteristic frequency, called the Larmor frequency. When the RF pulse is turned off, the spins relax toward their initial state determined by two tissuedependent relaxation times, T1 and T2, which vary with field strength. T1 (also known as the spin-lattice relaxation) and T2 (or the transverse relaxation time) along with proton density are the principal determinants of signal intensity. The image can emphasize either the T1 or T2 characteristics of the tissue, impacting tissue contrast. By applying a system of gradient coils, one can effectively spatially encode the distribution of Larmor frequencies. Different tissues have varying appearance often based on levels of fat and water content, reflected by their inherent T1 and T2 relaxation times. Tissue morphology is often characterized by their appearance on T1-weighted or proton density images: tendon, muscle, fat, marrow, cortical bone, articular, and fibrocartilage have characteristic appearances. Many pathologic states, alternatively, are characterized by increased mobile water or effective T2 lengthening. Examples include soft tissue edema, inflammatory infiltrates, and neoplasm (Fig. 243-5). Images that emphasize T2 contrast are therefore helpful to display most pathologic states. Selective maps of T2 have been used to characterize the state of articular cartilage in early degenerative disease. Other cartilage-specific

properties that relate to water content, glycosaminoglycan (GAG) content, and integrity of collagen architecture can be assessed using T2 and other parametric maps that can be derived from the MR data (Fig. 243-6). Measures and parametric images of apparent diffusion coefficient (ADC) provide a method to quantify soft tissue edema present in inflammatory states and may help to differentiate tissue inflammation from other potential sources of increased T2 contrast.6 The loss of restrictive diffusion has likewise been employed as a quantitative metric and imaging tool to assess collagen breakdown in articular cartilage that occurs in early osteoarthritis.7 The latter utilizes a technique known as diffusion tensor imaging (DTI) to estimate the loss asymmetry in the axial versus radial diffusion coefficients present following cartilage injury and has been shown to correlate with both alterations in proteoglycan content and altered collagen architecture. The widely used contrast for MRI studies is a neutral hydrophilic salt of the gadolinium chelate, gadolinium diethylenetriamine-penta-acetic acid (Gd-DTPA). Gadolinium can be injected intravenously or directly into the joint. IV injection (indirect magnetic resonance arthrography) carries the contrast in the vascular system to areas of hyperemia and inflammation (Fig. 243-7). It can be used for assessment of synovial activity in inflammatory joint diseases. Gadolinium is taken up in inflamed synovium and is able to demonstrate thickened pannus. The slope of the early time-signal intensity curve provides a measure of tissue perfusion and can quantify inflammatory activity. Contrast material excreted into the synovial fluid provides excellent depiction of intra-articular structures and can be used in lieu of arthrographic direct techniques. In GAG-depleted cartilage, there can be delayed uptake of contrast into the cartilage, which would normally be inhibited by the negatively charged GAG molecules. Patients with renal disease who receive IV injection of gadolinium can develop nephrogenic systemic fibrosis (NSF) (see Chapter 251). When the kidney cannot sufficiently clear out the gadolinium, it produces fibrosis of many tissues, including the skin, muscle, heart, nerves, and pleura. To date, NSF has been seen only in patients who have been given IV gadolinium with acute or chronic renal insufficiency. The changes in the skin with NSF are usually bilateral and symmetric, primarily involving the extremities and the trunk. These changes can mimic systemic sclerosis but, unlike that disease, the face is usually spared. If renal function improves, the skin lesions may

CHAPTER 243  Imaging Studies in the Rheumatic Diseases  

1683

B

A

FIGURE 243-5.  Magnetic resonance imaging of the right wrist in a female patient with advanced rheumatoid arthritis. A, Proton density coronal image shows loss of normally bright marrow signal within the scaphoid and lunate bones (arrows). The proximal scaphoid is eroded, and the lunate appears deformed and translocated and volarly tilted (not shown), giving rise to its triangular appearance. The distal ulna (u) is poorly visualized due to a large erosion. Intermediate-intensity material (appears dark gray) within the carpus and distal radioulnar joint is difficult to separate from the distal ulna, lunate, and scaphoid. The triquetrum (t), hamate (h), trapezium (tr1) and trapezoid (tr2), capitate (c), and radius (r) are labeled. B, Fluid-sensitive coronal image emphasizing T2 relaxation demonstrates increased signal intensity (bright) within the inflammatory pannus, compatible with increase in mobile water associated with inflammation. Increased signal intensity is evident within the lunate, scaphoid, and distal ulna, including focal areas within the distal row of carpal bones, corresponding to small erosions. Diffuse increased signal within the distal radius likely reflects reactive marrow edema (asterisk).

A

B

C

D

FIGURE 243-6.  Imaging of the soft tissues in a patient with retrocalcaneal pain demonstrating complementary nature of magnetic resonance imaging and ultrasound. A, Proton density sagittal image emphasizes anatomic detail, whereas (B) fluid-sensitive image depicts thickening and increased signal intensity within the distal Achilles tendon reflecting tendinosis, retrocalcaneal bursitis, or tear of the deep surface of the tendon. Surrounding increased signal intensity (bright areas) within the adjacent soft tissue reflects adjacent soft tissue edema. C, Long axis gray scale and (D) power Doppler images of the same patient obtained when the patient presented for ultrasound-guided therapeutic injection. The tendon (arrows) is inhomogeneous. A prominent hypoechoic collection deep to the tendon is compatible with retrocalcaneal bursitis (b). There is prominent increased vascularity on power Doppler imaging at the margin of the bursa and tendon. The calcaneus (calc) is labeled.

1684

CHAPTER 243  Imaging Studies in the Rheumatic Diseases  

A 0

(min–1)

0.38

0

0.57

0

0.05

100 90 80 70 60 50 40 30 20 10

B ms

FIGURE 243-7.  Functional magnetic resonance imaging. A, Parametric image derived from fitting a twocompartment model of soft tissue and synovial enhancement following intravenous administration of contrast material. Three parameters extracted from the time-intensity curves are displayed as parametric images: Ktrans (left) provides a measure of contrast exchange into the extravascular soft tissues; Ve (center) and Vp (right) reflect the relative distribution volumes for the extravascular space and plasma, respectively. The arrows depict a region of increased synovial volume and enhancement at the second metacarpophalangeal joint. Increased values of Ktrans and Ve illustrate increased vascular permeability at the site of inflammation. (Printed with permission of Dr. Luis Beltran.) B, Sagittal T2 map of the knee in which relative T2 relaxation is color encoded, showing regions of higher cartilage T2 values at the femoral condyle and tibial plateau. This reflects alterations in cartilage collagen architecture and water content and possibly early osteoarthritis. (Printed with permission of Dr. Gregory Chang.)

stabilize or get better, although in some patients, the process progresses, affecting mobility and causing severe pain. Injection of dilute gadolinium into the joint (direct MR arthrography) is helpful for outlining structures to determine whether there is morphologic damage. Injection is usually performed either under fluoroscopic or ultrasound guidance. This technique is particularly effective for visualization of small structures such as the labrum of the hip or shoulder if there is no joint effusion. It is also helpful for demonstrating breakdown of soft tissue structures that normally prevent communication between joint compartments such as the rotator cuff, triangular fibrocartilage of the wrist, and ligaments in the various joints. Newer techniques that enable image acquisition in near real time as well as the development of MR-compatible needles now permit a variety of percutaneous procedures to be performed directly under MR guidance.

  SCINTIGRAPHY

Scintigraphy by its nature represents physiologic imaging because it derives from labeling physiologically occurring substances with a gamma-emitting radionuclide and uses detectors in the form of gamma cameras arranged in a planar or circumferential configuration to determine the distribution of radionuclide within the tissue. Scintigraphy can provide a global assessment of abnormal tracer uptake or can be performed using a targeted approach (Fig. 243-8). Images often provide high tissue contrast but are of relatively poor spatial resolution. Commonly employed agents vary from tagged red blood cells to assess blood flow; agents that reflect bone metabolism (technetium99m methylene diphosphate [Tc-MDP]); agents that reflect glucose metabolism (18-fluorine deoxyglucose [18-FDG]), in the case of PET; and agents that concentrate at sites of inflammation, such as autologous white blood cells labeled with 111In (indium) and 67Ga-citrate (gallium). Clinical applications

include detection of a variety of malignancies, osteomyelitis, vascular graft infection, multifocal infectious disease, inflammatory diseases such as RA, vasculitis, inflammatory bowel disease, sarcoidosis, fever of unknown origin, and infection of joint prostheses. Traditional nuclear medicine involves use of single gamma photon emissions as a product of nuclear decay. The information can be displayed using planar imaging through a single (or multiple) pinhole camera or displayed tomographically in a manner similar to CT (single-photon emission CT [SPECT]). Bone scintigraphy employs Tc-MDP as the radioactive tracer. The isotope goes to areas of high bone turnover and vascular flow as well as areas of calcium or bone deposition. Three-phase bone scans are obtained at different intervals after injection, reflecting the early vascular phase, the intermediate blood pool phase, and the late phase. Each phase allows for further characterization of the disease process. Abnormal tracer uptake is seen in areas of inflammation, infection, neoplasm, osteonecrosis, and fracture. The scan is most useful to identify the location of lesions within the skeleton but is nonspecific. PET scans use the appearance of two simultaneously produced 511-KEV gamma rays following annihilation of a positron and electron pair to localize the distribution of radionuclide. The near-simultaneous detection of the photons (coincidence counting) provides an estimate of source tracer concentration. Newer PET scanners are often used in combination with either CT or MRI in order to achieve improved spatial registration, allow accurate estimates of soft tissue attenuation, provide high-quality anatomic images, and quantify metabolic activity.8 Combined PET-CT or PET-MRI provides high-resolution images of abnormal metabolic activity and may ultimately provide the most definitive maps of inflammatory activity in patients with rheumatic disease. Early results to date have been promising and are expected to provide sensitive evaluation of the response to DMARDs in patients with inflammatory arthritis.

A

B

C

D

FIGURE 243-8.  Rectilinear bone scan in a patient with back pain. A, B, Anterior and (C, D) posterior delayed images of the axial and appendicular skeleton demonstrate increased tracer uptake in the region of the sacral ala, left ankle, and right midfoot (arrows). Follow-up radiographs confirmed the presence of bilateral sacral ala fractures. Note that the central pooling of tracer in the expected location of the urinary bladder is normal. Bone scans provide a sensitive but nonspecific method to evaluate the appendicular and axial skeletal. Increased uptake in the feet in this patient was attributed to degenerative change. L = left; R = right.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 243  Imaging Studies in the Rheumatic Diseases  

GENERAL REFERENCES 1. Jamar F, Versari A, Galli F, et al. Molecular imaging of inflammatory arthritis and related disorders. Semin Nucl Med. 2018;48:277-290. 2. Barile A, Arrigoni F, Bruno F, et al. Computed tomography and MR imaging in rheumatoid arthritis. Radiol Clin North Am. 2017;55:997-1007. 3. Garner HW, Wessell DE. Current status of ultrasound and dual-energy computed tomography in the evaluation of gout. Rheumatol Int. 2018;38:1339-1344. 4. Forney MC, Delzell PB. Musculoskeletal ultrasonography basics. Cleve Clin J Med. 2018;85:283-300.

1685.e1

5. Möller I, Janta I, Backhaus M, et al. The 2017 EULAR standardised procedures for ultrasound imaging in rheumatology. Ann Rheum Dis. 2017;76:1974-1979. 6. Rubin DA. MRI and ultrasound of the hands and wrists in rheumatoid arthritis. I. Imaging findings. Skeletal Radiol. 2019. [Epub ahead of print.] 7. Hayashi D, Roemer FW, Jarraya M, et al. Imaging in osteoarthritis. Radiol Clin North Am. 2017;55:1085-1102. 8. Sollini M, Berchiolli R, Kirienko M, et al. PET/MRI in infection and inflammation. Semin Nucl Med. 2018;48:225-241.

1685.e2

CHAPTER 243  Imaging Studies in the Rheumatic Diseases  

REVIEW QUESTIONS 1. With regard to imaging of gout, the following statement(s) is (are) true: A . The presence of periarticular articular erosions and soft tissue nodules is pathognomonic. B. Osteopenia is a hallmark of the disease. C. Dual-energy computed tomography can be of value to assess disease extent. D. Chondrocalcinosis has a characteristic appearance on ultrasound in these patients. E. B and D are correct. Answer: E  Dual-energy CT is well suited to assess the distribution of tophaceous deposits in patients with gout due to differences of the absorption characteristics of sodium urate crystals in comparison with other types of deposition diseases. On ultrasound, calcification along the superficial margin of the cartilage is characteristic of gout, giving rise to a double-line sign. 2. A patient presents with radial-sided wrist pain that extends proximally into the forearm. What imaging test(s) would assess possible etiologies? A . Radiographs alone should be adequate. B. MRI C. Ultrasound D. A and B are correct. E. B and C are correct Answer: E  The clinical history suggests de Quervain tenosynovitis. Radiographs would allow assessment of the osseous structures but not the adjacent tendons. Although MRI would provide the most complete examination of the radial-sided structures, ultrasound is well suited to evaluate the soft tissues, including the first dorsal compartment tendons. One also could perform ultrasound-guided therapy at the time of diagnosis. 3. A patient is suspected of having sacroiliitis. What imaging study(ies) would be appropriate to evaluate the patient? A . Radiographs of the SI joints B. Computed tomography C. Nuclear scintigraphy (bone scan) D. MRI E. All the above Answer: E  Dedicated radiographic views of the sacroiliac joints should be the first study ordered to assess for possible erosions and may be sufficient to establish the diagnosis. Computed tomography would provide a more sensitive evaluation for subtle erosions as well as the adjacent soft tissues, particularly with the addition of intravenous contrast. Computed tomography provides an ideal method to perform guided therapy or aspiration. Bone scan would provide a sensitive evaluation of the SI joints, but would be abnormal for trauma, inflammatory, or degenerative etiologies. Bone scan would allow a global assessment of the axial and appendicular skeleton to determine whether other sites are potentially affected. MRI allows the best assessment of the bone, adjacent marrow space, and soft tissues.

4. A patient has shoulder pain and gives a history of prior dislocation. There is an equivocal abnormality on the humeral head on radiographic evaluation. Which additional study should be considered? A . Ultrasound to rule out rotator cuff tear B. Noncontrast computed tomography C. Additional specialized radiographs to evaluate the scapula D. Direct MR arthrography E. Nuclear scintigraphy Answer: D  Although ultrasound could evaluate the rotator cuff, it does not provide adequate assessment of the capsular labral complex. Cross-sectional imaging with intra-articular contrast would best accomplish this. MR arthrography would be optimal. Direct arthrography has been the method of choice in assessing the glenoid labrum, surrounding ligaments, and capsule. CT arthrography is of value, particularly in the postoperative shoulder, and provides indirect imaging of internal structures by coating them with contrast material. A noncontrast CT would be very limited in assessing the labroligamentous complex, even in the presence of a joint effusion due to poor soft tissue contrast. 5. A patient with early rheumatoid arthritis is being considered for placement on a DMARD. All of the imaging studies below could assess the level of disease activity and response to therapy except: A . 18FDG scan B. Radiographs of the hands and wrists C. Gray scale ultrasound with power Doppler D. MRI with gadolinium E. Parametric MRI of distribution volumes of contrast in the extravascular space Answer: B  Radiographic findings in rheumatoid arthritis usually occur when there has already been irreversible joint damage. Ideally, therapy would be instituted on a radiographically negative patient. The remaining examinations can provide sensitive evaluation of disease activity before the development of either bone or cartilage erosion.

CHAPTER 244  Inherited Diseases of Connective Tissue  

1685

storage disorders, termed mucolipidoses, are caused by a genetic defect in posttranslational modification of lysosomal enzymes and share features with the mucopolysaccharidoses.  

EPIDEMIOLOGY

All mucopolysaccharidosis disorders are rare, each with an incidence of one or fewer cases per 100,000 births, and are without ethnic predilection.

244  INHERITED DISEASES OF CONNECTIVE TISSUE REED E. PYERITZ

  MUCOPOLYSACCHARIDOSES  

DEFINITION

Proteoglycans are ubiquitous components of the extracellular matrix and the surfaces of cells, and they are among the largest and most complex of human molecules. Proteoglycans consist of a protein core to which are covalently bound glycosaminoglycans (GAGs; formerly called mucopolysaccharides) of several types: dermatan sulfate, heparan sulfate, keratan sulfate, and chondroitin sulfate. During normal degradation, these four polymeric molecules are cleaved from their protein core in lysosomes; then they, plus hyaluronan (a GAG lacking a protein core), are catabolized further in lysosomes in a stepwise fashion by more than a dozen enzymes. Genetic defects in any one of these enzymes lead to the accumulation of GAG metabolites in lysosomes, with profound disruption of cellular physiology. The phenotypes resulting from deficiencies of these catabolic enzymes are termed mucopolysaccharidoses and are classified into seven types (Table 244-1).1 Several additional



PATHOBIOLOGY

With the exception of mucopolysaccharidosis II (Hunter syndrome), which is X-linked, each of these disorders is autosomal recessive. All mucopolysaccharidoses are caused by deficiency of a single lysosomal enzyme responsible for a specific step in GAG metabolism. Catabolism of GAG proceeds normally until the step requiring the defective enzyme, when further normal metabolism halts. Although a minor degree of nonspecific breakdown occurs, resulting in urinary excretion of cleaved GAG that can be useful diagnostically, the accumulation of GAG within lysosomes of cells of mesenchymal origin, endothelium, and, in most cases, neurons causes widespread, progressive cellular dysfunction and clinical effects. Lysosomal enzymes are targeted to lysosomes by post-translational addition of mannose 6-phosphate. Deficiency of the phosphotransferase that catalyzes the first step in this reaction results in an inability to catabolize any GAG molecules. The catabolic enzymes, which normally would be transported into lysosomes, instead are secreted from the cell and are found in unusually high concentrations in plasma, providing one diagnostic test for mucolipidoses.

Pathology

All pathologic manifestations of mucopolysaccharidosis and mucolipidosis disorders worsen with age, and some are present from early developmental stages. Gross anatomic hallmarks are hepatosplenomegaly, marked skeletal alterations (termed dysostosis multiplex)2 that result in short stature and thoracic cage deformity, thickening and narrowing of airways and arteries, and coarsening of facial features. Although mental retardation is a prominent feature of some of these conditions, the brain may show only ventriculomegaly

CHAPTER 244  Inherited Diseases of Connective Tissue  

ABSTRACT

The inherited diseases of connective tissue are diverse in phenotypic characteristics, genetic and molecular causes, and effects on health. Among more than 200 conditions, some of the more common are the mucopolysaccharidoses, Marfan syndrome, the Ehlers-Danlos syndromes, the osteogenesis imperfecta syndromes, and pseudoxanthoma elasticum. In their severe forms, all can shorten life expectancy. Fortunately, therapies such as medication, enzyme replacement, and gene therapy are becoming more frequent and effective.

KEYWORDS

mucopolysaccharidoses Marfan syndrome Ehlers-Danlos syndromes osteogenesis imperfecta

1685.e3

1686

CHAPTER 244  Inherited Diseases of Connective Tissue  

TABLE 244-1 MUCOPOLYSACCHARIDOSES AND MUCOLIPIDOSES TYPE

EPONYM OR COMMON NAME

MPS IH

Hurler syndrome

DM and short stature; MR; corneal clouding; HS; heart disease; death in childhood

AR

252800 α-l-iduronidase

MPS IS

Scheie syndrome

Coarse facies; stiff joints, corneal clouding; aortic valve disease; normal intelligence and life span

AR

252800 α-l-iduronidase

MPS II

Hunter syndrome

Severe form: coarse facies, DM and short stature, HS; MR; no corneal clouding; death by late adolescence. Mild form: coarse facies, short stature; normal intelligence; survival to adulthood

XL

309900 Iduronate sulfatase

CLINICAL FEATURES

INHERITANCE OMIM*

ENZYMATIC DEFECT

MPS IIIA Sanfilippo A

Severe MR and hyperactivity; mild somatic changes

AR

252900 Heparan N-sulfatase

MPS IIIB

Sanfilippo B

Same as MPS IIIA

AR

252920 α-N-acetylglucosaminidase

MPS IIIC Sanfilippo C

Same as MPS IIIA

AR

252930 Acetyl-coenzyme A: α-glucosaminide acetyltransferase

MPS IIID Sanfilippo D

Same as MPS IIIA

AR

252940 N-acetylglucosamine 6-sulfatase

MPS IVA

Morquio A

Short stature and distinct skeletal dysplasia with odontoid hypoplasia and myelopathy; corneal clouding; normal intelligence; valvular heart disease

AR

253000 Galactose 6-sulfatase

MPS IVB

Morquio B

Same as MPS IVA

AR

253010 β-Galactosidase

MPS VI

Maroteaux-Lamy

DM and short stature; corneal clouding; normal intelligence; aortic stenosis; leukocyte inclusions; hydrocephalus in severe form

AR

253200 N-acetylgalactosamine

MPS VII

Sly syndrome

DM; HS; widely variable, including MR

AR

253220 β-Glucuronidase

MPS IX



Short stature; periarticular soft tissue masses

AR

601492 Hyaluronidase

ML II

I-cell disease

Similar to but more severe than MPS IH but with cellular inclusions; no mucopolysacchariduria

AR

252500 UDP-N-acetylglucosamine: lysosomal enzyme N-acetylglucosaminyl-1-phosphotransferase

ML III

Pseudo-Hurler polydystrophy

Short stature and mild DM; stiff joints, mild MR; survival to adulthood

AR

252500 Same as ML II arthropathy, coarse facies; variable but milder

*Entries in Online Mendelian Inheritance in Man, OMIM. McKusick-Nathans Institute of Genetic Medicine. Baltimore: Johns Hopkins University. http://omim.org. AR = autosomal recessive; DM = dysostosis multiplex; HS = hepatosplenomegaly; MR= mental retardation; UDP = uridine diphosphate; XL = X-linked.

secondary to communicating hydrocephalus. On microscopy, mesenchymal cells show a cytoplasm full of apparently empty vacuoles; these are lysosomes from which GAG has been removed by fixation. Cells cultured from patients show greatly enlarged lysosomes filled with granular material. In the severe form of mucolipidosis, dense inclusions are present, which gave rise to the common name, I-cell disease.  

CLINICAL MANIFESTATIONS

Each of the disorders in Table 244-1 shows a wide spectrum of clinical severity. This wide spectrum has led to a classification that gives the impression of separate disorders within some of the mucopolysaccharidosis and mucolipidosis types, but these represent the apparent ends of the continuum. Some of the disorders without treatment result in death by adolescence (Hurler syndrome, severe Hunter syndrome, mucolipidosis II), but others are commonly compatible with survival to adulthood. The latter group of disorders is emphasized here. The milder end of the mucopolysaccharidosis I spectrum, Scheie syndrome, may not be diagnosed until adulthood; patients present with stiffened joints, corneal clouding and glaucoma, carpal tunnel syndrome, and aortic valvular disease. Stature and intelligence are not affected. The main health risks are cardiac valvular involvement, thickening of meninges that can produce a myelopathy, and thickening of the upper airways that can produce obstructive symptoms and sleep apnea. The milder form of mucopolysaccharidosis II, Hunter syndrome, is distinctive because it is X-linked (affecting males almost exclusively), and the cornea shows little overt clouding. Cervical myelopathy, obstructive airway disease, and cor pulmonale are important concerns. A combined conductive and neurosensory hearing loss is common. Neither mucopolysaccharidosis IV (Morquio syndrome) nor mucopolysaccharidosis VI (Maroteaux-Lamy syndrome) affects intelligence. Both syndromes often are associated with severe skeletal changes, which are distinct radiographically but produce similar problems of kyphoscoliosis, pectus carinatum, restrictive lung disease, severe short stature, and joint degeneration. Cervical myelopathy resulting from a thickened dura is common to both disorders and is accentuated by odontoid hypoplasia in mucopolysaccharidosis IV.

Thickening of the aortic and mitral valves may produce severe dysfunction necessitating their replacement. General anesthesia is especially hazardous because of the narrow upper and middle airways and cervical instability. Patients with mucolipidosis III (pseudo-Hurler polydystrophy) resemble patients with mucopolysaccharidosis VI but often have mild to moderate mental retardation. Aortic regurgitation is common.  

DIAGNOSIS

Differential Diagnosis

Diagnosis of these conditions is difficult in young children, before most of the clinical features have progressed, but should be considered in any person with hepatosplenomegaly and coarsening of the facial features. Evaluation requires a pedigree analysis, ophthalmologic examination, skeletal radiographic survey, echocardiography, and analysis of the urine for excretion of GAGs. Often the specific mucopolysaccharidosis is evident from radiographs, the presence or absence of corneal clouding, and the pattern of mucopolysacchariduria. Enzymatic analysis of leukocytes confirms the diagnosis. Patients with mucolipidoses do not show mucopolysacchariduria but have marked elevation of all the GAG catabolic lysosomal enzymes in plasma.

TREATMENT  Ventriculoperitoneal shunting is necessary if intracranial pressure is elevated. Close attention to hearing and visual problems is essential throughout life. Many adults with mucopolysaccharidosis or mucolipidosis require surgery for carpal tunnel syndrome. Cardiovascular surgery for valvular or coronary disease may be necessary. All use of anesthesia is high risk because of the narrow airways and, in the case of mucopolysaccharidosis IV, atlantoaxial instability. For patients who remain ambulatory, selective joint replacement can be beneficial. Because of the morbidity associated with thoracic cage deformity, consideration should be given to stabilizing the spinal deformity before it becomes severe.

CHAPTER 244  Inherited Diseases of Connective Tissue  

Replacement of the deficient enzyme by intravenous infusion is being studied for most of the mucopolysaccharidosis disorders.3 Laronidase (Aldurazyme) has been approved in the United States for treatment of mucopolysaccharidosis I. An infusion every 2 weeks for 1 year in adolescent and adult patients resulted in substantial reduction in hepatosplenomegaly and modest improvement in pulmonary function, sleep apnea, and joint mobility. Whether early institution of therapy in young children modulates mental retardation in the Hurler variant of mucopolysaccharidosis I is uncertain. Galsulfase (Naglazyme) has been approved for the treatment of mucopolysaccharidosis VI, in which somatic rather than neurologic problems predominate. Bone marrow transplantation has been attempted in many of the mucopolysaccharidosis disorders,4 with mixed success. The earlier transplantation occurs, the better the outcome in terms of somatic problems, but prevention of mental retardation has not occurred. Current recommendations based on consensus in Europe calls for hematopoietic stem cell transplantation for patients with Hurler syndrome before the age of 2.5 years. Enzyme replacement should be started in all patients when diagnosed. Gene therapy has been successful in a mouse model of Hunter syndrome, thereby providing proof of concept for the treatment of Hunter syndrome patients with cognitive impairment.5

  MARFAN SYNDROME  

DEFINITION

Marfan syndrome is an autosomal dominant, pleiotropic disorder caused by defects in the principal component of the extracellular microfibril, the large glycoprotein fibrillin-1.6 The disease manifestations occur in multiple systems, especially the eye, skeleton, heart, aorta, lung, and integument. Notable features include dislocation of the ocular lens, tall stature with particularly long limbs and digits, deformity of the thoracic cage by pectus carinatum or excavatum with abnormal curvature of the spine, mitral and tricuspid valve prolapse, dilation of the sinuses of Valsalva and predisposition to aortic dissection, spontaneous pneumothorax, abnormal skin stretch marks, hernias, and dural ectasia. If untreated, patients often die before 30 or 40 years of age from aortic dissection or congestive heart failure.  

EPIDEMIOLOGY

Marfan syndrome is a common mendelian disorder, with an estimated incidence of about 1 per 5000 births. Marfan syndrome is found throughout the world, without ethnic or geographic predilection.  

PATHOBIOLOGY

Pathogenesis

Mutations in FBN1, which maps to human chromosome 15q21.1 and encodes fibrillin-1, cause Marfan syndrome and related connective tissue disorders. More than 1500 distinct mutations have been found, and few occur in more than one family. Patients are heterozygous for mutations in FBN1, leading to autosomal dominant inheritance. Extracellular microfibrils are polymers of many fibrillin-1 molecules and are ubiquitous in the extracellular matrix of most tissues. Latent transforming growth factor-β (TGF-β) binding protein, which keeps the cytokine inactive, bears striking homology to regions of fibrillin. Abnormalities of either the quality or the quantity of microfibrils disrupt normal signaling by TGF-β, especially during embryonic development and postnatal growth. Studies in mice engineered to harbor human mutations in FBN1 showed that excessive TGF-β signaling causes abnormal lung septation (the precursor to pneumothorax), bone overgrowth, mitral valve prolapse, muscular hypoplasia, and aortic dilation. This fundamental shift in understanding of the pathogenesis of Marfan syndrome has suggested novel therapies, such as with small molecules that affect the activity of TGF-β or its downstream signaling. The features of Marfan syndrome are highly variable, even among relatives who share the same mutation in FBN1. This variability persists after accounting for the effects of age. Men tend to be affected more severely, for unclear reasons.

Pathology

The features of Marfan syndrome are age dependent. Some severely affected infants have flagrant features and often die of mitral regurgitation and heart failure despite aggressive management. At the other end of the clinical spectrum, Marfan syndrome merges with several related disorders, and patients may not come to medical attention, let alone receive a definitive diagnosis, until adulthood.

1687

None of the gross or microscopic pathologic changes is specific for Marfan syndrome. The medial degeneration of the aortic wall, characterized by disarray and fragmentation of the elastic fibers and increased proteoglycan (often inappropriately termed cystic medial necrosis), also can be seen in other disorders and in older people with hypertension. Aortic dissection (Chapter 69) usually begins just superior to the aortic valve (type A) and often progresses to the bifurcation. Death usually results from retrograde dissection and hemopericardium. About 10% of dissections begin in the descending thoracic aorta (type B).  

CLINICAL MANIFESTATIONS

The lens tends to be displaced superiorly, and usually the zonules remain intact. The retina is at increased risk for detachment, especially in patients who are highly myopic. Tubular bones overgrow, accounting for the disproportionate tall stature (dolichostenomelia), long digits (arachnodactyly), and sternal deformity. Ligaments may be lax, causing scoliosis and joint hypermobility. Alternatively, congenital contractures are common, especially of the elbows. The palate typically is highly arched, and the dentition can be crowded and maloccluded. Mitral valve prolapse occurs in about 80% of cases, and the valve leaflets become progressively thickened (myxomatous on histopathology). The mitral annulus may dilate and calcify. Aortic root dilation begins in the sinuses of Valsalva and progresses with age, albeit at highly variable rates (Chapters 61 and 69).7,8 Most males with Marfan syndrome have an aortic root dimension above the upper limit of normal for their body surface area by adolescence. Some females show a slower progression and may have a root diameter near the upper limit of normal well into adulthood. The dilation usually does not involve the distal ascending aorta. Spontaneous pneumothorax, resulting from rupture of apical blebs, occurs in about 5% of patients. Stretch marks (striae atrophicae) occur over areas of flexural stress, such as the shoulders, breasts, and lower back. The neural canal in the lumbosacral region is enlarged in most people with Marfan syndrome; this may be visible on plain radiographs, especially if the neuroforamina are widened. Imaging by computed tomography or magnetic resonance imaging is diagnostic and should be used in patients with back pain and radicular symptoms. Dural ectasia progresses with age; large anterior meningoceles in the pelvis are a severe manifestation. Simple cysts in the liver and kidneys are common, increase with age, and seldom cause clinical problems. Sleep apnea is of increased frequency in adults.9  

DIAGNOSIS

Differential Diagnosis

The conditions that overlap clinically and genetically with Marfan syndrome include familial aortic aneurysm, familial ectopia lentis, mitral valve prolapse, mild aortic dilation, striae, skeleton (MASS) phenotype (which includes many families with mitral valve prolapse syndrome), and Loeys-Dietz syndrome. Most of these conditions are diagnosed clinically, so differentiating among them is arbitrary. A careful family history is essential to this process. Molecular genetic testing has a limited role. However, if the mutation in FBN1 is known in a family, analysis of DNA can be used effectively for presymptomatic or prenatal diagnosis. Loeys-Dietz syndrome, which is associated with generalized arterial tortuosity and susceptibility to dissection, is caused by mutation in either of two receptors for TGF-β, TGFBR1 and TGFBR2, and molecular analysis is clinically available. Most mutations in TGFBR1 and TGFBR2 do not cause Loeys-Dietz syndrome but do predispose to familial thoracic aneurysm and dissection. A question of Marfan syndrome arises most commonly in tall, lanky adolescents who have several minor skeletal features, nearsightedness, and athletic desires. A detailed ophthalmologic examination with full pupillary dilation and a transthoracic echocardiogram are essential components in the evaluation. If these test results are negative and no one in the family has a history of Marfan syndrome or aortic dissection, the patient probably can be reassured.10

TREATMENT  All patients should be seen at least annually by a physician who manages the overall care. Most patients require annual ophthalmologic and cardiologic consultation and orthopedic consultation as required by specific problems. Lens subluxation often requires surgical correction.11 A number of studies, but only one randomized clinical trial, support the prophylactic use of β-adrenergic blockade from an early age to slow the rate of aortic root dilation and protect

1688

CHAPTER 244  Inherited Diseases of Connective Tissue  

against aortic dissection. Based on studies of the Marfan mouse, therapies that interfere with excess signaling through pathways mediated by TGF-β are being studied in human clinical trials. One large European trial suggested a benefit of the angiotensin receptor blocker losartan on aortic root dilation rate, A1  but three trials found no benefit of losartan compared with atenolol. A2-A4  Prophylactic surgical repair of the aortic root has had the greatest beneficial impact. The composite graft, involving a prosthetic valve in a Dacron tube and implantation of the coronary ostia into the graft, was the first approach to produce markedly improved survival in these patients. More recently, replacement of the aneurysm and preservation of the native aortic valve have shown promise and should be considered first.12 For adults, aortic root surgery should be strongly considered when the maximal aortic diameter reaches 45 mm, and a family history of aortic dissection should prompt earlier repair (Chapter 61).



PROGNOSIS

Life expectancy for people with Marfan syndrome has improved markedly, to the point that many patients can expect survival to advanced years. Some data suggest that aortic dissection and cardiovascular death are more likely in patients with FBN1 haploinsufficiency with expression of the nonmutated allele than in patients with dominant-negative mutations in which abnormal fibrillin 1 coded by the mutated allele interacts with normal fibrillin 1 coded by the normal allele.13

  EHLERS-DANLOS SYNDROMES  

DEFINITION

The Ehlers-Danlos syndromes are clinically variable and genetically heterogeneous. Diagnoses still are based largely on the bedside examination. The unifying themes among these disorders are fragility of tissues, joint hypermobility, and skin hyperextensibility.14  

EPIDEMIOLOGY

No accurate data exist, but an incidence of about 1 in 5000 births is a reasonable estimate of how many individuals qualify for one of the Ehlers-Danlos syndrome diagnoses. Each type represents something of a clinical spectrum, with the mild end merging with what might be considered normal variation. Just as the diagnostic criteria are arbitrary, so would be any determination of prevalence based on phenotypic criteria. The extent to which normal variation in joint hypermobility, skin elasticity, and tissue fragility represents genetic variation at loci that encode collagen or other extracellular matrix genes requires considerable research.



PATHOBIOLOGY

Pathogenesis

Defects in collagen and other proteins in the extracellular matrix of various tissues underlie all forms of Ehlers-Danlos syndrome that have been elucidated so far. The specific mutations occur in a variety of genes, with the effect of altering the structure, synthesis, post-translational modifications, or stability of the collagens involved. The known molecular defects are listed in Table 244-2.

Pathology

Few findings in the routine pathologic evaluation distinguish among the various types of Ehlers-Danlos syndrome or even distinguish individual types from normal. Thickness of the dermis is decreased in some forms, especially the vascular type, and the walls of arteries are reduced in thickness in this type. By electron microscopy, the classic, hypermobile, and kyphoscoliotic types have abnormal collagen fibers, especially when viewed in cross section (variable and often increased fiber diameter with an irregular outline). In the vascular type, some patients have dilated endoplasmic reticulum consistent with aberrant secretion of type III collagen molecules.  

CLINICAL MANIFESTATIONS

The major and minor features of each Ehlers-Danlos syndrome are detailed in Table 244-2. Infants with classic Ehlers-Danlos syndrome often are born prematurely by 4 to 8 weeks because of rupture of fetal membranes. Diagnosis of the vascular and kyphoscoliotic types is important because of their cardiovascular features. The vascular type, previously termed Ehlers-Danlos syndrome IV, is characterized by spontaneous rupture of large arteries and hollow organs, especially the colon and uterus, and pneumothorax. Because these events carry considerable morbidity, life expectancy is reduced, on average, by more than half. During pregnancy, women with this form of Ehlers-Danlos syndrome are especially vulnerable to rupture of major arteries and the uterus. In the kyphoscoliotic type, aortic root dilation and aortic regurgitation can develop. Patients with most forms of Ehlers-Danlos syndrome are prone to develop mitral valve prolapse, and progression to mitral regurgitation (Chapter 66) occurs more often than in the common form of mitral valve prolapse.  

DIAGNOSIS

Differential Diagnosis

By careful adherence to the clinical features shown in Table 244-2 and judicious use of laboratory tests, the various defined types of Ehlers-Danlos syndrome can be differentiated. Many other specific syndromes need to be excluded.

TABLE 244-2 EHLERS-DANLOS SYNDROMES TYPE

FORMER NAME

CLINICAL FEATURES*

INHERITANCE

OMIM†

MOLECULAR DEFECT

Classic

EDS I and II

Joint hypermobility; skin hyperextensibility; atrophic scars; smooth, velvety skin; subcutaneous spheroids

AD

130000 130010

Structure of type V collagen caused by mutations in COL5A1 or COL5A2

Hypermobility

EDS III

Joint hypermobility; some skin hyperextensibility, with or without a smooth, velvety texture

AD AR

130020 225320

? Tenascin-X (TNX)

Vascular

EDS IV

Thin skin; easy bruising; pinched nose; acrogeria; rupture of large- and medium-caliber arteries, uterus, and large bowel

AD

130050 (225350) (225360)

Deficient type III collagen (COL3A1)

Kyphoscoliotic

EDS VI

Joint hypermobility; congenital, progressive rupture; scoliosis; scleral fragility with globe rupture; tissue fragility, aortic dilation, MVP

AR

225400

Deficiency of lysyl hydroxylase

Arthrochalasis

EDS VII A

Joint hypermobility, severe, with subluxations, congenital hip dislocation; and skin hyperextensibility; tissue fragility

AD

130060

No cleavage of amino terminus of type I procollagen caused by mutations in COL1A1 or COL1A2

Dermatosparaxis

EDS VII C

Severe skin fragility; decreased skin elasticity, easy bruising; hernias; premature rupture of fetal membranes

AR

225410

No cleavage of amino terminus of type I procollagen caused by deficiency of peptidase

Unclassified types

EDS V EDS VIII EDS X EDS XI EDS IX EDS, progeroid form

Classic features Classic features and periodontal disease Mild classic features, MVP Joint instability Classic features; occipital horns Classic features and premature aging

XL AD ? AD XL AR

305200 130080 225310 147900 309400 130700

? ? ? ? Allelic to Menkes syndrome Deficiency of galactosyltransferase I

*Listed in order of diagnostic importance. † Entries in Online Mendelian Inheritance in Man, OMIM. McKusick-Nathans Institute of Genetic Medicine. Baltimore: Johns Hopkins University. http://omim.org. AD = autosomal dominant; AR = autosomal recessive; EDS = Ehlers-Danlos syndrome; MVP = mitral valve prolapse; XL = X-linked.

CHAPTER 244  Inherited Diseases of Connective Tissue  

The kyphoscoliotic type of Ehlers-Danlos syndrome in infants shares some features with severe Marfan syndrome. Patients with Larsen syndrome may resemble patients with the arthrochalasis type of Ehlers-Danlos syndrome. The skin redundancy and loss of elasticity of the dermatosparaxis type of Ehlers-Danlos syndrome is reminiscent of autosomal dominant cutis laxa, which is not associated with easy bruising or tissue fragility. The most difficult decision is whether any diagnosis of Ehlers-Danlos syndrome is warranted.15 Patients who have only joint hypermobility without skin changes should not be labeled with Ehlers-Danlos syndrome; a diagnosis of familial joint hypermobility might be more appropriate. Familial joint instability involves a predisposition to dislocations of major joints that is rare in most types of Ehlers-Danlos syndrome except for arthrochalasis.

  OSTEOGENESIS IMPERFECTA SYNDROMES  

Management of most skin and joint problems should be conservative and preventive. Sutures need to be placed with careful attention to approximating the margins and avoiding tension; removable sutures should be left in place for twice the usual time. Most instances of joint hypermobility and pain in Ehlers-Danlos syndrome do not require surgical treatment. Benefit often is derived from physical therapy designed to strengthen the muscles that provide support for the loose ligaments. All patients should receive genetic counseling about the mode of inheritance and their risk for having children affected with Ehlers-Danlos syndrome. The possibility of prenatal diagnosis exists for all of the Ehlers-Danlos syndrome types with defined molecular or biochemical defects. The vascular type of Ehlers-Danlos syndrome requires particular surgical care; the ruptured arteries are difficult to repair because of the pronounced vascular fragility. Experienced vascular surgeons are having some success with prophylactic repair of vessels deemed to be at risk for dissection or rupture. One clinical trial suggested improved outcomes with prophylactic β-adrenergic blockade. A5  Rupture of the bowel is a surgical emergency. Because the risk for uterine and vascular rupture is especially high during pregnancy in women with the vascular form, affected women should be advised that there is a substantial risk for death related to pregnancy and delivery. Patients should be advised to avoid contact sports and to treat blood pressure elevations aggressively. Arteriography and arterial lines should be avoided if possible. Biochemical and genetic screening holds the potential for reassuring relatives at risk that they do not have a defect in type III collagen. The kyphoscoliotic type of Ehlers-Danlos syndrome may improve with large doses of vitamin C (1 to 4 g/day) because ascorbate is a cofactor for the enzyme that is deficient. No other metabolic or genetic therapy is effective in other forms of Ehlers-Danlos syndrome.

DEFINITION

The heterogeneous group of disorders called osteogenesis imperfecta includes, at one end of the severity spectrum, a type that is lethal prenatally or in the neonatal period and, at the other, such mild features that make distinguishing affected individuals from the general population is difficult.16,17 The unifying feature is hereditary osteopenia (insufficient bone), with primary defects in the protein matrix in bone and other tissues. The clinical syndromes all involve osteoporosis with liability to fracture (Chapter 230).  

TREATMENT 

1689

EPIDEMIOLOGY

No careful epidemiologic study has been performed, and the milder forms of type I osteogenesis imperfecta merge with the phenotypes of familial osteoporosis, fracture susceptibility, and joint hypermobility found in the general population. A crude estimate of the overall prevalence of osteogenesis imperfecta is 1 to 2 per 20,000 births. The neonatal lethal form (type II), which is almost always caused by a new mutation in a parental gamete, has an incidence of about 1 in 50,000 births.  

PATHOBIOLOGY

Pathogenesis

Most patients in whom mutations have been found usually have defects in the two genes that encode the procollagen chains of type I collagen, COL1A1 and COL1A2. Type I collagen is composed of two α1(I) and one α2(I) procollagen chains; the mature fiber requires considerable post-translational modification, which occurs appropriately only if the three procollagen chains have intertwined to form a triple helix that is perfect and completed at the right speed. A mutation that affects formation of the triple helix, such as substitution of one of the mandatory glycine residues that occurs at every third position, also has adverse effects on the modifications that render the molecule capable of forming effective mature fibers. As a result, a single nucleotide change resulting in a missense mutation can have profound effects on the extracellular matrix and produce a severe condition. Alternatively, and at first glance paradoxically, a mutation that eliminates an entire allele, or at least production of any product capable of intertwining with normal procollagen chains, has a much milder effect on the extracellular matrix and on the severity of osteogenesis imperfecta. Examples of the most common classes of mutations are shown in Table 244-3. Hundreds of mutations have been described. Patients with mutations in COL1A1 or COL1A2 are heterozygous, and thus the most common forms of osteogenesis imperfecta are inherited as autosomal

TABLE 244-3 OSTEOGENESIS IMPERFECTA TYPE

INHERITANCE

OMIM*

I

Fractures variable in number; little deformity; stature normal or nearly so; blue sclerae; hearing loss common but not always present; DI uncommon

CLINICAL FEATURES

AD

166200

Typically, one nonfunctional COL1A1 allele

BASIC DEFECTS

II

Lethal in utero or shortly after birth; many fractures at birth typically involving ribs (may appear “beaded”) and other long bones; little calvaria; pulmonary hypertension

AD

166210

AR

259400

COL1A1 or COL1A2: substitution of glycyl residues; occasionally deletions of a portion of the triple-helical domain Deletion in COL1A2 plus a nonfunctional allele

III

Fractures common, but long bones progressively deform starting in utero; stature markedly reduced; sclerae often blue but become lighter with age; DI and hearing loss common

AD AR (rare)

259420 259440

One single amino acid substitution Two mutations in COL1A1 and/or COL1A2 (rarely)

IV

Fractures common; stature usually reduced; bone deformity common but rarely severe; scleral hue normal to grayish; hearing loss variable; DI common

AD

166220 166240

Point mutations in COL1A1 or COL1A2 Exon skipping mutations in COL1A2

V

Similar to type IV without DI or blue sclerae; fractures develop hyperplastic callus; calcification of the interosseous membrane between the radius and ulna

AD

610967

?

VI

Similar to type IV without DI, blue sclerae or wormian bones; excess osteoid present in bone

?

610968

?

VII

Similar to types II or III with fractures at birth, blue sclerae, no DI; presence of rhizomelic limb shortening and coxa vara

AR

610682

Mutations in CRTAP

VIII

Similar to types II or III with fractures at birth

AR

610915

Mutations in LEPRE1

IX

Similar to types II or III with fractures at birth

AR

259440

Mutations in PPIB

*Entries in Online Mendelian Inheritance in Man, OMIM. McKusick-Nathans Institute of Genetic Medicine. Baltimore: Johns Hopkins University. http://omim.org. AD = autosomal dominant; AR = autosomal recessive; DI = dentinogenesis imperfecta.

1690

CHAPTER 244  Inherited Diseases of Connective Tissue  

dominant traits. Several autosomal recessive forms of osteogenesis imperfecta occur because of mutations in genes that encode enzymes that process type I collagen into mature fibrils.

Analysis of the specific enzymes defective in the recessive forms of osteogenesis imperfecta is useful for establishing the diagnosis and enabling reproductive counseling and prenatal diagnosis if desired.

Pathology

Other than the gross pathology associated with the clinical manifestations, the most characteristic pathology is a primary reduction in bone matrix with secondary undermineralization.  

CLINICAL MANIFESTATIONS

The major phenotypic features of osteogenesis imperfecta are shown in Table 244-3. Among the most common forms, the most severe type is type II, followed in decreasing order by types III, IV, and I. In type II, infants either are stillborn or die soon after birth of pulmonary failure secondary to the small thorax, which usually is compromised further by myriad rib fractures. A few infants have survived for at least a few years but require enormous attention to their medical needs. Type III osteogenesis imperfecta may be confused with type II at birth, but survival alone helps make the distinction. Bony deformity is pronounced and not necessarily caused by fractures. Mobility is impaired, and most patients require a wheelchair at an early age. Stature may be severely compromised. Because of progressive vertebral column deformity and rib fractures, restrictive lung disease is a common problem as patients age; many die of pulmonary complications. Basilar impression causing compression of the brain stem and the craniocervical junction can produce central sleep apnea, headache, and upper motor neuron signs. Patients with type IV osteogenesis imperfecta generally have reduced stature, some bony deformity, and abnormal teeth that are opalescent and wear easily (dentinogenesis imperfecta). As in type I osteogenesis imperfecta, the tendency to fracture is highest in childhood and lessens with adolescence. A distinguishing characteristic of type IV osteogenesis imperfecta is a normal scleral hue. Type I osteogenesis imperfecta is probably the most common form and is associated with a bluish or blue-gray scleral hue. People with type I osteogenesis imperfecta who also have dentinogenesis imperfecta tend to have more severe skeletal problems. The risk for fracture diminishes during adulthood but reemerges as a major concern for women after menopause. Hearing impairment in all forms of osteogenesis imperfecta is common and age related, being rare before adolescence. The deficits are of either a mixed or a predominantly conductive form. The recessive forms of osteogenesis imperfecta (types VI to IX) range in severity from type IV to type II and may have distinctive radiologic or histopathologic findings.  

DIAGNOSIS

Differential Diagnosis

The range of diagnostic possibilities in a person with multiple fractures largely depends on age. In infancy, the genetic conditions hypophosphatasia, severe osteochondrodysplasias (e.g., achondrogenesis and forms of spondyloepiphyseal dysplasia), and Menkes syndrome need to be excluded when a diagnosis of type II or type III osteogenesis imperfecta is considered. The radiographic features eventually become entirely diagnostic, but often the neonatologist has to arrive at a definitive answer in short order. Analysis of serum alkaline phosphatase and copper can be helpful. In childhood, the most common situation leading to consideration of a mild form of osteogenesis imperfecta is child abuse. In this situation, the pattern of fracture is usually distinct, and bone mineralization should be normal if the child is the object of nonaccidental or repeated accidental trauma. Abnormal scleral hue, dentinogenesis imperfecta, and wormian bones (microfractures along the cranial sutures) all support the diagnosis of osteogenesis imperfecta. The legal and child-protective systems often request exclusion of osteogenesis imperfecta by analysis of collagen production from cultured skin fibroblasts or analysis of DNA for a mutation. In older children, the disorder idiopathic juvenile osteoporosis should be considered in any patient seen initially with repeated fractures. Many osteochondrodysplasias are associated with short stature, skeletal deformity, and a tendency to fracture. Pyknodysostosis and osteopetrosis are associated with sclerotic bones rather than osteoporotic ones. In adulthood, early-onset osteoporosis may be confused with osteogenesis imperfecta (Chapter 230). Mutations in type I collagen also cause familial osteoporosis, and the skeletal phenotypes merge; patients with true osteogenesis imperfecta may have scleral, hearing, or dental abnormalities and a positive family history.

TREATMENT  Management of the skeletal complications largely depends on orthopedic, physical, and occupational therapy approaches. Risedronate (2.5 or 5 mg daily) increases bone mineral density and reduces both first and recurrent fractures in children with osteogenesis imperfecta. A6  A7  The long-term goals are for the patient to maintain function and independence as an individual. These goals can be advanced in some patients by judicious use of intramedullary rods in the long bones of the legs; if mobility and especially ambulation can be maintained, the demineralization associated with inactivity can be avoided. Unaffected parents of a child with osteogenesis imperfecta and all affected individuals should have genetic counseling. For the parents of a child with type II osteogenesis imperfecta, the possibility of germinal mosaicism (which has been well documented in this condition) should not be overlooked. If one parent has a “new” mutation in one of the type I procollagen genes and multiple gonadal cells carry this mutation, the risk for recurrence in future children is not negligible. If the mutation in the affected child can be defined, the risk for recurrence can be quantified (through molecular analysis of sperm) if the mutation arose in the father. ,

  PSEUDOXANTHOMA ELASTICUM  

DEFINITION

Pseudoxanthoma elasticum is a heritable disorder of connective tissue with pleiotropic manifestations wherever elastic fibers are found but primarily in the skin, eye, and vasculature.18 Life expectancy is reduced, on average, because of a predisposition to myocardial infarction and gastrointestinal hemorrhage.  

EPIDEMIOLOGY

The exact frequency of pseudoxanthoma elasticum is unknown, but it is probably underdiagnosed. Rough approximations suggest a prevalence of 1 in 25,000 to 100,000 births. Males and females are equally affected, although women are more likely to seek medical attention out of concern for the skin changes.  

PATHOBIOLOGY

Pathogenesis

In most families, pseudoxanthoma elasticum occurs as an autosomal recessive trait, which means, given relatively small sibships, that many patients will have no affected relatives. Apparent autosomal dominant inheritance may reflect expression in occasional heterozygotes. The gene for pseudoxanthoma elasticum maps to human chromosome 16 and encodes one of the adenosine triphosphate (ATP)-binding cassette transporters (ABCC6). Because of the prominent histopathologic feature of calcification of elastic tissue, this gene may be important in calcium homeostasis. It is unclear, however, whether calcification is a primary or a secondary phenomenon in pseudoxanthoma elasticum.

Pathology

The hallmark of pseudoxanthoma elasticum, and an important diagnostic clue, is the histopathologic finding of hyperproliferated elastic fibers in the middermis; these fibers become fragmented, clumped, and calcified. An arteriolar sclerosis develops in the media of muscular arteries and arterioles; the lumen may become progressively and concentrically narrowed. Alternatively, microaneurysms can form. Thickening of the endocardium, especially in the atria, develops in some patients. In the eye, Bruch membrane becomes calcified and fragmented.  

CLINICAL MANIFESTATIONS

Because of the pleiotropic nature of pseudoxanthoma elasticum, the diagnosis initially may be suspected by any of a variety of clinicians, especially dermatologists, ophthalmologists, cardiologists, and gastroenterologists. The condition gains its name from the dermatologic feature of yellowish papules that appear at areas of flexural stress, especially the neck, groin, and popliteal and cubital fossae; in periumbilical regions; and on the buccal mucosa. The

appearance of affected skin has been likened to that of a “plucked chicken.” Over time, affected areas coalesce and become thickened. Changes in the eye begin as a generalized, subtle, mottled pattern in the retina (peau d’orange) and progress to the characteristic angioid streaks.19 The latter changes are not specific for pseudoxanthoma elasticum and can be seen in diabetes mellitus, sickle cell disease, and a variety of other conditions. Streaks represent breaks in Bruch membrane, an elastic lamina that lies between the retinal vasculature and the choroid. Spontaneous hemorrhages, especially those involving the macula, lead to progressive visual loss. Involvement of arteries of various calibers produces problems because of occlusion and hemorrhage.20 The lifetime risk for serious gastrointestinal hemorrhage from any site, but especially the stomach, is about 10%. Hypertension is relatively common, in part because of involvement of the renal vasculature. Progressive occlusion of peripheral arteries leads to absence of pulses; acral ischemia is rare because of the development of collaterals. The risk for stroke, myocardial infarction, abdominal angina, and intermittent claudication is increased independent of other risk factors. Impaired left ventricular function is common in adults.  

DIAGNOSIS

Differential Diagnosis

Whole exome sequencing is an efficient and sensitive way to make the diagnosis.21 An acquired form of pseudoxanthoma elasticum has been reported and is also of unclear etiology. This form is difficult to differentiate from a sporadic case in a family because of heterozygosity in the parents, but it tends to affect only the skin. As suggested by the name, the cutaneous features of pseudoxanthoma elasticum need to be differentiated from those of true xanthoma, which results from a disorder of lipid metabolism (Chapter 195). The dermatologic manifestations need to be differentiated from those of Miescher elastoma, elastic tissue nevi (Buschke-Ollendorff syndrome), and solar elastosis.

TREATMENT  No cure for or means of preventing pseudoxanthoma elasticum is known. In many instances, careful attention to the ocular features by a retinal specialist experienced in pseudoxanthoma elasticum can delay but not prevent loss of vision. The risk for gastrointestinal hemorrhage suggests that patients should avoid gastric irritants such as aspirin, nonsteroidal anti-inflammatory drugs, and excessive alcohol. Stool should be checked regularly for occult blood, and angiography may be necessary to detect the source of bleeding. All standard risk factors for atherosclerosis should be managed aggressively. Complaints of chest pain should prompt a rigorous investigation for coronary artery disease. Angioplasty has not been reported to be effective, and the coronary lesions tend to be diffuse. Coronary artery bypass graft surgery has been performed, but long-term results have not been reported. It may be theoretically advantageous to use vein grafts rather than the internal mammary artery for bypass. The excessive wrinkling and pseudoxanthoma in exposed areas can be ameliorated by plastic surgery.



FUTURE DIRECTIONS

Each of these disorders poses special considerations in clinical diagnosis, utility of molecular testing, genetic counseling, and management. For the storage disorders, the clinical utility of enzyme replacement therapy is actively being pursued by several pharmaceutical companies. For several of the other conditions, somatic stem cell therapy offers some promise but is years away from routine clinical use. In Marfan syndrome, clinical trials of drugs that modulate activity of TGF-β are underway. Additionally, close medical management for individuals detected as being at heightened risk for cardiovascular, skeletal, and ocular complications will remain a mainstay.

  Grade A References A1. Groenink M, den Hartog AW, Franken R, et al. Losartan reduces aortic dilatation rate in adults with Marfan syndrome: a randomized controlled trial. Eur Heart J. 2013;34:3491-3500. A2. Lacro RV, Dietz HC, Sleeper LA, et al. Atenolol versus losartan in children and young adults with Marfan’s syndrome. N Engl J Med. 2014;371:2061-2071. A3. Milleron O, Arnoult F, Ropers J, et al. Marfan Sartan: a randomized, double-blind, placebo-controlled trial. Eur Heart J. 2015;36:2160-2166. A4. Teixido-Tura G, Forteza A, Rodríguez-Palomares J, et al. Losartan versus atenolol for prevention of aortic dilation in patients with Marfan syndrome. J Am Coll Cardiol. 2018;72:1613-1618.

A5. Ong KT, Perdu J, De Backer J, et al. Effect of celiprolol on prevention of cardiovascular events in vascular Ehlers-Danlos syndrome: a prospective, randomized, open, blinded-endpoints trial. Lancet. 2010;376:1476-1484. A6. Bishop N, Adami S, Ahmed SF, et al. Risedronate in children with osteogenesis imperfecta: a randomised, double-blind, placebo-controlled trial. Lancet. 2013;382:1424-1432. A7. Dwan K, Phillipi CA, Steiner RD, et al. Bisphosphonate therapy for osteogenesis imperfecta. Cochrane Database Syst Rev. 2014;7:CD005088.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 244  Inherited Diseases of Connective Tissue  

GENERAL REFERENCES 1. Peracha H, Sawamato K, Averill L, et al. Molecular genetics and metabolism, special edition: diagnosis and prognosis of mucopolysaccharidosis IVA. Mol Genet Metab. 2018;125:18-37. 2. James RA, Singh-Grewal D, Lee SJ, et al. Lysosomal storage disorders: a review of the musculoskeletal features. J Paediatr Child Health. 2016;52:262-271. 3. Dornelles AD, Artigalás O, da Silva AA, et al. Efficacy and safety of intravenous laronidase for mucopolysaccharidosis type I: a systematic review and meta-analysis. PLoS ONE. 2017;12:1-18. 4. Boelens JJ, van Hasselt PM. Neurodevelopmental outcome after hematopoietic cell transplantation in inborn errors of metabolism: current considerations and future perspectives. Neuropediatrics. 2016;47:285-292. 5. Motas S, Haurigot V, Garcia M, et al. CNS-directed gene therapy for the treatment of neurologic and somatic mucopolysaccharidosis type II (Hunter syndrome). JCI Insight. 2016;1:1-19. 6. Sakai LY, Keene DR, Renard M, De Backer J. FBN1: the disease-causing gene for Marfan syndrome and other genetic disorders. Gene. 2016;591:279-291. 7. Pyeritz RE. Etiology and pathogenesis of the Marfan syndrome: current understanding. Ann Cardiothorac Surg. 2017;6:595-598. 8. Plichta RP, Glower DD, Hughes GC. Valvular disease in Marfan syndrome: surgical considerations and management. Curr Cardiol Rep. 2019;21:1-9. 9. Gaisl T, Bratton DJ, Kohler M. The impact of obstructive sleep apnoea on the aorta. Eur Respir J. 2015;46:532-544. 10. Albuquerque EV, Scalco RC, Jorge AA. Management of endocrine disease: diagnostic and therapeutic approach to tall stature. Eur J Endocrinol. 2017;176:R339-R353.

1691.e1

11. Esfandiari H, Ansari S, Mohammad-Rabei H, et al. Management strategies of ocular abnormalities in patients with Marfan syndrome: current perspective. J Ophthalmic Vis Res. 2019;14:71-77. 12. David TE, David CM, Manlhiot C, et al. Outcomes of aortic valve-sparing operations in Marfan syndrome. J Am Coll Cardiol. 2015;66:1445-1453. 13. Franken R, Groenink M, de Waard V, et al. Genotype impacts survival in Marfan syndrome. Eur Heart J. 2016;37:3285-3290. 14. Sobey G. Ehlers-Danlos syndrome: how to diagnose and when to perform genetic tests. Arch Dis Child. 2015;100:57-61. 15. Henneton P, Albuisson J, Adham S, et al. Accuracy of clinical diagnostic criteria for patients with vascular Ehlers-Danlos syndrome in a tertiary referral centre. Circ Genom Precis Med. 2019. 16. Forlino A, Marini JC. Osteogenesis imperfecta. Lancet. 2016;387:1657-1671. 17. Palomo T, Viaça T, Lazaretti-Castro M. Osteogenesis imperfecta: diagnosis and treatment. Curr Opin Endocrinol Diabetes Obes. 2017;24:381-388. 18. Tsang SH, Sharma T. Inborn errors of metabolism: pseudoxanthoma elasticum. Adv Exp Med Biol. 2018;1085:187-189. 19. Roach ES, Islam MP. Pseudoxanthoma elasticum. Handb Clin Neurol. 2015;132:215-221. 20. Kranenburg G, Visseren FLJ, de Borst GJ, et al. Arterial stiffening and thickening in patients with pseudoxanthoma elasticum. Atherosclerosis. 2018;270:160-165. 21. Hosen MJ, Van Nieuwerburgh F, Steyaert W, et al. Efficiency of exome sequencing for the molecular diagnosis of pseudoxanthoma elasticum. J Invest Dermatol. 2015;135:992-998.

1691.e2

CHAPTER 244  Inherited Diseases of Connective Tissue  

REVIEW QUESTIONS 1. Osteogenesis imperfecta (OI) syndromes demonstrate: A . Intergenic heterogeneity B. Intragenic heterogeneity C. Variable expression D. Pleiotropy E. All of the above Answer: E  All four of the choices are correct. Mutations in multiple genes can cause OI (intergenic heterogeneity). With a single locus, many different mutant alleles have been discovered (intragenic heterogeneity). Within a given type of OI, relatives with the same mutation can demonstrate different features of varying severity (variable expression). Multiple organ systems are affected in OI (pleiotropy). 2. Which of the following is not a common feature of Marfan syndrome? A . Pulmonary arteriovenous malformation B. Aortic root dilation C. Mitral valve prolapse D. Pulmonic artery dilation E. Aortic dissection Answer: A  All four of the latter choices are common features. Arteriovenous malformations are common in hereditary hemorrhagic telangiectasia but not in Marfan syndrome. 3. Hypertension is a relatively common feature in which syndrome? A . Osteogenesis imperfecta B. Hurler syndrome C. Marfan syndrome D. Vascular Ehlers-Danlos syndrome E. Pseudoxanthoma elasticum Answer: E  Because of partial occlusion of the renal arteries, elevated blood pressure can occur in pseudoxanthoma elasticum. None of the other syndromes has a predisposition to hypertension.

4. Which of the following is not a currently accepted approach to management of Marfan syndrome? A . Prophylactic surgery of the aortic root B. Exercise restriction C. Annual echocardiography D. Gene therapy E. Chronic β-adrenergic blockade Answer: D  At the present time, no approach to correcting the specific mutation in FBN1 is feasible. All of the other approaches are thought to be beneficial. 5. If treatment is not offered, which of the following conditions is associated with the best prognosis? A . Ehlers-Danlos syndrome, hypermobility type B. Osteogenesis imperfecta, type II C. Marfan syndrome D. Hurler syndrome E. Ehlers-Danlos syndrome, vascular type Answer: A  The hypermobility form of Ehlers-Danlos syndrome has little to no added mortality. Osteogenesis imperfecta type II is usually lethal in infancy. Marfan syndrome has reduced life expectancy because of aortic dissection. Patients with Hurler syndrome rarely survive to their third decade. Patients with the vascular form of Ehlers-Danlos syndrome are at high risk for death from arterial or bowel rupture.

CHAPTER 245  The Systemic Autoinflammatory Diseases  

1691

245  THE SYSTEMIC AUTOINFLAMMATORY DISEASES RICHARD M. SIEGEL AND DANIEL L. KASTNER



DEFINITION

The systemic autoinflammatory diseases (Table 245-1) are a group of disorders characterized by seemingly unprovoked inflammation, without evidence of high-titer pathogenic autoantibodies or antigen-specific T cells, thus distinguishing them from the more classic autoimmune diseases.1 The first conditions recognized as autoinflammatory were the hereditary recurrent fevers, a group of mendelian disorders characterized by episodic or fluctuating degrees of fever and localized inflammation. The scope of autoinflammatory disease has been broadened to include other heritable illnesses, including disorders in which purulent or granulomatous inflammation predominates, as well as inherited disorders of the complement system (Chapter 44).2,3 In addition, in numerous autoinflammatory conditions, some of which manifest in childhood and others that occur later in life, there is a complex interaction of genetic susceptibilities and environmental factors. These illnesses include systemiconset juvenile idiopathic arthritis (Still disease), Behçet disease, and even the crystalline arthritides. Recent advances in the genetics and pathophysiology of the inherited autoinflammatory diseases suggest that these conditions are inborn errors of innate immunity, the phylogenetically more primitive part of the immune system that uses germline membrane and intracellular receptors expressed in granulocytes and macrophages to mount the body’s first line of defense against pathogens (Chapters 39 and 42). Autoinflammatory syndromes can be grouped by key pathogenic cytokines and inflammatory pathways that are dysregulated and may be targets for effective therapies. Here we group diseases into those driven by interleukin-1 (IL-1), type I interferons, and activation of the NF-κB inflammatory signaling pathway. For other diseases, such as the newly described deficiency of adenosine deaminase 2 (DADA2), therapies have been found to be effective empirically, even before the pathogenic signaling pathways have been worked out.

  INTERLEUKIN-1–RELATED PERIODIC

FEVER SYNDROMES

The IL-1–associated autoinflammatory diseases are linked by markedly increased expression of or cellular responsiveness to this cytokine, and resolution of symptoms with IL-1 blockade. Interleukin-1α and interleukin-1β (IL-1α and IL-1β) are structurally related cytokines released from cells triggered by a number of inflammatory stimuli, such as lipopolysaccharide. They mediate inflammatory responses by binding to a common receptor that is present on the surface of a wide variety of cell types and signals to activate inflammatory genes through the nuclear factor kappa B (NF-κB) transcription factor complex. IL-1 is part of a larger family of cytokines including IL-18, IL-33, and IL-36, which bind to related receptors and share the property of not having a characteristic signal peptide that normally targets cytokines to secretory vesicles. Because of this, IL-1 family cytokines may be secreted only by dead or dying cells, functioning as molecular markers of cellular stress, which can trigger beneficial inflammatory responses to infection and injury. IL-1β and IL-18 are unique in that they are not biologically active until cleaved by the protease caspase-1, which also cleaves intracellular gasdermin D to induce an inflammatory form of cell death known as pyroptosis. Caspase-1 is itself activated

CHAPTER 245  The Systemic Autoinflammatory Diseases  

ABSTRACT

The systemic autoinflammatory diseases are a group of disorders characterized by seemingly unprovoked inflammation, without evidence of high-titer pathogenic autoantibodies or antigen-specific T cells, thus distinguishing them from the more classic autoimmune diseases. The first conditions recognized as autoinflammatory were the hereditary recurrent fevers, a group of mendelian disorders characterized by episodic or fluctuating degrees of fever and localized inflammation. The scope of autoinflammatory disease has been broadened to include other heritable illnesses, including disorders in which purulent or granulomatous inflammation predominates, as well as inherited disorders of the complement system. Recent advances in the genetics and pathophysiology of the inherited autoinflammatory diseases suggest that these conditions are inborn errors of innate immunity. These diseases can be subdivided based on the cytokine or signaling pathway primarily affected. At present, these disorders can be divided into those affecting interleukin-1, type I interferon, and the NF-κB signaling pathway. These categories are also helpful in guiding therapeutic approaches.

KEYWORDS

autoinflammatory disease familial Mediterranean fever inflammasome interleukin-1 pyrin tumor necrosis factor periodic fever syndromes

1691.e3

1692

CHAPTER 245  The Systemic Autoinflammatory Diseases  

TABLE 245-1 SYSTEMIC AUTOINFLAMMATORY DISEASES: A PARTIAL LISTING INHERITED AUTOINFLAMMATORY DISEASES

INHERITANCE

GENES OR RISK FACTORS

OMIM*

INTERLEUKIN-1-BETA–RELATED DISORDERS Familial Mediterranean fever (FMF) Tumor necrosis factor receptor–associated periodic syndrome (TRAPS) Hyperimmunoglobulinemia D with periodic fever syndrome (HIDS) Neonatal-onset multisystem inflammatory disease (NOMID)/Muckle-Wells syndrome (MWS)/familial cold autoinflammatory syndrome (FCAS)

Autosomal recessive Autosomal dominant Autosomal recessive Autosomal dominant/de novo

MEFV† TNFRSF1A† MVK† NLRP3 (formerly CIAS1)†

Schnitzler syndrome Multiple self-healing palmoplantar carcinoma Familial keratosis lichenoides chronica Autoinflammation with arthritis and dyskeratosis Deficiency of interleukin-1 receptor antagonist (DIRA) Deficiency of the IL-36R antagonist (DITRA)

Sporadic Autosomal dominant

NLRP3 mosaic mutations (some) NLRP1

Autosomal recessive Autosomal recessive

IL1RN† IL36RN†

612852 605507

TREX1, RNASEH2A, 2B, 2C SAMHD1 ADAR (DRADA) IFIH1 (MDA5) TMEM173 PSMB8, genes encoding other proteasome subunits†

225750

249100 142680 260920 607115 191900 120100 606636

INTERFERON-RELATED AUTOINFLAMMATORY SYNDROMES (INTERFERONOPATHIES) Aicardi-Goutières syndrome

Autosomal recessive or Autosomal dominant

STING-associated vasculopathy with onset in infancy (SAVI) Chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature (CANDLE), Nakajo-Nishimura syndrome, JMP syndrome

Autosomal recessive Autosomal recessive

612374 256040

NF-KAPPA-B–RELATED AUTOINFLAMMATORY SYNDROMES Haploinsufficiency of A20 (HA20) syndrome OTULIN deficiency Immunodeficiency and autoinflammatory disease associated with C-terminal NEMO mutations Immunodeficiency, autoinflammation, and amylopectinosis

Autosomal dominant Autosomal recessive X-linked

TNFAIP3† FAM105B (OTULIN)† IKBKG

191163 615712 300248

Autosomal recessive

HOIL1/RBCK1, HOIP/RNF31

Pediatric-onset inflammatory polyarthritis

Autosomal dominant (de novo)

Myd88

610924 612487 602170

Sporadic, autosomal dominant Complex inheritance

NOD2/CARD15† NOD2/CARD15†

186580, 605956 266600

Autosomal dominant Autosomal recessive

C1NH CD55

106100 125240

Autosomal dominant, sporadic Complex inheritance

CFH (complement factor H) CFH (complement factor H)

235400 603075

Deficiency of ADA2 (DADA2) Syndrome of periodic fever with aphthous stomatitis, pharyngitis, and cervical adenopathy (PFAPA) Autoinflammatory disease associated with NLRC4 mutations Systemic-onset juvenile idiopathic arthritis (SOJIA)/adult-onset Still disease

Autosomal recessive Idiopathic

ADA2† —

607575 —

Autosomal dominant Complex inheritance

606831 604302

Behçet disease

Complex inheritance

Syndrome of pyogenic arthritis with pyoderma gangrenosum and acne (PAPA) Chronic recurrent multifocal osteomyelitis (CRMO)

Autosomal dominant Sporadic, autosomal recessive

Synovitis acne pustulosis hyperostosis osteitis syndrome (SAPHO) Crystalline arthropathies

Idiopathic Complex inheritance

NLRC4 HLA-DRB1*11, LACC1, IL6, MIF polymorphisms HLA-B*51, polymorphisms in IL10, IL23R, CCR1, STAT4, KLRC4, ERAP1, MEFV, TLR4, IL1A-IL1B, IRF8, CEPB-PTPN1, ADO-EGR2, RIPK2, LACC1, FUT2 PSTPIP1† LPIN2,† when associated with congenital dyserythropoietic anemia (Majeed syndrome) — SLC2A9/GLUT9, ABCG2

GRANULOMATOUS DISORDERS Early-onset sarcoidosis/Blau syndrome Crohn disease COMPLEMENT DISORDERS Hereditary angioedema Early-onset protein-losing enteropathy, and thrombosis, intestinal lymphangiectasia, and intestinal inflammation Hemolytic-uremic syndrome Age-related macular degeneration OTHER AUTOINFLAMMATORY SYNDROMES

109650

604416 259680 — —

*Online Mendelian Inheritance in Man, an online catalogue of genetic disorders, available at http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=OMIM. Accessed September 29, 2014. † An updated list of disease-associated mutations is available online at http://fmf.igh.cnrs.fr/infevers. Accessed January 7, 2018. AGS https://www.nature.com/articles/nri3850 HOIP http://mirror.omim.org/entry/612487?search=HOIP&highlight=hoip#1 HOIL-1 https://www.ncbi.nlm.nih.gov/pubmed/23104095 OTULIN https://www.ncbi.nlm.nih.gov/pubmed/27523608 https://www.ncbi.nlm.nih.gov/pubmed/27559085 ADA2 = adenosine deaminase 2; A20 = TNF-γ–induced protein 3; JMP = joint contractures, muscle atrophy, microcytic anemia, panniculitits-induced lipodystrophy; NEMO = NF-κB essential modulator.

in cytoplasmic protein complexes containing various sensor proteins such as NLRP3 and the adapter protein ASC. These complexes are referred to as inflammasomes because of their ability to trigger IL-1–mediated inflammation. Autoinflammatory diseases described below are caused by mutations in genes encoding proteins that process or sense IL-1 or affect this process indirectly.



Familial Mediterranean Fever  

DEFINITION

Familial Mediterranean fever (FMF) is a recessively inherited disease that typically manifests with 12- to 72-hour episodes of fever and localized serosal, synovial, or cutaneous inflammation. Between attacks, patients usually feel completely well, although biochemical evidence of inflammation may remain,

CHAPTER 245  The Systemic Autoinflammatory Diseases  

and some patients eventually develop systemic amyloidosis. Before the identification of the causative gene, FMF was defined purely clinically; clinical features remain an important part of the diagnosis, because some patients with typical disease have only one, or sometimes no, demonstrable mutation in MEFV, the only known causative gene.  

EPIDEMIOLOGY

FMF is most common in individuals of Jewish, Arab, Armenian, Turkish, and Italian ancestry. The frequency of asymptomatic carriers of a single MEFV mutation in these populations is as high as 1 in 5, a finding that suggests a selective advantage for heterozygotes. With genetic testing, FMF is now frequently recognized in both Ashkenazi (eastern European) and non-Ashkenazi Jewish populations, as well as in Mediterranean populations previously thought not to be at risk. Mutation-positive individuals with typical symptoms have been documented worldwide. FMF usually manifests in childhood, sometimes even in infancy, although approximately 10% of patients experience their first attack as adults; infrequently, FMF first occurs in persons older than 40 years.  

PATHOBIOLOGY

MEFV, the gene for FMF, was identified by positional cloning in 1997. It encodes a 781–amino acid protein denoted pyrin (or marenostrin) that is expressed in granulocytes, monocytes, and dendritic cells, as well as in peritoneal, synovial, and dermal fibroblasts. The N-terminal 92 amino acids of pyrin are the prototype for a motif, the PYRIN domain, that is involved in protein-protein interactions; this domain defines a family of more than 20 human proteins, including pyrin itself, involved in the regulation of cytokine production (particularly the IL-1 family), NF-κB activation, and cell death. Most of the pathogenic FMF-associated mutations in pyrin reside in the C-terminal domain encoded by exon 10 of MEFV. An even larger number of variants of unknown significance have been described in individual patients with a spectrum of inflammatory phenotypes, and some patients with clinical presentations similar to FMF have been found to have pyrin mutations on only one allele. The assembly of pyrin-containing inflammasomes is activated by bacterial toxins that inactivate the GTPase RhoA, including toxins derived from Clostridia, Yersinia, and Vibrio species.4 RhoA inactivates pyrin by inducing phosphorylation of pyrin on N-terminal serine residues, which leads to sequestration of pyrin by 14-3-3 proteins and inhibition of pyrin’s ability to promote processing of the cytokine IL-1β. These discoveries identified a physiological role for pyrin in host defense against bacteria, and as FMFassociated mutations in MEFV render pyrin resistant to RhoA-mediated inactivation, provided a mechanism for the heightened secretion of IL-1β by cells harboring FMF-associated pyrin mutations.5 These findings also suggest that heightened resistance to pandemic bacterial infections, such as Yersinia pestis (the agent of plague in humans), may explain the extraordinarily high FMF carrier frequencies in certain populations. Additional evidence for the importance of pyrin phosphorylation in the regulation of IL-1β production came from the identification of mutations in one of the phosphorylation sites of pyrin associated with a severe early-onset syndrome of autoinflammation and neutrophilic dermatosis (PAAND).6,7  

CLINICAL MANIFESTATIONS

Episodes of FMF are more properly termed recurrent than periodic, and some patients associate attacks with psychological stress or physical exertion. Women of childbearing age sometimes experience their attacks with menses, with remissions during pregnancy. Some patients are unaware of fever during the attacks, but it is almost always observed when sought. Serosal involvement in FMF is usually peritoneal or pleural. Abdominal attacks are the most frequent, and they may vary from mild discomfort to frank peritonitis, with boardlike rigidity, direct and rebound tenderness, and air-fluid levels on upright films of the abdomen. Regardless of the severity of the abdominal attack, constipation is much more common than diarrhea. If a laparotomy or laparoscopy is performed during an attack, a small amount of sterile exudate rich in polymorphonuclear leukocytes may be found. Except for serosal inflammation, the appendix is normal. Repeated abdominal attacks may cause peritoneal adhesions, but ascites is rare. Pleurisy, usually unilateral, may accompany abdominal pain, or it may occur independently. Physical findings, if present, may include diminished breath sounds and a pleural friction rub, whereas chest radiograms may show a small effusion or atelectasis. With multiple attacks, pleural thickening may develop. Symptomatic nonuremic pericardial involvement in FMF has been reported but is unusual. In adults, the arthritis of FMF typically manifests as monoarticular involvement of the knee, hip, or ankle, and attacks of arthritis may persist for up to

1693

1 week at a time. In children, oligoarticular or polyarticular joint involvement may occur. Large joint effusions are sometimes present, and the synovial fluid may have as many as 100,000 leukocytes/mL. In approximately 5% of patients who are not treated with prophylactic colchicine, chronic arthritis (usually of the hip or knee) may develop, often necessitating joint replacement surgery. Regardless of colchicine treatment or a particular human leukocyte antigen (HLA-B27) status, some patients with FMF develop sacroiliitis. Arthralgia without frank arthritis is common in FMF. Cutaneous manifestations of FMF tend to be less common than serosal or synovial involvement. The characteristic skin lesion of FMF is erysipeloid erythema, a painful, demarcated erythematous area most often seen on the lower leg, ankle, or dorsum of the foot. This rash may occur independently, or it may accompany an episode of arthritis. Histologically, a mixed perivascular cellular infiltrate is seen. Other acute manifestations of FMF include unilateral scrotal inflammation (the tunica vaginalis is an embryologic remnant of the peritoneal membrane) and myalgia, either with fever or, especially in children, without fever and induced by vigorous exercise. Various forms of vasculitis also have been associated with FMF; Henoch-Schönlein purpura may occur in children with FMF; less frequently, polyarteritis nodosa is seen.  

COMPLICATIONS

Before the widespread use of colchicine prophylaxis, systemic AA amyloidosis (Chapter 179) was a frequent complication of FMF, caused by the ectopic deposition of a misfolded fragment of serum amyloid A (SAA), an acute phase reactant, in the gastrointestinal tract, kidneys, spleen, lung, testes, and adrenals. Malabsorption and nephrotic proteinuria leading to renal failure are the most common manifestations of AA amyloidosis. Cardiomyopathy is less common, and neuropathy and arthropathy are rare. Several risk factors for amyloidosis development in FMF have been identified, including late diagnosis of FMF, colchicine noncompliance, male gender, and specific genotypes of the MEFV and SAA genes. Amyloidosis in FMF is less common in the United States than in the Middle East. Abdominal fat aspirates are much less sensitive than rectal or renal biopsy in detecting the amyloidosis of FMF. The latter procedure may be preferred, because of the increasing recognition of nonamyloid glomerular disease in FMF. With early diagnosis, aggressive suppression of the acute phase response with colchicine or adjunctive agents may lead to improvement, but for patients with renal failure, early renal transplantation is preferred.  

DIAGNOSIS

Based on a simple recessive model of inheritance, two mutations in MEFV, in trans, should be identified to establish the genetic diagnosis of FMF. Nevertheless, the interpretation of genetic testing is complicated by complex alleles consisting of various combinations of mutations in cis, as well as by the observations that as many as one third of patients with clinically typical FMF have only one demonstrable mutation in MEFV, and a few patients with typical disease have no identifiable MEFV mutations. These latter two findings suggest that, under some circumstances, one MEFV mutation may be sufficient for symptoms or that additional genes for FMF exist. For these reasons, clinical data remain an essential part of the diagnosis of FMF, and genetic testing plays an adjunctive role in settings in which clinical experience is limited.8 Clinical criteria emphasize attack duration (12 to 72 hours); recurrence of symptoms (three or more episodes); documented fever (rectal temperature >38° C); painful manifestations in the abdomen, chest, joints, or skin; and the absence of other causative factors. The differential diagnosis includes the other hereditary recurrent fever syndromes (Table 245-2), as well as other conditions specific to the clinical setting. For patients with recurrent abdominal pain, considerations include gynecologic disorders, porphyria (Chapter 199; which can be distinguished by hypertension during attacks, dominant inheritance, and urine porphyrins), and hereditary angioedema (Chapter 237; which usually does not cause fever). The syndrome of periodic fever with aphthous stomatitis, pharyngitis, and cervical adenopathy is probably the most common cause of unexplained recurrent fever in children and is also included in the differential diagnosis. In patients presenting primarily with recurrent monoarthritis, joint aspiration for cultures and crystals may aid in excluding bacterial and crystalline arthritis. Still disease in children (systemic-onset juvenile idiopathic arthritis) and adults (adult-onset Still disease) is also considered in the differential diagnosis. Adult-onset Still disease9 (see Table 245-1) is an uncommon autoinflammatory condition of unknown cause that is not considered to be hereditary. It is characterized by spiking fever, an evanescent salmon-pink maculopapular rash, arthritis, and neutrophilic leukocytosis. It can be clinically distinguished from FMF by the pattern of fever (intermittent quotidian in Still disease vs. discrete episodes in FMF),

1694

CHAPTER 245  The Systemic Autoinflammatory Diseases  

TABLE 245-2 CLINICAL CHARACTERISTICS OF SELECTED AUTOINFLAMMATORY DISEASES CLINICAL FEATURE

FMF

TRAPS

HIDS

FCAS/MWS/NOMID

DADA2

Typical ethnicity

Arab, Armenian, Italian, Jewish, Turkish

Any ethnicity

Dutch, other North European

European

Georgian Jewish origin for PAN-like disease

Age of onset

Childhood or adult

Childhood or adult

Childhood, especially tied to immunizations

Childhood or adult (FCAS) Childhood (MWS) Infancy (NOMID)

Usually childhood

Attack duration

12-72 hr

Days to weeks

3-7 days

12-24 hr (FCAS) 1-2 days (MWS) Continuous, with flares (NOMID)

Occasional fevers Can have long asymptomatic periods

Abdominal involvement

Sterile peritonitis, constipation > diarrhea

Severe pain, vomiting, peritonitis

Sterile peritonitis, diarrhea, rarely constipation

Nausea (FCAS) Abdominal pain (MWS) Uncommon (NOMID)

Can be seen associated with abdominal vasculitis and infarctions, portal hypertension

Pleural attacks

Common

Common

Rare

Rare (MWS, NOMID)

Rare

Joint/bone involvement

Monoarthritis, rarely protracted arthritis in knee or hip

Arthritis in large joints, arthralgia

Arthralgia, symmetrical polyarthritis

Polyarthralgia (FCAS, MWS) Oligoarthritis (MWS) Clubbing (MWS, NOMID) Epiphyseal overgrowth, contractures, intermittent or chronic arthritis (NOMID)

Uncommon

Skin rash

Erysipeloid erythema on lower leg, ankle, foot

Migratory rash, underlying myalgia

Diffuse maculopapular rash, urticaria

Urticaria-like rash (cold-induced in FCAS)

Livedo reticularis, cutaneous vasculitis, scarring lesions, and necrosis can occur

Hematologic

Splenomegaly, occasional lymphadenopathy

Splenomegaly, occasional lymphadenopathy

Cervical adenopathy in children

Hepatosplenomegaly, adenopathy (NOMID, rare in MWS)

Cytopenias including pure red-cell aplasia, hypo-Ig, hepatosplenomegaly

Neurologic involvement

Aseptic meningitis?

Controversial

Headache

Headache (FCAS) Sensorineural deafness (MWS, NOMID) Chronic aseptic meningitis, intellectual disability (NOMID)

Recurrent ischemic strokes

Ophthalmologic involvement

Rare

Conjunctivitis, periorbital edema, rarely uveitis

Uncommon

Conjunctivitis (all) Uveitis (MWS, NOMID) Progressive vision loss (NOMID)

Retinal artery occlusion

Vasculitis

Henoch-Schönlein purpura (HSP), polyarteritis nodosa

HSP, lymphocytic vasculitis

Cutaneous vasculitis common, rarely HSP

Not seen Occasional in NOMID

Small and medium vessels, can present as polyarteritis nodosa

Systemic amyloidosis

Risk depends on MEFV and SAA genotypes; more common in Middle East

Occurs in ≈10%; risk increased with cysteine mutations

Rare

Rare (FCAS) 2-5%(MWS) May develop in adulthood (NOMID)

Can occasionally develop

Autoantibodies

Not usually seen

Not usually seen

Not usually seen

Not usually seen

ANCA usually negative

Effective treatments

Colchicine, IL-1 blockade

IL-1 blockade Etanercept

IL-1 blockade

IL-1 blockade

TNF blockade

the pattern of arthritis (chronic polyarthritis vs. intermittent monoarthritis), the characteristic skin involvement (evanescent rash vs. erysipeloid erythema), and the presence of lymphadenopathy (more common in Still disease).

for which there may be a slightly increased risk. Use of colchicine in lactating women is considered safe. Intravenous colchicine should be used with extreme caution, if at all, in FMF, because fatal toxicity has been reported in patients already receiving oral colchicine who are given the drug intravenously. IL-1 inhibitors are usually effective in patients who are poorly responsive to colchicine or who cannot tolerate therapeutic doses. A2  Canakinumab, a recombinant human anti–human-IL-1β antibody, was recently approved by the Food and Drug Administration (FDA) for the treatment of FMF in adults and children. In a randomized, placebo-controlled trial in patients with genetically confirmed colchicine-resistant FMF, mevalonate kinase deficiency, or TRAPS at the time of a flare, canakinumab, 150 mg subcutaneously, with the option for an add-on injection at the same dose, was found to be effective in controlling and preventing flares. A3 

TREATMENT  The mainstay of therapy for FMF is daily oral colchicine, which can prevent both acute attacks of FMF and the development of systemic amyloidosis. A1  Although colchicine has been used as an anti-inflammatory medication since ancient times, and its particular efficacy in FMF was discovered empirically, new insights into pyrin’s negative regulation by the RhoA GTPase may explain its mechanism of action in FMF. As a consequence of its ability to depolymerize microtubules, colchicine activates RhoA, probably through the release of the RhoA guanine-nucleotide-exchange factor GEF-H1 from microtubules. Activated RhoA leads to increased phosphorylation of pyrin, inactivating its pro-inflammatory function, even in cells harboring FMF-associated pyrin mutations. In adults, the therapeutic dose is 1.2 to 1.8 mg/day, and nearly 90% of patients note significant improvement at this dose. The major side effects are gastrointestinal, and they can usually be minimized by gradually increasing the dosage and avoiding milk products in patients who develop lactose intolerance. Most experts continue to prescribe colchicine to patients during pregnancy, with the recommendation that amniocentesis be performed to exclude trisomy 21,

Hyperimmunoglobulinemia D with Periodic Fever Syndrome (Mevalonate Kinase Deficiency)  

Hyperimmunoglobulinemia D with periodic fever syndrome (HIDS) was first described in 1984 as an FMF-like illness seen in six patients of Dutch ancestry. Besides the difference in ethnicity, a key distinction was the observation of extremely high levels of immunoglobulin D (IgD) in the serum of these patients, thus prompting the HIDS nomenclature. HIDS is now recognized

CHAPTER 245  The Systemic Autoinflammatory Diseases  

in a broader ethnic distribution, although northern Europeans still predominate. Overall, HIDS is still quite rare. Family studies documented autosomal recessive inheritance. The elevated IgD levels seen in HIDS appear to be an epiphenomenon and do not correlate with disease severity either among patients or in a given patient over time, although IgD may contribute to the release of pro-inflammatory cytokines in vitro. In 1999, patients with HIDS were found to have mutations in MVK, which encodes the mevalonate kinase enzyme involved in the biosynthesis of cholesterol and nonsterol isoprenes.10,11 Enzyme activity in patients is markedly reduced but not absent. Recent data link the isoprenoid deficiency resulting from MVK mutations to the manifestations of HIDS and make an interesting connection with FMF and other IL-1–related autoinflammatory diseases. Reduced prenylation inactivates the GTPase RhoA, abrogating its negative regulation of the pyrin inflammasome, resulting in increased IL-1β production. IL-1β and increased body temperature can further decrease mevalonate kinase enzymatic activity, thereby creating a vicious circle in which infection or immunization can precipitate HIDS attacks. One of the well-recognized clinical characteristics of HIDS is the provocation of attacks by immunizations. Other distinguishing clinical features include a very early age of onset (average age, 6 months), a duration of attacks intermediate between FMF and TRAPS (3 to 7 days), prominent cervical lymphadenopathy during attacks, polyarticular joint involvement, a diffuse maculopapular rash, the predominance of diarrhea over constipation with abdominal attacks, and the infrequency of pleuritic attacks or systemic amyloidosis. The diagnosis of HIDS can be established in a patient with recurrent episodes of fever and typical associated findings by documenting either two mutations in MVK or elevated levels of mevalonic acid, the substrate for mevalonate kinase, in the urine during attacks. Approximately 10% of patients with otherwise typical disease have only a single identifiable MVK mutation. As modestly increased IgD levels can be seen in other inflammatory conditions, and up to 20% of patients (particularly young children) with typical recurrent fevers and MVK mutations can have normal serum IgD levels, IgD levels alone are an unreliable diagnostic tool. Consequently, some experts prefer the term mevalonate kinase deficiency (MKD) to denote this illness. Nonsteroidal anti-inflammatory drugs (NSAIDs) or corticosteroids are sometimes useful in the treatment of the arthritic manifestations of HIDS. Colchicine is generally not effective. IL-1 inhibitors are generally effective in HIDS, and recently the FDA approved canakinumab for the treatment of this disorder. Patients with HIDS have a normal lifespan, and attacks may become somewhat less frequent in adulthood.

1695

oxygen species, constitutive activation of mitogen-activated protein (MAP) kinases, and increased production of pro-inflammatory cytokines by myeloid cells harboring the mutant receptor.12,13  

DIAGNOSIS

Although genetic testing is necessary for the diagnosis of TRAPS, certain clinical clues can help distinguish TRAPS from FMF. These include ethnicity (FMF is seen predominantly in Mediterranean and Middle Eastern populations, whereas TRAPS has a more widespread distribution), mode of inheritance (autosomal recessive in FMF, dominant in TRAPS), and duration of attacks, which tends to be longer in TRAPS and sometimes approaches continuous symptoms. The rash of FMF is typically erysipeloid erythema on the lower extremity, whereas patients with TRAPS often have a distinctive erythematous rash, often with underlying myalgia, which may migrate on the trunk or centrifugally on the extremities. Ocular involvement, with periorbital edema, conjunctivitis, and occasionally even uveitis, is observed in TRAPS but not in FMF. Finally, whereas colchicine is much more effective than corticosteroids in FMF, the opposite is true in TRAPS. Nevertheless, aside from the difference in duration and susceptibility to pharmacologic intervention, the abdominal, pleural, synovial, and even scrotal manifestations of the two diseases are rather similar. The usual age of onset for TRAPS is also in childhood, and systemic AA amyloidosis is seen in approximately 10% of untreated patients with TRAPS. As in FMF, life expectancy in TRAPS is normal in patients whose disease is not complicated by amyloidosis. As noted earlier, the diagnosis of TRAPS is established by the identification of TNFRSF1A mutations in the appropriate clinical setting. One variant, the substitution of glutamine for arginine at residue 92 (R92Q), is present in more than 1% of individuals with European ancestry and may be associated with a broader spectrum of symptoms than is typically seen in TRAPS, including early inflammatory arthritis or, in some cases, no symptoms at all. The substitution of lysine for proline at residue 46 (P46L) has been described in African American patients with TRAPS and is associated with a receptor shedding defect, but it is also seen among healthy West African controls. These findings establish a “gray zone” for the diagnosis of TRAPS and emphasize the potential role of polymorphisms in the recurrent fever genes in other more common inflammatory phenotypes.

TREATMENT 

Tumor Necrosis Factor Receptor–Associated Periodic Syndrome  



The treatment of TRAPS depends on the frequency and severity of attacks.14 Patients with relatively infrequent, mild episodes may respond to NSAIDs. Patients with more severe attacks that occur infrequently may be treated with corticosteroids, although increasing doses may be required as the episodes become more frequent and toxicities may become limiting. For patients with severe attacks occurring once a month or more frequently, treatment with etanercept, the soluble p75 TNF receptor:Fc fusion protein, may be warranted. This may be a unique effect of etanercept, because there is anecdotal evidence that monoclonal antibodies against TNF may actually exacerbate TRAPS. Consistent with a model implicating upregulation of components of the NLRP3 inflammasome by the aberrant inflammatory signaling in TRAPS, IL-1 inhibitors have also been found to be highly effective in TRAPS, and canakinumab recently gained regulatory approval for this indication following a randomized trial that demonstrated efficacy in controlling flares (see Treatment section under FMF).

DEFINITION

Worldwide, the tumor necrosis factor (TNF) receptor–associated periodic syndrome (TRAPS) is the second most frequently diagnosed hereditary recurrent fever syndrome, behind FMF. TRAPS is defined by recurrent episodes of fever and localized inflammation, in many ways resembling FMF, but differing in key details (noted later) and caused by mutations in TNFRSF1A, encoding the 55-kD receptor for TNF (TNFR1, p55, CD120a). Whereas a positive genetic test is not necessary to diagnose FMF, the diagnosis of TRAPS requires the identification of a TNF receptor mutation. One of the first wellcharacterized families with what was later defined as TRAPS was of Irish ancestry, and the condition was termed familial Hibernian fever to emphasize the ethnic background and clinical differences from FMF. However, with the discovery of TNF receptor mutations in families of other ancestries, the ethnically neutral TRAPS nomenclature was proposed.  

PATHOBIOLOGY

The p55 TNF receptor comprises four cysteine-rich extracellular domains, a transmembrane region, and an intracellular death domain. To date, nearly all of the coding mutations described are in the extracellular domains and approximately one third are missense substitutions of cysteine residues that abolish highly conserved disulfide bonds. The initial description of TRAPS documented a defect in activation-induced ectodomain cleavage of the p55 receptor in patients with the C52F TNFRSF1A mutation, possibly leading to a defect in homeostasis by impaired downregulation of membrane receptors and diminished shedding of potentially antagonistic soluble receptor molecules. However, not all TRAPS-associated TNFR1 mutations exhibit this shedding defect, and recent studies indicate a more complex pathogenetic picture, with the extracellular mutations causing defective trafficking and intracellular accumulation of mutant TNFR1 protein. Spontaneous signaling and possibly activation of the unfolded protein response leads to production of mitochondrial reactive

Cryopyrin-Associated Periodic Syndromes: The Cryopyrinopathies  

Three rare, recurrent febrile disorders usually beginning early in life have been associated with mutations in NLRP3 (formerly CIAS1), the gene encoding a protein variously named cryopyrin, NLRP3, NALP3, PYPAF1, or CATERPILLER 1.1, a key component of the NLRP3 inflammasome that activates caspase-1. These disorders are referred to as cryopyrinopathies or cryopyrinassociated periodic syndromes (CAPS). The least severe clinical phenotype is familial cold autoinflammatory syndrome (FCAS; formerly called familial cold urticaria), which is dominantly inherited and is notable for day-long attacks of chills, fever, headache, diffuse urticarial skin rash, arthralgia, and conjunctivitis, precipitated by generalized cold exposure. Amyloidosis is rare in FCAS. Of intermediate severity is Muckle-Wells syndrome (MWS), also dominantly inherited, in which 1- to 2-day episodes of chills, fever, urticarial rash, limb pain, and arthritis occur independently of cold exposure. Sensorineural

1696

CHAPTER 245  The Systemic Autoinflammatory Diseases  

hearing loss is common in MWS, and systemic amyloidosis may occur. The most severe NLRP3-associated phenotype is neonatal-onset multisystem inflammatory disease (NOMID), known in Europe as chronic infantile neurologic cutaneous and articular (CINCA) syndrome. It is usually sporadic owing to the reduced reproductive fitness of most affected individuals. Fever and constitutional symptoms occur almost daily, often from birth, with generalized urticarial skin rash, a peculiar arthropathy characterized by epiphyseal overgrowth of the long bones, and central nervous system (CNS) involvement that includes chronic aseptic meningitis, uveitis, and cochlear inflammation, which may lead to intellectual disability, blindness, and deafness. In all three cryopyrinopathies, the rash is not true urticaria because there is a neutrophilic rather than a mast cell infiltrate and serum histamine levels are normal. Because there are patients with FCAS, MWS, and NOMID/CINCA without demonstrable germline NLRP3 mutations, these diagnoses remain clinical, although genetic testing serves as a valuable adjunct and has greatly increased the recognition of all three conditions. Deep sequencing has identified somatic NLRP3 mutations in some patients with symptoms consistent with CAPS who are negative for mutations by standard genetic testing. In addition, overlap syndromes that are intermediate between FCAS and MWS and NOMID/ CINCA have been reported. NLRP3 is only one of a number of inflammasomes nucleated by other members of the NLR gene family, and gain-of-function mutations in NLRP1, encoding an NLR family member prominently expressed in the skin, underlie a spectrum of diseases characterized by keratotic skin lesions, arthritis, systemic inflammation and self-healing palmoplantar carcinomas.15 Gain-of-function mutations in NLRC4, another intracellular NLR family member, have been reported in patients with a spectrum of symptoms ranging from FCAS-like disease to severe inflammatory disease with infantileonset enterocolitis and macrophage activation syndrome. IL-18, a cytokine related to IL-1 which is also intracellularly processed by NLR inflammasomes, is particularly elevated in patients with NLRC4 mutations, and blockade of IL-18 has been successful in ameliorating symptoms in one severely affected individual resistant to other therapies.16

  THE INTERFERONOPATHIES:

TYPE-I-INTERFERON ASSOCIATED AUTOINFLAMMATORY DISEASES

A group of hereditary syndromes marked by systemic inflammation has emerged that share induced gene expression of type I interferon target genes and cytokines and have been termed interferonopathies.17 Although a wide variety of genetic lesions can cause these diseases, the common induction of this cytokine signaling pathway has raised hopes that blocking type I interferons or interferon signaling may be beneficial for these patients. These diseases mimic some aspects of the antiviral state and congenital infection syndromes without a role for a known pathogen and also make an interesting counterpoint to systemic lupus erythematosus (SLE) and Sjögren syndrome, polygenic autoimmune diseases where type I interferon–induced gene expression is also seen in peripheral blood. Unlike most other autoinflammatory syndromes, autoantibodies can be found in some type I interferonopathies, placing them on the spectrum between autoinflammatory and autoimmune disease.  



TREATMENT  Blockade with anakinra, a recombinant IL-1 receptor antagonist, is effective in controlling fever and acute phase reactants in all three cryopyrinopathies, and longitudinal analysis of a large series of patients at the U.S. National Institutes of Health showed that long-term treatment with anakinra markedly decreased CNS inflammation and end-organ damage in NOMID/CINCA, which led to the regulatory approval of anakinra for the treatment of this condition in the United States and Europe. More recent studies have also documented the efficacy of rilonacept, another soluble IL-1 blocker, and canakinumab in FCAS and MWS. The latter two agents may be less effective against NOMID/ CINCA because of reduced penetration into the CNS.



Deficiencies of IL-1 and IL-36 Receptor Antagonists

Deficiency of the IL-1 receptor antagonist (DIRA) is characterized by the neonatal onset of a pustular skin rash, multifocal osteomyelitis, periostitis, and, rarely, vasculitis. Fever is not a prominent finding, although acute phase reactants are markedly elevated. DIRA is caused by recessively inherited lossof-function mutations in IL1RN, which encodes the IL-1 receptor antagonist (IL-1Ra). Patients usually present within the first 2 weeks of life with skin lesions ranging from discrete crops of pustules to generalized severe pustulosis or ichthyosiform lesions. Histologic examination demonstrates extensive neutrophilic infiltrates in the dermis and epidermis. Typical radiographic findings include multifocal osteolytic lesions, periosteal elevation of the long bones, heterotopic ossification of the proximal femurs, and widening of the anterior rib ends. Bone biopsies demonstrate sterile purulent osteomyelitis, fibrosis, and sclerosis. In DIRA, the lack of IL-1Ra leads to unopposed IL-1β and IL-1α signaling, whereas in the cryopyrinopathies, NLRP3 mutations lead to inflammasome activation and increased IL-1β production. DIRA patients respond dramatically to anakinra, a recombinant form of the protein they lack. Loss-of-function mutations in IL36RN, coding for a protein that serves a similar function as IL1RA in blocking the activity of the IL-1 family member IL-36, are associated with autosomal recessive inheritance of a syndrome termed DITRA (for deficiency of the IL-36 receptor antagonist) characterized by generalized pustular psoriasis, periodic fevers, and elevation of systemic inflammatory markers.

Aicardi-Goutières Syndrome and Associated Diseases

A neonatal syndrome characterized by severe neurologic dysfunction and CSF pleocytosis, Aicardi-Goutières syndrome (AGS) was first associated with mutations in the TREX1 gene encoding a 3′–5′ DNA exonuclease. The definition of this syndrome has expanded to encompass mutations in a number of other genes associated with DNA or RNA modification or sensing; delayed onset past 1 year of age in some patients, particularly those with IFIH1 or ADAR mutations; and extracerebral features such as glaucoma, hypothyroidism, and inflammatory bowel disease. Certain features, such as “chilblains,” cold-induced vascular-based inflammatory lesions in the hands and feet, overlap with those seen in pediatric lupus, and a small number of AGS patients meet clinical criteria for SLE.

Proteasome Associated Autoinflammatory Syndromes

A constellation of diseases have been described linked to recessive loss-offunction mutations in PSMB8, which encodes the β5i subunit of the proteasome, also known as LMP7. An autosomal recessive syndrome in adults characterized by recurrent fevers, progressive lipodystrophy, joint contractures, and cardiac manifestations has been linked to homozygous missense mutations in PSMB8.18 Patients with a syndrome termed CANDLE (chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature) were found to have homozygous missense and nonsense mutations in PSMB8, with some patients having only one known PSMB8 mutation, or compound mutations with other proteasome subunits.19 It is not yet clear whether these syndromes represent identical diseases related to loss of function of β5i. The β5i proteasome subunit is one of the subunits that are induced in immune cells through immune stimuli such as interferons, altering the proteasome so that it more efficiently processes peptides for antigen presentation to T cells. However, there is no indication of a T-cell component to this disease, and studies have shown that the β5i proteasome subunit can be expressed in nonimmune cells such as adipocytes. A striking interferon transcriptional signature, similar to that seen in SLE, has been observed in circulating blood cells from patients with CANDLE. Defective degradation of proteins in cells lacking β5i may result in buildup of ubiquitinated proteins, which somehow triggers interferon production, or PSMB8 deficiency may enhance interferon signaling by stabilizing components of the interferon signal transduction machinery that are negatively regulated by ubiquitin-proteasome degradation. Whichever the mechanism, the link to interferon hyperactivity suggests that blocking interferons with antibodies or inhibitors of interferon signal transduction may be effective in the therapy of CANDLE and possibly other proteasomeassociated autoinflammatory syndromes (PRAAS).

  AUTOINFLAMMATORY DISEASES ASSOCIATED

WITH ACTIVATION OF NF-κB

Although the NF-κB family of transcription factors was one of the first to be described that transduces extracellular signals into induction of inflammatory gene expression, until recently, no mutations in this pathway were known to cause human autoinflammatory disease. Loss-of-function mutations in IKBKG, encoding NEMO (NF-kappa-B essential modulator), a structural component of the I-κB kinase (IKK) complex, cause an X-linked immunodeficiency syndrome, and inherited mutations in TNFRSF13B, encoding TACI, a member of the TNF-family cytokine receptor gene family, cause some cases of common variable immunodeficiency. More recently, a number of kindreds

CHAPTER 245  The Systemic Autoinflammatory Diseases  

The protein mutated in all three disorders is NLRP3, a critical component of the eponymous NLRP3 inflammasome, which serves as an intracellular scaffold for the processing of IL-1β. The alternative name of this protein, cryopyrin, refers to its amino terminal PYRIN domain, the basis for a structural and functional relationship to the protein mutated in FMF. Disease-associated cryopyrin mutations are thought to decrease the threshold for inflammasome

1696.e1

activation, thereby increasing IL-1β production. The discovery that the NLRP3 inflammasome is also necessary for IL-1β production in response to crystalline forms of monosodium urate and calcium pyrophosphate connected the pathophysiology of these rare autoinflammatory diseases to crystal-induced arthritis (Chapter 257), which shares some clinical features, such as episodic, self-limited attacks, with autoinflammatory diseases.

CHAPTER 245  The Systemic Autoinflammatory Diseases  

have been described with heterozygous loss-of-function mutations in TNFAIP3, encoding the A20 protein, which negatively regulates induction of NF-κB through promoting K48 linked ubiquitination and degradation of intracellular components of TNF-family receptor signaling complexes such as RIP1.20 In these kindreds, individuals with these mutations develop symptoms with some features of Behçet disease, such as oral and genital ulcers, uveitis, and skin abscesses. Biochemical hyperinduction of NF-κB and inflammatory cytokine production by stimuli such as LPS and TNF could be seen in peripheral blood myeloid cells from these patients. Mutations in IKBKG that lead to truncations in NEMO that are unable to bind A20 have also been found to lead to inflammatory disease in addition to the immunodeficiency resulting from defects in other functions of NEMO. Recently, a patient with a de novo gain-of-function mutation in the MYD88 gene coding for the signal transduction protein Myd88 was described with pediatric-onset severe destructive polyarthritis.21 This mutation is similar to those seen in somatic activating mutations associated with lymphoma. Myd88 activates NF-κB downstream of TLR and IL-1 receptors, and cells from this patient exhibit increased NF-κB activation and enhanced production of pro-inflammatory cytokines. These findings and other syndromes listed in Table 245-1 have validated the role of NF-κB and A20 in regulating clinically relevant inflammatory responses and point to the need for better therapeutic agents which can effectively regulate NF-κB activation in genetic and acquired diseases where this pathway is hyperactive.

OTHER INHERITED SYSTEMIC AUTOINFLAMMATORY DISEASES

Syndrome of Pyogenic Arthritis with Pyoderma Gangrenosum and Acne  

The syndrome of pyogenic arthritis with pyoderma gangrenosum and acne (PAPA) is a rare, dominantly inherited autoinflammatory disease characterized by intermittent episodes of sterile pyogenic arthritis, pyoderma gangrenosum, and severe cystic acne. It is caused by mutations in the gene encoding prolineserine-threonine phosphatase–interacting protein 1 (PSTPIP1), also known as CD2BP1. PSTPIP1 is a cytoskeletal protein that interacts with certain other proteins involved in the immune response, including CD2; the WiskottAldrich syndrome protein (WASP); a phosphatase denoted PTP-PEST; and pyrin, the FMF protein. PAPA mutations abrogate the binding of PSTPIP1 to PTP-PEST, leading to hyperphosphorylation of PSTPIP1 and increased binding to pyrin. Both in patients and in cell lines, this finding is associated with markedly increased IL-1β production. Early in life, PAPA tends to present with monoarticular or pauciarticular pyogenic arthritis, sometimes induced by trauma. In the absence of treatment, arthritis may progress to severe joint damage and ankylosis. As patients reach puberty, skin manifestations begin to predominate, including disfiguring cystic acne. Pathergy also may develop, and extensive pyoderma gangrenosum may require opiates for pain control. The diagnosis of PAPA syndrome is made by documenting PSTPIP1 mutations in the appropriate clinical setting. High doses of corticosteroids have been used in PAPA, with varying success, and patients with arthritis sometimes require aspiration, intra-articular corticosteroids, or open drainage. Newer investigational approaches for PAPA syndrome focus on the use of targeted cytokine inhibitors. Anecdotal evidence supports the use of anakinra or canakinumab for the arthritis and monoclonal anti-TNF antibodies for the pyoderma gangrenosum of PAPA.

Granulomatous Inflammatory Arthritis, Dermatitis, and Uveitis (Blau Syndrome)  

Blau syndrome is a rare, dominantly inherited illness characterized by the following features: early-onset granulomatous synovitis often complicated by cyst formation and camptodactyly (flexion contractures of the fingers and toes); granulomatous anterior and posterior uveitis, sometimes causing retinal detachment, glaucoma, cataracts, and blindness; and an intermittent papular rash with noncaseating granulomas. Lung or other visceral involvement is generally not present. However, visceral involvement of the liver and spleen is observed in early-onset sarcoidosis (Chapter 89), which is phenotypically quite similar to Blau syndrome. Both Blau syndrome and some cases of earlyonset sarcoidosis are caused by mutations in NOD2/CARD15. Distinct variants of NOD2/CARD15 have been associated with susceptibility to Crohn disease, which manifests as granulomatous inflammation of the gastrointestinal tract (Chapter 132). The protein encoded by this gene is thought to be an intracellular sensor of bacterial products. Crohn disease–associated mutations in the ligand-binding, leucine-rich repeat region of the protein may alter responses

1697

to bacterial products in the gastrointestinal tract to cause inflammation, whereas Blau syndrome mutations in the nucleotide binding domain may lead to constitutive extraintestinal inflammation. Topical and systemic corticosteroids are currently the mainstay of treatment of Blau syndrome. There are case reports of the efficacy of TNF and IL-1 inhibitors in this disease.  

Deficiency of ADA2 (DADA2)

A syndrome characterized by fevers, early-onset strokes, and vasculopathy or frank vasculitis mimicking polyarteritis nodosa has been found to be caused by autosomal recessive mutations in ADA2 (formerly CECR1), encoding adenosine deaminase 2 (ADA2).22 Although sharing enzymatic activity with intracellular adenosine deaminase encoded by the ADA gene, ADA2 is not necessary for lymphocyte development and DADA2 patients do not initially present with severe combined immunodeficiency (SCID), as is the case for ADA deficiency. ADA2 is a secreted extracellular protein, and functional studies have found that ADA2 plays a role in vascular development and regulation of macrophage differentiation. Although this disease is generally diagnosed in childhood, it should be considered in the differential diagnosis of young adults with unexplained lacunar strokes and/or vasculitic skin lesions. As ADA2 is expressed by myeloid cells, bone marrow transplantation may be therapeutic, and TNF inhibition has also shown efficacy in preventing recurrent strokes, though the underlying mechanisms are not yet clear.23

  NEW AUTOINFLAMMATORY SYNDROMES AND

THE PROMISE OF WHOLE-EXOME SEQUENCING

Recent years have seen a dramatic acceleration in the pace of discovery of new mendelian inflammatory diseases as a result of the availability of whole-exome sequencing, which allows unbiased identification of disease-causing mutations in protein coding sequences, although it should be noted that accurate clinical description of these syndromes is as important as the genetic tools for identification of new syndromes. These discoveries have confirmed the role of gene products in human inflammation that were identified in animal model systems and identified new genes and proteins not previously thought to be involved in the regulation of inflammation. More sensitive deep sequencing techniques have also allowed discovery of subgenomic mosaic mutations in NLRP3 and TNFRSF1A, causing similar or related syndromes to those associated with de novo or inherited mutations in these genes. For example, mosaic mutations in NLRP3 account for some cases of cryopyrinopathies that were found to be “negative” for these mutations by standard whole-exome sequencing analysis, and low-percentage NLRP3 mutations have been found in Schnitzler syndrome, which is characterized by monoclonal gammopathy, neutrophilic skin lesions resembling those in cryopyrinopathies, and systemic inflammation.24 Early-onset, apparently sporadic, cases of inflammatory syndromes have often turned out to be due to de novo mutations in a child when screened against parental DNA. For example, inherited gain-of-function mutations in CARD14, encoding an adapter protein in innate immune sensing, cause dominantly inherited familial psoriasis, and a more severe gain-of-function de novo mutation in the same gene causes infantile-onset severe pustular psoriasis. Recessive mutations in HOIL1/RBCK1 that impair the addition of linear ubiquitin chains to receptor signaling complexes cause a complex syndrome marked by autoinflammation and immunodeficiency and intramuscular glycogen deposition. Gainof-function mutations in PLCG2, encoding phospholipase Cγ2, an enzyme with essential functions in B-cell receptor and Fc receptor signaling, cause a dominantly inherited autoinflammatory syndrome characterized by blistering skin lesions, bronchiolitis, arthralgia, ocular inflammation, and enterocolitis in the absence of autoantibodies.

  Grade A References A1. Ozen S, Demirkaya E, Erer B, et al. EULAR recommendations for the management of familial Mediterranean fever. Ann Rheum Dis. 2016;75:644-651. A2. Hashkes PJ, Spalding SJ, Giannini EH, et al. Rilonacept for colchicine-resistant or -intolerant familial Mediterranean fever: a randomized trial. Ann Intern Med. 2012;157:533-541. A3. De Benedetti F, Gattorno M, Anton J, et al. Canakinumab for the treatment of autoinflammatory recurrent fever syndromes. N Engl J Med. 2018;378:1908-1919.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 245  The Systemic Autoinflammatory Diseases  

GENERAL REFERENCES 1. Georgin-Lavialle S, Fayand A, Rodrigues F, et al. Autoinflammatory diseases: state of the art. Presse Med. 2019;48:e25-e48. 2. Hoffman HM, Broderick L. The role of the inflammasome in patients with autoinflammatory diseases. J Allergy Clin Immunol. 2016;138:3-14. 3. Manthiram K, Zhou Q, Aksentijevich I, et al. The monogenic autoinflammatory diseases define new pathways in human innate immunity and inflammation. Nat Immunol. 2017;18:832-842. 4. Park YH, Wood G, Kastner DL, et al. Pyrin inflammasome activation and RhoA signaling in the autoinflammatory diseases FMF and HIDS. Nat Immunol. 2016;17:914-921. 5. Chung LK, Park YH, Zheng Y, et al. The Yersinia virulence factor YopM hijacks host kinases to inhibit type III effector-triggered activation of the pyrin inflammasome. Cell Host Microbe. 2016;20:296-306. 6. Masters SL, Lagou V, Jeru I, et al. Familial autoinflammation with neutrophilic dermatosis reveals a regulatory mechanism of pyrin activation. Sci Transl Med. 2016;8:1-10. 7. Moghaddas F, Llamas R, De Nardo D, et al. A novel pyrin-associated autoinflammation with neutrophilic dermatosis mutation further defines 14-3-3 binding of pyrin and distinction to familial Mediterranean fever. Ann Rheum Dis. 2017;76:2085-2094. 8. Federici S, Vanoni F, Ben-Chetrit E, et al. An international Delphi survey for the definition of new classification criteria for familial Mediterranean fever, mevalonate kinase deficiency, TNF receptorassociated periodic fever syndromes, and cryopyrin-associated periodic syndrome. J Rheumatol. 2019;46:429-436. 9. Govoni M, Bortoluzzi A, Rossi D, et al. How I treat patients with adult onset Still’s disease in clinical practice. Autoimmun Rev. 2017;16:1016-1023. 10. van der Meer JW, Simon A. The challenge of autoinflammatory syndromes: with an emphasis on hyper-IgD syndrome. Rheumatology (Oxford). 2016;55:ii23-ii29. 11. Tanaka T, Yoshioka K, Nishikomori R, et al. National survey of Japanese patients with mevalonate kinase deficiency reveals distinctive genetic and clinical characteristics. Mod Rheumatol. 2018;1-7. 12. Borghini S, Ferrera D, Prigione I, et al. Gene expression profile in TNF receptor-associated periodic syndrome reveals constitutively enhanced pathways and new players in the underlying inflammation. Clin Exp Rheumatol. 2016;34:S121-S128.

1697.e1

13. Rigante D, Frediani B, Cantarini L. A comprehensive overview of the hereditary periodic fever syndromes. Clin Rev Allergy Immunol. 2018;54:446-453. 14. ter Haar NM, Oswald M, Jeyaratnam J, et al. Recommendations for the management of autoinflammatory diseases. Ann Rheum Dis. 2015;74:1636-1644. 15. Zhong FL, Mamai O, Sborgi L, et al. Germline NLRP1 mutations cause skin inflammatory and cancer susceptibility syndromes via inflammasome activation. Cell. 2016;167:187-202. 16. Canna SW, Girard C, Malle L, et al. Life-threatening NLRC4-associated hyperinflammation successfully treated with IL-18 inhibition. J Allergy Clin Immunol. 2017;139:1698-1701. 17. Crow YJ, Manel N. Aicardi-Goutieres syndrome and the type I interferonopathies. Nat Rev Immunol. 2015;15:429-440. 18. Honda-Ozaki F, Terashima M, Niwa A, et al. Pluripotent stem cell model of Nakajo-Nishimura syndrome untangles proinflammatory pathways mediated by oxidative stress. Stem Cell Reports. 2018;10:1835-1850. 19. Brehm A, Liu Y, Sheikh A, et al. Additive loss-of-function proteasome subunit mutations in CANDLE/ PRAAS patients promote type I IFN production. J Clin Invest. 2015;125:4196-4211. 20. Zhou Q, Wang H, Schwartz DM, et al. Loss-of-function mutations in TNFAIP3 leading to A20 haploinsufficiency cause an early-onset autoinflammatory disease. Nat Genet. 2016;48:67-73. 21. Sikora KA, Bennett JR, Vyncke L, et al. Germline gain-of-function myeloid differentiation primary response gene-88 (MYD88) mutation in a child with severe arthritis. J Allergy Clin Immunol. 2018;141:1943-1947. 22. Moens L, Hershfield M, Arts K, et al. Human adenosine deaminase 2 deficiency: a multi-faceted inborn error of immunity. Immunol Rev. 2019;287:62-72. 23. Ombrello AK, Qin J, Hoffman PM, et al. Treatment strategies for deficiency of adenosine deaminase 2. N Engl J Med. 2019;380:1582-1584. 24. de Koning HD, van Gijn ME, Stoffels M, et al. Myeloid lineage-restricted somatic mosaicism of NLRP3 mutations in patients with variant Schnitzler syndrome. J Allergy Clin Immunol. 2015;135:561-564.

1698

CHAPTER 246 Osteoarthritis  

246  OSTEOARTHRITIS VIRGINIA BYERS KRAUS AND TONIA L. VINCENT



DEFINITION

Osteoarthritis (OA) presents in many phenotypic forms and can involve any joint in the body. Although it is the most prevalent form of arthritis, it has been likened to a collection of orphan diseases because of the heterogeneity of presentations.1 Characteristic patterns of joint involvement are recognizable and can often provide clues to specific etiologies of disease (Table 246-1). For instance, there is clear evidence for a polyarticular subset of hand OA in women. There are three major determinants of the pattern of polyarticular involvement: symmetry, clustering by row, and clustering by ray. There are also indications that subtle morphometric abnormalities of the hip, such as acetabular depth and shape and femoral head morphometry, can increase the risk for hip OA.2 It is estimated that at least half of the susceptibility to the various major forms of OA (spine, hands, knees, hip) in the population is explained by genetic factors. In contrast, any joint can develop OA as a consequence of a severe injury or repetitive overuse. Although the characteristic radiographic features (joint space narrowing and osteophytes) have long been accepted as the sine qua non of OA, a long preradiographic stage is now recognized. This has prompted a new focus on the disease process, relating to underlying endotypes and distinct pathobiologic mechanisms. Despite different patterns of joint involvement, the end stage of OA represents a common pathology. Despite its prevalence, OA has attracted relatively modest research attention over recent decades compared with the autoimmune arthritides, and this has contributed to the perpetuation of a number of unhelpful paradigms such as, “OA is due to wearing down of joint surfaces,” “OA is an inevitable consequence of aging,” “non–weight-bearing joints do not get true OA,” and “damaged joints do not have the ability to repair themselves.” In recent years, much progress has been made in understanding the pathophysiology of disease. In efforts to further the understanding of the pathophysiology of the disease, to facilitate communication across the field, and to help advance research and drug development for OA, the Osteoarthritis Research Society International (OARSI) supported an initiative to create a pathobiologic definition of OA. The current definition is as follows: “Osteoarthritis is a disorder involving movable joints characterized by cell stress and extracellular matrix degradation initiated by micro and macro injury that activates maladaptive repair responses including pro-inflammatory pathways of innate immunity. The disease manifests first as a molecular derangement (abnormal joint tissue metabolism) followed by anatomic, and/or physiologic derangements (characterized by cartilage degradation, bone remodeling, osteophyte formation, joint inflammation and loss of normal joint function), that can culminate in illness.”3 The proposed definition was not intended to distinguish an OA patient uniquely from patients with other forms of arthritis, but rather to spur scientific advances and serve as the building blocks for defining OA phenotypes and molecular endotypes.

TABLE 246-1 RECOGNIZED DISTINCT PATTERNS OF OSTEOARTHRITIS JOINT SITE

COMMON PRESENTATION

INFORMATIVE VARIATIONS

Knee

Medial dominant: idiopathic or injury related

Lateral dominant: injury related

Hip

Central (medial, concentric) cartilage loss in women, commonly bilateral, associated with hand OA, with less tendency to progress

Superolateral pole in men, unilateral and progressive

Hand

Nodal generalized OA—“arthritis of menopause” involving multiple finger joints and predominating in women

Predisposition to OA of knee, hip, and spine; erosive subset*

Ankle

Injury related

Increased ankle OA risk contralateral to knee OA

Spine

Lumbar and cervical spine associated with hand OA

*Erosion is often present in hand osteoarthritis (OA) if the detection method is sensitive, e.g., magnetic resonance imaging. Information in part taken from Doherty M, ed. Color Atlas and Text of Osteoarthritis. Barcelona: Wolfe Publishing; 1994.



EPIDEMIOLOGY

Prevalence and Societal Burden

Among the approximately 100 different arthritic conditions, OA is the most prevalent. It affects more than 320 million individuals globally, including an estimated 30.8 million adults in the United States (13.4% of the civilian adult U.S. population).4 Because the global burden estimates only consider hip and knee OA, and not OA at other sites, it is highly likely that the real burden of OA has been underestimated. Estimates may also be conservative because they do not account for the current trend of increased obesity, which contributes to OA, and the “graying” (aging) of society, a leading contributor to the increasing prevalence of OA. Recent estimates suggest that arthritis prevalence in the United States has also been substantially underestimated, especially among adults younger than 65 years of age, owing to the low sensitivity of the single screening question on the National Health Interview Survey for physician-diagnosed arthritis.5 Given that OA is by far the most prevalent of all arthritides, these findings underscore the major underestimated societal impact of OA in the U.S. population and, by analogy, worldwide.

Osteoarthritis as a Serious Disease

OA has all the hallmarks of a serious condition. Lower extremity OA is the leading cause of mobility impairment in older adults in the United States. Disability and loss of function associated with OA are higher in women, those with lower education levels, the socially disadvantaged, and those reliant on manual labor, weight bearing, or positions that involve walking or knee bending for their livelihood. Pain from arthritis is one of the key barriers to maintaining physical activity. The more severe the walking disability, the higher the risk for death, largely due to cardiovascular disease. Overall, knee OA is associated with a 1.55-fold increased risk for all-cause mortality compared with the general population.  

PATHOBIOLOGY

OA is the oldest known disease, affecting ancient hominids millions of years ago through modern-day humans. Some joints are commonly affected, whereas others are rarely involved. An evolutionary perspective attributes the pattern of joint involvement in generalized OA to the fact that some joints have undergone recent rapid evolutionary change and are underdesigned to cope with the new mechanical stresses placed on them.

Mechanical Factors

Mechanical factors are the most important etiological agents in OA development. The epidemiology is compelling: mechanical joint injury or overuse, joint malalignment, and dysplasia are strong independent risk factors for developing disease, even in young individuals. Chondrocytes have several mechanosensing mechanisms and can perceive loads above a defined threshold as injurious. As we age, our ability to reduce the impact of joint loading during normal activities is reduced by loss of muscle mass (e.g., quadriceps strength across the knee), loss of gait reflexes, and poor response times. In vitro, threshold strains above 10% at 0.5 Hz for 12 hours of loading leads to a predominance of cartilage catabolism. In essence, OA will occur when abnormal loads traverse a normal joint or when normal loads are experienced by a joint that has lost its mechanoprotective mechanisms (Table 246-2), pointing to a “Goldilocks principle” with respect to loading that is context specific (Fig. 246-1). This understanding of the prime role of mechanosensing by chondrocytes in the etiology of OA fits well with the traditional adage that OA is an “inside-out” disease, in contrast to rheumatoid arthritis, which is an “outside-in” disease (having its origin in synovitis). Conversely, joints that are immobilized do not develop disease. This is evident in individuals who have sustained a stroke (Fig. 246-2) or previous polio as well as in animal studies where casting of limbs after induction of disease halts development. Importantly, the most significant load comes from

TABLE 246-2 MECHANICAL ETIOLOGIES OF OSTEOARTHRITIS ABNORMAL LOAD ON NORMAL JOINT Direct articular trauma Obesity Repetitive occupational load, e.g., “coal miner’s back,” “cotton-picker’s thumb” Joint malalignment, e.g., valgus and varus deformities

NORMAL LOAD ON AN UNPROTECTED JOINT Aging by loss of muscle support and gait reflexes Chondrodysplasia through maladapted joint shape and weakened joint tissues Joint destabilization, e.g., ruptured anterior cruciate ligament, meniscal tear Cartilage weakened by previous arthritis, e.g., gout, rheumatoid arthritis, sepsis (in the past referred to as secondary osteoarthritis)

CHAPTER 246 Osteoarthritis  

ABSTRACT

Osteoarthritis (OA) is the most prevalent of all arthritides and one of the most prevalent conditions in the world. Although OA represents a common endstage joint pathology, it can present in many phenotypic forms and involve any joint in the body. Mechanical factors are the most important etiological agents in OA development. The disease process involves pro-inflammatory pathways of innate immunity. Recent evidence also demonstrates an important role of failed tissue repair and regeneration in the pathogenesis of OA. OA has all the hallmarks of a serious condition including associated disability, loss of function, increased mortality, likely due to limitations of mobility, and high economic burden. OA is similar to other chronic diseases, such as osteoporosis and heart disease, with a long, indolent and silent initial stage. When progressive, OA eventually presents clinically with disease (anatomic, physiological, and serological abnormalities) and illness (pain, aching, stiffness) manifestations. Despite the magnitude of the health burden and disability of OA, current treatments focus on analgesia, although many attempts are currently underway to develop disease modifying OA drugs (DMOADs).

KEYWORDS

osteoarthritis cartilage bone synovium mechanics inflammation joint pain

1698.e1

CHAPTER 246 Osteoarthritis  

• Production of proteases • Production of cytokines and chemokines • Release of matrix fragments (’DAMPs’) • Cell senescence

Cartilage thickening Increased

1699

Macro and Micro injury

Mechanical load

Injurious

Synovial and immune cell activation/recruitment

Joint cells perceive injurious load

Reduced

Endogenous repair

Cartilage thinning

• Weakened cartilage matrix • Reduced subchondral bone compliance

Homeostasis

• Potentiation of catabolic damage • Suppression of matrix synthesis • Inhibition of repair

Disease

FIGURE 246-1.  Vicious cycle of mechanically induced sterile inflammation in osteoarthritis.

FIGURE 246-2.  Asymmetrical osteoarthritis (OA) in a patient with left-sided hemiparesis. This demonstrates that joints that are immobilized do not develop OA.

the action of muscle across the joint rather than from weight, and this explains why OA affects upper limb joints such as the fingers and why these are also protected by immobility (see Fig. 246-2). Thresholds of perceived load can be observed in exercise dosage studies in rodents; increasing exercise in a modest way increases the thickness of cartilage, consistent with a hypertrophic tissue response, but forced treadmill exercise leads to cartilage degradation and disease.6

Aging

Age is a very strong risk factor for disease. The contribution of age to disease occurs at a number of levels. Probably the most important impact comes from loss of mechanoprotective muscle support across joints with aging (mentioned previously). It is also the case that the tissues of the joint become stiffer because of increased cross-linking of the matrix, and this changes how cells of the joint are able to respond both to physiologic and injurious mechanical load. Cellular aging also changes the metabolic phenotype of the cell, making it more susceptible to free radical–induced damage and driving cell death or cellular senescence. These features likely contribute to a reduction in regeneration potential with age.

Genetic Factors

Genetic factors play an important role in risk because OA is highly heritable. Although rare monogenic forms of OA exist (often associated with chondrodysplasia—abnormal joint formation), its most common forms are

complex, that is, polygenic involving multiple genes, each contributing a low risk of disease.7 Although genetic etiologies are thought to account for as much as 60% of hip OA, 39% to 65% of hand and knee OA, and 70% of spine OA, it is estimated that only 22.5% of the heritable causes of OA have been discovered to date in a genome-wide association study for OA of 77,052 cases and 378,169 controls.8 Genome-wide association studies in large-joint OA have identified over 60 candidates that reach genome-wide significance, although the clinical relevance of each of these is yet to be determined; some of these may turn out to be genes that determine developmental joint shape rather than pathways that might be modifiable, nonetheless, certain strong pathways are emerging (e.g., members of the TGF-beta family TGFβ-signaling-pathway). In general, genetic factors involved in risk for OA often act by modulating gene regulation and may be characterized by polymorphisms that lead to transcriptional differences—termed allelic expression imbalance—whereby one allele produces fewer transcripts than the other allele.9

Metabolic Factors

Metabolic factors play a role in OA risk. The jury is still out on whether “metabolic OA” exists as a discrete entity. This terminology is variously applied to features associated with the metabolic syndrome (obesity, dyslipidemia, diabetes, and hypertension). Epidemiology appears to show that body mass index (BMI) alone accounts for the increased risk in this group, although hypertension may also be an independent risk factor.10 Obesity contributes to disease risk through mechanical overload as well as poor joint muscle support as a result of sedentary behavior. Despite low BMI, elderly subjects in Beijing, China had an equal (men) or higher (women) prevalence of radiographic and symptomatic knee OA compared with individuals in the Framingham, Massachusetts, cohort, pointing to genetic and environmental factors (such as squatting or heavy physical activity) that underlie these differences. Adiposity is also associated with increased systemic inflammation driven in part by inflammatory cytokines, so-called adipokines, that are made by adipocytes.11 This systemic inflammatory component of obesity appears to be tied to the microbiome and the ability of very pro-inflammatory microbiome-related factors, such as lipopolysaccharide, to evade clearance by a fatty liver. Thus, the presence of nonalcoholic fatty liver disease may be a superior indicator of adverse inflammatory effects of obesity compared with BMI alone.

Gender

Gender influences both radiographic and symptomatic OA incidence that is more common in women after menopause. This association is largely absent before menopause, indicating that female sex hormones protect joints from OA. Hand OA in particular has a peak incidence around the time of menopause in women.

1700

CHAPTER 246 Osteoarthritis  

Molecular Pathophysiology: From Bench to Bedside The Role of Proteases in Osteoarthritis

The finding of specific fragments of aggrecan in OA synovial fluid was the first indication that proteases were important in the pathophysiology of OA. Although several matrix metalloproteinases (MMPs) were, by then, characterized, the fragments in the joint did not appear to resemble those generated by known MMPs, and the term aggrecanase was coined. In 1999, the first aggrecanase, a disintegrin and metalloproteinase with thrombospondin motif 5 (ADAMTS5), was identified. The subsequent demonstration that mice deficient in ADAMTS5 were protected from surgically induced OA provided definitive evidence that OA was a potentially targetable disease. Aggecanase inhibitors were shelved, but there has been recent renewed interest in this enzyme as a therapeutic target for OA. Unwanted side effects were partly to blame, as was the challenge of taking these types of drugs into clinical trial in the absence of good disease biomarkers.

Mechanobiology in Osteoarthritis

The discovery of proteases as targets in disease appeared, at first, to be at odds with the strong mechanical etiology. However, the observation that cells of cartilage and synovium could induce proteases in response to mechanical injury provided a mechanism to explain how abnormal mechanical load could provoke disease. This may also explain how the elimination of surface chondrocytes also blocks OA development after injury. In mice, joint immobilization after induction of OA not only protects joints from cartilage loss but also abrogates the up-regulation of catabolic genes.12 A number of intracellular pathways are activated in response to mechanical injury, driving both reparative and degradative responses. Mechanisms include release of growth factors from the cartilage matrix in response to tissue injury and activation of mechanosensitive receptors on the cell surface (see E-Fig. 246-1). Once the cartilage matrix is weakened by proteolytic activity, even noninjurious load may then be perceived as injurious by the chondrocyte, setting up a cycle of chronic, mechanically induced matrix degradation (Fig. 246-1).

Inflammatory Cytokines in Osteoarthritis

Inflammation is recognized as a hallmark of OA, although whether it is a cause or consequence of disease is still debated. Sensitive imaging (by ultrasound, magnetic resonance imaging (MRI), and etarfolatide that identifies sites of activated macrophages) demonstrates inflammation (effusions and synovial thickening) in the majority (~75%) of knees with radiographic OA. MRI effusion synovitis and Hoffa synovitis (inflammation of the intra-articular fat pad) predict about a three-fold increased risk for incident radiographic OA with a year. Unlike rheumatoid arthritis, OA, like many other chronic diseases, involves the innate immune system. All the previously discussed factors suggest a robust whole joint model of a failed wound healing response that integrates the key role of mechanics as an inciting factor and innate immune wound healing response and inflammation as a perpetuating process (see E-Fig. 2461). A normal wound healing response entails an initial autodébridement phase followed by a cellular proliferation and fibrotic wound remodeling phase. A joint with active OA appears to be in a perpetual state of autodébridement, with macrophages taking on dual roles of both pro- and anti-inflammatory (M1 and M2) phenotypes simultaneously. Mouse strains with less susceptibility to OA after injury have an attenuated inflammatory response. Despite the fact that inflammatory cytokines such as interleukin-1 (IL-1) and tumor necrosis factor-α (TNF-α) are potent inducers of proteolytic activity in chondrocytes and are capable of driving proteoglycan and type II collagen degradation in cartilage explants in vitro, to date there is little evidence from knockout mice or clinical trials that IL-1β, IL-1α, or TNF-α are drivers of cartilage breakdown in vivo.

Senescence in Osteoarthritis

Cellular senescence is a collection of cellular phenotypes that often coexist in a stressed cellular environment.12 Senescence, a well-established hallmark of aging, is a biological state in which cells have lost the ability to divide but remain metabolically active; they produce senescence-associated secretory proteins (SASPs) with pro-inflammatory and cytotoxic functions. In addition to replicative senescence (from telomere shortening), the concept of cellular senescence has now been expanded to include premature senescence (from cell stress). Senolytics are currently being explored as antiarthritics; preliminary evidence suggests that clearing senescence cells and alleviating a cell checkpoint control (through elimination of p21) are promising strategies based on decreased susceptibility to post-traumatic OA and ability to turn a nonhealer mouse strain to a healer for ear wound regeneration, respectively.

Regeneration of Articular Cartilage in Osteoarthritis

OA incidence and progression are due to the net effects of joint breakdown and repair in response to adverse mechanics and micro and macro injury events. Although it was long believed that cartilage had no innate repair capacity, cartilage regeneration is seen in OA patients undergoing procedures in which the abnormal mechanical stress in the joint is corrected or temporarily removed. Examples of this include high tibial osteotomy and joint distraction, a procedure in which the OA joint is held rigid and under tension by an external metal frame secured into the bone above and below the joint. Both procedures demonstrate regrowth of articular cartilage on MRI and sustained clinical improvement. Injury can also trigger a tissue repair response due to release of growth factors from the matrix and recruitment of intra- and extraarticular progenitor cells. The ability to repair is affected by genetics, joint site, depth of cartilage lesion, age, and inflammation (both positively and negatively according to inflammatory phenotype). Just like humans, mouse strains differ in their susceptibility to OA after joint injury. Repair of focal cartilage lesions in the mouse demonstrates both genetic and age dependence. Analyses of a range of selected healer and nonhealer mouse strains demonstrate a range of regeneration phenotypes and correlation of wound healing capacities of ear and knee cartilages; genes representing DNA repair and Wnt signaling pathways are related to regenerative phenotype suggesting the existence of underlying mechanisms for regeneration that significantly affect susceptibility to OA. Direct evidence for an innate cartilage repair capacity is also provided by analyses of the cartilage content of post-translationally modified proteins (indicative of “older” protein forms with long residence time in vivo) relative to their native nonmodified forms (indicative of “younger,” more recently synthesized protein forms in vivo); compared with healthy cartilage, OA ratio of is characterized by a site-specific (ankle > knee > hip) regenerative capacity with a lower older/younger protein forms.13 The clinical importance of endogenous repair in the pathogenesis of OA is also supported by the knowledge that polymorphic variants in GDF5, TGFβ1 (as well as other family members), and FGF18 (and its receptor FGFR3) have all emerged from the recent genome-wide association study described earlier and loss of stress defense mechanisms, such as autophagy, are associated with increased risk for OA.14 This points to the possibility that disease is a failure of repair and regeneration rather than increased degradation. In addition, the mechanisms that drive cartilage repair likely involve injury-induced release of matrix-bound growth factors that contribute to the recruitment and differentiation of local mesenchymal progenitor cells. Delivery of intra-articular fibroblast growth factor 18 (FGF18) to promote cartilage regeneration in OA is in clinical trial and shows early success. Mechanical load has both mechanoadaptive (homeostatic) and pathogenic consequences in the joint. A reduction in load will lead to reversible cartilage thinning or atrophy. Increased load on an otherwise healthy joint will cause an increase in the cartilage volume. By contrast, micro or macro injury (mechanical load that exceeds a defined injury threshold) triggers a cascade of events driven by the cells of the joint (chondrocytes, synoviocytes, osteocytes) to incite an inflammatory sterile tissue injury response, including production and activation of matrix degrading enzymes. Subsequent tissue remodeling leads to weakened cartilage and reduced bone compliance, and this loss of protection means that even physiologic load is perceived as injurious. When the catabolic process overwhelms the endogenous repair response (that differs by joint site), the cycle continues (E-Fig. 246-1).  

CLINICAL MANIFESTATIONS

Clinical Course

OA has traditionally been identified by two major characteristic radiographic alterations of joints; these are joint space narrowing, representing cartilage erosion due to extracellular matrix catabolism, and osteophyte formation, representing an anabolic cartilage response (chondrophyte) that ossifies (Fig. 246-3). These two major radiographic characteristics are often scored using the 1957 Kellgren and Lawrence criteria (see Fig. 246-3 legend) or generally some variation of it. However, as with many other chronic diseases, such as osteoporosis, OA also has a long “silent” phase during which the aneural articular cartilage transitions from a healthy joint tissue to a disease state involving the whole joint organ (see Fig. 246-3) and that is recognizable as abnormalities on a radiograph. Rates of progression vary by joint site. It is therefore probable that progression from subclinical to clinically manifest OA is influenced by mechanical, biologic, developmental, and genetic factors. The reclassification of disease—to one that is no longer a purely radiographic entity, but a disease process that includes molecular, preradiographic, and radiographic stages provides a scenario amenable to the development of primary

CHAPTER 246 Osteoarthritis  

1. Activation of mechanosensitive ion channels

Ca mobilization

Cellular response: 1. Induce inflammatory response genes 2. Control activity of degradative enzymes 3. Initiate tissue repair response 4. Drive cell senescence?

1700.e1

2. Release of growth factors from the pericellular matrix e.g., FGF2, TGFβ

3. Other

Chondrocyte

E-FIGURE 246-1.  Schematic showing mechanisms of activation and cellular response of the chondrocyte to mechanical injury. FGF2 = fibroblast growth factor 2; TGF-β = transforming growth factor-β.

1701

CHAPTER 246 Osteoarthritis  

Normal Joint

Osteoarthritic joint Synovial membrane hyperplasia, inflammation, fibrosis

Synovial membrane Cartilage

Collagen fibrils Calcified cartilage

Subchondral bone

Proteoglycan complexes

A

Leukocyte

Osteophyte

Osteophyte

Roughening and loss of cartilage matrix

Cartilage Thickened subchondral bone

B

Normal

Osteoarthritis Synovial membrane

Synovial membrane

Cartilage Cartilage

Bone

Bone

C

D

E

F

FIGURE 246-3.  Pathologic features of osteoarthritic joint tissues. A, Features of a normal adult synovial joint. Healthy adult articular cartilage is characterized by a smooth surface and extracellular matrix (ECM) composed of a collagen type II fibrillar network and large proteoglycan complexes. The ECM is produced and maintained by the cellular components of cartilage, chondrocytes. The subchondral bone consists of a thin cortical layer and underlying trabecular bone. The synovial membrane lines the joint capsule and attaches at the cartilagebone interface. In the normal state, it consists of a lining layer one or two cells thick, with underlying vascularized loose connective tissue. B, Typical changes to tissues seen in osteoarthritis (OA). Enzymatic activities (ADAMTS4,5 and MMP-13 in particular) cleave proteoglycan and collagen components of the ECM, leading to loss of these molecules from the matrix. As the process advances, the articular cartilage thins and fibrillates, and eventually fissures down to the underlying bone are seen. Simultaneously, a remodeling response in the bone is observed. Thickening of the cortical subchondral bone layer occurs, and new bone growth at the margins appears as osteophytes. The synovial membrane changes observed in OA patients include lining layer hyperplasia, inflammation in the form of leukocyte infiltration, and fibrosis, which can be seen to varying degrees. Photomicrographs of human joint tissues showing these features are depicted in C (normal tissues) and D (OA tissues). (C and D, Courtesy Edward F. DiCarlo, MD, Hospital for Special Surgery, New York, NY.) E, F, Radiographic features of osteoarthritic joint. Patient baseline (E) vs. 3 years later (F) showing typical features of OA progression including the development of medial joint space narrowing and osteophyte (spur outgrowth) at joint margins. The classical Kellgren & Lawrence grading system scores radiographs by 5 categories (scores 0-4) as follows: 0 = no osteoarthritis; 1 = small osteophyte of doubtful significance; 2 = definite osteophyte(s), possible joint space narrowing; 3 = multiple osteophytes, definite joint space narrowing, some subchondral sclerosis and possible deformity of bony ends; 4 = large osteophytes, marked narrowing of joint space, severe sclerosis of subchondral bone and definite deformity of bone ends.

(for pre-illness), secondary (for early disease), and tertiary (for late disease) prevention strategies.

Clinical Symptoms (Illness)

Illness symptoms of OA are protean and include joint pain and/or aching at rest or on use or motion, loss of normal joint motion and function, short duration (5 cm). Shoulder pain, weakness on abduction, and loss of motion occur in varying degrees, ranging from severe pain and mild weakness to no pain and marked weakness. A positive drop-arm sign with inability to maintain actively 90 degrees of passive shoulder abduction may be present in patients with large or massive tears. Small complete tears and incomplete tears of the rotator cuff are treated conservatively with rest, physical therapy, and NSAIDs. Although its role has not yet been established by careful studies, a subacromial injection of a corticosteroid may relieve pain. Surgical repair may be indicated in younger patients. Bicipital tendinitis is manifested by pain, most often in the anterior region of the shoulder and occasionally more diffusely. The pain may be acute but is usually chronic and is related to impingement of the biceps tendon by the acromion. Tenosynovitis of the long head of the biceps is present, and the tendon may be frayed and fibrotic. Palpation over the bicipital groove reveals localized tenderness. The patient’s response should be compared with the response to palpation of the opposite side (i.e., tendon with normal tenderness). Pain may be reproduced over the bicipital tendon in some cases by supination of the forearm against resistance (Yergason sign), shoulder flexion against resistance (Speed test), or extension of the shoulder. Treatment of bicipital tendinitis consists of rest, hot packs, ultrasound, and, as pain subsides, passive and then active ROM exercises. NSAIDs may be helpful, and occasionally a small amount of corticosteroid carefully injected into the tendon sheath may be of benefit. Rupture of the biceps tendon can occur at the superior edge of the bicipital groove, producing a characteristic bulbous enlargement of the lateral half of the muscle belly. Adhesive capsulitis (frozen shoulder) is associated with generalized pain and tenderness and severe loss of active and passive motion in all planes. It is rare before 40 years of age but may occur secondary to any type of shoulder problem. However, not every stiff and painful shoulder is necessarily adhesive capsulitis. Inflammatory arthritis and diabetes can cause adhesive capsulitis. Additional factors such as immobility, low pain threshold, depression, and ,

Olecranon bursitis occurs frequently and involves the subcutaneous olecranon bursa, either secondary to trauma or as an idiopathic condition. The bursa is characteristically swollen and tender on pressure, but pain may be minimal and usually no motion is lost. Aspiration may yield clear or blood-tinged fluid with a low viscosity or grossly hemorrhagic fluid. Inflammatory olecranon bursitis may be caused by gout, RA, or calcium pyrophosphate deposition disease, and infection can also cause a bursitis. Aspiration alone and protection from trauma are usually sufficient to resolve the condition. A small dose of corticosteroid may be injected into the bursa. With septic olecranon bursitis, localized erythema is the major clue. Heat, pain, and a positive culture are also frequently present. Lateral epicondylitis, or tennis elbow, is a common condition in those who overuse their arms.7,8 Localized tenderness directly over or slightly anterior to the lateral epicondyle is the hallmark of this disorder. Pain may occur during handshakes, while lifting a briefcase, or with other similar activities. Probably less than 10% of patients actually acquire lateral epicondylitis through playing tennis. Job and recreational activities, including gardening and athletics, are the usual causes. Pathologically, the condition consists of degeneration of the common extensor tendon, particularly of the extensor carpi radialis brevis tendon. Treatment is aimed at altering activities and preventing overuse of the forearm musculature. Ice packs, heat, and NSAIDs are of some benefit. A forearm brace also can be used. A local corticosteroid injection with a 25-gauge needle over the lateral epicondyle often produces satisfactory initial relief. Isometric strengthening is important as the initial part of a rehabilitation program. Medial epicondylitis, or golfer’s elbow, which mainly involves the flexor carpi radialis, is less common and less disabling than lateral epicondylitis. Local pain and tenderness over the medial epicondyle are present, and resistance to wrist flexion exacerbates the pain.

  DISORDERS OF THE WRIST AND HAND

A ganglion is a cystic swelling that arises from a joint or tendon sheath and occurs most commonly over the dorsum of the wrist. It is synovial lined and contains thick, jelly-like fluid. Ganglia apparently develop secondary to trauma or prolonged wrist extension. Usually, the only symptom is swelling, but occasionally a large ganglion produces discomfort on wrist extension. De Quervain tenosynovitis may result from repetitive activity that involves pinching with the thumb while moving the wrist. The symptoms are pain, tenderness, and occasionally swelling over the radial styloid. Pathologic findings include inflammation and narrowing of the tendon sheath around the abductor pollicis longus and extensor pollicis brevis. A positive Finkelstein test result is usually seen; pain increases when the thumb is folded across the palm and the fingers are flexed over the thumb as the examiner passively deviates the wrist toward the ulnar side. However, this test also may be positive in patients with osteoarthritis (OA) of the first carpometacarpal joint and must be differentiated from this common condition. Treatment involves splinting, local corticosteroid injection (Fig. 247-4), and NSAIDs as indicated. Rarely, surgical removal of the inflamed tenosynovium is needed. Volar flexor tenosynovitis consists of inflammation of the tendon sheaths of the flexor digitorum superficialis and flexor digitorum profundus tendons in the palm. It is extremely common but often unrecognized. Pain in the palm is felt on finger flexion, but in some cases the pain radiates to the proximal interphalangeal (PIP) and metacarpophalangeal (MCP) joints on the dorsal side, misleading the examiner. The diagnosis is made by palpation and identification of localized tenderness and swelling of the volar tendon sheaths. The middle and index fingers are most commonly involved, but the ring and little fingers also can be affected. Often a nodule composed of fibrous tissue

CHAPTER 247  BURSITIS, TENDINITIS, AND OTHER PERIARTICULAR DISORDERS  

FIGURE 247-4.  Injection of de Quervain tenosynovitis.

can be palpated in the palm just proximal to the MCP joint on the volar side. The nodule interferes with the normal tendon gliding and can cause a triggering or locking, which may be intermittent and may produce an uncomfortable sensation. Similar involvement can occur at the flexor tendon of the thumb. The most common cause is overuse trauma of the hands from gripping with increased pull on the flexor tendons. It may be part of inflammatory conditions, such as RA, psoriatic arthritis, or apatite crystal deposition disease. It is seen frequently in conjunction with OA of the hands. Injection of a longacting steroid into the tendon sheath usually relieves the problem, although surgery on the tendon sheath may be needed in unremitting cases. Gamekeeper’s thumb (skier’s thumb) is caused by trauma to the thumb resulting in instability of the first MCP joint. This instability is due to laxity or rupture of the ulnar collateral ligament. It is treated by immobilization, but surgical repair may be necessary. Avulsion of flexor digitorum profundus (jersey finger) may result from trauma, usually in football, when a player grabs onto a jersey. The distal phalanx, usually the fourth, is hyperextended while the digitorum profundus is contracting maximally. The avulsion of the tendon results in an inability to flex the distal phalanx of that digit. Surgery is required to correct the problem.

  DISORDERS OF THE HIP REGION

Although trochanteric bursitis is common, it frequently goes undiagnosed. It occurs predominantly in middle-aged to elderly people, and somewhat more often in women. The main symptom is aching over the trochanteric area and lateral thigh. Walking, various hip movements, and lying on the involved hip may intensify the pain. Onset may be acute, but more often it is gradual, with symptoms lasting for months. In chronic cases, the patient may fail to locate or describe the pain adequately, or the physician may fail to note the symptoms or interpret them correctly. Occasionally, the pain has a pseudoradiculopathic quality, radiating down the lateral aspect of the thigh.9 In a few cases, the pain is so severe that the patient cannot walk and complains of diffuse pain of the entire thigh. The best way to diagnose trochanteric bursitis is to palpate over the trochanteric area and elicit point tenderness. In addition to specific pain on deep pressure over the trochanter, other tender points may be noted throughout the lateral aspect of the thigh muscle. Pain may be worse with external rotation and abduction against resistance. Although bursitis has historically been described as the principal problem, the condition may actually arise at the insertions of the gluteus medius and gluteus minimus tendons. Local trauma and degeneration play a role in the pathogenesis, leading to tendinosis and/or tendon tears. Conditions that may contribute to trochanteric bursitis, apparently by adding stress to the area, include OA of the lumbar spine or of the hip, leg-length discrepancy, and scoliosis. Treatment consists of local injection of depot corticosteroid using a 22-gauge, 3.5-inch needle to ensure that the bursal area is reached (Fig. 247-5). NSAIDs, weight loss, and strengthening and stretching of the gluteus medius muscle and iliotibial band help in management. Coccydynia is manifested by pain in the coccyx area when pressure is applied to the area. This most notably occurs on sitting. The patient squirms from buttock to buttock to relieve the pressure and consequent pain and often chooses to sit on a cushion. The symptoms may be chronic and severe. The condition may relate to a fall on the coccyx, dropping to a hard chair when

1707

FIGURE 247-5.  Injection of trochanteric bursitis.

sitting, or some related trauma to the coccyx. However, at times no obvious cause can be detected. Women are much more frequently affected, perhaps because the lordosis that often occurs in women exposes the coccyx to more trauma. The diagnosis is confirmed by finding localized tenderness over the coccyx on palpation. A plain x-ray film can be obtained to exclude a fracture or dislocation of the coccyx. Treatment with a local injection of 1 mL of a long-acting corticosteroid and 2 mL of a 2% lidocaine solution is usually very effective. The exact nature of the pathology of coccydynia has not been studied, but it is presumed to be a bone bruise. In iliopsoas bursitis, groin and anterior thigh pain are present and worsen on passive hip hyperextension and sometimes on flexion, especially with resistance. Tenderness is palpable over an involved bursa. The patient may hold the hip in flexion and external rotation to eliminate pain and may limp to prevent hyperextension of the hip. The iliopsoas bursa lies behind the iliopsoas muscle, anterior to the hip joint and lateral to the femoral vessels. It communicates with the hip in 15% of cases. The diagnosis is more apparent if a cystic mass is seen (approximately 30% of cases); however, other causes of cystic swelling in the femoral area must first be excluded. A bursal mass can cause femoral venous obstruction or femoral nerve compression. As with most cases of bursitis, acute or recurrent trauma and inflammatory conditions such as RA may lead to iliopsoas bursitis (also called iliopectineal bursitis). Iliopsoas tendinitis may overlap with the bursitis or occur independently in a similar clinical picture. The diagnosis is confirmed by plain x-ray with injection of a contrast medium into the bursa, or by ultrasonography, computed tomography, or MRI. Iliopsoas bursitis/tendinitis usually responds to conservative treatment including physical therapy and corticosteroid injections. With recurrent involvement, excision of the bursa may be necessary. Ischial or ischiogluteal bursitis is caused by trauma or by prolonged sitting on hard surfaces, as evidenced by the name weaver’s bottom. Pain is often exquisite when sitting or lying down. The hamstring muscles originate from the ischial tuberosity, and the ischiogluteal bursa is superficial to the tuberosity. Because the bursa is superficial to the tuberosity, separating the gluteus maximus from the tuberosity, the pain may radiate down the back of the thigh. Point tenderness over the ischial tuberosity is present. Use of cushions, hamstring stretching, and local injection of a corticosteroid are helpful.

  DISORDERS OF THE KNEE REGION

Anserine bursitis is seen predominantly in overweight, middle-aged to elderly women with large legs and OA of the knees. The symptoms are pain and tenderness over the medial aspect of the knee approximately 2 inches below the joint margin, with the pain worsened by climbing stairs. The pes anserinus (Latin for “goose foot”) is composed of the conjoined tendons of the sartorius, gracilis, and semitendinosus muscles. The bursa extends between the above tendons and the tibial collateral ligament. Tendinitis of these tendons, rather than bursitis, is the predominant cause of the syndrome. The diagnosis is made by eliciting exquisite tenderness over the bursal area. Anserine bursitis is often overlooked because it frequently occurs concomitantly with OA of the knee, which, when present, is the assumed cause of pain; however, in some cases of dual involvement, anserine bursitis is the principal source of pain.

1708

CHAPTER 247  BURSITIS, TENDINITIS, AND OTHER PERIARTICULAR DISORDERS  

The treatment is rest, stretching of the adductor and quadriceps muscles, and a corticosteroid injection into the bursa and tendon insertion site. Prepatellar bursitis manifests as a swelling superficial to the kneecap and results from trauma such as frequent kneeling, leading to the name housemaid’s knee. The prepatellar bursa lies anterior to the lower half of the patella and the upper half of the patellar ligament. The pain is generally slight unless pressure is applied directly over the bursa. The infrapatellar bursa, which lies between the patellar ligament and the tibia, is also subject to trauma and swelling. Chronic prepatellar bursitis can be treated by protecting the knee from the irritating trauma. Patellar tendinitis (jumper’s knee) is seen predominantly in athletes engaging in activities such as repetitive running, jumping, or kicking. Pain and tenderness are present over the patellar tendon. Iliotibial band syndrome manifests by lateral knee pain caused by friction between the iliotibial band and the lateral femoral condyle. It is an overuse injury and is seen in runners, cyclists, and other athletes performing repetitive knee flexion activities. Popliteal cysts, also known as Baker cysts, are not uncommon, and the clinician should be well aware of the possibility of their dissection or rupture. A cystic swelling behind the knee with mild or no discomfort can be the only initial finding. With further distention of the cyst, however, a greater awareness and discomfort are experienced, particularly on full flexion or extension. The cyst is best seen when the patient is standing and examined from behind. Any knee disease having a synovial effusion can develop into a popliteal cyst. Popliteal cysts are most common secondary to RA, OA, or internal derangements of the knee. There are a few reported cases secondary to gout and reactive arthritis. A syndrome of pseudothrombophlebitis may occur as a result of cyst dissection into the calf or actual rupture of the cyst. Findings include diffuse swelling of the calf, pain, and sometimes erythema and edema of the ankle. An ultrasound or arthrogram of the knee confirms both the cyst and the possible dissection or rupture.10 A cyst related to an inflammatory arthritis is treated by injection of a depot corticosteroid into the knee joint, and possibly into the cyst itself, which usually resolves the problem. If the cyst results from OA or an internal derangement of the knee, surgical repair of the underlying joint lesion is usually necessary to prevent a recurrence of the cyst. In the knee area, tendon ruptures may occur, and quadriceps tendon rupture is involved approximately 50% of the time; otherwise, patellar tendon rupture occurs. Quadriceps tendon rupture is generally caused by sudden violent contractions of the quadriceps muscle when the knee is flexed. A hemarthrosis of the knee joint may follow. Patients with chronic renal failure, RA, hyperparathyroidism, or gout and patients with SLE taking steroids have been reported to have spontaneous ruptures of the quadriceps tendon. The patient experiences a sudden sharp pain and cannot extend the leg. X-ray studies may show a high-riding patella. The tendon is usually found to be degenerated, and surgical repair is often indicated. Rupture of the patellar tendon has been associated with a specific episode of trauma, repetitive trauma from sporting activities, and systemic diseases. Meniscal tears are common causes of knee “locking” and pain. Physical examination may show pain, with or without clicking, when the hip and knee are bent to 90 degrees. MRI is the diagnostic test of choice. Physical therapy is often as effective as surgery, A5-A7  even for patients who complain of knee catching or occasional locking. A8  Intra-articular injection of mesenchymal stem cells is of limited if any benefit. A9 

  DISORDERS OF THE ANKLE AND FOOT REGION

Achilles tendinitis usually results from trauma, athletic overactivity, or improperly fitting shoes with a stiff heel counter, but it also can be caused by inflammatory conditions such as ankylosing spondylitis, reactive arthritis, gout, RA, and calcium pyrophosphate dihydrate crystal deposition disease.11 Pain, swelling, and tenderness occur over the Achilles tendon at its attachment and in the area proximal to the attachment. Crepitus on motion and pain on dorsiflexion may be present. Management includes NSAIDs, rest, shoe corrections, heel lift, gentle stretching, and sometimes a splint with slight plantar flexion. Local injection of platelet-rich plasma (PRP) has become an increasingly used treatment for releasing growth factors into degenerative tendons12; however, more recent randomized, placebo-controlled trials for treatment of chronic Achilles and other tendinopathies have found PRP injections to be ineffective in improving pain and activity. A10  The Achilles tendon is vulnerable to rupture when involved with tendinitis, and treatment with a corticosteroid injection could increase this possibility. Achilles tendon rupture is well known and occurs with a sudden onset of pain during forced dorsiflexion. An audible snap may be heard, followed by

difficulty in walking and standing on toes. Swelling and edema over the area usually develop. Diagnosis can be made with the Thompson test, in which the patient kneels on a chair with the feet extending over the edge and the examiner squeezes the calf and pushes toward the knee. Normally this produces plantar flexion, but in a ruptured tendon, no plantar flexion occurs. Achilles tendon rupture usually occurs during athletic events or with trauma from jumps or falls. The tendon is more prone to tear in people with preexisting Achilles tendon disease and in those taking corticosteroids. Orthopedic consultation should be obtained, and immobilization or surgery may be selected, depending on the situation. For acute, severe ankle sprain, a below-knee cast or Aircast produces a faster recovery than a tubular compression bandage, but there is no difference in outcomes at 9 months. Plantar fasciitis, which is seen primarily in persons between 40 and 60 years of age, is characterized by pain in the plantar area of the heel. The onset may be gradual, or it may occur with trauma or overuse from some activity, such as athletics, prolonged walking, using improper shoes, or striking the heel with some force. Plantar fasciitis may be idiopathic; it also is likely to be present in younger patients with spondyloarthritis (Chapter 249). The pain characteristically occurs in the morning on arising and is most severe for the first few steps. After an initial improvement, the pain may worsen later in the day, especially after prolonged standing or walking. The pain is burning, aching, and occasionally lancinating. Palpation typically reveals tenderness anteromedially on the medial calcaneal tubercle at the origin of the plantar fascia. Treatment includes relative rest with a reduction in stressful activities, NSAIDs, use of heel pad or heel cup orthosis, arch support, and stretching of the heel cord and plantar fascia. A local corticosteroid injection, using a 25-gauge needle, is often of help. In posterior tibial tendinitis, pain and tenderness occur just posterior to the medial malleolus; it can be caused by trauma, excessive pronation, RA, or spondyloarthropathy. Extension and flexion may be normal, but pain is present on resisted inversion or passive eversion. The discomfort is usually worse after athletic activity, and swelling and localized tenderness may be present. Treatment usually includes rest, NSAIDs, and possibly a local injection of corticosteroid. Immobilization with a splint is sometimes needed. Posterior tibialis tendon rupture, which is not commonly recognized, is a cause of progressive flat foot. It can result from trauma, chronic tendon degeneration, or RA. An insidious onset of pain and tenderness may be noted along the course of the tendon just distal to the medial malleolus, along with swelling medial to the hind foot. The unilateral deformity of hind foot valgus and forefoot abduction is an important finding. The forefoot abduction can best be seen from behind; more toes are seen from this position than would be seen normally. The result of the single heel rise test is positive when the patient is unable to rise onto the ball of the affected foot while the contralateral foot is off the floor. Treatment usually includes rest, NSAIDs, and possibly an orthosis. Surgical repair of the tendon is sometimes indicated. Manifestations of retrocalcaneal bursitis include pain at the back of the heel, tenderness of the area anterior to the Achilles tendon, and pain on dorsiflexion. Local swelling is present, with bulging on the medial and lateral aspects of the tendon. Retrocalcaneal bursitis, also called sub-Achilles bursitis, may coexist with Achilles tendinitis, and distinguishing the two is sometimes difficult. This condition may be secondary to RA, spondylitis, a reactive arthritis, gout, or trauma. Turf toe is an injury of the big toe originally described during play on artificial turf. It results from hyperextension of the first metatarsophalangeal (MTP) joint when a fixed, dorsiflexed foot is forced into the ground. The plantar capsular ligament may be sprained or torn. Stress fracture is also known as march fracture or fatigue fracture because it was first associated with spontaneous fracture after long marches in army recruits. Pain, swelling, tenderness, and occasionally erythema develop over the metatarsal area, usually without any clear history of trauma. On questioning, however, the episode of spontaneous pain related to onset of the fracture can be identified in some cases. The neck of the second metatarsal bone is most frequently involved, but the third metatarsal is also a site of fracture and less commonly seen in the fourth and fifth metatarsals. Aside from prolonged marching, other athletic events with overactivity, including jogging, are common causes. Stress fractures may be seen in patients with RA and in elderly people. The difficulty in diagnosing stress fractures is that the initial x-ray films usually show no abnormalities or, at most, only a faint fracture line. A repeat x-ray examination several weeks later shows healing with callus formation. Bone scans aid the early diagnosis of stress fractures by showing an increase in uptake over the fracture site. Usually these fractures heal spontaneously, and rest and strapping of the foot are helpful. Occasionally, a cast is needed.

  Grade A References A1. Coombes BK, Bisset L, Vicenzino B. Efficacy and safety of corticosteroid injections and other injections for management of tendinopathy: a systematic review of randomised controlled trials. Lancet. 2010;376:1751-1767. A2. Rhon DI, Boyles RB, Cleland JA. One-year outcome of subacromial corticosteroid injection compared with manual physical therapy for the management of the unilateral shoulder impingement syndrome: a pragmatic randomized trial. Ann Intern Med. 2014;161:161-169. A3. Wu YC, Tsai WC, Tu YK, et al. Comparative effectiveness of nonoperative treatments for chronic calcific tendinitis of the shoulder: a systematic review and network meta-analysis of randomized controlled trials. Arch Phys Med Rehabil. 2017;98:1678-1692. A4. Bannuru RR, Flavin NE, Vaysbrot E, et al. High-energy extracorporeal shock-wave therapy for treating chronic calcific tendinitis of the shoulder: a systematic review. Ann Intern Med. 2014;160:542-549. A5. Katz JN, Brophy RH, Chaisson CE, et al. Surgery versus physical therapy for a meniscal tear and osteoarthritis. N Engl J Med. 2013;368:1675-1684. A6. Sihvonen R, Paavola M, Malmivaara A, et al. Arthroscopic partial meniscectomy versus sham surgery for a degenerative meniscal tear. N Engl J Med. 2013;369:2515-2524. A7. Kise NJ, Risberg MA, Stensrud S, et al. Exercise therapy versus arthroscopic partial meniscectomy for degenerative meniscal tear in middle aged patients: randomised controlled trial with two year follow-up. BMJ. 2016;354:i3740. A8. Sihvonen R, Englund M, Turkiewicz A, et al. Mechanical symptoms and arthroscopic partial meniscectomy in patients with degenerative meniscus tear: a secondary analysis of a randomized trial. Ann Intern Med. 2016;164:449-455. A9. Kim SH, Ha CW, Park YB, et al. Intra-articular injection of mesenchymal stem cells for clinical outcomes and cartilage repair in osteoarthritis of the knee: a meta-analysis of randomized controlled trials. Arch Orthop Trauma Surg. 2019;139:971-980. A10. Moraes VY, Lenza M, Tamaoki MJ, et al. Platelet-rich therapies for musculoskeletal soft tissue injuries. Cochrane Database Syst Rev. 2014;4:CD010071.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 247  BURSITIS, TENDINITIS, AND OTHER PERIARTICULAR DISORDERS  

GENERAL REFERENCES 1. Villasenor-Ovies P, Navarro-Zarra JE, Saavedra MA, et al. A survey of anatomical items relevant to the practice of rheumatology: upper extremity, head, neck spine, and general concepts. Clin Rheumatol. 2016;35:3025-3030. 2. Bolon B. Mini-review: toxic tendinopathy. Toxicol Pathol. 2017;45:834-837. 3. Sconfienza LM, Alabano D, Allen G, et al. Clinical indications for musculoskeletal ultrasound updated in 2017 by European Society of Musculoskeletal Radiology (ESSR) consensus. Eur Radiol. 2018;28:5338-5351. 4. Allen GM. The diagnosis and management of shoulder pain. J Ultrason. 2018;18:234-239. 5. Huegel J, Williams AA, Soslowsky LJ. Rotator cuff biology and biomechanics: a review of normal and pathological conditions. Curr Rheumatol Rep. 2015;17:1-9. 6. Whittle S, Buchbinder R. In the clinic. Rotator cuff disease. Ann Intern Med. 2015;162: ITC1-ITC15.

1709.e1

7. Shillito M, Soong M, Martin N. Radiographic and clinical analysis of lateral epicondylitis. J Hand Surg Am. 2017;42:436-442. 8. Kani K, Porrino J, Dahiya N, et al. Visualization of the soft tissues at the lateral and medial epicondyles of the elbow. PMR. 2017;533-539. 9. Zibis AH, Mitrousias VD, Klontzas ME, et al. Great trochanter bursitis vs sciatica, a diagnosticanatomic trap: differential diagnosis and brief review of the literature. Eur Spine J. 2018;27: 1509-1516. 10. Shah A, James SL, Davies AM, et al. A diagnostic approach to popliteal fossa masses. Clin Radiol. 2017;72:323-337. 11. van der Vlist AC, Breda SJ, Oei EHG, et al. Clinical risk factors for Achilles tendinopathy: a systematic review. Br J Sports Med. 2019. [Epub ahead of print.] 12. Le ADK, Enweze L, DeBaun MR, et al. Platelet-rich plasma. Clin Sports Med. 2019;38:17-44.

1709.e2

CHAPTER 247  BURSITIS, TENDINITIS, AND OTHER PERIARTICULAR DISORDERS  

REVIEW QUESTIONS 1. A 78-year-old woman who has had periodic discomfort in her neck and radiating to both shoulders for several years had a fall, landing on her outstretched right arm. She had pain in the right shoulder and weakness on use. At the urgent care center she was found to have some pain on passive motion and was unable to actively abduct the right shoulder. She had a positive drop arm sign of the right shoulder but not of the left one. All upper extremity deep tendon reflexes were normal. Which of the following is the most likely diagnosis? A . Rotator cuff tendinitis is the main diagnosis. B. X-ray examination of the humerus of the injured side is not indicated in this case because tendon injuries are not visualized on plain x-ray film of the shoulder. C. Bicipital tendinitis is the likely diagnosis. D. A tear of the rotator cuff best fits as the diagnosis. E. Cervical spine disc herniation is the most likely diagnosis based on the history of neck pain and arm weakness. Answer: D  Acute trauma is often the cause of rotator cuff tears, although most tears occur gradually through chronic tendinitis. The positive drop arm sign is typical of a significant rotator cuff tear. Sometimes with trauma, a humeral fracture can occur along with the rotator cuff tear. Pain on abduction and especially on resisted abduction are seen in rotator cuff tendinitis; also, the drop arm sign is typically negative. Bicipital tendinitis usually does not cause shoulder weakness. The clinical picture does not fit a cervical disc herniation, and the reflexes are normal. (Biundo JJ. Regional Rheumatic Pain Syndromes, Primer on the Rheumatic Diseases, 13th ed. New York: Springer; 2006.) 2. A 51-year-old male insurance agent, whose hobby is wood working, complains of right hand pain of approximately 4 months’ duration. He had been doing more wood working during that period until the hand pain began to limit this activity. On examination there were no Heberden nodes of the DIP joints, and the PIP and MCP were not tender or swollen. Moderate-to-severe tenderness on palpation was present over the palm side of the hand at the region of the second and third metacarpal bones. Some swelling was detected at these sites. Crepitus was noted on palpation on flexion of these two digits. Which of the following is the most likely diagnosis? A . Because rheumatoid arthritis is a likely diagnosis, you should obtain RF, ANA, and hand x-ray films. B. Because of his age, he most likely has osteoarthritis of the hand. C. He fits the picture of volar flexor tenosynovitis. D. A C7 radiculopathy would best explain pain in this part of his hand. E. Overuse of his hands has caused carpal tunnel syndrome. Answer: C  Volar flexor tenosynovitis is one of the most common musculoskeletal syndromes and is frequently overlooked. It is manifest by pain in the palm of the hand. The volar flexor tendon sheath is tender on palpation and thickened. Triggering or snapping of the digit sometimes occurs, especially when more chronic. Overuse of the hand is the usual cause, but is also seen in association with OA of the hands and diabetes. Stretching of the involved digit and an injection of a small amount of a corticosteroid into the tendon sheath is usually helpful. (Biundo JJ. Regional Rheumatic Pain Syndromes, Primer on the Rheumatic Diseases, 13th ed. New York: Springer; 2006.)

3. A 48-year-old male professional cellist, whose hobby is gardening, has had a 4-month history of right shoulder pain. Pain occurs on various movements of the shoulder, and there is some limitation of movement. Night pain in bed is also reported. On examination, pain limits active abduction to 90 degrees. Less pain is noted on passive abduction, and 180 degrees of abduction is present. The drop arm sign is negative, but pain occurs on active internal rotation of the shoulder. There is no crepitus. Which of the following is the most likely diagnosis? A . Rotator cuff tendinitis is the most likely diagnosis. B. Rotator cuff tear best fits this clinical picture. C. Rupture of the biceps tendon is the diagnosis because the drop arm sign is negative. D. Adhesive capsulitis (frozen shoulder) is the diagnosis. E. OA of the shoulder (glenohumeral joint) is the likely diagnosis, and a plain x-ray film of the shoulder is needed to confirm the diagnosis. Answer: A  The typical physical examination findings of rotator cuff tendinitis are pain on active abduction, less pain on passive abduction, and more pain on resistive abduction. Loss of motion on active movements can occur when more severe. Also, pain may occur on active internal and on external rotation. In addition to the physical examination, ultrasonography and MRI can detect tendinitis, but not plain x-ray examination. In adhesive capsulitis restriction, the range of motion in all directions is usually seen. The full range on abduction and other motions in this case are not typical of a frozen shoulder. OA of the shoulder would not have had a pain history of only 4 months, and it is a much less frequent entity than rotator cuff tendinitis. Crepitus on range of motion is very common in OA of the glenohumeral joint. (Biundo JJ. Regional Rheumatic Pain Syndromes, Primer on the Rheumatic Diseases, 13th ed. New York: Springer; 2006.) 4. A 76-year-old woman with a known history of moderate OA of both knees and a partially torn right medial meniscus experienced pain in her right knee after stooping down to clean something off the floor. The next day she had discomfort and fullness behind her right knee and swelling of her lower leg and ankle. She was able to walk but had a limp. She was seen that day by her primary care physician for evaluation. Which of the following is the best course of action? A . Begin treatment with heparin or enoxaparin for a deep vein thrombosis. B. Give oral prednisone for 5 days as a tapering dose for an acute flare of the OA of her knee. C. Obtain a new x-ray film of the knees, with a standing AP and lateral view of the involved leg. D. Because of the acute flare of knee symptoms, give colchicines for a possible acute attack of gout. E. Obtain a Doppler ultrasound of the right leg for a deep vein thrombosis and an ultrasound image of the right knee to identify a Baker cyst that has dissected, causing a pseudothrombophlebitis picture. Answer: E  Popliteal cysts, also known as Baker cysts, are often asymptomatic. However, at times they may enlarge as a result of increasing pressure from a synovial effusion of the knee, flowing through a one-way valve from the knee to the popliteal cyst. The cyst can at times dissect downward, causing swelling of the leg and simulating thrombophlebitis. Thus the name given to the syndrome is pseudothrombophlebitis. The swelling of the leg may resolve on its own after several days. Often, a corticosteroid injection into the knee joint may help by decreasing the inflammation and synovial effusion of the knee. OA of the knee should not produce swelling of the lower leg, nor should gout. Treatment with anticoagulants should not be given unless the diagnosis of DVT is confirmed. (Biundo JJ. Regional Rheumatic Pain Syndromes, Primer on the Rheumatic Diseases, 13th ed. New York: Springer; 2006.)

CHAPTER 247  BURSITIS, TENDINITIS, AND OTHER PERIARTICULAR DISORDERS  

5. A 52-year-old moderately obese man complains of pain in his right foot that has been present for approximately 4 months. The pain has become worse recently, and he has had to stop the walks he was taking for exercise. The pain is described to be more in the rear foot and is particularly bad when he first gets out of bed in the morning. He has no history of gout or other arthritis except for some back pain. On examination there is no joint swelling. He does have pes cavus, and tenderness of the plantar surface of the rear foot. His calf muscles are also tight. What is the most likely diagnosis and/or the best course of evaluation and treatment? A . Order an MRI of the foot or technetium bone scan for possible stress fracture. B. Inject the calcaneus with a steroid, because that is the best initial treatment for heel pain. C. Make a tentative diagnosis of plantar fasciitis and treat with analgesics, stretching of calf and plantar fascia, and use of a heel cushion as the initial treatment. D. Obtain uric acid level and start treatment with colchicine. E. Obtain orthopedic consultation for possible Achilles tendon tear.

1709.e3

Answer: C  The complaint of morning foot pain, which is called “first step pain,” is typical of plantar fasciitis. Tenderness often elicited on pressure exerted on the plantar surface of the calcaneus. Trauma to the foot from prolonged walking, tight calf muscles, high arched foot with tight plantar fascia, and obesity are factors related to acquiring plantar fasciitis. A steroid injection of the calcaneal area is often helpful but is not the first line of treatment. With Achilles tendinitis/tear the pain is in the posterior heel and not the plantar surface. A stress fracture is more likely to occur in the metatarsal area and only very rarely occur in the calcaneus. (Biundo JJ. Regional Rheumatic Pain Syndromes, Primer on the Rheumatic Diseases, 13th ed. New York: Springer; 2006.)

CHAPTER 248  Rheumatoid Arthritis  

1709

increases with each decade. The incidence of rheumatoid arthritis may be decreasing though the reasons for this are unclear. Geographic variations in prevalence and phenotype can be remarkable. Most notably, cohorts have been described in rural Nigeria in which no individuals are affected with rheumatoid arthritis (though case ascertainment bias needs to be excluded); in contrast, a prevalence of 5% has been found in some studies of Chippewa, Yakima, and Inuit Native American tribes. Such extreme phenotypes are as yet poorly understood but likely will reflect the impact of genetics, the gastrointestinal or respiratory microbiome, or other environmental influences. Detailed study of such populations is now generating significant insight into pathogenesis.  

PATHOBIOLOGY

Genetics

248  RHEUMATOID ARTHRITIS IAIN MCINNES AND JAMES R. O’DELL



DEFINITION

Rheumatoid arthritis is a chronic systemic inflammatory disease of unknown etiology that primarily targets synovial tissues. It is relatively common, with a prevalence of slightly less than 1% in adults all over the world. Rheumatoid arthritis shortens survival and significantly affects quality of life in many patients.1 Essentially all patients exhibit some systemic features such as fatigue, low-grade fevers, anemia, and elevations of acute phase reactants (erythrocyte sedimentation rate [ESR] or C-reactive protein [CRP]). This systemic inflammation is believed to be responsible for a variety of coexistent comorbid conditions including cardiovascular disease, osteoporosis, cognitive dysfunction and psychiatric disease, and cancer.2 However, the primary target of rheumatoid arthritis is the joint. Synovial tissues proliferate in an uncontrolled fashion, resulting in excess fluid production, destruction of cartilage, erosion of marginal bone, and mechanical disruption of the tendons and ligaments. Such damage predicates long-term disability and increased mortality. In the past two decades, the treatment of rheumatoid arthritis has changed dramatically. Current therapeutic strategies should result in over 50% of patients achieving clinical remissions with treatment with appropriate disease-modifying antirheumatic drugs (DMARD), biologic DMARDs, and targeted synthetic DMARDs used either as monotherapy or in combination and according to rigorous treat-to-target principles.  

EPIDEMIOLOGY

Rheumatoid arthritis is a global disease with a variable geographic prevalence of 0.5 to 1% of adults. For reasons that are still unclear, the prevalence in women is around three times greater than that in men prior to the menopause. Rheumatoid arthritis can occur at any age. Overall, the annualized incidence of rheumatoid arthritis is approximately 40 per 100,000 for women and about half that for men. Because rheumatoid arthritis is a lifelong disease and its incidence increases or is stable with age, the prevalence of rheumatoid arthritis

A combination of candidate association, twin, and genome-wide association studies (GWAS) has established a strong genetic component to the risk for developing rheumatoid arthritis and the severity of the disease.3 Twin studies reveal a concordance rate of approximately 12 to 15% for monozygotic twins and 2 to 5% for dizygotic twins. Heritability is reported to be around 60%, with lower values for patients who are seronegative. GWAS in European, North American, and recently Asian populations, combined with subsequent meta-genome analyses, demonstrate that rheumatoid arthritis is a polygenic disorder. The majority of more than 100 thus far identified informative single nucleotide polymorphisms (SNPs) implicate immune genes, indicative of a primary immunologic etiology of disease a priori. Around 40% of genetic risk accrues in the HLA region. Certain HLA-DR alleles (e.g., DRB*0401, DRB*0404, DRB*0101, DRB*1402) associate with an increased risk for developing rheumatoid arthritis and thereafter progressing to more severe disease. Hypervariable regions on DR molecules are particularly important for antigen recognition by binding to antigenic peptides and presenting them to the T-cell receptor (TCR). The amino acid sequence of disease associated DRβ1 chains share a common structural motif, called the shared epitope (E-Table 248-1). Peptides derived from post-translationally modified proteins (e.g., via citrullination, acetylation, carbamylation) may bind with altered avidity to the shared epitope, providing a potential mechanism whereby this genetic factor can mediate disease risk at the molecular level. Thus, altered binding of such peptides to the HLA molecule (expressed on dendritic cells or B cells) may alter the dynamics of HLA-TCR interactions and thereby confer loss of self-tolerance upon (self-reactive) T cells inappropriately activated by this abnormal receptor engagement. Many other immune related genes are now implicated and add considerable depth to our understanding of pathogenetic pathways. Although individual SNPs usually exhibit modest risk contribution (odds ratios ≈1.05 to 1.2-fold) they nevertheless could confer significant functional impact. Risk genes can be usefully defined on the basis of their contribution to functional immune system compartments. Adaptive immunity associated SNPs include a functional polymorphism in protein tyrosine phosphatase nonreceptor 22 (PTPN22) that has been reproducibly associated with rheumatoid arthritis and a number of other autoimmune diseases, including type 1 diabetes, systemic lupus erythematosus, Graves disease, and Hashimoto thyroiditis. The co-stimulation receptors, CTLA, CD28, CD40, and signal protein kinase TYK2 have similarly been associated with disease. These could alter the maintenance of central thymic or peripheral tolerance mechanisms, or could confer altered quantitative and functional outcomes from interactions between T cells, B cells, and other antigen presenting cells with increased potential for autoimmunity to emerge. Recently, an association with peptidyl arginase deiminase has been recognized. Because this enzyme converts arginine to citrulline, this raises the possibility of a genetically predisposed risk conferred via increased generation of a major rheumatoid arthritis autoantigen. Pathways that regulate innate immune pathways are also identified (e.g., TRAF1-C5, STAT4, TNF-AIP3, IRAK1) together with pathways concerned with cell migration (ELMO1) and fetal development (LBH). Several SNPs identify cytokine or cytokine receptor coding loci, including TNF, IL-6R. Finally, there is increased interest in the role of epigenetics in disease risk and especially propagation. Thus, altered DNA methylation, histone modification (e.g., via acetylation), and microRNA expression patterns have emerged that associate with increased or perpetuation of inflammation. These have been best defined in fibroblast-like synoviocytes (e.g., differential methylation and thus gene expression has been found particularly for pathways related to cell growth and differentiation). Recently the presence of differentially expressed microRNAs has also attracted attention in myeloid lineage cells, T cells, and

1709.e4

ABSTRACT

CHAPTER 248  Rheumatoid Arthritis  

Rheumatoid arthritis is a chronic inflammatory disease affecting primarily the joints, but associated with substantial comorbidities that include cardiovascular, psychologic, oncologic, and bone disease. The underlying pathogenesis of rheumatoid arthritis is incompletely understood but comprises an early breach of tolerance to a range of post-translationally modifications, typically citrullinated proteins, leading to the detection of autoantibodies, cytokines, and chemokines, and dysregulated metabolism up to 10 years prior to clinically detectable disease onset (pre-RA). The primary target tissue is the inflamed synovium that contains a range of activated leukocytes and stromal cells, that under the influence of inflammatory cytokines drive local cartilage and bone destruction. Therapeutics include the sequential administration of disease modifying antirheumatic drugs (DMARDs), initially including methotrexate, escalating thereafter to the use of DMARDS, biologic agents (including TNF and IL-6 receptor inhibitors), rituximab, abatacept, and most recently Janus kinase inhibitors. Parallel use of NSAIDs, and judicious use of corticosteroids, is also common practice. Therapeutics should be adjusted based on formally assessed, intercurrent disease activity levels, to seek the lowest level of disease activity or remission, whichever is more appropriate for a given patient. Longterm outcomes have improved remarkably in the last decade though a smaller group of patients with difficult to treat disease remains, defining an ongoing significant unmet clinical need.

KEYWORDS

rheumatoid arthritis immune therapeutics synovitis comorbidity disease modifying anti-rheumatic drug biologic kinase inhibitor autoimmunity

CHAPTER 248  Rheumatoid Arthritis  

1709.e5

E-TABLE 248-1 HUMAN LEUKOCYTE ANTIGEN ASSOCIATIONS WITH RHEUMATOID ARTHRITIS THIRD HYPERVARIABLE REGION AMINO ACID SEQUENCES

HLA TYPES (ALLELES) AND METHODS OF DETECTION ALLOANTISERA (DR)

MLC (DW)

DNA (DRβ1)

70

71

72

73

74

MOST COMMON ETHNIC GROUPS

DR4

Dw4

*0401

Q

K

R

A

A

Whites (Western Europe)

DR4

Dw14

*0404

R

Whites (Western Europe)

DR4

Dw15

*0405

R

Japanese, Chinese

DR1

Dw1

*0101

R

Asian Indians, Israelis

DR6 (14)

Dw16

*1402

R

Yakima Native Americans

DR10



*1001

R

R

Spanish, Greeks, Israelis

DR4

Dw10

*0402

D

E

Whites (Eastern Europe)

DR4

Dw13

*0403

DR2

Dw2

*1501

D

A

DR3

Dw3

*0301

ASSOCIATED WITH RA

NOT ASSOCIATED WITH RA R

E

Polynesians Whites

G

R

Whites

A = alanine; D = aspartic acid; E = glutamic acid; HLA = human leukocyte antigen; K = lysine; MLC = mixed leukocyte cultures; Q = glutamine; R = arginine; * = the same amino acid in that position as for DRB1*0401.

1710

CHAPTER 248  Rheumatoid Arthritis  

Genetic background Some HLA-DRBI alleles Other HLA alleles Non-HLA alleles

RA initiated

Trigger ? Bacterial antigens ? Viral antigens ? Smoking ? Others

Ongoing autoimmune response Synovial proliferation and joint destruction FIGURE 248-1.  Initiation of rheumatoid arthritis (RA). HLA = human leukocyte antigen.

fibroblast-like synoviocytes. For example, dysregulation of miR146 and miR155 promotes cytokine and inflammatory pathways in macrophages.

Etiology

Increasing evidence supports a substantial role for environmental factors including smoking, other pulmonary exposures (e.g., silica), obesity, vitamin D deficiency, and lower educational attainment as enhancing factors for rheumatoid arthritis risk. Low levels of alcohol intake may be protective (Fig. 248-1). Smoking in particular has long been associated with a significant increase in the risk for developing rheumatoid arthritis, but this is true only for anticitrullinated peptide antibody (ACPA)-positive disease, especially those who have the shared epitope. Induction of epigenetic changes by smoking provides the most likely explanation for this etiologic link. Infectious triggers are also implicated, such as Mycobacteria, Streptococcus, Mycoplasma, Escherichia coli, Helicobacter pylori, and viruses (rubella, Epstein-Barr, parvovirus). The mechanisms whereby infections drive disease are poorly understood. Rheumatic fever (Chapter 274) and reactive arthritis (Chapter 249) are prime examples. Reactive arthritis in particular may occur following one of a myriad of different but specific infectious triggers presented to a specific location in the body (e.g., the gastrointestinal or genitourinary tract) on a predisposed genetic background, in most cases HLA-B27. In this syndrome, age and gender and hence immune system maturity may be critical in the development of clinical disease, which occurs primarily between the ages of 15 and 40 years in males. More recently, Lyme arthritis (Chapter 305) is an example of an arthritic syndrome for which an infectious trigger has clearly been demonstrated. Many other examples exist in animal models of arthritis, including arthritis induced by mycobacteria and streptococci. Currently most interest resides in the role of the microbiome in induction of rheumatoid arthritis. Studies of the gastrointestinal and oral microbiome (the latter particularly in periodontal disease) have identified dysbiosis associated with early rheumatoid arthritis. Specific microbes (e.g., P. gingivalis, P. coprie, A. actinomycetemcomitans) have all been implicated. Whether this operates via molecular mimicry, altered immune homeostasis (the GI tract is a vital area for maintenance of peripheral immunologic tolerance) or direct stimulation of immune function (e.g., via enhanced citrullination) remains under investigation. Finally, the use of oral contraceptives has been associated with a decrease in the incidence of rheumatoid arthritis. Because the effect appears to be strongest for oral contraceptives that have high estrogen content, it is postulated that estrogen is responsible for this protective effect. Studies that have tried to address the question of postmenopausal estrogen use and its effect on rheumatoid arthritis have yielded conflicting results.

Pathogenesis and Tissue Pathology

The overarching pathogenesis of rheumatoid arthritis remains unknown. Taking the foregoing together, the best current model is as follows. The earliest signs of inflammation often occur in the lung: abnormal high-resolution CT scans of pre-RA patients have been reported. Pulmonary mucosal citrullination is increased, perhaps reflecting smoking or other pulmonary irritants. Increased

local inflammation is observed, supporting initial B-cell and then T-cell activation, with consequent “breach of tolerance” to post-translationally modified self proteins. Similar events in the oral mucosa, in periodontal disease, or conceivably in the gastrointestinal tract may also contribute in some patient subgroups. The physico-chemical properties of these self-protein derived peptides that confer increased immune activation presumably arise within the genetic architecture described above and especially the structure of the shared epitope. A prearticular clinical phase (“Pre-RA”) ensues that may last up to several years, during which rheumatoid factor and autoantibodies against post-translationally modified proteins (AMPA), particularly anti-citrullinated peptide antibodies (ACPA), rise in titre and with increasingly broad specificity (epitope spreading). ACPAs, for example, can recognize citrullinated residues on a variety of self-proteins (e.g., type II collagen, vimentin, α-enolase, fibronectin, fibrinogen, and histones). The abnormalities thus appear to represent a broad failure to regulate immune homeostasis, rather than one individual (auto)antigen driving disease overall. One autoantigen may, however, serve as a trigger. In parallel, elevated levels of serum cytokines and chemokines are detected, indicative of generalized increased levels of systemic inflammation, along with lipid dysmetabolism. Biopsies of synovial tissue during this pre-RA phase are essentially normal, consistent with its systemic nature. Thereafter, a transitional event occurs that heralds the onset of clinically detectable arthritis. Systemic disease becomes localized to the joint. This process is ill-defined.4 ACPA specifics are broadest (they expand during pre-RA), and cytokine concentrations are highest immediately prior to disease onset (imminent RA). Early changes in synovial vascularization, deposition of immune complexes, altered neurologic supply, local infection, and microtrauma have all been postulated as articular localizing mechanisms. An elegant recent notion has ACPA specific for citrullinated vimentin binding to and activating osteoclasts, in turn leading to local bone erosion and pain and the release of chemokines, particularly IL-8. The latter may initiate synovial cellular recruitment. Established rheumatoid arthritis is associated with development of a dense synovial cellular infiltrate that has some degree of organization (Fig. 248-2). A lining layer forms of 4 to 8 cells in depth including macrophages and fibroblast-like synoviocytes (FLS). The interstitial area contains large numbers of T cells, B cells, plasma cells, mast cells, and FLS. In some patients, lymphocytic aggregates form (ectopic germinal centers) that confer a poorer clinical prognosis. These likely serve as a source of autoantibody production. Intriguing prospective biopsy studies have identified discrete synovial appearances across patient subgroups; myeloid, lymphocytic, and fibroblastic patho­ types have been described at the transcriptional and histologic level. These pathotypes may represent clinically discrete endotypes that in the future could define useful biomarkers to determine optimal choices of immune targeted therapies. The synovial T-cell response is primarily of TH1 and TH17 type. T cells are activated via antigen presented in a CD28/CD80/86 dependent manner by macrophages, B cells, or FLS, by the local cytokine milieu (e.g., IL-7, IL-15, TNF, IL-6), or by cognate cellular activation through cell contact with macrophages. Abatacept that blocks costimulation mediates its effects via interference with some of these pathways. The T cells secrete cytokines (e.g., IL-17 and GM-CSF) that drive further synovial proliferation. The predominant source of synovial cytokines is, however, macrophages, mast cells, and FLS. Macrophage-derived cytokines, particularly interleukin-1 (IL-1), IL-6, and tumor necrosis factor-α (TNF-α), play central roles in this ongoing inflammatory process. TNF and IL-6 in particular appear to have hierarchical functional dominance, reflected in successful therapeutic targeting (e.g., TNF inhibitors, tocilizumab, and JAK inhibitors) and commensurate with their functional profiles that include leukocyte activation, endothelial activation, angiogenesis, or osteoclast activation. The humoral immune system also plays a role, reflected in the success of rituximab therapy. Rheumatoid factor (RF) and ACPA presence correlate with more severe disease, including erosions of bone, and with the presence of extra-articular features. RF and ACPAs likely have a pathologic role. Via immune complex formation, or perhaps functioning in isolation, they increase complement activation, promote macrophage activation to release lysosomal enzymes, kinins, prostanoids, and oxygen/nitrogen free radicals via Fc receptor (FcR) binding and may activate osteoclasts. ACPAs also activate macrophages via TLR/FcR cross-talk mediated via citrullinated fibronectin. Articular damage is driven primarily by FLS and osteoclasts. FLS are partially transformed cells, with a distinct epigenetic profile, that exhibit anchorage independence, loss of contact inhibition, low-grade proliferation, and TLR expression, rendering them immunologically competent to sense tissue damage. They release prostanoids, cytokines, chemokines, and matrix metalloproteinases

CHAPTER 248  Rheumatoid Arthritis  

Inhibition of proteoglycan synthesis (IL-1)

Bone Joint capsule

Bone resorption (IL-1) Collagenase and PGE2 secreted by synoviocytes (induced by IL-1 and TNF-α)

Cartilage Normal synovium Synoviocyte hyperplasia

Vascular injury Influx of immune cells and plasma Initiation

1711

M

Cytokines C T TNF-α, IL-1, IL-6, IL-8, GM-CSF B P

Prostaglandin Iysosomal enzymes O2 radicals

Rheumatoid pannus

IgM=RF IgG Phagocytosis Cellular reactivation and of immune proliferation complex by cytokines

Immune Responses Cellular Humoral

Neutrophils attracted and activated (GM-CSF, IL-8, TNF-α, TGF-β) Destruction

Inflammation

FIGURE 248-2.  Events involved in the pathogenesis of rheumatoid synovitis (progressing from left to right). B = B lymphocyte; C = complement; GM-CSF = granulocyte-macrophage colony-stimulating factor; IgG, IgM = immunoglobulin G, M; IL = interleukin; M = macrophage; P = plasma cell; PGE2 = prostaglandin E2; RF = rheumatoid factor; T = T lymphocyte; TGF-β = transforming growth factor-β; TNF-α = tumor necrosis factor-α.

(MMPs) such as MMP1, MMP3, and MMP13, in disproportion to the release of tissue inhibitors of MMPs (TIMPs). In consequence, FLS promote cartilage damage; this local tissue damage area is known as the cartilage-pannus junction. Macrophages and mast cells are also likely to contribute to this local process. Bone damage requires cells with capacity to acidify the local milieu—osteoclast maturation and activation is a localized feature of rheumatoid arthritis synovium, arising as a consequence of RANKL, IL-1, TNF, and IL-17 activity. Thus activated osteoclasts localize in periarticular bone and in the adjacent bone marrow, leading to the characteristic erosions detected on plain radiography. The inflammatory process underlying erosion is detected as “bone edema” on MRI. Synovitis thus results in the destruction of cartilage in marginal bone and in the stretching or rupture of the joint capsule or tendons and ligaments. Damage is directly related to disability. Finally, generalized inflammation promotes systemic comorbidities. Circulating cytokines and immune complexes are proposed to activate endothelium and accelerate atherosclerosis, may provoke systemic osteoporosis, and can drive cognitive impairment, fatigue, and frank psychiatric presentations (e.g., depression). Inhibition of such processes should reduce comorbidities in the clinic.  

Rheumatoid Arthritis

Osteoarthritis

CLINICAL MANIFESTATIONS

Articular Manifestations

Rheumatoid arthritis can affect any of the synovial (diarthrodial) joints (Fig. 248-3). Most commonly, clinically apparent disease starts in the metacarpophalangeal (MCP), proximal interphalangeal (PIP), and metatarsophalangeal (MTP) joints, followed by the wrists, knees, elbows, ankles, hips, and shoulders, in roughly that order. Early treatment limits the joints involved. Less commonly, and usually later, rheumatoid arthritis may involve the temporomandibular, cricoarytenoid, and sternoclavicular joints. Rheumatoid arthritis may involve the upper part of the cervical spine, particularly the C1-C2 articulation (E-Fig. 248-1), but, unlike the spondyloarthropathies (Chapter 249), it rarely involves the rest of the spine. Patients are also at an increased risk for osteoporosis (Chapter 230), and this risk should be considered and dealt with early.

Hands

The hands are a major site of involvement; a significant proportion of the disability caused by rheumatoid arthritis is due to damage and dysfunction of the hands. Typically disease starts with swelling of the PIPs and MCPs. The distal interphalangeal (DIP) joints are almost never involved; significant involvement of the DIP joints should suggest the possibility of a different diagnosis (i.e., osteoarthritis or psoriatic arthritis). Figure 248-4 illustrates the classic ulnar deviation of the MCP joints and swan-neck deformities (hyperextension of the PIP joints) that are commonly seen in late disease. Boutonnière (or buttonhole) deformities also occur as a result of hyperflexion of the PIP joints. If the clinical disease remains active, hand function deteriorates. Sudden loss of function of individual fingers may occur as a result of tendon rupture, which requires the expertise of a hand surgeon to repair.

FIGURE 248-3.  Distribution of involved joints in the two most common forms of arthritis: rheumatoid arthritis and osteoarthritis. Black circles are shown over the involved joint areas.

Feet

Feet, particularly the MTP joints, are involved early in most patients. Radiographic erosions occur at least as early in the feet as in the hands. Subluxation of the toes is common and leads to the dual problem of breakdown of the skin and ulcers on the top of the toes and malalignment of the MTP heads. Painful ambulation develops owing to dislocation of the cushioning pads that usually protect the heads of the MTP joints.

Wrists

The wrist joints are involved in most patients with rheumatoid arthritis. Radial deviation is the rule, and patients with severe involvement may progress to volar subluxation. Even early in the course of the disease, synovial proliferation in and around the wrists may compress the median nerve, causing carpal tunnel syndrome (Fig. 248-5). Later, this synovial proliferation may invade tendons and lead to rupture, most commonly of extensor tendons.

Large Joints

Involvement of knees, ankles, elbows, hips, and shoulders is common. Characteristically, the whole joint surface is involved in a symmetrical fashion.

CHAPTER 248  Rheumatoid Arthritis  

A

1711.e1

B

E-FIGURE 248-1.  Subluxation of the cervical spine in patients with rheumatoid arthritis. A, In a lateral radiograph of the cervical spine, the body of C2 and its odontoid process are outlined by the broken lines and the posterior aspect of the anterior segment of C1 is indicated by a solid line. Normally, a space of only 2 to 3 mm separates C1 from C2. The space between C1 and the odontoid of C2 is markedly increased, indicative of subluxation of C1 and C2. B, Lateral view of a pathologic specimen from a patient who died of C1-C2 subluxation. The horizontal arrow shows the odontoid process that subluxed posteriorly, severely compressing and almost severing the cord. The vertical arrow shows a bone graft that had been put in place posteriorly in an attempt to prevent subluxation. Below the arrow, a nonhealing area is present through the bone graft, and inferior to that a wire fixation suture is still in place.

1712

CHAPTER 248  Rheumatoid Arthritis  

FIGURE 248-4.  Severe advanced rheumatoid arthritis of the hands. There is massive tendon swelling over the dorsal surface of both wrists, severe muscle wasting, ulnar deviation of the metacarpophalangeal joints, and swan-neck deformity of the fingers. (From Forbes CD, Jackson WF. Color Atlas and Text of Clinical Medicine, 3rd ed. London: Mosby; 2003.)

A

B

FIGURE 248-6.  Radiographs of the knees in the two most common forms of arthritis: rheumatoid arthritis and osteoarthritis. A, Severe involvement in rheumatoid arthritis, with almost complete symmetrical loss of joint space in both the medial and the lateral compartments, but with little subchondral sclerosis or osteophyte formation. B, Typical osteoarthritis, with severe, near-total loss of joint space of one compartment and a normal or actually increased joint space of the other compartment. Note also the significant subchondral sclerosis in the involved area, typical of osteoarthritis.

Neck

Although most of the axial skeleton is spared in rheumatoid arthritis, the cervical spine is commonly involved, particularly the C1-C2 articulation. Bony erosions and ligament damage can occur in this area and may lead to subluxation (see E-Fig. 248-1). Most often, subluxation at C1-C2 is minor and without accompanying symptoms; patients and caregivers need only be cautious and avoid actively forcing the neck into positions of flexion. Occasionally, subluxation at C1-C2 is severe and leads to compromise of the cervical cord with neurologic symptoms and in some cases death. If surgery is planned for a patient with long-standing rheumatoid arthritis, preoperative flexion and extension films of the cervical spine should be obtained looking for significant subluxation. Median nerve in carpal tunnel Tapping produces paresthesias in the shaded area (Tinel sign) FIGURE 248-5.  Carpal tunnel syndrome. Distribution of pain and/or paresthesias (shaded area) when the median nerve is compressed by swelling in the wrist (carpal tunnel).

Other Joints

Wherever synovial tissue exists, rheumatoid arthritis can cause problems. The temporomandibular, cricoarytenoid, and sternoclavicular joints are examples of other joints that may be involved. The cricoarytenoid joint is responsible for abduction and adduction of the vocal cords. Involvement of this joint may lead to a feeling of fullness in the throat, to hoarseness, and, rarely, when the cords are essentially fused in a closed position, to a syndrome of acute respiratory distress with or without stridor. In this latter situation, emergent tracheotomy may be life-saving.

Extra-Articular Manifestations Rheumatoid arthritis is symmetrical not only from one side of the body to the other but also within the individual joint. In the case of the knee (Fig. 248-6A), the medial and lateral compartments are both severely narrowed. In contrast, in patients with osteoarthritis (Fig. 248-6B) variably only one compartment of the knee may be involved. Synovial cysts may occur around any of the joints (large or small), and they occasionally manifest as soft, fluctuant masses that present diagnostic challenges. When the knee produces excess synovial fluid, it may accumulate in the popliteal fossa (popliteal or Baker cyst) (E-Fig. 248-2). These cysts can compress the popliteal nerve, artery, or veins. Baker cysts may dissect into the tissues of the calf (usually posteriorly), or they may rupture. Dissection may produce only minor symptoms, such as a feeling of fullness; rupture of the cyst with extravasation of its inflammatory content produces significant pain and swelling and may be confused with thrombophlebitis, the so-called pseudothrombophlebitis syndrome. Ultrasonography of the popliteal fossa and calf is useful to establish the correct diagnosis and rule out thrombophlebitis, which may be precipitated by popliteal cysts. Treatment of popliteal or any other cyst should be directed at interrupting the inflammatory process initially with an intra-articular injection of corticosteroid into the associated joint.

Systemic features of rheumatoid arthritis such as fatigue, weight loss, and low-grade fevers occur frequently. As with all the other extra-articular features, they are more common in those patients who are positive for rheumatoid factor (RF) or ACPA antibodies or both (Table 248-1) and respond to treatment.

Skin

Subcutaneous nodules are seen in approximately 20% of patients with rheumatoid arthritis, almost exclusively in those who are RF or ACPA positive. Patients with nodules who are seronegative should be carefully scrutinized for a different diagnosis, such as chronic tophaceous gout. Nodules may occur almost anywhere (e.g., lungs, heart, eye), but most commonly develop subcutaneously on extensor surfaces (particularly the forearms) (Fig. 248-7), over joints, or over pressure points. Rheumatoid nodules are firm on examination, usually are not tender, have a characteristic histologic picture, and are thought to be initiated by small vessel vasculitis. A syndrome of increased nodulosis, despite good control of the joint disease, has been described with methotrexate therapy (Fig. 248-8). Small vessel vasculitis,5 manifested as digital infarcts or leukocytoclastic vasculitis, may occur in rheumatoid arthritis (Fig. 248-9) and should prompt more aggressive DMARD treatment. A vasculitis of small and medium arteries

CHAPTER 248  Rheumatoid Arthritis  

E-FIGURE 248-2.  Arthrogram with a radiocontrast agent injected into the knee. The dye flows into the popliteal space and through a narrow channel into a large synovial cyst (Baker cyst) that has dissected into the soft tissue of the calf.

1712.e1

CHAPTER 248  Rheumatoid Arthritis  

1713

TABLE 248-1 EXTRA-ARTICULAR MANIFESTATIONS OF RHEUMATOID ARTHRITIS Skin

Nodules, fragility, vasculitis, pyoderma gangrenosum

Heart

Pericarditis, premature atherosclerosis, vasculitis, valve disease, and valve ring nodules

Lung

Pleural effusions, interstitial lung disease, bronchiolitis obliterans, rheumatoid nodules, vasculitis

Eye

Keratoconjunctivitis sicca, episcleritis, scleritis, scleromalacia perforans, peripheral ulcerative keratopathy

Neurologic

Entrapment neuropathy, cervical myelopathy, mononeuritis multiplex (vasculitis), peripheral neuropathy

Hematopoietic

Anemia, thrombocytosis, lymphadenopathy, Felty syndrome, large granular lymphocyte syndrome

Kidney

Amyloidosis, vasculitis

Bone

Osteopenia

FIGURE 248-10.  Rheumatoid nodules in the lung. Chest radiograph demonstrates discrete rheumatoid nodules in both right and left lower lobes. (Courtesy Dr. Martin Lidsky, Houston, TX.)

that is indistinguishable from polyarteritis nodosa also can be seen and requires aggressive systemic therapy. Finally, pyoderma gangrenosum occurs with increased frequency in association with rheumatoid arthritis.

Cardiovascular Involvement

FIGURE 248-7.  Rheumatoid nodules. Large rheumatoid nodules are seen in a classic location along the extensor surface of the forearm and in the olecranon bursa.

Cardiac involvement directly related to rheumatoid arthritis is uncommon; however, patients with rheumatoid arthritis have significantly increased morbidity and mortality from coronary artery disease and heart failure. A metaanalysis of observational studies has shown that the risk for incident cardiovascular disease is increased by 48% in patients with rheumatoid arthritis compared with that in the general population. The reasons are not clear, but chronic inflammation appears to be the major cause, superimposed upon the prevalence of conventional risk factors (e.g., smoking, obesity). Some of the medications used to treat rheumatoid arthritis like glucocorticoids and a sedentary lifestyle may be additional risk factors for the development of coronary artery disease. Pericardial effusions are common (50% by echocardiography) but usually are asymptomatic. Rarely, long-standing pericardial disease may result in a fibrinous pericarditis, and patients may present clinically with constrictive pericarditis (Chapter 68). A population-based inception cohort of rheumatoid arthritis patients in Olmstead County, Minnesota, has shown an increased incidence of venous thromboembolism compared with controls.

Pulmonary Manifestations

FIGURE 248-8.  Rheumatoid nodulosis. In this patient, multiple rheumatoid nodules are present over joints. In some cases, nodules may dominate the clinical picture. Rarely, this may be seen as a side effect of methotrexate therapy.

Pulmonary manifestations include pleural effusions, rheumatoid nodules, and parenchymal lung disease. Pleural effusions occur more commonly in men and are usually small and asymptomatic. Of interest, pleural fluid in rheumatoid arthritis is characterized by low levels of glucose and low pH and, therefore, may at times be confused with empyema. Rheumatoid nodules may occur in the lung, especially in men (Fig. 248-10); these are usually solid but may calcify, cavitate, or become infected. Rarely, pulmonary nodules rupture and produce a pneumothorax. If patients with rheumatoid arthritis are exposed to coal or silica dust, diffuse nodular densities may occur (Caplan syndrome). Differentiating rheumatoid nodules from lung cancer can be problematic, particularly if the lesion is solitary. Therefore, the presence of pulmonary nodules in a patient with rheumatoid arthritis should precipitate an aggressive diagnostic evaluation. Diffuse interstitial fibrosis occurs in rheumatoid arthritis and rarely may occur prior to joint manifestations. It may progress to a honeycomb appearance on radiography with increasing dyspnea. Rarely, bronchiolitis obliterans can be seen, with or without organizing pneumonia.

Ophthalmologic Manifestations

FIGURE 248-9.  Small vessel vasculitis. A and B, Rheumatoid vasculitis with small brown infarcts of palms and fingers in chronic rheumatoid arthritis. (Courtesy Dr. Martin Lidsky, Houston, TX.)

The most common eye manifestation of rheumatoid arthritis is keratoconjunctivitis sicca (dry eyes) from secondary Sjögren syndrome (Chapter 252). Patients may have associated xerostomia (dry mouth), parotid gland swelling, or, occasionally, lymphadenopathy. Scleritis also can occur and may be painful, with progression to thinning of the sclera, with deep pigment showing through on physical examination, and may progress to perforation (scleromalacia perforans). Rarely, tendonitis of the superior oblique muscles can result in double vision (Brown syndrome).

1714

CHAPTER 248  Rheumatoid Arthritis  

Neurologic Manifestations

Peripheral nerve entrapment syndromes, including carpal tunnel syndrome (median nerve at the wrist), and tarsal tunnel syndrome (anterior tibial nerve at the ankle), are common in rheumatoid arthritis. Vasculitis can lead to a stocking and glove neuropathy or mononeuritis multiplex, both of which may require aggressive therapy. Subluxations at C1-C2 may produce myelopathy (see E-Fig. 248-1). Rheumatoid nodules in the central nervous system have been described but are rare and usually asymptomatic.

Felty Syndrome

Felty syndrome is the triad of rheumatoid arthritis, splenomegaly, and neutropenia. This complication is seen in patients with severe RF/ACPA-positive disease and may be accompanied by hepatomegaly, thrombocytopenia, lymphadenopathy, and fevers. Most patients with Felty syndrome do not require special therapy; instead, treatment should be directed toward their severe rheumatoid arthritis and when this is done white blood counts improve. If severe neutropenia (Chapter 158) exists (10 joints (at least one small joint)

0 1 2 3 5

Serology (0-3) Neither RF or ACPA positive At least one test low positive titre At least one test high positive titre

0 2 3

Duration of Synovitis (0-1) 1 & ≤3 × ULN High +: >3 × ULN

Medium/Large joints Knee, hip, shoulders, elbow and ankle Small joints PIPs, MCPs, wrists and MTPs

E-FIGURE 248-3.  American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) classification criteria for rheumatoid arthritis (RA). The classification criteria are shown as a flow diagram for initial decision making and also in tabular form to calculate American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) classification criteria for rheumatoid arthritis (RA). ACPA = anti-citrullinated protein antibody; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate; MCP = metacarpal phalangeal joint; MTP = metatarsal phalangeal joint; PIP = proximal interphalangeal joint; RA = rheumatoid arthritis; RF = rheumatoid factor; ULN = upper limit of normal. (From Aletaha D, Neogi T, Silman AJ, et al. 2010 Rheumatoid arthritis classification criteria. An American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum. 2010;62:2569-2581.)

1714.e1

1715

CHAPTER 248  Rheumatoid Arthritis  

normalize the hemoglobin levels. Other causes of anemia should also be considered in rheumatoid arthritis, particularly iron deficiency anemia from gastrointestinal blood loss. Thrombocytosis is common, with platelet counts returning to normal as the inflammation is controlled. Acute phase reactants such as ESR and CRP levels parallel the activity of the disease, and their persistent elevation portends a poor prognosis in terms of both joint destruction and mortality. White blood cell counts may be elevated, normal, or, in the case of Felty syndrome, depressed. Eosinophilia is present in some patients with rheumatoid arthritis. Synovial fluid in rheumatoid arthritis is characterized by white blood cell counts in the range of 5000 to 100,000/µL, with approximately two thirds of the cells being polymorphonuclear leukocytes. There are no synovial fluid findings that are pathognomonic of rheumatoid arthritis.

TABLE 248-4 KEYS TO OPTIMIZE OUTCOME OF TREATMENT OF RHEUMATOID ARTHRITIS Early, accurate diagnosis Early DMARD therapy Strive for remission in all patients Monitor carefully for treatment toxicities Consider and treat comorbid conditions* *Important comorbid conditions include cardiovascular disease, increased susceptibility to infections, and osteoporosis. DMARD = disease-modifying antirheumatic drug.

Differential Diagnosis

The accurate diagnosis of rheumatoid arthritis early in its course, although challenging, is critical if patients are to benefit maximally from therapeutic intervention. Once disease has been present and active for years and the characteristic deformities and radiographic changes have occurred, the diagnosis is all too obvious. Once rheumatoid arthritis has progressed to that point, deformities may no longer be amenable to medical therapy. Many diseases can mimic rheumatoid arthritis (see Table 248-3). Early in the course of disease, self-limited viral syndromes need to be considered, especially hepatitis B and C, parvovirus, rubella (infection or vaccination), and Epstein-Barr virus. At any time, systemic lupus erythematosus, psoriatic arthritis, and reactive arthritis may present differential diagnostic challenges. In the case of these three mimics, a targeted history and examination to elucidate their associated clinical features, such as rashes, oral ulcers, nail changes, dactylitis, urethritis, and renal, pulmonary, gastrointestinal, or ophthalmologic involvement, is critical. Especially in elderly patients with fulminant-onset rheumatoid arthritis, remitting RF-negative symmetrical synovitis with pitting edema (the so-called RS3PE syndrome) and paraneoplastic syndromes should be considered. Chronic tophaceous gout also may mimic severe nodular rheumatoid arthritis. Hypothyroidism not only causes many rheumatic manifestations but also occurs commonly in conjunction with rheumatoid arthritis and, therefore, should be considered.

TABLE 248-5 TREATMENT OPTIONS FOR RHEUMATOID ARTHRITIS MEDICATION

DOSE

NONSTEROIDAL ANTI-INFLAMMATORY DRUGS* Celecoxib

100 mg PO twice daily or 200 mg PO daily

Diclofenac/misoprostol

50 mg/200 mg PO 2 to 4 times daily

Etodolac

300 mg PO 2 or 3 times daily; or 400 or 500 mg PO twice daily

Ibuprofen

800 mg PO three times daily

Ketoprofen

75 mg PO three times daily; or 50 mg PO four times daily

Naproxen

500 mg PO twice daily

Naproxen/esomeprazole

375 mg/20 mg or 500 mg/20 mg PO twice daily

Tolmetin

400 mg PO three times daily

GLUCOCORTICOIDS* Prednisone

5-10 mg PO once daily

DISEASE-MODIFYING ANTIRHEUMATIC DRUGS (DMARDs) Conventional DMARDs

TREATMENT  General Measures

Rheumatoid arthritis is a lifelong disease process that has no known cure; the diagnosis is made based on clinical criteria, and many different options exist for treatment. These factors magnify the importance of the patient-physician relationship and place a premium on the art as well as the science of medicine. Optimal patient care requires effective ongoing interactions between primary care physicians and rheumatologists, and, in some cases, physical therapists, occupational therapists, and orthopedic surgeons. A1  Because of the serious nature of the disease, the rapid introduction of new treatments, and the need for expertise in monitoring these therapies, all patients with rheumatoid arthritis should be evaluated early and followed closely by a rheumatologist. The goal of therapy, which is disease remission (Table 248-4) or very low disease activity, should always be kept in mind.8 When rheumatoid arthritis is treated early, remission is possible in over 50% of patients and low disease activity in at least another 35%. However, continued response requires the ongoing use of DMARDs. Essentially all rheumatoid arthritis patients should be treated with DMARDs.9 In many patients, combinations of different DMARDs (conventional and biologic) are necessary for optimal control. A2  Therapy should be escalated rapidly to ensure maximal suppression of disease while minimizing toxicity and expense. Patients with rheumatoid arthritis should be educated about their disease and its treatment. Patients should have an opportunity to spend time with physical therapists and occupational therapists to learn about range-of-motion exercises, joint protection, and assistive devices.

Medical Therapy

Three types of medical therapies are used for rheumatoid arthritis: NSAIDs, glucocorticoids, and DMARDs (both conventional and biologic)10 (Table 248-5). Initial therapy should always include a DMARD.

Nonsteroidal Anti-inflammatory Drugs

NSAIDs are important for the symptomatic relief they provide; however, they play only a minor role in altering the underlying disease process. Therefore, NSAIDs should rarely, if ever, be used to treat rheumatoid arthritis without the concomitant use of DMARDs. Many clinicians waste valuable time switching from one NSAID to another before starting DMARD therapy.

Hydroxychloroquine

200-400 mg PO daily (5 mg/kg)

Leflunomide

10-20 mg PO once daily

Methotrexate

7.5-25 mg PO† once weekly (orally or subcutaneously)

Sulfasalazine

500 mg PO once or twice daily for first two weeks; then 2 g daily in two evenly divided doses

Tofacitinib

5 mg PO twice daily

Azathioprine

1 mg/kg (50-100 mg) PO as a single dose or twice daily; may be increased up to 2.5 mg/kg

Minocycline

100 mg PO twice daily

Biologic DMARDs Abatacept

500-1000‡ mg IV at 0, 2, and 4 weeks, then every 4 weeks§

Adalimumab

40 mg SC every other week

Certolizumab

400 mg SC at weeks 2 and 4, followed by 200 mg every other week

Etanercept

50 mg SC once weekly

Golimumab

50 mg SC once a month¶

Infliximab

In conjunction with methotrexate, 3-5 mg/kg at 0, 2, and 6 weeks, then every 4-8 weeks

Rituximab

In combination with methotrexate, two 1000-mg intravenous infusions separated by 2 weeks (one course) every 16-24 weeks

Tocilizumab

4 mg/kg IV every 4 weeks followed by 8 mg/kg every 4 weeks**

*Initial treatment should always include a DMARD; see text. † May be administered SC to reduce side effects and when higher doses are needed ‡ 500 mg if 100 kg § May be administered SC once weekly with or without an intravenous loading dose ¶ May be administered IV at a dose of 2 mg/kg at weeks 0 and 4, then every 8 weeks **May be administered SC at a starting dose of 162 mg every other week if 100 kg DMARD = disease-modifying antirheumatic drug; PO = orally; IV = intravenously; SC = subcutaneously.

1716

CHAPTER 248  Rheumatoid Arthritis  

TABLE 248-6 GUIDELINES FOR USE OF GLUCOCORTICOIDS Avoid use of glucocorticoids without DMARDs Prednisone, >10 mg/day, is rarely indicated for articular disease Taper to the lowest effective dose Use as “bridge therapy” until DMARD therapy is effective Remember prophylaxis against osteoporosis DMARD = disease-modifying antirheumatic drug.

Much has been written about the gastrointestinal toxicity of NSAIDs, and these concerns are particularly relevant to patients with rheumatoid arthritis, who often have significant risk factors, including age and concomitant steroid use. Therefore, cyclooxygenase-2 (COX2)-selective agents have been a popular choice for rheumatoid arthritis patients. The evidence linking these agents to increased cardiovascular toxicity is, however, particularly troubling for patients with rheumatoid arthritis, who are already at high risk for myocardial infarction. In a randomized clinical trial, a cyclooxygenase-2 (COX-2) selective NSAID, celecoxib, was compared to two nonselective NSAIDs, naproxen or ibuprofen, in patients with rheumatoid arthritis and osteoarthritis, and celecoxib was found to be noninferior with regard to cardiovascular outcomes. A3  If COX2-selective agents are used, they should be kept at a low dose. Consideration should be given to low-dose aspirin prophylaxis, but this therapy may increase the gastrointestinal toxicity of NSAIDs. The use of concomitant misoprostol or proton pump inhibitors should be considered in all patients with rheumatoid arthritis who are taking NSAIDs. Additionally, the potential for NSAIDs to decrease renal blood flow and to increase blood pressure should be kept in mind.

Glucocorticoids

Glucocorticoids have had a significant role in the treatment of rheumatoid arthritis for more than half a century (Chapter 32). Indeed, it was chosen as the first disease to be treated with this new therapy. As was the case with the first patient treated in 1948, glucocorticoids are dramatically and rapidly effective in patients with rheumatoid arthritis. Not only are glucocorticoids useful for symptomatic improvement but they also significantly decrease the radiographic progression of rheumatoid arthritis. However, the toxicities of long-term therapy are extensive and potentially devastating. Therefore, the optimal use of these drugs requires an understanding of several principles (Table 248-6). Glucocorticoids remain among the most potent anti-inflammatory treatments available; for this reason, and because of their rapid onset of action, they are ideally suited to help control the inflammation in rheumatoid arthritis while the much slower-acting DMARDs are starting to work. Prednisone, the most commonly used glucocorticoid, should rarely be used in doses higher than 10 mg/day to treat the articular manifestations of rheumatoid arthritis. At this dose at the start of methotrexate-based treatment, the addition of prednisone reduces erosive joint damage, disease activity, physical disability, and the use of biologic treatment at 2 years. The dose should be slowly tapered to the lowest effective dose, and the concomitant DMARD therapy should be adjusted to make this possible. Glucocorticoids should rarely, if ever, be used without concomitant DMARD therapy. The paradigm is to shut off inflammation rapidly with glucocorticoids and then to taper them as the DMARD is taking effect (“bridge therapy”). The clear goal is to have all patients off or on very low doses of glucocorticoids with their disease controlled by DMARDs. In all patients receiving glucocorticoids, strong measures should be taken to prevent osteoporosis. Bisphosphonates have been shown to be particularly effective in this regard but are contraindicated in women of childbearing age. Higher doses of glucocorticoids may be necessary to treat extra-articular manifestations, especially vasculitis and scleritis.

Disease-Modifying Antirheumatic Drugs

DMARDs are a group of medications that have the ability to halt the disease process in the synovium and to modify or change the disabling potential of rheumatoid arthritis.11 These drugs also can halt or slow radiographic progression of the disease.

Conventional Disease-Modifying Antirheumatic Drugs

Included in this group of medications are methotrexate, sulfasalazine (Azulfidine), gold, antimalarials (hydroxychloroquine [Plaquenil] and others), leflunomide (Arava), azathioprine (Imuran), minocycline, and the recently approved tofacitinib (Xeljanz). It is critically important that clinicians and patients understand that conventional DMARDs take 2 to 6 months to exert their maximal effect, and all require some monitoring (Table 248-7). Monitoring of serum drug levels of biologic agents may be especially useful in patients in remission, because higher blood levels predict a longer relapse-free period.12 Other measures, such as glucocorticoid therapy, may be needed to control the disease while DMARDs are starting to work. These DMARDs have been shown to be effective in treating both early and more advanced rheumatoid arthritis. Until additional research elucidates factors that allow selection of the best initial therapy for each patient, the choice will depend on patient and physician concerns about toxicity and monitoring issues,

TABLE 248-7 CAVEATS FOR MONITORING DISEASE-MODIFYING ANTIRHEUMATIC DRUG THERAPIES* MEDICATION

CAVEATS

Prednisone

Use as bridge to effective DMARD therapy. Prophylaxis for osteoporosis? (see Table 248-6)

Hydroxychloroquine

Keep dosage lower than 5.0 mg/kg/day. Yearly eye checkup by ophthalmologist after 5 years of therapy

Sulfasalazine

CBC for neutropenia, initially every month, then every 6 mo

Methotrexate

CBC and SGOT/SGPT every 8-12 wk when dose is stable. Many toxicities respond to folic acid or small dose reduction. If pneumonitis, stop and do not restart. Decreasing renal function may precipitate toxicities. Absolute contraindication in pregnancy

Leflunomide

CBC and SGOT/SGPT every 4-8 wk; long half-life may require cholestyramine washout; absolute contraindication in pregnancy

TNF inhibitors

If fevers or infectious symptoms of any kind, stop until symptoms resolve; aggressively work up and treat possible infections. May precipitate congestive heart failure, demyelinating syndromes, or lupus-like syndromes

*Patients receiving DMARDs, both conventional and biologic, should be monitored by a rheumatologist. CBC = compete blood count; DMARD = disease-modifying antirheumatic drug; SGOT = serum glutamate oxaloacetate transaminase (aspartate aminotransferase); SGPT = serum glutamate pyruvate transaminase (alanine aminotransferase); TNF = tumor necrosis factor.

as well as the activity of disease and presence of comorbid conditions. The critical issue is not which DMARD to start first but rather getting the DMARD therapy started early in the disease process.

Methotrexate

Methotrexate should be the initial DMARD for most patients; it is economical, serious toxicities are rare, and when used in combination it makes essentially all other DMARDs work better. A4  Methotrexate is contraindicated in pregnancy and in patients with significant renal impairment. Methotrexate is dramatically effective in slowing radiographic progression and is usually given orally in doses ranging from 5 to 30 mg as a single weekly dose. This once-per-week administration is worthy of emphasis; prior experience with daily therapy in psoriasis has demonstrated the importance of allowing the liver time to recover between doses. Oral absorption of methotrexate is variable; subcutaneous injections of methotrexate are often effective when oral treatment is not and should be strongly considered before giving up on methotrexate. Side effects of methotrexate include oral ulcers, nausea, hepatotoxicity, bone marrow suppression, and pneumonitis. With the exception of pneumonitis, these toxicities respond to dose adjustments. Monitoring of blood counts and liver blood tests (albumin and aspartate aminotransaminase [AST] or alanine aminotransferase [ALT]) should be done every 3 months, with adjustments in the dose of methotrexate as needed. Renal function is critical for clearance of methotrexate; previously stable patients may experience severe toxicities if renal function deteriorates. Pneumonitis, although rare, is less predictable and can be fatal, particularly if the methotrexate is not stopped or is restarted. Folic acid, 1 to 4 mg/day, can significantly decrease most methotrexate toxicities without interfering with efficacy. If methotrexate alone does not sufficiently control disease, it is combined with other DMARDs. A5  Methotrexate in combination with virtually any of the other DMARDs (conventional or biologic) has been shown to be more effective than either drug alone.

Leflunomide

Leflunomide, a pyrimidine antagonist, has a very long half-life and is most commonly started at 10 to 20 mg/day orally. Diarrhea is the most common toxicity and responds to dose reduction. Doses of leflunomide of 10 to 20 mg three to five times per week are frequently used. Also, because of the long half-life and teratogenic potential of leflunomide, women wishing to become pregnant who have previously received leflunomide, even if therapy was stopped years ago, should have blood levels drawn. If toxicity occurs or if pregnancy is being considered, leflunomide can be rapidly eliminated from the body by treatment with cholestyramine. Laboratory monitoring for hematologic and hepatic toxicity should be done during treatment with leflunomide, as recommended for methotrexate.

Antimalarial Drugs

The antimalarial drugs hydroxychloroquine (Plaquenil) and chloroquine are frequently used to treat rheumatoid arthritis. They have the least toxicity of

CHAPTER 248  Rheumatoid Arthritis  

any of the DMARDs and do not require monitoring of blood tests. Yearly monitoring by an ophthalmologist after 5 years of therapy is recommended to detect any signs of retinal toxicity (rare). Hydroxychloroquine is the most commonly used preparation and is given orally at 200 to 400 mg/day (5 mg/kg). These drugs are frequently used in combination with other DMARDs, particularly methotrexate. Hydroxychloroquine decreases cholesterol levels, improves HDL function, and recently has been shown to decrease the incidence of diabetes in patients with rheumatoid arthritis.

Sulfasalazine

Sulfasalazine is an effective treatment when given in doses of 1 to 3 g/day. Monitoring of blood counts, particularly white blood cell counts, in the first 6 months is recommended. Sulfasalazine and hydroxychloroquine are often combined with methotrexate, a regimen referred to as triple therapy, which has been shown to have equal efficacy to the combination of a TNF inhibitor and methotrexate and is much more economical.

Minocycline

Minocycline 100 mg twice daily is an effective treatment for rheumatoid arthritis, particularly when used in early RF-positive disease. Chronic therapy (>2 years) with minocycline may lead to cutaneous hyperpigmentation which reverses when the drug is stopped. Minocycline has also been associated with drug-induced lupus.

Tofacitinib

Tofacitinib (Xeljanz) was the first JAK kinase inhibitor to be approved for rheumatoid arthritis. It is given orally at a dose of 5 mg twice daily, and complete blood count and liver function tests should be monitored. Additional toxicity concerns include infections, including tuberculosis, malignancies, and reactivation of herpes zoster. Tofacitinib has been shown to be effective as initial DMARD therapy, A6  when combined with methotrexate in patients who have had incomplete responses to methotrexate, and in patients who have failed TNF inhibitors. Another JAK inhibitor is baricitinib. In initial reports in patients whose rheumatoid arthritis had not responded to a DMARD, it has shown a consistent beneficial effect compared with placebo. A7  Furthermore, baricitinib is better than placebo or adalimumab for patients who had inadequate response to methotrexate. A8  Emerging drugs target distinct members of the JAK family (namely JAK1, JAK2, JAK3, and TYK2). For example, upadacitinib (15 or 30 mg weekly) is effective either as monotherapy or when added to other DMARDs. A9  A10  ,

1717

patient should be advanced to triple therapy. If the patient does not achieve adequate control after 3 to 6 months on triple therapy, either a TNF inhibitor or another biologic should be added to methotrexate.14 Among patients whose rheumatoid arthritis was previously treated with anti-TNF drugs but who had an inadequate primary response, a non-TNF biologic agent is more effective in achieving a good or moderate response than is a second anti-TNF medication. A18 

Treatment of Underlying Conditions

Optimal care of patients with rheumatoid arthritis requires recognition of the associated comorbid conditions, including an increased risk for cardiovascular death, osteoporosis, infections (especially pneumonia), and certain cancers such as lymphomas and lung cancers.

Cardiovascular Disease

Cardiovascular disease is being recognized as the cause of much of the excess mortality in rheumatoid arthritis. Various factors contribute to this mortality, including sedentary lifestyle and glucocorticoid therapy. However, a strong association between chronic inflammation and cardiovascular disease has been identified, and it is likely that this is the most significant factor. Therapies that control rheumatoid arthritis earlier and better can be expected to decrease cardiovascular morbidity and mortality. Both methotrexate and TNF inhibitors have been shown to significantly decrease cardiovascular mortality. Clinicians should consider rheumatoid arthritis a risk factor for cardiovascular disease and should aggressively address other cardiovascular risk factors (Chapter 45).

Other Associated Diseases

Osteoporosis is common in patients with rheumatoid arthritis, and early treatment results in long-term dividends. Patients with rheumatoid arthritis are at an increased risk for infections, and some forms of treatment further increase this risk. Patients should be cautioned to seek early medical attention for even minor symptoms suggestive of infection, especially if receiving biologic therapy. All patients with rheumatoid arthritis should receive a pneumococcal vaccine, yearly influenza vaccinations (Chapter 15), and one of the zoster vaccinations. Finally, patients with rheumatoid arthritis have an increased risk for lymphoma. Occasionally, B-cell lymphomas are associated with immunosuppression and regress after immunosuppression is discontinued. Patients with rheumatoid arthritis have significantly decreased risk of developing colon cancer, probably because of chronic inhibition of COX by NSAIDs.

Biologic Disease-Modifying Antirheumatic Drugs

Cytokines, most notably TNF-α, IL-1, and IL-6, play a central role in the pathophysiology of rheumatoid arthritis (Chapter 42). This understanding led to the development and clinical use of biologic agents directed against TNF-α (etanercept [Enbrel], infliximab [Remicade], adalimumab [Humira], golimumab [Simponi], and certolizumab [Cimzia]), IL-1 (anakinra [Kineret]), and IL-6 (tocilizumab [Actemra] and sarilumab [Kevzara]). Additionally, monoclonal antibodies that deplete B cells (anti-CD20, rituximab [Rituxan]) A11  and that block the second signal for T-cell activation (abatacept [Orencia]) are effective treatments for rheumatoid arthritis. All patients with rheumatoid arthritis receiving biologic therapies should be monitored by a rheumatologist, and their physicians should be aware of the risk for infections that are often atypical.13 All the biologics, when combined with methotrexate, have been shown to decrease disease activity and slow radiographic progression in patients with active disease despite methotrexate. A12  Early treatment with tocilizumab plus methotrexate was shown to result in greater sustainable clinical, functional, and radiographic benefits than methotrexate alone, with acceptable safety and tolerability, in early erosive rheumatoid arthritis. Sirukumab, which is another anti-IL-6 agent, also may become a useful agent in patients who are refractory in anti-TNF therapy. A13  However, standard-dose and high-dose biological drugs (with or without traditional DMARDs) are associated with an increase in serious infections. A14  Biologic agents also should not be used in combination with each other because they significantly increase the risk of infections (see Chapter 33).

The Order of Therapy in Rheumatoid Arthritis

Several randomized double-blind trials have elucidated the order of therapy in rheumatoid arthritis. The Treatment of Early Aggressive Rheumatoid Arthritis (TEAR) trial nicely showed that initial therapy with methotrexate in patients with poor-prognosis rheumatoid arthritis was not inferior at 2 years to initial combinations of either conventional DMARDs or the combination of methotrexate and etanercept. A15  The Rheumatoid Arthritis: Comparison of Active Therapies (RACAT) trial has also shown that in those patients who are not controlled on methotrexate alone, the strategy of initially adding sulfasalazine and hydroxychloroquine to methotrexate (triple therapy) was not inferior to the addition of etanercept to methotrexate. A16  Patients who have not been controlled by methotrexate and other standard treatments can achieve equivalent benefits, with no demonstrable difference in the prevention of joint damage, with combined conventional disease-modifying drugs as with biologics. A17  Therefore, because of the huge economic advantages, the typical rheumatoid arthritis patient should be started on methotrexate monotherapy, and, if not controlled after 3 to 6 months on maximum methotrexate, the



PROGNOSIS

Rheumatoid arthritis is not a benign disease and is not limited to the joints. Once established, it is a lifelong progressive disease that produces significant morbidity in most patients and premature mortality in many. However, allcause mortality in patients with rheumatoid arthritis is now similar to the general population.15 Older studies found that 50% of patients have had to stop working after 10 years (≈10 times the average rate). By comparison, a recent cohort study has shown that the aggressive use of disease-modifying agents and biologic agents, if needed, is associated with substantial reductions in disability. Therefore, early DMARD therapy is critical. Although limited long-term data are available, the current information strongly suggests that patients have the opportunity to benefit greatly if the newer principles of therapy are practiced. Patients who are RF or ACPA positive and those who are positive for the shared epitope have a worse prognosis, with more erosions and more extraarticular disease (see E-Table 248-1). The HLA-DRB1 locus, which is associated with disease susceptibility, may also be associated with radiologic severity, mortality, and treatment response.16 Once deformities are found on examination or erosions on radiography, the damage is largely irreversible. Erosions usually occur in the first 1 to 2 years, and the rate of radiographic damage can be reduced by early therapy.  

FUTURE DIRECTIONS

Significant advances in the effective treatment of rheumatoid arthritis have come from an understanding of the cytokine imbalance that accompanies this disease. Much research is focused on the further development of biologic products to modulate this balance. There remains a critical need for a cytokine thermostat that titrates the desired cytokine balance to control disease without altering critical immune functions. Even with existing therapies, many different effective options are available for rheumatoid arthritis. The challenge for the clinician is to pick the right option for each patient. Few data are currently available to aid in this choice, and the establishment of parameters, genetic or otherwise, that would allow selection of the best initial option for each patient would be a major

breakthrough. In this respect the option to adopt precision medicine principles is most enticing. Finally, elucidation of the trigger or triggers for rheumatoid arthritis may allow the development of strategies to prevent the onset of clinical disease.

  Grade A References A1. Lamb SE, Williamson EM, Heine PJ, et al. Exercises to improve function of the rheumatoid hand (SARAH): a randomised controlled trial. Lancet. 2015;385:421-429. A2. Nam JL, Takase-Minegishi K, Ramiro S, et al. Efficacy of biological disease-modifying antirheumatic drugs: a systematic literature review informing the 2016 update of the EULAR recommendations for the management of rheumatoid arthritis. Ann Rheum Dis. 2017;76:1113-1136. A3. Nissen SE, Yeomans ND, Solomon DH, et al. Cardiovascular safety of celecoxib, naproxen, or ibuprofen for arthritis. N Engl J Med. 2016;375:2519-2529. A4. Lopez-Olivo MA, Siddhanamatha HR, Shea B, et al. Methotrexate for treating rheumatoid arthritis. Cochrane Database Syst Rev. 2014;6:CD000957. A5. Moreland LW, O’Dell JR, Paulus HE, et al. A randomized comparative effectiveness study of oral triple therapy versus etanercept plus methotrexate in early aggressive rheumatoid arthritis: the treatment of Early Aggressive Rheumatoid Arthritis Trial. Arthritis Rheum. 2012;64:2824-2835. A6. Lee EB, Fleischmann R, Hall S, et al. Tofacitinib versus methotrexate in rheumatoid arthritis. N Engl J Med. 2014;370:2377-2386. A7. Genovese MC, Kremer JM, Kartman CE, et al. Response to baricitinib based on prior biologic use in patients with refractory rheumatoid arthritis. Rheumatology (Oxford). 2018;57:900-908. A8. Taylor PC, Keystone EC, van der Heijde D, et al. Baricitinib versus placebo or adalimumab in rheumatoid arthritis. N Engl J Med. 2017;376:652-662. A9. Genovese MC, Fleischmann R, Combe B, et al. Safety and efficacy of upadacitinib in patients with active rheumatoid arthritis refractory to biologic disease-modifying anti-rheumatic drugs (SELECTBEYOND): a double-blind, randomised controlled phase 3 trial. Lancet. 2018;391:2513-2524. A10. Burmester GR, Kremer JM, Van den Bosch F, et al. Safety and efficacy of upadacitinib in patients with rheumatoid arthritis and inadequate response to conventional synthetic disease-modifying anti-rheumatic drugs (SELECT-NEXT): a randomised, double-blind, placebo-controlled phase 3 trial. Lancet. 2018;391:2503-2512. A11. Porter D, van Melckebeke J, Dale J, et al. Tumour necrosis factor inhibition versus rituximab for patients with rheumatoid arthritis who require biological treatment (ORBIT): an open-label, randomised controlled, non-inferiority, trial. Lancet. 2016;388:239-247. A12. Nam JL, Ramiro S, Gaujoux-Viala C, et al. Efficacy of biological disease-modifying antirheumatic drugs: a systematic literature review informing the 2013 update of the EULAR recommendations for the management of rheumatoid arthritis. Ann Rheum Dis. 2014;73:516-528. A13. Aletaha D, Bingham CO 3rd, Tanaka Y, et al. Efficacy and safety of sirukumab in patients with active rheumatoid arthritis refractory to anti-TNF therapy (SIRROUND-T): a randomised, double-blind, placebo-controlled, parallel-group, multinational, phase 3 study. Lancet. 2017;389: 1206-1217. A14. Genovese MC, Kremer J, Zamani O, et al. Baricitinib in patients with refractory rheumatoid arthritis. N Engl J Med. 2016;374:1243-1252. A15. O’Dell JR, Curtis JR, Mikuls TR, et al. Validation of the methotrexate-first strategy in patients with early, poor-prognosis rheumatoid arthritis: results from a two-year randomized, double-blind trial. Arthritis Rheum. 2013;65:1985-1994. A16. O’Dell JR, Mikuls TR, Taylor TH, et al. Therapies for active rheumatoid arthritis after methotrexate failure. N Engl J Med. 2013;369:307-318. A17. Scott DL, Ibrahim F, Farewell V, et al. Tumour necrosis factor inhibitors versus combination intensive therapy with conventional disease modifying anti-rheumatic drugs in established rheumatoid arthritis: TACIT non-inferiority randomised controlled trial. BMJ. 2015;350:1-9. A18. Gottenberg JE, Brocq O, Perdriger A, et al. Non-TNF-targeted biologic vs a second anti-TNF drug to treat rheumatoid arthritis in patients with insufficient response to a first anti-TNF drug: a randomized clinical trial. JAMA. 2016;316:1172-1180.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 248  Rheumatoid Arthritis  

GENERAL REFERENCES 1. Smolen JS, Aletaha D, Barton A, et al. Rheumatoid arthritis. Nat Rev Dis Primers. 2018;4:1-23. 2. England BR, Thiele GM, Anderson DR, et al. Increased cardiovascular risk in rheumatoid arthritis: mechanisms and implications. BMJ. 2018;361:1-17. 3. Yarwood A, Huizinga TW, Worthington J. The genetics of rheumatoid arthritis: risk and protection in different stages of the evolution of RA. Rheumatology (Oxford). 2016;55:199-209. 4. Guo Q, Wang Y, Xu D, et al. Rheumatoid arthritis: pathological mechanisms and modern pharmacologic therapies. Bone Res. 2018;6:1-14. 5. Makol A, Matteson EL, Warrington KJ. Rheumatoid vasculitis: an update. Curr Opin Rheumatol. 2015;27:63-70. 6. Nordberg LB, Lillegraven S, Aga AB, et al. Comparing the disease course of patients with seronegative and seropositive rheumatoid arthritis fulfilling the 2010 ACR/EULAR classification criteria in a treat-to-target setting: 2-year data from the ARCTIC trial. RMD Open. 2018;4:11-17. 7. Schett G. The role of ACPAs in at-risk individuals: early targeting of the bone and joints. Best Pract Res Clin Rheumatol. 2017;31:53-58. 8. Aletaha D, Smolen JS. Diagnosis and management of rheumatoid arthritis: a review. JAMA. 2018;320:1360-1372. 9. Chatzidionysiou K, Emamikia S, Nam J, et al. Efficacy of glucocorticoids, conventional and targeted synthetic disease-modifying antirheumatic drugs: a systematic literature review informing the 2016

1718.e1

update of the EULAR recommendations for the management of rheumatoid arthritis. Ann Rheum Dis. 2017;76:1102-1107. 10. Smolen JS, Aletaha D, McInnes IB. Rheumatoid arthritis. Lancet. 2016;388:2023-2038. 11. Sparks JA. Rheumatoid arthritis. Ann Intern Med. 2019;170:ITC1-ITC16. 12. Martin-Lopez M, Carmona L, Balsa A, et al. Serum drug levels of biologic agents in the management of rheumatoid arthritis and spondyloarthritis: a systematic review. Rheumatol Int. 2018;38: 975-983. 13. Rutherford AI, Subesinghe S, Hyrich KL, et al. Serious infection across biologic-treated patients with rheumatoid arthritis: results from the British Society for Rheumatology Biologics Register for Rheumatoid Arthritis. Ann Rheum Dis. 2018;77:905-910. 14. Smolen JS, Landewé R, Bijlsma J, et al. EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2016 update. Ann Rheum Dis. 2017;76:960-977. 15. Lacaille D, Avina-Zubieta JA, Sayre EC, et al. Improvement in 5-year mortality in incident rheumatoid arthritis compared with the general population-closing the mortality gap. Ann Rheum Dis. 2017;76:1057-1063. 16. Viatte S, Plant D, Han B, et al. Association of HLA-DRB1 haplotypes with rheumatoid arthritis severity, mortality, and treatment response. JAMA. 2015;313:1645-1656.

1718.e2

CHAPTER 248  Rheumatoid Arthritis  

REVIEW QUESTIONS 1. A 32-year-old woman presents with 3 weeks of symmetrical swelling and stiffness in her PIPs and MCPs. The most helpful laboratory test result in establishing a diagnosis of RA in this patient would be: A . ESR of 65 mm/hour. B. RF of 120 IU (normal 3 mo that improve with exercise but are not relieved by rest Limitation of motion of the lumbar spine in both sagittal and frontal planes Limitation of chest expansion RADIOLOGIC CRITERIA Sacroiliitis: grade ≥2 bilateral or grade 3 or 4 unilateral GRADING Definite AS if the radiologic criterion is associated with at least one clinical variable Probable AS if: The three clinical criteria are present The radiologic criterion is present without the clinical criteria AS = ankylosing spondylitis.

FIGURE 249-2.  Bilaterally symmetrical sacroiliitis in ankylosing spondylitis.

CHAPTER 249  The Spondyloarthropathies  

Radiographs of the spine may reveal squaring of the vertebral bodies (loss of the normal anterior concavity of the lumbar vertebra) and “shiny corners” (subchondral sclerosis at the upper edge of the vertebral body), both of which are manifestations of enthesitis. Syndesmophytes, which represent marginal bridging of the vertebrae (Figs. 249-3 and 249-4), eventually develop and make the diagnosis clear. Because ankylosis of the apophyseal joints may occur without syndesmophyte formation, it is important to assess the posterior joints on the lateral lumbosacral spine views, as well as the anterior margin of the vertebrae. Eventually, the changes may result in a “bamboo spine,” so called because the bridging syndesmophytes can mimic the appearance of bamboo. It is now appreciated that osteoporosis (Chapter 230) is a significant feature of ankylosing spondylitis, probably reflecting both the local chronic inflammation and the abnormal biomechanical loading of the vertebrae as the disease progresses.

1721

Differential Diagnosis

The differential diagnosis of ankylosing spondylitis includes the following: osteitis condensans ilii; diffuse idiopathic skeletal hyperostosis (DISH; Chapter 243); the syndrome of synovitis, acne, pustulosis, hyperostosis, and osteomyelitis (SAPHO); and some induced hyperostotic states (vitamin A intoxication, fluorosis). New bone formation occurs in degenerative disc disease, but the bulky horizontal appearance of osteophytes is usually easily distinguished from that of syndesmophytes, and narrowing of the disc space is not a feature of ankylosing spondylitis. Osteoarthritis of the sacroiliac joint has recently been recognized as having a higher prevalence than previously appreciated. The clinical course and severity of ankylosing spondylitis are highly variable. Inflammatory back pain and stiffness dominate the picture in the early stages, whereas chronic pain and deformity may develop over time. In both early and late phases of the disease, there may be a significant impact on work disability and quality of life. In only a minority of patients does the full-blown picture of a bamboo spine eventually develop, but there are few variables that can reliably aid in prognosticating the course. At present, the strongest predictor of new syndesmophyte formation is the presence of syndesmophytes at baseline. In ankylosing spondylitis patients in whom new, refractory spinal pain develops, an intervertebral fracture should be considered, which can occur after only minimal trauma. Additional late complications may include cauda equina syndrome, osteoporotic compression fractures, spondylodiscitis, and restrictive lung disease.

Reactive Arthritis  

DEFINITION

Reactive arthritis is an aseptic arthritis that occurs subsequent to an extraarticular infection, most typically of the GI or GU tract.6 In the GI tract, the key pathogens are Salmonella typhimurium, Yersinia enterocolitica, Shigella flexneri, and Campylobacter jejuni. In the GU tract, Chlamydia trachomatis is the most common offender.  

FIGURE 249-3.  Left, Lumbar spondylitis in ankylosing spondylitis, with symmetrical marginal bridging syndesmophytes and calcification of the spinal ligament. Right, The bulky, nonmarginal, asymmetrical syndesmophytes of reactive arthritis with lumbar spondylitis.



FIGURE 249-4.  A 34-year-old man who has had ankylosing spondylitis for 9 years and neck pain. Radiographs demonstrate narrowing of the C2-C3 apophyseal joints posteriorly and anterior bridging marginal syndesmophytes extending from C2 to C5.

EPIDEMIOLOGY

The true incidence and prevalence of ReA are not well defined. In epidemics involving Salmonella (Chapter 292) or Yersinia (Chapter 296), it is estimated that reactive arthritis develops in 2 to 7% of infected individuals but in as many as 20% of B27-positive infected individuals. In such epidemic studies, B27 confers risk not only for the onset of arthritis but also for axial involvement and chronicity. Genetic variants in toll-like receptor 2 (TLR-2) are associated with acute reactive arthritis, thus implicating host innate immunity as central in reactive arthritis. The variability in the rate of reactive arthritis is determined by the heterogeneity of the cohorts reported, which introduces confounding variables of different genetic backgrounds in the population and different species of pathogens. Even in the setting of an epidemic point source outbreak, the inoculum varies widely among the exposed individuals, and the genetic makeup of the population at risk (e.g., the prevalence of B27) may differ greatly among different studies. Case ascertainment and relative risk are even more difficult to determine for postChlamydia reactive arthritis. Young adults in the United States have a high prevalence of asymptomatic Chlamydia carriage in the GU tract, and establishing a causal link between Chlamydia and synovitis can be difficult. Nevertheless, it is with Chlamydia that reactive arthritis has been most intensively studied.

PATHOBIOLOGY

Although immunofluorescence studies have identified bacterial antigens in the joints of patients with reactive arthritis after both GI and GU infections, it is primarily in post-Chlamydia reactive arthritis that results of polymerase chain reaction studies on synovial tissues have most consistently been positive, suggesting that viable Chlamydia may persist in the joints of such patients, albeit in a metabolically altered state. Typically, the onset of arthritis occurs 1 to 3 weeks after the GI or GU infection, but the temporal details are often difficult to define precisely. Although the definition of aseptic arthritis after an extra-articular infection may include a broader range of pathogens (e.g., Chlamydia pneumoniae), sites of infection (e.g., streptococcal pharyngitis), and types of infections (e.g., Giardia infections of the GI tract), these clinical scenarios have not generally been included in the category of reactive arthritis. They lack the other associated clinical features of the spondyloarthritis group of diseases, and they lack an association with B27.

1722

CHAPTER 249  The Spondyloarthropathies  

FIGURE 249-5.  Keratoderma blennorrhagicum of the feet in reactive arthritis.



DIAGNOSIS

The pattern of joint involvement in reactive arthritis is one of asymmetrical oligoarthritis with a predilection for the lower extremity, a pattern shared by most spondyloarthritis syndromes. Enthesitis may present as Achilles tendinitis or plantar fasciitis. Dactylitis, appearing as a sausage digit, may also be seen. Dactylitis is the net result of inflammatory changes affecting the joint capsule, entheses, periarticular structures, and periosteal bone. Sacroiliitis may be seen in the acute phase, but radiographic changes are seen largely in patients with a more chronic course. When reactive arthritis is accompanied by certain extra-articular features such as urethritis, conjunctivitis, or mucocutaneous lesions, the term Reiter syndrome has been applied historically, but it is no longer in common use. The urethritis may manifest as dysuria or discharge, and the rash as circinate balanitis, which appears as vesicles or shallow ulcerations on the glans penis. Painless lingual or oral ulcerations may also be seen. The fact that the cervicitis may be less symptomatic could partially account for the underdiagnosis in women. The classic skin manifestation of reactive arthritis is keratoderma blennorrhagicum, a painless papulosquamous eruption on the palms or soles (Fig. 249-5). Occasionally, nail dystrophy with pitting and onycholysis or subungual keratosis can be seen. The conjunctivitis can be bilateral and painful; in contrast, the acute anterior uveitis that can also be seen in this setting tends to be less painful and unilateral. Radiographic changes of reactive arthritis can be seen in the involved peripheral joints, with early findings consisting of soft tissue swelling and juxtaarticular osteopenia. Areas of periostitis and new bone formation may develop in peripheral joints. When changes in the sacroiliac joints are seen, they are typically asymmetrical (Fig. 249-6), in contrast to the symmetrical pattern seen in ankylosing spondylitis. In the chronic phase, syndesmophytes may develop, but they are described as bulky, nonmarginal, often asymmetrical formations that differ from the classic syndesmophytes of ankylosing spondylitis. The frequency with which reactive arthritis evolves into bona fide ankylosing spondylitis has not been determined definitively.

Differential Diagnosis

The most important differential diagnosis for such reactive arthropathies is septic arthritis. Both Yersinia and Salmonella can cause septic arthritis, so an appropriate culture of synovial fluid should precede the diagnosis of reactive arthritis whenever possible. The course of reactive arthritis is variable, and few prognostic markers are available for the clinician to predict the course in an individual case. The majority of patients have an initial episode lasting 2 to 3 months, but synovitis may persist for a year or longer. In one 5-year follow-up of a point source cohort of post-Salmonella reactive arthritis, 20% of patients had ongoing inflammatory joint disease, and some degree of functional disability was observed in 30% of patients 5 years after the onset of disease.

REACTIVE ARTHRITIS AND HUMAN IMMUNODEFICIENCY VIRUS

An aggressive form of spondyloarthritis may be seen in patients who are concomitantly infected with HIV. There is no increased frequency of reactive

FIGURE 249-6.  Bilaterally asymmetrical sacroiliitis in reactive arthritis. Erosions, pseudowidening, and ileal sclerosis are present.

TABLE 249-4 ENTEROPATHIC ARTHRITIS FEATURE

PERIPHERAL ARTHRITIS

SACROILIITIS, SPONDYLITIS

CROHN DISEASE Frequency in Crohn disease

10-20%

2-7%

HLA-B27 associated

No

Yes

Pattern

Transient, symmetrical

Chronic

Course

Related to activity of Crohn disease

Unrelated to activity of Crohn disease

Effect of surgery

Remission of arthritis uncommon

No effect

Effect of anti-TNF therapy

Effective

Effective

ULCERATIVE COLITIS Frequency in ulcerative colitis 5-10%

2-7%

HLA-B27 associated

No

Yes

Pattern

Transient

Chronic

Course

More common in pancolitis Unrelated than proctitis; related to activity of ulcerative colitis

Effect of surgery

Remission of arthritis

No effect

HLA = human leukocyte antigen; TNF = tumor necrosis factor.

arthritis in patients with HIV, but HIV may alter the course of these arthropathies, with a tendency for a more aggressive and more refractory joint disease. Aggressive skin and joint disease may be seen in patients in whom psoriatic arthritis develops in the setting of HIV infection. Most North American patients with the HIV-reactive arthritis constellation are B27 positive, but studies of comparable patients in Africa have found a sizable B27-negative component in such patients. The arthritis in these patients falls into two clinical patterns: (1) an additive, asymmetrical polyarthritis or (2) an intermittent oligoarthritis that most commonly affects the lower extremities. Enthesitis, fasciitis, conjunctivitis, and urethritis can all be seen in such patients. Sacroiliitis can occur, although extensive spinal syndesmophyte formation is not common.

Enteropathic Arthritis  

DEFINITION

Enteropathic arthritis refers to the arthritis associated with Crohn disease or ulcerative colitis (Chapter 132; Table 249-4).  

PATHOBIOLOGY

The association of bowel inflammation and arthritis is supported by ileocolonoscopic studies in which subclinical inflammation of the bowel has been demonstrated in diseases covering the entire spectrum of spondyloarthritis.

CHAPTER 249  The Spondyloarthropathies  

1723

Histologic evaluation demonstrates that changes of acute ileitis are seen in postdysenteric reactive arthritis, whereas chronic inflammatory changes are more likely to be seen in patients with ankylosing spondylitis. Altered bowel permeability, with enhanced bacteremia or antigenemia, may provide the link in both cases.  

CLINICAL MANIFESTATIONS

All extraenteric manifestations, including arthritis, occur more commonly in Crohn disease than in ulcerative colitis. Peripheral arthritis occurs in 10 to 20% of Crohn disease patients and in 2 to 7% of ulcerative colitis patients. This pattern of arthritis occurs more commonly in patients with other extraenteric features (e.g., erythema nodosum, iritis). It is typically an inflammatory nonerosive polyarthritis, predominantly of large joints. In general, the clinical activity of the peripheral arthritis parallels the activity of the gut inflammation, and measures that control the GI disease usually control the joint disease as well. The peripheral arthritis of enteropathic arthritis is not associated with B27. In contrast, the sacroiliitis or spondylitis of enteropathic arthritis follows a pattern in which the joint inflammation waxes and wanes independently of the bowel inflammation. Axial disease occurs in 2 to 7% of both Crohn disease and ulcerative colitis patients. HLA-B27 is found in 50% of patients with axial arthritis. The course tends to be chronic, as opposed to the transient course of peripheral arthritis.  

DIAGNOSIS

It is important to recognize that the musculoskeletal features of enteropathic arthritis may precede any GI symptoms or signs. Conversely, the diarrhea preceding the onset of peripheral or axial arthritis in a young patient could just as likely represent a food-borne pathogen (e.g., Salmonella, Yersinia), with secondary reactive arthritis as inflammatory bowel disease and accompanying enteropathic arthritis. In the initial assessment of such a patient, it is important to carry out careful and complete stool cultures. If the GI symptoms persist, diagnostic colonoscopy is often required to resolve the issue.

Psoriatic Arthritis  

EPIDEMIOLOGY

One in five patients with psoriasis also have psoriatic arthritis.7 Although most cases arise in patients with established cutaneous disease, some patients (particularly children) have arthritis that antedates the appearance of the skin lesions. Although the extent of psoriatic skin disease correlates poorly with the development of arthritis, the risk for psoriatic arthritis increases with a family history of spondyloarthritis. The age at onset can range from 30 to 55 years, with an equal predilection for psoriatic arthritis in women and men. Psoriatic spondylitis has a slight male preponderance. Large prospective studies also suggest that obesity is a significant risk factor for psoriatic arthritis.  

PATHOBIOLOGY

The genetic associations with psoriatic arthritis are complex. Psoriasis itself is associated with several HLA loci; some B alleles have been reported, but the dominant element is HLA-Cw6. HLA-B39 and HLA-B27 have been associated with sacroiliitis and axial involvement. No etiologic agent has been proved in psoriatic arthritis, although some investigators have proposed that the disease process represents reactive arthritis in response to cutaneous bacteria. The histopathology of the synovitis of psoriatic arthritis is comparable to that of the other forms of spondyloarthritis, with the absence of the local production of immunoglobulin and rheumatoid factor differentiating this disease from rheumatoid arthritis.8 There is the potential for aggressive osteolysis, fibrous ankylosis, and heterotopic new bone formation to occur in psoriatic arthritis. As mentioned earlier, the coexistence of HIV and psoriatic arthritis seems to set the stage for an aggressive course of joint destruction in some patients. Patients with psoriatic arthritis can experience substantial physical impairment.9  

DIAGNOSIS

Psoriatic arthritis has a variable manifestation and disease course, but several clinical patterns have been identified in prospectively monitored cohorts of patients.10 The clinical subsets are not mutually exclusive, nor are they static over time. The most common form, which affects 30 to 50% of patients, is an asymmetrical oligoarthritis that may involve both large and small joints. Dactylitis, arising as sausage digits, can be seen in fingers and toes and actually represents an enthesitis. In the second subset there is selective targeting of the distal interphalangeal joints, seen in 10 to 15% of patients. These changes

FIGURE 249-7.  Nail pitting in psoriasis. The pits are more discrete and regular compared with pits affecting the nail plate in dermatitis.

are strongly associated with nail dystrophy, of which the features are onycholysis, subungual keratosis, pitting, and oil drop–like staining (Fig. 249-7). The third subset (15 to 30% of patients) has a symmetrical polyarthritis that mimics rheumatoid arthritis in many ways, except for the absence of rheumatoid nodules and rheumatoid factor. The fourth clinical variant is psoriatic spondylitis, which occurs in 20% of patients; 50% of such patients are B27 positive. Finally, arthritis mutilans (5% of patients) is a destructive, erosive arthritis that affects large and small joints and can be associated with marked deformities and significant disability. Radiographic changes in psoriatic arthritis involve soft tissue swelling (particularly in the case of dactylitis), erosions, and periostitis. Axial involvement may lead to the appearance of asymmetrical sacroiliitis with syndesmophytes that are bulky, asymmetrical, and nonmarginal. The classic “pencil-in-cup” deformity may be seen in patients with distal interphalangeal joint disease or arthritis mutilans. Acro-osteolysis is noted in a minority of patients and reflects an aggressive erosive process.

Differential Diagnosis

The diagnosis of psoriatic arthritis depends on finding the typical skin or nail changes in association with one of the articular variants described previously. The differential diagnosis for the skin lesions can include seborrheic dermatitis, dyshidrotic eczema, fungal infection, keratoderma blennorrhagicum, and palmoplantar pustulosis.

Undifferentiated Spondyloarthritis Despite careful clinical and radiographic assessment, there are still a substantial number of patients who do not fall into one of the classic diagnostic subsets of spondyloarthritis outlined previously. These patients are often defined as having undifferentiated spondyloarthritis with peripheral enthesitis, asymmetrical arthritis or sacroiliitis, or iritis in the absence of identifiable antecedent infection or concurrent inflammatory bowel disease or psoriasis. The natural history of undifferentiated spondyloarthritis has not been well defined, and case heterogeneity and diagnostic dilemmas plague a systematic or multicenter approach to the problem. When the clinical course is examined, a number of patients may finally meet the diagnostic criteria for ankylosing spondylitis, but many retain a distinct undifferentiated spondyloarthritis pattern for prolonged periods.

TREATMENT  General Measures

Spondyloarthritis necessitates a global approach to management in which education of patients is the cornerstone.11 Because the typical onset is during young adulthood, these patients may experience significant frustration or depression if their acute arthritis evolves into a chronic disease that significantly impairs their functional capabilities and quality of life. A clinician managing patients with spondyloarthritis should be aware that these psychosocial aspects are an important part of the burden of illness. Similarly, there may be important implications for the workplace, particularly if a job demands significant bending or twisting. It is important to include the mechanical demands of the workplace in the global assessment of patients with spondyloarthritis.

1724

CHAPTER 249  The Spondyloarthropathies  

Exercise is an important part of the treatment plan for patients with ankylosing spondylitis.12 Generally, high-impact sports should be avoided, whereas swimming is an ideal exercise. Stretching to maintain mobility and maintenance of posture should be emphasized, and an experienced physiotherapist can greatly assist in instructing patients in daily exercises. Long car trips and air travel should include periodic stretching. Sleep position should emphasize a straight back position rather than one curled on the side. Deep breathing exercises and avoidance of smoking should be stressed. One key area of concern for patients is prognosis, because spondyloarthritis, particularly reactive arthritis, often occurs in young, active individuals for whom athletic activity is a priority. There is general recognition that reactive arthritis has a greater propensity for chronicity than was previously appreciated, and this should temper an overly optimistic projection of the disease’s natural history. At the 5-year follow-up of a cohort of patients with Salmonella-induced reactive arthritis, two thirds continued to have subjective complaints, and one third demonstrated objective changes in their joints. The variability in prognosis for the large group of patients falling into the undifferentiated spondyloarthritis group is perplexing. At present, there is a lack of reliable predictors of progression in patients with this heterogeneous cluster of articular and extra-articular features.

Medical Therapies

Nonsteroidal Anti-inflammatory Drugs

In general, joint inflammation in spondyloarthritis improves significantly after the introduction of nonsteroidal anti-inflammatory drugs (NSAIDs). Indomethacin and diclofenac (up to 150 mg/day in divided doses) or naproxen (up to 1000 mg/day in divided doses) are generally well tolerated in this population. A1  These agents have to be used with caution in enteropathic arthritis because of concern about exacerbating possible underlying inflammatory bowel disease. In the case of ankylosing spondylitis, the goal with anti-inflammatory treatment is to achieve sufficient relief of pain and stiffness to allow an active, sustained program of exercise and physical activity that maintains posture and improves quality of life. Some studies have suggested that NSAIDs have diseasemodifying capability, but this effect appears to be restricted to those patients with elevated C-reactive protein (CRP) or erythrocyte sedimentation rate (ESR).

Biologic Therapies

The pathogenic role of immunomodulatory cytokines in the pathogenesis of spondyloarthritis has remained unresolved, but the advent of biologic agents has changed the landscape for spondyloarthritis. Biologic agents such as monoclonal antibodies to TNF-α (infliximab, adalimumab, golimumab, and certolizumab) or the soluble TNF receptor (etanercept) have been used in the treatment of spondyloarthritis. So far, these five anti-TNF agents have been comparably effective in trials of ankylosing spondylitis and psoriatic arthritis. A2  These studies have generally reported a prompt response in clinical outcome measures and in laboratory indicators of inflammation, and MRI evaluations have shown improvement in local inflammation in the sacroiliac joints and spine. The anti-TNF treatments have been well tolerated, with no significant incidence of serious adverse events, but patients appear to relapse when treatment is discontinued. In patients whose active nonradiographic axial spondyloarthritis went into sustained remission with adalimumab, continued therapy reduces subsequent flares compared with withdrawing treatment. A3  Experience with longer-term treatment with anti-TNF agents has been encouraging with regard to the persistence of the therapeutic effect and the infrequency of late adverse events. A4  These biologic agents also have been shown to retard radiographic progression.13 Other potential FDA-approved alternatives for ankylosing spondylitis include the interleukin-17A inhibitors secukinumab (given intravenously 10 mg/kg at weeks 0, 2, and 4, followed by 150 mg or 75 mg subcutaneously every 4 weeks starting at week 8) and ixekizumab (80 mg subcutaneously every 2 or 4 weeks), each of which significantly reduces the signs of ankylosing spondylitis. A5  A6  Filgotinib (an oral Janus kinase inhibitor, given as 200 mg daily for 12 weeks) is a promising potential option. A7  ,

Corticosteroids

The response to the intra-articular injection of steroids in the peripheral joints of patients with spondyloarthritis is often neither as dramatic nor as sustained as in those with rheumatoid arthritis. Corticosteroid injection into the sacroiliac joints is usually performed under imaging guidance (fluoroscopy or computed tomography [CT]). One study found that such injections resulted in a good response in 79% of patients and that the improvement could persist for many months. Systemic corticosteroids (either orally or via an intravenous bolus protocol) are not generally recommended but have been used for severe symptomatic flares, despite few controlled trials to validate their effectiveness. The goal should be prompt tapering of the dose when symptomatic control is achieved. The recognition that osteoporosis (Chapter 230) is a significant problem in ankylosing spondylitis provides further impetus to use corticosteroids sparingly. Topical steroids are usually effective for the treatment of the mucous membrane and skin manifestations of reactive arthritis. For uveitis, topical corticosteroid eye drops are an integral component of management, and treatment should be monitored jointly with an ophthalmologist.14

Drugs

Sulfasalazine

Randomized placebo-controlled trials have provided some support for the use of sulfasalazine, mostly in psoriatic arthritis. Three 36-week randomized double-blind multicenter studies of patients with ankylosing spondylitis, psoriatic arthritis, and reactive arthritis, respectively, were undertaken to compare sulfasalazine (2 g/day) with placebo in each case. An analysis of these studies stratified the patients into those having axial disease and those having peripheral disease. In patients with only axial disease, response criteria were met equally in the sulfasalazine group and the placebo group. In patients with peripheral arthritis, significantly superior responses were seen with sulfasalazine: 59% of the sulfasalazine group and 43% of the placebo group responded (P < .0005). A8  These findings are useful in guiding the selection of patients for sulfasalazine treatment. A recent study comparing sulfasalazine with etanercept in ankylosing spondylitis demonstrated the superiority of tumor necrosis factor (TNF) inhibitor therapy with respect to symptomatic improvement as well as MRI evidence of inflammation.

Methotrexate

Concurrent with the widespread use of methotrexate in patients with rheumatoid arthritis, there has been increasing use of methotrexate in patients with spondyloarthritis, but responses have been good only for peripheral joint disease. There is no evidence that methotrexate is effective for the spinal inflammation characteristic of ankylosing spondylitis, nor is there evidence that methotrexate changes the course of axial involvement in ankylosing spondylitis. Experience with long-term methotrexate therapy in patients with psoriatic arthritis has increased, although there has been little in the way of randomized controlled trials. Long-term follow-up may be required to resolve whether methotrexate has a joint-sparing effect in psoriatic arthritis.

Antibiotic Therapy

The current concept of the pathogenesis of reactive arthritis postulates that a bacterial infection, usually gastrointestinal (GI) or genitourinary (GU), is the triggering event in an immunogenetically susceptible host. For the other subsets of spondyloarthritis, there is less compelling evidence that infection plays a causal role. It is sound clinical practice to treat any culture-proven chlamydial urethritis in conjunction with treatment of the sexual partner. For this indication, a single 1-g dose of azithromycin is as effective as doxycycline 100 mg twice a day for 7 days. The role of antibiotics in the management of reactive arthritis has been controversial in reactive arthritis, A9  but rifampin/azithromycin or rifampin/doxycycline therapy may be useful for Chlamydia-induced reactive arthritis. A10 

  APPROACH TO SPECIFIC SPONDYLOARTHROPATHIES

Ankylosing Spondylitis

Consensus guidelines from the American College of Rheumatology/Spondylitis Association of America/Spondyloarthritis Research and Treatment Network can help guide medical therapy (Table 249-5).15

Psoriatic Arthritis Patients typically receive aggressive treatment for psoriasis (Chapter 409), and tight control of inflammation significantly improves outcomes. A11  The advent of biologic agents has had a major impact on the treatment of psoriatic arthritis. The anti-TNF agents have been studied most extensively, indicating the efficacy of infliximab, etanercept, adalimumab, and golimumab.16 For example, treatment of psoriatic arthritis with subcutaneous golimumab (50 mg to 100 mg every 4 weeks) inhibits the progression of structural damage, with continued clinical efficacy and safety through 1 year. A12  Other FDA-approved biologic agents can also be efficacious. Ustekinumab (a monoclonal antibody against interleukin-12/23) is safe, well tolerated, and reduces the extent and severity of psoriasis, as is guselkumab (a monoclonal antibody against the P19 subunit of interleukin-23, given as 100 mg subcutaneously at weeks 0 and 4, 12, 20, and 28, and then every 8 weeks). A13  Brodalumab, a human monoclonal antibody against interleukin-17 receptor A (IL17RA), significantly improves response rates among patients with psoriatic arthritis. A14  Secukinumab (10 mg/kg intravenously at weeks 0, 2, and 4, followed by either 150 mg or 75 mg subcutaneously every 4 weeks) A15  is another alternative. In patients whose psoriatic arthritis has responded inadequately to TNF inhibitors, tofacitinib (an oral Janus kinase inhibitor at 5 mg or 10 mg twice daily) is more effective than placebo over 3 months for reducing disease activity. A16  Filgotinib (a Janus kinase inhibitor at 200 mg orally daily for 16 weeks) is a potential future option. A17 

TABLE 249-5 TREATMENT RECOMMENDATIONS FOR THE SPONDYLOARTHROPATHIES Nonsteroidal anti-inflammatory drugs (NSAIDs) on demand (and chronically only if needed) If inadequate response to trials of two different NSAIDs, add a tumor necrosis factor (TNF)-α inhibitor and try to discontinue NSAIDs if the response is good; options are: etancercept (50 mg subcutaneously once weekly) infliximab* (5 mg/kg intravenously at 0, 2, and 6 weeks, then every 6 or 8 weeks†) adalimumab* (40 mg subcutaneously every other week) golimumab* (50 mg subcutaneously once a month) certolizumab (400 mg subcutaneously initially and at 2 and 4 weeks, followed by 200 mg every other week or 400 mg every 4 weeks) If no response to trials of two different TNF-α inhibitors or contraindication to using them, add a slow-acting anti-rheumatic drug sulfasalazine (2 to 3 g orally daily in evenly divided doses with dosage intervals not exceeding 8 hours) or methotrexate (10 to 25 mg once weekly) for psoriatic arthritis and enteropathic arthritis Short-term treatment with locally injected corticosteroids is useful, but systemic corticosteroids are not recommended *Preferred over etanercept in patients with inflammatory bowel disease. † 6 weeks for ankylosing spondylitis; 8 weeks for psoriatic arthritis. Adapted from Ward MM, Deodhar A, Akl EA, et al. American College of Rheumatology/ Spondylitis Association of America/Spondyloarthritis Research and Treatment Network 2015 recommendations for the treatment of ankylosing spondylitis and nonradiographic axial spondyloarthritis. Arthritis Rheumatol. 2016;68:282-298.

As a general approach, nonsteroidal anti-inflammatory drugs are recommended to relieve musculoskeletal signs and symptoms. Treatment with disease-modifying drugs—such as methotrexate, sulfasalazine, or leflunomide— is recommended in patients with swollen joints, structural damage in the presence of inflammation, or clinically relevant extra-articular manifestations. Anti-TNF agents are recommended in patients with active enthesitis or dactylitis and insufficient response to other medications. Patients should be switched to another anti-TNF agent if the first is not successful. Interleukin-17 receptor antibodies and other newer agents currently are usually reserved for otherwise poorly responsive patients.

Reactive Arthritis The treatment of reactive arthritis begins with therapy of the triggering infection. For example, in patients who are PCR-positive either in blood or joint fluid for C. trachomatis or C. pneumonia, a 6-month course of combination therapy with rifampicin (300 mg/day) plus either doxycycline (200 mg/day) or azithromycin (500 mg/day followed by 5 days of 2 to 500 mg once a week) yields a response in 63% of patients (versus 22% with placebo), and complete remission is observed in 20% (versus 0% with placebo). For spondyloarthritis, NSAIDs and glucocorticoid injections are effective. In more severe cases, sulfasalazine is often effective if started in the first three months. Biologic therapies are often used in chronic HLA-B27 positive spondyloarthritis, although their efficacy is unproven, especially since about 50% of patients recover within about six months.

Enteropathic Arthritis For enteropathic arthritis, sulfasalazine is more helpful in ulcerative colitis than in Crohn disease (Chapter 132), whereas the opposite is generally true for methotrexate. TNF-α inhibitors (see Table 249-5) are effective in spondyloarthritis related to both ulcerative colitis and Crohn disease.17 A multidisciplinary approach has been shown to improve the management of rheumatic disease in patients with inflammatory bowel disease, allowing more comprehensive care.18

  Grade A References A1. Kroon FP, van der Burg LR, Ramiro S, et al. Non-steroidal anti-inflammatory drugs (NSAIDs) for axial spondyloarthritis (ankylosing spondylitis and non-radiographic axial spondyloarthritis). Cochrane Database Syst Rev. 2015;7:CD010952. A2. Maxwell LJ, Zochling J, Boonen A, et al. TNF-alpha inhibitors for ankylosing spondylitis. Cochrane Database Syst Rev. 2015;4:CD005468.

A3. Landewé R, Sieper J, Mease P, et al. Efficacy and safety of continuing versus withdrawing adalimumab therapy in maintaining remission in patients with non-radiographic axial spondyloarthritis (ABILITY-3): a multicentre, randomised, double-blind study. Lancet. 2018;392:134-144. A4. Song IH, Hermann KG, Haibel H, et al. Consistently good clinical response in patients with early axial spondyloarthritis after 3 years of continuous treatment with etanercept: longterm data of the ESTHER trial. J Rheumatol. 2014;41:2034-2040. A5. Baeten D, Sieper J, Braun J, et al. Secukinumab, an interleukin-17A inhibitor, in ankylosing spondylitis. N Engl J Med. 2015;373:2534-2548. A6. van der Heijde D, Cheng-Chung Wei J, Dougados M, et al. Ixekizumab, an interleukin-17A antagonist in the treatment of ankylosing spondylitis or radiographic axial spondyloarthritis in patients previously untreated with biological disease-modifying anti-rheumatic drugs (COAST-V): 16 week results of a phase 3 randomised, double-blind, active-controlled and placebo-controlled trial. Lancet. 2018;392:2441-2451. A7. van der Heijde D, Baraliakos X, Gensler LS, et al. Efficacy and safety of filgotinib, a selective Janus kinase 1 inhibitor, in patients with active ankylosing spondylitis (TORTUGA): results from a randomised, placebo-controlled, phase 2 trial. Lancet. 2018;392:2378-2387. A8. Braun J, van der Horst-Bruinsma IE, Huang F, et al. Clinical efficacy and safety of etanercept versus sulfasalazine in patients with ankylosing spondylitis: a randomized, double-blind trial. Arthritis Rheum. 2011;63:1543-1551. A9. Barber CE, Kim J, Inman RD, et al. Antibiotics for treatment of reactive arthritis: a systematic review and metaanalysis. J Rheumatol. 2013;40:916-928. A10. Carter JD, Espinoza LR, Inman RD, et al. Combination antibiotics as a treatment for chronic Chlamydia-induced reactive arthritis: a double-blind, placebo-controlled, prospective trial. Arthritis Rheum. 2010;62:1298-1307. A11. Coates LC, Moverley AR, McParland L, et al. Effect of tight control of inflammation in early psoriatic arthritis (TICOPA): a UK multicentre, open-label, randomised controlled trial. Lancet. 2015;386:2489-2498. A12. Kavanaugh A, van der Heijde D, McInnes IB, et al. Golimumab in psoriatic arthritis: one-year clinical efficacy, radiographic, and safety results from a phase III, randomized, placebo-controlled trial. Arthritis Rheum. 2012;64:2504-2517. A13. Deodhar A, Gottlieb AB, Boehncke WH, et al. Efficacy and safety of guselkumab in patients with active psoriatic arthritis: a randomised, double-blind, placebo-controlled, phase 2 study. Lancet. 2018;391:2213-2224. A14. Mease PJ, Genovese MC, Greenwald MW, et al. Brodalumab, an anti-IL17RA monoclonal antibody, in psoriatic arthritis. N Engl J Med. 2014;370:2295-2306. A15. Mease PJ, McInnes IB, Kirkham B, et al. Secukinumab inhibition of interleukin-17A in patients with psoriatic arthritis. N Engl J Med. 2015;373:1329-1339. A16. Gladman D, Rigby W, Azevedo VF, et al. Tofacitinib for psoriatic arthritis in patients with an inadequate response to TNF inhibitors. N Engl J Med. 2017;377:1525-1536. A17. Mease P, Coates LC, Helliwell PS, et al. Efficacy and safety of filgotinib, a selective janus kinase 1 inhibitor, in patients with active psoriatic arthritis (EQUATOR): results from a randomised, placebocontrolled, phase 2 trial. Lancet. 2018;392:2367-2377.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 249  The Spondyloarthropathies  

GENERAL REFERENCES 1. Brown MA, Wordsworth BP. Genetics in ankylosing spondylitis—current state of the art and translation into clinical outcomes. Best Pract Res Clin Rheumatol. 2017;31:763-776. 2. Taurog JD, Chhabra A, Colbert RA. Ankylosing spondylitis and axial spondyloarthritis. N Engl J Med. 2016;374:2563-2574. 3. Khmelinskii N, Regel A, Baraliakos X. The role of imaging in diagnosing axial spondyloarthritis. Front Med (Lausanne). 2018;5:1-11. 4. Duba AS, Mathew SD. The seronegative spondyloarthropathies. Prim Care. 2018;45:271-287. 5. Dashti N, Mahmoudi M, Aslani S, et al. HLA-B*27 subtypes and their implications in the pathogenesis of ankylosing spondylitis. Gene. 2018;670:15-21. 6. Schmitt SK. Reactive arthritis. Infect Dis Clin North Am. 2017;31:265-277. 7. Alinaghi F, Calov M, Kristensen LE, et al. Prevalence of psoriatic arthritis in patients with psoriasis: a systematic review and meta-analysis of observational and clinical studies. J Am Acad Dermatol. 2019;80:251-265. 8. Veale DJ, Fearon U. The pathogenesis of psoriatic arthritis. Lancet. 2018;391:2273-2284. 9. Mease P, Strand V, Gladman D. Functional impairment measurement in psoriatic arthritis: importance and challenges. Semin Arthritis Rheum. 2018;48:436-448. 10. Ritchlin CT, Colbert RA, Gladman DD. Psoriatic arthritis. N Engl J Med. 2017;376:957-970.

1725.e1

11. Gossec L, Smolen JS, Ramiro S, et al. European League Against Rheumatism (EULAR) recommendations for the management of psoriatic arthritis with pharmacological therapies: 2015 update. Ann Rheum Dis. 2016;75:499-510. 12. Zão A, Cantista P. The role of land and aquatic exercise in ankylosing spondylitis: a systematic review. Rheumatol Int. 2017;37:1979-1990. 13. Van den Bosch F, Coates L. Clinical management of psoriatic arthritis. Lancet. 2018;391:2285-2294. 14. Gupta N, Agarwal A. Management of uveitis in spondyloarthropathy: current trends. Perm J. 2018;22:1-19. 15. Ward MM, Deodhar A, Akl EA, et al. American College of Rheumatology/Spondylitis Association of America/Spondyloarthritis Research and Treatment Network 2015 recommendations for the treatment of ankylosing spondylitis and nonradiographic axial spondyloarthritis. Arthritis Rheumatol. 2016;68:282-298. 16. Singh JA, Guyatt G, Ogdie A, et al. Special Article: 2018 American College of Rheumatology/ National Psoriasis Foundation guideline for the treatment of psoriatic arthritis. Arthritis Rheumatol. 2019;71:5-32. 17. Peluso R, Manguso F, Vitiello M, et al. Management of arthropathy in inflammatory bowel diseases. Ther Adv Chronic Dis. 2015;6:65-77. 18. Conigliaro P, Chimenti MS, Ascolani M, et al. Impact of a multidisciplinary approach in enteropathic spondyloarthritis patients. Autoimmun Rev. 2016;15:184-190.

1725.e2

CHAPTER 249  The Spondyloarthropathies  

REVIEW QUESTIONS 1. A 28-year-old man presents with a 2-year history of low back pain and intermittent arthritis in the lower extremity. Physical examination demonstrates bilateral Achilles tendonitis and tenderness to direct pressure over the sacroiliac joints. Which of the following would not support the diagnosis of spondyloarthritis (SpA)? A . Presence of HLA-B27 B. History of anterior uveitis C. Positive rheumatoid factor D. Elevated CRP E. History of inflammatory bowel disease Answer: C  The history and the other optional features would be entirely consistent with a diagnosis of SpA. HLA-B27 is strongly associated with SpA in general, and its strongest association is with ankylosing spondylitis (AS). Anterior uveitis occurs in approximately 30% of AS cases, and is itself associated with HLA-B27 even when AS is not present. Positive rheumatoid factor is characteristic of rheumatoid arthritis but not of SpA, so it would not be expected to be positive in such patients. Elevated C-reactive protein (CRP) is a nonspecific marker of inflammation and would be commonly seen in patients with active SpA. Inflammatory bowel disease occurs in approximately 10% of AS cases, and its presence strongly suggests that the underlying arthritis is part of the SpA spectrum. 2. A 32-year-old man presents with pain and swelling of one ankle and one knee. The history reveals that he had just returned from a trip to South America and developed diarrhea that was now subsiding. He denied any eye inflammation or skin lesions. Physical examination confirmed tenderness and swelling of an ankle and a knee. The next appropriate diagnostic test would be which of the following? A . Aspiration of the knee swelling and culture of the synovial fluid B. Initiate a course of broad-spectrum antibiotics. C. X-rays of the lumbar and cervical spine D. Check a blood test for the presence of antinuclear antibodies. E. Initiate an empirical trial of methotrexate. Answer: A  A patient with new-onset swelling in the joints, particularly when one or two joints are affected, should always raise the possibility of septic arthritis. When there has been a recent infection elsewhere in the body, it is possible the inciting infection could spread by a hematogenous route to seed the joint. Although the history is suggestive of reactive arthritis (ReA), it is important to rule out septic arthritis with certainty before entertaining that diagnosis. Culture of synovial fluid is the definitive test to rule out septic arthritis. Empirical use of broad-spectrum antibiotics rarely has a place in the management of new-onset arthritis. The first step is to make a diagnosis. X-rays of the spine are important in the evaluation of inflammatory back pain, but in this case there is only peripheral joint involvement. If the course of the arthritis becomes chronic or back pain develops over time, x-rays at that point would be appropriate. Antinuclear antibodies in the serum are the hallmark of many autoimmune rheumatic diseases—in particular, lupus. But with the clinical presentation in this case, there are no features to implicate lupus or related conditions, and it would not be appropriate to search for antinuclear antibodies. The immediate priority is to rule out septic arthritis. It would be inappropriate to institute methotrexate before that is resolved.

3. A 30-year-old man presents with worsening back pain. The history reveals that this has been present for the past 3 years and is increasing in severity. It is characterized by early-morning stiffness, nocturnal pain, and modest improvement with exercise. He initially noted an improvement with an NSAID, but subsequently that effect was lost, and he has tried two other NSAIDs without effect. Physical examination reveals limitation in forward flexion of the lumbar spine. An AP x-ray of the pelvis reveals bilateral erosive changes in the sacroiliac joints. A lateral x-ray of the lumbar spine reveals squaring of several vertebral bodies. The next step in pharmacologic treatment of this condition is which of the following? A . Methotrexate B. Leflunomide C. Sulfasalazine D. TNF inhibitor biologic agents E. Azathioprine Answer: D  The patient meets the diagnostic criteria for ankylosing spondylitis (AS). He has long-standing back pain that has features of inflammatory back pain and has limitation in spinal mobility. The presence of bilateral erosive sacroiliitis provides imaging support for the diagnosis of AS. He has had an adequate trial of NSAIDs, and his back pain has not responded adequately to these agents. The classical DMARDs used in rheumatoid arthritis—methotrexate, leflunomide, sulfasalazine, and azathioprine—have proved ineffective in controlling the spinal inflammation seen in AS. On the other hand, the TNF inhibitors (infliximab, etanercept, adalimumab, golimumab) have established efficacy in controlling the signs and symptoms of AS. 4. A 35-year-old woman is under your care for ankylosing spondylitis (AS). She is HLA-B27 positive and has no extra-articular features accompanying her back pain. Her symptoms have been controlled with NSAIDs in combination with physiotherapy and exercises. She has an 8-year-old son, and she is inquiring about his current and future health prospects. Which of the following would be advisable in this circumstance? A . The son should be checked for HLA-B27 status. B. The mother should be advised that the likelihood of her son developing AS is less than 10%. C. The son should be advised to avoid sports and physical activities. D. Preventive low-dose NSAIDs should be instituted for the son. E. The son should be advised to avoid foreign travel in order to avoid food-borne pathogens. Answer: B  There is a 50% chance that any particular child of a B27-positive patient will be B27 positive, but there is only a 20% chance of a B27-positive individual with a positive family history developing AS. So the likelihood of a child developing AS when he has a B27-positive parent with AS is maximally 10%. It would not be advisable to test an asymptomatic child for HLA-B27, because it has low predictive value (as discussed earlier) and could introduce chronic anxiety about future disease, which is unwarranted. There is no reason to limit the sports and recreation of the child. Maintaining general fitness and muscle strength and posture is sound advice even if there is a greater chance of later arthritis than in the general population. There is no evidence that sports-related injuries trigger AS. There is no evidence that NSAIDs prevent the subsequent development of SpA in individuals who are genetically predisposed to SpA. Chronic NSAID therapy carries its own risks and would not be justified in an asymptomatic individual. There is a theoretical concern that being B27 positive could increase the risk of reactive arthritis after a food-borne pathogen illness. This might lead to advising a patient with prior diagnosis of ReA to minimize exposure to food-borne pathogens, but it would be inappropriate to restrict travel for anyone who has never had an episode of ReA.

CHAPTER 249  The Spondyloarthropathies  

5. A 19-year-old female patient has developed a painful swollen knee over the course of 7 days. There is no history of trauma and no associated skin lesions or ocular complaints. The patient relates a history that she has just returned from a trip abroad with the family and had a transient diarrheal episode while traveling. Further questioning reveals that she has had two prior episodes of diarrhea in the past year, each associated with crampy abdominal pain, but on each occasion the diarrhea was self-limited and she did not seek medical attention. Which of the following would not be an appropriate next step in the management of this patient? A . Referral to a gastroenterologist for ileocolonoscopy B. Culture of a stool sample C. Aspiration of the knee and culture of synovial fluid D. Check hemoglobin and ESR. E. Institute a 2-week course of ciprofloxacin.

1725.e3

Answer: E  The patient has developed arthritis in the setting of a recurrent diarrheal course. The key differential is between reactive arthritis (ReA) and enteropathic arthritis (EA) as part of inflammatory bowel disease (IBD). In either case, it would be inappropriate to institute an empirical course of antibiotics before a diagnosis is made. In neither postdysenteric ReA nor in IBDassociated arthritis is there evidence that antibiotics alter the course of the arthritis. It would be appropriate in this case to determine whether the recurrent diarrhea could be a manifestation of Crohn disease or ulcerative colitis, and ileocolonoscopy would be indicated to address that question. Recurrent diarrhea might reflect an infectious agent such as Salmonella. Stool cultures would be an appropriate test to address that. Culture of synovial fluid is the definitive test to rule out septic arthritis. Both Salmonella and Yersinia can cause a septic arthritis, so it is appropriate to exclude an infection in the joint before proceeding to other management decisions. Anemia of chronic disease might raise the suspicion that a chronic process, such as IBD, has been present for some time. Elevation in CRP or ESR is appropriate in the work-up of this patient, although this would not differentiate ReA from IBD-related arthritis. These acute phase reactants can serve as useful surrogate markers to monitor response to treatment.

1725

CHAPTER 250  Systemic Lupus Erythematosus  

250  SYSTEMIC LUPUS ERYTHEMATOSUS MARY K. CROW



DEFINITION

Systemic lupus erythematosus (SLE) is a multisystem autoimmune disease that results from immune system–mediated tissue damage. Manifestations of SLE can involve the skin, joints, kidney, central nervous system (CNS), cardiovascular system, serosal membranes, and hematologic and immune systems. The disease is highly heterogeneous, with individual patients manifesting variable combinations of clinical features. In most patients with SLE, the disease is characterized by a waxing and waning clinical course, although some demonstrate a pattern of chronic activity. The molecular triggers of the disease are not known, but the pathogenesis involves the production of autoantibodies specific for nucleic acids and nucleic acid–binding proteins. Immune complexes, along with immune system cells and soluble mediators, generate inflammation and tissue damage. Therapeutic approaches generally involve immunosuppression, although promising biologic agents targeting specific molecular mechanisms are in development.  

EPIDEMIOLOGY

A notable feature of SLE is that it occurs much more frequently in females than in males. Like Hashimoto thyroiditis and Sjögren syndrome, the femaleto-male ratio is approximately 8 : 1 to 10 : 1 in adults, and most cases are diagnosed between the ages of 15 and 44 years. In children and women older than

1725.e4

ABSTRACT

CHAPTER 250  Systemic Lupus Erythematosus  

Systemic lupus erythematosus (SLE) is a highly complex systemic autoimmune disease that involves nearly all components of the immune system and is associated with damage to multiple organs. SLE is notable for the female skewing of disease, with about 8 to 10 females for every 1 male affected. The disease is characterized by autoimmunity specific for nucleic acids and nucleic acid–binding proteins. Current concepts in disease pathogenesis involve nucleic acid–mediated activation of the innate immune system, production of cytokines such as type I interferon, extensive immune system dysregulation, and production of autoantibodies that form immune complexes. Therapy typically depends on immunosuppression with corticosteroids and other agents. Current studies are investigating the potential for biologic therapies to provide improved efficacy over current treatment options.

KEYWORDS

autoimmunity autoantibodies type I interferon lupus nephritis malar rash premature atherosclerosis

1726

CHAPTER 250  Systemic Lupus Erythematosus  

55 years, the ratio is closer to 2 : 1. The prevalence of SLE in the United States is estimated to be approximately 62.2 to 84.8 per 100,000, and the incidence of new cases is 4.6 to 5.6 per 100,000 per year.1,2 The prevalence, severity, and characteristics of disease differ in different ethnic groups, with SLE being three- to four-fold more prevalent in African American and American Indian women than in white women.3 The severity of disease is also greater in Hispanic individuals than in whites, although data for Hispanic populations are less abundant.4 Asians may also have a higher prevalence of disease than whites. Studies of lupus in minority populations indicate that socioeconomic factors are major contributors to the increased prevalence and severity of disease in African Americans and Hispanic Americans.

nucleic acids to innate immune system activation and lupus pathogenesis.11 The available data suggest a common theme: the genes associated with lupus confer either increased activation or impaired regulation of innate or adaptive immune responses, with increased type I interferon often observed in association with the risk genotype.

Environmental Triggers

Current understanding of lupus pathogenesis incorporates roles for genetic susceptibility based on a cumulative hit model involving multiple genes5; environmental triggers, including microbial infection, sunlight, and certain drugs; and altered immune system function. Recent advances in immunology have focused attention on the mechanisms that account for innate immune system activation.6 At least some of the genetic and environmental contributions to lupus are likely to promote innate immune system activation and subsequent autoimmunity. Others may contribute to inflammation and tissue damage. Induction of cellular stress responses, including oxidative modification of cell proteins, is of current interest as a mechanism that links environmental triggers to altered immune function. Murine models have proved useful in identifying genes that could contribute to lupus susceptibility or define patterns of disease. Production of autoantibodies characteristic of SLE and development of nephritis and accelerated death have been demonstrated in numerous murine strains in which immune system genes have been modified. In most cases, no alterations have been noted in the homologous human genes. The ease of induction of lupus-like disease in murine models suggests that there are numerous possible pathogenic paths that might lead to the clinical manifestations of the disease. One prominent immune mechanism involves components of the immune response to viral infection, particularly the type I interferon response, that are associated with lupus in both murine and human systems and are likely to be important in disease pathogenesis.7

Several classes of potential environmental triggers for lupus have been studied.12 Although the female preponderance of SLE implies a role for hormonal factors in the disease, recent concepts describe a contribution of epigenetic modification or dosage effects of the X chromosome as accounting for at least some of the sex skewing. A role for microbial triggers—particularly virus infection— has been postulated, consistent with the constitutional symptoms that often characterize the earliest stage of the disease. Epstein-Barr virus has garnered particular interest among investigators because the frequency of previous infection in SLE patients is significantly higher than in the general population (99 vs. 94%). Evidence of exposure to other viruses, including cytomegalovirus, is equivalent between SLE patients and healthy control subjects. Recent data implicate orthologs of human Ro60, a common target of lupus autoantibodies, in commensal bacteria and demonstrate cross-reactivity of human Ro-specific T-cell clones with bacteria-derived Ro60.13 Ultraviolet light exposure is a welldescribed trigger of lupus flares. Possible mechanisms include DNA damage, induction of cellular stress responses, and induction of apoptosis of skin cells, which result in concentration of nucleic acids and associated proteins in cell membrane blebs that can be processed by antigen-presenting cells. Data also support an association between current tobacco use and anti-double-stranded DNA antibodies and lupus disease activity. Certain drugs, including procainamide and hydralazine, can induce a lupus-like syndrome, but the symptoms usually abate after discontinuing use of the drug. These agents may promote demethylation of DNA, thereby altering gene expression and potentially increasing the availability of immunostimulatory DNA. Sulfa antibiotics have been reported to induce lupus flare in some patients. Administration of recombinant interferon-α to patients with hematologic malignancies or hepatitis C infection has been associated with induction of a lupus-like syndrome. In addition, anti–tumor necrosis factor agents have induced lupus autoantibodies and occasionally clinical lupus in patients with rheumatoid arthritis.

Genetics

Immunologic Triggers



PATHOBIOLOGY

An important role for a genetic contribution to lupus susceptibility is suggested by the high concordance of disease in monozygotic twins (24%). Rare mutations in genes encoding components of the complement pathway, including C1q, C2, and C4A, contribute to increased lupus susceptibility or severity. Impaired production of these early complement components may decrease the clearance of apoptotic cells, thereby augmenting the pool of available autoantigens, or decrease the solubility of immune complexes. Association of SLE with the major histocompatibility complex (MHC) class II alleles human leukocyte antigen (HLA)-DR3 (DRB1*03:01-DQA1*05:01-DQB1*02:01) and HLA-DR15 (DRB1*15:01/03-DQA1*01:02-DQB1*06:01) shows the strongest risk, and these class II alleles are associated with the production of particular autoantibodies. A large transancestral genome-wide association study (GWAS) of subjects of European, African, and Hispanic Amerindian ancestry has identified 80 non-HLA single-nucleotide polymorphisms associated with SLE at a false discovery rate P value < .001 in a meta-analysis. Some associations at a high level of significance were related to ancestry (E-Table 250-1).8,9 Polymorphic variants in components of the toll-like receptor (TLR) pathways that regulate type I interferon production, including interferon regulatory factor 5 (IRF5) and IRF7, are associated with a diagnosis of SLE and increased plasma interferon activity in some populations. Polymorphisms in the Fc receptor genes FCGR2A and FCGR3A have been associated with SLE nephritis, possibly based on altered clearance of immune complexes. Variants of the PTPN22 gene, which encodes a phosphatase that regulates T-cell activation, are also associated with SLE. GWASs have identified variations in regulators of innate immune system activation (e.g., TNFAIP3, ITGAM, IFIH1) and signaling molecules important in lymphocyte activation (e.g., STAT4, BANK1, and BLK). Rare mutations in genes encoding proteins that regulate nucleic acid integrity and degradation, including TREX1, encoding a DNase; SAMHD1, a triphosphohydrolase; RNASEH2A, B and C; and ADAR, an RNA-specific adenosine deaminase, have been documented in some patients with a lupus-like disorder called Aicardi-Goutieres syndrome, characterized by skin lesions, CNS disease, autoantibodies, and high levels of interferon.10 Mutations in these genes have also been documented in rare patients with SLE and have provided new insights into the likely contribution of endogenous

Genetic and environmental factors that increase the probability of development of SLE are likely to act on the immune system to induce autoimmunity and consequent tissue inflammation and damage.14 In addition to mechanisms that increase the availability of self-antigens (such as ultraviolet light), altered expression of gene products that mediate or regulate apoptosis, or impaired clearance of apoptotic debris, results in generalized activation of the immune system and contributes to autoimmunity in lupus. In parallel with the events that account for effective immune responses directed at exogenous microbes, the autoimmunity that occurs in SLE patients is likely to require activation of both innate and adaptive immune responses. The innate immune system (Chapter 39) recognizes common molecular patterns expressed on the microbe and augments antigen-presenting cell capacity and successful generation of an antigen-specific adaptive immune response. The characterization of the TLR family of pattern recognition receptors has provided new understanding of the mechanisms through which the innate immune system is activated by exogenous and endogenous stimuli, including nucleic acid–containing immune complexes, and promotes induction of a self-directed adaptive immune response (Chapter 40).

Type I Interferon

Studies of gene expression in peripheral blood mononuclear cells of SLE patients using microarray and RNA sequencing technology have demonstrated a sustained and broad “signature” of type I interferon–induced gene transcripts that reflect innate immune system activation. Interferon-α (IFN-α), along with other type I IFNs (e.g., IFN-β, IFN-ω) may be responsible for many of the immunologic alterations observed in SLE and is identified as a promising therapeutic target. Immune complexes containing DNA or RNA are postulated to induce the production of type I interferon in SLE. Demethylated CpG-rich DNA or RNA associated with nucleic acid–binding proteins can activate plasmacytoid dendritic cells and other immune system cells through TLRs and thereby result in the production of type I interferon (IFN-α or IFN-β) and other proinflammatory cytokines (E-Fig. 250-1). Sensing of intracellular RNA or DNA by cytosolic nucleic acid sensors represents another potential molecular pathway leading to type I interferon production. Diverse effects of

CHAPTER 250  Systemic Lupus Erythematosus  

E-TABLE 250-1 NON-HLA GENOMIC REGIONS ASSOCIATED WITH SYSTEMIC LUPUS ERYTHEMATOSUS BY ANCESTRY7 GENE* EUROPEAN ATG5 BANK1 BLK CLEC16A DEF6 DGKQ ETS1 FCGR2A GRB2 GTF2I IFIH1 IKZF1 IKZF3 IL10 IL12A IRF5 IRF7 IRF8 ITGAM JAZF1 MSRA NMNAT2 OLIG3 PKIA PTPN22 PTTG1-MIR146A PXK SLC15A4 SLC17A4 STAT4 TMEM39A

PROTEIN

TIMMDC1 TNFAIP3 TNFSF4 TNIP1 TYK2 UBE2L3 WDFY4

Autophagy related 5 B-cell scaffold protein with ankyrin repeats 1 B-lymphocyte-specific tyrosine kinase C-type lectin domain containing 16A DEF6, guanine nucleotide exchange factor Diacylglycerol kinase theta ETS proto-oncogene 1 Fc fragment of IgG receptor IIa Growth factor receptor bound protein 2 General transcription factor IIi Interferon induced with helicase C domain 1 IKAROS family zinc finger 1 IKAROS family zinc finger 3 Interleukin 10 Interleukin 12A Interferon regulatory factor 5 Interferon regulatory factor 7 Interferon regulatory factor 8 Integrin subunit alpha M JAZF zinc finger 1 Methionine sulfoxide reductase A Nicotinamide nucleotide adenylyltransferase 2 Oligodendrocyte transcription factor 3 CAMP-dependent protein kinase inhibitor alpha Protein tyrosine phosphatase nonreceptor type 22 Pituitary tumor-transforming 1 PX domain containing serine/threonine kinase like Solute carrier family 15 member 4 Solute carrier family 17 member 4 Signal transducer and activator of transcription 4 Transmembrane protein 39A-translocase of inner mitochondrial membrane Mitochondrial membrane domain containing 1 TNF alpha induced protein 3/A20 TNF superfamily member 4/Ox40 ligand TNFAIP3 interacting protein 1 Tyrosine kinase 2 Ubiquitin conjugating enzyme E2 L3 WDFY family member 4

AFRICAN BLK IRF5 ITGAM PLAT PTTG1

B-lymphocyte-specific tyrosine kinase Interferon regulatory factor 5 Integrin subunit alpha M Plasminogen activator, tissue type Pituitary tumor-transforming 1

HISPANIC CLEC16A GALC IRF5 ITGAM NCF2 TNIP1 STAT4

C-type lectin domain containing 16A Galactosylceramidase Interferon regulatory factor 5 Integrin subunit alpha M Neutrophil cytosolic factor 2 TNFAIP3 interacting protein 1 Signal transducer and activator of transcription 4

*Genes listed have P value < 5 × 10−8. From Langefeld CD, Ainsworth HC, Cunninghame Graham DS, et al. Transancestral mapping and genetic load in systemic lupus erythematosus. Nat Commun. 2017 Jul 17;8:1-18.

1726.e1

1726.e2

CHAPTER 250  Systemic Lupus Erythematosus  

Cytosol Endosome TLR3

TLR7

TLR8

TLR9

PP PPP RIG-1 Mitochondrion

MYD88 IRAK4 IRAK1 or IRAK2

TRIF

ER

TBK1 NF-κB

Natural ligands: ssRNA dsRNA DNA

IRF3

Agonists in development: Low molecular weight compound Oligonucleotide

cGAS AIM2 MDA5 ATP cGAMP GTP MAVS

ASC

TBK1

IRF7

Pro-inflammatory cytokines

STING

IRF3

Type 1 IFNs Nucleus

Caspase 1

IL-1β and IL-18

E-FIGURE 250-1.  Model for induction of innate immune system activation in systemic lupus erythematosus. Both exogenous and endogenous stimuli can induce toll-like receptor (TLR) and cytosolic nucleic acid sensor activation and thereby result in new gene transcription. Among potential endogenous ligands are immune complexes containing DNA or RNA or matrix-derived components and cytosolic DNA or RNA. TLR ligands trigger the activation of intracellular adaptors, including TIR domain-containing adapter-inducing interferon-β (Trif), Trif-related adaptor molecule (TRAM), TIR domain-containing adapter protein (TIRAP), or myeloid differentiation primary response protein 88 (MyD88), and induce transcription of type I interferons or inflammatory cytokines. Cytosolic RNA activates RIG-I or MDA5, which signal through MAVS and TBK1. Cytosolic DNA activates cGAS, which signals through cGAMP, STING, and TBK1. Both pathways can induce type I interferon transcription. Type I interferons mediate diverse effects on immune system cells, including maturation of dendritic cells, increased immunoglobulin (Ig) class switching, and induction of cytokines that promote autoimmunity and inflammation, including B-lymphocyte stimulator (BLyS), interleukin-10 (IL-10), interferon-γ, and chemokines. ds = double-stranded; ER = endoplasmic reticulum; ss = single-stranded. (From Junt T, Barchet W. Translating nucleic acid-sensing pathways into therapies. Nat Rev Immunol. 2015;15:529-44.)

1727

CHAPTER 250  Systemic Lupus Erythematosus  

type I interferon on immune system function are consistent with the altered immune responses observed in SLE patients, including maturation of dendritic cells, increased immunoglobulin class switching to mature immunoglobulin isotypes (immunoglobulin G [IgG] and IgA), induction of soluble mediators that increase B-cell differentiation and inflammatory responses, such as B-lymphocyte stimulator (BLyS) and IFN-γ, and modulation of effector T-cell programs. Induction of an immunostimulatory microenvironment by IFN-α may support the development of a humoral immune response directed at self-antigens, particularly intracellular particles that contain nucleic acids and nucleic acid–binding proteins. It is not known why some individuals initiate immune system activation directed at self-antigens and others do not. In addition to its effects on immune system function, type I interferon has been associated with altered endothelial cell function and microglial function in the brain and may contribute to the development of atherosclerotic vascular pathology and CNS disease in patients with lupus.15,16

Autoantibodies

The most characteristic lupus autoantibodies target intracellular particles containing both nucleic acid and nucleic acid–binding proteins. Understanding the significance of induction of these particular autoantibody specificities may provide clues to the etiology of SLE. An analysis of the spectrum of autoantibodies present in the sera of individuals in whom SLE is later diagnosed has suggested that autoantibodies reactive with certain RNA-binding proteins, including the Ro protein, occur early in the preclinical stage of the disease, along with a positive antinuclear antibody (ANA) test. These are often followed by anti-double-stranded DNA antibodies and, finally, by the development of antibodies specific for the spliceosomal proteins Smith (Sm) and ribonucleoprotein (RNP) at approximately the time of diagnosis (Fig. 250-1). These observations suggest that individuals who demonstrate progression from antibodies targeting RNA to those targeting DNA and spliceosomal proteins are those in whom sufficient autoimmunity develops to manifest clinical symptoms. Approximately one third of SLE patients have autoantibodies reactive with phospholipids or the proteins associated with them, particularly β2-glycoprotein I. These autoantibody specificities can also be present independently of SLE in primary antiphospholipid antibody syndrome (Chapter 162).

Immune Complexes and Complement

Tissue and organ damage in SLE is mediated by the deposition or in situ formation of immune complexes and subsequent complement activation and inflammation. The complement system (Chapter 44), composed of more than 30 proteins that act in concert to protect the host against invading organisms, initiates inflammation and tissue injury. Complement activation promotes

Diagnosis

chemotaxis of inflammatory cells and generates proteolytic fragments that enhance phagocytosis by neutrophils and monocytes. The classical complement pathway is activated when antibodies bind to antigen and generate potent effectors. Alternative pathway activation mechanisms differ in that they are initiated by the binding of spontaneously activated complement components to the surfaces of pathogens or self-tissues. C3a, an anaphylatoxin that binds to receptors on leukocytes and other cells, causes activation and release of inflammatory mediators. C5a is a potent soluble inflammatory, anaphylatoxic, and chemotactic molecule that promotes recruitment and activation of neutrophils and monocytes and mediates endothelial cell activation through its receptor. The release of reactive oxygen and nitrogen intermediates is an additional mechanism that contributes to tissue damage. Tissues targeted by immune system activity in lupus include the skin, where immune complexes and complement are deposited in a linear pattern (as demonstrated in the lupus band test, in which deposited antibodies are identified by a fluorescent tag), the glomeruli, and heart valves. Antibodies reactive with hippocampal neurons in the brain can mediate excitotoxic death. Immune and inflammatory mechanisms responsible for the vasculopathy of lupus are multifactorial and not clearly defined. Microvascular damage is observed in splenic arteries and is characterized by the typical onion-skin pattern of concentric connective tissue deposition. In addition to vascular damage mediated by inflammation, thrombosis, including microthrombi, contributes to ischemia and cell necrosis in the brain and other organs.  

CLINICAL MANIFESTATIONS

Symptoms and Signs Constitutional Symptoms

SLE is a disease that involves virtually all components of the immune system and can be accompanied by constitutional symptoms similar to those seen in the setting of microbial infection. Fatigue, headaches, weight loss, and fevers are common, along with generalized arthralgias, myalgias, and lymphadenopathy. The level of activity of lupus typically follows a pattern of flares and remissions, although some patients sustain active disease for prolonged periods. Careful monitoring for the development of major organ system disease is important to ensure timely adjustments in medical therapy.

Cutaneous and Mucous Membranes

The skin and mucous membranes are affected in most lupus patients (Table 250-1). The erythematous facial rash with a butterfly distribution across the malar and nasal prominences and sparing of the nasolabial folds is the classic rash of SLE and is seen in 30 to 60% of patients (Fig. 250-2). The butterfly rash is often triggered by sun exposure, but photosensitivity can also be demonstrated diffusely in other areas of the body. The discoid skin lesions are erythematous plaques with central scarring and may be covered with scale. These lesions are seen in about 25% of patients,

Patients with positive test

100 80

TABLE 250-1 CLINICAL MANIFESTATIONS OF SYSTEMIC LUPUS ERYTHEMATOSUS

60

MANIFESTATION ANA Anti-Ro Anti-La APL Anti-ds DNA Anti-Sm Anti-nRNP

40 20 0 −5

−4

−3

−2

−1 0 1 Time (yr)

2

3

4

5

FIGURE 250-1.  Proportion of patients with positive antibody tests relative to the

time of diagnosis or appearance of the first clinical manifestation of systemic lupus erythematosus (SLE). For each autoantibody, the proportion of patients testing positive relative to the time of diagnosis or to the time of appearance of the first clinical criterion was assessed. In analyses of the time from antibody development to the diagnosis of SLE, antinuclear antibodies (ANAs) appeared significantly earlier than anti-Sm antibodies and antinuclear ribonucleoprotein (anti-nRNP) antibodies, but not significantly earlier than anti-Ro, anti-La, antiphospholipid (APL), or anti-double-stranded DNA antibodies (anti-ds DNA). (From Arbuckle MR, McClain MT, Rubertone MV, et al. Development of autoantibodies before the clinical onset of systemic lupus erythematosus. N Engl J Med. 2003;349:16.)

APPROXIMATE FREQUENCY (%)

Cutaneous

88

Arthritis/arthralgias

76

Neuropsychiatric

66

Pleurisy/pericarditis

63

Anemia

57

Raynaud phenomenon

44

Vasculitis

43

Atherosclerosis

37

Nephritis

31

Thrombocytopenia

30

Sensorimotor neuropathy

28

Cardiac valvar disease

18

Pulmonary alveolar hemorrhage

12

Pancreatitis

10

Myositis

5

Myocarditis

5

1728

CHAPTER 250  Systemic Lupus Erythematosus  

FIGURE 250-2.  Malar rash in a patient with systemic lupus erythematosus. Note that the rash does not cross the nasolabial fold. (From Gladman DD, Urowitz MB. Systemic lupus erythematosus: clinical features. In: Klippel JH, Dieppe PA, eds. Rheumatology. 2nd ed. London: Mosby; 1998.)

involve the scalp or the face and ears, and may be associated with alopecia. Discoid lesions can be present in the absence of systemic manifestations of SLE (discoid lupus). In addition to the scarring alopecia of discoid lupus, more transient alopecia may be a clinical sign of increased disease activity and is associated with apoptosis of cells in the hair follicle. Inflammation of the deep dermis and subcutaneous fat can result in lupus panniculitis, with firm painful nodules that sometimes adhere to the epidermis, causing irregularities in the superficial skin. Subacute cutaneous lupus erythematosus is seen in sun-exposed areas and can involve erythematous plaques or psoriasiform lesions. It is associated with autoantibodies to the Ro (SSA) RNA-binding protein. Mucosal ulcerations, especially of the buccal mucosa and upper palate, result from mucositis and are typical of SLE. Manifestations of vasculopathy are also common in SLE, including arteriolar spasm or infarcts in the nail folds, a diffuse lacey pattern over the skin described as livedo reticularis, and petechial-purpuric or urticarial lesions on the extremities. Vasculopathy in SLE is often associated with the presence of antiphospholipid antibodies.

Musculoskeletal System

Arthralgias and nonerosive arthritis are among the most common clinical features of SLE and are experienced by more than 85% of patients. The proximal interphalangeal and metacarpophalangeal joints of the hand are most commonly symptomatic, along with the knees and wrists. In some patients (≈10%), deformities resulting from damage to periarticular tissue can occur, a condition termed Jaccoud arthropathy. The heavy use of corticosteroids in many lupus patients can be accompanied by the development of osteoporosis, including osteoporotic fractures or osteonecrosis, most commonly of the hips, although the underlying vasculopathy can also contribute to joint damage. Inflammation of the muscles with elevated creatine phosphokinase can occur rarely in SLE, and myopathy may be observed as a consequence of corticosteroid therapy. Fibromyalgia, characterized by painful trigger points at characteristic locations, commonly accompanies SLE and can contribute to fatigue and depression.

Renal System

Kidney involvement in SLE (Chapter 113) is common, with 74% of patients being affected at some time in the course of disease, and is a poor prognostic indicator. Renal pathology is generally attributed to the deposition of circulating immune complexes or in situ formation of these complexes in glomeruli and results in the activation of complement and subsequent recruitment of inflammatory cells. In addition to glomerular inflammation, necrosis, and scarring, renal pathology is characterized by vascular lesions, including thrombotic microangiopathy and extraglomerular vasculitis. Tubulointerstitial disease, including infiltration of the interstitium with mononuclear cells, tubular atrophy, and interstitial fibrosis, is increasingly recognized as associated with a poor prognosis for persistent nephritis and renal survival.17 Hypertension may be a consequence of significant renal involvement.

Most cases of lupus nephritis present a complex immunopathologic picture, but in general, the pattern of renal disease reflects the site of deposition of immunoglobulins and the quality of the effector mechanisms they induce. Mesangial deposition of immunoglobulin induces mesangial cell proliferation and is associated with microscopic hematuria and mild proteinuria (Fig. 250-3). Subendothelial deposition of immune complexes results in proliferative and exudative inflammation, together with hematuria, mild to moderate proteinuria, and reduced glomerular filtration rate. Subepithelial deposition of immune complexes adjacent to podocytes and along the glomerular basement membrane can result in membranous nephritis with nephrotic-range proteinuria. In addition, antiphospholipid antibodies may support the development of thrombotic or inflammatory vascular lesions within or external to glomeruli. A World Health Organization classification of lupus nephritis lesions was first published in 1975, with subsequent revisions. These classifications were reviewed and rigorously reexamined in the revised International Society of Nephrology and Renal Pathology Society classification criteria for lupus glomerulonephritis (GN) published in 2004, with additional review in 2018.18 (Table 113-7 in Chapter 113, and also E-Table 250-2). Class I and II GN involves mesangial deposition of immune complexes (class I without and class II with mesangial hypercellularity); class III describes focal GN involving less than 50% of total glomeruli; class IV includes diffuse GN involving 50% or more of glomeruli; class V designates membranous lupus nephritis; and class VI is characterized by advanced sclerotic lesions. Classes III and IV have subdivisions for active and sclerotic lesions, and class IV currently also has subdivisions for segmental and global involvement. Recent recommendations from these societies include elimination of the IV-S and IV-G subdivisions and replacement of the active and chronic designations for class III/IV lesions with application of activity and chronicity indices for all classes. They also suggest eliminating the term “endocapillary proliferation” and are considering a more appropriate definition of endocapillary hypercellularity. Validation of candidate revised classification criteria is planned by these groups. Pathologic diagnosis should include descriptions of tubulointerstitial and vascular disease as well as glomerular involvement. Several renal pathologic lesions seen in SLE patients that are not encompassed in the classification scheme for lupus GN include lupus podocytopathy, collapsing glomerulopathy, and thrombotic microangiopathy, the latter often associated with antiphospholipid syndrome. The prognosis of class I and class II disease is usually good, whereas class IV, the most common form of lupus nephritis, has the worst prognosis, particularly when the serum creatinine level is elevated at the time of diagnosis. Class V nephritis occurs in 10 to 20% of patients, and the implication for long-term outcome depends on the degree of proteinuria, with mild proteinuria having a good prognosis and nephrotic syndrome with chronic edema having a more negative prognosis. It should be noted that renal veins can occasionally become involved with thrombosis, which then also contributes to nephrotic syndrome. This complication can be evaluated by renal ultrasound (Chapter 113).

Cardiovascular System

Pericarditis and valve nodules were among the first clinical manifestations described in SLE. It is now recognized that premature atherosclerotic disease represents a significant contributor to morbidity and mortality in SLE patients. Pericarditis (Chapter 68) is the most common cardiac manifestation, but it is sometimes recognized only on imaging studies or at autopsy. It is a component of the generalized serositis that is often a feature of SLE and is associated with local autoantibodies and immune complexes. Pericarditis is usually manifested as substernal chest pain that is improved by bending forward and can be exacerbated by inspiration or coughing. The symptoms and effusions associated with pericarditis are quite responsive to moderate-dose (20 to 30 mg/day of prednisone) corticosteroid treatment. Structural valve abnormalities in SLE range from sterile nodules (originally described by Libman and Sacks) to nonspecific valve thickening (Chapter 66). The nodules are immobile and usually located on the atrial side of the mitral valve and sometimes on the arterial side of the aortic valve. Right-sided lesions are rare. These structural changes may in some cases result in valvular regurgitation. Although valve nodules are detected in most patients with SLE at autopsy, clinically significant valvular heart disease is much less common (1 to 18%). The verrucous valvular lesions of Libman and Sacks are most likely inflammatory in nature and may be associated with the presence of antiphospholipid antibodies. Premature and accelerated atherosclerosis is prevalent in lupus patients, and preclinical atherosclerotic carotid plaque has been documented in 37% of SLE patients as opposed to 15% of age- and sex-matched controls. Traditional

CHAPTER 250  Systemic Lupus Erythematosus  

1728.e1

E-TABLE 250-2 INTERNATIONAL SOCIETY OF NEPHROLOGY/RENAL PATHOLOGY SOCIETY 2004 CLASSIFICATION OF LUPUS NEPHRITIS Class I

Minimal mesangial lupus nephritis Normal glomeruli by light microscopy, but mesangial immune deposits by immunofluorescence

Class II

Mesangial proliferative lupus nephritis Purely mesangial hypercellularity of any degree or mesangial matrix expansion by light microscopy, with mesangial immune deposits. May be a few isolated subepithelial or subendothelial deposits visible by immunofluorescence or electron microscopy, but not by light microscopy

Class III

Focal lupus nephritis* Active or inactive focal, segmental, or global endocapillary or extracapillary glomerulonephritis involving 0.5 g/day or >3+ if quantitation is not performed or B. Cellular casts—may be red cell, hemoglobin, granular, tubular, or mixed

8. Neurologic disorder

A. Seizures—in the absence of offending drugs or known metabolic derangements (e.g., uremia, ketoacidosis, electrolyte imbalance) or B. Psychosis—in the absence of offending drugs or known metabolic derangements (e.g., uremia, ketoacidosis, electrolyte imbalance)

9. Hematologic disorder

A. B. C. D.

Hemolytic anemia—with reticulocytosis or Leukopenia—100 µm) can occur in systemic sclerosis. Manifestations include occlusion of the digital and ulnar arteries, leading to ischemic ulcerations and even loss of digits or limbs. Epidemiologic studies indicate increased risk of coronary artery disease in patients with systemic sclerosis. Dry eyes and dry mouth are common in systemic sclerosis, but in contrast to Sjögren syndrome (Chapter 252), salivary gland biopsy in such cases shows fibrosis rather than focal lymphocytic infiltration. Hypothyroidism due to thyroid fibrosis is common and may be associated with antithyroid autoantibodies. Although the central nervous system is generally spared in systemic sclerosis, autonomic neuropathy, as well as a primarily sensory neuropathy of the trigeminal nerve due to fibrosis or vasculopathy, can occur. Pregnancy in women with active systemic sclerosis has been associated with an increased rate of adverse fetal outcomes. Furthermore, cardiopulmonary involvement might worsen during pregnancy, and scleroderma renal crisis can occur. Inability to attain or maintain penile erection due to vascular insufficiency and fibrosis is frequent and may be the presenting disease manifestation in males with systemic sclerosis.

With the exception of ACE inhibitors used for scleroderma renal crisis, no therapy to date has been shown to significantly alter the natural history of systemic sclerosis, and none have been approved specifically for this indication. In contrast, organ-based treatments are commonly effective in alleviating symptoms and slowing progression of the cumulative organ damage. Treatment must be tailored to each patient’s unique needs.19 Because of the marked variability in clinical presentation, a thorough and individualized baseline evaluation is paramount. Optimal management should be guided by the following principles: prompt diagnosis, accurate classification and risk stratification, early recognition and assessment of organ-based complications, and long-term monitoring of progression, disease activity, and response to therapy. Management of disease complications should be proactive, with regular screening and initiation of appropriate intervention at the earliest possible opportunity. Given the multisystemic nature of systemic sclerosis, an integrated team-based management approach, typically at specialized medical centers, is most desirable. The team should incorporate appropriate medical specialists and facilitate coordinated holistic care of the patient.

Systemic Sclerosis and Cancer

Disease-Modifying Therapy

Patients with systemic sclerosis have an increased risk of cancer.18 In these patients, lung cancer and esophageal adenocarcinoma typically occur in the setting of long-standing interstitial lung disease or gastroesophageal reflux disease, and chronic inflammation and tissue damage may be contributing factors. In contrast, breast, lung, and ovarian carcinoma and lymphoma in systemic sclerosis tend to occur in close temporal association with the onset

TREATMENT AND PREVENTION 

Immunosuppressive Agents

Immunosuppressive agents, often highly effective in other connective tissue diseases, have generally shown modest or no benefit in systemic sclerosis. Corticosteroids alleviate stiffness, fatigue, and aching in early-stage disease, but do not slow disease progression and are associated with an increased risk for scleroderma renal crisis. Therefore corticosteroids should be avoided if

CHAPTER 251  Systemic Sclerosis (Scleroderma)  

possible; when absolutely necessary, they should be given at the lowest dose possible and for brief periods only. Cyclophosphamide was shown to reduce the progression of symptomatic interstitial lung disease in early systemic sclerosis. A1  Compared with placebo, patients treated with oral cyclophosphamide showed stabilization and, rarely, modest improvement in respiratory symptoms, pulmonary function, and abnormalities on chest HRCT after 1 year of treatment, but these benefits were shortlived. The use of cyclophosphamide in systemic sclerosis needs to be balanced against its potential for side effects, including bone marrow suppression, opportunistic infections, hemorrhagic cystitis, bladder cancer, and premature ovarian failure. Rituximab is an equally effective and safe alternative. A2  In small clinical trials, methotrexate was associated with a modest improvement in skin involvement. Mycophenolate mofetil was shown to improve skin involvement and stabilize lung disease as well as cyclophosphamide, and it was well tolerated in a randomized controlled clinical trial. A3  There is some support in the literature for the use of immunomodulatory agents including, tocilizumab, intravenous immunoglobulin, and extracorporeal photopheresis for the treatment of systemic sclerosis. Recent reports suggest that rituximab might be effective in ameliorating skin and lung involvement. In patients with severe systemic sclerosis who fail to respond to other treatments (Chapter 168), autologous hematopoietic stem cell transplantation improves long-term, event-free survival despite an increased treatment-related mortality in the first year. A4  More recently, adults with severe systemic sclerosis were randomized to undergo myeloablative autologous hematopoietic stem cell transplantation (36 patients) or to receive cyclophosphamide immunosuppression by means of 12 monthly infusions (39 patients). The rate of event-free survival at 54 months was 79% in the transplantation group and 50% in the cyclophosphamide group (P = 0.02); at 72 months it was 74% versus 47% (P = 0.03). Overall survival at 72 months also favored transplantation (86% vs. 51%, P = 0.02). Treatment-related mortality in the transplantation group was 3% at 54 months and 6% at 72 months, compared with 0% in the cyclophosphamide group. A5  Because of potential morbidity and mortality and its substantial cost, hematopoietic stem cell transplantation is presently considered only for carefully selected systemic sclerosis patients with aggressive or treatment-unresponsive disease.

Antifibrotic Therapy

Because tissue fibrosis causes progressive and irreversible organ damage, drugs that block or slow the fibrotic process represent a rational approach to therapy. To date, however, no antifibrotic drug has been shown to be effective in modifying disease course in SSc. D-Penicillamine has been extensively used as an antifibrotic agent. However, in a randomized controlled clinical trial, there was no difference in the extent of skin involvement between patients treated with standard-dose (750 mg/day) or very low-dose (125 mg every other day) D-penicillamine. Minocycline, bosentan, relaxin, interferon-γ, and inhibitors of tumor necrosis factor are putative antifibrotic agents that have failed to show meaningful benefit in systemic sclerosis clinical trials. Small-molecule tyrosine kinase inhibitors used in malignancies (e.g., imatinib, nilotinib, and dasatinib) block signaling by TGF-β and PDGF and thereby prevent fibrotic responses in vitro and in vivo.17 Antifibrotic drugs that had been recently approved for the treatment of idiopathic pulmonary fibrosis (nintedanib and pirfenidone) may have a role in the treatment of systemic sclerosis-associated interstitial lung disease. These two drugs are currently in clinical trials for this indication.

Treatment of Organ-Specific Complications Gastrointestinal Complications

Because significant gastroesophageal reflux may be asymptomatic, all patients with systemic sclerosis should be treated for this complication. Proton pump inhibitors may need to be given in relatively high doses and for prolonged periods, and patients should be instructed to elevate the head of the bed and eat frequent small meals. Recurrent gastrointestinal bleeding due to gastric vascular ectasia can be treated with laser or argon plasma photocoagulation. Bacterial overgrowth due to small bowel hypomotility causes bloating and diarrhea and may lead to malabsorption, weight loss, and malnutrition. Treatment with short courses of rotating broad-spectrum antibiotics such as metronidazole, erythromycin, and tetracycline can sometimes eradicate bacterial overgrowth. However, many patients relapse when antibiotics are stopped. In patients with malnutrition but intact small bowel function, enteral nutrition via a jejunostomy can be effective. In others, total parenteral nutrition may be indicated. Refractory hypomotility of the small bowel may respond to subcutaneous octreotide injections. Anorectal complications may respond to sacral neuromodulation.

Vascular Therapy and Raynaud Phenomenon

The goal of vascular therapy in systemic sclerosis is to reduce the frequency and duration of vasospastic episodes, prevent ischemic complications, and enhance their healing, and slow the progression of obliterative vasculopathy. Patients should dress warmly, minimize cold exposure, and avoid drugs that could precipitate or exacerbate vasospastic episodes. Calcium-channel blockers such as nifedipine and diltiazem are used commonly for Raynaud phenomenon but show only moderate benefit, and their use is often limited by side effects (palpitations, dependent edema, lightheadedness). Angiotensin II receptor

1741

blockers such as losartan are effective and generally well tolerated. Patients with severe Raynaud phenomenon require α1-adrenergic receptor blockers (e.g., prazosin), 5-phosphodiesterase inhibitors (e.g., sildenafil), topical nitroglycerine, intradigital botulin toxin injections, or intravenous prostaglandins. Low-dose aspirin and dipyridamole prevent platelet activation and may have a role as adjunctive agents but must be used with caution in light of the risk of bleeding from gastric vascular ectasia lesions. The endothelin-1 receptor antagonist bosentan reduces development of new ischemic ulcers, A6  and sildenafil may promote ulcer healing. A7  Patients with ischemic digital ulcerations may require surgical débridement, especially if necrotic tissue is present. Empirical long-term therapy with statins and antioxidants may slow the progression of vascular damage.

Pulmonary Arterial Hypertension

All patients with systemic sclerosis should be screened for pulmonary hypertension at initial evaluation, and those at high risk on a yearly basis. Treatment for symptomatic pulmonary hypertension should be started with an endothelin-1 receptor antagonist or a 5-phosphodiesterase inhibitor. Diuretics, oral anticoagulation, and digoxin, and supplemental oxygen, may be used when appropriate. If clinical response is inadequate, 5-phosphodiesterase inhibitors may be used in combination with endothelin-1 receptor antagonists. Prostacyclin analogs can be administered intravenously, by continuous subcutaneous infusion, or by frequent inhalations. Lung transplantation remains an option for selected patients with systemic sclerosis–associated pulmonary hypertension or interstitial lung disease who fail medical therapy.

Treatment and Prevention of Scleroderma Renal Crisis

Prompt recognition of impending or early scleroderma renal crisis is essential. Because patients with early-stage systemic sclerosis and progressive skin involvement are at highest risk, they should be instructed to monitor their blood pressure daily and report significant alterations immediately. Corticosteroids should be used only when absolutely necessary and at the lowest possible doses. When scleroderma renal crisis occurs, patients should be hospitalized and treatment with short-acting ACE inhibitors started immediately to achieve prompt blood pressure normalization. There is no evidence that “prophylactic” use of ACE inhibitors can prevent the development of scleroderma renal crisis or ameliorate its severity. Although up to two thirds of patients who develop renal crisis require dialysis, delayed recovery of renal function can occur. Kidney transplantation is appropriate for patients unable to discontinue dialysis after 2 years. Survival with renal transplantation in systemic sclerosis is comparable to that in other connective tissue diseases, and recurrence of scleroderma renal crisis in the kidney graft is rare.

Skin Care

Skin involvement in early systemic sclerosis is inflammatory and may respond to systemic antihistamines or short-term low-dose corticosteroids. Because of the increased risk for scleroderma renal crisis, blood pressure should be carefully monitored. Cyclophosphamide, methotrexate, D-penicillamine, and mycophenolate have been associated with modest improvement in skin induration in early-stage systemic sclerosis.20 Skin dryness can be managed with the use of hydrophilic ointments and emollient bath oils. Fingertip ulcerations should be protected by occlusive dressing to promote healing and prevent infection. Infected skin ulcers are treated with topical or oral antibiotics and may necessitate surgical débridement. No medical therapy has been shown to be effective in preventing soft tissue calcification or in promoting its dissolution, and surgical therapy and lithotripsy are only occasionally effective.



PROGNOSIS AND NATURAL HISTORY

Patients with diffuse cutaneous systemic sclerosis have a more rapidly progressive disease course, greater internal organ involvement, and generally worse prognosis compared with those with limited cutaneous systemic sclerosis. However, the outcome of the disease is difficult to predict. Early inflammatory symptoms of diffuse cutaneous systemic sclerosis such as fatigue, edema, arthralgia, and pruritus commonly subside after 2 to 4 years. Skin thickening typically reaches a plateau, followed by slow regression, which characteristically occurs in an order that is the reverse of initial involvement, with softening on the trunk followed by proximal and finally the distal extremities. Sclerodactyly and finger contractures generally fail to resolve. Relapse or recurrence of skin thickening may occur. Visceral organ involvement develops and progresses most rapidly during the initial 2 to 4 years of the disease. New organ involvement rarely occurs once the skin involvement has reached a plateau. Similarly, scleroderma renal crisis almost invariably occurs within the first 4 years of disease. In patients with limited cutaneous systemic sclerosis, Raynaud phenomenon may precede other disease manifestations by years or even decades, and visceral organ complications such as pulmonary hypertension and primary biliary cirrhosis generally occur late in the course of the disease.

Age- and gender-adjusted standardized mortality ratios (SMRs) in patients with systemic sclerosis have ranged from 1.05 to 5.40 across studies, but an overall SMR of 2.72 higher than in the general population has been determined.21 The 10-year survival rate is 55% for patients with diffuse cutaneous systemic sclerosis and 75% for patients with limited cutaneous systemic sclerosis. Survival correlates with the extent of skin involvement, which represents a surrogate for visceral organ involvement. The leading causes of death are pulmonary fibrosis, pulmonary hypertension, renal and severe gastrointestinal involvement, and cardiac disease. Markers of poor prognosis include male sex, African American race, older age of disease onset, low body mass index, extensive skin thickening with truncal involvement, and evidence of significant or progressive visceral organ involvement. Autoantibodies to topoisomerase-I or absence of anticentromere antibodies are markers of poor prognosis. In one study, systemic sclerosis patients who had extensive skin involvement, vital capacity less than 55% of predicted, significant gastrointestinal involvement, and clinically evident cardiac involvement or scleroderma renal crisis had a less than 40% 10-year survival. The severity of pulmonary hypertension is correlated with mortality, and systemic sclerosis patients with a mean pulmonary arterial pressure of 45 mm Hg or higher had a 33% 3-year survival rate. In scleroderma renal crisis, therapy with ACE inhibitors has had a dramatic effect on survival, increasing from less than 10% at 1 year in the pre-ACE inhibitor era to better than 70% 3-year survival at the present time. The topic of immunoglobulin (Ig)G4–related disease is discussed in Chapter 259.

  Grade A References A1. Barnes H, Holland AE, Westall GP, et al. Cyclophosphamide for connective tissue disease-associated interstitial lung disease. Cochrane Database Syst Rev. 2018;1:CD010908. A2. Sircar G, Goswami RP, Sircar D, et al. Intravenous cyclophosphamide vs rituximab for the treatment of early diffuse scleroderma lung disease: open label, randomized, controlled trial. Rheumatology (Oxford). 2018;57:2106-2113. A3. Tashkin DP, Roth MD, Clements PJ, et al. Mycophenolate mofetil versus oral cyclophosphamide in scleroderma-related interstitial lung disease (SLS II): a randomised controlled, double-blind, parallel group trial. Lancet Respir Med. 2016;4:708-719. A4. van Laar JM, Farge D, Sont JK, et al. Autologous hematopoietic stem cell transplantation vs intravenous pulse cyclophosphamide in diffuse cutaneous systemic sclerosis: a randomized clinical trial. JAMA. 2014;311:2490-2498. A5. Sullivan KM, Goldmuntz EA, Keyes-Elstein L, et al. SCOT study investigators. Myeloablative autologous stem-cell transplantation in severe scleroderma. N Engl J Med. 2018;378:35-47. A6. Matucci-Cerinic M, Denton CP, Furst DE, et al. Bosentan treatment of digital ulcers related to systemic sclerosis: results from the RAPIDS-2 randomised, double-blind, placebo-controlled trial. Ann Rheum Dis. 2011;70:32-38. A7. Hachulla E, Hatron PY, Carpentier P, et al. Efficacy of sildenafil on ischaemic digital ulcer healing in systemic sclerosis: the placebo-controlled SEDUCE study. Ann Rheum Dis. 2016;75:1009-1015.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 251  Systemic Sclerosis (Scleroderma)  

GENERAL REFERENCES 1. Denton CP, Khanna D. Systemic sclerosis. Lancet. 2017;390:1685-1699. 2. Tratenberg M, Gutwein F, Rao V, et al. Localized scleroderma: a clinical review. Curr Rheumatol Rev. 2017;13:86-92. 3. Mostmans Y, Cutolo M, Giddelo C, et al. The role of endothelial cells in the vasculopathy of systemic sclerosis: a systematic review. Autoimmun Rev. 2017;16:774-786. 4. Launay D, Sobanski V, Hachulla E, et al. Pulmonary hypertension in systemic sclerosis: different phenotypes. Eur Respir Rev. 2017;26:1-6. 5. Sakkas LI, Simopoulou T, Daoussis D, et al. Intestinal involvement in systemic sclerosis: a clinical review. Dig Dis Sci. 2018;63:834-844. 6. Denazas K, Ladas SD, Karamanolis GP. Evaluation and management of esophageal manifestations in systemic sclerosis. Ann Gastroenterol. 2018;31:165-170. 7. Rangarajan V, Matiasz R, Freed BH. Cardiac complications of systemic sclerosis and management: recent progress. Curr Opin Rheumatol. 2017;29:574-584. 8. Zanatta E, Polito P, Favaro M, et al. Therapy of scleroderma renal crisis: state of the art. Autoimmun Rev. 2018;17:882-889. 9. Cutolo M, Smith V, Furst DE, et al. Points to consider—Raynaud’s phenomenon in systemic sclerosis. Rheumatology (Oxford). 2017;56(suppl 5):v45-v48. 10. Wigley FM, Flavahan NA. Raynaud’s phenomenon. N Engl J Med. 2016;375:556-565. 11. Brown M, O’Reilly S. The immunopathogenesis of fibrosis in systemic sclerosis. Clin Exp Immunol. 2019;195:310-321.

1742.e1

12. Knobler R, Moinzadeh P, Hunzelmann N, et al. European Dermatology Forum S1-guideline on the diagnosis and treatment of sclerosing diseases of the skin, Part 1: localized scleroderma, systemic sclerosis and overlap syndromes. J Eur Acad Dermatol Venereol. 2017;31:1401-1424. 13. Ferreli C, Gasparini G, Parodi A, et al. Cutaneous manifestations of scleroderma and scleroderma-like disorders: a comprehensive review. Clin Rev Allergy Immunol. 2017;53:306-336. 14. Belch J, Carlizza A, Carpentier PH, et al. ESVM guidelines—the diagnosis and management of Raynaud’s phenomenon. VASA. 2017;46:413-423. 15. Das A, Kumar A, Arrossi AV, et al. Scleroderma-related interstitial lung disease: principles of management. Expert Rev Respir Med. 2019;13:357-367. 16. Woodworth TG, Suliman YA, Li W, et al. Scleroderma renal crisis and renal involvement in systemic sclerosis. Nat Rev Nephrol. 2016;12:678-691. 17. Smolenska Z, Barraclough R, Dorniak K, et al. Cardiac involvement in systemic sclerosis: diagnostic tools and evaluation methods. Cardiol Rev. 2019;27:73-79. 18. Dolcino M, Pelosi A, Fiore FF, et al. Gene profiling in patients with systemic sclerosis reveals the presence of oncogenic gene signatures. Front Immunol. 2018;9:1-20. 19. Kowal-Bielecka O, Fransen J, Avouac J, et al. Update of EULAR recommendations for the treatment of systemic sclerosis. Ann Rheum Dis. 2017;76:1327-1339. 20. Herrick AL, Pan X, Peytrignet S, et al. Treatment outcome in early diffuse cutaneous systemic sclerosis: the European Scleroderma Observational Study (ESOS). Ann Rheum Dis. 2017;76: 1207-1218. 21. Poudel DR, Jayakumar D, Danve A, et al. Determinants of mortality in systemic sclerosis: a focused review. Rheumatol Int. 2018;38:1847-1858.

1742.e2

CHAPTER 251  Systemic Sclerosis (Scleroderma)  

REVIEW QUESTIONS 1. Which of the following does not contribute to the pathogenesis of systemic sclerosis (SSc)? A . Vascular wall remodeling B. Tissue fibrosis C. Vasculitis D. T cells E. Hypoxia Answer: C  Vasculitis does not typically occur in SSc. Vascular remodeling leading to hypoxia, inflammation, and fibrosis are the hallmarks of the disease. 2. Which is a characteristic autoantibody seen in SSc patients? A . Anticentromere B. Anti-CCP C. Antihistone D. Antiphospholipid E. Anti-Smith Answer: A  Anticentromere is a hallmark SSc-associated autoantibody, whereas the other autoantibodies are seen in rheumatoid arthritis (B), lupus (C, E), and the antiphospholipid syndrome (D). 3. Which is a major factor in SSc morbidity? A . Glomerulonephritis B. Episcleritis C. Amyloidosis D. Pulmonary hypertension E. Bowel infarction Answer: D  Pulmonary hypertension develops in up to 15% of SSc patients and is a major cause of morbidity and mortality. The other clinical features are uncommon in SSc and indicate other autoimmune or rheumatic diseases.

4. In patients with SSc-associated lung disease, lung biopsy may show: A . Honeycombing B. Interstitial fibrosis C. Plasma cell accumulation D. Intimal proliferation in the small vessels E. All of the above Answer: E  All of the listed pathologic features can be noted in lung biopsies from patients with SSc-associated interstitial lung disease. 5. Which of the following drugs may contribute to scleroderma renal crisis? A . NSAIDs B. Glucocorticoids C. Mycophenolate D. Rituximab E. Penicillamine Answer: B  The use of glucocorticoids has been shown to increase the risk of new-onset scleroderma renal crisis in patients with SSc.

1742

CHAPTER 252  Sjögren Syndrome  

autoimmune disease, the presentation of secondary Sjögren syndrome closely resembles that of primary Sjögren syndrome.  

EPIDEMIOLOGY

Primary Sjögren syndrome is a common disease that affects 0.1 to 0.6% of the general adult female population.2,3 A higher prevalence of the disease has been reported (0.5 to 2%), but this must be considered with caution because the reported prevalence of Sjögren syndrome depends on the classification criteria used in the various studies, and the prevalence of sicca symptoms in the general population is high. Conversely, in recent studies with strict criteria, a lower prevalence has been found: 1.02 per 10,000 adults.4 Primary Sjögren syndrome has a female preponderance (female-to-male ratio at least 9 : 1). The age peak of the disease occurs after menopause in the mid-50s.  

252  SJÖGREN SYNDROME XAVIER MARIETTE AND GAETANE NOCTURNE



DEFINITION

Sjögren syndrome is a systemic autoimmune disease characterized by lymphocytic infiltrates of salivary and tear glands, leading to oral and ocular dryness, and by autoantibody secretion. It can be encountered either alone (primary Sjögren syndrome) or in the presence of other systemic autoimmune diseases (secondary Sjögren syndrome) like rheumatoid arthritis, systemic lupus erythematosus, inflammatory myositis, and systemic sclerosis.1 Sjögren syndrome in the setting of rheumatoid arthritis usually follows the diagnosis of rheumatoid arthritis by many years and is mainly manifested by keratoconjunctivitis sicca, with systemic features being rather uncommon. Associated with other systemic

PATHOPHYSIOLOGY

Recent years have witnessed major advances in the pathophysiologic mechanisms of the disease. Several studies have confirmed the role of innate immunity, genetics, and B-cell activation and the relation between abnormalities in them. The presence of an interferon (IFN) signature (the expression of type 1 IFN–inducible genes) has been shown both in salivary glands and blood. Plasmacytoid dendritic cells, the professional cells secreting type 1 IFN, are present within the glands. Type 2 IFN–dependent genes can be overexpressed in salivary glands. Natural killer (NK) cells, another actor of innate immunity able to secrete type 2 IFN, are present in salivary glands of patients and play a role in the disease. In line with this IFN signature, multiple viral agents have been incriminated as etiologic factors for either the development or the modulation of Sjögren syndrome; these include Epstein-Barr virus, retroviruses, and coxsackieviruses, but in all cases the data remain controversial. The genetics of primary Sjögren syndrome5 are now better understood with the reports of two genome-wide association studies (GWASs). Like in other systemic autoimmune diseases, human leukocyte antigen (HLA) is the most important region associated with the disease, especially HLA-DR3-DQ1 in patients with autoantibodies. Other genes associated with the disease are involved in the IFN response. These include IFN regulatory factor 5 (IRF5), a pivotal transcription factor in the type 1 IFN pathway; signal transducer and activator of transcription 4 (STAT4); and IL12A, involved in the type 2 IFN pathway. Other genes found to be associated with the disease are TNIP1, playing a role in control of nuclear factor (NF)-κB activation, and CXCR5, involved in germinal center formation. The presence of ectopic salivary gland germinal centers demonstrates the importance of B-cell activation in primary Sjögren syndrome. Different cytokines may explain this B-cell activation. Several studies have focused on the role of BAFF (B-cell activating factor of the tumor necrosis factor [TNF] family), a cytokine that promotes B-cell maturation, proliferation, and survival. It has been shown that BAFF is enhanced in sera and in salivary glands from primary Sjögren syndrome patients. Interestingly, BAFF can be secreted by salivary gland epithelial cells, the target of autoimmunity, after stimulation by the innate immune system (type 1 or type 2 IFN, or viral infections). Thus, this cytokine is likely to be a link between innate immunity and autoimmunity. The current hypothetical scenario for the development of primary Sjögren syndrome is based on the successive activation of innate and adaptive immune systems (Fig. 252-1). Environmental factors such as viral infections or hormonal imbalance may act at the initial stage of the disease by activating epithelial cells. This epithelial cell activation is promoted in patients who carry susceptibility factors in the genes for IFN pathway proteins. These patients experience a greater degree of IFN pathway activation, which leads to BAFF overproduction, B- and T-cell activation,6 and secretion of autoantibodies, especially in predisposed patients. These autoantibodies constitute immune complexes that participate in the maintenance of IFN-α production. Altogether, these steps promote a vicious cycle of immune system activation leading to tissue damage.  

CLINICAL MANIFESTATIONS

Glandular

Decreased salivary secretion results in mouth dryness and increased incidence of oral infections, mucosal friability, and dental caries due to loss of the lubricating, buffering, and antimicrobial capacities of saliva.7,8 Fungal infections (primarily candidiasis) are also common. Parotid salivary gland or other major salivary gland enlargement can also occur. Persistent enlargement should be carefully followed, however, to exclude bacterial superinfection and, more important, the development of lymphoma.

CHAPTER 252  Sjögren Syndrome  

ABSTRACT

Sjögren syndrome is a systemic autoimmune disease characterized by lymphocytic infiltrates of salivary and tear glands, leading to oral and ocular dryness, and by autoantibody secretion. In one third of the patients, systemic manifestations may occur, development of B-cell lymphoma being the most severe. Recent years have witnessed major advances in understanding the pathogenesis of the disease, including the role of interferon signature and B-cell activation. This progress has allowed moving into a promising and more targeted approach to therapeutic intervention.

KEYWORDS

Sjögren syndrome B cells interferon (IFN) signature B-cell lymphoma chronic B-cell stimulation targeted therapies keratoconjunctivitis sicca mucosa-associated lymphoid tissue (MALT) lymphoma

1742.e3

CHAPTER 252  Sjögren Syndrome  

1743

Viral or hormonal components PDC TLR stimulation IRF5* STAT4* * Autoimmune associated Other TF polymorphisms Female prevalence IFNα and other inflammatory genes

B Cell

HLA Class II restrictions T Cell

BAFF Apoptosis

DC Blebs

Lymphoid hyperplasia

Autoantibodies α-M3R α-Ro α-La

n

tio

ltra

i Inf

Oligoclonal expansion

Apoptosis

Glandular dysfunction FIGURE 252-1.  Hypothetical scenario for development of primary Sjögren syndrome. An environmental factor (e.g., virus) causes epithelial cell and dendritic cell (DC) activation. Plasmacytoid DCs are also activated by immune complexes, promoting interferon (IFN) pathway activation, which leads to BAFF overproduction and to B- and T-cell activation. B-cell activation leads to autoantibody production within germinal center–like structures. Interleukin-12 secreted by myeloid DCs leads to natural killer cell and helper T-cell type 1 activation, which promotes tissue damage and IFN-γ production. IFN-α and IFN-γ enhance BAFF secretion. Epithelial cells release autoantigens that participate in immune complex formation and perpetuate the vicious cycle of immune system overactivation. BAFF = B-cell activating factor of the tumor necrosis factor family; IRF5 = interferon regulatory factor 5; PDC = plasmacytoid dendritic cell; STAT4 = signal transducer and activator of transcription 4; TF = transcription factors; TLR = toll-like receptors.

Decreased lacrimal flow and impaired lacrimal composition lead to damage of the corneal and conjunctival epithelia, a condition known as keratoconjunctivitis sicca. As a result of keratoconjunctivitis sicca, Sjögren syndrome patients might experience foreign-body sensation, grittiness, irritation, photosensitivity, and thick rope-like secretions at the inner canthus, all leading to increased discomfort and possibly visual impairment, with considerable functional disability. Furthermore, ocular complications include corneal ulceration and scarring, bacterial keratitis, and eyelid infections that require continuous ophthalmologic care and treatment.

Systemic

In addition to the sicca features, systemic manifestations occur in approximately 20 to 30% of primary Sjögren syndrome patients.9,10 Of note, it has been increasingly appreciated that the extraglandular manifestations in Sjögren syndrome can be divided into two major types according to the underlying pathophysiologic mechanism. Thus, lymphocytic infiltration of the epithelia of organs beyond the exocrine glands (e.g., renal, liver, and bronchial epithelial cells) results in interstitial nephritis, autoimmune cholangitis, and obstructive bronchiolitis, respectively. These clinical features appear early and usually have a benign course. On the other hand, immune complex deposition as a result of the ongoing B-cell hyperreactivity can give rise to the extraepithelial manifestations—palpable purpura, glomerulonephritis, interstitial pneumonitis, and peripheral neuropathy—that are linked to increased morbidity and risk for lymphoma development. The main systemic manifestations are listed in Table 252-1. Peripheral neuropathy may occur through various mechanisms. Vasculitis may be present with cryoglobulinemia, leading to both sensory and motor symptoms. More frequently, pure sensory neuropathy is present, sometimes purely ataxic and sometimes in the form of small-fiber neuropathy. This latter entity is difficult to diagnose because clinical and electromyographic examinations are normal. The diagnosis may be made by skin biopsy showing rarefaction of sensory small fibers.

Sjögren Syndrome and Non-Hodgkin Lymphomas

Chronic polyclonal B-cell activation is commonly present in primary Sjögren syndrome, which may explain why this autoimmune disease has the strongest association with the development of B-cell lymphoma (relative risk, 15 to 20). More recent studies have estimated this risk at a lower level: 6 in Denmark and Sweden, 7 in Taiwan, and 9 in Norway. Lymphomas complicating primary Sjögren syndrome have specific features (Chapter 176).11,12 They are mostly B-cell non-Hodgkin lymphomas with a predominance of low-grade, marginal-zone histologic type. Mucosal localization is predominant, notably as mucosa-associated lymphoid tissue (MALT) lymphomas. Interestingly, lymphomas often develop in organs where primary Sjögren syndrome is active, such as salivary glands. In the setting of Sjögren syndrome, chronic autoimmune B-cell activation plays the major role in the lymphomagenesis process, and the identified predictors of lymphoma development in primary Sjögren syndrome are in line with this phenomenon. The main clinical predictors are permanent swelling of salivary glands, splenomegaly, lymphadenopathy, and palpable purpura. The main biologic predictors are positivity of rheumatoid factor (RF), cryoglobulinemia, lymphopenia (especially CD4 lymphopenia), low complement levels, and a monoclonal component in serum or urine. Three novel predictive factors for lymphoma development have been recently described: (1) the presence of ectopic germinal centers associated with the occurrence of lymphoma in primary Sjögren syndrome patients; (2) demonstration that BAFF levels are increased in primary Sjögren syndrome patients with current or previous lymphoma compared with patients without lymphoma; and (3) abnormalities of the gene TNFAIP3 coding for the A20 protein that regulates NF-κB activation, found in up to 77% of MALT lymphomas complicating primary Sjögren syndrome. In half of the cases, TNFAIP3 mutations or deletions occur within lymphoma cells; in the other 50%, they involve germline TNFAIP3 mutations with functional consequences.

1744

CHAPTER 252  Sjögren Syndrome  

TABLE 252-1 EXTRAGLANDULAR MANIFESTATIONS OF PRIMARY SJÖGREN SYNDROME

TABLE 252-2 DRUGS AND TOXINS THAT MIGHT DECREASE LACRIMAL AND SALIVARY SECRETION

CONSTITUTIONAL SYMPTOMS

STRONG EFFECT

Fatigue Low-grade fever

Atropine, atropinic antiparkinsonian drugs, anticholinergic antihistaminic drugs Antidepressants: imipramine (amitriptyline) and inhibitors of monoamine oxidase Neuroleptics Morphine, codeine, tramadol A-type botulinum toxin Class IA antiarrhythmic (disopyramide) Isotretinoin Toxins and psychotropic drugs: tobacco, ecstasy, cannabis, cocaine

SKIN AND VASCULAR Small vessel vasculitis Palpable purpura Raynaud phenomenon Photosensitivity reactions similar to subacute cutaneous systemic lupus erythematosus Xerosis UPPER AND LOWER AIRWAYS Pyogenic sialoadenitis or parotitis Interstitial pneumonitis or fibrosis Chronic bronchitis Bronchiectasis Bronchiolitis obliterans with organizing pneumonia Chronic obstructive pulmonary disease MUSCULOSKELETAL Polyarthralgia, polyarthritis Myopathy, polymyositis RENAL Type I renal tubular acidosis Tubular interstitial nephritis Cryoglobulinemia-associated glomerulonephritis NEUROLOGIC Peripheral motor sensory neuropathy Pure sensory neuropathy (including pure ataxic neuropathy) Small-fiber sensitive neuropathy Multiple sclerosis–like focal lesions Spinal cord dysfunction, including transverse myelitis HEPATOBILIARY Autoimmune primary biliary cholangitis NEOPLASIA Lymphadenopathy, MALT (mucosa-associated lymphoid tissue) lymphoma

Laboratory Findings

The most common serologic finding in primary Sjögren syndrome is hypergammaglobulinemia. The elevated γ-globulins contain several autoantibodies directed against non-organ-specific antigens,13 such as RF and antinuclear antibody (ANA).14 Specific ANA, anti-SSA/Ro, and anti-SSB/La antibodies are present in 60 to 80% and 30 to 40% of patients, respectively, and anti-SSB/ La is never present without anti-SSA/Ro. Of note, the presence of anti-SSA/ Ro, possibly with anti-SSB/La, may mediate complete heart block of newborns owing to cross-mimicry between specific fetal myocardial antigens and epitopes of the SSA/Ro-SSB/La complex. Anemia of chronic inflammation and high erythrocyte sedimentation rates (due to hypergammaglobulinemia) are frequently encountered, whereas C-reactive protein levels are usually within normal limits. Cytopenias (most frequently lymphopenia and neutropenia) can also occur. In the setting of interstitial nephritis, the presence of hypokalemic, hyperchloremic acidosis might reveal distal renal tubular acidosis. A monoclonal immunoglobulin can be detected in 10 to 15% of patients with Sjögren syndrome, depending on the technique used. Approximately 20% of patients with Sjögren syndrome have cryoglobulins in their sera. Complement levels may be decreased, especially C4. This low C4 level may be either genetically determined or secondary to consumption (in immune complexes or cryoglobulinemia).  

DIAGNOSIS

Differential Diagnosis

The definition of primary Sjögren syndrome had suffered for a long time from the absence of accurate and consensus-driven diagnostic criteria. This is important because the patients’ main symptoms (dryness, fatigue, and pain) are frequent in the general population. They can be caused by numerous drugs (Table 252-2), anxiety and/or depression, other comorbidities, or aging (Table 252-3). Sarcoidosis can mimic the clinical picture of Sjögren syndrome.

MODERATE EFFECT β-Adrenergic blockers α-Adrenergic blockers

Calcium-channel blockers Benzodiazepines Inhibitors of serotonin reuptake (very slight effect) Histamine-1 antihistaminic drugs Diuretics Some antiretroviral drugs

TABLE 252-3 THE DIFFERENT CAUSES OF SICCA SYMPTOMS Drugs, particularly psychotropic drugs (see Table 252-2) Aging, postmenopausal estrogen deficiency Prolonged use of contact lenses Fibromyalgia and chronic fatigue syndrome Anxiodepressive syndromes Head and neck radiotherapy Diabetes (uncontrolled) Severe hyperlipidemia Amyloidosis Sarcoidosis Lymphoma Graft-versus-host disease Some viral infections (HIV, HCV, HTVL-1) IgG4-related sialoadenitis Sjögren syndrome HCV = hepatitis C virus; HIV = human immunodeficiency virus; HTVL-1 = human T-lymphocytic virus-1; IgG4 = immunoglobulin G4.

However, in sarcoidosis, minor salivary gland biopsy reveals noncaseating granulomas, and autoantibodies are typically absent. Other Sjögren syndrome mimickers include chronic graft-versus-host disease, amyloidosis, infection with viruses such as HIV, human T-lymphotropic virus 1 (HTLV-1), hepatitis C virus, and immunoglobulin G4 (IgG4)-related disease (Chapter 259). The latter disease is important in the differential diagnosis of Sjögren syndrome. It more often involves men with salivary or lacrimal gland enlargement (previously called Mikulicz syndrome) with previous organ-specific autoimmune disease (like autoimmune pancreatitis) without anti-SSA/SSB antibodies. Sicca symptoms without salivary lymphoid infiltrate and without anti-SSA/ SSB antibodies may be part of the fibromyalgia syndrome (Chapter 258), and several acronyms have been proposed for designating these patients: sicca asthenia polyalgia syndrome or dry eyes and mouth syndrome. The arthralgias and arthritis can also sometimes mimic rheumatoid arthritis (Chapter 248).15

Diagnostic Criteria

International agreement has established a definition of Sjögren syndrome based on the American-European Consensus Group (AECG) criteria, which require the presence of either focal lymphocytic infiltrates in minor salivary glands with a focus score of 1 or more, or anti-SSA/SSB autoantibodies (Table 252-4). A new set of preliminary criteria for Sjögren syndrome classification was proposed by an expert consensus panel (American College of Rheumatology [ACR]-Sjögren International Collaborative Clinical Alliance [SICCA]).16 According to these criteria, classification of an individual as a patient with primary Sjögren syndrome requires the presence of two out of three of the following objective items: (1) a positive serum test for anti-Ro/SSA and/or anti-La/SSB antibodies, or positive RF and ANA (titer >1: 320); (2) presence of keratoconjunctivitis sicca, defined by an ocular staining score higher than 3; and (3) presence of focal lymphocytic sialoadenitis, defined by a focus score of 1 focus/4 mm2 or higher in a labial salivary gland biopsy.

Assessment of Activity of the Disease

An international expert group recently set up a Sjögren syndrome activity score under the umbrella of the European League Against Rheumatism. (EULAR). Two indices have been developed: (1) a patient-administered questionnaire to assess subjective features, the EULAR Sjögren Syndrome

TABLE 252-4 2016 ACR/EULAR SS CLASSIFICATION CRITERIA OF PRIMARY SJÖGREN SYNDROME* INCLUSION CRITERIA

ITEM

• At least one symptom of ocular or oral dryness (based on AECG questions), or • Suspicion of SS from ESSDAI questionnaire (at least one domain with positive item)

SCORE

Focus Score ≥ 1

3

Anti-SSA Ab+

3

Ocular Staining Score ≥ 5

1

Schirmer test ≤ 5 mm/5 min

1

Unstimulated Salivary Flow ≤ 0.1 mL/min

1

Total

9.0

SS CLASSIFICATION CRITERIA pSS case defined by a score ≥ 4 Based on 98% consensus among clinician experts Note: • Exclusion criteria similar to AECG and ACR criteria except past lymphoma and cured HCV infection • Patients taking anticholinergic drugs should be evaluated for OSS, Schirmer, and UWS flow after a sufficient interval off these medications *2016 American College of Rheumatology/European League Against Rheumatism Sjögren’s Syndrome Classification Criteria for Primary Sjögren’s Syndrome AECG = American-European Consensus Group; Anti-SSA = anti-Ro antibody; ESSDAI = EULAR Sjögren’s Syndrome (SS) Disease Activity Index; HCV = hepatic C virus; UWS = unstimulated whole saliva. Shiboski CH, Shiboski SC, Seror R, et al; International Sjögren’s Syndrome Criteria Working Group. 2016 American College of Rheumatology/European League Against Rheumatism classification criteria for primary Sjögren’s syndrome: a consensus and data-driven methodology involving three international patient cohorts. Ann Rheum Dis. 2017;76:9-16.

Patient Reported Index, based on three different visual analogue scores: dryness, fatigue, and limb pain; and (2) a systemic activity index to assess systemic complications, the EULAR Sjögren Syndrome Disease Activity Index. The latter index comprises 12 domains with three or four levels of activity for each domain. Determination of the threshold of moderate activity as well as the minimal clinically important improvement is in progress, with the objective to base inclusion criteria and primary end points of future clinical studies on EULAR Sjögren Syndrome Disease Activity Index levels. Ultrasound findings in major salivary glands correlate with subjective and objective oral and ocular dryness and with systemic autoimmune features in patients with primary Sjögren syndrome. Ultrasound also can be useful for follow-up assessments.17

TREATMENT  Symptomatic Treatment

Muscarinic agonists (pilocarpine hydrochloride and more recently cevimeline hydrochloride) are effective for treating sicca features (oral dryness and, to a lesser extent, ocular dryness). A1  A2  Topical cyclosporine collyrium (0.05%) also was effective for moderate or severe ocular dryness and inflammation in a randomized controlled trial versus placebo, as were 0.1% clobetasone butyrate eye-drops. A3  Environmental measures (avoidance of hot air heating systems or excessive air conditioning, use of a humidifier, appropriate glasses to protect the eye from evaporating air flow) and “little means” (sugar-free chewing gums, regular water drinking, salivary substitutes) might be useful. Regular dental examinations and oral hygiene are crucial for reducing subsequent oral health issues (i.e., caries and periodontal disease associated with xerostomia). To treat pain, simple analgesics should be used first, particularly acetaminophen/ paracetamol, which does not cause dryness.

Biologics

Randomized trials of infliximab and etanercept did not show any efficacy of TNF-blocker agents in primary Sjögren syndrome on a composite primary outcome including limb pain, fatigue, and dryness visual analogue scales. B-cell targeting appears to be a promising strategy in primary Sjögren syndrome.19 Although cohort registries and several randomized controlled trials have reported at least some short-term efficacy using rituximab (a monoclonal anti-CD20 antibody), A5  A6  other trials and a meta-analysis have shown no benefit. A7  A8  Rituximab may be useful in cases of persistent parotid swelling or systemic complications, especially in cryoglobulinemia-induced vasculitis. Data are also inconclusive for belimumab (an anti-BAFF monoclonal antibody). ,



,

FUTURE DIRECTIONS

Sjögren syndrome is a model of autoimmune disease because it can be primary or associated with other autoimmune diseases; it represents autoimmunity where the risk for lymphoma is most important and where increased risk for cardiovascular and cerebrovascular disease has been found in a systematic review and meta-analysis. Sjögren syndrome is the autoimmune disease for which the target tissue of autoimmunity is the most easily available, with the lip biopsy being necessary for diagnosis. Recent progress in pathophysiology has emphasized a number of similarities with systemic lupus erythematosus that support consideration of Sjögren syndrome as a form of lupus of the mucosa. Even if the pathogenetic mechanisms of the disease remain largely unknown, improved knowledge of the effector mechanisms will allow identification of new targets for future therapy. Moreover, with the validated composite activity scores of EULAR Sjögren Syndrome Patient Reported Index and EULAR Sjögren Syndrome Disease Activity Index,20 the tools are now available to begin new clinical trials with novel drugs for this disease that will improve the poor quality of life currently associated with it.

  Grade A References A1. Hamad A, Lodi G, Porter S, et al. Interventions for dry mouth and hyposalivation in Sjögren’s syndrome: a systematic review and meta-analysis. Oral Dis. 2019;25:1027-1047. A2. Shih KC, Lun CN, Jhanji V, et al. Systematic review of randomized controlled trials in the treatment of dry eye disease in Sjögren syndrome. J Inflamm (Lond). 2017;14:1-11. A3. Aragona P, Spinella R, Rania L, et al. Safety and efficacy of 0.1% clobetasone butyrate eyedrops in the treatment of dry eye in Sjögren syndrome. Eur J Ophthalmol. 2013;23:368-376. A4. Wang SQ, Zhang LW, Wei P, et al. Is hydroxychloroquine effective in treating primary Sjögren’s syndrome? A systematic review and meta-analysis. BMC Musculoskelet Disord. 2017;18:1-13. A5. Meijer JM, Meiners PM, Vissink A, et al. Effectiveness of rituximab treatment in primary Sjögren’s syndrome: a randomized, double-blind, placebo-controlled trial. Arthritis Rheum. 2010;62:960-968. A6. Devauchelle-Pensec V, Mariette X, Jousse-Joulin S, et al. Treatment of primary Sjögren syndrome with rituximab: a randomized trial. Ann Intern Med. 2014;160:233-242. A7. Bowman SJ, Everett CC, O’Dwyer JL, et al. Randomized controlled trial of rituximab and costeffectiveness analysis in treating fatigue and oral dryness in primary Sjögren’s syndrome. Arthritis Rheumatol. 2017;69:1440-1450. A8. Letaief H, Lukas C, Barnetche T, et al. Efficacy and safety of biological DMARDs modulating B cells in primary Sjögren’s syndrome: systematic review and meta-analysis. Joint Bone Spine. 2018;85:15-22.

,

Immunomodulatory Drugs

To date, no immunomodulatory drug has proved efficacious in primary Sjögren syndrome.18 Severe organ manifestations of primary Sjögren syndrome have to be treated in accordance with treatment modalities used in systemic lupus erythematosus or other connective tissue diseases. Randomized trials have assessed hydroxychloroquine in primary Sjögren syndrome and failed to demonstrate any clinical efficacy. A4  Despite these negative results on clinical outcomes, hydroxychloroquine is frequently used in primary Sjögren syndrome, especially to treat arthralgia with or without synovitis or purpura. Controlled studies are needed to assess the use of methotrexate, leflunomide, mycophenolate sodium, azathioprine, and cyclosporine. Intravenous immunoglobulin (IVIG) has been used in the treatment of Sjögren syndrome–associated sensorimotor neuropathies or nonataxic sensory neuropathy without any necrotizing vasculitis.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 252  Sjögren Syndrome  

GENERAL REFERENCES 1. Alani H, Henty JR, Thompson NL, et al. Systematic review and meta-analysis of the epidemiology of polyautoimmunity in Sjögren’s syndrome (secondary Sjögren’s syndrome) focusing on autoimmune rheumatic diseases. Scand J Rheumatol. 2018;47:141-154. 2. Rischmueller M, Tieu J, Lester S. Primary Sjögren’s syndrome. Best Pract Res Clin Rheumatol. 2016;30:189-220. 3. Qin B, Wang J, Yang Z, et al. Epidemiology of primary Sjögren’s syndrome: a systematic review and meta-analysis. Ann Rheum Dis. 2015;74:1983-1989. 4. Maciel G, Crowson CS, Matteson EL, et al. Prevalence of primary Sjögren’s syndrome in a US population-based cohort. Arthritis Care Res (Hoboken). 2017;69:1612-1616. 5. Teos LY, Alevizos I. Genetics of Sjögren’s syndrome. Clin Immunol. 2017;182:41-47. 6. Nocturne G, Mariette X. B cells in the pathogenesis of primary Sjögren syndrome. Nat Rev Rheumatol. 2018;14:133-145. 7. Bhattarai KR, Junjappa R, Handigund M, et al. The imprint of salivary secretion in autoimmune disorders and related pathological conditions. Autoimmun Rev. 2018;17:376-390. 8. Mariette X, Criswell LA. Primary Sjögren’s syndrome. N Engl J Med. 2018;378:931-939. 9. Goules AV, Tzioufas AG. Primary Sjögren’s syndrome: clinical phenotypes, outcome and the development of biomarkers. Autoimmun Rev. 2016;15:695-703. 10. Brito-Zerón P, Baldini C, Bootsma H, et al. Sjögren syndrome. Nat Rev Dis Primers. 2016;2: 1-20.

1745.e1

11. Alunno A, Leone MC, Giacomelli R, et al. Lymphoma and lymphomagenesis in primary Sjögren’s syndrome. Front Med (Lausanne). 2018;5:1-7. 12. Vasaitis L, Nordmark G, Theander E, et al. Comparison of patients with and without pre-existing lymphoma at diagnosis of primary Sjögren’s syndrome. Scand J Rheumatol. 2018;1-6. 13. Tong L, Koh V, Thong BY. Review of autoantigens in Sjögren’s syndrome: an update. J Inflamm Res. 2017;10:97-105. 14. Bowman SJ. Primary Sjögren’s syndrome. Lupus. 2018;27:32-35. 15. Mirouse A, Seror R, Vicaut E, et al. Arthritis in primary Sjögren’s syndrome: characteristics, outcome and treatment from French multicenter retrospective study. Autoimmun Rev. 2019;18:9-14. 16. Shiboski CH, Shiboski SC, Seror R, et al. 2016 American College of Rheumatology/European League Against Rheumatism classification criteria for primary Sjögren’s syndrome: a consensus and datadriven methodology involving three international patient cohorts. Ann Rheum Dis. 2017;76:9-16. 17. Martire MV, Santiago ML, Cazenave T, et al. Latest advances in ultrasound assessment of salivary glands in Sjögren syndrome. J Clin Rheumatol. 2018;24:218-223. 18. Romão VC, Talarico R, Scire CA, et al. Sjögren’s syndrome: state of the art on clinical practice guidelines. RMD Open. 2018;4:1-8. 19. Andreu Sanchez JL, Fernandez Castro M, Del Campo Fontecha PD, et al. SER recommendations on the use of biological drugs in primary Sjögren’s syndrome. Reumatol Clin. 2019. [Epub ahead of print.] 20. Del Papa N, Vitali C. Management of primary Sjögren’s syndrome: recent developments and new classification criteria. Ther Adv Musculoskelet Dis. 2018;10:39-54.

1745.e2

CHAPTER 252  Sjögren Syndrome  

REVIEW QUESTIONS 1. A 59-year-old woman is referred to your office because of dry mouth, grittiness of eyes, and a rash on both legs. Physical examination discloses unilateral parotid enlargement and a purpuric rash of the lower extremities. Past medical history is unremarkable to date. Among the following diagnostic tests, which one is the most useful to establish the diagnosis of Sjögren syndrome? A . Complement levels B. Cryoglobulins C. Serum protein electrophoresis D. Salivary flow E. Anti-Ro/SSA antibodies Answer: E  Anti-Ro/SSA positivity is one of the two required 2002 European/ American classification criteria for Sjögren syndrome (SS) and one out of the three 2012 preliminary ACR/SICCA 2012 criteria. Low complement levels, especially C4, and cryoglobulinemia have been designated as adverse prognostic factors for lymphoma development and mortality among SS patients. However, they are not included in the classification criteria. Hypergammaglobulinemia and monoclonal gammopathy are likewise not included in the classification criteria of SS. Salivary flow rate of less than 1.5 mL/15 minutes is an objective criterion of oral dryness, included in the 2002 European/ American classification SS criteria but not in the 2012 preliminary ACR/ SICCA criteria. It is not as specific as anti-Ro/SSA antibodies. 2. A 55-year-old woman presents with mild pain affecting her hands as well as ocular and mouth dryness for 10 years. Her examination reveals positive ocular staining for both eyes, revealing keratoconjunctivitis sicca, positive anti-Ro/SSA antibodies, and negative hepatitis C virus and HIV serology. Additional testing included complete blood count, revealing lymphopenia with normal hemoglobin levels and platelet counts. Mixed monoclonal cryoglobulinemia and low C4 levels were also detected. Which of the following statements is true? A . The patient fulfills both the 2002 American/European Consensus Group (AECG) criteria and the 2012 preliminary ACR/SICCA criteria for primary SS (pSS). B. The patient fulfills only the 2002 AECG criteria for pSS. C. A minor salivary gland biopsy is mandatory to classify the patient as having SS according to 2002 AECG group criteria for pSS. D. A minor salivary gland biopsy is mandatory to classify the patient as having SS according to 2012 preliminary ACR/SICCA criteria for pSS. E. The patient suffers from sicca asthenia polyalgia syndrome. Answer: A  Two out of three 2012 preliminary ACR/SICCA criteria are fulfilled as well as four out of six 2002 AECG criteria. Because anti-Ro/SSA antibodies are present, no histopathologic confirmation is required for the classification of pSS according to 2002 AECG criteria in the presence of subjective and objective symptoms and signs of salivary and lacrimal gland involvement. Sicca asthenia polyalgia syndrome is a diagnosis of exclusion and refers to the presence of fibromyalgia-like features in association with sicca symptomatology.

3. Which of the following is the best option for improving a patient’s ocular dryness? A . Methotrexate B. Pilocarpine C. Hydroxychloroquine D. Ocular cyclosporine drops E. Infliximab Answer: D  Ocular cyclosporine drops have proved useful in the treatment of dry eye. The efficacy of methotrexate on ocular dryness has never been demonstrated. Ocular dryness is less frequently improved than salivary dryness with secretagogues (i.e., pilocarpine). The efficacy of hydroxychloroquine on dryness has not been demonstrated. No efficacy of infliximab has been demonstrated in primary Sjögren syndrome. 4. Which of the following features is not considered an adverse predictor for lymphoma development in Sjögren syndrome? A . Presence of splenomegaly B. Presence of purpura C. Salivary flow rate less than 1 mL/15 minutes D. Presence of persistent parotid gland enlargement E. Presence of CD4 lymphopenia Answer: C  All the other items are predictive factors of lymphoma. An association has not been reported between salivary flow rate and risk for lymphoma development. 5. A 55-year-old woman is referred to your department for polysynovitis and palpable purpura. She fulfills the AECG 2002 criteria for Sjögren syndrome. Which of the following treatments would you recommend to treat these systemic manifestations? A . Infliximab B. Intravenous immunoglobulin (IVIG) C. Pilocarpine D. Rituximab E. Cyclophosphamide Answer: D  Three randomized controlled trials (RCTs) and one registry study suggested possible efficacy of rituximab in primary Sjögren syndrome, at least in some systemic manifestations. Two RCTs did not demonstrate any efficacy of tumor necrosis factor blockers (i.e., infliximab) in primary Sjögren syndrome. IVIG has been proposed in open studies in the treatment of Sjögren syndrome– associated neuropathies without vasculitis but never has demonstrated any efficacy in other systemic manifestations. The secretagogue pilocarpine has no effect on systemic manifestations. The systemic manifestation presented by the patient is not severe enough to propose cyclophosphamide.

CHAPTER 253  Inflammatory Myopathies  

1745

253  INFLAMMATORY MYOPATHIES STEVEN A. GREENBERG



OVERVIEW

The inflammatory myopathies are a heterogeneous group of acquired disorders in which the immune system is thought to play a major pathogenic role.1 Though some genetic disorders affecting muscle also have significant involvement of the immune system and are treated with immunosuppressive therapy as standard of care (e.g., treatment of Duchenne muscular dystrophy with

CHAPTER 253  Inflammatory Myopathies  

ABSTRACT

The inflammatory myopathies are a heterogeneous group of acquired disorders in which the immune system is thought to play a major pathogenic role. The four major subtypes of inflammatory myopathy are dermatomyositis, polymyositis, immune-mediated necrotizing myopathy, and inclusion body myositis (also called sporadic inclusion body myositis). These disorders have distinct clinical and pathologic features and pathophysiologies.

KEYWORDS

inflammatory myopathies myositis inclusion body myositis dermatomyositis polymyositis

1745.e3

1746

CHAPTER 253  Inflammatory Myopathies  

TABLE 253-1 CLASSIFICATION OF INFLAMMATORY MYOPATHIES DISORDER

AGE RANGE

CLINICAL FEATURES

MUSCLE PATHOLOGY

Dermatomyositis

Juvenile and adult forms

Proximal weakness plus skin

Perimysial and perivascular inflammation, perifascicular atrophy

Polymyositis

Adult (rare in childhood)

Proximal weakness

Endomysial inflammation with invasion of non-necrotic muscle fibers

Immune-mediated necrotizing myopathy

Adult

Proximal weakness

Multifocal necrotic muscle fibers

Inclusion body myositis

Adult >40 years old

Prominent quadriceps and finger flexor weakness; treatment refractory

Endomysial inflammation with invasion of non-necrotic muscle fibers plus rimmed vacuoles

Overlap syndromes

Adult

Myositis plus defined connective tissue disease

Nonspecific inflammation

Other (granulomatous myositis, eosinophilic myositis)

All ages

Proximal or distal weakness

Specific to type (e.g., granulomas present with granulomatous myositis)

corticosteroids), these genetic disorders are not classified as inflammatory myopathies. The four major subtypes of inflammatory myopathy are dermatomyositis, polymyositis, immune-mediated necrotizing myopathy, and inclusion body myositis (also called sporadic inclusion body myositis). These disorders have distinct clinical and pathologic features and pathophysiologies (Table 253-1). Whereas dermatomyositis and polymyositis have been described in the medical literature for over 100 years, immune-mediated necrotizing myopathy and inclusion body myositis have only become defined as syndromes distinct from polymyositis within the last few decades. The original classification of inflammatory myopathies (collectively also referred to as “myositis”2) was developed in 1975 (the Bohan and Peter classification). Since then much progress has been made with the discovery of myositis-specific and myositis-associated autoantibodies, improvements in defining their morphologic features and histopathology, and refining the phenotypic distinctions between the different inflammatory myopathies. These developments necessitated the new 2017 European League Against Rheumatism/ American College of Rheumatology diagnostic criteria and classification for adult and juvenile inflammatory myopathies and their major subgroups.3-5  

EPIDEMIOLOGY

The prevalence of dermatomyositis has been estimated at 100 to 210 per million. The estimated prevalence of polymyositis is confounded by frequent misdiagnosis of inclusion body myositis and muscular dystrophies as polymyositis. Traditionally, polymyositis has been considered more prevalent (70 per million), but comparative studies with attention to inclusion body myositis have found a prevalence of polymyositis of 35 per million, approximately half the prevalence of inclusion body myositis of 70 per million. Inclusion body myositis is the most common acquired muscle disease after age 50.6 The prevalence of immune-mediated necrotizing myopathy is unknown.7 Dermatomyositis has biphasic peaks in prevalence in childhood (7 to 15 years) and in midlife (30 to 50 years), whereas polymyositis peaks in prevalence in midlife. Inclusion body myositis is rarely diagnosed before the age of 40 and is most common after the age of 50. Dermatomyositis and polymyositis have female predominance; inclusion body myositis has male predominance. Ethnicity and worldwide distribution influence the development of various inflammatory myopathies.  

PATHOBIOLOGY

The pathophysiologies of various forms of inflammatory myopathy are poorly understood. These disorders do share in common injury to muscle by the immune system. Much of the theory of pathophysiology of these disorders comes from microscopic examination of muscle biopsies and the distinct pathologies of these disorders (Fig. 253-1). The muscle pathology of dermatomyositis involves loss of blood vessels and injury to myofibers at the edges of muscle fascicles (i.e., perifascicular atrophy; see Fig. 253-1). The relationship of these two features to each other is uncertain but has been postulated to be due to a primary injury to muscle capillaries, followed by ischemic injury to myofibers. An alternative view is that a common factor injures both myofibers and capillaries. Skin pathology shows features analogous to that of muscle, with an interface dermatitis consisting of injury to the basal layer of keratinocytes. Much evidence points toward dermatomyositis as mediated by the type 1 interferon cytokine family, consisting mainly of interferon (IFN)-α and IFN-β.

Studies of dermatomyositis skin and muscle samples show marked upregulation of type 1 IFN-inducible transcripts and proteins uniquely in dermatomyositis among muscle diseases and similarly to systemic lupus erythematosus among skin diseases. The presence of autoantibodies in some patients with dermatomyositis, such as antibodies to the type 1 IFN-inducible protein MDA5, is of uncertain significance but seems likely due to an immune reaction to proteins that are not normally expressed at high levels or exposed to the immune system.8 The paraneoplastic associations of dermatomyositis suggest that in such patients an immune reaction against an underlying malignancy results in bystander injury to muscle and skin. Because polymyositis is a diverse group of disorders, the mechanisms involved are likely to be varied. Pathologically, there is an appearance of invasion of muscle fibers by adaptive immune system cells (T cells) that appears to be antigen driven, so that cytotoxic T cell–mediated autoimmunity directed against an unknown target has been a favored hypothesis. The antigens targeted by this process and the fundamental cause are unknown. Immune-mediated necrotizing myopathy is also a poorly understood disorder.9 It can also be paraneoplastic, suggesting cross-reactions by the immune system with the underlying malignancy and with muscle antigens. More commonly, immune-mediated necrotizing myopathy occurs in association with treatment with statin drugs. The identification of autoantibodies against the target of statins, 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR), in the majority of patients who develop immune-mediated necrotizing myopathy in association with statin use suggests that the upregulation of HMGCR in muscle is directly toxic to muscle and triggers an immune reaction against it. Anti–signal recognition particle (SRP) autoantibodies have also been identified in some patients with immune-mediated necrotizing myopathy. Anti-SRP myopathy, anti-HMGCR myopathy, and autoantibody-negative myopathy have been considered to be three distinct subtypes of immune-mediated necrotizing myopathy.10 The pathogenesis of inclusion body myositis is complex. Two dual pathologies have been noted: degeneration of myofibers and of myonuclei in particular, evident as formation of rimmed vacuoles (see Fig. 253-1A and B) and involvement of the immune system.11 The accumulation of more than 75 different proteins into sarcoplasmic aggregates in a small percentage of inclusion body myositis myofibers has been reported and has given rise to a number of molecular toxicity hypotheses in which certain specific protein aggregates are theorized as injurious to myofibers. The immune system involvement in inclusion body myositis is notable in that whereas most other forms of inflammatory myopathy are generally responsive to immunomodulatory treatments, inclusion body myositis is refractory to treatment. This is particularly remarkable in that inclusion body myositis has the greatest evidence of all the inflammatory myopathies of a highly refined antigen-driven adaptive immune system involvement. Pathology shows very chronic and often marked but variable inflammatory infiltrates of T cells, myeloid dendritic cells, and plasma cells in muscle. Studies of the T-cell receptors have strongly suggested that T-cell autoimmunity is driven by one or more specific antigens, though the identity of any of these antigens is unknown. Studies of a B-cell pathway in inclusion body myositis have led to identification of an autoantibody that is highly specific to inclusion body myositis among muscle diseases. Circulating autoantibodies against a 43-kD muscle protein have been identified as cytoplasmic 5′ nucleotidase 1A (cN1A; NT5C1A). cN1A is a nucleotidase that is most abundant in skeletal muscle and involved

CHAPTER 253  Inflammatory Myopathies  

1747

Rimmed vacuoles

Inclusion bodies

A

B Necrotizing myopathy

C Dermatomyositis

D Polymyositis

Perimysium

E

Endomysium

F

FIGURE 253-1.  Pathologies of inflammatory myopathies. A and B, Rimmed vacuoles (arrowheads) of inclusion body myositis (IBM). C, Invasion of non-necrotic muscle fiber in IBM. D, Scattered necrotic and regenerating myofibers in immune-mediated necrotizing myopathy. E, Perivascular and perimysial inflammation (arrows), with perifascicular atrophy (arrowheads), in dermatomyositis. F, Endomysial inflammation in polymyositis. (With permission from the Inclusion Body Myositis Foundation, Inc.)

in the metabolism of nucleic acids. Serum anti-cN1A autoantibodies are present in 50 to 70% of patients with inclusion body myositis, depending on which assays and what cutoffs are used, and they are highly specific to inclusion body myositis (>90 to 95%) among muscle diseases. The role of blood testing for anti-cN1A autoantibodies in the diagnosis and management of patients with suspected inclusion body myositis is currently being defined, potentially shortening the time to diagnosis, reducing the misdiagnosis rate, and avoiding more invasive muscle biopsy in some patients. A genetic basis of inclusion body myositis has been suggested by its significant association with the class II MHC allele HLA-DRB1*03:01.12  

CLINICAL MANIFESTATIONS AND DIAGNOSIS

A diagnosis of inflammatory myopathy is considered when a patient presents with proximal or distal weakness without sensory symptoms or in patients with the characteristic skin lesions of dermatomyositis. Less frequently, asymptomatic elevated creatine kinase (CK) levels lead to a diagnosis of inflammatory myopathy.13 Most patients with dermatomyositis, polymyositis, or immune-mediated necrotizing myopathy present with subacute proximal weakness of the arms and legs progressing over months, though these diseases may present acutely. Patients with inclusion body myositis present later in

life, usually symptomatic from slowly progressive weakness of knee extensors and finger flexors. More specific diagnostic features of these disorders are considered individually (Table 253-2). Most patients undergo muscle biopsy, or skin biopsy in the case of suspected dermatomyositis, as part of the diagnostic evaluation.

Dermatomyositis

Patients with dermatomyositis typically present with characteristic skin lesions or muscle weakness.14,15 Virtually pathognomonic skin features are a heliotrope rash, a violaceous periorbital macular erythema, sometimes with edema, and Gottron papules, violaceous papules over dorsal metacarpophalangeal and interphalangeal joints of the hands (Fig. 253-2). Periungual telangiectasias and thrombosed capillaries, poikiloderma over photoexposed areas such as the upper back (“shawl sign”), nonscarring alopecia, and subcutaneous calcification are other suggestive signs. Prominent pruritus is also a common feature of dermatomyositis. Muscle weakness in dermatomyositis is less specific, occurring in a pattern indistinguishable from many other muscle diseases. Useful laboratory studies for the evaluation of suspected dermatomyositis include serum CK (though CK can be normal or even below typical laboratory lower limits of normal in patients with highly active disease) and

1748

CHAPTER 253  Inflammatory Myopathies  

TABLE 253-2 CLINICAL DIAGNOSTIC CRITERIA FOR INFLAMMATORY MYOPATHIES DISORDER

DIAGNOSIS

Dermatomyositis

1. Diagnostic skin involvement (heliotrope rash, Gottron papules) OR diagnostic muscle biopsy finding of perifascicular atrophy OR 2. All of the following: • Suggestive skin involvement • Subacute or chronic proximal or distal weakness • Muscle biopsy showing perimysial or perivascular inflammation without features suggesting another disorder (e.g., endomysial inflammation, rimmed vacuoles) OR skin biopsy showing interface dermatitis along with clinical exclusion of systemic lupus erythematosus

Polymyositis

All of the following: 1. Subacute or chronic proximal weakness 2. Elevated serum creatine kinase (CK) 3. Muscle biopsy showing invasion of endomysial inflammation 4. Response to immunotherapy OR appropriate consideration and exclusion of limb-girdle muscular dystrophies and inclusion body myositis

Immune-mediated necrotizing myopathy

Both of the following: 1. Subacute or chronic proximal weakness 2. Muscle biopsy showing necrotizing myopathy, with scattered necrotic or regenerating myofibers and a lack of inflammation other than macrophage invasion of necrotic muscle fiber

Inclusion body myositis

All of the following: 1. Adult >40 years old 2. Finger flexion or quadriceps weakness 3. Muscle biopsy showing endomysial inflammation OR the presence of serum anti-cN1A autoantibodies 4. Muscle biopsy showing rimmed vacuoles OR invasion of non-necrotic muscle fibers OR the presence of serum anti-cN1A autoantibodies

cNIA = cytoplasmic 5′ nucleotidase 1A.

dermatomyositis-associated autoantibody studies (e.g., anti-Jo-1, anti-Mi2, and anti-MDA5). Occasional patients have abnormal serum aldolase but normal serum CK. Skin biopsy showing a cell-poor interface dermatitis supports the diagnosis of dermatomyositis. Muscle biopsy showing perimysial and perivascular inflammation also supports a diagnosis of dermatomyositis, whereas the presence of perifascicular atrophy in a muscle biopsy is pathognomonic for dermatomyositis. Because dermatomyositis is associated with malignancy, appropriate laboratory and radiologic studies should be performed to search for underlying malignancy in all newly diagnosed patients. The most common dermatomyositis-associated malignancies tend to reflect the overall age and gender cancer rates within the individual patient’s population (i.e., breast, lung, and colorectal cancer in Western countries; nasopharyngeal cancer in Asian populations). This observation supports the notion of dermatomyositis as a paraneoplastic process that can develop in virtually any kind of cancer. Clinical diagnostic criteria are outlined in Table 253-2. The clinical features of muscle weakness in dermatomyositis are entirely nonspecific, with no particular pattern indicative of dermatomyositis compared with other muscle diseases. In practice, certain dermatologic clinical findings (heliotrope rash, Gottron papules) or muscle biopsy findings (perifascicular atrophy) are considered nearly pathognomonic for dermatomyositis.

Polymyositis

The diagnosis of polymyositis is often problematic, with historically many patients with genetically defined limb-girdle muscular dystrophies and inclusion body myositis being misdiagnosed as polymyositis. The 1975 criteria for polymyositis that are frequently cited allow for a diagnosis of “definite” polymyositis without a muscle biopsy. In clinical practice today, the core criteria for the diagnosis of polymyositis are subacute proximal weakness, elevated serum CK, and muscle biopsy showing endomysial inflammation without features suggestive of another diagnosis such as inclusion body myositis (see Table 253-2). Patients with defined connective tissue disorders such as Sjögren syndrome or mixed connective tissue disease have “overlap syndromes,” often also classified as polymyositis.16 Patients with immune-mediated necrotizing myopathy have historically been classified as polymyositis but are increasingly classified separately. Patients with inclusion body myositis are frequently misdiagnosed as polymyositis because of a lack of appreciation of the characteristic inclusion body myositis finger flexor weakness and because muscle biopsies

A

B

C

D

FIGURE 253-2.  Clinical findings in dermatomyositis. A, Erythematous to violaceous raised papules overlying the metacarpal and interphalangeal joints, known as Gottron papules. These are considered the hallmark finding in dermatomyositis. B, Cuticular overgrowth and periungual capillary changes, which include dilated and tortuous blood vessels with areas of atrophy, telangiectasia, vessel dropout, and bushy loop formation along the fingernail bed. C, Erythema and minimal edema involving the upper eyelids, with occasional telangiectasia, known as the heliotrope rash. D, Subcutaneous calcification erupting through skin (arrowhead), seen clinically and by x-ray.

CHAPTER 253  Inflammatory Myopathies  

show endomysial inflammation. The presence of autoantibodies such as anti-Jo-1 argue more for polymyositis than inclusion body myositis, though these may be seen in dermatomyositis as well.

Immune-Mediated Necrotizing Myopathy

Immune-mediated necrotizing myopathy has increasingly been separated from the polymyositis category. Acute or subacute proximal weakness indistinguishable from that of polymyositis or dermatomyositis and an elevated CK are nonspecific, but muscle biopsy showing scattered necrotic or regenerating myofibers without inflammation other than macrophages invading these necrotic myofibers is typical of immune-mediated necrotizing myopathy. The presence of anti-HMGCR (3-hydroxy-3-methylglutaryl-coenzyme A reductase) or antiSRP (signal recognition particle) antibodies both suggest immune-mediated necrotizing myopathy. Immune-mediated necrotizing myopathy, particularly when associated with anti-SRP antibodies, may be paraneoplastic, and laboratory and radiologic evaluation for malignancy should be considered.

Inclusion Body Myositis

Inclusion body myositis has a clinical presentation distinct from other inflammatory myopathies.17 Inclusion body myositis weakness is always slowly progressive rather than the acute or subacute weakness more typically seen in other forms of inflammatory myopathies. Clinical diagnostic criteria are shown in Table 253-2. Inclusion body myositis has a high misdiagnosis rate, estimated at approximately 50% of patients. Symptoms of inclusion body myositis rarely are present before the age of 40 years and most commonly occur after the age of 50. The distribution of weakness is usually in finger flexors or quadriceps rather than proximal arms (shoulder abduction) or proximal legs (hip flexion), more typical of polymyositis or dermatomyositis. Inclusion body myositis is a highly atrophying muscle disease,18 and loss of bulk in medial and lateral anterior thighs and ventral forearms is characteristic. Patients present with difficulty walking, buckling of knees, or weakness of grip. The diagnosis of inclusion body myositis can be highly suspected in such patients of appropriate age and findings on examination of quadriceps atrophy and weakness of finger flexors, especially flexor digitorum profundi, responsible for flexion of distal fingertips. Examination of the strength in these distal fingertips, which needs to be done one finger at a time, is often the single most helpful approach to the diagnosis of inclusion body myositis. Serum CK is either normal or modestly elevated (typically less than five times the upper limit of normal). A serum autoantibody, anti-cN1A (also

1749

called anti-NT5C1A), appears highly specific to inclusion body myositis among muscle diseases and may be of diagnostic value. Most patients undergo muscle biopsy, with characteristic features being the presence of rimmed vacuoles seen on hematoxylin and eosin (H&E) and Gomori trichrome staining, along with endomysial inflammation or invasion of non-necrotic muscle fibers. Immunohistochemical stains detecting p62 or TDP-43 are of additional highly specific diagnostic value.

TREATMENT  Generally, most patients with dermatomyositis, polymyositis, and immunemediated necrotizing myopathy respond to immunomodulatory therapies, whereas patients with inclusion body myositis are almost universally refractory. A general approach to treatment is shown in Fig. 253-3.

Treatment of Dermatomyositis and Polymyositis

Most patients with muscle involvement from dermatomyositis and polymyositis are treated with and respond to corticosteroids.19,20 Dosing is typically prednisone at 1 mg/kg/day orally until significant improvement occurs (typically 1-3 months), followed by gradual taper of 10 mg/day/month. Second-line agents include methotrexate, azathioprine, cyclosporine, and intravenous immunoglobulin. Second-line agents are used for two reasons: they may have a better side-effect profile than chronic higher doses of corticosteroids, and they may be necessary for patients whose responses are insufficient to corticosteroids alone. An important decision is whether to start second-line agents concurrently with initial corticosteroid treatment or wait and see how low a dose of corticosteroids offers satisfactory control and then add agents only if the corticosteroid dose cannot be lowered sufficiently. Thus in the former approach, prednisone 60 mg/day and methotrexate 7.5 mg PO weekly might be started concurrently, and the methotrexate dose increased weekly to 15 to 20 mg PO weekly. Once improvement is substantial, the dose of prednisone may be tapered over 3 to 6 months. Stability on methotrexate alone would then be followed by gradual reduction in its dose. For patients with severe initial presentations, the combination of corticosteroids and periodic intravenous immunoglobulin (1 g/kg every 2 weeks) may offer a better chance for more rapid improvement. A number of randomized placebo-controlled trials have shown no benefit for treating dermatomyositis or polymyositis. A1  A2  These studies have almost always used the Bohan and Peter criteria for the diagnosis, which may result in inclusion of patients with limb-girdle muscular dystrophies and inclusion body myositis misdiagnosed as having polymyositis. In the single largest trial, which used rituximab, all subjects received active drug, but the comparison ,

Suspected Inflammatory Myopathy Typical proximal > distal weakness

Finger flexor and quadriceps weakness

Suspect DM or PM or necrotizing myopathy

Suspect IBM Muscle biopsy

Muscle biopsy (+/– skin biopsy)

Not-confirmed IBM +/– treatment

Therapy Muscle weakness

DM skin only

Topical agents; hydroxychloroquine

Prednisone 1 mg/kg/day x 1-3 months; slow taper or Methylprednisolone 500-1000 mg IV/day × 3-5 days, followed by oral prednisone

Confirmed IBM No medical treatment

Consider 2nd-line agent initially, or wait to see if relapse after taper

2nd-line agents: Methotrexate 7.5 mg PO weekly, increasing by 2.5 mg weekly to 20 mg PO weekly as needed; with folic acid 1 mg daily except on day of methotrexate or IVIG 0.4 g/kg/day × 5 days each month or 1 g/kg/day every 2 weeks or Azathioprine 1-3 mg/kg/day

Refractory—consider IBM

FIGURE 253-3.  Approach to treatment of suspected inflammatory myopathy. DM = dermatomyositis; IBM = inclusion body myositis; IVIG = intravenous immunoglobulin; PM = polymyositis. (With permission from the Inclusion Body Myositis Foundation, Inc.)

was treated “early” or “late” (8 weeks later), and no significant differences were found. A3 

Treatment of Inclusion Body Myositis

No therapies have demonstrated efficacy for inclusion body myositis, with negative results for prednisone, intravenous immunoglobulin, methotrexate, antithymocyte globulin, etanercept, interferon-β, and alemtuzumab. Current management of patients with inclusion body myositis is supportive, involving avoidance of falls and the use of ankle supports and gait assistive devices. Tendon transfer to improve hand function has been used.



PROGNOSIS

Most patients with adult dermatomyositis, polymyositis, and statin-associated immune-mediated necrotizing myopathy have a good prognosis but require long-standing immunomodulatory therapy. Many patients with juvenile dermatomyositis may go into long-standing remission or cure with aggressive initial treatment. Patients with anti-SRP–associated immune-mediated necrotizing myopathy may have severe and difficult-to-treat disease. Patients with inclusion body myositis generally have a slowly progressive course, with one series showing a mean time to loss of ambulation of 12 years.

  Grade A References A1. Gordon PA, Winer JB, Hoogendijk JE, et al. Immunosuppressant and immunomodulatory treatment for dermatomyositis and polymyositis. Cochrane Database Syst Rev. 2012;8:CD003643. A2. Vermaak E, Tansley SL, McHugh NJ. The evidence for immunotherapy in dermatomyositis and polymyositis: a systematic review. Clin Rheumatol. 2015;34:2809-2095. A3. Oddis CV, Reed AM, Aggarwal R, et al. Rituximab in the treatment of refractory adult and juvenile dermatomyositis and adult polymyositis: a randomized, placebo-phase trial. Arthritis Rheum. 2013;65:314-324.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 253  Inflammatory Myopathies  

GENERAL REFERENCES 1. Malik A, Hayat G, Kalia JS, et al. Idiopathic inflammatory myopathies: clinical approach and management. Front Neurol. 2016;7:1-19. 2. McHugh NJ, Tansley SL. Autoantibodies in myositis. Nat Rev Rheumatol. 2018;14:290-302. 3. Lundberg IE, Tjärnlund A, Bottai M, et al. 2017 European League Against Rheumatism/American College of Rheumatology classification criteria for adult and juvenile idiopathic inflammatory myopathies and their subgroups. Ann Rheum Dis. 2017;76:1955-1964. 4. Lundberg IE, de Visser M, Werth VP. Classification of myositis. Nat Rev Rheumatol. 2018;14: 269-278. 5. Leclair V, Lundberg IE. New myositis classification criteria—what we have learned since Bohan and Peter. Curr Rheumatol Rep. 2018;20:1-8. 6. Jabari D, Vedanarayanan VV, Barohn RJ, et al. Update on inclusion body myositis. Curr Rheumatol Rep. 2018;20:1-8. 7. Meyer A, Meyer N, Schaeffer M, et al. Incidence and prevalence of inflammatory myopathies: a systematic review. Rheumatology (Oxford). 2015;54:50-63. 8. Palterer B, Vitiello G, Carraresi A, et al. Bench to bedside review of myositis autoantibodies. Clin Mol Allergy. 2018;16:1-17. 9. Pinal-Fernandez I, Casal-Dominguez M, Mammen AL. Immune-mediated necrotizing myopathy. Curr Rheumatol Rep. 2018;20:1-10. 10. Allenbach Y, Mammen AL, Benveniste O, Immune-Mediated Necrotizing Myopathies Working Group, et al. 224th ENMC International Workshop: clinico-sero-pathological classification of immunemediated necrotizing myopathies. Zandvoort, The Netherlands, 14-October 16, 2016. Neuromuscul Disord. 2018;28:87-99.

1750.e1

11. Herbelet S, De Bleecker JL. Immune checkpoint failures in inflammatory myopathies: an overview. Autoimmun Rev. 2018;17:746-754. 12. Britson KA, Yang SY, Lloyd TE. New developments in the genetics of inclusion body myositis. Curr Rheumatol Rep. 2018;20:1-8. 13. Jones J, Wortmann R. Idiopathic inflammatory myopathies-a review. Clin Rheumatol. 2015;34: 839-844. 14. Mainetti C, Terziroli Beretta-Piccoli B, Selmi C. Cutaneous manifestations of dermatomyositis: a comprehensive review. Clin Rev Allergy Immunol. 2017;53:337-356. 15. Concha JSS, Patsatsi A, Marshak-Rothstein A, et al. Advances in cutaneous lupus erythematosus and dermatomyositis: a report from the 4th International Conference on Cutaneous Lupus Erythematosus—an ongoing need for international consensus and collaborations. J Invest Dermatol. 2019;139:270-276. 16. Lepreux S, Hainfellner JA, Vital A. Idiopathic inflammatory myopathies overlapping with systemic diseases. Clin Neuropathol. 2018;37:6-15. 17. Greenberg SA. Inclusion body myositis: clinical features and pathogenesis. Nat Rev Rheumatol. 2019. [Epub ahead of print.] 18. Pérez-Baos S, Prieto-Potin I, Román-Blas JA, et al. Mediators and patterns of muscle loss in chronic systemic inflammation. Front Physiol. 2018;9:1-14. 19. de Souza FHC, de Araujo DB, Vilela VS, et al. Guidelines of the Brazilian Society of Rheumatology for the treatment of systemic autoimmune myopathies. Adv Rheumatol. 2019;59:6. 20. Kohsaka H, Mimori T, Kanda T, et al. Treatment consensus for management of polymyositis and dermatomyositis among rheumatologists, neurologists and dermatologists. J Dermatol. 2019;46: e1-e18.

1750.e2

CHAPTER 253  Inflammatory Myopathies  

REVIEW QUESTIONS 1. A 65-year-old man developed slowly progressive difficulty arising from a chair and experienced buckling of the knees while walking, resulting in several falls. Muscle biopsy showed endomysial inflammation and rimmed vacuoles. The correct diagnosis is: A . Polymyositis B. Dermatomyositis C. Inclusion body myositis D. Immune-mediated necrotizing myopathy E. Overlap syndrome Answer: C  Inclusion body myositis (IBM). IBM is a later-onset, slowly progressive disease presenting with prominent quadriceps or finger flexion weakness. In this case the knee buckling indicates quadriceps weakness. Endomysial inflammation can be seen in either polymyositis or IBM, but rimmed vacuoles confirm IBM as the correct diagnosis. 2. A 35-year-old woman developed proximal weakness and a purplish papular rash over the dorsum of the hands. A skin biopsy might be expected to show what feature and lead to which diagnosis? A . Lymphocytoclastic vasculitis AND polymyositis B. Interface dermatitis AND dermatomyositis C. Granulomas AND granulomatous myositis D. Necrotic cells AND necrotizing myopathy E. Necrotic cells AND dermatomyositis Answer: B  This patient has Gottron papules (see Fig. 253-2A), virtually pathognomonic for dermatomyositis. Skin biopsy in dermatomyositis shows an interface dermatitis, with pathology of the basal layer of keratinocytes lying at the border (interface) between the epidermis and the dermis. 3. A 60-year-old man with proximal weakness and mildly elevated serum creatine kinase (CK) is diagnosed with polymyositis and treated with highdose prednisone 80 mg/day for several months, before slow taper of prednisone to 30 mg/day by month 6. His serum CK improves, but no improvement in strength is present, and he appears mildly weaker. Which statement is most likely true? A . He has refractory polymyositis and needs more aggressive therapy. B. He has developed steroid myopathy in addition to his polymyositis. C. The diagnosis of polymyositis should be reconsidered, and he could undergo a second muscle biopsy or blood diagnostic testing for anticN1A autoantibodies as the next step. D. The diagnosis of polymyositis should be reconsidered; he may have inclusion body myositis or a limb-girdle muscular dystrophy. E. C and D

Answer: E  Patients with IBM in particular, and sometimes limb-girdle muscular dystrophy, are often misdiagnosed as having polymyositis. Serum CK, modestly elevated in IBM, may lower with prednisone treatment, but patients rarely have improved strength. 4. Which syndrome can be paraneoplastic and should prompt thorough investigation for an underlying malignancy? A . Dermatomyositis B. Polymyositis C. Inclusion body myositis D. Granulomatous myositis E. All of the above Answer: A  Dermatomyositis. A new diagnosis of dermatomyositis should prompt a thorough investigation for an underlying malignancy, with reported rates estimated at 15 to 23%.12 The types of malignancy found in patients with dermatomyositis generally reflect those found in age-, sex-, and populationmatched persons without dermatomyositis. Therefore breast, lung, and colorectal cancer are the three most common cancers from Western country cohorts, whereas nasopharyngeal carcinoma is the most common dermatomyositisassociated cancer in Asian studies. This finding further supports the role of dermatomyositis as a paraneoplastic process that can occur with virtually any kind of cancer. (Ungprasert P, Bethina NK, Jones CH. Malignancy and idiopathic inflammatory myopathies. N Am J Med Sci. 2013;5:569-572.) 5. Normal serum CK in a patient with proximal muscle weakness excludes the diagnosis of: A . Dermatomyositis B. Inclusion body myositis C. Muscular dystrophy D. Granulomatous myositis E. None of the above Answer: E  None of the above. Serum CK is a helpful indicator of muscle disease when elevated, but normal CK does not exclude muscle disease. In particular, patients with very active dermatomyositis frequently have normal serum CK.

1750

CHAPTER 254  The Systemic Vasculitides  

TABLE 254-1 NAMES FOR VASCULITIDES ADOPTED BY THE 2012 INTERNATIONAL CHAPEL HILL CONSENSUS CONFERENCE ON THE NOMENCLATURE OF VASCULITIDES LARGE-VESSEL VASCULITIS Takayasu arteritis Giant cell arteritis MEDIUM-VESSEL VASCULITIS Polyarteritis nodosa Kawasaki disease Buerger disease* SMALL-VESSEL VASCULITIS Antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis Microscopic polyangiitis Granulomatosis with polyangiitis (formerly Wegener granulomatosis) Eosinophilic granulomatosis with polyangiitis (formerly Churg-Strauss syndrome) Immune complex small-vessel vasculitis Antiglomerular basement membrane disease Cryoglobulinemic vasculitis Immunoglobulin (Ig)A vasculitis (Henoch-Schönlein purpura) Hypocomplementemic urticarial vasculitis VARIABLE-VESSEL VASCULITIS Behçet syndrome Cogan syndrome SINGLE-ORGAN VASCULITIS Cutaneous leukocytoclastic angiitis Cutaneous arteritis Primary central nervous system vasculitis Isolated aortitis VASCULITIS ASSOCIATED WITH SYSTEMIC DISEASE Lupus vasculitis Rheumatoid vasculitis Sarcoid vasculitis Others (e.g., IgG4-related aortitis) VASCULITIS ASSOCIATED WITH PROBABLE ETIOLOGY

254  THE SYSTEMIC VASCULITIDES JOHN H. STONE



*Buerger disease (thromboangiitis obliterans) is not always considered to be a primary form of vasculitis and was not included in this consensus statement on nomenclature. From Jennette JC, Falk RJ, Bacon PA, et al. 2012 Revised International Chapel Hill Consensus Conference Nomenclature of Vasculitides. Arthritis Rheum. 2013;65:1-11.

DEFINITION

The vasculitides are a heterogeneous group of disorders linked by the common finding of destructive inflammation within blood vessel walls. The most current nomenclature scheme identifies at least 27 different forms of primary vasculitis (Table 254-1). The major forms of vasculitis are discussed in this chapter.  

Hepatitis C virus–associated cryoglobulinemic vasculitis Hepatitis B virus–associated vasculitis Syphilis-associated aortitis Drug-associated immune complex vasculitis Drug-associated ANCA-associated vasculitis Cancer-associated vasculitis Others

CLASSIFICATION

Classification by Vessel Size

The etiology of most forms of vasculitis remains unknown, and major gaps exist in our understanding of the pathophysiologic processes. The most valid basis for classification of the vasculitides is the size of the predominant blood vessels involved. The vasculitides are categorized initially by whether the vessels affected are primarily large, medium, or small (Table 254-2). Large vessels are considered the aorta, its primary branches, and any vessel that is not located within an organ such as a muscle, kidney, nerve, or the skin. Medium-sized vessels, in contrast, consist of the main visceral arteries and their branches. (Thus, the renal artery is considered a large vessel, but its intrarenal branches—the interlobar and arcuate arteries—are medium-sized vessels). Finally, small vessels include smaller intraparenchymal arteries as well as arterioles, capillaries, and venules. Medium-vessel vasculitis and even large-vessel vasculitis can also affect small arteries. However, large-vessel vasculitis affects large arteries more often than medium- or small-vessel vasculitis, medium-vessel vasculitis affects predominantly medium arteries, and small-vessel vasculitis affects predominantly arterioles, capillaries, and venules.

TABLE 254-2 CONSIDERATIONS IN THE CLASSIFICATION OF SYSTEMIC VASCULITIS Size of predominant blood vessels affected Epidemiologic features: Age Sex Ethnic background Pattern of organ involvement Pathologic features: Granulomatous inflammation Immune complex deposition vs. “pauci-immune” histopathology Presence of ANCA in serum ANCA = antineutrophil cytoplasmic antibody.

Additional Considerations in Classification

Several considerations other than blood vessel size are relevant to the classification of vasculitis (see Table 254-2). These are (1) age, sex, and ethnic background of the patient; (2) tropism for particular organs; (3) presence or absence of granulomatous inflammation; (4) participation of immune complexes in the pathophysiologic process; and (5) detection of characteristic autoantibodies in the patients’ serum, such as antineutrophil cytoplasmic antibodies (ANCAs).

CHAPTER 254  The Systemic Vasculitides  

ABSTRACT

The vasculitides are a heterogeneous group of disorders linked by the common finding of destructive inflammation within blood vessel walls. The current Chapel Hill Consensus on nomenclature identifies at least 27 different forms of primary vasculitis. The vasculitides are classified first according to the size of the blood vessel involved—small, medium, or large—but also according to whether their pathology is characterized by granulomatous inflammation or immune complex deposition. Patients with some forms of vasculitis (e.g., the group associated with antineutrophil cytoplasmic antibodies) typically have disease-specific autoantibodies in their blood.

KEYWORDS

1750.e3

systemic vasculitis antineutrophil cytoplasmic antibody (ANCA) positive vasculitis giant cell arteritis Takayasu vasculitis (arteritis) polyarteritis nodosa Kawasaki disease granulomatosis with polyangiitis (formerly Wegener) eosinophilic granulomatosis with polyangiitis (EGPA; formerly Churg-Strauss)

1751

CHAPTER 254  The Systemic Vasculitides  

Age, sex differences, and ethnic variation are discussed later in the section on Epidemiology. The organ tropisms of these disorders are illustrated by the following examples. Whereas immunoglobulin (Ig)A vasculitis (IgAV, also known as Henoch-Schönlein purpura) typically affects the skin, joints, kidneys, and gastrointestinal (GI) tract, granulomatosis with polyangiitis (GPA; formerly Wegener granulomatosis) classically involves the upper airways, lungs, and kidneys. In contrast to both IgAV and GPA, Cogan syndrome involves the eyes, the audiovestibular apparatus of the inner ear, and (in 10 to 15% of cases) the large arteries. The presence or absence of granulomatous inflammation is a crucial element of vasculitis diagnosis and classification. Granulomatous inflammation implicates a small number of vasculitides that bear this hallmark, including GPA, giant cell arteritis, Takayasu arteritis, and eosinophilic granulomatosis with polyangiitis (EGPA; Churg-Strauss syndrome). Immune complexes are essential to the pathophysiologic mechanism of some forms of small- and medium-vessel vasculitis. Complexes of IgA1, for example, are found in IgAV. Immune complexes consisting of IgG, IgM, complement components, and the hepatitis C virion characterize most cases of mixed cryoglobulinemia. In contrast, “pauci-immune” types of small- and mediumvessel vasculitis, such as GPA and microscopic polyangiitis, have little immunoglobulin or complement deposition within diseased tissues. Many but not all patients with pauci-immune forms of vasculitis are ANCA positive.  

EPIDEMIOLOGY

The epidemiologic features of individual forms of systemic vasculitis vary tremendously by geography (Table 254-3). This may reflect genetic influences, variation in environmental exposures, and other unknown disease risk factors. For example, whereas Behçet syndrome is rare in North Americans, affecting only 1 person in approximately 300,000, this condition is several hundred times more common among inhabitants of countries bordering the ancient Silk Route. Similarly, although Takayasu arteritis is rare in the United States— on the order of 3 new cases per million people per year—this disease is reportedly the most common cause of renal artery stenosis in India, where the incidence may be as high as 200 to 300 per million per year. Age is an important consideration in the epidemiology of vasculitis. Eighty percent of patients with Kawasaki disease are younger than 5 years. In contrast,

giant cell arteritis virtually never occurs in patients younger than 50 years, and the mean age of patients with this disease is 72. Age may also have an impact on disease severity and outcome. In IgAV, the overwhelming majority of cases in children (who represent 90% of all cases) have self-limited courses, resolving within several weeks. In adults, however, IgAV has a higher likelihood of chronicity and a poor renal outcome. The distribution of sex varies across many forms of vasculitis. Buerger disease is the only form of vasculitis with a striking male predominance. The greater prevalence of smoking among males in most societies probably explains this predilection. In contrast, Takayasu arteritis has an overwhelming tendency to occur in females (a 9 : 1 female-to-male ratio). The pauci-immune forms of vasculitis, such as GPA, EGPA, and microscopic polyangiitis, occur in males and females with approximately equal frequencies, but the phenotypic expression of these conditions may be affected by both age and sex. The strongest link between any single gene and vasculitis is the association of HLA-B51 with Behçet syndrome. In Behçet syndrome, 80% of Asian patients have the HLA-B51 gene. The prevalence of HLA-B51 is significantly higher among patients with Behçet syndrome in Japan than among nondisease control subjects (55% versus 50 yr, mean age 72 yr; females 3 : 1; northern European ancestry

Takayasu arteritis

Incidence: 3/million

200-300/million (India)

Age > blacks

*Among children younger than 5 years. † Now named granulomatosis with polyangiitis. From Gonzalez-Gay MA, Garica-Porrua C. Epidemiology of the vasculitides. Rheum Clin North Am. 2001;27:729-749.

TABLE 254-4 PATHOLOGIC CHARACTERISTICS OF SELECTED FORMS OF VASCULITIS TAKAYASU ARTERITIS

POLYARTERITIS NODOSA

GRANULOMATOSIS WITH POLYANGIITIS (WEGENER GRANULOMATOSIS)

Vessels involved

Elastic (large) or muscle (mediumsized) arteries

Organ involvement

Aorta, aortic arch and Skin, peripheral nerves, Upper respiratory tract, major branches, and gastrointestinal tract, lungs, kidneys, skin, pulmonary arteries and other viscera eyes

Type of vasculitis and Granulomatous with inflammatory cells some giant cells; fibrosis in chronic stages

*Formerly named Churg-Strauss syndrome.

EOSINOPHILIC GRANULOMATOSIS WITH POLYANGIITIS*

Medium-sized and Small arteries and veins; Small arteries and veins; small muscle arteries sometimes sometimes medium-sized vessels medium-sized vessels

Necrotizing, with mixed cellular infiltrate

Necrotizing or granulomatous (or both); mixed cellular infiltrate plus occasional eosinophils

Upper respiratory tract, lungs, heart, peripheral nerves

HENOCHSCHÖNLEIN PURPURA

CUTANEOUS LEUKOCYTOCLASTIC ANGIITIS

Capillaries, venules, and arterioles

Capillaries, venules, and arterioles

Skin, joints, gastrointestinal tract, kidneys

Skin, joints

Leukocytoclastic, with Leukocytoclastic, with Necrotizing or granulomatous (or some lymphocytes occasional both); prominent and variable eosinophils eosinophils and other eosinophils; IgA mixed infiltrate deposits in affected tissues

1752

CHAPTER 254  The Systemic Vasculitides  

include leukocytoclasis (degranulation and destruction of neutrophils within blood vessel walls), granulomatous findings (with or without giant cells), lymphoplasmacytic infiltrates, varying degrees of eosinophilic infiltration, necrosis, and combinations of all these findings.  

PATHOPHYSIOLOGY

Some pathophysiologic mechanisms are common to many different forms of vasculitis, regardless of the size of the predominant blood vessels involved. Immune complex deposition, for example, is present in several types of vasculitis that involve both medium-sized and small blood vessels. In this section, the general concepts related to the pathogenesis of large-vessel vasculitides are discussed separately from those of medium- and small-vessel vasculitides.

Large-Vessel Vasculitides

The pathologic process in large-vessel vasculitis appears to begin in the adventitia. In both Takayasu arteritis and giant cell arteritis, abundant numbers of activated T lymphocytes are found within inflamed arterial walls, centering on the adventitia. In Takayasu arteritis, most of these T cells appear to be of the CD8+ subtype. Current evidence suggests that the cytotoxic functions of these cells, mediated by perforin and granzyme B, contribute to smooth muscle cell damage in this disease. CD4+ T-cell responses in Takayasu arteritis have not been well defined. In giant cell arteritis (Chapter 255), much evidence now suggests an antigendriven disease, with the site of immunologic recognition events being the adventitia. CD4+ T cells that secrete interferon (IFN)-γ appear to be recruited to the adventitia by a specific antigen(s), the identity of which remains unknown. Both the T cells that orchestrate the transmural inflammation and the inciting antigens are theorized to reach the adventitia through the vasa vasorum. Subsequently, T-cell signals from the adventitia stimulate macrophages and multinucleated giant cells to elaborate an array of downstream mediators, including metalloproteinases and platelet-derived growth factor. Interleukin (IL)-6, known to be a crucial cytokine in giant cell arteritis and probably Takayasu arteritis as well, is produced by macrophages residing in the blood vessel wall. The results of this inflammatory cascade are granulomatous inflammation, destruction of the internal elastic lamina, arterial wall hyperplasia, smooth muscle cell proliferation, intimal thickening, vascular occlusion, and in some cases, weakening of the vessel wall, leading to dilation and aneurysm formation. Matrix metalloproteinases appear to play important roles in destruction of the internal elastic lamina, damage to other vascular tissues, and weakening of the arterial wall.

Medium- and Small-Vessel Vasculitides

Several different pathophysiologic mechanisms are operative among the medium- and small-vessel vasculitides. In many cases, the mechanisms outlined in the following sections overlap.

Immune Complex–Mediated Vascular Injury

Immune complex–mediated tissue injury does not produce a single clinical syndrome but rather applies to many forms of vasculitis and overlaps with injuries caused by other immune mechanisms. Numerous variables influence immune complex–mediated injury, including the physical properties of the immune complexes (e.g., their size), the ability of the immune complexes to activate complement, the antigen-to-antibody ratio, and the hemodynamic features of specific vascular beds. Immune complexes participate in the pathophysiologic process of some forms of both medium- and small-vessel vasculitis, including polyarteritis nodosa, cryoglobulinemia, IgAV, cutaneous leukocytoclastic angiitis, and rheumatoid vasculitis.

Role of Antineutrophil Cytoplasmic Antibodies

ANCAs are directed against antigens that reside within the primary granules of neutrophils and monocytes. Two types of ANCA are relevant to vasculitis: (1) those directed against proteinase 3 (PR3), known as PR3-ANCA; and (2) those directed against myeloperoxidase (MPO), termed MPO-ANCA. ANCA interact with cytokines, neutrophils, monocytes, and other elements of the immune system to amplify ongoing inflammation in certain forms of vasculitis. A striking and still unexplained feature of ANCA-associated vasculitis (AAV) is that patients with primary forms of these conditions virtually never have antibodies to both PR3 and MPO. Despite the specificity of these antibodies, however, evidence for a primary role of ANCA in the etiology of human disease is still tenuous. In GPA, abnormal cytokine regulation interacts with the production of ANCA to fuel the inflammatory response. TH1 cytokines such as interferon

(IFN)-γ, interleukin (IL)-12, and tumor necrosis factor (TNF) appear to play important roles. Under the direction of IL-12, CD4+ T cells from patients with GPA produce elevated levels of TNF, and peripheral blood mononuclear cells secrete increased amounts of IFN-γ. Serum levels of soluble receptors for TNF are elevated in patients with active GPA and normalize with the induction of remission. In vitro priming of activated neutrophils with TNF markedly enhances the ability of ANCA to stimulate neutrophil degranulation. Despite the strong rationale for anti-TNF strategies in GPA, however, a randomized trial of etanercept showed no efficacy in the maintenance of disease remissions. B-cell depletion is a more effective approach to the treatment of AAV. The efficacy of this treatment strategy probably relates to the removal of several B-cell functions beyond their evolution into plasma cells and the production of ANCA. Such other functions include cytokine production, antigen presentation, and B cell–T cell crosstalk.

Superantigen Model

The degree of immune activation in Kawasaki disease and the acute but generally self-limited nature of this illness imply a potential role for superantigens. Superantigens are proteins produced by microbial pathogens (e.g., Staphylococcus aureus or Streptococcus species) that are capable of stimulating large populations of T cells in a manner unrestricted by the class II major histocompatibility complex (MHC). Superantigens bind directly to conserved amino acid residues outside the antigen-binding groove on class II MHC molecules, thereby selectively stimulating T cells that express particular β-chain variable gene segments. Through the binding of this MHC-superantigen complex to its cognate T-cell receptors, as many as 20% of circulating lymphocytes may become activated, leading to a potentially enormous outpouring of cytokines. With regard to the etiology of Kawasaki disease, substantial attention has focused on toxic shock syndrome toxin 1, an exotoxin produced by S. aureus. Superantigens have also been postulated to play roles in the susceptibility to disease flares in GPA. Nasal carriage of S. aureus and superantigens associated with these organisms has been linked to a greater likelihood of disease flares in some studies.

Anti–Endothelial Cell Antibodies

Anti–endothelial cell antibodies can induce endothelial cell injury and lysis through either complement-mediated cytotoxicity or antibody-dependent cellular cytotoxicity. Both of these mechanisms have been demonstrated to cause endothelial injury in in vitro assays employing sera from patients with systemic vasculitis. The ability of these antibodies to damage endothelial cells is an appealing argument for their potential role in forms of vasculitis in which the endothelium is the focus of the inflammation (as opposed to the more external vessel wall layers). However, the true relevance of anti–endothelial cell antibodies to human disease and their importance within the larger context of other disease mechanisms remain unclear.  

CLINICAL MANIFESTATIONS

Large-Vessel Vasculitides Takayasu Arteritis

Takayasu arteritis (Chapter 69) affects the aorta and its major branches. In contrast to atherosclerosis, which is characterized by focal irregular lesions, the lesions of Takayasu arteritis are long, smooth, tapered stenoses (E-Fig. 254-1). The most commonly involved arteries are the subclavian and innominate arteries. Takayasu arteritis has been termed “pulseless disease” because of its ability to obliterate peripheral pulses (particularly in the upper extremities). Exuberant collateral circulation develops over time in response to the gradual narrowing of major arteries, making the loss of digits or limbs from ischemia extremely rare. The extensive development of collateral circulation usually renders unnecessary any attempts to revascularize stenoses of primary aortic branches, such as the subclavian artery. The pulmonary circulation is involved in approximately 50% of cases of Takayasu arteritis. Patients with severe narrowing of the aortic arch vessels supplying the head may develop Takayasu retinopathy, the hypotensive retinopathy leading to neovascularization originally described by Takayasu. In contrast, patients with prolonged hypertension associated with renal artery stenosis demonstrate the classic ocular features of hypertension: “copper wiring” and multiple retinal infarctions. This complication is particularly difficult to diagnose1 and dangerous because vascular narrowings of large arteries to the arms and legs often lead to underestimations of the true central aortic pressure. Takayasu arteritis involvement of the ascending aorta may lead to aortic dilation, aortic regurgitation, aneurysm formation, and aortic rupture.

CHAPTER 254  The Systemic Vasculitides  

A

1752.e1

C

B E-FIGURE 254-1.  Large-vessel disease in Takayasu arteritis. Long, smooth tapering in the left common femoral artery (A) and the right renal artery (B). Dilation of the ascending aorta (C). Aortic regurgitation necessitated an aortic valve replacement and replacement of the ascending aorta with a Gore-Tex graft.

CHAPTER 254  The Systemic Vasculitides  

1753

TREATMENT  For patients with marked symptoms and signs of an inflammatory phase, glucocorticoids (prednisone 1 mg/kg/day) are usually effective in controlling the disease. The toxicity of high doses of glucocorticoids in young women, however, urges the early consideration of alternative agents.2 IL-6 inhibition, shown to be highly effective in giant cell arteritis, A1  also appears to be effective in Takayasu arteritis. A2  Patients can be treated with tocilizumab, administered either intravenously (8 mg/kg each month) or subcutaneously (162 mg each week). Once tocilizumab has been initiated, prednisone should be tapered to low doses (10 mg daily or less) within 3 months and ultimately discontinued altogether, if possible. The optimal duration of tocilizumab in Takayasu arteritis is uncertain. Consideration may be given to tapering (to either 4 mg/kg intravenously each month or 162 mg subcutaneously every other week) after one year. Large-vessel imaging may be useful in guiding decisions regarding the duration of therapy. About 50% of patients will relapse within 10 years,3 so some Takayasu arteritis patients may require chronic treatment with tocilizumab plus (possibly) low-dose glucocorticoids.

A

Giant Cell Arteritis

Giant cell arteritis is the other primary form of vasculitis that involves arteries far larger than vasculitides of any other category.4 This disease is discussed in detail elsewhere (Chapter 255).

Medium-Vessel Vasculitides Polyarteritis Nodosa

Polyarteritis nodosa has a striking predilection for certain organs, particularly the skin, peripheral nerves, GI tract, and kidneys.5 This disease usually begins with nonspecific symptoms such as malaise, fatigue, fever, myalgias, and arthralgias. Overt signs of vasculitis may not occur until weeks or months after onset of the first symptoms. Skin lesions of polyarteritis nodosa include livedo reticularis, subcutaneous nodules, ulcers, and digital gangrene. A majority of patients with polyarteritis nodosa (>80% in some series) have vasculitic neuropathy, typically in the pattern of a mononeuritis multiplex. The classic GI manifestation of polyarteritis nodosa is “intestinal angina,” the occurrence of postprandial abdominal pain. Polyarteritis nodosa can also affect individual GI tract organs such as the gallbladder or appendix, presenting as cholecystitis or appendicitis, respectively. The typical renal manifestation of polyarteritis nodosa is vasculitic involvement of the medium-sized intrarenal arteries, leading to renin-mediated hypertension and renal infarctions. Cardiac lesions, which usually remain subclinical, may lead to myocardial infarction or congestive heart failure. Polyarteritis nodosa usually spares the lungs. The diagnosis of polyarteritis nodosa requires either a tissue biopsy or an angiogram that demonstrates microaneurysms (Fig. 254-1).6 Simultaneous nerve and muscle biopsies (e.g., sural nerve and gastrocnemius muscle) are of high yield if there is a clinical suspicion of vasculitic neuropathy. Symptoms suggestive of a neuropathy can be confirmed by electrodiagnostic studies that demonstrate a sensorimotor axonal neuropathy, often in a mononeuritis multiplex pattern. The pathologic changes in polyarteritis nodosa are limited to the arterial circulation, and the lesions are segmental, favoring the branch points of arteries. In gross pathologic specimens, aneurysmal bulges of the arterial wall may be visible. Histologic sections reveal infiltration and destruction of the blood vessel wall by inflammatory cells, accompanied by fibrinoid necrosis. Granulomatous inflammation is absent.

TREATMENT  Therapeutic regimens involving entecavir or tenofovir combined with plasma exchange and short-term (two-week) courses of glucocorticoids have substantially improved the treatment of hepatitis B virus (HBV)-associated polyarteritis nodosa. Because of increasing use of the HBV vaccine, however, fewer than 10% of polyarteritis nodosa cases now are associated with HBV infections. Patients with idiopathic polyarteritis nodosa and multiorgan involvement require high-dose glucocorticoid treatment initially. Approximately half of patients with idiopathic polyarteritis nodosa achieve remissions or cures with high doses of glucocorticoids alone. Severe cases of multiorgan disease, particularly vasculitic neuropathy, should be treated with cyclophosphamide (2 mg/kg/day, adjusted for renal dysfunction). For patients with idiopathic polyarteritis nodosa limited to the skin, tumor necrosis factor inhibitors may be effective glucocorticoidsparing agents. Most patients with either HBV-associated or idiopathic polyarteritis nodosa ultimately achieve cures of their disease.

B FIGURE 254-1.  Vasculitis of medium-sized arteries in polyarteritis nodosa. A, Mesenteric angiogram showing numerous aneurysms in medium-sized arteries. B, Fibrinoid necrosis (arrows) in a jejunal artery from a patient who required surgical resection of necrotic bowel.

Kawasaki Disease

Kawasaki disease occurs exclusively in young children. Because of its striking mucocutaneous findings and lymphadenopathy, Kawasaki disease is also known as mucocutaneous lymph node syndrome. Features of Kawasaki disease include high fevers, cervical adenopathy, conjunctival congestion, buccal erythema, prominence of the tongue papillae (“strawberry tongue”), a polymorphous truncal rash, erythema of the palms and soles, and desquamation of skin from the fingertips occurring days to weeks into the illness.7,8 In its acuity and severity, Kawasaki disease resembles toxic shock syndrome and scarlet fever, both of which are mediated by superantigens (see Pathophysiology). In a small number of patients with Kawasaki disease, panvasculitis in the coronary vessels leads to acute cardiac complications. Coronary arteritis leads to narrowing of the vessel lumen by the migration of myointimal cells from the media through the fragmented internal elastic lamina. Direct complications include aneurysmal dilation and thrombosis of the coronary arteries, leading to myocardial infarction and possibly to death (in 1 to 2% of patients with Kawasaki disease during the acute illness). Late mortality from myocardial infarction may occur from the thrombosis of coronary artery aneurysms formed during the initial inflammatory stage. Such myocardial infarctions have been reported in middle-aged individuals who had febrile illnesses consistent with Kawasaki disease in childhood.

TREATMENT  The recommended therapeutic regimen in Kawasaki disease is the combination of intravenous immune globulin (IVIG; 400 mg/kg/day on 4 consecutive days) and acetylsalicylic acid (100 mg/kg/day, lowered to 3 to 5 mg/kg/day after resolution of the fever). IVIG prevents the formation of coronary aneurysms in most cases.9 For patients who are predicted to be unresponsive to IVIG, the combination of IVIG plus cyclosporine (5 mg/kg for 5 days) is better than IVIG alone. A3  Glucocorticoids are reserved for salvage therapy in patients whose treatment with IVIG and acetylsalicylic acid has failed. A4 

1754

CHAPTER 254  The Systemic Vasculitides  

Buerger Disease

Buerger disease, also known as thromboangiitis obliterans (Chapter 72), is not considered to be a primary form of vasculitis and was not included in the most recent consensus statement on nomenclature. Buerger disease has a remarkably strong yet poorly understood association with cigarette smoking. In short, it does not occur in the absence of exposure to tobacco. The vessels affected by Buerger disease are the distal medium-sized arteries and veins, particularly vessels at the levels of the ankles and wrists. The disease is characterized by thrombotic obliterations that begin distally and proceed proximally. Buerger disease tends to be segmental in nature, involving 5- to 10-cm lengths of blood vessels. Arterial obliteration leads to the development of collateral vessels with a “corkscrew” appearance on angiography. Vascular occlusion in Buerger disease often leads to the loss of digits and, if smoking persists, to loss of larger amounts of tissue (e.g., hands or feet). Despite the intense involvement of the extremities in Buerger disease, internal organ disease almost never occurs.

TREATMENT  Complete abstinence from tobacco is essential to the treatment of Buerger disease. Failure to stop smoking is associated with a dramatic increase in the risk of limb loss by amputation. No other therapeutic interventions, including glucocorticoids and anticoagulation, have dramatic effects on Buerger disease.

Small-Vessel Vasculitides Antineutrophil Cytoplasmic Antibody–Associated Vasculitides Granulomatosis with Polyangiitis

Classic GPA (formerly Wegener granulomatosis) involves the upper respiratory tract, lungs, and kidneys. Distinctive features may also occur in the eyes, ears, and other organs. The three pathology hallmarks of GPA are (1) granulomatous inflammation in the upper or lower respiratory tract, (2) necrotizing vasculitis affecting arteries or veins, and (3) segmental glomerulonephritis associated with necrosis and thrombosis of capillary loops, with or without granulomatous lesions. Approximately 90% of patients with GPA have nasal involvement, including crusting, bleeding, and obstruction. Cartilaginous inflammation may lead to nasal septal perforation and collapse of the nasal bridge (“saddle nose” deformity). Erosive sinus disease and subglottic stenosis (narrowing of the trachea just below the vocal cords) are highly characteristic of GPA. Both conductive and sensorineural hearing loss can occur in GPA, though conductive lesions caused by middle ear disease are more common. Orbital masses (“pseudotumors” that develop behind the eye), scleritis, and peripheral ulcerative keratitis are the most dangerous ocular lesions. Episcleritis and conjunctivitis also occur. Uveitis is rare. The clinical manifestations of GPA in the lung range from asymptomatic nodules to fulminant alveolar hemorrhage. The most common radiographic findings are pulmonary infiltrates, nodules, and cavitary lesions. Large-airway disease leading to bronchial narrowing is a challenging diagnosis to establish because patients present with few symptoms until advanced disease is present. The clinical presentation of renal disease in GPA is usually rapidly progressive glomerulonephritis: hematuria, red blood cell casts, and proteinuria (usually non-nephrotic). Without appropriate therapy, end-stage renal disease may ensue within weeks. Sixty percent of patients with GPA have musculoskeletal symptoms during their disease course. The presenting complaint is frequently arthralgias or an oligoarthritis that is migratory in nature. Skin lesions in GPA include the full panoply of lesions associated with cutaneous vasculitis, including purpura (Fig. 254-2). Cutaneous nodules over the extensor surfaces of joints, particularly the elbow, may mimic rheumatoid nodules. These lesions are known as cutaneous extravascular necrotizing granulomata or Churg-Strauss lesions. Meningeal inflammation, presenting with headaches, cranial neuropathies, and a clinical picture compatible with chronic meningitis, is perhaps the most common central nervous system (CNS) manifestation of GPA. Mononeuritis multiplex may affect the peripheral nervous system. GPA is the prototype of conditions associated with ANCA.10 Approximately 75 to 80% of GPA patients have antibodies directed against proteinase 3 (PR3), which lead to positive cytoplasmic (C-ANCA) staining on immunofluorescence testing of serum against human neutrophils. Another 10 to 15% have antibodies directed against myeloperoxidase (MPO), which cause perinuclear staining on immunofluorescence. An ANCA-negative assay sample does not exclude GPA, because a substantial minority of patients are ANCA-negative.

FIGURE 254-2.  Cutaneous small-vessel vasculitis showing palpable purpuric lesions with necrosis and crusting.

ANCA titers do not correlate precisely with disease activity and, indeed, should generally not be used as the sole guide to treatment.

TREATMENT  Manifestations of GPA that constitute immediate threats either to the function of a vital organ or to the patient’s life require treatment urgently.11 From the late 1960s until 2010, the combination of cyclophosphamide (2 mg/kg orally daily) and high doses of glucocorticoids (prednisone 1 mg/kg orally daily, tapered during 6 to 12 months) was the standard of care for GPA. Intermittent administration of cyclophosphamide by IV infusion is also effective in remission induction. However, a multicenter clinical trial that compared rituximab to cyclophosphamide in patients with either GPA or microscopic polyangiitis demonstrated that rituximab (375 mg/m2 weekly times four) is at least as effective as the conventional regimen. A5  A6  Rituximab appears to be more effective for AAV patients who present with disease flares. An alternative dosing regimen of rituximab, 1 g times two separated by 2 weeks, may also be effective.12 Limited forms of GPA may respond to the combination of methotrexate (up to 25 mg/ week) and glucocorticoids, but rituximab is now often employed in this setting as well. Rituximab (e.g., 500 mg every 6 months) is more effective than azathioprine for maintaining remission in patients who demonstrate a tendency to flare. A7  In patients with end-stage renal disease, renal transplantation can substantially reduce subsequent mortality.13 ,

Microscopic Polyangiitis

Microscopic polyangiitis is characterized by (1) nongranulomatous necrotizing vasculitis with few or no immune deposits, (2) involvement of small (and possibly medium-sized) blood vessels in the arterial or venous circulation, and (3) tropism for the kidneys and lungs. Many cases of small-vessel vasculitis once regarded as polyarteritis nodosa are now classified more properly as microscopic polyangiitis. In contrast to polyarteritis nodosa, an ANCA-negative disorder, 70% of microscopic polyangiitis patients are ANCA positive. Thus, microscopic polyangiitis is considered to be a form of AAV. The ANCAs in microscopic polyangiitis are usually directed against myeloperoxidase, leading to a perinuclear pattern of staining on immunofluorescence testing (P-ANCA). Microscopic polyangiitis is not characterized by granulomatous inflammation, and upper respiratory tract symptoms, if present at all, are much milder than those associated with GPA.

TREATMENT  The approach to the treatment of microscopic polyangiitis is similar to the treatment of GPA. The combination of rituximab and glucocorticoids is the treatment regimen of choice for most patients with microscopic polyangiitis.

CHAPTER 254  The Systemic Vasculitides  

Eosinophilic Granulomatosis with Polyangiitis

EGPA (formerly named Churg-Strauss syndrome) is an eosinophil-rich form of granulomatous inflammation that involves the respiratory tract and other organs. The disease is associated with necrotizing vasculitis of small to mediumsized vessels. Two hallmarks of EGPA are asthma and eosinophilia. Several phases of EGPA are described: • A prodromal phase characterized by the presence of allergic disease (typically asthma or allergic rhinitis), which may last months to many years • An eosinophilia–tissue infiltration phase in which remarkably high peripheral eosinophilia may occur and tissue infiltration by eosinophils is observed in the lung, GI tract, and other tissues • A vasculitic phase in which systemic necrotizing vasculitis afflicts a wide range of organs, ranging from the heart and lungs to peripheral nerves and skin

TREATMENT  Patients with mild disease may be treated with prednisone. The addition of mepolizumab, 300 mg subcutaneously each month, has also been shown to have an important glucocorticoid-sparing role. A8  Patients with evidence of neurologic, cardiac, renal, or GI involvement should be treated with cyclophosphamide or rituximab14 in addition to glucocorticoids. Although clinical remissions are obtained in more than 90% of patients with EGPA, the majority of patients are unable to discontinue glucocorticoids entirely (primarily because of asthma symptoms), and disease recurrences are seen in 25%. In most cases, relapses are heralded by the return of eosinophilia. Approximately 50% of cases of EGPA are associated with ANCA, usually directed against myeloperoxidase, but the percentage may be higher among untreated patients.

Immune Complex–Mediated Vasculitides

Anti–Glomerular Basement Membrane Disease

Anti–glomerular basement membrane (anti-GBM) disease is vasculitis affecting glomerular capillaries, pulmonary capillaries, or both, accompanied by the deposition of anti–basement membrane autoantibodies within basement membranes. Anti-GBM disease is discussed in detail elsewhere (Chapter 113). Immunoglobulin A Vasculitis/Henoch-Schönlein Purpura

IgA vasculitis (IgAV) is characterized by nonthrombocytopenic purpura, arthritis, abdominal pain, and glomerulonephritis. The histopathologic findings are those of a leukocytoclastic vasculitis with IgA deposition. IgAV can develop at any age, but 80 to 90% of the cases occur in children. Although the cause is unknown, the disease’s seasonal variation and the fact that two thirds of patients with IgAV experience antecedent acute upper respiratory illnesses suggest an infectious trigger in the majority of cases. The diagnosis of IgAV can be confirmed only by demonstration of IgA deposition within and around blood vessel walls. The classic IgAV patient presents with the acute onset of fever, palpable purpura on the lower extremities and buttocks, abdominal pain, arthritis, and hematuria. The clinician must be alert to the possibility of IgAV even when only parts of the syndrome are present. Most patients with IgAV, especially children, have a self-limited disease that lasts an average of 4 weeks.

TREATMENT  Glucocorticoids ameliorate the GI, joint, and skin symptoms in many cases, but some patients respond surprisingly poorly to conventional doses of glucocorticoids, even in doses on the order of 40 to 60 mg/day. Anecdotal evidence suggests that pulse glucocorticoids (e.g., methylprednisolone 500 to 1000 mg/ day times three doses) may abort persistent bouts of IgAV. The efficacy of glucocorticoids in the glomerulonephritis associated with this condition is controversial. Uncontrolled studies suggest that methylprednisolone pulses (1 g/day for three doses), followed by oral prednisone combined with azathioprine or mycophenolate mofetil may be useful in severe glomerulonephritis associated with IgAV.

Hypocomplementemic Urticarial Vasculitis

At least three subtypes of urticarial vasculitis are known: (1) normocomplementemic, a form that is generally idiopathic and benign (which may be viewed as a manifestation of cutaneous leukocytoclastic angiitis); (2) hypocomplementemic, a form that is often associated with a systemic inflammatory disease;

1755

and (3) hypocomplementemic urticarial vasculitis syndrome (HUVS), a potentially severe condition usually associated with autoantibodies to the collagen-like region of C1q. Most patients with the hypocomplementemic subtype have an underlying systemic disorder, such as systemic lupus erythematosus (Chapter 250) or Sjögren syndrome (Chapter 252). Many HUVS patients have C1q “precipitins,” IgG autoantibodies to the collagen-like region of C1q that trigger the classical pathway of complement activation. The role of anti-C1q antibodies in disease pathogenesis remains unclear. The lesions of urticarial vasculitis must be distinguished from the far more common chronic idiopathic urticaria (Chapters 237 and 411). Unlike idiopathic urticaria, the lesions of urticarial vasculitis last more than 48 hours, often have a purpuric component (i.e., they do not blanch), and resolve with postinflammatory hyperpigmentation. In urticarial vasculitis, lesions associated with vasculitis are often accompanied by stinging or burning. Urticarial vasculitis affects the capillaries and postcapillary venules, showing leukocytoclastic vasculitis on light microscopy. Direct immunofluorescence studies reveal both immunoglobulin and complement deposition in or around blood vessels of the upper dermis or the dermoepidermal junction.

TREATMENT  Patients with urticarial vasculitis whose serum complement levels remain normal during attacks often have self-limited disease and require little therapy. Other cases, especially HUVS, may cause life-threatening involvement of the lungs or other organs and require periods of intensive immunosuppression. Treatment decisions in HUVS must be individualized according to the patient’s clinical status.

Cryoglobulinemia

Cryoglobulins are antibodies that precipitate from serum under conditions of cold and resolubilize on rewarming.15 Cryoglobulins are classified into types I, II, and III on the basis of whether monoclonality and rheumatoid factor activity (the ability to bind to the Fc portion of IgG) are present. Type I cryoglobulins, which are monoclonal but lack rheumatoid factor activity, are associated with certain hematopoietic malignant neoplasms (e.g., multiple myeloma) and often lead to hyperviscosity rather than to vasculitis (Chapter 178). In contrast, type II and type III cryoglobulins may be associated with systemic vasculitis involving small (and often medium-sized) blood vessels. Vasculitis results from the deposition of cryoglobulin-containing immune complexes within blood vessel walls and the activation of complement. Cryoglobulin types II and III are termed mixed cryoglobulins because they consist of complexes of both IgG and IgM antibodies. The IgM components in both type II and type III cryoglobulinemia possess rheumatoid factor activity (i.e., assays for rheumatoid factor are positive, indicating binding of the IgM antibody to the Fc portion of IgG). Whereas the IgM component in type II cryoglobulin is monoclonal, the IgM in type III cryoglobulin is polyclonal. Ninety percent of patients with vasculitis secondary to mixed cryoglobulins are hypocomplementemic, with C4 levels characteristically more depressed than C3. Infection with hepatitis C virus (HCV) accounts for at least 80% of the vasculitis cases associated with mixed cryoglobulins.16

TREATMENT  The optimal therapy for most cases of cryoglobulinemic vasculitis is successful treatment of the underlying HCV infection. For cryoglobulinemic patients with relatively mild disease (e.g., frequent purpuric lesions, shallow cutaneous ulcers), short courses of prednisone followed by the institution of effective therapy for HCV may be sufficient. For patients with severe cutaneous ulcers, mononeuritis multiplex, glomerulonephritis, or other manifestations of severe disease, glucocorticoids, rituximab, and possibly a short course of plasma exchange may be indicated.

Variable-Vessel Vasculitides

The variable-vessel vasculitides have no predominant type of vessel involved but rather can affect vessels of any size (small, medium, and large) and any type (arteries, veins, and capillaries).

Cogan Syndrome

The combination of inflammatory eye disease and vestibuloauditory dysfunction is the sine qua non of Cogan syndrome.17 In addition to inflammatory

1756

CHAPTER 254  The Systemic Vasculitides  

disease of the eyes and ears, up to 15% of patients with Cogan syndrome have vasculitis involving medium-sized to large blood vessels. Although the ocular manifestations vary, the classic presentation is the combination of interstitial keratitis and sensorineural hearing loss. Cogan syndrome may appear first in either the eyes or the ears. Although intervals as long as 1 to 2 years have been described between the start of disease in one organ and the appearance of disease in the other, the time between disease manifestations in these organs is usually only a matter of months. Patients usually present with photophobia and blurry vision, sometimes accompanied simultaneously by auditory or vestibular dysfunction. The vascular disease associated with Cogan syndrome remains poorly described but typically involves the primary branches of the thoracic or abdominal aorta.

TREATMENT  Rapidly progressive sensorineural hearing loss requires early and aggressive therapy with high doses of systemic glucocorticoids. Some otolaryngologists also perform intratympanic injections of glucocorticoids. Cyclophosphamide, mycophenolate mofetil, and biologic agents such as tumor necrosis inhibitors or rituximab can be considered for patients with suboptimal responses to glucocorticoids who still have salvageable hearing. Nevertheless, attempts at treatment are begun too late in some patients, and it is important to realize when the risk of further immunosuppression outstrips the likelihood of longterm benefit. Many Cogan syndrome patients become candidates for cochlear implants.

Behçet Syndrome

Behçet syndrome may affect small, medium, and large vessels in either the venous or the arterial circulation.18 The most typical lesions in Behçet syndrome are mucocutaneous, reflecting the involvement of small blood vessels. The triad of recurrent mouth ulcers, genital ulcers, and eye inflammation is the classic presentation. The criteria for diagnosis of the International Study Group for Behçet Syndrome consist of one required manifestation—recurrent oral ulceration—plus at least two of the following: recurrent genital ulceration, characteristic eye or skin lesions, or a pathergy reaction (see later). However, the spectrum of Behçet syndrome encompasses many manifestations not included in these criteria. In addition, oral ulcers are not invariably the first disease manifestation, so the diagnosis of Behçet syndrome must be considered in the occurrence of disease features otherwise highly characteristic (e.g., bilateral panuveitis). Large-vessel complications of Behçet syndrome may include aneurysms in the pulmonary and systemic arterial systems. Venous complications include thromboses of the deep venous system, vena cava, portohepatic vein, and cerebral sinus. Pathergy—the development of pustules at the sites of sterile needle pricks—is a distinctive feature in many patients with Behçet syndrome, particularly those of Turkish origin. The arthritis of Behçet syndrome is a nondeforming, oligoarticular, asymmetrical arthritis of large joints. GI lesions in Behçet syndrome typically consist of ulcerations of the distal ileum or cecum. Crohn disease (Chapter 132), which can cause genital ulcers as well as GI tract disease, may be particularly difficult to distinguish from Behçet syndrome.

TREATMENT  Low-dose glucocorticoids are effective for intransigent mucocutaneous disease and may have a better side-effect profile than other medications used for this purpose (e.g., thalidomide). Intermittent courses of glucocorticoids during periods of particular mucocutaneous disease activity may be sufficient for patients with mild disease.19 Apremilast, an oral phosphodiesterase 4 inhibitor, (at 30 mg twice daily for 12 weeks) has shown effectiveness in treating oral ulcers in a preliminary study. A9  Colchicine is often considered for mucocutaneous disease, but its efficacy is doubtful. Severe disease in any organ system always requires initial therapy with high doses of prednisone (e.g., 1 mg/kg/day). TNF inhibition with infliximab (5 mg/kg IV every 4 to 6 weeks) or adalimumab (40 mg every other week) is the treatment of choice for patients with the most severe forms of uveitis or meningoencephalitis.

Selected Single-Organ Vasculitides

Single-organ vasculitis is defined as vasculitis within the vessels of any type or size of a single organ, in the absence of any features (e.g., ANCA) suggesting one of the systemic forms of vasculitis.

Cutaneous Leukocytoclastic Angiitis

Cutaneous leukocytoclastic angiitis has also been termed hypersensitivity vasculitis. Cutaneous leukocytoclastic angiitis is the preferred name because no hypersensitivity or allergy is evident in many cases. Histories of exposure to new medications or to infections may be elicited. An immune complex deposition is central to the pathophysiologic process. Although it is occasionally associated with synovitis, other signs of systemic involvement are absent. The skin lesions in cutaneous leukocytoclastic angiitis occur in “crops,” coinciding with some period of elapsed time following exposure to the inciting antigen. The usual time between the exposure and the onset of clinically evident vasculitis is 10 to 14 days. The lesions typically occur first in dependent regions, such as on the lower extremities or buttocks. The rash may be asymptomatic but is usually accompanied by burning or tingling sensations.

TREATMENT  Keys to the management of cutaneous leukocytoclastic angiitis include (1) exclusion of any underlying form of vasculitis that may cause subclinical involvement of other organs and (2) removal of any agent (e.g., a medication) that may have triggered the vasculitis. For patients in whom a precipitant can be identified, elimination of the offending agent usually leads to resolution of the vasculitis within days to weeks. The type, intensity, and duration of therapy for cutaneous leukocytoclastic angiitis are based on the degree of disease severity. Mild cases may be treated simply with leg elevation, H1 antihistamines, or lowdose prednisone. For persistent disease not associated with cutaneous gangrene, colchicine, hydroxychloroquine, or dapsone may be tried. For severe cases, high doses of glucocorticoids are indicated to suppress inflammation quickly and prevent skin ulceration.

Vasculitis of the Central Nervous System

CNS vasculitis20 includes two major categories of disease, one of which is not a true vasculitis. These conditions are primary angiitis of the CNS (PACNS) and reversible cerebral vasoconstriction syndrome (RCVS). The diagnosis and management of these two conditions differ dramatically. The clinical, radiologic, and pathologic characteristics of PACNS and RCVS are shown in Table 254-5. Primary Angiitis of the Central Nervous System

PACNS typically develops in a subacute fashion, with the evolution of multifocal strokes, encephalopathy, headache, and other clinical features over months. Headache is often the first symptom. As the condition progresses, most patients develop lethargy, confusion, and memory loss. Some patients develop multifocal strokes, seizures, evidence of increased intracranial pressure, or myelopathy. The results of routine laboratory tests (e.g., erythrocyte sedimentation rate) are often normal in PACNS. Lumbar puncture demonstrates abnormalities of the cerebrospinal fluid in approximately 80% of cases, usually a modest monocytosis and elevated protein. Lumbar punctures should be performed in all patients in whom the diagnosis of PACNS is considered seriously. Although the findings on lumbar puncture in PACNS patients are nonspecific, a normal lumbar puncture argues against PACNS, and the procedure frequently identifies important PACNS mimickers such as infection or malignancy. Magnetic resonance imaging (MRI) is the critical imaging modality in PACNS. Because of the subacute nature of the disorder, MRI studies reveal multifocal CNS infarctions in most cases. Strokes, hemorrhagic lesions, and mass lesions typically occur in more than one vascular territory. A normal brain MRI argues strongly against the diagnosis of PACNS. Angiography is less helpful in the evaluation of patients with PACNS for two main reasons. First, the sizes of blood vessels involved in PACNS are often too small to be resolved adequately, even by conventional angiography. The false-negative rate of angiography in PACNS is on the order of 35%. Second, the “classic” string-of-beads abnormality on angiography, produced by segmental arterial narrowing alternating with dilations, is nonspecific and can be mimicked perfectly by a host of nonvasculitic conditions (the most common of which is RCVS). No angiographic pattern is pathognomonic for PACNS, and there

CHAPTER 254  The Systemic Vasculitides  

TABLE 254-5 PRIMARY ANGIITIS OF THE CENTRAL NERVOUS SYSTEM (PACNS) VERSUS REVERSIBLE CEREBRAL VASOCONSTRICTION SYNDROME (RCVS) PACNS

RCVS

Female-to-male ratio

1 : 1

2-3 : 1

Onset

Subacute (weeks to months)

Sudden (seconds to minutes)

Headache

Insidious, dull

Thunderclap

Typical lumbar puncture findings

Abnormal in 50-80%: lymphocytic pleocytosis; elevated protein

Normal

Typical MRI findings

Multifocal subacute infarctions

Normal Watershed infarcts in minority

Typical angiogram findings

Normal in up to 40% of cases Abnormal angiographic features when present cannot be distinguished from RCVS

Multifocal stenoses/ dilatations

Utility of brain biopsy

Reasonable sensitivity in appropriately selected patients Important for excluding disease mimickers

Little to no role Helpful if confusing clinical situation confounds differentiation from PACNS or PACNS mimickers

MRI = magnetic resonance imaging.

is a significant tendency to overdiagnose “vasculitis” on angiographic grounds alone. A normal brain MRI in the setting of an abnormal angiogram suggests RCVS, not PACNS. When employed in appropriately selected patients whose history and radiologic studies suggest PACNS, brain biopsy is associated with reasonable positive and negative predictive values and frequently identifies important PACNS mimickers.

1757

TABLE 254-6 MAJOR DISEASE CATEGORIES IN THE DIFFERENTIAL DIAGNOSIS OF VASCULITIDES Other forms of vasculitis Simultaneous occurrence of common medical problems in the same patient Infections Bacterial, viral, mycobacterial, fungal Occlusive processes Hypercoagulable states Livedoid vasculopathy (atrophie blanche) Atheroembolic disease Malignant neoplasms Lymphoma (including lymphomatoid granulomatosis) Castleman disease Amyloidosis Paraproteinemias Connective tissue disorders Systemic lupus erythematosus, mixed connective tissue disease Systemic sclerosis Rheumatoid arthritis Miscellaneous Atrial myxoma Calciphylaxis Fibromuscular dysplasia Neutrophilic dermatoses Pyoderma gangrenosum Sarcoidosis Reversible cerebral vasoconstriction syndrome

TREATMENT  Several approaches to the treatment of RCVS are reasonable. First, one may opt for watchful waiting. It is not clear that immunosuppression is either necessary or helpful. Moreover, attempts to treat vasospasm with calcium-channel blockers may lead to a vascular steal phenomenon, potentially causing harm. Second, because it is frequently difficult to do nothing for a patient with possibly serious CNS disease, calcium-channel blockers (e.g., nifedipine 30 mg three times daily) may be tried. Third, because of the frequent diagnostic uncertainty at the time of presentation, some clinicians opt to treat empirically with glucocorticoids (prednisone 1 mg/kg/day) for 1 month, followed by a taper over several weeks. Fourth, combinations of calcium-channel blockers and glucocorticoids are also reasonable. Cytotoxic therapy is not indicated in RCVS.

TREATMENT  Prednisone and cyclophosphamide are appropriate for treatment of patients who have abnormal findings on brain biopsy. Treatment courses of 6 to 12 months are recommended.

Reversible Cerebral Vasoconstriction Syndrome

RCVS is probably far more common than PACNS. Overtreatment of patients with RCVS who are misdiagnosed as having PACNS leads to substantial morbidity. Eighty percent of patients with RCVS are women. A careful history is the most important part of the evaluation. In contrast to the subacute course that typifies PACNS, RCVS usually begins in a more dramatic fashion with a “thunderclap” headache.21 Compared with PACNS, the neurologic signs are less severe in RCVS (e.g., encephalopathy is less common). RCVS frequently occurs in the setting of precipitants associated with vasospasm, such as in the postpartum setting or following the use of vasoactive agents such as nasal decongestants and recreational drugs. The lumbar puncture is usually normal in RCVS, and brain MRI usually does not show multifocal CNS infarctions, with the exception of watershed infarctions mentioned earlier. The typical angiographic findings in RCVS— vascular narrowing and beading—are generally indistinguishable from those of PACNS and conditions that mimic PACNS. Multifocal vascular narrowing is particularly characteristic of RCVS. The most distinctive angiographic feature of RCVS is that the abnormalities are completely reversible, usually within 4 to 8 weeks. These abnormalities in RCVS are caused by vasospasm rather than true vasculitis. In the evaluation of patients with potential RCVS, a diagnostic strategy that can clinch the diagnosis is a follow-up angiogram 4 to 8 weeks after the first. Angiographic abnormalities due to RCVS will resolve in this interval.



DIAGNOSIS

Differential Diagnosis

The major categories of diseases that can mimic vasculitis are displayed in Table 254-6. Certain features of a patient’s case should raise the diagnostic suspicion for vasculitis. First, most cases of vasculitis do not begin suddenly but rather unfold subacutely during weeks or months. Second, pain is usually a prominent feature of vasculitis, resulting from arthritis or arthralgias, myalgias, headaches, neuropathy, testicular infarction, digital ischemia, sinusitis, otalgia, back pain (caused by aortic inflammation), postprandial abdominal pain (caused by mesenteric vasculitis), or other disease manifestations. Third, signs of inflammation such as fever, rash, weight loss, and elevated acute phase reactants are highly characteristic. Finally, multiorgan system involvement is the rule in vasculitis. The diagnosis of vasculitis should be established through biopsy of an involved organ whenever possible. Diagnoses based on angiography alone have many potential pitfalls, as discussed in the sections on PACNS and RCVS. Angiographic findings that are “consistent with vasculitis” must be interpreted in the proper context. A diverse array of other diseases, ranging from atherosclerosis to vasospasm to pheochromocytoma, may mimic the angiographic appearance of vasculitis. Systemic vasculitis can also be mimicked by two or more common medical problems or treatment complications occurring simultaneously in the same patient. Finally, high on the differential diagnosis of any individual form of vasculitis are other forms of vasculitis. For example, digital ischemia and splinter hemorrhages may be secondary to idiopathic polyarteritis nodosa. They may also be caused by polyarteritis nodosa associated with HBV infection, GPA, EGPA, microscopic polyangiitis, cryoglobulinemia, Buerger disease, or some other form of vasculitis. Because the appropriate interventions for these conditions vary widely, careful distinction among these potential etiologies is essential.

TREATMENT  Current treatment approaches to specific vasculitides are described under their “Clinical Manifestations” for each disease. General points regarding therapy are addressed here. The intensity of treatment in patients with vasculitis must be guided by the degree of disease activity. Specifically, the treatment of vasculitis should be predicated not only on abnormal laboratory test results but also on clear evidence of active disease. In addition, the intensity of treatment must be adapted to the type of vasculitis. Whereas giant cell arteritis responds to high doses of glucocorticoids in most cases, for example, GPA nearly always requires an additional agent (rituximab, cyclophosphamide, or methotrexate) for disease control. In contrast, despite the dramatic fashion in which they sometimes present, most cases of IgA vasculitis and cutaneous leukocytoclastic angiitis require no immunosuppressive treatment at all. Conventional therapies such as glucocorticoids, immunomodulating agents, and cytotoxic drugs induce remissions and control vasculitis in most cases. Moreover, in some cases—a variable percentage, depending on the type of vasculitis—the disease is curable. Unfortunately, the treatments of vasculitis have enormous potential for toxicity. Regular monitoring of patients’ bone marrow, renal, and hepatic function is essential to avoid treatment-induced toxicity. Prophylaxis against opportunistic infections, particularly Pneumocystis pneumonia (Chapter 321), is an important part of many vasculitis treatment regimens. During the tapering of immunosuppressive medications, disease flares are common in many forms of vasculitis. A common error is treating patients with high doses of immunosuppressive agents for too long. The most appropriate use of medications such as cyclophosphamide and glucocorticoids is to induce remission as quickly as possible with early, aggressive treatment regimens, and then to convert patients to safer treatments for the maintenance of remission. Rituximab is replacing cyclophosphamide as the drug of choice for some forms of vasculitis, particularly AAV. Patients with AAV who demonstrate a tendency to flare are often retreated with rituximab (500 mg or 1 g) every 4 to 6 months, at least until lengthy periods of disease control are established.



PROGNOSIS

Assuming that the diagnosis is made before the patient has become catastrophically ill, the prognosis in systemic vasculitis is determined largely by the answers to four questions: 1. Was the diagnosis established before the occurrence of major irreversible organ damage? 2. Was aggressive (but appropriately dosed) treatment begun in a timely fashion? 3. Was there careful monitoring during treatment, and were specific steps taken to avoid drug-induced toxicity (e.g., opportunistic infection)? 4. Were the potentially toxic medications that induced remission stopped at an appropriate juncture and replaced with less dangerous medications (or was treatment stopped altogether)? For most forms of vasculitis, the factors that determine long-term drug-free remissions remain poorly understood. The likelihood of achieving sustained remissions after discontinuation of all medications (or cures) varies according to the specific type of vasculitis.  

FUTURE DIRECTIONS

Compelling laboratory and naturally occurring animal models of disease, combined with the known associations among HBV, HCV, and vasculitis in humans, suggest that additional links between infection and systemic vasculitis may be established in the future. Important strides have been made in the description of cytokine and chemokine pathways that are operative in vascular inflammation, but relevant anticytokine interventions remain to be defined for clinical therapies. B-cell depletion is emerging rapidly as the treatment of choice for some forms of severe vasculitis. IL-6 inhibition strategies also now play important roles in the large-vessel vasculitides. Additional studies are required to define the full spectrum of clinical utility of these and other biologic agents.

  Grade A References A1. Stone JH, Tuckwell K, Dimonaco S, et al. Trial of tocilizumab in giant-cell arteritis. N Engl J Med. 2017;377:317-328. A2. Nakaoka Y, Isobe M, Takei S, et al. Efficacy and safety of tocilizumab in patients with refractory Takayasu arteritis: results from a randomised, double-blind, placebo-controlled, phase 3 trial in Japan (the TAKT study). Ann Rheum Dis. 2018;77:348-354.

A3. Hamada H, Suzuki H, Onouchi Y, et al. Efficacy of primary treatment with immunoglobulin plus ciclosporin for prevention of coronary artery abnormalities in patients with Kawasaki disease predicted to be at increased risk of non-response to intravenous immunoglobulin (KAICA): a randomised controlled, open-label, blinded-endpoints, phase 3 trial. Lancet. 2019;393:1128-1137. A4. Wardle AJ, Connolly GM, Seager MJ, et al. Corticosteroids for the treatment of Kawasaki disease in children. Cochrane Database Syst Rev. 2017;1:CD011188. A5. Stone JH, Merkel PA, Spiera R, et al. Rituximab versus cyclophosphamide for remission induction in ANCA-associated vasculitis. N Engl J Med. 2010;363:221-232. A6. Specks U, Merkel PA, Seo P, et al. Efficacy of remission-induction regimens for ANCA-associated vasculitis. N Engl J Med. 2013;369:417-427. A7. Guillevin L, Pagnoux C, Karras A, et al. Rituximab versus azathioprine for maintenance in ANCAassociated vasculitis. N Engl J Med. 2014;371:1771-1780. A8. Wechsler ME, Akuthota P, Jayne D, et al. Mepolizumab or placebo for eosinophilic granulomatosis with polyangiitis. N Engl J Med. 2017;376:1921-1932. A9. Hatemi G, Melikoglu M, Tunc R, et al. Apremilast for Behçet’s syndrome—a phase 2, placebocontrolled study. N Engl J Med. 2015;372:1510-1518.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 254  The Systemic Vasculitides  

GENERAL REFERENCES 1. Kim ESH, Beckman J. Takayasu arteritis: challenges in diagnosis and management. Heart. 2018;104:558-565. 2. Barra L, Yang G, Pagnoux C, et al. Non-glucocorticoid drugs for the treatment of Takayasu’s arteritis: a systematic review and meta-analysis. Autoimmun Rev. 2018;17:683-693. 3. Comarmond C, Biard L, Lambert M, et al. Long-term outcomes and prognostic factors of complications in Takayasu arteritis: a multicenter study of 318 patients. Circulation. 2017;136: 1114-1122. 4. van der Geest KSM, Sandovici M, van Sleen Y, et al. What is the current evidence for disease subsets in giant cell arteritis? Arthritis Rheumatol. 2018;70:1366-1376. 5. Forbess L, Bannykh S. Polyarteritis nodosa. Rheum Dis Clin North Am. 2015;41:33-46. 6. Karadag O, Jayne DJ. Polyarteritis nodosa revisited: a review of historical approaches, subphenotypes and a research agenda. Clin Exp Rheumatol. 2018;36 Suppl 111:135-142. 7. Sosa T, Brower L, Divanovic A. Diagnosis and management of Kawasaki disease. JAMA Pediatr. 2019. [Epub ahead of print.] 8. Sundel RP. Kawasaki disease. Rheum Dis Clin North Am. 2015;41:63-73. 9. Liu YC, Lin MT, Wang JK, et al. State-of-the-art acute phase management of Kawasaki disease after 2017 scientific statement from the American Heart Association. Pediatr Neonatol. 2018;59: 543-552. 10. Lamprecht P, Kerstein A, Klapa S, et al. Pathogenetic and clinical aspects of anti-neutrophil cytoplasmic autoantibody-associated vasculitides. Front Immunol. 2018;9:1-10. 11. Lynch JP 3rd, Derhovanessian A, Tazelaar H, et al. Granulomatosis with polyangiitis (Wegener’s granulomatosis): evolving concepts in treatment. Semin Respir Crit Care Med. 2018;39:434-458.

1758.e1

12. Cornec D, Kabat BF, Mills JR, et al. Pharmacokinetics of rituximab and clinical outcomes in patients with anti-neutrophil cytoplasmic antibody associated vasculitis. Rheumatology (Oxford). 2018;57: 639-650. 13. Wallace ZS, Wallwork R, Zhang Y, et al. Improved survival with renal transplantation for end-stage renal disease due to granulomatosis with polyangiitis: data from the United States Renal Data System. Ann Rheum Dis. 2018;77:1333-1338. 14. Raffray L, Guillevin L. Treatment of eosinophilic granulomatosis with polyangiitis: a review. Drugs. 2018;78:809-821. 15. Kolopp-Sarda MN, Miossec P. Cryoglobulins: an update on detection, mechanisms and clinical contribution. Autoimmun Rev. 2018;17:457-464. 16. Roccatello D, Sciascia S, Rossi D, et al. The challenge of treating hepatitis C virus-associated cryoglobulinemic vasculitis in the era of anti-CD20 monoclonal antibodies and direct antiviral agents. Oncotarget. 2017;8:41764-41777. 17. Durtette C, Hachulla E, Resche-Rigon M, et al. Cogan syndrome: characteristics, outcome and treatment in a French nationwide retrospective study and literature review. Autoimmun Rev. 2017;16: 1219-1223. 18. Greco A, De Virgilio A, Ralli M, et al. Behçet’s disease: new insights into pathophysiology, clinical features and treatment options. Autoimmun Rev. 2018;17:567-575. 19. Hatemi G, Christensen R, Bang D, et al. 2018 update of the EULAR recommendations for the management of Behcet’s syndrome. Ann Rheum Dis. 2018;77:808-818. 20. Byram K, Hajj-Ali RA, Calabrese L. CNS vasculitis: an approach to differential diagnosis and management. Curr Rheumatol Rep. 2018;20:1-7. 21. Cappelen-Smith C, Calic Z, Cordato D. Reversible cerebral vasoconstriction syndrome: recognition and treatment. Curr Treat Options Neurol. 2017;19:1-15.

1758.e2

CHAPTER 254  The Systemic Vasculitides  

REVIEW QUESTIONS 1. Immunosuppressive therapy is known to be futile for which disease associated with vascular inflammation? A . Buerger disease (thromboangiitis obliterans) B. Granulomatosis with polyangiitis (formerly Wegener granulomatosis) C. IgA vasculitis (Henoch-Schönlein purpura) D. Giant cell arteritis E. Mixed cryoglobulinemia Answer: A  The only effective intervention known for Buerger disease is smoking cessation. Neither immunosuppression nor anticoagulation is of any demonstrable benefit in relieving the severe digital ischemia associated with this disease. 2. Which form of systemic vasculitis is most likely to present with an orbital pseudotumor? A . Takayasu arteritis B. Granulomatosis with polyangiitis (formerly Wegener granulomatosis) C. Microscopic polyangiitis D. Polyarteritis nodosa E. Behçet syndrome Answer: B  Two common ocular manifestations of granulomatosis with polyangiitis are scleritis and orbital pseudotumor. The latter can lead to proptosis and vision loss if blood circulation to the eye is compromised by the retrobulbar mass. 3. A 24-year-old man from Lebanon presents with multiple aphthous ulcers, bilateral ocular erythema, tender erythematous nodules over his anterior legs, and a deep venous thrombosis. An ophthalmologic examination reveals severe bilateral uveitis affecting both the anterior and posterior compartments of the eye. In addition to anticoagulation and the institution of glucocorticoid therapy, what is the most appropriate approach to treating his disease at this time? A . Intravenous cyclophosphamide B. Azathioprine C. Intravenous immunoglobulin D. Tumor necrosis factor inhibition E. B-cell depletion Answer: D  Tumor necrosis factor inhibition is highly effective in Behçet disease and is employed in the presence of severe disease manifestations such as posterior uveitis or central nervous system (CNS) disease. 4. A 32-year-old woman presents with the rapid onset of the worst headache of her life 1 day after giving birth to a healthy baby boy. The delivery followed an uneventful pregnancy. The patient has no history of migraine headaches, and her past medical history is unremarkable. The patient’s blood pressure is 140/90 mm Hg, and although her neurologic examination is nonfocal, the severity of the headache prompts a computed

tomographic (CT) scan of the brain. This is negative for an intracranial bleed, and the headache resolves after approximately 24 hours, during which time it was profoundly disabling. Approximately 18 hours later, the headache returns, again rising to a crescendo within several minutes of onset. A repeat head CT is negative and a lumbar puncture is also performed, revealing a normal opening pressure, 1 white blood cell/mL (100% lymphocytes), and a cerebrospinal fluid protein of 40 mg/dL (normal 30 to 50 mg/dL). The headache improves following round-the-clock narcotic treatment but then worsens again on the third postpartum day. A magnetic resonance imaging study is negative for hemorrhage, cerebral infarctions, and masses, but a four-vessel cerebral angiogram shows alternating areas of vascular narrowing and dilation (beading) in multiple vascular distributions. What is the most appropriate intervention for this patient now? A . Tighter blood pressure control, with transfer to the intensive care unit for sodium nitroprusside therapy B. Addition of a baby aspirin to her anticoagulation with low-molecularweight heparin C. Pulse methylprednisolone (1000 mg/day for 3 days) followed by prednisone 60 mg/day and cyclophosphamide 100 mg/day D. Nifedipine 30 mg three times daily E. Proceed to brain biopsy Answer: D  Nifedipine. This patient has a history that is classic for reversible cerebral vasoconstriction syndrome (RCVS), an entity that mimics CNS vasculitis but is not an inflammatory condition. The features suggesting RCVS in this patient are the presence of a “thunderclap” headache, the nonfocal neurological examination, the normal cross-sectional imaging and lumbar puncture, and the diffuse beading on angiography. This beading represents vasospasm, which is the essential pathophysiology of RCVS. Calcium-channel blockers are often used in therapy, but prolonged courses of immunosuppression are not indicated. Brain biopsy can be diagnostic in primary angiitis of the CNS, but would not be indicated in a patient like this who has typical clinical and radiographic features of RCVS. 5. A 4-year-old boy presents with 3 days of a febrile illness with temperatures to nearly 105° Fahrenheit. Physical examination revealed an erythematous “strawberry” tongue, swollen lips, and a pink macular rash on the trunk. There is also tender cervical and axillary lymphadenopathy and diffuse swelling of the distal upper and lower extremities. Which treatment is indicated immediately to prevent the occurrence of coronary aneurysms? A . Intravenous immunoglobulin B. High-dose aspirin therapy C. Pulse methylprednisolone D. Plasma exchange E. Anticoagulation with heparin Answer: A  This child has Kawasaki disease, and the most feared complication of this disorder is formation of coronary aneurysms, which can lead to myocardial infarction. If administered in a timely manner, intravenous immunoglobulin prevents the occurrence of such aneurysms.

1758

CHAPTER 255  Giant Cell Arteritis and Polymyalgia Rheumatica  

255  GIANT CELL ARTERITIS AND POLYMYALGIA RHEUMATICA ROBERT SPIERA



DEFINITION

Polymyalgia rheumatica (PMR) and temporal arteritis, also called giant cell arteritis (GCA), are companion systemic inflammatory disorders of unknown etiology that represent a spectrum from severe proximal aches and pains and constitutional symptoms to an occlusive granulomatous vasculitis of medium and large vessels that can lead to permanent blindness or other organ and tissue damage. These disorders occur primarily in patients older than 50 years, in women more than in men; they are propagated by antigen-driven, cellmediated (TH1) immune mechanisms that may be associated with specific genetic markers, and they are highly responsive to corticosteroids.  

EPIDEMIOLOGY

In the United States, the average annual incidence of PMR is 52.5 per 100,000 patients aged 50 years and older and increases with age. The prevalence is about 0.5 to 0.7%. Internationally, the frequency varies, with the highest rates occurring in the Scandinavian countries.1,2 The incidence and prevalence of GCA are approximately one third those of PMR.  

PATHOBIOLOGY

The etiology of PMR and GCA are unknown, but both demonstrate familial aggregation and have a genetic association with human leukocyte antigen (HLA)-DR4 and a demonstrated sequence polymorphism encoded within the hypervariable region of the HLA-DRβ1*04 gene. Other genetic associations have been suggested, including polymorphisms that may be seen in increased frequency in patients with the disease. Infectious triggers of GCA have been postulated. Varicella-zoster virus antigen has been detected in temporal artery biopsies of patients with GCA.3 Abundant bacterial and viral DNA has been demonstrated in the arterial wall of patients with GCA using advanced DNA imaging techniques. Disease in genetically predisposed patients may also be triggered by endogenous antigens such as elastin, and the inflammatory manifestations are directed by specific patterns of cell-mediated, TH1-associated cytokines. The cytokine production by the mononuclear cells in the involved tissues appears to influence the clinical phenotype. In PMR, mononuclear cell inflammation can be found not only in the proximal joints, such as the shoulders, but also in the surrounding tendons, bursae, and soft tissues consistent with enthesitis. Although muscle pains may be present, no muscle inflammation is found.  

CLINICAL MANIFESTATIONS

PMR and GCA are systemic inflammatory disorders that occur primarily in patients older than 50 years, in women more than in men (2:1), and in whites.

CHAPTER 255  Giant Cell Arteritis and Polymyalgia Rheumatica  

ABSTRACT

Giant cell arteritis (GCA) and polymyalgia rheumatica (PMR) are related inflammatory diseases affecting older individuals. Both are characterized by evidence of systemic inflammation including elevations of acute phase reactants such as CRP and erythrocyte sedimentation rate. PMR is a clinical syndrome characterized by symmetric pain and stiffness in the shoulder and thigh girdle, with prominent morning stiffness, and exquisite responsiveness to low dose corticosteroid therapy. GCA is a large vessel vasculitis in which polymyalgic symptoms can be present in the majority of patients but characterized more specifically by large vessel inflammation that can result in ischemic complications including headaches, jaw claudication, and most concerning, vision loss. Timely diagnosis of GCA is essential, as vision loss, once incurred, is generally irreversible, whereas therapy with corticosteroids is protective against that dreaded complication, and typically affords prompt resolution of symptoms. Diagnosis of GCA is generally supported by temporal artery biopsy, but large vessel imaging including ultrasound, MRI/MRA, and PET scanning are emerging as useful diagnostic modalities. Therapy of both disorders is primarily with corticosteroids, which are effective in low doses in PMR, and in modest to high doses in GCA. The morbidity of corticosteroids particularly with prolonged use in this vulnerable population is substantial leading to the exploration for other potential steroid sparing modalities, particularly as the underlying pathophysiology of these disorders has become better understood. Most recently, tocilizumab, an anti-interleukin 6 receptor antibody has been shown to be effective in GCA and is approved for the treatment of this disorder.

KEYWORDS

giant cell arteritis polymyalgia rheumatica pathophysiology therapy complication temporal arteritis

1758.e3

1758.e4

CHAPTER 255  Giant Cell Arteritis and Polymyalgia Rheumatica  

GCA tissue contains the T-lymphocyte products interferon-γ and interleukin (IL)-2 and the macrophage products IL-1β, IL-6, and transforming growth factor-β. In PMR vascular tissue, transcripts are found for transforming growth factor-β, IL-1, and IL-2 but not for interferon-γ. Patients with GCA who present with fever of unknown origin and who do not have ischemic symptoms, such as visual loss, have low interferon-γ levels. Arteries that express high interferon-γ levels typically have multinucleated giant cells present; these cells remove debris and secrete cytokines that stimulate intimal hyperplasia and lead to angiogenesis. IL-17-producing TH17 cells have also been found in involved vascular tissue and peripheral blood of patients with untreated GCA but disappear rapidly with institution of corticosteroid therapy, whereas TH1 cells are more persistent, speaking to the possibility that more than one antigenic trigger may be involved. The adventitia is considered the immunologic center in the pathogenesis of GCA where toll-like receptors (TLRs) are highly expressed on dendritic cells and resident tissue macrophages. Macrophages and T lymphocytes enter the vessel wall through the vasa vasorum with the aid of adhesion molecules and come into contact with an inciting antigen. Here, it is likely that clonal proliferation of CD4+ T cells is triggered by the presentation of unknown antigens by antigen-presenting cells. The activated CD4 cells produce interferon-γ that attracts macrophages to the arterial wall. Some of these macrophages fuse at the intima-media to form multinucleated giant cells. These

cells produce vascular endothelial growth factor, which triggers neovascularization, both at the intima-media junction and at the level of the vasa vasorum, sprouting from the adventitia to the media. The subsequent immunologic events lead to a characteristic topography of mononuclear cells throughout the vessel wall. Products of the giant cells and macrophages at the intimamedia junction include collagenase and nitric oxide, both of which probably contribute to tissue damage. The pathologic impact of cytokines leads not only to the characteristic medial damage but also to significant intimal hyperplasia that eventually, if it is not treated, may cause luminal narrowing and tissue ischemia. In GCA, a transmural (involving all layers of the vessel) inflammatory infiltrate, comprised of predominantly mononuclear cells and commonly with giant cells, is found in the superficial temporal arteries as well as in other large and medium-sized arteries. In elderly patients, fragmentation of the internal elastica is characteristic and helps differentiate this vascular lesion from that of atherosclerosis. Often, macrophages containing fragments of elastic tissue are found at the intima-media junction, the histologic center of the inflammatory process. As mentioned earlier, immunochemical techniques demonstrate differing patterns of cells and their proinflammatory and profibrotic products in the adventitia, media, and intima. Intimal proliferation may be prominent and lead to luminal narrowing. Fibrinoid necrosis, a common histologic feature in polyarteritis nodosa, is not seen in GCA.

CHAPTER 255  Giant Cell Arteritis and Polymyalgia Rheumatica  

PMR and GCA are particularly uncommon in African Americans. Shared characteristics of the two disorders include significant cytokine-driven constitutional symptoms, such as fever, fatigue, and weight loss, as well as a markedly elevated erythrocyte sedimentation rate (ESR), anemia, and thrombocytosis. The musculoskeletal hallmark of PMR is proximal, severe, and symmetrical morning and even day-long stiffness, soreness, and pain in the shoulder, neck, and pelvic girdles. Fifty percent of patients with GCA share this characteristic proximal pain syndrome. Carpal tunnel syndrome and hand and knee synovitis may be seen in patients with PMR, but the overall presentation remains predominantly proximal, as opposed to rheumatoid arthritis, in which distal synovitis dominates. Whereas patients with PMR may appear to have proximal muscle weakness, this is invariably due to pain and not muscle inflammation (Table 255-1). Magnetic resonance imaging (MRI) and ultrasound studies4,5 in patients with PMR have confirmed the presence of inflammation of extra-articular synovial structures, in particular subacromial and subdeltoid bursae in the shoulders. Specific signs and symptoms of GCA are best appreciated in their anatomic and physiologic contexts. GCA preferentially affects certain blood vessels, including the branches of the external carotid artery, the ophthalmic artery and particularly its posterior ciliary branches, and the large arteries that arise from the aortic arch and abdominal aorta. Headache and scalp pain are probably the most frequent symptoms, occurring in 50 to 75% of patients. Headache is often the first manifestation of GCA and is described as boring, severe, and constant, unresponsive to simple pain medications and persisting through the night. Classically, patients complain of persistent and prominent temporal headaches, but occipital pains can also occur. Ear, pinna, or parotid gland pain may occur secondary to involvement of the posterior auricular artery. Jaw claudication and pain due to masseter muscle ischemia on chewing occur in 50% of patients. Lingual and maxillary artery involvement can lead to jaw or tongue pain on chewing or talking. The superficial temporal artery may become tortuous, prominent, nodular, or tender, but these findings are not invariable, and an abnormal temporal artery may be found on biopsy in vessels that appear normal. It is important to note that a dry, nonproductive cough can be a feature of the disease because this often may direct the clinician away from considering GCA and more toward consideration of an infectious or neoplastic respiratory cause of the symptoms. Rarely, mononeuritis multiplex or sensorineural hearing loss can occur but should lead the clinician to consider other possible vasculitides such as antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis or polyarteritis nodosa.

TABLE 255-1 GIANT CELL ARTERITIS: CLINICAL FEATURES INFLAMMATORY Polymyalgia rheumatica: constitutional symptoms Fever Weight loss Fatigue Laboratory abnormalities Hematologic: anemia, thrombocytosis Elevated sedimentation rate, C-reactive protein ISCHEMIC Ocular Diplopia Amaurosis fugax Fixed vision loss Complete blindness Cranial symptoms Headache Jaw claudication Scalp tenderness Scalp or lingual necrosis (rare) Cerebrovascular accidents Large vessel disease Leg or arm claudication Diminished pulses, blood pressure asymmetry Aortic aneurysms LATE COMPLICATIONS Aortic aneurysms Thoracic aorta Abdominal aorta Corticosteroid complications Osteoporosis Fractures Cataracts

1759

Fixed or intermittent symptoms related to vasculitic involvement of the ophthalmic arteries and its branches are the most dreaded in this illness and demand immediate therapeutic intervention. These symptoms are related to vascular narrowing due to both active inflammation and endothelial injury–mediated vasospasm. Decreased vision secondary to arteritis is the most common serious consequence of GCA, occurring in 20 to 50% of patients who present to ophthalmologists. It is the presenting symptom in 60% of patients with GCA who develop visual loss. A careful history of most patients who present with “sudden” visual loss reveals that preceding headache, constitutional symptoms, and PMR occurred in approximately 40% of patients. Even the evolution of the visual loss is often staggered, with amaurosis fugax in 10% and a partial field defect progressing to complete blindness over days. If GCA remains untreated, the second eye may become involved within 1 to 2 weeks. The posterior ciliary arteries are the most frequently involved; thus, anterior ischemic optic neuropathy is the most common lesion, which can be easily defined by an ophthalmologist. Occlusion of the central retinal artery and its branches is uncommon; thus, exudates, hemorrhages, and frank vasculitis are infrequent. Five percent of patients with GCA may present with diplopia or ptosis, which may precede visual loss. The final visual abnormality can be a composite of many ischemic events occurring together in the optic nerve, the extraocular muscles, the chiasm, and the brain itself. Because GCA primarily involves arteries that contain elastica and the elastic lamina is lost as vessels pierce the dura, intracerebral lesions such as strokes are uncommon but not unheard of. Large artery involvement most commonly presents as arm or leg claudication; rarer manifestations are stroke, subclavian steal syndrome, intestinal infarction, and symptomatic aortic aneurysm. Thus, a subclinical arteritis can exist and demands long-term monitoring. There is an emerging appreciation that some older patients classified as having GCA can present with large vessel disease resembling Takayasu arteritis clinically, with a paucity of cranial ischemic symptoms but often the presence of PMR-like symptoms. Conversely, in patients presenting with typical GCA with cranial symptoms and a positive temporal artery biopsy, large vessel disease with aortic wall thickening is markedly more frequent even early in the disease course. Steroid-treated PMR and GCA are self-limited illnesses lasting 1 to 2 years in most patients. However, a subgroup of patients with both disorders can have active inflammatory disease as manifested by persistent symptoms and blood test signs of active inflammation for 7 to 10 years. Of note is the fact that thoracic aneurysms with giant cells in the tissue can develop as long as 15 years after the diagnosis, successful treatment, and discontinuation of steroids. Indeed, the incidence of thoracic and aortic aneurysms is markedly higher in patients with a prior history of presumably successfully treated GCA than in age-matched control subjects. In studies of repaired aortic aneurysms, pathologic findings consistent with GCA have been found in approximately 2 to 4% of specimens from individuals without previously recognized or suspected arteritis. In most studies, survival rates for patients with PMR and GCA are similar to those of unaffected persons of the same age. However, one study did show that survival was decreased in a group of patients with GCA who had permanent visual loss and required more than 10 mg of prednisone per day at 6 months. This probably supports the experience that the morbidity and mortality are caused by steroid-related treatment complications in this high-risk, elderly group of patients possessing many comorbid conditions.  

DIAGNOSIS

The diagnoses of PMR and GCA are based on clinical findings, with supporting but not diagnostic aid obtained from laboratory tests and temporal artery biopsy (Fig. 255-1).6,7 No physician should await an abnormal finding on temporal artery biopsy or demand the presence of an elevated ESR before making the definitive diagnosis of GCA in the setting of a characteristic clinical picture. That said, the laboratory hallmark of PMR and GCA is an elevation in IL-6-stimulated acute phase reactants such as the ESR and C-reactive protein. The ESR is usually in excess of 50 mm/hour and may exceed 100 mm/hour. An ESR in the low 20s or 30s, however, does not exclude a diagnosis of PMR or GCA if other characteristic clinical features are present and especially if the patient is already taking steroids. Normocytic, normochromic anemia and thrombocytosis occur in approximately 50% of patients with both disorders and are excellent guides to the state of inflammation. In both PMR and GCA, the frequency of rheumatoid factor, antinuclear antibody, ANCA, monoclonal proteins, and cryoglobulins is not higher than in age-matched control subjects, and complement is not reduced. Alkaline phosphatase activity may be elevated in one third of patients, primarily those with GCA. Although these tests are not indicated in PMR

1760

CHAPTER 255  Giant Cell Arteritis and Polymyalgia Rheumatica  

Elderly patient, PMR symptoms Exclude alternative explanations: infection, malignancy, connective tissue disease Elevated sedimentation rate/CRP?

Yes

No Consider diagnostic trial corticosteroids: prednisone 15-20 mg qd

Prednisone 15-20 mg qd

Dramatic clinical response?

No Alternatives: 1. Consider GCA, and do temporal artery biopsy and/or large vessel imaging

Dramatic clinical response?

Yes

Yes

Treat as PMR: taper corticosteroids to lowest dose that adequately controls symptoms

No Pursue alternative diagnosis such as: infection, malignancy, connective tissue disorder, hypothyroidism

2. Consider • Malignancy • Infection • Connective tissue disorder • Hypothyroidism FIGURE 255-1.  Diagnostic algorithm for polymyalgia rheumatica (PMR). CRP = C-reactive protein; GCA = giant cell arteritis.

and GCA, muscle enzymes and electromyography are normal, and muscle biopsy shows type II fiber atrophy but no inflammation.

Superficial Temporal Artery Assessment

Temporal artery tenderness, nodularity, and diminished pulsation are typical findings on physical examination in a patient with GCA. Color duplex ultrasonography has been used as an adjunctive noninvasive diagnostic tool in GCA. A hypoechoic halo around the superficial temporal artery has been reported in 73% of patients with biopsy-proven GCA. The halo, representing edema in the arterial wall, was observed bilaterally in a significant subset of patients and disappeared in a mean of 16 days after the initiation of steroids in one study. The presence of the halo in this study had a sensitivity of 73% and was 100% specific for GCA. Other groups have been unable to replicate this experience, however, finding Doppler ultrasonography to be no more sensitive or specific than physical examination in patients thought to have the disease. Findings of stenosis or occlusion of temporal arteries by Doppler ultrasound have also been recognized as being modestly sensitive and specific for the diagnosis of GCA in some studies. Operator dependency remains a challenge to the more widespread use of this modality diagnostically. [18F] Fluorodeoxyglucose–positron emission tomography (PET) may be helpful in identifying large vessel inflammation suggestive of GCA, but it is not helpful in assessing the temporal arteries themselves, given their relatively small size and high background uptake in that area. Conventional angiography is rarely used in the diagnosis of GCA. Some studies have suggested that MRI/magnetic resonance angiography (MRA) may be a helpful noninvasive diagnostic modality. Superficial cranial arteries can be visualized, and mural inflammatory changes and luminal narrowing can be identified. Large vessel involvement can also be assessed. Studies have suggested sensitivities and specificities of MRI/MRA similar to those of biopsy in the diagnosis of GCA. Nevertheless, temporal artery biopsy remains the diagnostic “gold standard” in GCA, and given the relatively easy accessibility of the artery and potentially significant morbidity of therapy in GCA, histologic confirmation is favored in most cases. Although temporal artery biopsy continues to be an important diagnostic test for the presence of GCA, a few caveats must be stated. First, in a patient in whom the clinical diagnosis is likely, treatment with steroids should be instituted immediately without waiting for the biopsy results. Second, because

of the skipped nature of the pathologic inflammatory lesions in the vessel wall, as many as 20 to 30% of biopsy specimens may be normal despite an overwhelming diagnostic likelihood of GCA. Because the biopsy is helpful in confirming the diagnosis of GCA, in which high doses of steroids are used, the following guidelines are given. Patients with pure PMR and no GCA signs or symptoms do not need a biopsy. However, because 10% of these patients may develop such clinical manifestations of GCA within the next year, they should be told to report such symptoms immediately. When GCA is likely, an outpatient biopsy should be performed on the symptomatic side of the head, preferably including inflamed areas with tenderness or nodularity and incorporating 2 to 3 cm of vessel. Multiple sections should be requested because of the segmental nature of the disease process. Some rheumatologists routinely request bilateral biopsies, which may increase the likelihood of obtaining an abnormal finding by up to 5%, whereas others perform a contralateral biopsy if the first specimen is normal. Diagnostic biopsy findings continue to be present for as long as 2 to 4 weeks after the clinical diagnosis is made and steroid treatment instituted, and may be seen even months into treatment.

Differential Diagnosis

The systemic nature of these disorders and the fact that they occur in elderly people demand careful diagnostic scrutiny to avoid missing a malignant neoplasm or major infection and possibly treating patients inappropriately with high-dose steroids. This is true in PMR because there is no diagnostic test and in GCA because the GCA biopsy finding may be normal in the face of active, vision-threatening vasculitis. Infections that must be considered and ruled out if clinically appropriate include tuberculosis, endocarditis, and hepatitis B and C. Malignant neoplasms such as lymphoma and multiple myeloma may mimic PMR, and an age-appropriate cancer evaluation is always indicated in this age group. Autoimmune disorders such as elderly-onset rheumatoid arthritis and systemic lupus erythematosus, as well as dermatomyositis and other types of vasculitis, must be considered in the differential diagnosis and sorted out by employing clinical information and serologic testing. There is support for the concept that elderly-onset rheumatoid arthritis is the same disorder as PMR with negative rheumatoid factor, a more proximal focus of joint inflammation, and a good response to low-dose prednisone. The distinction may be semantic because neither disorder tends to evolve into an erosive

arthritis. A more protracted clinical course, however, is often seen in patients in whom distal synovitis is a prominent feature, and those patients are classified as having elderly-onset rheumatoid arthritis. PMR and GCA should always be thought of in the setting of a fever of unknown origin because symptoms and signs can be occult or the history incomplete.

TREATMENT  Both PMR and GCA are highly responsive to corticosteroids, which are the treatment of choice.8 This response is so characteristic that an immediate and dramatic improvement in PMR and GCA symptoms within 1 to 3 days after steroid institution supports the diagnosis. Conversely, a lack of rapid and significant improvement in signs, symptoms, and function within 5 to 7 days should lead the clinician to suspect the initial impression and consider an alternative diagnosis (e.g., tumor or infection) or the presence of GCA in PMR patients that might require a higher steroid dose. Because the inflammatory burden of the two disorders is different, different doses of steroids are employed at the start of treatment. Whereas PMR usually responds to 15 mg of prednisone daily, GCA usually requires 40 to 60 mg of prednisone per day in divided doses or higher doses if organ or tissue damage is present or threatened. In GCA, if visual symptoms are present as a fixed loss or amaurosis fugax, the patient often should be treated with high-dose intravenous methylprednisolone with doses ranging from 40 mg every 8 hours to 1 g/day for 3 days, followed by high-dose oral steroids in divided doses. Within 2 to 3 days after the institution of steroids, most symptoms of PMR or GCA clear rapidly, and patients describe a miraculous improvement.9 The steroid dose is then maintained for 2 to 3 weeks, during which the ESR, C-reactive protein, hemoglobin, and platelet counts normalize. Steroid taper is then instituted and guided by the clinical response. In PMR, taper is commonly by 1 mg every 7 to 10 days; in GCA, taper is by 5 to 10 mg every 7 to 10 days. In GCA, the use of alternate-day corticosteroid regimens to minimize steroid side effects is generally not recommended because randomized controlled trials have demonstrated higher rates of treatment failure with alternate-day dosing schedules. It is important that the taper be guided primarily by clinical findings (e.g., PMR stiffness, headache, fatigue) and that the level of ESR elevation be considered within that clinical context. One should never “chase the ESR” because the elderly patient would be subjected inappropriately to a dangerously high cumulative dose of steroids with their attendant side effects. An increased dose of prednisone should be based on a change in symptoms, not solely on an increase in the ESR. One possible exception is in a patient with a history of GCA and prior abrupt vision loss in one eye, in whom any further compromise of vision would be catastrophic. The effective dose demanded for a flare often can be as low as 5 to 10 mg/day of prednisone, and uncommonly up to 60 mg/ day to control symptoms (e.g., visual abnormalities). A persistently elevated ESR (>50 mm/hour) without PMR or GCA symptoms should alert the physician to look for alternative causes, such as infection. Treatment is a careful balancing act between disease control and avoidance of steroidrelated toxicity. The overall goal of the patient and the physician is to attain the best disease control with the lowest dose of steroids. In most patients, prednisone can be tapered safely in 1 to 2 years. However, other patients may need to take low doses of steroids for 2 years or more. The higher the initial dose and cumulative dose, the greater the likelihood that the patient will develop a major steroid side effect such as sepsis, osteoporosis, osteonecrosis, diabetes, emotional lability, or myopathy (Chapter 32). Appropriate immunizations, osteoporosis regimens (calcium, vitamin D, and bisphosphonates), and metabolic monitoring are mandatory in all patients prescribed chronic steroid therapy. The major feared outcome in GCA is ischemic complications of the disease, most often vision loss or, less frequently, cerebrovascular accident. Vision loss is usually irreversible, and although it is uncommon after the diagnosis is suspected and glucocorticoid therapy is instituted, it can occur early in the course of treatment. A meta-analysis of retrospective studies showed that antiplatelet or anticoagulant therapy has a marginal benefit when used together with corticosteroids in patients with established GCA. Although this has not been demonstrated in prospective randomized controlled trials, in most patients, adjunctive therapy with low-dose aspirin should be considered unless there is a strong contraindication to its use. Alternative immunosuppressive agents have been tested in both PMR and GCA patients in an attempt to “spare steroids” and to control the inflammatory state. Studies examining the efficacy of methotrexate in GCA have yielded mixed results, with the largest, most recent study showing no incremental benefit from combined therapy. One individual patient meta-analysis of three randomized placebo-controlled trials suggested a modest benefit to methotrexate in GCA in terms of affording a steroid-sparing benefit and reducing likelihood of flares. Given the modest nature of the benefit demonstrated and the potential toxicities of methotrexate in this elderly population, methotrexate is not routinely incorporated as first-line therapy in GCA. In PMR, methotrexate has been shown in one study to afford a benefit in terms of steroid sparing and possibly reducing numbers of flares. The magnitude of the benefit appears modest, and no reduction in corticosteroid-related side effects was demonstrated. At present,

methotrexate is not routinely used in the management of either disease, but in individual patients with refractory disease or excessive corticosteroid morbidities, addition of weekly rheumatoid arthritis–level doses of methotrexate (7.5 to 20 mg/week) or azathioprine (2 mg/kg/day) is employed in selected instances. There have been case series suggesting cyclophosphamide may be of value in patients with refractory disease and/or unacceptable corticosteroid toxicity, but adverse events are common, and it is rare to use this agent in GCA. Tocilizumab (TCZ), a humanized monoclonal anti-IL-6 receptor antibody, has been shown in 2 randomized trials to have efficacy in GCA. A1  A2  TCZ use resulted in higher remission rates, less cumulative glucocorticoid use, and higher likelihood of sustained remission at 12 months than treatment with placebo.10 Relatively rapid glucocorticoid tapers were utilized (tapered to off by month 6), without occurrence of cranial ischemic complications in these trials. Tocilizumab has recently been approved by the FDA for the treatment of GCA based on these findings. Two prospective open label trials have similarly suggested TCZ to afford a significant steroid sparing benefit in PMR, but this has not yet been tested in randomized blinded placebo-controlled trials.11,12 Randomized controlled trials of TNF inhibitors including infliximab and adalimumab in GCA have failed to demonstrate benefit in terms of preventing relapses or affording a steroid-sparing benefit. A3  Treatment with abatacept, a costimulatory molecule blocker, has recently been shown in a randomized blinded placebo-controlled trial to result in a modest benefit in terms of relapsefree survival at 12 months A4  and is presently being evaluated in a larger trial. ,



FUTURE DIRECTIONS

Better understanding of the disease-causing roles of immunologically active cells and their cytokine products, along with genetics and correlations with clinical subsets, will lead to more focused treatment modalities and the avoidance of the need for long-term treatment with steroids. A recently published cohort study revealed that GCA is associated with increased risks for myocardial infarction, stroke, and peripheral vascular disease,13 suggesting that greater attention should be paid to cardiovascular risk reduction in patients with this disease.

  Grade A References A1. Villiger PM, Adler S, Kuchen S, et al. Tocilizumab for induction and maintenance of remission in giant cell arteritis: a phase 2, randomised, double-blind, placebo-controlled trial. Lancet. 2016;387:1921-1927. A2. Stone J, et al. Tocilizumab for sustained glucocorticoid-free remission in giant cell arteritis. N Engl J Med. 2017;377:317-328. A3. Yates M, Loke YK, Watts RA, et al. Prednisolone combined with adjunctive immunosuppression is not superior to prednisolone alone in terms of efficacy and safety in giant cell arteritis: meta-analysis. Clin Rheumatol. 2014;33:227-236. A4. Langford CA, et al. A randomized double-blind trial of abatacept and glucocorticoids for the treatment of giant cell arteritis. Arthritis Rheumatol. 2015;67:3949-3951.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 255  Giant Cell Arteritis and Polymyalgia Rheumatica  

GENERAL REFERENCES 1. Gonzalez-Gay MA, Mattesor EL, Castaneda S. Polymyalgia rheumatic. Lancet. 2017;390: 1700-1712. 2. Dejaco C, Brouwer E, Mason JC, et al. Giant cell arteritis and polymyaligia rheumatic. Nat Rev Rheumatol. 2017;13:578-592. 3. Gilden D, White T, Boyer PJ, et al. Varicella zoster virus infection in granulomatous arteritis of the aorta. J Infect Dis. 2016;213:1866-1871. 4. Rinagel M, Chatelus E, Jousse-Joulin S, et al. Diagnostic performance of temporal artery ultrasound for the diagnosis of giant cell arteritis: a systematic review and meta-analysis of the literature. Autoimmun Rev. 2019;18:56-61. 5. Prieto-Gonzalez S, Villarreal-Compagny M, Cid MC. Imaging in the management of giant cell arteritis. Med Clin (Barc). 2019. [Epub ahead of print.] 6. Baig IF, Pascoe AR, Kini A, et al. Giant cell arteritis: early diagnosis is key. Eye Brain. 2019;11: 1-12. 7. Keser G, Aksu K. Diagnosis and differential diagnosis of large-vessel vasculitides. Rheumatol Int. 2019;39:169-185.

1761.e1

8. Dejaco C, Singh YP, Perel P, et al. 2015 Recommendations for the management of polymyalgia rheumatica: a European League Against Rheumatism/American College of Rheumatology collaborative initiative. Ann Rheum Dis. 2015;74:1799-1807. 9. Buttgereit F, Dejaco C, Matteson EL, et al. Polymyalgia rheumatica and giant cell arteritis: a systematic review. JAMA. 2016;315:2442-2458. 10. Strand V, Dimonaco S, Tuckwell K, et al. Health-related quality of life in patients with giant cell arteritis treated with tocilizumab in a phase 3 randomised controlled trial. Arthritis Res Ther. 2019;21:1-9. 11. Lally L, Forbess L, Hatzis C, et al. Efficacy and safety of tocilizumab for the treatment of polymyalgia rheumatic. Arthritis Rheumatol. 2016;68:2550-2554. 12. Devauchelle-Pensec V, et al. Efficacy of first-line tocilizumab therapy in early polymyalgia rheumatica: a prospective longitudinal study. Ann Rheum Dis. 2016;75:1506-1510. 13. Li L, Neogi T, Jick S. Giant cell arteritis and vascular disease-risk factors and outcomes: a cohort study using UK Clinical Practice Research Datalink. Rheumatology (Oxford). 2017;56:753-762.

1761.e2

CHAPTER 255  Giant Cell Arteritis and Polymyalgia Rheumatica  

REVIEW QUESTIONS 1. A 74-year-old gentleman presents with a 7-week history of increasing generalized shoulder and hip pain, morning stiffness, low-grade fevers, and a dry cough. He was found to be anemic with a hemoglobin of 9.7 and had a thrombocytosis with a platelet count of 642 and an elevated sedimentation rate of 114. A search for infection and neoplasia has been unrevealing, including a bone marrow biopsy, which did not suggest a myeloproliferative neoplasm. He denies headaches, visual symptoms, or jaw claudication. The patient has reluctance regarding any further interventional procedures. Which of the following would be the best option to establish his diagnosis? A . A computed tomographic (CT) angiogram with imaging of the aortic arch and its major branches B. A conventional angiogram C. Duplex ultrasound of his temporal arteries and axillae D. A positron emission tomography (PET) scan E. Bilateral temporal artery biopsies Answer: E  The patient likely has GCA. Although PET scanning, duplex ultrasound of temporal arteries, and CT angiography may be helpful in some patients, a temporal artery biopsy remains the gold standard and is the most specific if the biopsy can be established. Duplex sonography has yet to be proved reliably accurate. PET scanning and angiography, although potentially helpful, can provide false-positive results in patients with severe atherosclerosis and false-negative results in patients without clear evidence of large vessel disease. Polymyalgia rheumatica (PMR) may be a possibility here (and this patient indeed has PMR), but the presence of dry cough and degree of inflammatory response makes concurrent GCA a pressing concern. 2. A 77-year-old woman with a history of diabetes, osteoporosis, and modest alcohol use presents with 3 weeks of headaches and feeling systemically unwell and then loss of vision in her left eye 2 days ago. She is found to be anemic with a markedly elevated sedimentation rate, and a temporal artery biopsy confirms the diagnosis of GCA. She is started on 60 mg of daily prednisone. Which of the following would be an appropriate additional intervention at this point? A . Add Infliximab 5 mg/kg monthly as a steroid-sparing intervention. B. Add methotrexate 10 mg weekly with the hope of titrating that to 15 mg weekly to afford a steroid-sparing benefit. C. Change her corticosteroid regimen to 120 mg on alternate days and plan an alternate day taper from there. D. Add low-dose aspirin daily to her regimen. E. Initiate treatment with tocilizumab 162 mg subcutaneously weekly, and discontinue corticosteroid therapy. Answer: D  Patients with GCA often struggle with comorbidities that raise concerns regarding prolonged exposure to corticosteroids. There is at this time, however, no proven steroid-sparing intervention that can decrease the risk for cranial ischemic complications. Tocilizumab has been demonstrated to afford a corticosteroid sparing benefit, and may be appropriate as initial therapy for GCA in such patients, but has been used in conjunction with corticosteroids in the trials which have demonstrated efficacy. Encouragingly, relatively brief courses of corticosteroids (6 months) were used without resultant ischemic complications in the largest clinical trial. Low-dose aspirin has been shown in two retrospective studies to be likely protective against cranial ischemic complications, so if there is no contraindication, that would be an appropriate intervention. Methotrexate has been shown in one trial in one individual meta-analysis to have a corticosteroid benefit in terms of cumulative steroid dose and also reduced risk for flares, but in the largest multicenter study addressing its efficacy, it was not shown to be superior to placebo. Moreover, in this patient with comorbidities including diabetes and alcohol use, methotrexate could be morbid, so this would not seem an appropriate intervention at this point. Infliximab has been shown in a randomized, doubleblinded, placebo-controlled trial not to be helpful in the treatment of GCA and can increase risk for infection in this patient already on high doses of corticosteroids with underlying diabetes. Alternate-day corticosteroid therapy has been shown to increase the risk for later flare and is not generally employed in GCA, particularly early in the course of treatment.

3. A 72-year-old man presents with a 2-month history of morning stiffness, mild synovitis in his knees, puffiness in his hands in the early morning hours, and severe shoulder and thigh girdle myalgias. He is mildly anemic with hemoglobin of 11.6 and has a markedly elevated sedimentation rate at 82. He is seronegative for rheumatoid factor, citric citrullinated peptide (CCP), and antineutrophil cytoplasmic antibodies. He has no evidence of peripheral joint erosions on plain radiographs. Which of the following would be the most appropriate therapeutic intervention at this point? A . Begin naproxen 500 mg twice daily. B. Begin methotrexate 10 mg weekly with the hope of bringing that dose up to 15 mg weekly in the next few weeks. C. Begin methylprednisolone 12 mg daily. D. Begin hydroxychloroquine 200 mg twice daily. E. Begin methylprednisolone 12 mg daily along with methotrexate 10 mg weekly with a plan to taper the corticosteroids to off over the next 4 weeks. Answer: C  The patient described has PMR but presents with some peripheral synovitis, which can be seen in up to 10 to 15% of patients with PMR. The patient clinically is unlikely to have rheumatoid arthritis in the absence of seropositivity for rheumatoid factor and CCP, has typical shoulder and thigh girdle myalgias and morning stiffness, and does not have evidence of erosive disease radiographically. An initial trial of methylprednisolone 12 mg daily would have both therapeutic and likely diagnostic value because if his response is spectacular, that would strongly argue for PMR and continuing that approach to therapy. Although methotrexate and hydroxychloroquine have been agents of interest in polymyalgia rheumatica, their use has not been well established in double-blinded, placebo-controlled trials. One study suggesting a benefit to methotrexate in PMR did not demonstrate a benefit in terms of reducing numbers of corticosteroid-related adverse events, and at longer term followup, the initial benefits seen by the addition of methotrexate did not seem to persist. Naproxen would be unlikely to afford adequate relief to this patient and, given his comorbidities, would likely be as problematic if not more problematic than a trial of corticosteroids. 4. An 80-year-old gentleman with a prior history of GCA successfully treated 7 years earlier presents with searing chest pain. His past medical history is otherwise only notable for borderline hypertension. He was successfully treated with a 1-year course of steroids but has been free of corticosteroids for 3 years. He has been feeling systemically well until recently. He had not seen a physician in 2 years. Which of the following explanations for his chest pain is of primary concern? A . Dissecting thoracic aneurysm B. Acute pericarditis C. A myocardial infarction related to his increased risk for atherosclerosis given his prior history of GCA D. Costochondritis related to a flare of PMR E. Rib fracture Answer: A  Patients with a prior history of GCA are at increased risk for thoracic aortic aneurysms, even several years after completing therapy without evidence of residual active disease. The searing quality of his pain and his prior history of treated GCA make this a concern that should not be missed. The quality of his pain did not sound suggestive of a myocardial infarction, and patients with GCA have not been demonstrated to be at increased risk for atherosclerotic heart disease, although the possibility of that association has been of interest. Patients with GCA are not particularly at a higher risk for pericarditis. Although PMR can arise even years after successfully treated GCA (or even years after successfully treated PMR), the quality of his pain does not sound suggestive thereof, and costochondritis would not typically be a feature of PMR. The quality of the pain described did not sound suggestive of a rib fracture, but moreover that would be a diagnosis that could be overlooked at least initially without putting the patient at increased risk for a serious cardiovascular compromise.

CHAPTER 255  Giant Cell Arteritis and Polymyalgia Rheumatica  

5. A 72-year-old woman had presented with an 8-week history of progressively increasing morning stiffness, shoulder and thigh girdle myalgias and arthralgias, and fatigue. An extensive search for underlying infection and ageappropriate screening for underlying neoplasm were unrevealing. A serum immunofixation did not reveal a monoclonal gammopathy. She was found to be anemic with a hemoglobin of 10.7 and had a sedimentation rate of 108. She denied any headaches, scalp sensitivity, jaw claudication, or visual symptoms. She admitted to low-grade fevers but not greater than 100° F. PMR was suspected, and she was treated with methylprednisolone 12 mg daily in divided doses. Three days later, she called to report no significant improvement in her symptoms. It was suggested that her dose be raised to 16 mg daily in divided doses. She comes in for evaluation still complaining of similar myalgias and arthralgias, and her sedimentation rate remains elevated at 62, although her hemoglobin improved slightly to 10.9. Which of the following would be an appropriate next intervention? A . Add methotrexate as a disease-controlling/steroid-sparing intervention. B. Add infliximab to help achieve disease control. C. Refer her for a bone marrow biopsy to exclude underlying myeloproliferative or lymphoproliferative disease. D. Add hydroxychloroquine 400 mg daily. E. Refer her for a temporal artery biopsy and raise the methylprednisolone to 32 mg daily.

1761.e3

Answer: E  This patient presented with fairly typical PMR symptoms but in addition had low-grade fevers, which can be seen but are not as common as they are in GCA. Moreover, she was somewhat anemic. Her lack of response to 12 mg and then even 16 mg of daily methylprednisolone makes PMR unlikely unless in the context of a more substantial inflammatory disease such as GCA. GCA is found in approximately 10% of patients with “pure” PMR, and conversely, up to 50% of patients with GCA can present with PMR-like symptoms.

1761

CHAPTER 256  Infections of Bursae, Joints, and Bones  

256  INFECTIONS OF BURSAE, JOINTS, AND BONES ERIC L. MATTESON AND DOUGLAS R. OSMON

  INFECTION OF BURSAE



Septic Bursitis  

DEFINITION

Bursae are the satellite structures that form to protect tissues from bony prominences. The superficial bursae, including the olecranon, prepatellar, infrapatellar, and bursae over the first metatarsophalangeal bunions, are more likely to become infected than are the deep bursae, such as the subacromial, trochanteric, and iliopsoas bursae.

1761.e4

ABSTRACT

CHAPTER 256  Infections of Bursae, Joints, and Bones  

Septic bursitis, arthritis, and osteomyelitis most commonly result from direct inoculation through the skin, by hematologic spread, or following bone and joint surgery. Predisposing factors include immune-compromised state, autoimmune diseases, diabetes mellitus, alcoholism, and antecedent surgery, including bowel and bladder procedures and joint arthroplasty. Diagnosis is based on diagnostic suspicion and frequently elevated acute phase reactants such as the sedimentation rate and C-reactive protein, elevated white blood cell count, and identification of causative organisms by culture and serologic identification in joint or bursal fluid or blood, or in some cases, especially fungal infections and tuberculosis, on tissue biopsy. Magnetic resonance imaging, computed tomography, and radiography, as well as nuclear medicine techniques, aid in identification of infection foci. Initial treatment of suspected bacterial and fungal infection is empirical, guided by the clinical presentation and informed by identification of the causative organism. Surgical treatment includes débridement of the affected musculoskeletal structure, and management of prosthetic joint infections requires débridement with initial retention of the prosthesis, resection arthroplasty, or sometimes amputation. Successful management is dependent on the medical and surgical therapies employed, ensuring adequate vascularization of the affected tissues, and ability to manage comorbidities such as diabetes mellitus.

KEYWORDS

septic arthritis joints bones bursa osteomyelitis debridement antibiotics prosthetic joint infection

1762



CHAPTER 256  Infections of Bursae, Joints, and Bones  

consist of an oral antistaphylococcal penicillin or first-generation cephalosporin. If community-acquired methicillin-resistant S. aureus (MRSA) is suspected, cotrimoxazole or minocycline may be added to one of these agents. In patients who are allergic to penicillin, oral clindamycin or linezolid may be used. Patients who have severe inflammation, are septic, or are immunocompromised may require hospitalization for initiation of treatment with intravenous nafcillin, oxacillin, or cefazolin; if methicillin-resistant Staphylococcus aureus (MRSA) is suspected, intravenous vancomycin, daptomycin, or linezolid should be used (Chapter 272). Vancomycin can also be used in patients who are allergic to penicillin. The duration of antimicrobial therapy is guided by the clinical response and comorbid conditions. It should be continued until there is no longer bursal inflammation. This may require several weeks of intravenous or oral therapy and multiple aspirations. Failure of the septic bursitis to respond to initial antibiotic therapy mandates a second course of therapy; recurrence thereafter or inability to adequately drain the bursa with needle aspiration is an indication for surgical intervention.

EPIDEMIOLOGY

Olecranon bursitis may occur in as many as 10 in 100,000 persons. The majority of cases occur in men, and antecedent trauma to the skin is frequent.  

PATHOBIOLOGY

Septic bursitis of superficial bursae is most commonly due to direct inoculation through the overlying skin; less commonly, it is secondary to overlying cellulitis. Most cases of deep septic bursitis are due to contiguous spread from adjacent infected joints or hematogenous seeding. Predisposing risk factors for septic bursitis include trauma to the skin. For example, olecranon septic bursitis may occur in plumbers, athletes, and patients with chronic obstructive pulmonary disease (COPD) who frequently lean on their elbows; prepatellar or infrapatellar septic bursitis can occur in housecleaners, gardeners, and carpet layers. At least one third of patients with septic bursitis have an underlying comorbid illness such as diabetes mellitus, rheumatoid arthritis, gout, COPD, or alcoholism.  

CLINICAL MANIFESTATIONS

In immune-competent patients, septic bursitis often but not always presents with fever and erythema and warmth of the overlying skin; there may be swelling of the bursae. In contrast to those with septic arthritis, patients with septic bursitis of superficial bursae have intact range of motion of the joints, which may be limited only at the extremes of flexion.1 Pain on motion of the joint and restriction of joint range of motion are highly suggestive of septic arthritis. Acute phase reactants such as C-reactive protein, the sedimentation rate, and the white blood cell count (WBC) may be elevated.  

DIAGNOSIS

Radiography should be performed to look for a foreign body and to evaluate the surrounding bones. Aspiration of bursal fluid is helpful in the diagnosis of patients who have pain, erythema, and/or swelling of an affected area. However, given the risk for contaminating the bursa if the aspiration occurs through cellulitic skin, many clinicians choose to aspirate a bursa only if empirical antimicrobial therapy has failed. Ultrasound or computed tomography (CT) guidance greatly enhances the successful aspiration of superficial bursae. Care must be taken not to violate the joint space when aspirating a bursa to avoid inoculating it. The leukocyte count of the bursal fluid is generally lower than that seen in septic arthritis, with a mean of 13,500 cells/μL. Even in immune-competent hosts, cell counts can range from less than 1500/μL to greater than 100,000/μL. A leukocyte count greater than 2000/μL has a sensitivity of 94% and a specificity of 79% for superficial (olecranon or prepatellar) bursitis. Bacterial culture and in vitro susceptibilities must be obtained; if additional fluid is available, a Gram stain may be obtained, although its sensitivity may be as low as 15%. The presence of crystals does not exclude the possibility of septic bursitis (Chapter 257). Staphylococcus aureus (Chapter 272) is the most common cause of septic bursitis, present in more than 80% of culture-proven cases, followed by β-hemolytic streptococci. Aerobic gram-negative bacilli, including Escherichia coli, Campylobacter jejuni, and Pseudomonas species, are rare causes of septic bursitis. Chronic bursitis may be associated with systemic infections due to Brucella abortus, atypical mycobacteria, or Mycobacterium tuberculosis, as well as fungi; the presence of these infections should raise the possibility of systemic infection.

Differential Diagnosis

In the immune-competent host, nonseptic bursitis (Chapter 247) may have a somewhat more indolent presentation than septic bursitis. The differential diagnosis includes gout, pseudogout, arthritis, and trauma with hemobursa. An overlying cellulitis may be confused with bursitis. Fever is usually not present in nonseptic bursitis because of mechanical or friction trauma.

TREATMENT  Treatment of septic bursitis is guided by knowledge of the putative underlying organisms, in most cases, S. aureus. Because the Gram stain is positive in less than two thirds of patients and cultures may be delayed, empirical therapy is guided by the clinical presentation. Most patients can be treated as outpatients, but those who are immunocompromised may require hospitalization for intravenous antibiotic therapy. Initial ambulatory treatment in patients without comorbidities may



PREVENTION

Because superficial septic bursitis is often associated with occasional or avocational activities involving kneeling or resting on the elbows, using protective padding may be helpful.  

PROGNOSIS

The optimal duration of therapy is unknown, but prognosis of superficial bursitis is generally excellent. The presence of comorbid conditions, especially those associated with deep bursal infections, including septic arthritis, bacteremia, and osteomyelitis, is associated with more intractable and difficult disease.

  INFECTION OF JOINTS



Septic Arthritis  

DEFINITION

Septic arthritis refers to infection of a joint by a microorganism. It is associated with increased morbidity and mortality as well as loss of articular integrity and function.2 Septic arthritis is usually caused by a bacterial infection. Other microorganisms can cause infections with clinical characteristics that differ from those of bacterial infections; these are reviewed separately.  

NONGONOCOCCAL SEPTIC ARTHRITIS  

EPIDEMIOLOGY

The incidence of septic arthritis affecting native joints is about 5 to 8 in 100,000 patient years. Among patients presenting with an acutely swollen and painful joint, the prevalence of bacterial arthritis ranges widely, from less than 10% to as high as 27%, depending on the source population. Nongonococcal septic arthritis is the most common form of septic arthritis and is somewhat more common in men than in women.  

PATHOBIOLOGY

More than 90% of cases of septic arthritis are due to staphylococci or streptococci (Table 256-1). Septic arthritis can result from direct inoculation (e.g., accidents, bites, surgery) or by extension from infected bone into an adjacent joint space. Approximately 75% of cases are due to hematogenous spread, particularly in patients with indwelling catheters and immunocompromised patients. Septic arthritis due to needle arthrocentesis (4 days) fever

• Daily examination and history • Blood cultures – repeat on limited basis • Cultures for any suspected sites of infection

Unexplained fever • Clinically stable • Myeloid recovery not imminent • Consider CT scan sinuses and lungs

Unexplained fever • Clinically stable • Rising ANC: Myeloid recovery imminent

Observe; No antimicrobial changes unless clinical, microbiologic, or radiographic data suggest new infection

Receiving fluconazole (antiyeast) prophylaxis

Preemptive approach; start antifungal based upon results of: • CT scans chest/sinuses • Serial serum galactomannan tests

Documented infection • Clinically unstable • Worsening signs and symptoms of infection

Receiving anti-mold prophylaxis

Empirical antifungal therapy with anti-mold coverage: • echinocandin • voriconazole • amphotericin B preparation

• Examine and re-image (CT, MRI) for new or worsening sites of infection • Culture/biopsy/drain sites of worsening infection: assess for bacterial, viral, and fungal pathogens • Review antibiotic coverage for adequacy of dosing and spectrum • Consider adding empirical antifungal therapy • Broaden antimicrobial coverage for hemodynamic instability

Empirical antifungal therapy: • Consider switch to a different class of mold active antifungal

FIGURE 265-4.  High-risk patient with fever after 4 days of empirical antibiotics. ANC = absolute neutrophil count; CT = computed tomography; IV = intravenous; MRI = magnetic resonance imaging. (Modified from Freifeld AG, Bow EJ, Sepkowitz KA, et al. Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 update by the Infectious Diseases Society of America. Clin Infect Dis. 2011;52:e56-e93.)

associated with fever, rash, and manifestations of hemophagocytosis, although there are other infectious causes of hemophagocytic syndromes as well (Chapter 160). The characteristic vesicular rash of varicella reactivation is relatively frequent in stem cell and solid organ transplant recipients with chronic T-cell deficiencies, especially in the absence of antiviral prophylaxis. Disseminated disease in immunocompromised patients is associated with high mortality rates unless prompt antiviral therapy is initiated. Antivirals administered as prophylaxis in high-risk HSCT and solid organ transplant recipients can decrease both early morbidity and late mortality associated with herpes simplex virus (HSV), varicella-zoster virus (VZV), and CMV disease, although drugrelated toxicities must be considered in risk-benefit calculations. There are numerous noninfectious causes of rashes and lesions that are common in immunocompromised individuals, particularly drug-induced hypersensitivity syndromes that can be both mild and severe, and rarely progress to life-threatening toxic epidermal necrolysis. Sweet syndrome, or acute febrile neutrophilic dermatosis, is characterized by skin lesions with neutrophilic infiltration in the dermis (see Fig. 411-23). It may appear during impending neutrophil recovery, treatment with granulocyte colony-stimulating factor, associated with numerous drugs, or as a paraneoplastic manifestation, particularly in hematologic malignancies (most commonly acute myelogenous leukemia). Biopsy with appropriate microbial stains and culture is essential to distinguish these lesions from infectious causes of ecthyma gangrenosum and other disseminated infections, such as those caused by mycobacteria, Nocardia, and fungi.

  RESPIRATORY SYNDROMES

Pneumonia should be suspected in a patient who has respiratory symptoms as manifested by cough, shortness of breath, chest pain, and hypoxia, even in

the absence of radiographic abnormalities initially. While opportunistic infections are always a consideration, pneumonia in immunocompromised patients is more often caused by community-acquired pathogens such as pneumococci and Haemophilus influenzae that can cause lobar or diffuse pneumonia. Patients on ventilators are at risk for secondary gram-negative bacillary pneumonia or staphylococcal pneumonia. Opportunistic fungi have been increasingly recognized as causes of lung infection in immunocompromised neutropenic patients and transplant recipients. Epidemic mycoses such as blastomycosis, coccidioidomycosis, and histoplasmosis may be manifested as acute pneumonia after recent exposure in immunocompromised patients, making a travel history mandatory. Although Candida species commonly colonize indwelling vascular and urinary catheters, candidal pneumonia is unusual in the absence of systemic candidiasis. Although classically associated with an “interstitial pattern” of lung infiltration, Pneumocystis species pneumonia can be manifested as local consolidation or pulmonary nodules, exhibiting granulomatous inflammation on pathologic examination. Filamentous fungi, including Aspergillus, Zygomycetes, Fusarium species, and Scedosporium species, are challenging to treat. From an initial focus, Aspergillus infection can spread through the pulmonary vasculature, which sets the stage for localized hemorrhage, creating a halo sign on CT scan, and infarction and necrosis. This can progress to cavitary lesions. Invasive pulmonary aspergillosis is usually caused by Aspergillus fumigatus and can be very difficult to identify specifically by conventional methods for detecting fungal infections. Fluorescent in situ hybridization–based molecular methods will be a promising approach.11 These organisms can also cause primary airway disease, presenting with features typical of tracheobronchitis, with or without findings apparent on CT scan. This infection is particularly well described in lung transplant recipients, who may also have involvement of the bronchial anastomosis.

1823

CHAPTER 265  SUSPECTED INFECTION IN THE IMMUNOCOMPROMISED HOST  

TABLE 265-8 COMMON INFECTIOUS AND NONINFECTIOUS SYNDROMES IN IMMUNOCOMPROMISED HOSTS PRIMARY ORGAN SYSTEM

BACTERIA

FUNGI

VIRUSES

PARASITES AND PROTISTS

NONINFECTIOUS

Cutaneous

Disseminated gram-positive and gram-negative bacteria, e.g., Staphylococcus aureus Pseudomonas aeruginosa Mycobacterium spp Nocardia spp

Candida spp. Filamentous fungi, e.g., Aspergillus spp Zygomycetes Fusarium spp Scedosporium spp Cryptococcus spp

Herpes simplex Varicella-zoster CMV HHV-6 Adenovirus Parvovirus B19

Leishmania spp Acanthamoeba Naegleria fowleri Balamuthia mandrillaris

Drug eruptions GVHD Sweet syndrome

Sinopulmonary

Gram-positive and gram-negative causes of sinusitis and pneumonia S. aureus Streptococcus pneumoniae P. aeruginosa Haemophilus influenzae Anaerobes Legionella Nocardia spp Mycobacterium spp

Filamentous fungi, e.g., Aspergillus spp Zygomycetes Fusarium spp Scedosporium spp Cryptococcus spp Endemic fungi, e.g., Histoplasma capsulatum Coccidioides immitis Blastomyces dermatitidis Pneumocystis jiroveci

Respiratory viruses, e.g., RSV Parainfluenza Influenza Adenovirus Reactivation herpes viruses, e.g., CMV, VZV

Toxoplasma gondii Strongyloides stercoralis hyperinfection syndrome

Drug-related pulmonary toxicities Pneumonitis (sirolimus) Diffuse alveolar damage Bronchiolitis obliterans syndromes

Gastrointestinal

Neutropenic enterocolitis (“typhlitis”) Mixed gram-positive, gramnegative, anaerobes (Bacteroides fragilis, Clostridium septicum) Clostridium difficile colitis Enteric diarrheal pathogens Salmonella spp Shigella spp Escherichia coli Campylobacter spp

Candida spp. Microsporidia

CMV EBV-PTLD Adenovirus Coxsackievirus Rotavirus Norovirus

Cryptosporidium Giardia lamblia Cystoisospora belli Cyclospora cayetanensis Strongyloides stercoralis

Drug-related toxicities, e.g., MMF

Neurologic

Gram-positive and gram-negative bacteria Listeria monocytogenes Pneumococcus Meningococcus Nocardia spp Mycobacterium tuberculosis Syphilis

Filamentous fungi Cryptococcus spp

Herpes viruses HSV HHV-6 VZV JC virus West Nile virus Miscellaneous viral encephalitides

Toxoplasma gondii Acanthamoeba Naegleria fowleri Balamuthia mandrillaris

Drug-related toxicities, e.g., calcineurin inhibitor toxicity PRES carbapenem-related seizures voriconazole CNS toxicity Cefepime-induced encephalopathy

CMV = cytomegalovirus; EBV-PTLD = Epstein-Barr virus–post-transplantation lymphoproliferative disorder; GVHD = graft-versus-host disease; HHV-6 = human herpes virus 6; HSV = herpes simplex virus; MMF = mycophenolate mofetil; PRES = posterior reversible encephalopathy syndrome; RSV = respiratory syncytial virus; VZV = varicella-zoster virus.

A

B

FIGURE 265-5.  Ecthyma gangrenosum. A 28-year-old woman with fever and neutropenia while receiving chemotherapy for acute leukemia developed several tender edematous papules on her thighs. A, Central crust and surrounding erythema are shown. B, The papules became necrotic during 1 to 2 days, with the formation of black, well-demarcated eschar. Cultures from blood and the necrotic eschar grew Pseudomonas aeruginosa. (© DermAtlas; http://www.DermAtlas.org.)

There is a high risk for severe pulmonary infection caused by reactivation of M. tuberculosis in people who are treated with biologic immune response modifiers (Chapter 33), especially TNF antagonists, in the setting of autoimmune or other inflammatory diseases, such as rheumatoid arthritis, psoriasis, and inflammatory bowel diseases. The same patients are also at increased risk for invasive fungal infections, including reactivation of endemic infections

such as histoplasmosis, warranting enhanced screening and a high level of suspicion for disease. A particularly common concern is reactivation pneumonitis caused by members of the herpesvirus family, especially CMV, which occurs most frequently in the setting of chronic T-cell depression associated with transplantation. Respiratory viruses, which infect immunocompromised hosts with the

1824

CHAPTER 265  SUSPECTED INFECTION IN THE IMMUNOCOMPROMISED HOST  

A

C

B

FIGURE 265-6.  Mycobacterium marinum. A 20-year-old man with Crohn disease receiving infliximab developed progressively worsening nonhealing foot ulcers following trauma and exposure to brackish water 2 years earlier. Biopsy cultures of the ulcer grew Mycobacterium marinum. A, Two necrotic ulcers with central eschars, surrounding erythema, and pedal edema at the time of diagnosis. B, After 1 month of treatment, granulation tissue is evident in the base of the ulcer, and erythema and edema are reduced. C, After 10 months of treatment, the ulcers have closed with residual scars and hyperpigmentation.

A

B

FIGURE 265-7.  Disseminated candidiasis. A 60-year-old woman with fever during neutropenia that developed after receipt of therapy for acute leukemia developed tender papular lesions on her extremities, trunk, and back. A, Blood cultures returned positive for Candida tropicalis. B, After resolution of neutropenia, lesions developed a more pustular appearance.

same frequency as the general population, cause lower respiratory tract disease and pneumonitis more frequently in hosts with suppressed cellular immunity. Patients who have defects in cellular immunity typically exhibit higher viral loads and prolonged shedding, causing concern for the emergence of antiviral resistance and infection control.12 Noninfectious causes of pulmonary infiltrates in immunosuppressed hosts include early complications of chemotherapy administration (e.g., diffuse alveolar damage and hemorrhage) and late complications of GVHD and organ rejection (e.g., bronchiolitis obliterans syndromes). Certain drugs that are frequently administered in these populations of patients, like sirolimus, can cause direct lung toxicity.

  GASTROINTESTINAL SYNDROMES

Diarrhea in an immunocompromised host can be caused by a wide variety of infectious and noninfectious etiologies,13 including conventional enteric pathogens such as Salmonella (Chapter 292), Shigella (Chapter 293), and Campylobacter (Chapter 287). Currently and recently hospitalized patients who have received courses of antibiotic treatment are commonly infected with Clostridium difficile (Chapter 280). C. difficile colitis can be both severe and persistent in immunosuppressed hosts. More recently, norovirus has emerged as an important cause of chronic gastroenteritis in immunocompromised patients, in whom diagnoses have been commonly elusive and misapplied to noninfectious syndromes (e.g., GVHD), and outcomes can be poor (Chapter 356). Several intestinal protozoa are associated with diarrhea in compromised patients.14 Three acid-fast staining protozoa, Cryptosporidium species, Cystoisospora belli, and Cyclospora cayetanensis, are associated with predisposing impairments in cell-mediated immunity (see Chapters 329 and 332). Giardia lamblia is classically associated with hypogammaglobulinemia. The Microsporidia are a family of eukaryotic obligate intracellular parasites once thought to be protozoa but more recently placed taxonomically closer to fungi. In addition to causing diarrhea in immune-suppressed patients, they are distinguished by the ability to cause extraintestinal disease in virtually any other organ.

Because acute and chronic diarrhea in immunocompromised patients can be caused by a wide array of bacterial, viral, and protozoal pathogens with broadly overlapping clinical presentations, multiplex molecular panels that have the capability of rapidly detecting one to two dozen potential gastrointestinal pathogens have become important diagnostic tools in recent years. Caution should be exercised in interpretation, however, because a positive result may reflect active infection or asymptomatic colonization. The small intestine is the primary location of calcineurin inhibitor (e.g., cyclosporine, tacrolimus) absorption and metabolism; elevated blood concentrations of calcineurin inhibitors to toxic levels may be a sign of enteritis, for example, due to rotavirus or Cryptosporidium, perhaps as a consequence of altered drug metabolism. CMV is a major cause of gastrointestinal disease in transplant recipients and increasingly recognized in patients with impaired cell-mediated immunity not due to transplantation or AIDS (Chapter 352). It can involve the entire digestive tract from esophagus to anus, and patients typically present with signs and symptoms of disseminated disease, including fever, malaise, and bone marrow suppression (i.e., a mononucleosis-like syndrome). Symptoms of CMV gastritis, enteritis, and colitis include anorexia, nausea, abdominal discomfort, and diarrhea. CMV infection also causes mild elevations in hepatic enzymes and, rarely, fulminant hepatitis or pancreatitis, typically in the context of primary infection after liver transplantation or pancreas transplantation, respectively. Most laboratories use molecular assays (CMV nucleic acid amplification assays or CMV pp65 antigenemia assays) to establish a diagnosis of CMV gastrointestinal disease. However, sensitivity of these assays can occasionally be very low or undetectable because of tissue-invasive disease that is isolated (“compartmentalized”) to the alimentary tract. In these cases, endoscopy is required to establish a diagnosis. Indeed, because colitis and diarrhea in immunocompromised patients can be caused by such a wide range of problems, including multiple infections, focal Epstein-Barr virus–associated post-transplantation lymphoproliferative disorder, GVHD, and drug-induced toxicity, uncertainty in diagnosis or failure of response to appropriate treatment requires further evaluation with endoscopy.

Individuals who have received long courses of chemotherapy, radiation, and antibiotics commonly experience Candida mucosal overgrowth in the mouth and esophagus. HSV and CMV can cause symptoms identical to those of Candida esophagitis. In severely neutropenic patients, anaerobic streptococci and gram-negative pathogens such as P. aeruginosa can cause severe mucositis and pharyngitis. In cancer patients, these organisms take advantage of the cytotoxic effects of chemotherapy, which promotes sloughing of mucosal surfaces and subsequently predisposes to infection. Neutropenic patients may also develop enterocolitis that can be of mixed anaerobic and aerobic bacterial origin. Neutropenic enterocolitis, also known as typhlitis or ileocecal syndrome, results from chemotherapeutic damage to the intestinal mucosa in the setting of neutropenia. Presentation usually includes fever, abdominal pain, nausea, vomiting, and diarrhea. Because neutropenic enterocolitis can rapidly progress to intestinal perforation, sepsis, and multisystem organ failure, prompt diagnosis and aggressive medical or surgical intervention are required.

  NEUROLOGIC SYNDROMES

Brain abscess or meningitis can be caused in immunocompromised patients by gram-positive or gram-negative bacteria as well as by anaerobes.15 Listeria monocytogenes is a particularly common cause of meningitis in these patients. Encapsulated bacteria such as pneumococci and staphylococci can cause metastatic CNS disease and meningitis. In patients with impaired cell-mediated immunity, C. neoformans is likewise a leading cause of CNS infection, usually presenting as cryptococcal meningitis or meningoencephalitis and sometimes as parenchymal mass lesions of the brain. Aspergillus species can infect the CNS both by direct sinus invasion and by hematogenous spread. Patients with impaired cellular immunity can develop CNS infection with Toxoplasma gondii, M. tuberculosis, H. capsulatum, or Nocardia species, either through severe infection by latent organisms or by reactivation. CNS syphilis should also be considered in patients with severe immunologic impairment. Among the many viral etiologies of meningoencephalitis in immunocompromised individuals are enteroviruses, measles, neurotropic herpesviruses (HSV-1, CMV, VZV), and human herpesvirus 6. Progressive multifocal leukoencephalopathy caused by JC polyomavirus may occur in those with chronic CD4 lymphocytopenia and in recipients of hematopoietic stem cell or solid organ transplants as well as individuals treated with natalizumab, rituximab, and possibly other immunomodulatory monoclonal antibodies. Transplant recipients are also at increased risk for meningoencephalitis caused by West Nile virus. Patients with CNS symptoms presenting early after transplantation might be suspected of having severe infections acquired from the donor, including West Nile virus, rabies, HIV, lymphocytic choriomeningitis virus, or other viruses known to be transmitted through organ donation. In addition to opportunistic infection, noninfectious causes of neurologic symptoms must be considered in immunocompromised patients, including drug toxicity, immunologic disorders, paraneoplastic syndromes, and GuillainBarré syndrome. Posterior reversible encephalopathy syndrome (PRES) should be in the differential diagnosis of neurologic complications in transplant recipients, patients with autoimmune diseases, and those receiving high doses of cancer chemotherapy, particularly when the clinical presentation includes the sudden onset of severe “thunderclap” headache, seizures, confusion, and visual disturbance. PRES can be confirmed by characteristic CT or magnetic resonance imaging patterns of predominantly posterior cerebral edema and angiographic evidence of reversible vasoconstriction. PRES can be caused by endothelial injury, vasospasm, or edema associated with certain drugs such as calcineurin inhibitors (e.g., cyclosporine, tacrolimus).

  Grade A References A1. Vidal L, Ben Dor I, Paul M, et al. Oral versus intravenous antibiotic treatment for febrile neutropenia in cancer patients. Cochrane Database Syst Rev. 2013;10:CD003992. A2. Beyar-Katz O, Dickstein Y, Borok S, et al. Empirical antibiotics targeting gram-positive bacteria for the treatment of febrile neutropenic patients with cancer. Cochrane Database Syst Rev. 2017;6: CD003914. A3. Nakane T, Tamura K, Hino M, et al. Cefozopran, meropenem, or imipenem-cilastatin compared with cefepime as empirical therapy in febrile neutropenic adult patients: a multicenter prospective randomized trial. J Infect Chemother. 2015;21:16-22. A4. Horita N, Shibata Y, Watanabe H, et al. Comparison of antipseudomonal beta-lactams for febrile neutropenia empiric therapy: systematic review and network meta-analysis. Clin Microbiol Infect. 2017;23:723-729. A5. Stern A, Carrara E, Bitterman R, et al. Early discontinuation of antibiotics for febrile neutropenia versus continuation until neutropenia resolution in people with cancer. Cochrane Database Syst Rev. 2019;1:CD012184.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 265  SUSPECTED INFECTION IN THE IMMUNOCOMPROMISED HOST  

GENERAL REFERENCES 1. DeWitt S, Chavez SA, Perkins J, et al. Evaluation of fever in the emergency department. Am J Emerg Med. 2017;35:1755-1758. 2. Fishman JA. Infection in organ transplantation. Am J Transplant. 2017;17:856-879. 3. White L, Ybarra M. Neutropenic fever. Hematol Oncol Clin North Am. 2017;31:981-993. 4. Pizzo PA. Management of patients with fever and neutropenia through the arc of time: a narrative review. Ann Intern Med. 2019;170:389-397. 5. Taplitz RA, Kennedy EB, Bow EJ, et al. Outpatient management of fever and neutropenia in adults treated for malignancy: American Society of Clinical Oncology and Infectious Diseases Society of America clinical practice guideline update. J Clin Oncol. 2018;36:1443-1453. 6. Moon H, Choi YJ, Sim SH. Validation of the Clinical Index of Stable Febrile Neutropenia (CISNE) model in febrile neutropenia patients visiting the emergency department. Can it guide emergency physicians to a reasonable decision on outpatient vs. inpatient treatment? PLoS One. 2018;13:1-12. 7. Schmidt-Hieber M, Teschner D, Maschmeyer G, et al. Management of febrile neutropenia in the perspective of antimicrobial de-escalation and discontinuation. Expert Rev Anti Infect Ther. 2019:1-13.

1825.e1

8. Crawford J, Becker PS, Armitage J, et al. Myeloid growth factors, version 2.2017, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2017;15:1520-1541. 9. Smith TJ, Bohlke K, Lyman GH, et al. Recommendations for the use of WBC growth factors: American Society of Clinical Oncology Clinical Practice Guideline Update. J Clin Oncol. 2015;33: 3199-3212. 10. Burke VE, Lopez FA. Approach to skin and soft tissue infections in non-HIV immunocompromised hosts. Curr Opin Infect Dis. 2017;30:354-363. 11. Moura S, Cerqueira L, Almeida A. Invasive pulmonary aspergillosis: current diagnostic methodologies and a new molecular approach. Eur J Clin Microbiol Infect Dis. 2018;37:1393-1403. 12. Bitterman R, Eliakim-Raz N, Vinograd I, et al. Influenza vaccines in immunosuppressed adults with cancer. Cochrane Database Syst Rev. 2018;2:CD008983. 13. Panarelli NC, Yantiss RK. Inflammatory and infectious manifestations of immunodeficiency in the gastrointestinal tract. Mod Pathol. 2018;31:844-861. 14. Siddiqui ZA. An overview of parasitic infections of the gastro-intestinal tract in developed countries affecting immunocompromised individuals. J Parasit Dis. 2017;41:621-626. 15. Levin SN, Lyons JL. Infections of the nervous system. Am J Med. 2018;131:25-32.

1825.e2

CHAPTER 265  SUSPECTED INFECTION IN THE IMMUNOCOMPROMISED HOST  

REVIEW QUESTIONS 1. A patient who has an absolute neutrophil count below 500 cells/µL for 2 weeks after induction therapy for acute myelogenous leukemia is at increased risk for: A . Escherichia coli bacteremia from the gastrointestinal tract B. Influenza infection involving the upper respiratory tract C. Varicella-zoster virus reactivation D. Cytomegalovirus disease involving the lungs E. Cryptococcus neoformans meningoencephalitis Answer: A  These patients do not have increased risk for influenza involving the upper respiratory tract compared with the population in general but have increased risk for complications (such as pneumonitis). Answers C to E are correct only for people with advanced deficits in T-cell immunity. 2. Early antibiotic therapy would be appropriate before documenting a microbial diagnosis in: A . A neutropenic patient who has developed fever, with normal hemodynamics B. A heart transplant recipient who has a new left lower lobe consolidation, fever, and leukocytosis C. A kidney transplant recipient with pyuria, fever, and elevated creatinine concentration D. None of the above E. All of the above Answer: E  These all outline high-risk situations in which empirical antibiotic therapy would be appropriate before documentation of disease to avoid rapid progression with sepsis. 3. A man presents with diffuse papular skin lesions 20 days after receipt of bone marrow transplantation, with fever. He has been receiving fluconazole, moxifloxacin, and acyclovir prophylactically. The least likely cause of his skin lesions is: A . Candida glabrata B. Sweet syndrome C. Aspergillus fumigatus D. Herpes simplex virus E. A vasculitic reaction to a drug Answer: D  Herpes simplex virus rarely disseminates to cause these types of skin lesions, especially in people receiving acyclovir.

4. People who have received tumor necrosis factor inhibitors as biologic therapy for an autoimmune disease have heightened risks for infections caused by which of the following? A . Viruses B. Bacteria C. Fungi D. All of the above E. None of the above Answer: D  Tumor necrosis factor inhibition may alter immune responses to multiple arms of the immune system. 5. What is the most likely cause of neutropenic enterocolitis that developed after receipt of cytotoxic therapy for acute myelogenous leukemia? A . Cytomegalovirus B. Clostridium difficile C. Norovirus D. Mixed anaerobic and aerobic bacteria E. Adenovirus Answer: D  These infections are typically caused by organisms that are “native” in the gastrointestinal tract during episodes of mucosal injury.

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

1825

266  PREVENTION AND CONTROL OF HEALTH CARE–ASSOCIATED INFECTIONS DAVID P. CALFEE

THE BURDEN OF HEALTH CARE–ASSOCIATED INFECTIONS

The Centers for Disease Control and Prevention (CDC) defines health care– associated infections (HAIs) as infections that patients acquire during the course of receiving health care treatment for other conditions. Nosocomial infection is a term that refers specifically to an HAI that develops in association with hospital care. The development of infection during the course of health care is not, however, limited to the acute care hospital setting. Thus, health care–associated infection is the preferred term in referring to the broader spectrum of infections that develop during the course of health care, wherever that care may be provided, including acute care hospitals, long-term care facilities, rehabilitation facilities, dialysis facilities, and even the patient’s home during the receipt of home care services. The most extensive data regarding the incidence of and outcomes associated with HAIs come from the acute care hospital setting.1 On the basis of a point prevalence study conducted in 183 U.S. hospitals in 2011, it was estimated that up to 1.4 million HAIs occur in hospital patients each year, with approximately 75,000 associated deaths (Table 266-1). By 2015, there was about a 16% decrease, from 4 to 3.2%, in prevalence.2 Previous European studies have estimated that 4.1 million HAIs occur in European acute care hospitals each year. Thus, approximately one of every 14 to 20 patients admitted to U.S. and European hospitals develops an HAI, making HAI one of the most common complications associated with the receipt of health care. Moreover, these data indicate that HAIs are one of the top 10 causes of death in the United States. Whereas many of these HAI-associated deaths occur among patients who are already severely ill and who have a high likelihood of death due to their underlying disease, a substantial proportion of HAI-related deaths occur among persons who were otherwise expected to survive their hospitalization. In a single-center study, 31% of unexpected in-hospital deaths were determined to be possibly or probably related to an HAI. In addition to an increased risk of death, patients who develop HAI suffer a number of other adverse outcomes, including prolonged hospital stays, additional medical interventions and antibiotic treatment, discomfort, and loss of function and income. These statistics are particularly concerning when they are considered with the knowledge that many of these infections are preventable. In fact, a systematic review found that 55 to 70% of four of the most common types of HAIs are preventable through the use of currently available, evidence-based preventive strategies (see Table 266-1). Although the majority of HAI statistics come from acute care hospitals, there are data to demonstrate that HAIs are significant problems in other health care settings as well. Point prevalence surveys conducted in European and the U.S. Veterans Affairs system long-term care facilities found that the prevalence of HAI among long-term care facility residents ranged from 2.4 to 5.2%. The overall burden of HAI among long-term care facility residents has been estimated to be 1.64 to 3.83 million infections per year in the United States and at least 2.6 million infections per year in Europe. Vascular access– related infections are the most common HAIs among patients requiring chronic hemodialysis for end-stage renal disease, with a reported incidence of 1.1 to 5.5 episodes per 1000 catheter days, and they are associated with increased morbidity, hospitalization, and death.3 The magnitude of HAIs related to care provided in other settings, such as ambulatory surgery and endoscopy centers, has not been as thoroughly studied, but such infections have been well described.

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

ABSTRACT

Health care–associated infections (HAIs) are a substantial cause of morbidity and mortality among persons receiving health care. These infections may be caused by organisms that are part of a patient’s normal microbiota (“endogenous” organisms) or by organisms acquired during the course of health care (“exogenous” organisms). Antimicrobial resistance among many of the common bacterial causes of HAIs (e.g., Klebsiella pneumoniae, Pseudomonas aeruginosa, Acinetobacter species) has become a major challenge in the treatment of HAIs and has been associated with higher rates of death and other adverse outcomes. Factors that increase the risk of HAI include disruption of normal host defenses as a result of invasive medical devices, surgical procedures, and immunosuppressive medications; alteration of the patient’s normal microbiota as can occur with receipt of antibiotics; and exposure to exogenous pathogens during the receipt of health care. Many underlying medical conditions can also predispose patients to the development of HAIs. While the epidemiology of HAIs is complex, many of these infections are the result of a failure to consistently implement basic infection prevention practices. Thus, a large proportion of HAIs could be prevented by improving adherence to currently recommended, evidence-based infection prevention strategies. These strategies include, but are not limited to, hand hygiene, use of aseptic or sterile technique during invasive procedures, safe injection practices, proper cleaning and disinfection of medical equipment and high-touch surfaces within health care facilities, prompt removal of medical devices that are no longer medically indicated, and judicious use of antibiotics.

KEYWORDS

nosocomial health care–associated infection antimicrobial resistance infection control Clostridioides difficile device-related infections prevention

1825.e3

1826

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

TABLE 266-1 ESTIMATES OF THE BURDEN, COSTS, AND PREVENTABILITY OF COMMON HEALTH CARE–ASSOCIATED INFECTIONS IN U.S. HOSPITALS NUMBER OF INFECTIONS PER YEAR†

AVERAGE ATTRIBUTABLE COST* PER INFECTION‡

CASE-FATALITY RATE§

PROPORTION PREVENTABLE‖

Urinary tract infection   Catheter-associated urinary tract infection

28,100-176,700 19,024-119,626

$749-$1007

2.3%

65-70%

Surgical site infection

50,800-281,400

$11,087-$34,670

2.8%

55%

Pneumonia   Ventilator-associated pneumonia

50,800-281,400 19,863-110,027

$14,806-$28,508

14.4%

55%

Blood stream infection   Central line–associated blood stream infection

20,700-140,200 17,388-117,768

$6461-$29,156

12.3%

65-70%

C. difficile infection

23,700-155,000

$5682-$9124

TYPE OF INFECTION

2.9%

*In 2007 U.S. dollars. † Magill SS, Edwards JR, Bamberg W, et al. Multistate point-prevalence survey of health care-associated infections. N Engl J Med. 2014;370:1198-1208. ‡ Scott RD. The direct medical costs of healthcare-associated infections in U.S. hospitals and the benefits of prevention. Centers for Disease Control and Prevention; 2009. Available at: http://www.cdc.gov/hai/ pdfs/hai/scott_costpaper.pdf. Accessed August 14, 2017. § Klevens RM, Edwards JR, Richards CL, Jr., et al. Estimating health care-associated infections and deaths in U.S. hospitals, 2002. Public Health Rep. 2007;122:160-166. ‖ Umscheid CA, Mitchell MD, Doshi JA, et al. Estimating the proportion of healthcare-associated infections that are reasonably preventable and the related mortality and costs. Infect Control Hosp Epidemiol. 2011;32:101-114.

PATHOBIOLOGY

HAIs can be caused by organisms that are a part of the patient’s normal flora (i.e., endogenous infection) or by pathogens acquired during exposure to health care (i.e., exogenous infection) through the contaminated hands of health care workers, the environment, contaminated medical equipment, other patients, or visitors. A variety of factors can contribute to the development of an HAI, and in many cases, HAIs are multifactorial in nature. These factors can be related to the pathogen, the host, the specific health care interventions that a patient receives, the setting in which health care is received, and the methods by which these interventions are made. HAI prevention strategies focus on eliminating, reducing, or modifying one or more of these risk factors.

Pathogen-Related Factors

A variety of pathogen-related factors contribute to the ability of an organism to cause infection. These factors include the organism’s normal reservoir, mode of transmission (e.g., direct or indirect contact transmission, respiratory droplets, airborne particles), ability to survive on inanimate objects and surfaces, ability to produce biofilm, virulence factors, and resistance to antimicrobial agents and, for some organisms (e.g., Clostridium difficile), disinfectants.

Host-Related Factors

Many host-specific factors are associated with an inherent increased risk of one or more types of infection, regardless of the receipt of health care; however, when a patient with one or more of these risk factors enters the health care system, these factors contribute to an increased risk of HAI. Such risk factors include age (with neonates and older adults having an increased risk of infection because of incomplete development or senescence of the immune system, respectively), obesity, smoking, severity of illness, and certain medical conditions (e.g., burns, end-stage liver or renal disease, poorly controlled diabetes, some cancers, congenital or acquired immune deficiency). These factors reflect suppression of the immune system or breaches of other normal host defense mechanisms. Whereas many of these factors are not amenable to intervention or cannot be effectively modified in the short term, interventions that address remediable risk factors (e.g., obesity, smoking, poorly controlled diabetes mellitus) have the potential to reduce the risk of HAI during future episodes of health care.

Health Care–Related Factors

Health care–related HAI risk factors are those resulting from interventions that are intended to treat or otherwise provide benefit for a patient’s existing medical conditions but that also introduce an increase in the risk of infection. These factors may disrupt normal host defenses or alter the patient’s normal microbiologic flora. Health care–related risk factors include the use of invasive devices (e.g., central venous catheters, urinary catheters, endotracheal tubes), surgical procedures, exposure to antibiotics, receipt of immunosuppressive medications, and prolonged hospitalization. Because each of these interventions poses at least some degree of increased risk of infection, the risk-to-benefit

ratio of each intervention must frequently be reassessed so that patients are not exposed to unnecessary risk. For example, central venous catheters and indwelling urinary catheters are major risk factors for primary blood stream infection and urinary tract infection, respectively. In a patient who has a true medical need for one of these devices, the benefits of the catheter exceed the risk of infection. However, once the patient recovers from the condition that necessitated the catheter, the risks associated with the device then outweigh the benefits. Exposure to antibiotics is a well-established risk factor for colonization and infection with multidrug-resistant organisms (MDROs) and development of C. difficile infection (CDI) through a mechanism known as antibiotic selection pressure. Antimicrobial use is common in acute care hospitals and other health care settings, such as long-term care and dialysis facilities and ambulatory care practices. A 2009 point prevalence survey of hospitals in 25 European countries found that 29% of hospitalized patients receive one or more antimicrobials during their hospital stay. In a 2011 study, the overall prevalence of antimicrobial use among patients in 183 U.S. hospitals was 49.9%. The prevalence of antimicrobial use, however, varied substantially among hospital locations, ranging from 2.9% in nursery wards to 77.3% in surgical critical care units. Even more important, studies have shown that 25 to 75% of antimicrobial use in acute care, long-term care, and hemodialysis facilities is unnecessary or inappropriate. Inappropriate and unnecessary antimicrobial use includes administration of antimicrobial regimens that are broader in spectrum or longer in duration than necessary, use of antibiotics that do not provide activity against the causative pathogen, treatment for test results that do not reflect the presence of infection (e.g., specimen contamination, asymptomatic colonization), use of antibacterial agents for treatment of conditions that are not due to bacterial infection (e.g., viral respiratory tract infections), and prescription of inappropriate doses of an antibiotic. Misuse and overuse of antimicrobial agents in the outpatient setting are also well-recognized problems in the United States and other countries, including many countries where antimicrobials can be obtained without a prescription. This inappropriate use of antimicrobial agents introduces unnecessary risk for the development of complications of antibiotic therapy, including CDI, MDRO infection, and toxicity, and represents an important target for intervention.

Health Care Delivery–Related Factors

This group of risk factors includes those that are introduced as a result of the way in which health care is delivered. These risk factors do not offer any potential benefit to patients but rather are associated only with risk. Health care delivery–associated risk factors include, among other things, failure to perform hand hygiene when indicated or to use aseptic or sterile technique during invasive procedures, unsafe injection practices (e.g., entering a multidose vial with a used needle), and failure to adequately clean and disinfect or sterilize the patient environment and medical equipment and instruments. These risks are all potentially modifiable and are thus important targets for HAI prevention initiatives. Antibiotic use, which has already been discussed as a patient-specific health care–related risk factor, can also be considered a health

1827

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

care delivery–related risk factor. Unlike with other types of drugs, use and misuse of antibiotics in one patient or population can introduce risks among the larger population through changes in microbial ecology (i.e., selection and increased prevalence of antimicrobial-resistant pathogens). Many of these health care delivery–related factors are the result of poor adherence to recommended, evidence-based infection prevention practices. Despite recognition that poor hand hygiene practice is a leading cause of pathogen transmission, the existence of major national and international guidelines, and initiatives to improve hand hygiene practices among health care workers, compliance with recommended hand hygiene practices among health care personnel remains unacceptably low. In the United States, average rates of health care worker compliance with recommended hand hygiene practices have been reported to be less than 50%, with some individual studies reporting rates as low as 20% in some intensive care units (ICUs). Similarly, unsafe injection practices continue to be identified as the cause of health care–related transmission of blood-borne pathogens, such as hepatitis B and C viruses. In recent years, there has been an increasing recognition of the role of environmental contamination in the transmission of health care–associated pathogens.4 Environmental contamination with these organisms is common, and many of these organisms can persist in the health care environment for prolonged periods. For example, environmental contamination with C. difficile has been detected in up to 100% of hospital rooms occupied by patients with CDI, whereas methicillin-resistant Staphylococcus aureus (MRSA) has been detected on environmental surfaces in approximately 70% of hospital rooms housing patients infected and colonized with MRSA. Some but not all studies have identified similarly high rates of environmental contamination with multidrug-resistant gram-negative pathogens. This contamination can result in patient-to-patient transmission through transient contamination of health care workers’ hands and equipment or by direct contact of the patient with the contaminated environment. Studies have shown that admission to a hospital room in which the prior occupant was colonized or infected with one of several MDROs is a significant risk factor for acquisition of that organism. Environmental contamination is, however, a potentially modifiable risk factor for HAI. Cleaning and disinfection of the environment and portable medical equipment that is shared among patients are often suboptimal. For example, one multicenter study conducted in 36 acute care hospitals in the United States found that at baseline, only 48% of high-risk environmental surfaces were cleaned during routine cleaning after discharge of the patient. Improvement in cleaning practices and other interventions to reduce the microbiologic burden in the health care environment has been shown to reduce the microbial burden of organisms in the environment and has been associated with a reduction in the risk of acquisition of MDROs and CDI. Identification of optimal strategies to reduce or eliminate the risk to patients that is posed by a contaminated health care environment is an area of active investigation, with study of methods to improve routine cleaning and disinfection as well as the study of novel methods of disinfection. A1  Recent outbreaks of infection due to contaminated medical

devices (e.g., duodenoscopes, heater-cooler devices used in cardiothoracic surgery procedures) highlight the critical role that cleaning and disinfecting or sterilizing of medical equipment plays in preventing health care–associated infections. In addition to cleaning and disinfection of environmental surfaces and disinfection or sterilization of shared medical equipment, environmental infection control interventions are important for preventing patients from acquiring pathogens due to exposure to water (e.g., Legionella species) and air (e.g., environmental fungi) within the health care setting.

Pathogens in Health Care–Associated Infections

The organisms most commonly identified in device- and procedure-associated infections (i.e., central line–associated blood stream infection, catheterassociated urinary tract infection, ventilator-associated pneumonia, and surgical site infection) vary somewhat among the different types and sites of infection. Overall, in the United States 10 pathogen groups accounted for more than 75% of pathogens identified in device- and procedure-related infections reported to the CDC through the National Healthcare Safety Network between 2011 and 2014.5 These pathogen groups and the proportion of reported pathogens that they represented include Escherichia coli (15%), Staphylococcus aureus (12%), Klebsiella species (8.0%), coagulase-negative staphylococci (8%), Enterococcus faecalis (7%), Pseudomonas aeruginosa (7%), Candida albicans (7%), Enterobacter species (4%), Enterococcus faecium (4%), and other Enterococcus species (4%). Whereas many of these pathogens represent patients’ endogenous flora, further discussion of several of the organisms that may be acquired during exposure to health care is warranted.

Multidrug-Resistant Organisms

An increasingly concerning problem is the emergence of acquired antimicrobial resistance among many of the bacterial pathogens that are common causes of HAIs (Table 266-2). MDROs represent a significant health threat because infections caused by many of these MDROs have been associated with worse outcomes than those caused by antimicrobial-susceptible strains of the same organism, including excess length of hospital stay, increased health care costs, and higher mortality, with mortality rates approaching 50% in some studies. Possible explanations for the observed increased rate of adverse outcomes associated with MDRO infections include the presence of more severe underlying disease, delays in initiating effective therapy, and the use of more toxic or less effective therapy for treatment of the infection. Regardless of their cause, the poor outcomes associated with MDRO infections highlight the critical need for effective preventive measures and development of new antibiotics with activity against these MDROs, particularly multidrug-resistant gram-negative bacilli (MDR-GNB). Organisms can develop resistance to antimicrobial agents to which they were previously susceptible through a variety of mechanisms, including induction, genetic mutation, and acquisition of new genetic material (e.g., conjugation with cell-to-cell transfer of genetic material by plasmids or transposons). In the health care setting, however, patient-to-patient transmission of MDROs

TABLE 266-2 RATES OF ANTIMICROBIAL RESISTANCE AMONG PATHOGENIC ISOLATES FROM HEALTH CARE–ASSOCIATED INFECTIONS PROPORTION OF ISOLATES RESISTANT TO ANTIBIOTIC

ANTIBIOTIC CLASS

United States (2011-2014)*

ICUs in 43 Asian, African, European, and Latin American Countries (2007-2012)†

ICUs in 13 European Countries (2014)‡

Staphylococcus aureus

Anti-staphylococcal penicillins (e.g., oxacillin, methicillin)

42-57%

36-62%

25%

Klebsiella species

Extended-spectrum cephalosporins Carbapenems

10-28% 3-13%

63-71% 14-20%

44% 8%

Pseudomonas aeruginosa

Carbapenems

8-28%

34-43%

28%

Enterococcus faecium

Glycopeptides (vancomycin)

58-86%

NR

NR

Acinetobacter baumannii

Carbapenems

33-69%

66-77%

64%

Escherichia coli

Extended-spectrum cephalosporins Carbapenems Fluoroquinolones

13-24% 1-2% 29-49%

62-66% 5-8% 64-70%

17% 1% NR

ORGANISM

ICUs, intensive care units; NR, not reported. *Weiner LM, Webb AK, Limbago B, et al. Antimicrobial-resistant pathogens associated with healthcare-associated infections: summary of data reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 2011-2014. Infect Control Hosp Epidemiol. 2016;37:1288-1301. † Rosenthal VD, Maki DG, Mehta Y, et al. International Nosocomial Infection Control Consortium (INICC) report, data summary of 43 countries, for 2007-2012. Am J Infect Control. 2014;42:942-956. ‡ European Centre for Disease Prevention and Control. Annual Epidemiological Report 2016—Healthcare-associated infections acquired in intensive care units. Stockholm: ECDC;2016. Available at: https://ecdc.europa.eu/en/publications-data/healthcare-associated-infections-acquired-intensive-care-units-annual. Accessed August 12, 2017.

1828

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

is more common than de novo development of resistance in a previously susceptible organism within the patient’s existing microbiome. Identified risk factors for acquisition of MDROs include exposure to antibiotics, frequent or prolonged exposure to health care facilities (e.g., hospitals, nursing homes), poor infection control practices among health care workers, environmental contamination with MDROs, and prevalence of MDROs among other patients within a health care facility. Among the pathogens reported to the CDC between 2011 and 2014 as causes of device-associated infections and surgical site infections, approximately 20% had antimicrobial susceptibility profiles that met the CDC’s definition of multidrug resistance (see Table 266-2). Of note, 58 to 86% of Enterococcus faecium isolates were vancomycin resistant (VRE), and 42 to 57% of S. aureus isolates were resistant to methicillin (MRSA). These multidrug-resistant grampositive pathogens have been recognized as significant health care–associated pathogens for several decades. More recently, the emergence of multidrug resistance among several gram-negative pathogens has been identified as a growing global health threat among persons receiving health care (Chapter 289). For example, approximately 55% of Acinetobacter baumannii isolates reported to the CDC in 2011-2014 demonstrated acquired resistance to at least one drug in three or more antibiotic classes. Such multidrug resistance definitions were also met by 13% of Klebsiella isolates and 14% of P. aeruginosa isolates. Antimicrobial resistance is an important problem in many regions of the world. For example, International Nosocomial Infection Control Consortium data from intensive care units in 43 countries in Asia, Africa, Europe, and Latin America collected between 2007 and 2012 demonstrated methicillin resistance in 61% of S. aureus isolates and extended-spectrum cephalosporin resistance in 66% of Klebsiella pneumoniae isolates and 63% of E. coli isolates. The lack of use of a standardized definition of multidrug resistance limits direct comparisons of resistance data from different populations. Standardized definitions for multidrug resistant, extensively drug resistant, and pandrug resistant have been proposed. Adoption of these or other standardized definitions is needed to allow a more thorough understanding of the global burden of antimicrobial resistance among health care–associated pathogens. One example of the emergence and rapid dissemination of MDR-GNB is carbapenem-resistant Enterobacteriaceae (CRE), particularly K. pneumoniae (Chapter 289). Carbapenem resistance among these organisms was rare in the United States before the year 2000, at which time less than 1% of K. pneumoniae isolates reported to the CDC demonstrated such resistance. By 2009-2010, 8 to 13% of K. pneumoniae isolates from hospital-associated infections were resistant to carbapenems. In the United States, carbapenem resistance among the Enterobacteriaceae is most commonly due to the production of K. pneumoniae carbapenemase (KPC), a class A serine β-lactamase enzyme that hydrolyzes all β-lactam antibiotics. The KPC enzyme, which was first described in 2001, is encoded by the blaKPC gene carried on a transmissible plasmid that also carries additional genes that confer resistance to several other classes of antimicrobial agents. Thus, in addition to carbapenem resistance, these organisms demonstrate resistance to other β-lactam antibiotics and to several other classes of antibiotics. Asymptomatic carriage of MDROs is relatively common among persons with health care exposures. In fact, patients with clinically apparent MDRO infections represent a relatively small proportion of the total burden of these pathogens. Reported rates of MRSA carriage have ranged from 4.6 to 13.6% among hospital patients, 2 to 22% among U.S. ambulatory dialysis patients, and 10 to 100% among residents of long-term care facilities. The prevalence of VRE carriage among hospital patients and ambulatory dialysis patients has been reported to range from 6.3 to 67% and 0 to 16%, respectively. Studies that have included a variety of MDR-GNB have reported carriage rates of 19 to 32% in ICU patients and hospital patients with diarrhea, 25% among residents of long-term care facilities, and 16% among chronic hemodialysis patients. Studies that have focused specifically on carbapenem-resistant Enterobacteriaceae have demonstrated prevalence rates ranging from 2 to 5.4% among high-risk hospital patients in the United States and 2 to 49% among post–acute care facility patients in Israel. These asymptomatic carriers play an important role in the epidemiology of MDRO infections. First, they are at substantial risk of subsequent infection with the colonizing organism, with up to one third of carriers of MRSA, VRE, and MDR-GNB developing symptomatic infection within 12 months. Second, asymptomatic MDRO carriers can contribute to MDRO transmission within the health care system through contamination of their surrounding environment and of health care workers’ hands, clothing, and medical equipment. In fact, several studies have demonstrated that the risk of acquiring an MDRO, such as MRSA, VRE, and MDR-GNB, during hospitalization is related to the prevalence, or colonization pressure,

of that MDRO among other patients. Finally, admission to a hospital room in which the prior patient was colonized or infected with the MRSA, VRE, or MDR-GNB has been associated with an increased risk of acquisition of those organisms. There are, however, some encouraging data related to the incidence of some MDROs, particularly MRSA. In the United States, the incidence of hospitalonset invasive MRSA infections decreased 54% between 2005 and 2011. During the same time period, the incidence of invasive MRSA infections among dialysis patients, a group with a rate of invasive MRSA infection that is approximately 100 times greater than that of the general population, was also observed to have significantly decreased in the United States. These changes occurred despite the emergence of community-associated MRSA as a significant cause of skin and soft tissue infections among persons without typical health care–associated risk factors and the introduction of community-associated MRSA as a health care–acquired pathogen.6 In England, an 86% reduction in the number of cases of MRSA bacteremia reported through a mandatory reporting system was observed between 2004 and 2012. The specific cause of these observed decreases in MRSA HAIs is uncertain and may be the result of improvements in basic infection control practices, introduction of specific MRSA prevention practices, or other changes in the epidemiology of this pathogen. There are also examples of successful containment of other MDROs. For example, the spread of CRE within Israeli hospitals has been substantially reduced with the introduction of a nationwide intervention.

Clostridium Difficile

C. difficile, the etiologic agent of pseudomembranous colitis, is the most common cause of health care–associated infectious diarrhea (Chapter 280). Although community-associated CDI occurs,7 most cases are associated with receipt of health care. Analysis of CDC data from 2010 found that 94% of cases of CDI were associated with health care exposure but that 75% of cases had their onset outside of the hospital (e.g., in the community or in long-term care facilities). Clinical manifestations of CDI range from asymptomatic carriage to mild diarrhea to life-threatening colitis, toxic megacolon, and sepsis. The overall mortality associated with CDI has been reported to range from 2 to 6%, with substantially higher mortality among patients who develop toxic megacolon and other severe manifestations of the disease. Compared with patients without CDI, hospital patients who develop CDI experience an extended duration of hospitalization, and approximately 15 to 30% of these patients will experience at least one recurrence of the disease, typically within 1 to 2 months of the initial episode. Persons who experience one recurrence have a 50 to 60% chance of additional recurrences. Estimates of CDI-attributable hospital costs range from $5682 to $9124 per case (in 2007 U.S. dollars), amounting to a total cost of $1 billion to $4.8 billion per year. More HAIs are now due to C. difficile than to MRSA. The incidence of CDI among hospitalized patients in the United States more than doubled between 2000 and 2009. Data from the CDC indicate that approximately 250,000 cases of C. difficile occur each year in the United States. In addition to an increased incidence of CDI, the rate of C. difficile–related deaths increased more than four-fold in the United States during the last decade, causing approximately 14,000 deaths per year. The observed increases in the incidence of and mortality associated with CDI were temporally associated with the emergence and dissemination of the 027/NAP1/BI strain of C. difficile, which has been associated with greater mortality and higher rates of recurrence than other circulating strains. Initially identified in North America, this epidemic strain has now disseminated globally. The problem of C. difficile extends beyond the acute care hospital setting. For instance, data from Ohio indicate that 62% of CDI cases in 2006 occurred in nursing homes. Data from the CDC have shown that 20% of hospital-onset CDI occurs among persons with recent residence in a nursing home and that two-thirds of nursing home–onset CDI cases occur among residents who have been recently discharged from an acute care hospital. These findings demonstrate the complex epidemiology of CDI. The development of CDI is a two-step process that first requires acquisition of C. difficile through fecal-oral transmission and then the presence or introduction of factors that allow progression to symptomatic disease. Although some healthy individuals without health care exposure are intestinal carriers of C. difficile, acquisition of the organism during contact with the health care system plays a critical role in the epidemiology of CDI. Transmission in the health care setting can occur as the result of exposure to organisms on health care workers’ hands, environmental surfaces, or medical equipment. Hand contamination of health care workers is common after contact with the skin or environment of persons with CDI.8 In the absence of effective hand hygiene

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

practices, contaminated health care workers may transmit the organism to other patients. Contamination of the environment and medical equipment with C. difficile is also common. Testing performed in hospital rooms has identified C. difficile on environmental surfaces in up to 100% of rooms housing patients with active CDI and in as many as 33% of non-CDI patient rooms. Commonly contaminated surfaces include bed rails, bedside tables, telephones, call buttons, and blood pressure cuffs. C. difficile spores are resistant to killing by many common hospital disinfectants and can persist in the environment for long times, further contributing to the risk of exposure to a contaminated environment. Patients who are admitted to a hospital room in which the previous occupant had CDI have been found to be at greater risk for development of CDI than patients admitted to hospital rooms in which the previous occupant did not have CDI. Once C. difficile has been ingested, several factors are known to be associated with development of symptomatic CDI. The major risk factor is receipt of antibiotics. Such exposures disrupt the normal intestinal flora and allow C. difficile to multiply to larger numbers and to produce toxins that result in disease. Although exposure to any antimicrobial agent may increase the risk for development of CDI, clindamycin, third-generation cephalosporins, penicillins, and fluoroquinolones may present the highest risk. Other factors that have been associated with an increased risk for development of CDI include receipt of cytotoxic chemotherapeutic agents and gastric acid suppressive medications such as proton pump inhibitors, failure to develop an antibody response to C. difficile, and older age (i.e., age older than 64 years).

Viruses Respiratory Viruses

Common respiratory viruses, such as influenza, can be transmitted in the health care setting by health care workers, visitors, and patients, resulting in health care–acquired disease. Higher rates of morbidity and mortality have been observed among those who acquire infection during hospitalization, probably due to the presence of significant underlying medical illness. Despite several studies that have associated higher influenza immunization rates of health care workers with lower rates of nosocomial influenza transmission, the uptake of influenza vaccination among health care workers remains relatively low. This has led many public health agencies and professional societies to call for mandatory influenza vaccination policies for all eligible health care workers. The health care–associated transmission of newly emerged respiratory viruses, such as the severe acute respiratory syndrome coronavirus (SARSCoV) in 2003, pandemic influenza in 2009, and the Middle East respiratory syndrome coronavirus (MERS-CoV) in 2013, highlights the importance of syndromic surveillance to allow rapid identification of patients with potentially communicable diseases and implementation of appropriate infection control precautions.

Blood-borne Viruses

Although routine screening of the blood supply for the blood-borne pathogens (BBPs) hepatitis B virus, hepatitis C virus, and human immunodeficiency virus has dramatically decreased the incidence of health care–associated BBP infections, transmission of these pathogens within health care settings continues to occur. Most health care–associated BBP transmission that occurs now is due to failure to adhere to recommended basic infection control practices. Unsafe injection practices (e.g., reuse of syringes, contamination of multidose vials, improper use and disinfection of blood glucose monitoring devices that are used for multiple patients) and inadequate cleaning, disinfection, and sterilization of medical equipment and the health care environment (e.g., dialysis facilities) have been identified in several recent outbreaks of patientto-patient transmission of BBPs. A large proportion of documented health care–associated hepatitis B virus and hepatitis C virus transmission events has occurred in outpatient settings and long-term care facilities, highlighting the importance of infection prevention programs throughout the entire health care system. Transmission of BBPs from health care worker to patient is uncommon but can occur, typically in the setting of “exposure-prone” invasive procedures but also as a result of drug diversion by health care workers. Guidelines are available to assist health care workers and health care facilities in minimizing the risk posed to patients by a BBP-infected health care worker while allowing most such health care workers to remain involved in patient care activities (E-Table 266-1).

Fungi

Candida albicans and other Candida species accounted for approximately 9.3% of all pathogens reported to the CDC between 2011 and 2014 as causes of

1829

device-associated infections and surgical site infections, placing them among the most common pathogens implicated in HAIs. The recent emergence of multidrug-resistant C. auris as a cause of health care facility–associated outbreaks further highlights the importance of Candida species in HAI epidemiology. Exposure to environmental fungi, such as Aspergillus species, in the health care setting can result in HAI, particularly in immunocompromised hosts. Such exposure and resulting infection are most commonly associated with inadequate environmental control measures during construction, demolition, or water damage within the health care facility. A multistate outbreak of invasive fungal infections in the United States, mostly due to Exserohilum rostratum associated with contaminated methylprednisolone injections, demonstrates that contaminated medications and other medical products are additional potential sources of exposure to fungal pathogens during health care.

Device-Associated Infections Central Line–Associated Blood Stream Infections

Central line–associated blood stream infections (CLABSIs) are blood stream infections that occur in patients with a central venous catheter and in whom there is no other identified source of the infection. Development of CLABSI has been associated with longer hospital stays, increased risk of death, and greater hospital costs than those observed among otherwise similar patients who do not develop CLABSI (see Table 266-1). Rates of morbidity and mortality, however, vary substantially, depending on the causative pathogen and characteristics of the patient in whom the infection occurs. Although CLABSIs are often thought of as a complication that occurs among ICU patients, the use of central venous catheters in non-ICU hospital wards has expanded substantially during the past few decades. Thus, the incidence of and the overall burden of CLABSI in some non-ICU wards now often exceeds that in ICUs. CLABSIs are also important problems among nonhospitalized persons with central venous catheters, such as persons receiving chronic total parenteral nutrition, chemotherapy, or dialysis in the outpatient or home care setting. Blood stream infections due to central venous catheters are largely the result of contamination or colonization of the external surface or the intraluminal surface of the catheter. This contamination can occur either during catheter insertion or after insertion, related to a number of aspects of catheter use and care. Effective strategies have been identified to reduce the risk of catheter contamination during insertion and throughout the time that the catheter remains in situ.9 This research has led to the development of evidence-based guidelines for prevention of vascular catheter–related infections (see E-Table 266-1). The “central line bundle” refers to a small number of evidence-based practices that, when used together, can reduce the risk of CLABSI even more than would be expected when the components are introduced individually. The central line bundle includes hand hygiene, maximal barrier precautions during insertion (i.e., use of sterile gown and gloves and a surgical cap and mask by the operator and covering the patient with a sterile, full-body drape), chlorhexidine skin antisepsis, optimal insertion site selection (i.e., avoidance of the femoral site in adult patients), and daily review of catheter necessity with immediate removal of catheters that are no longer necessary. Regarding the use of a cutaneous antiseptic prior to central venous catheter insertion, a recent randomized trial found that a chlorhexidine–alcohol skin antiseptic solution provided greater protection against short-term catheter-related infections than did a povidone iodine–alcohol solution. A2  Other interventions that have been associated with reductions in CLABSI rates include covering of the insertion site with a sterile gauze or semipermeable transparent dressing, cleansing of the catheter insertion site with an antiseptic such as chlorhexidine, use of aseptic technique to access and to manipulate the catheter, scrubbing of the catheter hub with a disinfectant before accessing the catheter lumen for administration of medications or other products or for aspiration of blood, use of antimicrobial- or antiseptic-coated catheters, and use of chlorhexidine rather than regular soap and water for daily bathing of ICU patients. Widespread adoption of the central line bundle and other CLABSI prevention strategies has been associated with a substantial reduction in the incidence of CLABSI in U.S. hospitals. Data from the CDC demonstrate a 50% decrease in the incidence of CLABSI in U.S. hospitals between 2008 and 2014.

Catheter-Associated Urinary Tract Infections

A catheter-associated urinary tract infection (CAUTI) is a urinary tract infection that develops in a patient who has or who recently had an indwelling urinary catheter. Similar to the pathogenesis of catheter-related blood stream infections, CAUTIs develop by the introduction of pathogens into the bladder as a result of contamination and colonization of the internal or external surface of the catheter. Among patients with indwelling urinary catheters, the incidence

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

1829.e1

E-TABLE 266-1 SELECT GUIDELINES AND EVIDENCE-BASED RECOMMENDATIONS FOR THE PREVENTION AND MANAGEMENT OF HEALTH CARE–ASSOCIATED INFECTIONS GENERAL TOPIC Device-associated infections

SPECIFIC TOPIC Vascular catheter–associated infections

Catheter-associated urinary tract infections

Ventilator-associated pneumonia/health care–associated pneumonia

Cardiovascular implantable electronic devices Surgical site infections

Health care– associated pathogens

GUIDELINES AND RECOMMENDATIONS Marschall J, Mermel LA, Fakih M, et al. Strategies to prevent central line-associated bloodstream infections in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol. 2014;35:753-771. O’Grady NP, Alexander M, Burns LA, et al. Guidelines for the prevention of intravascular catheter– related infections. Clin Infect Dis. 2011;52:1087-1099. Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter–related infection: 2009 update by the Infectious Diseases Society of America. Clin Infect Dis. 2009;49:1-45. Lo E, Nicolle LE, Coffin SE, et al. Strategies to prevent catheter-associated urinary tract infections in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol. 2014;35:464-479. Gould CV, Umscheid CA, Agarwal RK, et al. Guideline for prevention of catheter-associated urinary tract infections 2009. http://www.cdc.gov/infectioncontrol/guidelines/cauti/index.html Hooton TM, Bradley SF, Cardenas DD, et al. Diagnosis, prevention, and treatment of catheterassociated urinary tract infection in adults: 2009 International Clinical Practice Guidelines from the Infectious Diseases Society of America. Clin Infect Dis. 2010;50:625-663. Centers for Disease Control and Prevention. Guideline for preventing health care–associated pneumonia, 2003. MMWR Recomm Rep. 2004;53(RR-3):1-36. Klompas M, Branson R, Eichenwald EC, et al. Strategies to prevent ventilator-associated pneumonia in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol. 2014;35:915-936. Kalil AC, Metersky ML, Klompas M, et al. Management of adults with hospital-acquired and ventilator-associated pneumonia: 2016 clinical practice guidelines by the Infectious Diseases Society of America and the American Thoracic Society. Clin Infect Dis. 2016;63:e61-e111. Baddour LM, Epstein AE, Erickson CC, et al. Update on cardiovascular implantable electronic device infections and their management: a scientific statement from the American Heart Association. Circulation. 2010;121:458-477. Berríos-Torres SI, Umscheid CA, Bratzler DW, et al. Centers for Disease Control and Prevention guideline for the prevention of surgical site infection, 2017. JAMA Surg. 2017;152:784-791. Anderson DJ, Podgorny K, Berrios-Torres SI, et al. Strategies to prevent surgical site infections in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol. 2014;35:605-627. Bratzler DW, Dellinger EP, Olsen KM, et al. Clinical practice guidelines for antimicrobial prophylaxis in surgery. Am J Health Syst Pharm. 2013;70:195-283.

Clostridium difficile

Multidrug-resistant organisms Methicillin-resistant Staphylococcus aureus

Carbapenem-resistant Enterobacteriaceae Blood-borne pathogens

Mycobacterium tuberculosis

Dubberke ER, Carling P, Carrico R, et al. Strategies to prevent Clostridium difficile infections in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol. 2014;35:628-645. Cohen SH, Gerding DN, Johnson S, et al. Clinical practice guidelines for Clostridium difficile infection in adults: 2010 update by the Society for Healthcare Epidemiology of America (SHEA) and the Infectious Diseases Society of America (IDSA). Clin Infect Dis. 2010;31:431-455. Siegel JD, Rhinehart E, Jackson M, et al. Management of multidrug-resistant organisms in healthcare settings. http://www.cdc.gov/infectioncontrol/guidelines/MDRO/index.html Calfee DP, Salgado CD, Milstone AM, et al. Strategies to prevent methicillin-resistant Staphylococcus aureus transmission and infection in acute care hospitals: 2014 update. Infect Control Hosp Epidemiol. 2014;35:772-796. Liu C, Bayer A, Cosgrove SE, et al. Clinical practice guidelines by the Infectious Diseases Society of America for the treatment of methicillin-resistant Staphylococcus aureus infections in adults and children. Clin Infect Dis. 2011;52:1-38. Centers for Disease Control and Prevention. Facility guidance for control of carbapenem-resistant Enterobacteriaceae (CRE)—November 2015 update CRE toolkit. www.cdc.gov/hai/organisms/ cre/cre-toolkit/index.html Henderson DK, Dembry L, Fishman NO, et al. SHEA guideline for management of healthcare workers who are infected with hepatitis B virus, hepatitis C virus, and/or human immunodeficiency virus. Infect Control Hosp Epidemiol. 2010;31:203-232. Centers for Disease Control and Prevention. Updated CDC recommendations for the management of hepatitis B virus–infected health-care providers and students. MMWR Recomm Rep. 2012;61:1-12. Centers for Disease Control and Prevention. Guidelines for preventing the transmission of Mycobacterium tuberculosis in health-care settings, 2005. MMWR Recomm Rep. 2005;54(RR-17):1137.

Antimicrobial stewardship

Barlam TF, Cosgrove SE, Abbo LM, et al. Implementing an antibiotic stewardship program: guidelines by the Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America. Clin Infect Dis. 2016;62:e51-e77.

General infection prevention

Centers for Disease Control and Prevention. Guide to infection prevention in outpatient settings: minimum expectations for safe care. 2011. http://www.cdc.gov/hai/settings/outpatient/ outpatient-care-guidelines.html Centers for Disease Control and Prevention. Guideline for hand hygiene in health-care settings: recommendations of the Healthcare Infection Control Practices Advisory Committee and the HICPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. MMWR Recomm Rep. 2002;51(RR16):1-45. World Health Organization. WHO guidelines on hand hygiene in health care. 2009. http:// whqlibdoc.who.int/publications/2009/9789241597906_eng.pdf Siegel JD, Rhinehart E, Jackson M, Chiarello L; Healthcare Infection Control Practices Advisory Committee. 2007 guideline for isolation precautions: preventing transmission of infectious agents in healthcare settings. 2007. http://www.cdc.gov/infectioncontrol/guidelines/isolation/index.html Centers for Disease Control and Prevention. Guidelines for environmental infection control in health-care facilities. MMWR Recomm Rep. 2003;52(RR-10):1-42. Rutala WA, Weber DJ; Healthcare Infection Control Practices Advisory Committee. Guideline for disinfection and sterilization in healthcare facilities, 2008. http://www.cdc.gov/hicpac/pdf/ guidelines/Disinfection_Nov_2008.pdf

1830

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

of bacteriuria is 3 to 8% per day, and 10 to 25% of those with bacteriuria will subsequently develop symptoms consistent with a urinary tract infection. A number of basic practices, such as aseptic technique during insertion, maintenance of proper cleanliness and hygiene, securement of the catheter to avoid piston-like movement of the catheter within the urethra, and maintenance of a closed system with unobstructed flow of urine from the bladder into the collection system, are recommended to reduce the risk of CAUTI.10 It has been estimated that with implementation of such practices, 65 to 70% of CAUTIs that occur in acute care hospitals could be prevented. Several studies have demonstrated that indwelling urethral catheters are often inserted for inappropriate reasons and that many catheters that were initially inserted for an appropriate indication remain in place even after the initial indication for catheterization has resolved. This may be due in part to lack of familiarity with appropriate indications for catheter insertion or with options that exist regarding alternatives to the use of indwelling urethral catheters. Additional studies have found that physicians are often unaware that their patient has a urinary catheter. Thus, perhaps the greatest opportunity for CAUTI prevention is avoidance of unnecessary catheter insertion and prompt removal of catheters that are no longer necessary. Development of protocols that explicitly define appropriate indications for insertion of urinary catheters, introduction of interventions that remind clinicians to reassess the appropriateness of a patient’s urinary catheter, and nurse-driven protocols that allow nurses to remove unnecessary urinary catheters have been associated with reduced catheter use and lower rates of CAUTI.

Ventilator-Associated Pneumonia

In hospitalized patients, mechanical ventilation is one of the most common risk factors for the development of pneumonia. Mechanical ventilation and the interventions required to provide mechanical ventilation (e.g., endotracheal intubation, sedation) increase the risk of pulmonary infection through a variety of mechanisms, including an increased risk of aspiration of oropharyngeal and gastrointestinal secretions and impairment of the cough reflex. From both a clinical and epidemiologic standpoint, ventilator-associated pneumonia (VAP) is a difficult diagnosis to establish with certainty because of the subjective nature of many of the variables considered (e.g., chest radiograph findings, changes in the characteristics of respiratory tract secretions), alternative explanations for clinical and radiographic abnormalities (e.g., acute respiratory distress syndrome, atelectasis), and difficulty in determining whether the results of respiratory tract cultures represent true infection or colonization of the airway. To minimize subjectivity involved in making the diagnosis of VAP, a new surveillance classification has been proposed that characterizes all adverse ventilator-associated events (VAE) to replace the current VAP definition.11 Based on a point prevalence study, pneumonia is one of the two most common types of HAI and approximately 39% of cases are associated with mechanical ventilation. These infections are associated with mortality rates and health care costs that are among the highest observed among all HAIs (see Table 266-1). Studies suggest that at least 55% of VAP cases are preventable. As with other device-associated infections, avoiding the use of the device is the most effective means of preventing infection. For preventing VAP and other complications of mechanical ventilation, the use of noninvasive methods of ventilation, minimized use of sedation (e.g., manage patients without sedative if possible, daily sedation interruption and assessment of readiness to be extubated), and early mobilization can eliminate or at least reduce the duration of mechanical ventilation. For patients who do require mechanical ventilation, the following are routinely recommended for VAP prevention: proper cleaning, disinfection, sterilization, use, and maintenance of respiratory equipment; minimization of pooling of secretions above the endotracheal tube cuff; and elevation of the head of the bed (unless it is medically contraindicated). Routine oral hygiene, typically with an antiseptic such as chlorhexidine, is also a commonly used strategy that may be beneficial, but supporting data are limited. In some randomized, controlled trials, selective oropharyngeal or digestive decontamination has been associated with a significant reduction in VAP, but this approach has not yet been widely adopted as a standard of care in the United States, at least in part because of concerns that this intervention may lead to selection of antimicrobial-resistant organisms.

Other Device-Related Infections

With the advances that have occurred in medical technology in recent years, there have been substantial improvements in the capabilities of existing medical devices and the development of new implantable devices to treat and to manage a variety of medical conditions, particularly of the cardiovascular and nervous systems. These devices include cardiovascular implantable electronic devices

such as pacemakers and implantable cardioverter-defibrillators, ventricular assist devices, deep brain stimulators, and intrathecal pumps. The potential benefits that these devices may offer to patients come with at least some degree of risk of device-related infection.

Surgical Site Infections

Surgical site infections (SSIs) are infections that develop at the site of an operative procedure. Although only a relatively small proportion of patients who undergo surgery subsequently develop SSI and the overall mortality rate associated with SSI is relatively low, the absolute number of SSIs that occur and the overall cost and burden of morbidity and mortality associated with SSI are large because of the volume of surgical procedures performed each year (see Table 266-1). A number of factors contribute to the risk of SSI: the specific site and type of surgical procedure; duration of the procedure; tissue hypoxia at the surgical site; wound contamination with endogenous or exogenous organisms; surgical technique; perioperative infection prevention practices; environmental controls related to temperature, humidity, and air purity within the operating room; and the patient’s underlying medical conditions, smoking status, and other factors that may increase the inherent risk of infection. Although some of these factors are not amenable to corrective intervention, many of them are, and a number of interventions have been proven to reduce the risk of SSI. In fact, it has been estimated that 55% of SSIs could be prevented by routine application of evidence-based preventive measures. The use of sterile technique, preoperative skin preparation with an antiseptic agent, A3  administration of antimicrobial prophylaxis within 60 minutes before the surgical incision, maintenance of normothermia, and perioperative glucose control are practices that have been demonstrated to reduce the risk of infection associated with a wide variety of surgical procedures. There is also evidence that at least in some types of surgeries, the use of supplemental oxygen (e.g., 80% fraction of inspired oxygen) in the perioperative period may reduce the risk of SSI. Although substantial opportunities for improvement remain, surveillance data reported to the CDC’s National Healthcare Safety Network by U.S. hospitals between 2008 and 2014 showed a 17% reduction in the incidence of SSI occurring in association with 10 common surgical procedures.

HEALTH CARE–ASSOCIATED INFECTION PREVENTION STRATEGIES

Many HAIs are preventable. This has been recognized since at least as early as the mid-1800s, when Semmelweiss demonstrated a dramatic reduction in sepsis-related deaths among maternity ward patients after an intervention to improve hand hygiene among health care workers. The degree to which HAIs are preventable was more clearly quantified in a multicenter study conducted in the 1970s. In that study, it was found that 32% of HAIs could be prevented through establishment of an effective infection control program. More recent estimates that take into account newer research and the technology that has been developed during the following three decades suggest that at least 55 to 70% of some of the most common HAIs can be prevented. Scientific research and clinical experience have led to the publication of evidence-based guidelines for the prevention of HAIs (see E-Table 266-1).

Antimicrobial Stewardship

Antimicrobial stewardship refers to interventions designed to improve the appropriateness of antimicrobial use by promoting the selection of an optimal antimicrobial regimen (i.e., the most appropriate drug, dose, duration, and route of administration) for a specific patient. The goals of an antimicrobial stewardship program are to optimize clinical outcomes (e.g., cure of infection related to antimicrobial use, minimize toxicity and other adverse events) and to limit the antimicrobial selection pressure that drives the emergence of antimicrobial-resistant strains. Multidisciplinary antimicrobial stewardship programs have been associated with several desirable outcomes, including significant reductions in antimicrobial use, reduced rates of antimicrobial resistance among health care–associated pathogens, reduced incidence of adverse outcomes associated with antibiotic use (e.g., toxicity, C. difficile infection), and significant reductions in hospital antimicrobial-associated costs. A variety of approaches have been used by successful antimicrobial stewardship programs, and guidelines describing these strategies have been published (see E-Table 266-1).12,13 One of the most commonly used and most effective strategies includes formulary restriction and requirement of preauthorization before prescribing certain antibiotics (e.g., antibiotics that are broad in their antimicrobial spectrum, are associated with significant toxicity, or are expensive). A second approach that has been considered to be a core strategy for antimicrobial stewardship activities is prospective audit of the appropriateness

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

Strategies to prevent HAI are designed to interrupt one or more of the factors associated with pathogen transmission or the development of infection due to endogenous or exogenous organisms. HAI prevention strategies are sometimes described by one of two categorization schemes that classify the intervention on the basis of either the applicability of the intervention for the prevention of a variety of HAIs or the priority an intervention should be given with regard to the overall HAI prevention program within a health care facility. The first categorization scheme classifies interventions as either “horizontal” or “vertical” infection prevention strategies. Horizontal prevention strategies are those that have the potential to prevent a wide variety of HAIs or to prevent transmission of a variety of pathogens. Vertical strategies, on the other hand, are those that target a specific pathogen. Examples of horizontal prevention strategies include hand hygiene, education of health care workers, environmental cleaning and disinfection, antimicrobial stewardship, and use of aseptic technique for invasive procedures. Vertical or pathogen-specific prevention strategies include the use of interventions that are applied to patients known or suspected to be colonized or infected with a specific pathogen to prevent transmission of that pathogen. Preventing its transmission could involve interrupting its normal routes of transmission (e.g., contact, droplet, airborne), active surveillance testing for the pathogen (e.g., MRSA) to detect asymptomatic carriers who may serve as unrecognized reservoirs for its transmission, targeted decolonization therapy for persons identified as carriers of the pathogen of interest (e.g., MRSA), and vaccination programs (e.g., health care worker influenza vaccination campaigns).

1830.e1

The second categorization scheme classifies interventions as either basic or advanced (or special) preventive practices. Basic practices are those that should routinely be implemented in all health care facilities. Basic practices for acute care hospitals include most of the horizontal approaches described before as well as a few vertical approaches, such as transmission-based precautions. Special or advanced interventions are additional interventions that a facility may implement when basic practices fail to adequately control or prevent HAIs or transmission of one or more specific pathogens within the facility. This approach is consistent with recommendations and guidelines from the CDC and several professional societies that promote a tiered approach to the implementation of various HAI prevention strategies (see E-Table 266-1). The selection of specific strategies to be implemented within a specific health care facility should be based on an assessment of risk, local data and epidemiology, scientific evidence and guidelines, regulatory and accreditation requirements, and facility-specific cost-benefit calculations. Many of the prevention strategies that have been proven effective for prevention of deviceassociated infections and SSIs have already been discussed. Several additional preventive strategies are discussed in more detail here.

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

of prescribed antimicrobial therapy with provision of feedback to the prescribing clinician if opportunities for further optimization of therapy are available (e.g., narrowing or broadening spectrum of therapy, discontinuing antimicrobial therapy, or altering drug dose or dosing interval on the basis of available clinical data). Additional approaches that have been included in successful antimicrobial stewardship programs include education, development of guidelines and clinical pathways, computer-assisted decision support, and protocols to optimize conversion from the parenteral to the oral route of administration when appropriate. In addition to the standard challenges associated with implementation of interventions that require a change in human behavior and clinical practice, antimicrobial stewardship programs must also address the complex and constantly changing problems and issues associated with antimicrobial resistance. Although an increasing proportion of acute care hospitals and long-term care facilities have introduced antimicrobial stewardship programs, many of these programs are not adequately resourced to reach their full potential. There is also a recognized need for development of such programs in other health care settings, such as long-term acute care hospitals, dialysis facilities, and outpatient practices.

Decolonization Therapy

Decolonization refers to the administration or application of antimicrobial or antiseptic agents to a person to eliminate or to reduce the burden of carriage of one or more pathogens. For example, selective oropharyngeal or digestive decontamination has been used for prevention of VAP and SSI after colorectal surgery, respectively. More recently, topical decolonization has been studied as a horizontal intervention for prevention of a variety of HAIs and prevention of transmission of a variety of pathogens. Several quasi-experimental, before-after studies have associated the use of chlorhexidine, compared with nonantimicrobial soap, for daily bathing of patients with significant reductions in rates of blood stream infections, including CLABSI, acquisition of MDROs, blood culture contamination, and contamination of the environment and health care workers. Thus far, two higher quality studies of daily chlorhexidine bathing have been published. A multicenter, cluster-randomized trial conducted in eight adult ICUs and one bone marrow transplant unit in the United States found that daily bathing with chlorhexidine, compared with nonmedicated soap, was associated with a 23% reduction in the combined outcome of MRSA and VRE acquisition. A4  A similar study in pediatric ICUs showed a significant reduction in the incidence of bacteremia in the per-protocol analysis, although the difference observed in the intention-to-treat analysis did not reach statistical significance. A5  A third cluster-randomized trial conducted in 74 adult ICUs in the United States found that providing all ICU patients with decolonization therapy that consisted of intranasal application of mupirocin for 5 days and daily chlorhexidine bathing significantly reduced the MRSA-positive clinical cultures attributable to the ICU by 37% and was associated with a lower incidence of all-cause blood stream infections compared with the use of active surveillance and contact precautions for MRSA-colonized patients without decolonization therapy. A6  More recently published, single-center, randomized studies have yielded differing results. In one study, daily bathing with chlorhexidine was not associated with a reduction in the incidence of health care– associated infections in adult intensive care unit patients, A7  while in another, every other day bathing with chlorhexidine was associated with a significant reduction in hospital-acquired infections. A8  Thus, the role of routine chlorhexidine bathing as an HAI prevention strategy remains an unresolved issue. The inconsistent results obtained from clinical trials suggest that factors such as compliance and bathing technique as well as hospital-specific factors related to the epidemiology of HAI may influence the outcomes observed at the individual hospital and individual patient levels. In another cluster-randomized trial conducted in ICUs, selective decontamination of the digestive tract (SDD) and selective oropharyngeal decontamination (SOD) were compared for their impact on the prevalence of antimicrobial-resistant gram-negative bacteria, mortality, ICU-acquired bacteremia, and length of ICU stay. SDD was associated with a lower incidence of ICU-acquired bacteremia but was also associated with increased carriage of aminoglycoside-resistant gram-negative bacteria, thus raising concerns about long-term antimicrobial resistance risks with use of this strategy. A9  Postdischarge MRSA decolonization with chlorhexidine and mupirocin can reduce subsequent MRSA infection by 30%. A10 

Active Surveillance Testing

Active surveillance testing identifies asymptomatic carriers of a pathogen of interest (e.g., MRSA, VRE, MDR-GNB) with the intention to introduce additional interventions for identified carriers to prevent infection in the carrier

1831

or transmission to others. The interventions that may be applied to the identified carriers include transmission-based precautions (e.g., contact precautions), decolonization therapy (mostly applicable to S. aureus), and altered antimicrobial therapy (e.g., surgical antimicrobial prophylaxis). The role of active surveillance has long been the subject of debate and investigation. It is commonly used in outbreak control efforts in conjunction with other interventions. In the non-outbreak setting, there are numerous reports of use of active surveillance as part of a comprehensive program to reduce transmission of or infection with MDROs in individual hospitals, large hospital systems, and health care facilities within specific geographic regions, such as a number of northern European countries. A cluster-randomized trial conducted in U.S. ICUs found that there was not a significant difference in the incidence of colonization or infection with MRSA and VRE in ICUs that performed active surveillance testing with introduction of contact precautions for patients found to be carriers of MRSA or VRE compared with ICUs that did not conduct active surveillance testing. A11  However, there was a delay in reporting the results of surveillance testing that was longer than would be anticipated in normal clinical practice, which may limit the ability to generalize the study findings to all settings. Routine use of gowns and gloves for all patients is another intervention that has been used for its potential ability to reduce the risk of transmission of pathogens between patients by reducing contamination of health care workers’ hands and clothing. A cluster-randomized study found that universal use of gowns and gloves for all patients in intensive care units was not associated with a reduction in the primary outcome of acquisition of either MRSA or VRE. There was, however, a significantly lower incidence of MRSA acquisition, a secondary study outcome, associated with universal gown and glove use. A12  Thus, the utility of this strategy for preventing pathogen transmission remains uncertain.

GOALS AND INCENTIVES FOR HEALTH CARE– ASSOCIATED INFECTION PREVENTION

In addition to the obvious altruistic incentive of preventing patient harm, there are a number of additional incentives for health care providers and health care facilities to prevent HAIs. In the United States, the amount of attention given to HAIs and their prevention has greatly increased since the 1999 release of the Institute of Medicine report To Err Is Human: Building a Safer Health System. With this attention has come increasing pressure from the public, regulatory and accreditation agencies, and health care payers for the health care industry to improve its efforts and outcomes related to HAI prevention. The Centers for Medicare and Medicaid Services (CMS) Conditions of Participation list specific requirements for infection prevention and control processes in health care facilities that receive funding from CMS. Hospitals found to be deficient in the implementation of the requirements are at risk of losing the funding that they receive from the agency. Similarly, hospital accreditation agencies such as The Joint Commission have established standards for infection prevention programs and periodically survey hospitals to determine if such programs are in place. In addition, public reporting of hospital-specific HAI data by CMS and other agencies provides further incentive for hospitals to optimize their HAI prevention efforts.

  Grade A References A1. Anderson DJ, Chen LF, Weber DJ, et al. Enhanced terminal room disinfection and acquisition and infection caused by multidrug-resistant organisms and Clostridium difficile (the Benefits of Enhanced Terminal Room Disinfection study): a cluster-randomized, multicenter, crossover study. Lancet. 2017;389:805-814. A2. Mimoz O, Lucet JC, Kerforne T, et al. Skin antisepsis with chlorhexidine-alcohol versus povidone iodine-alcohol, with and without skin scrubbing, for prevention of intravascular-catheter-related infection (CLEAN): an open-label, multicentre, randomised, controlled, two-by-two factorial trial. Lancet. 2015;386:2069-2077. A3. Tuuli MG, Liu J, Stout MJ, et al. A randomized trial comparing skin antiseptic agents at cesarean delivery. N Engl J Med. 2016;374:647-655. A4. Climo MW, Yokoe DS, Warren DK, et al. Effect of daily chlorhexidine bathing on hospital-acquired infection. N Engl J Med. 2013;368:533-542. A5. Milstone AM, Elward A, Song X, et al. Daily chlorhexidine bathing to reduce bacteraemia in critically ill children: a multicentre, cluster-randomised, crossover trial. Lancet. 2013;381:1099-1106. A6. Huang SS, Septimus E, Kleinman K, et al. Targeted versus universal decolonization to prevent ICU infection. N Engl J Med. 2013;368:2255-2265. A7. Noto MJ, Domenico HJ, Byrne DW, et al. Chlorhexidine bathing and health care-associated infections: a randomized clinical trial. JAMA. 2015;313:369-378. A8. Swan JT, Ashton CM, Bui LN, et al. Effect of chlorhexidine bathing every other day on prevention of hospital-acquired infections in the surgical ICU: a single-center, randomized controlled trial. Crit Care Med. 2016;44:1822-1832.

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

As previously mentioned in the discussion of decolonization therapy, a more recently published cluster-randomized trial found that providing universal decolonization to all ICU patients without the use of active surveillance testing was associated with a significantly greater reduction in MRSA clinical cultures attributable to the ICU than was active surveillance testing with isolation of MRSA-positive patients. Although data from these cluster-randomized trials of active surveillance testing and the emergence of data supporting the use of horizontal measures such as universal decolonization therapy have contributed substantial new data regarding the control of multidrug-resistant gram-positive pathogens (e.g., MRSA and VRE) in the ICU in non-outbreak situations, additional study of these interventions in other settings and for other pathogens, such as MDR-GNB, is needed.

Interfacility and Regional Collaboration

Given the frequency with which patients are shared among hospitals and other health care settings within geographic regions and the evidence of

1831.e1

interfacility patient transfer as a risk factor of introduction of MDROs and C. difficile into a health care facility, implementation of comprehensive prevention programs in all facilities within a region may be more effective in preventing HAIs and MDRO transmission than implementation of the same interventions in a single health care institution. The establishment of regional infection prevention initiatives offers several potential benefits, including the ability to more effectively address the complex epidemiology of HAIs that involve multiple health care facilities (e.g., MDRO acquisition, transmission, and infection) than single-institution interventions and opportunities for collaboration and sharing of information, experiences, and resources. A number of successful regional and national initiatives for the prevention of HAI, including CLABSI, dialysis-related blood stream infections, MRSA, VRE, CRE, and C. difficile, have been described.

A9. Oostdijk EA, Kesecioglu J, Schultz MJ, et al. Effects of decontamination of the oropharynx and intestinal tract on antibiotic resistance in ICUs: a randomized clinical trial. JAMA. 2014;312:1429-1437. A10. Huang SS, Singh R, McKinnell JA, et al. Decolonization to reduce postdischarge infection risk among MRSA carriers. N Engl J Med. 2019;380:638-650. A11. Huskins W, Huckabee C, O’Grady N, et al. Intervention to reduce transmission of resistant bacteria in intensive care. N Engl J Med. 2011;364:1407-1418. A12. Harris AD, Pineles L, Belton B, et al. Universal glove and gown use and acquisition of antibioticresistant bacteria in the ICU: a randomized trial. JAMA. 2013;310:1571-1580.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

GENERAL REFERENCES 1. Magill SS, Dumyati G, Ray SM, et al. Evaluating epidemiology and improving surveillance of infections associated with health care, United States. Emerg Infect Dis. 2015;21:1537-1542. 2. Magill SS, O’Leary E, Janelle SJ, et al. Changes in prevalence of health care-associated infections in U.S. hospitals. N Engl J Med. 2018;379:1732-1744. 3. Miller LM, Clark E, Dipchand C, et al. Hemodialysis tunneled catheter-related infections. Can J Kidney Health Dis. 2016;3:1-10. 4. Cobrado L, Silva-Dias A, Azevedo MM, et al. High-touch surfaces: microbial neighbours at hand. Eur J Clin Microbiol Infect Dis. 2017;36:2053-2062. 5. Weiner LM, Webb AK, Limbago B, et al. Antimicrobial-resistant pathogens associated with healthcareassociated infections: summary of data reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 2011-2014. Infect Control Hosp Epidemiol. 2016;37: 1288-1301. 6. Hassoun A, Linden PK, Friedman B. Incidence, prevalence, and management of MRSA bacteremia across patient populations—a review of recent developments in MRSA management and treatment. Crit Care. 2017;21:211.

1832.e1

7. Balsells E, Shi T, Leese C, et al. Global burden of Clostridium difficile infections: a systematic review and meta-analysis. J Glob Health. 2019;9:1-20. 8. Jullian-Desayes I, Landelle C, Mallaret MR, et al. Clostridium difficile contamination of health care workers’ hands and its potential contribution to the spread of infection: review of the literature. Am J Infect Control. 2017;45:51-58. 9. Bell T, O’Grady NP. Prevention of central line-associated bloodstream infections. Infect Dis Clin North Am. 2017;31:551-559. 10. Saint S, Greene MT, Krein SL, et al. A program to prevent catheter-associated urinary tract infection in acute care. N Engl J Med. 2016;374:2111-2119. 11. Spalding MC, Cripps MW, Minshall CT. Ventilator-associated pneumonia: new definitions. Crit Care Clin. 2017;33:277-292. 12. Goff DA, Kullar R, Bauer KA, et al. Eight habits of highly effective antimicrobial stewardship programs to meet the Joint Commission standards for hospitals. Clin Infect Dis. 2017;64:1134-1139. 13. Rennert-May E, Chew DS, Conly J, et al. Clinical practice guidelines for creating an acute care hospitalbased antimicrobial stewardship program: a systematic review. Am J Infect Control. 2019. [Epub ahead of print.]

1832.e2

CHAPTER 266  Prevention and Control of Health Care–Associated Infections  

REVIEW QUESTIONS 1. What percentage of device- and procedure-associated health care– associated infections that occur in U.S. hospitals are potentially preventable? A . 0-15% B. 20-30% C. 40-50% D. 55-70% E. 85-100% Answer: D  A systematic review of published studies found that 55 to 70% of central line–associated blood stream infections, catheter-associated urinary tract infections, ventilator-associated pneumonia, and surgical site infections are preventable with the use of currently available evidence-based prevention strategies and technologies. Unfortunately, many of these strategies are not consistently implemented. 2. Bathing of intensive care unit (ICU) patients with chlorhexidine on a daily basis has been associated with which of the following outcomes? A . Reduction in MRSA-positive clinical cultures attributable to the ICU B. Reduction in overall blood stream infection rates in ICU patients C. Reduction in central line–associated blood stream infection rates in ICU patients D. Reduction in environmental contamination with VRE E. All of the above Answer: E  Several quasi-experimental studies and three cluster-randomized controlled trials have demonstrated the potential for daily bathing with chlorhexidine to reduce a variety of health care–associated infection outcomes in ICU patients. This intervention has generally been well tolerated. Further study of the risk of development of chlorhexidine resistance among common health care–associated pathogens as the use of chlorhexidine becomes more common is warranted. Daily chlorhexidine bathing in other settings has not been subject to such extensive study, and thus the role of the horizontal infection control intervention in the non-ICU setting is not yet known. 3. The increased rates of morbidity and mortality associated with health care–associated infections caused by multidrug-resistant organisms (MDROs), compared with those caused by susceptible organisms of the same species, are likely due to all except which of the following? A . Greater virulence of antimicrobial-resistant organisms compared with antimicrobial-susceptible organisms B. Delays in initiating effective antimicrobial therapy C. The presence of more severe underlying disease in patients who develop MDRO infections D. The availability of only less effective or more toxic antimicrobial agents for treatment of some MDRO infections compared with agents available for the treatment of more susceptible organisms Answer: A  The higher rates of morbidity and mortality associated with MDRO infections, compared with infections caused by susceptible organisms of the same species, are likely to be multifactorial in nature. In general, however, MDROs are not more virulent than susceptible organisms of the same species.

4. Which of the following scenarios does not represent an avoidable health care–associated risk factor for health care–associated infection or MDRO acquisition? A . The presence of an indwelling urinary catheter in an otherwise healthy patient who underwent elective knee replacement surgery 2 days earlier B. Insertion of a central venous catheter for administration of cytotoxic chemotherapy C. Insertion of an indwelling urinary catheter to reduce the need for nursing assistance in a 73-year-old patient with urinary incontinence D. Administration of antimicrobial prophylaxis for 48 hours after an elective colectomy E. Continuation of empirically prescribed piperacillin-tazobactam therapy for a patient admitted in septic shock who has now demonstrated a clinical response to current therapy and from whom pan-susceptible E. coli has been isolated from blood and urine specimens obtained at the time of admission Answer: B  The appropriateness of indwelling invasive devices and ongoing antimicrobial therapy should be assessed on a regular (i.e., at least once daily) basis with discontinuation of the device when an appropriate indication is not present and adjustment of antimicrobial therapy based on the results of diagnostic testing and clinical assessments. 5. Recommendations for the prevention of central line–associated blood stream infections (CLABSIs) include all of the following except: A . Use of maximal barrier precautions during catheter insertion (e.g., use of a cap, mask, sterile gown, sterile gloves, and sterile full-body drape) B. Routine replacement of central venous catheters every 7 days C. Preparation of the catheter insertion site with an antiseptic agent (e.g., >0.5% chlorhexidine preparation with alcohol) before catheter insertion D. Daily review of catheter necessity with immediate removal of catheters that are no longer necessary E. Scrubbing of the access port with an antiseptic before accessing the port Answer: B  Currently available evidence-based guidelines for prevention of CLABSI provide recommendations for catheter insertion, use, and maintenance. Routine catheter replacement is not a recommended strategy for CLABSI prevention.

1832

CHAPTER 267  Approach to the Patient with Suspected Enteric Infection  

of traveler’s diarrhea worldwide are the diarrheagenic Escherichia coli enterotoxigenic E. coli (ETEC) and enteroaggregative E. coli (EAEC). The invasive bacteria (Shigella, Salmonella, and Campylobacter species) cause diarrhea among travelers to all regions but are more common in Asia. Infection with Cyclospora species should be suspected when persistent or recurrent diarrhea follows

TABLE 267-1 EPIDEMIOLOGIC FEATURES IMPORTANT IN DETERMINING POTENTIAL CAUSE OF ENTERIC INFECTION IN A PERSON WITH DIARRHEA EPIDEMIOLOGIC FEATURE

267  APPROACH TO THE PATIENT WITH SUSPECTED ENTERIC INFECTION HERBERT L. DUPONT AND PABLO C. OKHUYSEN



EPIDEMIOLOGY

Enteric infections are second only to respiratory tract infections as common infectious medical problems. In certain populations, enteric infections are hyperendemic: poorly nourished infants living in developing tropical countries showing excessive rates of mortality; infants in certain daycare centers; unhygienic residents of custodial institutions for persons with severe intellectual disabilities; immunosuppressed persons; and visitors from industrialized areas to developing regions with traveler’s diarrhea.  

Giardia spp

Travel to Russia (especially St. Petersburg)

Cryptosporidium, Giardia spp

Travel to Nepal

Cyclospora spp

Travel to the developing tropical/ semitropical world from an industrialized region

Enterotoxigenic Escherichia coli, enteroaggregative E. coli; Shigella, Campylobacter, Salmonella spp; other bacterial causes; Giardia, Cyclospora, Cryptosporidium spp and noroviruses

Presence of associated cases (an outbreak)

Use incubation period and clinical features to determine probable cause

Antibiotic, chemotherapy, or proton pump inhibitor use in the past 2 months, particularly with a history of recent or current hospitalization

Clostridium difficile

Contact with daycare centers

Any enteropathogen, often the low-dose organisms: Giardia, Cryptosporidium, Shigella spp or viral pathogens

Anal-genital, oral-anal, or digital-anal contact

Any organism spread by fecal-oral route; in those with proctitis, suspect Neisseria gonorrhoeae, Chlamydia trachomatis, herpes simplex, or Treponema pallidum

Immunosuppressed person

Any agent, especially C. difficile, norovirus, Cryptosporidium, Cyclospora, Cystoisospora, Shigella, and Salmonella spp; C. jejuni, Mycobacterium avium-intracellulare, microsporidia, herpes simplex virus, and cytomegalovirus

Recent or current cruise ship travel

Norovirus, less frequently enterotoxigenic E. coli

ETIOLOGY

In approaching a patient with an enteric infection, epidemiologic (Table 267-1) and clinical (Table 267-2) features suggest a potential etiologic agent responsible for illness and a plan for evaluation (Table 267-3) and management (Table 267-4).1 Recent travel (Chapter 270) to mountainous regions or recreational lakes of North America should raise the suspicion of infection caused by Giardia species.2 When diarrhea occurs during or after travel to a developing tropical region, a bacterial enteropathogen should be suspected.3 The leading causes

ETIOLOGIC AGENT TO SUSPECT

Travel to mountainous areas of North America

TABLE 267-2 CLINICAL FEATURES OF ENTERIC INFECTION CLINICAL SYNDROME

ETIOLOGIC AGENTS SUSPECTED

SPECIAL CONSIDERATIONS

Sustained fever, often with systemic toxicity (enteric or typhoid fever)

Salmonella typhi, nontyphoid Salmonella spp, Campylobacter spp, Shigella spp, Yersinia enterocolitica

Stool and blood cultures; empirical antibiotics generally indicated. For this and other syndromes, multiplexed nucleic acid amplification tests are accurate and provide fast turn around. Confirmation with culture and sensitivity testing is recommended for bacterial agents of diarrhea

Acute watery (secretory) diarrhea

Any agent. Consider Vibrio cholerae (if water losses are major), enterotoxigenic or enteroaggregative Escherichia coli, Shigella spp, Salmonella spp, Campylobacter jejuni, viral or parasitic protozoal pathogen

Fluid and electrolyte therapy crucial for recovery in dehydration

Recurrent vomiting (gastroenteritis)

Viral agents (rotavirus or noroviruses) or preformed toxin (Staphylococcus aureus or Bacillus cereus)

In case of an outbreak, incubation period suggests the etiology

Bloody diarrhea (dysentery)

Shigella spp, C. jejuni, Salmonella spp, Shiga toxin– producing E. coli (e.g., O157:H7 or other serotype) or enteroinvasive E. coli, Aeromonas hydrophila, noncholera Vibrio spp, Yersinia enterocolitica, Entamoeba histolytica, or inflammatory bowel disease

Stool culture and occasionally parasite examination important to determining cause; hemolytic-uremic syndrome may complicate diarrheal disease caused by Shiga toxin–producing E. coli or rarely Shigella dysenteriae

Diarrhea lasting ≥2 weeks (persistent diarrhea)

Giardia spp and other protozoal parasites, bacterial overgrowth, Tropheryma whipplei, lactase deficiency, Brainerd diarrhea, postinfectious irritable bowel syndrome (PI-IBS), unmasked inflammatory bowel disease (IBD), or celiac sprue. Rarely due to conventional agents of bacterial diarrhea

Stool culture and parasite examination indicated; empirical anti-Giardia therapy may be useful; remove milk from diet; prior raw milk or untreated (well or surface) water consumption or international travel may predispose to Brainerd diarrhea; with illness lasting >30 days, consider Brainerd diarrhea, PI-IBS, celiac disease, or IBD

CHAPTER 267  Approach to the Patient with Suspected Enteric Infection  

ABSTRACT

Enteric infections are a common medical problem and may present as gastroenteritis, secretory diarrhea, dysentery, enteric fever, or persistent diarrhea (lasting ≥15 days). Patients with chronic diarrhea (lasting ≥4 to 6 weeks) seldom have an infectious disease as the underlying cause. Clinical and epidemiologic features help in developing a plan for evaluation and management. Noroviruses are the major cause of foodborne gastroenteritis worldwide. Enterotoxigenic and enteroaggregative Escherichia coli are the most common causes of traveler’s diarrhea. Travel to mountainous areas is a risk factor for giardiasis. Recreational water parks are associated with cryptosporidiosis. Health care exposure and antibiotic therapy are risk factors for Clostridium difficile, although many cases are now acquired in the community with no antibiotic exposure. Fecal leukocytes are of value in identifying colonic inflammation and suggest infection with shigella, salmonella, campylobacter, or shiga toxin–producing E. coli or C. difficile. A two-step procedure (glutamate dehydrogenase followed by enzyme immunoassay) is recommended for the diagnosis of C. difficile infection. Sensitive nucleic acid amplification tests (NAATs) lack specificity as stand-alone tests. Multiplexed NAATs can assist in the diagnosis of enteric infections but do not provide information on antimicrobial susceptibility or organism relatedness (clonality) that may assist in antibiotic selection and outbreak investigations and may produce false-positive results in colonized patients. Until NAATs are fully validated, they should be used alongside traditional culture and immunoassays. Rehydration and symptomatic drugs are preferred for afebrile, nondysenteric diarrhea. Antibiotic therapy is reserved for enteric fever, severe dysenteric disease, and severe cases of traveler’s diarrhea.

KEYWORDS

enteric infection enteric pathogen traveler’s diarrhea gastroenteritis dysentery infectious diarrhea norovirus

1832.e3

CHAPTER 267  Approach to the Patient with Suspected Enteric Infection  

1833

TABLE 267-3 LABORATORY TESTS AND PROCEDURES USEFUL IN THE DIAGNOSIS OF INFECTIOUS DIARRHEA SPECIFIC TEST OR PROCEDURE

WHEN INDICATED

CLINICAL SIGNIFICANCE

Fecal leukocyte test

For moderate to severe cases

When present, indicates diffuse colonic inflammation, often due to Shigella, Salmonella, Campylobacter spp, Shiga toxin–producing Escherichia coli, or Clostridium difficile

Fecal lactoferrin

For moderate to severe cases to help identify inflammatory forms of enteric infection, to use in health care–associated diarrhea to help determine whether C. difficile toxin test should be performed

More sensitive test than fecal leukocytes and will pick up the same pathogens as fecal leukocytes but also pathogens associated with less striking degrees of inflammation (enteroaggregative E. coli and C. difficile)

C. difficile toxins A and B

Diarrhea associated with use of antibiotics, chemotherapy, or proton pump inhibitors, especially associated with current or recent hospitalization. Cases of community acquisition without health care exposure are increasingly being reported.

Most sensitive tests are culture and tissue culture toxin neutralization assay. Nucleic acid amplification tests are sensitive but lack specificity. Most specific tests are enzyme immunoassay for toxins A and B; a two-step procedure can be used: sensitive but nonspecific C. difficile glutamate dehydrogenase antigen test followed by toxin assay or nucleic acid amplification test followed by toxin immunoassay

Multiplexed nucleic acid amplification tests (NAAT)

Patients with persistent diarrhea, recent travel to developing countries, immunocompromised patients (HIV, solid organ transplant, hematopoietic stem cell transplant recipients)

Several platforms commercially available. Can detect up to 22 viral, bacterial, and protozoa targets. May detect incidental carriers or produce false positive results due to low target numbers or probe nonspecificity. The role of certain pathogens tested by these platforms such as enteropathogenic and enteroaggregative E. coli has not been validated in U.S. adults.

Stool culture for Shigella, Salmonella, Campylobacter spp, and Shiga toxin–producing E. coli (O157:H7 and others)

Moderate to severe diarrhea and when stools are positive for inflammatory markers or contain gross blood and mucus (dysentery)

The four mucosa-inflammatory bacteria are the only bacteria routinely sought by most laboratories. Culture offers the advantage of antimicrobial susceptibility testing and the study of outbreaks

Specialized stool culture for Vibrio spp

For cases of profuse watery diarrhea in cholera-endemic areas and outbreaks of seafood-associated diarrhea or dysentery

Cholera cases may need aggressive fluid therapy. Non–cholera vibrios can cause dysentery.

Parasite examination: (1) NAAT, (2) enzyme immunoassay for Giardia spp, Cryptosporidium spp, or Entamoeba histolytica; (3) acid-fast stain for Cyclospora or Cryptosporidium spp or Cystoisospora; or (4) trichome stain and microscopic examination

In any patient with persistent diarrhea and when diarrhea follows visits to mountainous or recreational lakes in North America, Nepal, Haiti, Peru, or Russia

NAATs are increasingly used for detection of Cyclospora, Cryptosporidium and Giardia. If microscopic evaluation is performed, experience of the laboratory personnel is important. The commercially available enzyme immunoassay tests are sensitive.

Esophagogastroduodenoscopy and flexible sigmoidoscopy

Persistent diarrhea in patients without evidence of cause of illness

Identified cause of diarrhea is treated; without diagnosis, subjects may be treated symptomatically.

TABLE 267-4 THERAPY FOR AND PREVENTION OF INFECTIOUS DIARRHEA THERAPEUTIC OPTION

INDICATION

PHARMACOLOGIC AGENT

Oral fluid and electrolyte therapy

For infants, elderly patients, and anyone with profuse watery diarrhea

Soups, soft drinks, and saltine crackers are sufficient; formal oral replacement therapy may be needed with dehydrating forms of diarrhea

Diet: easily digestible foods

In all forms of diarrhea to facilitate enterocyte renewal and recovery

Soups and broth, saltine crackers, steamed vegetables, baked or broiled meats

Nonspecific therapy

For temporary (≤48 hours) control of diarrhea in older children and adults without evidence of severe diarrhea caused by an invasive or inflammatory bacterial or parasitic pathogen

Loperamide is the most effective symptomatic treatment and will decrease number of stools passed by 60%; bismuth subsalicylate is much less effective and will reduce number of stools by 40%; the antisecretory agent crofelemer is of value in HIV-associated diarrhea

Empirical antibacterial drugs

Enteric fever with toxicity Febrile dysenteric diarrhea

Fluoroquinolones for 7-10 days Azithromycin is recommended when fever or dysentery complicates illness Rifaximin 200 mg for 3 days, azithromycin 1000 mg in a single dose, or fluoroquinolone for 1-3 days when traveling to areas with low antibiotic resistance

Traveler’s diarrhea Specific antibacterial therapy

Shigellosis, campylobacteriosis, cholera

See Chapters 286, 287, and 293

Antiparasitic drugs

Giardiasis, amebiasis, cryptosporidiosis, cyclosporiasis

See Chapters 329, 330, and 332

Prophylaxis in traveler’s diarrhea

Persons traveling to developing areas on tight schedules, those with history of prior traveler’s diarrhea, persons with unstable underlying medical disorders, and those interested in prophylaxis

Rifaximin, 200 mg twice daily with meals, while in a high-risk region

travel to Nepal, Haiti, or Peru or other regions of the developing world (travelrelated infections are discussed in detail in Chapter 270). A specific food or water vehicle cannot be suspected unless multiple cases of illness with a common exposure occur. All too frequently, persons assume that food consumed during their last meal before an illness onset is responsible

for the symptoms. The highly variable incubation period for diarrheal disease, which may be as short as 2 hours after eating a food for preformed toxins, to a week or even longer for microbial enteropathogens, makes the determination of a specific food or beverage in a single case of illness impossible. When an outbreak of diarrhea results in multiple cases, a category of etiology (preformed

1834

CHAPTER 267  Approach to the Patient with Suspected Enteric Infection  

toxin versus enteric infection) can be determined by calculating the incubation period after looking at timing of the common exposure and the time of first symptoms. Short incubation periods are characteristic of food poisoning associated with enterotoxins (2 to 7 hours for cases caused by Staphylococcus aureus, 2 to 4 hours for Bacillus cereus enterotoxin food poisoning). Longer incubation periods (usually 12 to 72 hours or longer) are associated with most cases of intestinal infection.4 The clinical expression of diarrheal illness will give clues to the etiologic agent involved in disease (see Table 267-2). In the patient with diarrhea who is receiving or recently has completed a course of an antimicrobial drug, a proton pump inhibitor, or an anticancer drug, particularly with recent or current hospitalization, Clostridium difficile infection (Chapter 280) should be suspected. An increasing number of cases of C. difficile diarrhea are occurring in the community without prior health care contact or antimicrobial use. When a person has close contact with an infant or infants attending a daycare center, a number of low–infectious dose pathogens are found in this setting (e.g., Giardia, Cryptosporidium, or Shigella species or viral pathogens, particularly norovirus) should be suspected. Persons practicing oral-anal, or digital-anal contact experience higher rates of enteric infection acquired through fecal-oral contamination, often associated with infection by multiple pathogens or through the practice of unprotected receptive anal intercourse leading to proctitis due to sexually transmitted organisms. When persons have advanced acquired immunodeficiency syndrome (AIDS) or other forms of severe immunodeficiency associated with chemotherapy, hematopoietic stem cell or solid organ transplantation, chronic use of immunosuppressive drugs, and depressed intestinal immunity may lead to enteric infection with a variety of parasitic, bacterial, or viral pathogens (see Table 267-1) (Chapter 265). Infants with malnutrition may develop persistent diarrhea and substantial long-term morbidity due to protozoal parasites, including Giardia and Cryptosporidium. Enteric infection syndromes may be divided into at least five groups on the basis of the clinical presentation: (1) febrile systemic disease (enteric fever); (2) acute watery diarrhea (secretory diarrhea); (3) recurrent vomiting as the

primary manifestation of enteric disease (gastroenteritis); (4) passage of many small-volume stools containing blood and mucus (dysentery); and (5) diarrhea lasting 2 weeks or longer (persistent diarrhea). Table 267-2 lists the major syndromes along with the expected cause.5 Noroviruses (Chapter 356) have become the major cause of food-borne gastroenteritis and the most commonly identified cause of waterborne enteric disease.6 They have been identified as causes of persistent diarrhea in immunocompromised patients, especially in those undergoing hematopoietic stem cell transplantation. Campylobacter species (Chapter 287) is a commonly reported bacterial enteropathogen in industrialized countries and is the most important definable cause of Guillain-Barré syndrome (Chapter 392), often resulting in severe disease requiring assisted ventilation, intensive care unit confinement, and permanent neurologic sequelae. E. coli O157:H7 and other Shiga toxin–producing E. coli (STEC) (Chapter 288) are important causes of food-borne and waterborne colitis complicated by hemolytic-uremic syndrome in children and occasionally in elderly people. The most commonly detected pathogens in endemic diarrhea in the United States in one study were noroviruses (26%), rotavirus (18%), and Salmonella species (5.3%). EAEC strains are being shown to be important causes of pediatric diarrhea in the United States as diagnostic methods improve. In May and June of 2011, there was a large outbreak of diarrhea and hemolyticuremic syndrome reported from Germany and France due to an O104:H4 strain of enteroaggregative E. coli (EAEC) that had picked up the STEC phage controlling production of Shiga toxin. In the future, with improved diagnostic tools, we are likely to see more of these hybrid superpathogens with multiple virulence properties.  

DIAGNOSIS

Laboratory Findings

Laboratory tests (Fig. 267-1; see Table 267-3) can be useful and are of particular value in the more severely ill patients, when subjects are forced by their illness to alter activities or are totally disabled and confined to bed or when

A B

C

D

FIGURE 267-1.  Laboratory tests to diagnose causes of diarrhea. A, Dysenteric stool. B, Stool culture and biochemical tests confirm Salmonella. C, Giardia trophozoites. D, Many leukocytes in diffuse colonic inflammation.

CHAPTER 267  Approach to the Patient with Suspected Enteric Infection  

1835

E

F FIGURE 267-1, cont’d.  E, Entamoeba histolytica trophozoite with ingested red blood cells. F, Oocysts of Cryptosporidium (left) and Cyclospora (right). (From the CDC Public Health Information Library. http://phil.cdc.gov/phil/home.asp: images 7829 and 7827.)

many patients are afflicted during an outbreak. In each of these situations, the laboratory may help establish cause and allow development of a proper plan of treatment (see Table 267-3). Standard laboratory tests include procedures looking for fecal inflammatory markers, such as microscopic detection of fecal leukocytes or the more sensitive, commercially available test, fecal lactoferrin or calprotein. These tests are particularly helpful to suggest the presence of the invasive bacterial pathogens Shigella, Salmonella, and Campylobacter species or the noninvasive but inflammatory C. difficile. Enteropathogens associated with infectious diarrhea can be rapidly identified with new molecular methods, including real-time polymerase chain reaction, quantitation of pathogen load, and next-generation sequencing.7-11 Stool culture is performed in more severe cases of sporadic diarrhea and in disease outbreaks and is carried out with blood culture in a patient with fever and systemic toxicity. Other indications for stool culture are presence of dysentery (passage of grossly bloody stools) and when fecal inflammatory markers are found. In dysenteric diarrhea, particularly in the presence of an outbreak, the laboratory should also be instructed to look for E. coli O157:H7 and other Shiga toxin–producing E. coli. Parasite examination is indicated by diarrhea and persistent (≥14 days) illness; evidence that the subject practices oral-anal sex or unprotected receptive anal intercourse; or associated immunosuppression. Other tests are indicated in special situations, including stool culture for Vibrio cholerae in a patient with severe watery diarrhea with excessive fluid losses in or returning from a cholera-endemic area and culture for Mycobacterium avium complex, herpes simplex virus, and cytomegalovirus in those with immunosuppression. For patients with persistent diarrhea without etiologic diagnosis when routine tests are employed, endoscopy (esophagogastroduodenoscopy and flexible sigmoidoscopy or colonoscopy) may be indicated in attempting to determine the nature and cause of illness.

TREATMENT  Treatment of diarrhea should be tailored to the clinical syndrome. Oral rehydration therapy with fluids and electrolytes is used to treat acute watery diarrhea and gastroenteritis and all forms of enteric infection, especially when complicated with any degree of dehydration. A1  Oral rehydration therapy is particularly important in infants; it can be life-saving in developing countries for infants with severe diarrhea. Patients with diarrhea should be fed easily digestible foods to facilitate enterocyte renewal and to speed up disease recovery. In afebrile, nondysenteric diarrhea, symptomatic drugs may allow older children and adults with illness to return earlier to school or work. Loperamide is the most active drug for improvement of symptoms. Bismuth subsalicylate can reduce diarrhea and is mildly effective in reducing nausea and vomiting associated with viral gastroenteritis. For enteric fever, febrile dysenteric disease, and moderate to severe cases of traveler’s diarrhea, empirical antimicrobial therapy is indicated (see Table 267-4). For outbreaks of dysenteric diarrhea, particularly in children in whom fever is not significant, antibacterial and antimotility drugs should be initially withheld while the etiology of the outbreak is being established to prevent patients infected by STEC strains from being predisposed to hemolytic-uremic syndrome. For bacterial and parasitic pathogen-specific diarrhea, antimicrobial therapy is often advised (see other chapters in the text for specific treatments). Because of the importance of diarrhea when persons travel from industrialized regions to developing countries, prophylaxis with the orally administered, poorly absorbed rifaximin can be employed for some groups (see Table 267-4), with expected protection rates exceeding 70%. In sporadic cases of acute or persistent diarrhea, infectious agents are not always responsible. Table 267-5 offers a partial list of the noninfectious causes of diarrhea that should be considered.

TABLE 267-5 NONINFECTIOUS CAUSES OF DIARRHEA Running Fecal impaction Drugs and laxatives Enteral feeding Irradiation Pancreatic insufficiency Intestinal lymphangiectasia Foods (especially dietetic) Cirrhosis and biliary obstruction Diabetic diarrhea Alcoholism Collagenous colitis Microscopic colitis VIPoma Ischemic bowel disease Irritable bowel syndrome

Small bowel bacterial overgrowth Systemic mastocytosis and eosinophilic gastroenteritis Tropical sprue Celiac sprue Dermatitis herpetiformis Intestinal graft-versus-host disease Thyrotoxicosis Adrenal insufficiency Factitious Inflammatory bowel disease Food allergy Carcinoid Villous adenoma Stress with autonomic stimulation

  PERSISTENT DIARRHEA

Persistent diarrhea for more than 2 weeks develops in about 3% of individuals who travel to developing countries. Common causes include parasitic infections (e.g., Giardia [Chapter 330], Cryptosporidium [Chapter 329]), and bacterial infections (e.g., enteroaggregative E. coli [Chapter 288]). The evaluation should include culture for bacterial pathogens and methods for detecting bacterial, viral, and protozoal infections (e.g., nucleic acid amplification tests), including multiplexed NAATs, as well as microscopy for protozoal infections. Antimicrobial therapy can be empirically given to patients who have recently returned from the developing world but should preferably be given based on the results of laboratory testing.12 Therapy should be limited to the more severe cases to avoid post-antibiotic carriage of multidrug-resistant enterobacteriaceae, which has been shown to spread to family members and persist for as long as 12 months.

  Grade A Reference A1. Riddle MS, DuPont HL, Connor BA. ACG clinical guideline: diagnosis, treatment and prevention of acute diarrheal infections in adults. Am J Gastroenterol. 2016;111:602-622.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 267  Approach to the Patient with Suspected Enteric Infection  

GENERAL REFERENCES 1. Shane AL, Mody RK, Crump JA, et al. 2017 Infectious Diseases Society of America clinical practice guidelines for the diagnosis and managment of infectious diarrhea. Clin Infect Dis. 2017;65: 1963-1973. 2. Einarsson E, Ma’ayeh S, Svard SG. An up-date on Giardia and giardiasis. Curr Opin Microbiol. 2016;34:47-52. 3. Steffan R, Hill DR, DuPont HL. Traveler’s diarrhea: a clinical review. JAMA. 2015;313:71-80. 4. Acree M, Davis AM. Acute diarrheal infections in adults. JAMA. 2017;318:957-958. 5. Lübbert C. Antimicrobial therapy of acute diarrhea: a clinical review. Expert Rev Anti Infect Ther. 2016;14:193-206.

1836.e1

6. Bányai K, Estes MK, Martella V, et al. Viral gastroenteritis. Lancet. 2018;392:175-186. 7. Jiang ZD, DuPont HL. Etiology of travellers’ diarrhea. J Travel Med. 2017;24:S13-S16. 8. Byrne S, Robson JM. Advances in molecular diagnosis of parasitic enteropathogens. Pathology. 2015;47:234-242. 9. Binnicker MJ. Multiplex molecular panels for the diagnosis of gastrointestinal infection: performance, result interpretation and cost-effectiveness. J Clin Microbiol. 2015;53:3723-3728. 10. Eckbo EJ, Yansouni CP, Pernica JM, et al. New tools to test stool: managing travelers’ diarrhea in the era of molecular diagnostics. Infect Dis Clin North Am. 2019;33:197-212. 11. Hirvonen JJ. Comparison of three multiplex real-time PCR assays for detection of enteric viruses in patients with diarrhea. Eur J Clin Microbiol Infect Dis. 2019;38:241-244. 12. DuPont HL. Persistent diarrhea: a clinical review. JAMA. 2016;315:2712-2723.

1836.e2

CHAPTER 267  Approach to the Patient with Suspected Enteric Infection  

REVIEW QUESTIONS 1. A 30-year old traveler to Goa, India, for vacation develops diarrhea and fever on the fifth day at the resort. After 36 hours of diarrhea, stools begin to contain bright red blood. She visited a U.S. travel medicine clinic before heading for India and was given two antibiotics, one for the common watery diarrhea and a second in the less likely event that she developed bloody diarrhea and fever. Which of the following antibiotics would she have correctly received from the clinic to treat dysenteric traveler’s diarrhea? A . Oral ciprofloxacin, 500 mg twice a day for 3 days B. Azithromycin, 1000-mg single dose C. Levofloxacin, 500 mg once a day for 3 days D. Metronidazole, 500 mg three times a day for 10 days E. Rifaximin, 200 mg three times a day for 3 days Answer: B  The invasive pathogens (Shigella, Salmonella, and Campylobacter) occur more commonly in travelers to Asia compared with destinations in Africa and Latin America. Ciprofloxacin and levofloxacin would not be the best choice in this patient because of the importance of fluoroquinoloneresistant Campylobacter as a causative agent of dysenteric traveler’s diarrhea occurring in Asia. Metronidazole would be appropriate for persistent diarrhea caused by Giardia or Entamoeba histolytica but would not be useful in most patients with traveler’s diarrhea where the typical enteropathogenic agents are bacterial in origin. Rifaximin is of value for the common watery diarrhea of travelers but is of no value for the treatment of mucosally invasive enteropathogens as the drug is poorly absorbed (50 mL per min, with necessary modifications if creatinine is less than 50 mL per min) are also FDAlicensed for the treatment of urinary tract infections due to P. aeruginosa. Cefiderocol (S-649266) is a siderophore cephalosporin with in vitro activity against gram-negative bacteria including P. aeruginosa (carbapenem-resistant and carbapenem-susceptible) strains. It has been tested clinically in a Phase II noninferiority study (2 g every 8 hours IV for 7 to 14 days with adjustment to renal function) against imipenem/cilastatin (1 g every 8 hours with adjustment to renal function) for complicated and uncomplicated (up to 30% of total patients) urinary tract infections caused by gram-negative bacteria (including E. coli, Pseudomonas spp, Klebsiella spp, and Acinetobacter spp) in 452 patients. Cefiderocol met noninferiority versus imipenem/cilastatin for the composite clinical cure and microbiological eradication end point. Serious adverse events were 4.7% for cefiderocol and 8.1% for imipenem/cilastatin. Cefedirecol has been designated as a Qualified Infectious Diseases Product by the U.S. FDA, which allows for priority review and provides eligibility for fast-track approval status. Hot tub folliculitis is usually self-limited; in difficult to resolve infections oral ciprofloxacin 500 mg twice daily for 7 days is effective. Toe web infections responds to topical use of acetic acid (3 to 5%) compresses two to four times daily and in more severe cases to oral ciprofloxacin 500 mg twice daily for 7 days; concurrent treatment of tinea pedis if present is necessary. The presence of dryness helps significantly in the elimination of the infection. Application of acetic acid 5% compresses 2 to 4 times daily for 4 to 16 weeks and débridement of the onycholytic part of the nail are effective in the majority of patients. Topical polymyxins have also been used. Removal of the nail and oral ciprofloxacin are occasionally used if topical therapy fails. Management of P. aeruginosa–infected burn wounds is both surgical and medical. Extensive débridement of colonized eschar or necrotic tissue is required in addition to antibiotic treatment. Medical treatment follows the principles detailed in the bacteremia treatment section. Management of osteochondritis due to P. aeruginosa complicating puncture wounds of the foot requires intravenous antibiotics (cefepime, ceftazidime, or piperacillin-tazobactam) or oral ciprofloxacin for 7 to 14 days in addition to surgical intervention for acquisition of tissue for biopsy and removal of necrotic cartilage. Management of native vertebral osteomyelitis due to P. aeruginosa necessitates a 6-week course of the following antibiotics: as first choice (adjusted for renal function) are cefepime 2 g IV every 8-12 hours or meropenem 1 g IV every 8 hours or doripenem (doripenem suggested by IDSA guidelines but no FDA clearance) 500 mg IV every 8 hours. Alternatives include ciprofloxacin 750 mg po every 12 hours or 400 mg IV every 8 hours or ceftazidime 2 g IV every 8 hours. For penicillin allergy and quinolone-resistant strains, aztreonam 2 g IV every 8 hours is suggested. Potential consideration of combination therapy for treatment of natural vertebral osteomyelitis with use of an antipseudomonal β-lactam and ciprofloxacin or antipseudomonal β-lactam and an aminoglycoside is advocated by the current guidelines.16 For treatment of chronic osteomyelitis of various bones due to P. aeruginosa, ciprofloxacin has been used extensively. Experts infer that oral treatment with ciprofloxacin 750 mg twice daily is acceptable as an alternative to parenteral treatment. There are relatively few data regarding the appropriate duration of treatment. A duration of 6 to 12 weeks is most commonly reported. Chronic osteomyelitis as well as sternoarticular osteomyelitis often require débridement of necrotic bone tissue. The treatment of prosthetic joint infection in addition to antibiotic treatment may necessitate removal of the prosthesis (two-stage exchange usually in the United States, but one-stage exchange has also been used); nevertheless, the physician should identify the patients that are not candidates for a new prosthesis.

B. cepacia is a plant and human pathogen. Its classification has evolved, as has its name. It was previously known as P. cepacia and then Xanthomonas cepacia. B. cenocepacia, B. multivorans, and B. cepacia belong to the B. cepacia complex and area pathogens mainly of patients with CF, chronic granulomatous disease, sickle cell disease, patients with burns, patients with bronchiectasis, and patients with malignancy. B. cepacia complex is implicated in infections of the lung, blood, and other sites in immunocompromised patients and even in immunocompetent patients and children in the hospital environment due to outbreaks. Aerosolized medications, chlorhexidine solutions, napkins, and prefabricated clothes along with horizontal transmission have led to significant morbidity and mortality. Patients with CF typically present initially with asymptomatic carriage. Nevertheless, pneumonia and progressive decline of lung function is associated with B. cepacia complex infections in patients with CF. The most dramatic presentation of the lung ailment is the cepacia syndrome, a fulminant necrotizing pneumonia often accompanied by bacteremia. Other types of infection include meningitis, endophthalmitis, pericarditis, endocarditis, burn wound infection, cholangitis, peritonitis, abscesses in the abdomen, perineum, or scrotum, and post-lung transplant pneumonia.

TREATMENT  High-quality data in the literature regarding the treatment of B. cepacia complex infections are scarce and suggested empiric treatment is further hampered by the variable antimicrobial resistance of the bacterium. Trimethoprimsulfamethoxazole (10 to 15 mg/kg/day based on trimethoprim component) has been considered as first-line treatment (note that there are B. cepacia complex isolates resistant to TMP/SMX), but other first-line options are meropenem (1 to 2 g every 8 hours) or ceftazidime (2 g every 8 hours). Second-line options include minocycline (100 mg IV every 12 hours), ciprofloxacin, ticarcillin/clavulanate, and piperacillin/tazobactam. Although ciprofloxacin has also been used in the treatment of B. cepacia complex infections, the European Committee on Antimicrobial Susceptibility Testing (EUCAST) reports B. cepacia to be inherently resistant to ciprofloxacin. Furthermore, the Clinical and Laboratory Standards Institute reports intrinsic resistance for piperacillin/tazobactam and EUCAST reports intrinsic resistance for ticarcillin/clavulanate. Because of the variable resistance of B. cepacia complex, combinations of first- and second-line antibiotics (dual or even triple combinations) have been used for the treatment of critically ill patients.



PREVENTION

Primary prevention of Pseudomonas infections includes preventing pollution of water by Pseudomonas. This applies to both the public environment and the hospital environment. Outbreaks have been linked to aquatic environments such as whirlpools, swimming pools, and spas. Thus, control of growth of this organism in the recreational environment by proper antibacterial treatment of water is essential, comparable to the control practiced in hospitals. Contamination of various equipment and devices (i.e., breast implants, ocular implants, and sinus irrigation devices) must be avoided. Handwashing cannot be emphasized enough in the prevention of infections due to P. aeruginosa, S. maltophilia, and B. cepacia. Hospital point-of-use water filtration has also been employed in the battle against P. aeruginosa and S. maltophilia infections. Barrier nursing practices may decrease horizontal transmission, especially in the critical care environment. Isolation of patients infected with extensively drugresistant and pandrug-resistant P. aeruginosa and strict infection control measures (dedicated personnel and equipment, use of gowns and gloves) may be needed to avoid intrahospital spread. Nevertheless, even antiseptic solutions may be contaminated by P. aeruginosa, S. maltophilia, and B. cepacia. Factors that shorten hospital length of stay and decrease the use of antibiotics are likely to decrease the incidence of these infections.  

PROGNOSIS

P. aeruginosa infections carry a high mortality, even with treatment. Mortality rates of 30% have been reported for P. aeruginosa bacteremia.18 Differences also exist between appropriately and inappropriately treated P. aeruginosa infections, with mortality rates even double when inappropriate antibiotics are used (usually for multidrug-resistant strains). Furthermore, P. aeruginosa infections are associated with increased length of hospitalization and medical costs. Loss of vision is the grave outcome of ophthalmic infection. Antibiotic resistance associated with P. aeruginosa will most likely continue to pose an enormous burden on human lives and financial resources in years to come.

S. maltophilia infections are associated with attributable mortality of up to about 37% if they are not appropriately treated. Melioidosis is associated with a 14% mortality rate in Australia and up to 40% in countries in Southeast Asia when it is treated. Fortunately, reports of glanders are rare in humans. Mortality of glanders with treatment is about 40 to 50%. B. cepacia has a predilection for CF patients and often signals impaired overall prognosis in these patients. Furthermore, B. cepacia seems to adversely affect transplantation. Indeed, in patients presenting with the cepacia syndrome, survival is unusual. Furthermore, many transplant centers do not support transplantation when B. cepacia is present because of the grave prognosis associated with the infection in the post-transplantation period.

Grade A References A1. Smith S, Rowbotham NJ, Regan KH. Inhaled anti-pseudomonal antibiotics for long-term therapy in cystic fibrosis. Cochrane Database Syst Rev. 2018;3:CD001021. A2. Chetchotisakd P, Chierakul W, Chaowagul W, et al. Trimethoprim-sulfamethoxazole versus trimethoprim-sulfamethoxazole plus doxycycline as oral eradicative treatment for melioidosis (MERTH): a multicentre, double-blind, non-inferiority, randomised controlled trial. Lancet. 2014;383:807-814.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 290  PSEUDOMONAS AND RELATED GRAM-NEGATIVE BACILLARY INFECTIONS  

GENERAL REFERENCES 1. Moradali MF, Ghods S, Rehm BH. Pseudomonas aeruginosa lifestyle: a paradigm for adaptation, survival, and persistence. Front Cell Infect Microbiol. 2017;7:1-29. 2. Wiener LM, Webb AK, Limbago B, et al. Antimicrobial-resistant pathogens associated with healthcareassociated infections: summary of data reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 2011-2014. Infect Control Hosp Epidemiol. 2016;37: 1288-1301. 3. Corlouer C, Lamy B, Desroches M, et al. Stenotrophomonas maltophilia healthcare-associated infections: identification of two main pathogenic genetic backgrounds. J Hosp Infect. 2017;96:183-188. 4. Perual Samy R, Stiles BG, Sethi G, et al. Melioidosis: clinical impact and public health threat in thetropics. PLoS Negl Trop Dis. 2017;11:1-28. 5. Sylvester MJ, Sanghvi S, Patel VM, et al. Malignant otitis externa hospitalizations: analysis of patient characteristics. Laryngoscope. 2017;127:2328-2336. 6. Kalil AC, Metersky ML, Klompas M, et al. Management of adults with hospital-acquired and ventilatorassociated pneumonia: 2016 clinical practice guidelines by the Infectious Diseases Society of America and the American Thoracic Society. Clin Infect Dis. 2016;63:575-582. 7. Crull MR, Somayaji R, Ramos KJ, et al. Changing rates of chronic Pseudomonas aeruginosa infections in cystic fibrosis: a population-based cohort study. Clin Infect Dis. 2018;67:1089-1095. 8. Sujimoto S, Miyoshi K, Yamane M, et al. Lung transplantation for diffuse panbronchiolitis: 5 cases from a single centre. Interact Cardiovasc Thorac Surg. 2016;22:679-681. 9. Baddour LM, Wilson WR, Bayer AS, et al. Infective endocarditis in adults: diagnosis, antimicrobial therapy, and management of complications: a scientific statement for healthcare professionals from the American Heart Association. Circulation. 2015;132:1435-1486.

1936.e1

10. Rhodes A, Evans LE, Alhazzani W, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock: 2016. Intensive Care Med. 2017;43:304-377. 11. Pang Z, Raudonis R, Glick BR, et al. Antibiotic resistance in Pseudomonas aeruginosa: mechanisms and alternative therapeutic strategies. Biotechnol Adv. 2019;37:177-192. 12. Tsuji BT, Pogue JM, Zavascki AP, et al. International consensus guidelines for the optimal use of the polymyxins: endorsed by the American College of Clinical Pharmacy (ACCP), European Society of Clinical Microbiology and Infectious Diseases (ESCMID), Infectious Diseases Society of America (IDSA), International Society for Anti-infective Pharmacology (ISAP), Society of Critical Care Medicine (SCCM), and Society of Infectious Diseases Pharmacists (SIDP). Pharmacotherapy. 2019;39:10-39. 13. Falagas ME, Rafailidis PI. Fosfomycin: the current status of the drug. Clin Infect Dis. 2015;61: 1144-1146. 14. Wood GC, Swanson JM. An update on aerosolized antibiotics for treating hospital-acquired and ventilator-associated pneumonia in adults. Ann Pharmacother. 2017;51:1112-1121. 15. Tunkel AR, Hasbun R, Bhimraj A. 2017 Infectious Diseases Society of America’s clinical practice guidelines for healthcare-associated ventriculitis and meningitis. Clin Infect Dis. 2017;64:e64-e65. 16. Berbari EF, Kanj SS, Kowalski TJ, et al. 2015 infectious diseases society of America (IDSA) clinical practice guidelines for the diagnosis and treatment of native vertebral osteomyelitis in adults. Clin Infect Dis. 2015;61:e26-e46. 17. Hadano Y. Imported melioidosis in Japan: a review of cases. Infect Drug Resist. 2018;11:163-168. 18. Pena C, Cabot G, Gomez-Zorrilla S, et al. Influence of virulence genotype and resistance profile in the mortality of Pseudomonas aeruginosa bloodstream infections. Clin Infect Dis. 2015;60:539-548.

1936.e2

CHAPTER 290  PSEUDOMONAS AND RELATED GRAM-NEGATIVE BACILLARY INFECTIONS  

REVIEW QUESTIONS 1. Which of the following bacteria is most commonly associated with hospitalacquired infections in the United States? A . Stenotrophomonas maltophilia B. B. cepacia C. P. aeruginosa D. B. mallei E. B. pseudomallei Answer: C  Data from the National Healthcare Safety Network at the Centers for Disease Control and Prevention indicate that during 2009 to 2010, P. aeruginosa was responsible for 8% of a total of 69,475 hospital-acquired infections and was the sixth most frequent culprit among 81,139 pathogens. S. maltophilia is a pathogen responsible for hospital-acquired infections but ranks lower in frequency than P. aeruginosa. B. pseudomallei and B. mallei are not causes of hospital-acquired infections in the United States. However, these pathogens are bioterrorism factors B. B. cepacia is an important emerging pathogen that is associated with hospital-acquired infections in immunocompromised patients beyond its established role as a pathogen in cystic fibrosis patients. B. cepacia hospital-acquired infections rank much lower in frequency than hospital-acquired infections due to P. aeruginosa. 2. Which of the following statements regarding P. aeruginosa is true? A . It is a nonmotile gram-negative bacterium. B. It is not the only cause of ecthyma gangrenosum. C. Endocarditis due to P. aeruginosa is always associated with high fever. D. Ertapenem is effective as treatment against P. aeruginosa. E. Tigecycline is effective as treatment against P. aeruginosa. Answer: B  Other causes of ecthyma gangrenosum include Staphylococcus aureus, Aeromonas hydrophila, Klebsiella pneumoniae, Morganella morganii, Stenotrophomonas maltophilia, Citrobacter freundii, and fungi (Candida albicans, Aspergillus fumigatus). P. aeruginosa is a motile bacterium. Ertapenem and tigecycline are not effective treatment against P. aeruginosa. Endocarditis due to P. aeruginosa is usually more indolent than endocarditis due to S. aureus. It is not always associated with high fever.

3. Which of the following patient groups is at increased risk for the development of P. aeruginosa infections? A . Neutropenic patients B. Patients with cystic fibrosis C. Patients with Charcot arthropathy D. Patients requiring mechanical ventilation E. All the above Answer: E  Neutropenia, Charcot’s arthropathy, mechanical ventilation, and also burns, cystic fibrosis, cancer, transplantation, diabetes, steroid treatment, foreign bodies, hypogammaglobulinemia, and AIDS are associated with an increased risk for P. aeruginosa infection. 4. Which of the following is considered the drug of first choice for Stenotrophomonas maltophilia? A . Doxycycline B. Meropenem C. Trimethoprim-sulfamethoxazole D. Ciprofloxacin E. Ceftazidime Answer: C  S. maltophilia has inherent resistance to carbapenems. Trimethoprim-sulfamethoxazole is the choice for first-line treatment. Ciprofloxacin and ceftazidime have been used in the treatment of S. maltophilia infections either alone or in combination but are not first-choice treatment. Tetracyclines have some activity against S. maltophilia but are not first-choice options for treatment. 5. The following are true regarding infection due to B. pseudomallei except which one? A . Southeast Asia, Australia, the Indian subcontinent, and Sri Lanka are locations with an increased risk for this infection. B. Risk factors for infection include renal failure, diabetes mellitus, and heavy alcohol consumption. C. The infection is called melioidosis. D. An extended eradication regimen with oral trimethoprim-sulfamethoxazole for 1 month is necessary to follow initial intravenous treatment. E. Recurrence of melioidosis is due to reactivation. Answer: D  An extended oral eradication regimen with oral trimethoprimsulfamethoxazole for 3 to 6 months (not only 1 month) is recommended to follow initial intravenous treatment for the treatment of melioidosis.

1936

CHAPTER 291  DISEASES CAUSED BY ACINETOBACTER AND STENOTROPHOMONAS SPECIES  

291  DISEASES CAUSED BY ACINETOBACTER AND STENOTROPHOMONAS SPECIES KEITH S. KAYE AND ROBERT A. BONOMO

  ACINETOBACTER SPP  

DEFINITION

The Pathogen

Acinetobacter species are gram-negative aerobic bacteria that are coccobacillary in shape and are generally described as aerobic, non–lactose-fermenting, nonfastidious, nonmotile, catalase positive, and oxidase negative. Unique among microbial pathogens, the appearance of Acinetobacter species visualized with a Gram stain is highly dependent on the life cycle. In the early growth phases, Acinetobacter spp appear rod shaped. In the stationary phase, they acquire a coccobacillary Acinetobacter spp morphology. Their description of being “nonmotile” is seriously being questioned. In the context of this chapter, Acinetobacter spp refers specifically to A. baumannii and A. baumannii-calcoaceticus complex. A representative list of Acinetobacter species is provided in E-Table 291-1. Molecular methods used to identify and to classify Acinetobacter are listed in E-Table 291-2.  

EPIDEMIOLOGY

Acinetobacter spp can colonize many body surfaces and cause infection in almost any organ system.1 Consequently, there are a number of common clinical syndromes associated with infection by Acinetobacter. The most common infections are respiratory (pneumonia), blood stream (bacteremia), urinary tract, wound, skin and soft tissue, and burn infections; osteomyelitis secondary to trauma; and meningitis.2 In general, infection is observed only in critically ill, immunocompromised, or injured hosts. Recently, infections by Acinetobacter spp are being described in patients without significant medical problems from the community setting. During the past decade, Acinetobacter spp emerged in the United States from a pathogen that was primarily found in intensive care units (ICUs) to one that can affect patients in non-ICU wards, patients in long-term care settings,3,4 and military personnel with combat injuries acquired in the Middle East. Overall, there are few distinguishing features of Acinetobacter spp

CHAPTER 291  DISEASES CAUSED BY ACINETOBACTER AND STENOTROPHOMONAS SPECIES  

ABSTRACT

Acinetobacter and Stenotrophomonas spp are most frequently recovered from patients in the hospital, often from intensive-care units. The most common type of infections caused by these pathogens are infections of the respiratory tract. Recovery of these pathogens from the respiratory tract does not equal infection—colonization occurs, particularly among patients who are mechanically ventilated. Antimicrobial resistance is commonly detected among both pathogens, mediated by a variety of different mechanisms. Infections due to Acinetobacter spp are often treated with either sulbactam (given in the formulation ampicillin-sulbactam in the United States) or a carbapenem (meropenem or imipenem). Extremely-drug resistant (XDR) strains resistant to these antibiotics can be treated with minocycline, tigecyeline, or polymyxins. Combination therapy is often reserved to treat XDR Acinetobacter spp strains. For infections due to Stenotrophomonas spp, trimethoprim-sulfamethoxazole is the treatment of choice. Infection prevention is important in curtailing spread of these pathogens.

KEYWORDS

Acinetobacter Stenotrophomonas antibiotic resistance nosocomial polymyxin ventilator-associated pneumonia (VAP)

1936.e3

1936.e4

CHAPTER 291  DISEASES CAUSED BY ACINETOBACTER AND STENOTROPHOMONAS SPECIES  

E-TABLE 291-1 REPRESENTATIVE MEMBERS OF THE ACINETOBACTER GENUS* Acinetobacter albensis Acinetobacter apis Acinetobacter baumannii, Acinetobacter genome sp 2; reference strain ATCC 19606

Acinetobacter genome sp 3, reference strain ATCC 19004

Acinetobacter baylyi

Acinetobacter genome sp 6

Acinetobacter beijerinckii Acinetobacter bohemicus Acinetobacter bouvetii

Acinetobacter genome sp 10

Acinetobacter calcoaceticus, Acinetobacter genome sp 1, reference strain ATCC 23055

Acinetobacter genome sp 11

Acinetobacter courvalni Acinetobacter gerneri

Acinetobacter genome sp 13BJ or 14TU

Acinetobacter grimontii

Acinetobacter genome sp 14BJ

Acinetobacter haemolyticus, Acinetobacter genome sp 4, reference strain ATCC 17906

Acinetobacter genome sp 15BJ

Acinetobacter johnsonii, Acinetobacter genome sp 7, reference strain ATCC 17909

Acinetobacter genome sp 10

Acinetobacter junii

Acinetobacter genome sp 16

Acinetobacter lwoffii

Acinetobacter genome sp 17

Acinetobacter parvus

Acinetobacter genome sp 13TU, reference strain ATCC 17903 Now called Acinetobacter nosocomialis

Acinetobacter radioresistens

Acinetobacter genome sp 15TU

E-TABLE 291-2 IDENTIFICATION AND TYPING METHODS USED TO CHARACTERIZE ACINETOBACTER SPECIES 16S rRNA gene restriction analysis Amplified fragment length polymorphism

Acinetobacter genome sp 12 Acinetobacter seiferti Acinetobacter schindleri

Acinetobacter genome sp between 1 and 3

Acinetobacter tandoii

Acinetobacter genome sp close to 13TU

Acinetobacter tjernbergiae

Acinetobacter pittii

Acinetobacter townerii

Acinetobacter ursingii

Acinetobacter venetius *Genome species designations are added. Where appropriate, the American Type Culture Collection (ATCC) reference strain number is given.

blaOXA-51 gene amplification Polymerase chain reaction coupled with electrospray ionization mass spectrometry Multilocus sequence typing Ribotyping Restriction enzyme digestion of the 16S-23S rRNA intergenic spacer sequences Sequence analysis of the 16S-23S rRNA gene spacer region rpoB gene sequencing and its flanking spacers Matrix-assisted laser desorption/ionization time-of-flight Whole Genome Sequencing (WGS)

CHAPTER 291  DISEASES CAUSED BY ACINETOBACTER AND STENOTROPHOMONAS SPECIES  

1937

infection except for, in some instances, skin manifestations. The frequency of Acinetobacter infections is often greater in the summer than in other seasons.  

CLINICAL MANIFESTATIONS

Pneumonia

Because of colonization of the oropharynx and tracheostomy tubes in patients on ventilators, the upper respiratory tract is the most common site for infection by Acinetobacter species. The two distinct syndromes associated with respiratory tract infection due to Acinetobacter are community-acquired pneumonia (CAP) and health care–associated pneumonia (HCAP).5 In tropical regions of China, Asia, Australia, and the South Pacific, CAP due to Acinetobacter species is increasingly recognized. In some locations, the incidence can exceed 15%. Reports have highlighted the emergence of Acinetobacter as a common cause of CAP in western China. In Saudi Arabia, Acinetobacter is the most common pathogen associated with late-onset and recurrent ventilatorassociated pneumonia (VAP) in one adult ICU. The common comorbid conditions predisposing to CAP due to Acinetobacter species are mainly chronic obstructive pulmonary disease (emphysema), renal disease, diabetes mellitus, and alcoholism.6 CAP due to Acinetobacter species appears to be associated with a high incidence of bacteremia, acute respiratory distress syndrome, sepsis, and death (mortality rates in some reports are ≥50%). The reasons for these fulminant presentations are still unknown. Rarely, CAP due to Acinetobacter species can be manifested with consolidation and multiple lung abscesses. More frequently, Acinetobacter spp are a cause of pneumonia in patients who have risk factors for antimicrobial resistant organisms, often manifested as VAP. In the United States, Acinetobacter species are a leading cause of VAP. HCAP due to Acinetobacter largely resembles the clinical spectrum of gramnegative pneumonias (bilateral infiltrates, pleural effusion, cavitations, hypoxemia, and bacteremia). Most cases are described in patients on ventilators. The main factors associated with HCAP due to Acinetobacter species are mechanical ventilation, previous antibiotic exposure, ICU stay, surgery, and underlying pulmonary disease. The major challenge complicating nosocomial pneumonia due to Acinetobacter species is the frequent recovery of multidrug resistant (MDR) and sometimes extensively drug resistant (XDR) strains. When isolates are MDR or XDR, the options for treatment are limited, and complications quickly arise.7,8 HCAP due to MDR Acinetobacter has been associated with excess lengths of stay and mortality, although some studies have reported similar outcomes compared with control patients matched by severity of illness and duration of ICU stay.

Bacteremia

Blood stream infection due to Acinetobacter species is often a consequence of infection of intravenous catheters (i.e., central line–associated blood stream infection or CLABSI) or is secondary to HCAP due to Acinetobacter. Less commonly, wound infections can cause bacteremia. In most series, mortality associated with blood stream infection ranges from approximately 15 to greater than 50%.

Urinary Tract Infection

Urinary tract infections are most commonly caused by enteric gram-negative bacilli; only rarely are these infections caused by Acinetobacter species. The indwelling bladder catheter has been implicated as the major risk factor for urinary tract infection (cystitis and pyelonephritis) due to Acinetobacter species.

Wound, Burn, and Skin and Soft Tissue Infections

In many clinical series to date, traumatic or postoperative wounds, burns, and skin and soft tissue infections (SSTIs) are the most common causes of Acinetobacter infections. Most likely, the combination of antibiotic use, colonization, and compromised or devitalized tissues is responsible. The spectrum of infection can extend from cellulitis to necrotizing fasciitis. Because of the outbreak of A. baumannii among military personnel in Iraq and Afghanistan, reports of severe wound infections and SSTIs caused by this pathogen are increasing in frequency. Necrotizing SSTI associated with A. baumannii occurs in hosts with underlying comorbidities (e.g., trauma, cirrhosis) and is often accompanied by bacteremia. Acinetobacter infection has become increasingly common among patients residing in burn units, sometimes resulting in unit-wide outbreaks. Multiple drug resistance and the presence of co-pathogens frequently complicate treatment. Most cases require surgical débridement and lead to substantial mortality. In addition, the use of central venous catheters and total parenteral nutrition is more common among patients with SSTIs. Acinetobacter species–associated SSTIs can be manifested with a peau d’orange appearance, with overlying tiny vesicles (Fig. 291-1), and, when

FIGURE 291-1.  Cellulitis caused by Acinetobacter baumannii. There is characteristic edematous peau d’orange erythema, with associated vesicles that may coalesce to form nonhemorrhagic bullae. (From Guerrero DM, Perez F, Conger NG, et al. Acinetobacter baumannii–associated skin and soft tissue infections: recognizing a broadening spectrum of disease. Surg Infect [Larchmt]. 2010;11:49-57.)

untreated, can progress to necrotizing infection with bullae (hemorrhagic and nonhemorrhagic).

Osteomyelitis

The conflicts in Iraq and Afghanistan highlighted the first cases of osteomyelitis due to Acinetobacter species. Before this time, rare cases of osteomyelitis occurring in soldiers were reported during the Korean and Vietnam wars. Most of the initial reports from the Middle East described “contiguous focus” osteomyelitis. These patients had open fractures or exposed bone, with gross findings of infection: purulence, necrotic tissue, or environmental contamination with exposed bone; temperature higher than 38° C; leukocyte count greater than 12,000/µL; and Acinetobacter species identified from culture of deep wound tissue obtained intraoperatively. Frequently, these infections require multiple surgical débridements of necrotic bone.

Meningitis

Acinetobacter meningitis is occasionally found in the post-neurosurgical setting, with mortality exceeding 15 to 30%. Infection often involves intraventricular catheters.9 Although Acinetobacter is an uncommon cause of meningitis from the community, the frequency is greater in cases of hospital-acquired meningitis, particularly among patients with meningitis involving an indwelling intraventricular catheter.  

DIAGNOSIS

There are now more than 50 different species in the genus Acinetobacter, and their classification and identification remain problematic for clinicians. Automated and biochemical systems are sometimes inaccurate for the identification of Acinetobacter species. The introduction of matrix-assisted laser desorption/ ionization time-of-flight into the clinical laboratory may facilitate better diagnosis.

TREATMENT  An increasing number of strains of Acinetobacter species are resistant to all antibiotics, and these strains are often responsible for outbreaks in large hospitals. These multidrug-resistant (MDR) or extremely drug-resistant (XDR) strains are resistant to three or more classes of antibiotics. Among the resistance genes found in Acinetobacter species are a large collection of genes encoding β-lactamases, aminoglycoside-modifying enzymes, and many efflux pumps. Even more concerning is the emergence of XDR strains of Acinetobacter that are resistant to carbapenems10 and sulbactam, leaving few treatment options.

1938

CHAPTER 291  DISEASES CAUSED BY ACINETOBACTER AND STENOTROPHOMONAS SPECIES  

Unfortunately, resistance to “last-line” agents, such as the cationic antimicrobial peptides colistin (polymyxin E)11 and its close relative polymyxin B, also occurs. Regrettably, colistin resistance is becoming more common as clinicians are forced to use this agent more frequently. In hospital environments, Acinetobacter species can withstand drying (desiccation) and may even be transmitted by aerosol/respiratory droplet. Combined with drug resistance, these characteristics create a formidable infection control challenge. Increasing resistance to a variety of antimicrobial agents complicates the treatment of Acinetobacter species infections. In general, infections due to more resistant strains of Acinetobacter are associated with less favorable outcomes than are infections due to more susceptible strains. These worse outcomes are likely due in part to limited and suboptimal treatment options as well as delays in the time to implementation of effective antimicrobial therapy for patients with infections due to MDR and XDR Acinetobacter strains. The treatment doses that follow all assume normal renal function; doses need to be adjusted based on the degree of renal insufficiency. When it is “susceptible,” Acinetobacter is usually treated with sulbactam or a carbapenem. A commonly used dose of sulbactam (in the formulation of ampicillin-sulbactam in the United States) is 3 g IV every 6 hours. Some experts recommend a higher dose of ampicillin-sulbactam (3 g IV every 4 hours). Commonly used carbapenem regimens include imipenem 500 to 1000 mg IV every 6 hours or meropenem 500 to 1000 mg IV every 8 hours. Some experts recommend that higher doses of meropenem be used (1 g IV every 6 hours or 2 g IV every 8 hours). One must be cautious: susceptibility to imipenem does not always translate to susceptibility to meropenem, and susceptibility to ceftazidime does not always indicate that cefepime can be used. Aminoglycosides are an option for treatment of urinary tract infection and potentially for systemic infections. Unfortunately, the presence of ribosomal methyl transferases limits the use of plazomicin. Treatment of infection with XDR Acinetobacter is challenging.12 Tigecycline (100 mg loading dose, then 50 mg IV every 12 hours) and minocycline (200 mg IV or PO initially, followed by 100 mg IV or PO every 12 hours) have good in vitro activity against many strains of XDR Acinetobacter. Some experts recommend higher maintenance dosages of tigecycline (100 mg IV every 12 hours) and minocycline (200 mg IV or PO every 12 hours). These agents are appropriate for treatment of SSTI, but because of low serum concentrations and lack of clinical experience in the treatment of MDR and XDR Acinetobacter, monotherapy with these should for the most part be avoided for invasive infections such as bacteremia and pneumonia. Sometimes, there is no choice but to use tigecycline as it might be the only available agent with in vitro activity against infecting Acinetobacter strains. Eravacycline, a new synthetic fluorocycline, may have a role against Acinetobacter spp as MICs are better. Invasive infections due to XDR strains, such as bacteremia, pneumonia, and deep wound infections, are often treated with a polymyxin antimicrobial, either colistimethate sodium (CMS, often referred to as colistin) or polymyxin B. If CMS is used, a loading dose of 300 mg colistin base activity should be administered, followed by maintenance dosing of 150 to 180 mg colistin base activity every 12 hours with subsequent renal dose adjustments made according to published guidelines.13 If polymyxin B is used, a loading dose of 2.5 mg/kg × 1 should be administered intravenously, followed by a dose of 2.5 to 3.0 mg/ kg/day by continuous IV infusion or in divided doses every 12 hours infused over a period of 60 minutes.14 Clinicians often treat invasive infections due to XDR Acinetobacter with combination therapy, although there is a lack of prospective controlled data demonstrating the superiority of combination therapy. Agents often combined with the polymyxins include rifampin (10 mg/kg/day, not to exceed 600 mg), imipenem or meropenem, sulbactam (ampicillin-sulbactam), tigecycline or minocycline, and aminoglycosides. In controlled trials, however, neither rifampin nor meropenem added measurable clinical benefit over colistin monotherapy for the treatment of serious infection due to MDR or XDR Acinetobacter. A1  A2  The role of combination therapy in preventing the emergence of polymyxin resistance during therapy remains uncertain. A prospective clinical trial which included genotypic analysis of infecting Acinetobacter strains would be necessary to address the impact of combination therapy on the emergence of resistance. In addition, the benefit of aerosolized colistin has not yet been established. The major adverse effect of polymyxin therapy is nephrotoxicity.15 Current advances have improved our understanding of the pharmacokinetics, pharmacodynamics, and dosing of the polymyxins, but this is still an emerging and changing field. There is promising interest in the use of antioxidants (e.g., ascorbic acid) to prevent nephrotoxicity, but adequately powered randomized controlled studies have not been performed. In cases of meningitis due to Acinetobacter species, treatment should be intravenous meropenem (2000 mg every 8 hours) plus the intraventricular/ intrathecal administration of an aminoglycoside (gentamicin or amikacin, depending on susceptibilities). Intraventricular/intrathecal gentamicin is administered at a dose of 4 mg once daily every 1 to 3 days until clinical and microbiologic improvement occurs. Amikacin can be used in place of gentamicin at the dose of 30 mg once a day, if required. For cases of meningitis due to XDR Acinetobacter, intravenous colistin or polymyxin B should be supplemented with intraventricular or intrathecal administration of colistin 10 mg/day, or polymyxin B 5 to 10 mg/day. ,



PREVENTION

Infection prevention practices, including hand hygiene, use of contact precautions, maintenance of environmental cleanliness, and implementation of antimicrobial stewardship, can help prevent spread of Acinetobacter in the hospital. During outbreaks, cohorting of patients and use of dedicated staff to care for cohorted patients might be necessary to control spread. Removal of indwelling devices from patients, including vascular catheters and endotracheal tubes, can help prevent colonization and infection with Acinetobacter.  

PROGNOSIS

Invasive infection due to Acinetobacter has been associated with crude mortality rates in excess of 50% and increases in duration of ICU stay of 6 days and duration of total hospital stay of more than 14 days, with data indicating that infection is a cause of, not a marker of, poorer prognosis.16 Fatal A. baumannii infections in relatively immunocompetent patients appear to be related to clade B.17 With appropriate early empiric therapy, however, the mortality rate for community-acquired infection can be as low as 11%.18 By comparison, infections due to strains of MDR and XDR Acinetobacter are associated with increases in mortality and durations of hospitalization compared with more susceptible Acinetobacter strains.19

  STENOTROPHOMONAS MALTOPHILIA DEFINITION  

S. maltophilia has attracted significant attention because it is often MDR due to intrinsic and acquired factors. S. maltophilia contributes significantly to morbidity, but usually not mortality, in immunocompromised patients.

The Pathogen

S. maltophilia are gram-negative bacteria that need methionine for growth and, like Pseudomonas aeruginosa and Acinetobacter species, are non–lactose fermenters. S. maltophilia possess flagella that are multitrichous (more than one flagellum arising from the pole) and are distinguished from P. aeruginosa as oxidase negative. Colonies of S. maltophilia may appear pale yellow or lavender-green on blood agar plates. The organism emits a mild ammonia-like odor that is used for preliminary identification. S. maltophilia can colonize inanimate surfaces in the hospital, including catheters, intravenous fluid, water supplies, and hospital equipment. This pathogen may also survive in hospital-grade disinfectant. Other health care– associated sources of S. maltophilia include contaminated intravenous fluids, hospital water and ice supplies, nebulizers, dialysis machines, ventilator circuits, thermometers, blood gas analyzers, intra-abdominal balloon pumps, and central venous or arterial pressure monitors.  

EPIDEMIOLOGY

Despite its harboring many resistance genes, little is known about the virulence of S. maltophilia. Nevertheless, a number of risk factors are associated with infection by S. maltophilia (Table 291-1). Although S. maltophilia is regarded as primarily a nosocomial pathogen (it is the third most common nonfermenting gram-negative bacillus health care–associated pathogen), communityacquired infection can occur.20 S. maltophilia infection primarily affects patients who are immunocompromised (including patients with hematologic malignant

TABLE 291-1 RISK FACTORS FOR INFECTION BY STENOTROPHOMONAS MALTOPHILIA Prolonged hospitalization or intensive care unit stay Intravascular catheters Indwelling devices Mechanical ventilation or tracheostomy Neutropenia or cytotoxic chemotherapy Solid organ transplantation Immunocompromised state Mucositis Malignant disease Chronic lung disease (especially cystic fibrosis and chronic obstructive pulmonary disease) HIV infection Hemodialysis Antibiotic exposure (especially to carbapenems, extended-spectrum cephalosporins, and fluoroquinolones) Exposure to other patients with S. maltophilia

neoplasms21 and patients who underwent solid organ transplantation), patients cared for in ICUs who are mechanically ventilated, hemodialysis patients with intravascular catheters, neonates, and patients with cystic fibrosis. Colonization or infection with S. maltophilia among patients with cystic fibrosis (Chapter 83) has been associated with reduction in lung function.  

CLINICAL MANIFESTATIONS

Respiratory Tract

The respiratory tract is the most common site of isolation of S. maltophilia in the hospital. Surveillance programs report a rate of culture-positivity of S. maltophilia from hospitalized patients with pneumonia of more than 3% in the United States. Nosocomial pneumonia due to S. maltophilia is often associated with significant pulmonary disease such as emphysema, bronchiectasis, lung transplantation, or endobronchial obstruction. In addition, many of these patients are on mechanical ventilators, have tracheostomy tubes in place, or are receiving broad-spectrum antibiotics. Among mechanically ventilated patients, it can be challenging to differentiate colonization due to S. maltophilia from infection. CAP due to S. maltophilia has been reported but is very rare. The incidence of respiratory tract infection due to S. maltophilia in patients with cystic fibrosis (Chapter 83) is increasingly being reported. There may be clear links with resistance to antipseudomonal antibiotics (tobramycin, imipenem, ceftazidime) used to treat these patients. The presence of chronic S. maltophilia infection has been associated with increases in cystic fibrosis exacerbations, decline in lung function, hospitalizations, need for lung transplantation, and mortality. It remains unclear whether S. maltophilia has a causative association with poor outcomes or is merely a marker for severe underlying disease.22

Blood Stream Infection and Endocarditis

Bacteremia may be secondary to a respiratory, urinary, or gastrointestinal source but is most commonly due to infection of an indwelling intravascular device. In many cases, except for those involving intravenous catheters, the portal of entry is not apparent. Immunocompromised patients with indwelling intravascular catheters, including those with hematologic malignant disease, are at particularly high risk for CLABSI due to S. maltophilia. In these cases, removal of the infected catheter is an important component of management. On occasion, an environmental reservoir or contaminated vascular access device is linked to the presence of bacteremia. Intravenous drug users are at especially high risk for contaminating prosthetic valves with S. maltophilia. In the case of endocarditis, favorable outcomes are reported with antimicrobial therapy, but surgery may also be required.

Urinary Tract Infection

Although S. maltophilia is frequently recovered from urine specimens in patients with indwelling urinary catheters, the role of this organism as a pathogen in this setting is unclear.

Meningitis

Cases of central nervous system infection due to S. maltophilia are reported rarely. These are often associated with central nervous system devices or antecedent neurosurgery.

Skin and Soft Tissue Infection

S. maltophilia can be isolated from postoperative wounds, but its role as a pathogen in this setting is unclear. In contrast, S. maltophilia can cause burn wound sepsis, manifested as a syndrome very similar to ecthyma gangrenosum (Chapter 412) in immunocompromised oncology patients.

TREATMENT  As a pathogen that is an increasingly important cause of nosocomial infections, S. maltophilia exhibits intrinsic resistance to many antibiotics.23 The complexity of resistance genes and mechanisms is summarized in Table 291-2. Resistance to imipenem, piperacillin-tazobactam, ceftazidime, and aminoglycosides is common. Despite significant resistance to many agents, trimethoprim-sulfamethoxazole remains the drug of choice (10 to 15 mg/kg/day IV, based on the trimethoprim component). Treatment duration is uncertain, but it is usually 7 to 14 days but might be longer for deep seated or endovascular infections. Although overall there is excellent susceptibility, resistance to trimethoprim-sulfamethoxazole has become more common, particularly in cystic fibrosis patients. In vitro studies indicate that ticarcillin-clavulanate (3.1 g every 6 hours), minocycline (200 mg followed by 100 mg every 12 hours, not to exceed 400 mg in 24 hours), some of

TABLE 291-2 MECHANISMS OF RESISTANCE IN STENOTROPHOMONAS MALTOPHILIA DRUG

MECHANISM OF RESISTANCE

β-Lactams, including imipenem

L1 and L2 β-lactamases Outer membrane permeability/efflux

Aminoglycosides

Aminoglycoside-modifying enzymes, transport

the fluoroquinolones (particularly levofloxacin and moxifloxacin), and tigecycline may be useful agents. Polymyxin-based regimens may serve as alternatives in the face of resistance to trimethoprim-sulfamethoxazole, but susceptibility should be confirmed prior to administration (see earlier for details regarding polymyxin treatment). Aztreonam is effective against strains exhibiting resistance to β-lactams by metallo-β-lactamases, but can be compromised if the co-production of other β-lactamases, most notably the chromosomal L2 β-lactamase, is present. The addition of the β-lactamase inhibitor avibactam to aztreonam has been explored in an effort to preserve aztreonam activity in the setting of β-lactamase production.24 In certain instances, ceftazidime combined with avibactam was added to aztreonam to restore susceptibility. Various combination regimens for invasive S. maltophilia infections have been studied and demonstrated synergy but their efficacy in clinical settings remains unclear. Inhaled antibiotics are a therapeutic option but their clinical effectiveness remains unknown.



PROGNOSIS

Among patients with VAP or bacteremia due to S. maltophilia, crude mortality rates in the range of 60% have been reported. Bacteremia has been associated with poor outcomes, particularly in patients with malignancy or severe immuncompromising conditions. The attributable mortality of invasive S. maltophilia infection has been reported to be between 12 and 37.5%, but patients who receive early appropriate antibiotic therapy appear to do better.  

PREVENTION

Standard infection control practices including hand hygiene and thorough environmental cleaning can help prevent spread of Stenotrophomonas in the hospital. In outbreak settings, contact precautions and possibly cohorting of patients have been used to control spread.

  Grade A References A1. Durante-Mangoni E, Signoriello G, Andini R, et al. Colistin and rifampicin compared with colistin alone for the treatment of serious infections due to extensively drug-resistant Acinetobacter baumannii: a multicenter, randomized clinical trial. Clin Infect Dis. 2013;57:349-358. A2. Paul M, Daikos GL, Durante-Mangoni E, et al. Colistin alone versus colistin plus meropenem for treatment of severe infections caused by carbapenem-resistant Gram-negative bacteria: an open-label, randomised controlled trial. Lancet Infect Dis. 2018;18:391-400.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 291  DISEASES CAUSED BY ACINETOBACTER AND STENOTROPHOMONAS SPECIES  

GENERAL REFERENCES 1. Wong D, Nielsen TB, Bonomo RA, et al. Clinical and pathophysiological overview of Acinetobacter infections: a century of challenges. Clin Microbiol Rev. 2017;30:409-447. 2. Oh DH, Kim YC, Kim EJ, et al. Multidrug-resistant Acinetobacter baumannii infection in lung transplant recipients: risk factors and prognosis. Infect Dis (Lond). 2019;51:493-501. 3. García-Patiño MG, García-Contreras R, Licona-Limón P. The immune response against Acinetobacter baumannii, an emerging pathogen in nosocomial infections. Front Immunol. 2017;8:1-10. 4. Huang H, Chen B, Liu G, et al. A multi-center study on the risk factors of infection caused by multidrug resistant Acinetobacter baumannii. BMC Infect Dis. 2018;18:1-6. 5. Wood GC, Swanson JM. An update on aerosolized antibiotics for treating hospital–acquired pneumonia in adults. Ann Pharmacother. 2017;51:1112-1121. 6. Dexter C, Murray GL, Paulsen IT, et al. Community-acquired Acinetobacter baumannii: clinical characteristics, epidemiology and pathogenesis. Expert Rev Anti Infect Ther. 2015;13:567-573. 7. Shin B, Park W. Antibiotic resistance of pathogenic Acinetobacter species and emerging combination therapy. J Microbiol. 2017;55:837-849. 8. Lee CR, Lee JH, Park M, et al. Biology of Acinetobacter baumannii: pathogenesis, antibiotic resistance mechanisms, and prospective treatment options. Front Cell Infect Microbiol. 2017;7:1-35. 9. Velkov T, Dai C, Ciccotosto GD, et al. Polymyxins for CNS infections: pharmacology and neurotoxicity. Pharmacol Ther. 2018;181:85-90. 10. Isler B, Doi Y, Bonomo RA, et al. New treatment options against carbapenem-resistant Acinetobacter baumannii infections. Antimicrob Agents Chemother. 2019;63:1-17. 11. Espinal P, Pantel A, Rolo D, et al. Relationship between different resistance mechanisms and virulence in Acinetobacter baumannii. Microb Drug Resist. 2019;25:752-760. 12. Garnacho-Montero J, Timsit JF. Managing Acinetobacter baumannii infections. Curr Opin Infect Dis. 2019;32:69-76. 13. Tsuji BT, Pogue JM, Zavascki AP, et al. International consensus guidelines for the optimal use of the polymyxins: endorsed by the American College of Clinical Pharmacy (ACCP), European Society

1939.e1

of Clinical Microbiology and Infectious Diseases (ESCMID), Infectious Diseases Society of America (IDSA), International Society for Anti-infective Pharmacology (ISAP), Society of Critical Care Medicine (SCCM), and Society of Infectious Diseases Pharmacists (SIDP). Pharmacotherapy. 2019;39:10-39. 14. Pogue JM, Ortwine JK, Kaye KS. Clinical considerations for optimal use of the polymyxins: a focus on agent selection and dosing. Clin Microbiol Infect. 2017;23:229-233. 15. Pogue JM, Ortwine JK, Kaye KS. Are there any ways around the exposure-limiting nephrotoxicity of the polymyxins? Int J Antimicrob Agents. 2016;48:622-626. 16. Leão AC, Menezes PR, Oliveira MS, et al. Acinetobacter spp. are associated with a higher mortality in intensive care patients with bacteremia: a survival analysis. BMC Infect Dis. 2016;16:1-8. 17. Jones CL, Clancy M, Honnold C, et al. Fatal outbreak of an emerging clone of extensively drugresistant Acinetobacter baumannii with enhanced virulence. Clin Infect Dis. 2015;61:145-154. 18. Vazquez Guillamet C, Kollef MH. Acinetobacter pneumonia: improving outcomes with early identification and appropriate therapy. Clin Infect Dis. 2018;67:1455-1462. 19. Choi SH, Cho EB, Chung JW, et al. Changes in the early mortality of adult patients with carbapenemresistant Acinetobacter baumannii bacteremia during 11 years at an academic medical center. J Infect Chemother. 2019;25:6-11. 20. Adegoke AA, Stenstrom TA, Okoh AI. Stenotrophomonas maltophilia as an emerging ubiquitous pathogen: looking beyond contemporary antibiotic therapy. Front Microbiol. 2017;8:1-18. 21. Chang YT, Lin CY, Chen YH, et al. Update on infections caused by Stenotrophomonas maltophilia with particular attention to resistance mechanisms and therapeutic options. Front Microbiol. 2015;6:1-20. 22. Berdah L, Taytard J, Leyronnas S, et al. Stenotrophomonas maltophilia: a marker of lung disease severity. Pediatr Pulmonol. 2018;53:426-430. 23. Wang Y, He T, Shen Z, et al. Antimicrobial resistance in Stenotrophomonas spp. Microbiol Spectr. 2018;6:1-14. 24. Mojica MF, Papp-Wallace KM, Taracila MA, et al. Avibactam restores the susceptibility of clinical isolates of Stenotrophomonas maltophilia to aztreonam. Antimicrob Agents Chemother. 2017;61:1-5.

1939.e2

CHAPTER 291  DISEASES CAUSED BY ACINETOBACTER AND STENOTROPHOMONAS SPECIES  

REVIEW QUESTIONS 1. Your patient has a blood culture turn positive. The preliminary result from the laboratory indicates possible Acinetobacter baumannii. Which of the following agents is most likely to have in vitro activity against A. baumannii? A . Ertapenem B. Ampicillin-sulbactam C. Piperacillin-tazobactam D. Trimethoprim-sulfamethoxazole E. Doxycycline Answer: B  Increasing resistance to a variety of antimicrobial agents complicates the treatment of Acinetobacter species infections today. When it is susceptible, Acinetobacter is typically treated with sulbactam (in the formulation of ampicillin-sulbactam in the United States at a dose of 3 g intravenously every 6 hours). Other options are the carbapenems imipenem and meropenem and the broad-spectrum cephalosporins. See treatment section under Acinetobacter species. 2. Your patient has a blood culture turn positive. The preliminary result from the laboratory indicates possible Stenotrophomonas maltophilia. Which of the following agents is most likely to have in vitro activity against S. maltophilia? A . Ertapenem B. Ampicillin-sulbactam C. Piperacillin-tazobactam D. Trimethoprim-sulfamethoxazole E. Meropenem Answer: D  S. maltophilia is an increasingly important cause of the nosocomial infections, and it exhibits intrinsic resistance to many antibiotics. Nevertheless, trimethoprim-sulfamethoxazole remains the drug of choice, at a dose of 10 to 15 mg/kg/day intravenously, based on the trimethoprim component, and it is given for 14 days or longer with positive blood cultures. See treatment section under Stenotrophomonas maltophilia. 3. Your patient has been admitted to the intensive care unit because of respiratory failure. He is intubated and mechanically ventilated. The intensive care unit is in the midst of an Acinetobacter outbreak. What would be the most effective intervention to decrease risk for development of an infection due to Acinetobacter in your patient? A . Extubate the patient and remove him from mechanical ventilation as soon as possible. B. Institute contact precautions (i.e., gowns and glove precautions). C. Prescribe prophylaxis for the patient with intravenous imipenem. D. Clean frequently touched objects in the patient’s room with bleach. E. Prescribe prophylaxis for the patient with Saccharomyces through a nasogastric tube. Answer: A  During outbreaks in the hospital, cohorting of patients and use of dedicated staff to care for the cohorted patients might be necessary to control spread. Removal of indwelling devices from the patients, including vascular catheters and endotracheal tubes, can help prevent colonization and infection with Acinetobacter. See prevention section under Acinetobacter species.

4. You are managing a patient with a deep wound infection resulting from a combat injury that occurred in Iraq. After incision and drainage, preliminary results from operative cultures indicate growth of possible A. baumannii. The patient has an anaphylactic reaction to β-lactam antibiotics. Which agent would be most likely to be an effective therapeutic agent for the patient? A . Aztreonam B. Ciprofloxacin C. Tigecycline D. Chloramphenicol E. Rifampin Answer: C  Since initial reports of the outbreak of A. baumannii among military personnel in Iraq and Afghanistan, reports of severe wound infections and skin and soft tissue infections by this pathogen have increased in frequency. Multiple drug resistance and the presence of copathogens frequently complicate treatment. Most cases require surgical débridement and are associated with substantial mortality. Tigecycline might be the only available agent with in vitro activity against this pathogen in this setting. See the section on wound, burn, and skin and soft tissues infections under clinical manifestations of Acinetobacter species. 5. Your patient with cystic fibrosis has colonization of the respiratory tract with Stenotrophomonas maltophilia. The patient asks you what the significance of this colonization is and what can be done. Which of the following is the most accurate answer? A . Because Stenotrophomonas is only a colonizer, there is no association with poor outcomes. B. Treatment of colonization with an active antimicrobial improves the outcomes of cystic fibrosis patients. C. Stenotrophomonas increases mortality risk among cystic fibrosis patients primarily by causing necrotizing pneumonia. D. Pneumonia due to S. maltophilia in a patient with cystic fibrosis is an indication for lung transplantation. E. Colonization with S. maltophilia among patients with cystic fibrosis has been associated with reduction in lung function. Answer: E  Colonization or infection with S. maltophilia among patients with cystic fibrosis has been associated with reduction in lung function. See reference 15 and the epidemiology section under Stenotrophomonas maltophilia.

CHAPTER 292  SALMONELLA INFECTIONS (INCLUDING ENTERIC FEVER)  

1939

292  SALMONELLA INFECTIONS (INCLUDING ENTERIC FEVER) JOHN A. CRUMP



DEFINITION

Many of the more than 2500 serovars (also called serotypes) of Salmonella enterica subspecies enterica infect humans and cause a range of clinical conditions from asymptomatic intestinal carriage to intestinal infection to invasive disease with extraintestinal complications. Each serovar designation follows the species name (e.g., Salmonella enterica subspecies enterica serovar Typhimurium) and is frequently abbreviated as simply Salmonella followed by the serovar name (e.g., Salmonella Typhimurium).

CHAPTER 292  SALMONELLA INFECTIONS (INCLUDING ENTERIC FEVER)  

ABSTRACT

Salmonella Typhi, Salmonella Paratyphi A, Salmonella Paratyphi B, Salmonella Paratyphi C, and Salmonella Sendai are either solely or almost exclusively pathogens of humans; they cause primarily enteric fever rather than diarrhea, and transmission is usually through water or food. Other serovars of Salmonella, described here as nontyphoidal Salmonella, have reservoirs in warm-blooded animals and cause human illness after the consumption of contaminated meat or animal products, contamination of produce or water by animal feces or animal products, or exposure to contaminated animals and their environments. Antimicrobial resistance is a growing problem among Salmonella, and resistance to ampicillin, chloramphenicol, and trimethoprim-sulfamethoxazole is widespread. Resistance to extended-spectrum cephalosporins and fluoroquinolones is emerging. Although diarrhea due to nontyphoidal Salmonella is usually managed symptomatically, enteric fever and invasive nontyphoidal Salmonella disease require timely and appropriate antimicrobials. Vaccines are available to prevent typhoid fever, and prevention efforts for other Salmonella infections focus on the provision of safe water and food.

KEYWORDS

serovar enteric fever bacteremia diarrhea paratyphoid fever typhoid fever Salmonella

1939.e3

1940

CHAPTER 292  SALMONELLA INFECTIONS (INCLUDING ENTERIC FEVER)  

The Pathogen

Members of the family Enterobacteriaceae (also see Chapter 289), salmonellae are gram-negative, non-spore-forming bacilli. They can be differentiated into more than 2500 serovars by their somatic (O) antigens, which are composed of lipopolysaccharides and are part of the cell wall, and by their flagellar (H) and capsular (Vi) antigens. Salmonella serogroups were traditionally designated by letters based on O antigens (e.g., A, B, C1, C2). The growing number of serogroups then made it necessary to move to a numeric designation. During the transition, the traditional letter-based serogroup were retained in brackets after the numeric designation (e.g., O:2 [A], O:4 [B], O:6,7 [C1], O:8 [C2]). Some of the important serovars and their serogroups are Typhi (group O:9 [D1]), Choleraesuis (group O:7 [C1]), Typhimurium (group O:4 [B]), and Enteritidis (group O:9 [D1]). Salmonella Enteritidis and Typhimurium are the most common nontyphoidal serovars causing human disease. Today, serovar can be inferred by whole genome sequencing and subdivision within serovars, useful for epidemiologic investigations, is usually achieved by molecular or genomic subtyping methods.1  

EPIDEMIOLOGY

Salmonella Typhi, Salmonella Paratyphi A, Salmonella Paratyphi B, Salmonella Paratyphi C, and Salmonella Sendai are either solely or almost exclusively pathogens of humans; they cause primarily enteric fever rather than diarrhea, and transmission is usually through water or food. As a result of modern sewage and water treatment facilities and improved food safety practices, typhoid fever and paratyphoid fever have become rare in high-income countries but remain a problem in countries that lack adequate sanitation and a safe water supply. There are usually fewer than 500 cases of typhoid fever each year in the United States, mainly acquired abroad2; in contrast, an estimated 10.9 million cases occurred globally in 2017.3 Other serovars of Salmonella (described here as nontyphoidal Salmonella) have reservoirs in warm-blooded animals and cause human illness after the consumption of contaminated meat or animal products, contamination of produce or water by animal feces or animal products, or exposure to animals and their environments. Some nontyphoidal Salmonella serovars appear frequently in particular animal species, and about 10% of Salmonella infections may be attributable to animal exposure, including small pet turtles.4 Salmonella Enteritidis has a reservoir in chickens, and infection is often linked to the consumption of undercooked eggs and poultry products or exposure to live chicks. Such a relationship is less clear for some other nontyphoidal serovars (e.g., Salmonella Typhimurium). Foodborne nontyphoidal Salmonella was estimated to be associated with approximately 1.0 million domestically acquired illnesses and 378 deaths in the United States in 2006. In the United States, a disproportionate number of infections occur in July through October, probably related to warm weather. Salmonella infections are most common among infants and children younger than 5 years of age. Worldwide, nontyphoidal Salmonella was estimated to cause 153.1 million illnesses and 57,000 deaths in 2010, with 52% of these estimated to be foodborne.5

Antimicrobial Resistance

Salmonellae have become increasingly resistant to antimicrobial agents, often by acquiring resistance transfer factors (e.g., plasmid mediated). Mutations in the waaY, phoP, and pmrB genes also confer resistance to antimicrobial peptides.6 It is thought that antimicrobial resistance in the human-restricted salmonellae (e.g., Salmonella Typhi) is driven primarily by antimicrobial use in humans, whereas antimicrobial resistance among the nontyphoidal salmonellae (e.g., Salmonella Typhimurium) is associated with the use of antimicrobial agents in farm animals. Among Salmonella Typhi isolated in the United States from 2008 through 2012, 10 to 13% were resistant to the traditional first-line antimicrobials ampicillin, chloramphenicol, and trimethoprim-sulfamethoxazole, whereas 69% showed decreased fluoroquinolone susceptibility. Resistance to extended-spectrum cephalosporins is emerging among Salmonella Typhi in south Asia but remains rare among Salmonella Paratyphi A. Among human nontyphoidal Salmonella bloodstream isolates in the United States from 2003 to 2013, resistance to three or more antimicrobial classes was common. In addition, 5.0% of human nontyphoidal Salmonella bloodstream isolates were resistant to ceftriaxone and 4.5% were resistant to ciprofloxacin.7 Susceptibility breakpoints and interpretive criteria were established for azithromycin in 2015 in response to reports of Salmonella with elevated azithromycin minimum inhibitory concentrations.



PATHOBIOLOGY

Etiology

Salmonellae are transmitted by the ingestion of fecally contaminated food or water; contact with animals, their environments, and other fomites; and, rarely, close contact with infected persons (e.g., oral-anal intercourse). The ultimate sources of contamination are humans or animals that are acutely ill or are shedding the organism without symptoms.

Contaminated Animal Products

Salmonella infection in humans usually occurs from ingestion of contaminated animal food products, most often eggs, poultry, and meat. Salmonella Choleraesuis is associated with pig products, Salmonella Dublin with cattle and unpasteurized milk from cattle, and Salmonella Enteritidis with poultry and poultry products, including eggs. Fecal material on poultry and other animal carcasses can spread at slaughterhouses, such as when many poultry carcasses are placed in the same hot-water tank to remove feathers. Salmonella contaminating carcasses can multiply to high levels if meat or other animal products are not refrigerated. Human illness may result if such animal products are inadequately cooked or if utensils or other uncooked foods are cross-contaminated during preparation. A wide range of foods can be contaminated with animal or human feces, from production on the farm through consumption in the home. Reports of produceassociated Salmonella outbreaks, due to contamination by animal or human feces during production, are increasing.8 Salmonella outbreaks have occurred from contaminated cheese, ice cream, vegetables, fruit, juice, and alfalfa sprouts.

Contaminated Food and Water Contamination by Pets

Salmonella infections may be acquired after contamination of food or water with the feces of pet turtles, chicks, ducks, birds, dogs, cats, and many other species. Contamination by Humans

Salmonella infection can also be acquired by eating food or by drinking water contaminated by human fecal shedders who have not adequately washed their hands. Infection has been spread by the fecal-oral route among children, by contaminated enema and fiberoptic instruments, by diagnostic and therapeutic preparations made from animal or insect products (e.g., pancreatic extract, carmine dye), and from intentional or unintentional contamination of restaurant salad bars. Outbreaks of salmonellosis may occur in institutionalized patients, who are probably more prone to the development of Salmonella infections for three reasons. First, within institutionalized populations, there is an increased prevalence of underlying diseases that decrease host defense mechanisms against salmonellae, such as disorders of gastric acidity and intestinal motility; second, the use of antimicrobial agents modifies the normal, protective intestinal flora; and third, institutional food prepared in bulk may be more likely to be contaminated than are individually prepared meals. Outbreaks in nurseries and among the elderly in nursing homes are associated with the highest case-fatality ratios (>5%).

Contact with Animals and Their Environments

Both healthy and sick animals may harbor and shed Salmonella. Transmission of Salmonella from animals and their environment to humans occurs primarily by the fecal-oral route. Animal hides and saliva often harbor fecal organisms, and transmission can occur when persons pet, touch, feed, or are licked by animals. Transmission has also been associated with contaminated animal bedding, flooring, barriers, other environmental surfaces, and clothing and shoes. Contact with calves, turtles and other reptiles, rodents, and young poultry and their environments has been associated with Salmonella outbreaks. Humans may also become infected when animals come into contact with their food or water. Infections can be prevented by education, supervision of animal contact, provision and promotion of handwashing facilities, and separation of food handling and consumption from animal areas.

Contact with Infected Persons

Close contact with persons shedding Salmonella is an occasional source of infection. Transmission has been documented among persons handling feces (e.g., parents changing the diapers of an infected infant) and is associated with certain sexual practices (e.g., oral-anal intercourse).

CHAPTER 292  SALMONELLA INFECTIONS (INCLUDING ENTERIC FEVER)  

Pathophysiology

After the ingestion of organisms, the likelihood for development of infection, as well as the severity of infection, is related to the dose, the virulence of the Salmonella strain, and the status of host defense mechanisms. Usually at least 102 to 103 bacteria are required to produce clinical infection in a normal host. Higher inocula are associated with increased disease severity, whereas smaller inocula are more likely to result in transient intestinal carriage. Gastric acid serves as a host defense mechanism by killing many of the ingested organisms, and intestinal motility is probably another host defense mechanism. In the absence of or with a decrease in gastric acidity (e.g., in infants and the elderly; after gastrectomy, vagotomy, or gastroenterostomy; or with the use of drugs that reduce gastric acidity) and with decreased intestinal motility (e.g., the use of antimotility drugs), much smaller inocula can produce infection, and the infection tends to be more severe. Administration of antimicrobial agents before the ingestion of salmonellae can markedly reduce the size of the inoculum needed to produce infection, presumably by reducing the concentration of protective bowel flora. Although any Salmonella serovar can produce any of the Salmonella syndromes (transient asymptomatic carrier state, enterocolitis, enteric fever, bacteremia, and chronic carrier state), each serovar tends to be associated with certain syndromes much more frequently than with others. For example, Salmonella Anatum usually causes asymptomatic intestinal infection, whereas Salmonella Typhimurium generally causes enterocolitis. Salmonella Choleraesuis is more likely to produce bacteremia (often with metastatic infection) than asymptomatic infection or enterocolitis, and some serovars such as Salmonella Typhi and Salmonella Paratyphi A are most likely to cause enteric fever as well as the chronic carrier state. Fortunately, most Salmonella serovars are of relatively low pathogenicity for humans. Therefore, although food products are commonly contaminated, large outbreaks tend to involve the more virulent serovars. To produce infection, invasion must occur across the mucosa of the intestine. When the organisms reach the lamina propria, an influx of polymorphonuclear leukocytes serves as a host defense mechanism to prevent invasion of the lymphatics. Certain serovars seem to have a greater ability than others to invade the lymphatics and subsequently to produce bacteremia (e.g., Salmonella Choleraesuis and Salmonella Dublin, which commonly produce bacteremia after intestinal infection). Both the small intestine and the colon are involved in the inflammatory process. In the case of Salmonella Typhi and other causes of enteric fever, salmonellae invade the mononuclear phagocytes in Peyer patches in the ileum and mesenteric lymph nodes. Some intracellular salmonellae form a nonreplicating population of “persisters” that could provide a reservoir for relapsing infection; intracellular persistence is determined by conditions in the vacuolar environment of the infected cells. Others multiply intracellularly and are carried through the lymphatic system and blood stream to the liver, spleen, bone marrow, and other parts of the reticuloendothelial system. When in the reticuloendothelial system, they multiply intracellularly in mononuclear phagocytes and produce the systemic manifestations of enteric fever. The onset of fever is associated with bacteremia and the release of cytokines (e.g., tumor necrosis factor and interleukins) from mononuclear phagocytes. Ulcerations over Peyer patches are responsible for the intestinal manifestations of enteric fever, such as pain, perforation, and bleeding. In Salmonella enterocolitis, the organisms remain localized in the intestinal mucosa, and diarrhea results from the inflammation produced by polymorphonuclear leukocytes. In addition, watery diarrhea may occur, apparently the result of the secretion of water and electrolytes by small intestinal epithelial cells in response to an enterotoxin secreted by some of the Salmonella strains or in response to tissue mediators of inflammation.9 Patients with diseases that impair host defense mechanisms seem to have an increased frequency of severe Salmonella infection. A striking association has been recognized between diseases producing hemolysis and Salmonella bacteremia. Specifically, Salmonella bacteremia is common in patients with sickle cell disorders, malaria, and bartonellosis. In fact, because of the frequency of Salmonella bacteremia in sickle cell diseases and the underlying bone disease in these patients to which salmonellae localize, these organisms are the most common cause of osteomyelitis in patients with sickle cell disorders (Chapter 154). Prolonged Salmonella bacteremia may occur in patients with hepatosplenic schistosomiasis, possibly related to localization on and in the intravascular schistosomes. Patients with lymphoma and leukemia are also more prone to the development of Salmonella bacteremia. A markedly increased frequency and severity of Salmonella infections in general have been observed in patients with human immunodeficiency virus (HIV) infection, particularly

1941

those with CD4+ T-lymphocyte counts less than 200 cells/μL. Prolonged and recurrent, refractory Salmonella bacteremia is common among these patients. Other risk factors that increase the frequency and severity of Salmonella infection are extremes of age, immunocompromised states (e.g., from immunosuppressive agents), malnutrition, and probably diabetes. Nontyphoidal Salmonella serovars are a leading cause of community-acquired blood stream infection in sub-Saharan Africa, where children younger than 3 years and HIV-infected adults carry most of the burden of invasive disease.  

CLINICAL MANIFESTATIONS

Asymptomatic Intestinal Carrier State

The asymptomatic intestinal carrier state may result from inapparent infection (which is the most common form of Salmonella infection), or it may follow clinical disease (in which case the patient becomes a convalescent carrier). The carrier state is usually self-limited to several weeks to months, with the prevalence of positive stool cultures rapidly decreasing over time. By 1 year, far less than 1% of carriers still have positive stools. The main exception is Salmonella Typhi; about 3% of those infected excrete the organism for life. Women and older men are most likely to become chronic carriers of Salmonella Typhi, related to the presence of biliary tract disease, especially calculi. A patient who has had salmonellae in stool for 1 year (chronic carrier) is likely to become a lifelong carrier; the reservoir is in the biliary tree, usually in calculi in the gallbladder. Patients with Schistosoma haematobium infection are predisposed to become chronic urinary carriers of salmonellae.

Enterocolitis

After an incubation period of usually 12 to 48 hours, the illness starts suddenly with crampy abdominal pain and diarrhea. Nausea and vomiting may occur but are usually not prominent or persistent. The diarrhea may be watery and of large or small volume. Stools may contain mucus and are occasionally bloody. Polymorphonuclear leukocytes are present in the stool. Diarrhea may be mild or severe, with up to 20 to 30 stools a day. Fever is present in most patients, and the temperature may reach 40° C (104° F) or higher. The abdomen is tender to palpation. Transient bacteremia may occur and is most commonly seen in infants, the elderly, and patients with impaired host defense mechanisms. Symptoms generally improve during a period of days, with fever lasting no more than 2 to 3 days and diarrhea lasting no more than 5 to 7 days. However, these symptoms occasionally persist for up to 14 days. More severe disease is seen with malnutrition, inflammatory bowel disease, and HIV infection. Reactive arthritis may follow enterocolitis in up to 7% of cases, especially among those with the HLA-B27 phenotype.

Enteric Fever

Enteric fever is produced by Salmonella Typhi (typhoid fever), Salmonella Paratyphi A, B, and C (paratyphoid fever), and occasionally other serovars. Sometimes it immediately follows classic enterocolitis caused by the same organism. The syndrome is characterized by prolonged, sustained fever and may be associated with relative bradycardia, splenomegaly, rose spots, and leukopenia. Therapy aborts the course of the disease. The following is a description of untreated illness. After an incubation period of 5 to 21 days (generally 7 to 14 days), fever and malaise develop, often associated with cough. A small proportion of patients may have diarrhea during the incubation period. The fever tends to rise in a stepwise fashion during the first few days to a week and then becomes sustained, usually at 39.4° to 40.0° C (103° to 104° F) or higher. Relative bradycardia is seen in up to half of patients. Apathy, confusion, delirium, and even psychosis may occur. Abdominal distention, pain, and tenderness may occur in the first week and may be associated with diarrhea or constipation; these symptoms are generally more pronounced during the second week of fever. Most patients have abdominal tenderness during the course of the illness. In about 30% of patients, rose spots develop on the abdomen or chest (or both) toward the end of the first week or during the second week of fever. These faint, salmon-colored maculopapular lesions are subtle and may be difficult to see, particularly in dark-skinned patients. Salmonellae can be cultured from punch biopsies of these lesions. Hepatosplenomegaly occurs in about half of patients. Leukopenia and neutropenia are seen in about 20%. Abnormal liver function test results are common. After 2 weeks of illness, the severe complications of intestinal hemorrhage and perforation related to necrosis of Peyer patches may be observed in about 5% of patients. These perforations may require surgical as well as medical therapy and can occur even in a patient treated with antimicrobials. Intestinal perforation is the leading cause of death from enteric fever.

1942

CHAPTER 292  SALMONELLA INFECTIONS (INCLUDING ENTERIC FEVER)  

The illness usually resolves by the end of the fourth week in an untreated patient. Relapse may occur in untreated as well as in treated patients, but the illness is milder than the original episode. Rarely, some of the following complications may occur: pancreatitis, cholecystitis, infective endocarditis, meningitis, pneumonia, hepatic or splenic abscess, orchitis, or focal infection at virtually any site.

Bacteremia

Patients with Salmonella bacteremia usually complain of fever and chills lasting days to weeks. Gastrointestinal symptoms are unusual, but in some patients Salmonella bacteremia follows classic enterocolitis. Other symptoms are nonspecific, such as malaise, anorexia, and weight loss. Metastatic infection of bones, joints, aneurysms (particularly of the abdominal aorta), meninges (mainly in infants), pericardium, pleural space, lungs, heart valves, cysts, uterine myomas, malignant neoplasms, and other sites is common, and symptoms may be related to the site of metastatic infection. Stool cultures are often negative for salmonellae, but blood cultures are positive. Although any Salmonella serovar can produce bacteremia, Salmonella Dublin, Salmonella Choleraesuis, Salmonella Heidelberg, Salmonella Oranienburg, Salmonella Panama, and Salmonella Sandiego are associated with increased likelihood of bacteremia.10 Salmonella bacteremia occurs with increased frequency in infants, the elderly, and patients with diseases associated with hemolysis (e.g., sickle cell diseases, malaria, bartonellosis), HIV infection, lymphoma, leukemia, disseminated histoplasmosis, and perhaps systemic lupus erythematosus.11 Localization to bone is common in patients with sickle cell diseases (Chapter 154). In Africa, the common symptoms of invasive nontyphoidal Salmonella disease, which is seen predominantly in patients with HIV infection, malaria, and malnutrition, are fever, hepatosplenomegaly, and respiratory symptoms; features of enterocolitis are often absent. Prolonged Salmonella bacteremia lasting for months may occur in patients with hepatosplenic schistosomiasis. In patients with HIV infection, recurrent, relapsing Salmonella bacteremia may develop, which may be difficult to cure with antimicrobial agents.  

DIAGNOSIS

Although Salmonella enterocolitis is an invasive disease, the differential diagnosis includes all causes of acute diarrhea, including invasive bacteria such as Campylobacter jejuni, Shigella species, invasive Escherichia coli, Yersinia enterocolitica, and Vibrio parahaemolyticus; toxigenic bacteria such as Vibrio cholerae, enterotoxigenic E. coli, enterohemorrhagic E. coli (e.g., E. coli O157:H7), Staphylococcus aureus, Bacillus cereus, Clostridium perfringens, and Clostridium difficile; viruses; and protozoa such as Entamoeba histolytica, Giardia intestinalis, and Cryptosporidium species. The invasive bacterial causes of diarrhea, enterohemorrhagic E. coli and C. difficile infection, are also associated with polymorphonuclear leukocytes in stool, whereas bacterial toxigenic causes (other than C. difficile and enterohemorrhagic E. coli), viruses, and protozoa generally are not. The bacterial toxigenic causes of diarrhea other than C. difficile and enterohemorrhagic E. coli do not produce fever. Stool culture is definitive for the diagnosis of Salmonella enterocolitis, but by the time the results of stool culture are available, most patients are recovering. Culture-independent diagnostic testing is used increasingly, but positive results for Salmonella may require culture to yield an isolate for public health purposes. A stained smear of the stool usually demonstrates polymorphonuclear leukocytes. Serologic studies are of little clinical value in Salmonella enterocolitis, but they may be of use in epidemiologic studies. Culture-independent methods, including panel-based multiplex molecular diagnostics from stool specimens, are used increasingly by clinical laboratories. The differential diagnosis of Salmonella bacteremia includes virtually all acute infectious and noninfectious causes of fever, including bacteremia caused by other organisms. The diagnosis is proved by isolation of the microorganism from blood or from another normally sterile site. The differential diagnosis of enteric fever is broad and depends in part on the area of the world where the infection was acquired. All causes of sustained fever are in the differential diagnosis, including infective endocarditis, disseminated tuberculosis, brucellosis, tularemia, Mycoplasma pneumoniae infection, rickettsial infections, Q fever, and viral infections such as infectious mononucleosis. Depending on the site of acquisition, diseases such as malaria, amebic abscesses of the liver, and visceral leishmaniasis also enter into the differential diagnosis. The diagnosis of enteric fever is best proved by isolation of the microorganism from blood, stool, or bone marrow.12 During the first week of illness,

blood cultures are positive in about 90% of patients, but culture positivity decreases in the next 2 weeks to less than 50% during the third week of illness. Stool cultures are usually negative during the first week but are generally positive by the third week. Bone marrow cultures give the highest yield, with up to 95% being positive; they should be considered in suspected cases with negative blood cultures. Bone marrow cultures may be positive even after several days of antimicrobial treatment, when blood cultures have become negative. Urine cultures and cultures of punch biopsies of rose spots may also be positive. The string test to obtain samples of bile from the duodenum has likewise yielded positive cultures. The peripheral leukocyte count is usually normal, but leukopenia, which occurs in about 20% of cases, may be suggestive of enteric fever. Fecal leukocytes are generally present. The Widal and other serologic tests that detect serum antibodies against Salmonella Typhi are limited by shortcomings of both sensitivity and specificity and rarely provide useful information to guide management of the patient. Polymerase chain reaction and other molecular techniques lack sensitivity for diagnosis from blood and other specimens, but they have been used to determine the Salmonella serovar of bacterial isolates.

TREATMENT  Enterocolitis

The primary approach to the treatment of Salmonella enterocolitis is fluid and electrolyte replacement. Drugs with antiperistaltic effects, such as loperamide or diphenoxylate with atropine, can relieve cramps, but they should be used sparingly because they can prolong the diarrhea. Salmonella enterocolitis is self-limited, and antimicrobial therapy is usually not indicated, except perhaps in groups of patients at high risk for invasive disease. Antimicrobial therapy reportedly has little effect on the clinical course, and in some studies, it has prolonged the duration of Salmonella excretion in stool. In addition, most patients are improving by the time salmonellae or other bacterial pathogens are isolated from stool. The fluoroquinolones are active against virtually all bacterial pathogens that cause diarrhea (including salmonellae), except for C. difficile and many Campylobacter organisms. Thus, it is reasonable to use fluoroquinolones for patients with suspected or known Salmonella enterocolitis who are severely ill and suspected of being bacteremic. The threshold for antimicrobial treatment is also decreased in those at increased risk for severe illness (e.g., infants, the elderly, patients with sickle cell disease, immunosuppressed individuals). As an example, in adults, ciprofloxacin, 500 mg every 12 hours orally or 400 mg every 12 hours intravenously for 3 to 5 days, or until defervescence, has been widely used. An extended-spectrum cephalosporin such as ceftriaxone is an alternative. In the presence of gross bloody diarrhea, antimicrobial therapy should be withheld until the possibility of E. coli O157:H7 infection has been eliminated because antimicrobial therapy may increase the frequency of hemolytic-uremic syndrome. Other agents, such as amoxicillin and trimethoprim-sulfamethoxazole, have also been widely used in severely ill adults. However, many strains of Salmonella are now resistant to these agents.

Enteric Fever

Resistance to the traditional first-line antimicrobial agents (ampicillin, chloramphenicol, trimethoprim-sulfamethoxazole) has emerged worldwide among the salmonellae causing enteric fever. Consequently, alternative antimicrobial agents are now preferred. The fluoroquinolones remain agents of choice for the treatment of enteric fever in many parts of the world, A1  but resistance has become sufficiently common in other areas for alternative antimicrobials to be used routinely. Fluoroquinolones can be administered orally and have high bioavailability, they concentrate in bile and the bowel, and they often retain activity against multidrug-resistant strains of Salmonella Typhi and other causes of enteric fever. Most important, the fluoroquinolones have proved to be effective in the treatment of enteric fever, even with short courses (e.g., 3 to 7 days). The proportion of patients cured exceeds 95%, and relapse and chronic fecal carriage after therapy are uncommon. Ciprofloxacin (500 mg orally twice a day) for 7 to 14 days has been the fluoroquinolone of choice for enteric fever. If a patient cannot tolerate oral therapy, the fluoroquinolones can be administered intravenously. Reduced susceptibility and resistance to fluoroquinolones are increasingly reported in Salmonella Typhi and Salmonella Paratyphi A strains both in the United States and elsewhere and are associated with treatment failure. If decreased fluoroquinolone susceptibility or resistance is suspected or demonstrated, alternative agents include extended-spectrum cephalosporins (e.g., intravenous ceftriaxone) and azithromycin. Extended-spectrum cephalosporins such as ceftriaxone are recommended agents for the treatment of enteric fever in settings where fluoroquinolone

CHAPTER 292  SALMONELLA INFECTIONS (INCLUDING ENTERIC FEVER)  

resistance is common and where extended-spectrum cephalosporin resistance is rare. A2  A3  Ceftriaxone dosed at 1 to 2 g every 12 to 24 hours for adults and 75 mg/kg/day for children, given intravenously or intramuscularly for 10 to 14 days, results in cure of 95% of patients. Outbreaks of extensively drug resistant, including ceftriaxone resistant, Salmonella Typhi are occurring in south Asia. If the Salmonella isolate is shown to be susceptible by antimicrobial susceptibility testing, ampicillin, chloramphenicol, or trimethoprim-sulfamethoxazole may be considered. The ampicillin dose is 25 mg/kg intravenously every 6 hours. The use of chloramphenicol should be weighed against the risk for bone marrow toxicity. The chloramphenicol dose is 50 mg/kg/day orally, divided into four doses. Chloramphenicol can be given intravenously at the same dose if oral therapy is not possible. Trimethoprim-sulfamethoxazole (4/20 mg/kg intravenously or orally every 12 hours) is given for 10 to 14 days. Azithromycin (10 mg/kg/day orally for 7 days) is effective in the treatment of patients with uncomplicated typhoid fever caused by multidrug-resistant and some extensively drug-resistant strains. The oral route of administration makes it a particularly attractive choice in settings where multidrug resistance is common and intravenous extended-spectrum cephalosporins are impractical, unavailable, or too expensive. Severe disease due to extensively drug-resistant Salmonella Typhi may require carbapenems. Patients often require supportive care with intravenous saline, correction of electrolyte and acid-base disturbances, and, in the setting of intestinal bleeding, blood transfusion. If perforation is suspected, abdominal imaging should be performed to evaluate for free air. If perforation seems likely, laparotomy should be performed as soon as possible to repair the perforation. In the setting of perforation, antimicrobial therapy should be broadened to cover bowel flora. Two multicenter, international prospective cohort studies of 88 consecutive patients (adults and children) undergoing surgery for gastrointestinal typhoid perforation demonstrated high mortality and complication rates, with a 30-day mortality rate of 9.1%, surgical site infection in 67.0%, and organ space infection in 10.2% of patients.13 Steroid therapy is beneficial in some patients with severe enteric fever and coma, delirium, or shock. Dexamethasone is administered at doses of 3 mg/ kg initially, followed by 1 mg/kg every 6 hours for 48 hours. Steroids can mask the signs and symptoms of abdominal perforation and should not be continued for more than 48 hours. Salicylates should be avoided. Relapses of typhoid fever may be treated with the same antimicrobial regimen as the initial attack. ,

Bacteremia

The agents of choice to treat Salmonella bacteremia are the fluoroquinolones, such as ciprofloxacin, and the extended-spectrum cephalosporins, such as ceftriaxone. Typical doses are ciprofloxacin 400 mg every 12 hours intravenously and ceftriaxone 1 to 2 g every 12 to 24 hours intravenously. When the salmonellae are known to be susceptible, ampicillin 1 to 2 g intravenously every 4 to 6 hours or trimethoprim-sulfamethoxazole 8 mg/kg/day (of the trimethoprim component) intravenously can be used. Chloramphenicol is another option. Antimicrobial susceptibility testing is necessary because of the emergence of infections resistant to the fluoroquinolones or extended-spectrum cephalosporins and sometimes both. In cases of sustained bacteremia, the possibility of endovascular infection should be investigated. For transient bacteremia or bacteremia without localization, therapy is continued for 7 to 14 days. With localization to bone, aneurysms, heart valves, and various other sites, antimicrobial therapy should be given for much longer periods (e.g., 6 weeks). Surgical drainage, removal of foreign bodies, or resection of an aneurysm is often necessary to cure localized infection. The possibility of schistosomiasis should be considered and treated, when present, in patients with sustained Salmonella bacteremia (Chapter 334). Patients with HIV infection tend to experience repeated relapses after treatment courses for Salmonella bacteremia. In this group, initial treatment with ciprofloxacin for 2 weeks or longer is recommended. Long-term suppressive therapy has been suggested for those experiencing frequent relapses.14

Carriers

Chronic carriers (i.e., >1 year) of salmonellae other than Salmonella Typhi are uncommon. Stools of convalescent carriers spontaneously become negative during a period of weeks to months, and no therapy should be given. The rare chronic carrier of Salmonella serovars other than Typhi (usually infected with Salmonella Paratyphi A, B, or C) may be treated with a fluoroquinolone, amoxicillin, or trimethoprim-sulfamethoxazole in the doses listed later for 4 to 6 weeks. Patients who experience relapse usually have gallbladder disease (most often calculi) and will not be cured with antimicrobial therapy alone. Cholecystectomy plus antimicrobial therapy may cure these patients, but it is doubtful that the carrier state is a sufficient indication for cholecystectomy. Chronic carriers of Salmonella Typhi can be treated with ciprofloxacin (500 to 750 mg twice daily) for 6 weeks or with amoxicillin at doses of 6 g/day in three or four divided doses plus probenecid 2 g/day in divided doses for 6 weeks. Trimethoprim-sulfamethoxazole (160/800 mg twice daily) plus rifampicin (300 mg twice daily) for 6 weeks may be considered as an alternative regimen. Patients with persistent urinary carriage and S. haematobium infection should

1943

be treated with praziquantel before eradication of Salmonella Typhi is attempted. For patients with persistent carriage and anatomic abnormalities (e.g., gallstones), cholecystectomy combined with antimicrobial therapy is often necessary. For patients with persistent carriage despite adequate antimicrobial therapy and without an identifiable anatomic abnormality, chronic suppressive therapy may be considered. Chronic carriers who do not prepare food and who practice adequate personal hygiene usually do not constitute a public health hazard. Therefore, after the institution of appropriate personal hygienic precautions, and in the absence of evidence of a chronic carrier infecting others, cholecystectomy is probably not indicated to eradicate the carrier state.



PREVENTION

Salmonella infection is best prevented by protecting the water supply, preventing fecal contamination during food production, cooking and refrigerating foods, pasteurizing milk and milk products, and handwashing before preparing foods. Travelers should judiciously avoid consuming untreated water (including ice), raw vegetables, and fruits. Food should be cooked or peeled, and drinks should be boiled, carbonated, or commercially bottled. The widespread presence of salmonellae in the animal kingdom means that reducing the risk for Salmonella infections requires a multifaceted approach. There is no vaccine for Salmonella infection other than that for Salmonella Typhi. Travelers should be vaccinated before going to areas that are endemic for typhoid fever.15 Two types of vaccines are available in the United States. One is the typhoid Vi capsular polysaccharide vaccine, which is administered as a single intramuscular injection, with booster doses given every 2 years if needed. This vaccine provides a degree of herd protection against typhoid fever when it is used at the population level. The other licensed typhoid fever vaccine type is the oral live attenuated Ty21a vaccine. Revaccination is necessary every 5 years, if indicated. Ty21a vaccine should not be used in immunocompromised persons or those receiving antimicrobials. Both types of vaccines confer greater than 75% protective efficacy. More recently, new Vi polysaccharide protein-conjugate vaccines have been developed for widespread use and show greater promise as routine health interventions. A4  A typhoid Vi tetanus toxoid conjugate vaccine has been prequalified by the World Health Organization and is recommended for use as a single dose from age 6 months in typhoid endemic countries, but this vaccine is not yet licensed in the United States.16 Vaccines afford only partial immunity to typhoid fever. Persons who have been vaccinated should still restrict their diets to avoid potentially contaminated food and fluids. When cases of imported typhoid are identified in the United States, the local health department should be informed and will monitor stool cultures. Typhoid fever acquired in the United States is typically investigated by the public health department to identify potential sources and chronic carriers.  

PROGNOSIS

Mortality in patients with Salmonella enterocolitis is rare; infants and the elderly are at greatest risk, with death occurring as a result of dehydration and electrolyte imbalance. Mortality from Salmonella bacteremia is not uncommon and is most likely to occur in the very young, the very old, the malnourished, and the immunocompromised. Before the advent of antimicrobial therapy, typhoid fever had a case-fatality ratio of 15 to 20%. This has been reduced to less than 1% in industrialized countries. However, the case-fatality ratio remains high in some developing countries. The case-fatality ratio of invasive nontyphoidal Salmonella in Africa is approximately 20%. In treated patients, the temperature usually returns to normal after 3 to 5 days of therapy, but this may take longer in patients treated with extended-spectrum cephalosporins than in those treated with fluoroquinolones and in those infected with isolates with decreased fluoroquinolone susceptibility who are treated with ciprofloxacin. In the era before antimicrobial therapy, 5 to 10% of patients who recovered from typhoid fever had relapses. Relapses continued to occur in 10 to 15% of patients treated with chloramphenicol, ampicillin, and trimethoprimsulfamethoxazole, but this seemed to be much less frequent (40) times per day, abdominal cramps, and tenesmus. The most severe infections are due to S. dysenteriae type 1 and progress rapidly from watery to bloody diarrhea and frequently to frank dysentery. The severity of illness and variety of complications of S. flexneri and S. dysenteriae type 1 contributes to their importance as causes of death in children in developing countries. In this population, bacteremia with the infecting strain or other enteric flora can occur, and the intense mucosal inflammation can lead to toxic megacolon. Colonic perforation and pancolitis are rare complications that may require surgical intervention. In young children, in whom mesenteric

CHAPTER 293 Shigellosis  

The determinants of virulence are present in a large virulence plasmid found in all Shigella and also in enteroinvasive E. coli serotypes capable of causing similar clinical illness. The virulence plasmid encodes the proteins for a type III secretion system that facilitates the transfer of microbial effector molecules into the host cell; these proteins manipulate the actin cytoskeleton of the host cells to facilitate bacterial translocation to the host cell cytoplasm. The molecular mechanisms, which have been intensively studied and appear ever more complex, are mediated by engaging and coordinating multiple calcium-dependent and -independent cellular regulatory pathways. Once inside the epithelial cell, the bacteria are sequestered from host immune responses. Shigella also secrete proteins that interrupt immune signaling pathways and block innate immune responses. Whereas the short-term gain for the organism is microbial multiplication and subsequent excretion in feces in the hope of encountering another susceptible individual, the same mechanisms ultimately act to eliminate the organisms at later stages of infection when neutrophils play a major role in disease resolution by killing the bacteria and degrading Shigella virulence proteins. S. dysenteriae type 1 and certain serotypes of Shiga toxin–producing E. coli (notably O157:H7 and others) also produce similar protein toxins that inhibit protein synthesis and directly cause cell death. It is one important reason that infection due to these organisms is often severe. Shiga toxins play an important role in the pathogenesis of hemolytic-uremic syndrome (HUS; Chapter 163), a potentially lethal complication, due to its affinity for vascular endothelium within the kidneys.

1945.e1

1946

CHAPTER 293 Shigellosis  

TABLE 293-1 CLINICAL SYNDROMES AND COMPLICATIONS OF SHIGELLOSIS STAGE

TIME OF APPEARANCE AFTER ONSET OF ILLNESS

SYMPTOMS AND SIGNS

PATHOLOGY AND PATHOGENESIS

Prodrome

Earliest findings

Fever, chills, myalgias, anorexia

None or early colitis with cytokine response

Watery diarrhea

0-3 days

Fever, abdominal cramps, loose stools

Mild colitis with fecal leukocytes and erythrocytes

Bloody diarrhea

1-3 days

Frequent stools containing blood and mucus, abdominal cramps and tenderness, fever, anorexia

Colitis with fecal leukocytes and erythrocytes

Dysentery

1-5 days

Frequent small-volume stools consisting of blood, More extensive colitis with crypt abscesses and mucus, and pus; severe abdominal cramps; tenesmus mucosal ulcerations

Acute complications

3-7 days

Seizures, obtundation, bacteremia, colonic obstruction, Severe colitis, terminal ileitis mucosal perforation, peritonitis

Additional acute complications due primarily to infection with Shigella dysenteriae type 1

3-7 days

Toxic megacolon, leukemoid reaction, hemolyticuremic syndrome

Severe colitis, expression of Shiga toxin

Postinfectious syndromes

1-3 weeks

Reactive arthritis, with or without urethritis and conjunctivitis

Autoimmune inflammatory response, most common in individuals expressing HLA-B27 antigen

support for the rectosigmoid colon is not fully developed, the intense proctitis results in straining to pass stool and may lead to rectal prolapse. A rapid rise in temperature can cause a seizure in this age group, distinguishable from typical febrile seizures by the older age of the child and the rarity of multiple seizures. Other central nervous system manifestations, such as obtundation, suggests the presence of hypoglycemia, due to poor food intake and inadequate gluconeogenesis, or hyponatremia, secondary to inappropriate secretion of antidiuretic hormone. Anorexia may be intense and prolonged, and combined with ongoing catabolism of host muscle protein associated with excessive pro-inflammatory cytokine production and protein-losing enteropathy due to colitis, results in some degree of protein-energy malnutrition that may progress to kwashiorkor and death (Chapter 203). S. flexneri infection is also associated with reactive arthritis and other autoimmune inflammatory manifestations, such as tendinitis, conjunctivitis, uveitis, urethritis, or erythema nodosum, a constellation of findings commonly termed reactive arthritis (formerly Reiter syndrome) (Chapter 249). This occurs primarily in individuals positive for HLA-B27 antigen and may be a consequence of molecular mimicry. S. dysenteriae type 1 infection may result in a leukemoid reaction associated with thrombocytopenia and microangiopathic hemolytic anemia, often followed by HUS (Chapter 163) and acute renal failure.  

DIAGNOSIS

Shigellosis presenting with watery diarrhea is clinically indistinguishable from the many other causes of watery diarrhea, except for the frequency of fever in Shigella infections, unless the stool is examined microscopically to demonstrate the presence of red and white blood cells or infection is confirmed microbiologically or by molecular diagnostics (CIDT). Routine laboratory studies document leukocytosis with many immature “band” forms, especially in patients with bloody diarrhea or dysentery. In the developed world, these presentations are most commonly due to a Shiga toxin–producing E. coli (STEC), such as serotype O157:H7, or S. flexneri and less often Campylobacter jejuni, nontyphoidal Salmonella species, or rarely Yersinia enterocolitica or Entamoeba histolytica. In patients who have recently received antibiotics, bloody diarrhea is often the consequence of Clostridium difficile (Chapter 280). In contrast, a large multicountry study of moderate to severe pediatric diarrhea in low- and middle-income countries documented S. flexneri as the cause of most bloody diarrhea or dysentery, with no cases of S. dysenteriae detected among more than 10,000 individuals studied.16 Syndromic diagnosis may fail to identify subjects with Shigella infection. In a study in Kenya, only 7 of 63 children with Shigella isolated from stool had clinically identified dysentery (sensitivity 11%, CI 4.6 to 21.6%).17 Culture diagnosis not only takes days before results are available, but also often underestimates prevalence. Falsenegatives can be due to poor sample collection, delay in processing, or failure to use the best culture methods, as well as prior antibiotic administration which may limit detection of still viable but nonculturable organisms. In contrast, CIDT results can be available in hours, and for these reasons antigen capture or polymerase chain reaction diagnostic tests are becoming more attractive but may be more costly. In resource poor settings, the use of more than one selective culture medium and obtaining multiple specimens will increase the yield of positives. Initial screening is based on the failure of Shigella species to ferment lactose, with additional studies of lactose-negative colonies to detect anaerobic fermentation of glucose, and lack of motility or hydrogen

sulfide production to distinguish Shigella from Salmonella species. Specific diagnosis employs standard serologic methods for Shigella O antigens. This is why culture takes several days to confirm results. In the case of mild watery diarrhea due to S. sonnei, the patient is usually recovered by the time the laboratory results are available, so they are of little use in guiding treatment. Culture yields are maximized when only patients with bloody diarrhea or dysentery are studied. Results are valuable for selection of empirical treatment, and for epidemiologic purposes when antibiotic sensitivity is also determined to track the emergence of antibiotic resistance and help to select effective antibiotics when treatment is indicated.

TREATMENT  In healthy adults and children, most Shigella infections cause mild diarrhea, not requiring diagnostic work-up or antibiotic treatment. Clinical trial evidence supports the use of antibiotics to treat bloody diarrhea or dysentery due to Shigella to decrease the severity of symptoms and the duration of fever and diarrhea. A1  Over the years, successive antibiotics have been used to treat Shigella infections because of emerging drug resistance, including tetracycline, chloramphenicol, ampicillin, trimethoprim-sulfamethoxazole, nalidixic acid, pivmecillinam, ciprofloxacin, azithromycin, and ceftriaxone. For antibiotic susceptible strains, any of these antibiotics are effective for clinical and microbiologic cure and high-quality evidence regarding the superiority of one over the other is lacking. Single-dose or short-course treatment has shown promise but data remain limited and more studies are required. Some studies suggest that cefixime (an oral third-generation cephalosporin) is efficacious in children but is less effective in adults, and amoxicillin is less effective than ampicillin. Nonabsorbable antibiotics, such as oral furazolidone, rifaximin, and gentamicin possess in vitro activity against Shigella but are not recommended because optimal treatment requires therapeutic levels of the drugs in both the lumenal and the mucosal compartments, and these agents have shown variable efficacy in clinical use. Figure 293-2 illustrates the strategy for antibiotic therapy and recommended drugs for shigellosis in children and adults in high-income and low-income settings. Therapeutic options for treating shigellosis are increasingly limited because of increasing and wide-spread resistance in both high-income and low-income countries.18 In April 2017, the CDC issued guidance to limit antibiotic treatment of Shigella in the United States to patients who are immunocompromised or have severe illness, defined as requiring hospitalization, or suffering from invasive disease or complications.19 Because identification of the etiologic agent for diarrhea can take 3 to 5 days, empiric therapy pending culture results is necessary in these severe cases. Local susceptibility data should guide antibiotic selection, and this is the rationale to systematically collect drug sensitivity information. The CDC advisory describes the emergence and spread of Shigella strains in the United with elevated minimum inhibitory concentration (MIC) values for ciprofloxacin of 0.12 to 1 µg/ml. Clinicians should be aware that laboratory standards may categorize these strains as susceptible to ciprofloxacin, yet they can harbor quinolone resistance genes known to confer reduced susceptibility and result in potential treatment failure or prolonged shedding, thereby increasing the risk of secondary cases. For these reasons, the CDC recommends that Shigella strains with an MIC of 0.12 µg/ml or higher should not be treated with fluoroquinolones. MSM populations are at increased risk of harboring antibiotic-resistant strains. Resistance is more prominent in developing countries where both burden of infections and antibiotic use are high. As a consequence, oral therapy for shigellosis is becoming increasingly challenging, with severely limited therapeutic options. In South Asia, the majority of strains are now resistant to ciprofloxacin and resistance to azithromycin is rapidly emerging. It is possible that in

CHAPTER 293 Shigellosis  

Children*

1947

Adults**

First line:

First line:

• Oral therapy: - Ciprofloxacin 15 mg/kg twice daily for 3 days • Parenteral therapy for severe illness: - Ceftriaxone 50-100 mg once daily for 3 days IV or IM PLUS • Zinc 20 mg (10 mg for infants ≤6 months) for 10-14 days, by mouth (developing country only)

• Oral therapy: - Ciprofloxacin 500 mg once a day for 3 days • Parenteral therapy for severe illness: - Ceftriaxone 1 g every 24 hours for 3 days IV or IM

If local strains are ciprofloxacin-resistant*** Second line oral therapy:

Second line oral therapy:

• Azithromycin 12 mg/kg once on first day, then 6 mg/kg once daily on days 2 through 4 (total course: 4 days); OR • Cefixime 8 mg/kg once daily for 3 days; OR • Trimethoprim-sulfamethoxazole 4 mg/kg/day of TMP and 20 mg/kg/day SMZ twice a day for 5 days (if susceptibility known or likely based on local data)

• Azithromycin 1 g (or 500 mg every 12 hours) for 1 day; OR • Azithromycin 500 mg on day 1, followed by 250 mg once daily for 4 days; OR • Cefixime 400 mg once daily for 3 days; OR • Trimethoprim-sulfamethoxazole 160 mg of TMP and 800 mg SMZ twice a day for 5 days, (if susceptibility known or likely based on local data)

Improving after 48 hours of antibiotics?

Yes Finish therapy

No Is susceptibility known? Yes Adjust therapy according to susceptibility results

No Choose different 1st or 2nd line therapy not previously used

*The WHO expert committee on the use of essential medicines 201719 currently recommends no antibiotic treatment for non-bloody, non-febrile diarrheal episodes, including those due to Shigella. **CDC recommends limiting antibiotic treatment of Shigella to patients who are immunocompromised or have severe illness, defined as requiring hospitalization, or suffering from invasive disease or complications.18 ***CDC recommends that Shigella strains with an MIC of 0.12 µg/ml or higher should not be treated with fluoroquinolones.18 FIGURE 293-2.  Treatment of shigellosis.

the near future there could be no oral antibiotics available to which Shigella are susceptible, making common community-acquired childhood infections very difficult to manage. The CDC and WHO have recently declared antibiotic resistant Shigella a serious public health threat requiring new interventions. The WHO expert committee on the use of essential medicines 201720 currently recommends no antibiotic treatment for nonbloody, nonfebrile diarrheal episodes, including those due to Shigella. Ciprofloxacin is the first-choice antibiotic for bloody diarrhea or dysentery, with azithromycin, cefixime, and ceftriaxone among second-choice agents for ciprofloxacin-resistant strains. However, a recent large multicountry study has revealed a high residual mortality burden in children in developing countries in the 60- to 90-day period following clinically moderateto-severe acute watery diarrhea, with a substantial proportion of these infections attributed to Shigella. Whether empiric antibiotic treatment of suspected nondysenteric shigellosis in children presenting with severe dehydration or malnutrition in developing countries confers survival benefit is unknown, and a current multicountry clinical trial is assessing whether children less than 2 years old with severe watery diarrhea or malnutrition in countries with a high child mortality burden might benefit from empiric antibiotic therapy with azithromycin. In severe shigellosis, the prompt initiation of effective antibiotic treatment shortens the duration of illness and achieves a more rapid improvement in symptoms such as fever, cramps, and tenesmus. Evidence from Bangladesh demonstrates that prompt treatment of infections due to S. dysenteriae type 1 reduces the frequency of HUS and presumably other complications due to

colonic inflammation, such as megacolon, perforation, and rectal prolapse. Parenteral ceftriaxone has been most useful in severe or drug-resistant illness, children with immune deficiency including HIV, or inability to take oral medications due to vomiting or diminished consciousness. However, this requires hospitalization or outpatient administration and monitoring, and both decrease access to treatment and increase its cost. In addition, the increasing prevalence of extended spectrum β-lactamase (ESBL) isolates of S. flexneri and S. sonnei expressing the CTX-M type of β-lactamase that confers resistance to all β-lactam antibiotics except cephamycins and carbapenems is already compromising consideration of ceftriaxone as an effective treatment option. In more severe infections, complications such as hypoglycemia or hyponatremia can be managed with appropriate glucose or saline given intravenously, but patients must be monitored by trained clinical staff. Colitis with toxic megacolon and intestinal perforation represent difficult problems, primarily associated with S. dysenteriae type 1 disease. To correct rectal prolapse in infants the mucosa should be kept moist with saline until manual reduction can be achieved. However, because of the recent disappearance of S. dysenteriae type 1 disease such complications are also diminishing in frequency. The prolonged anorexia and catabolic responses in shigellosis, require close attention to nutritional rehabilitation, especially in malnourished children in developing countries where the diet is often poor. Even with adequate protein-containing and energy-dense diets after infection, replacement of nutrient stores may take as long as four times the duration of the clinical illness and full nutritional recovery may not be

possible before the next infection occurs, causing further nutritional deterioration. Administration of zinc is associated with a modest clinical benefit, including reduced duration of symptoms, as well as a reduction in subsequent diarrheal incidence. WHO guidelines recommend zinc supplementation in all children with diarrhea in resource-limited settings, where zinc deficiency is common. Seizures can occur but are usually self-limited, and other neurologic complications respond to fluid and electrolyte management and correction of hypoglycemia and hyponatremia. Reactive arthritis is a greater problem because it is an autoimmune response that occurs primarily in genetically susceptible individuals positive for HLA-B27 antigen. Chronic destructive arthritis may require nonsteroidal anti-inflammatory agents, steroids, or inflammatory cytokine inhibitors, in addition to effective antibiotic treatment to eliminate the causative organism, as well as ongoing medical management. Some evidence suggests that antimotility agents are contraindicated in shigellosis, particularly in cases of bloody diarrhea or dysentery in young children, because these drugs may slow peristalsis and prolong contact between the organisms and the mucosa, thereby increasing microbial invasion, pathologic changes, and severity, including toxic megacolon.

treatment with antibiotics. When treatment is not effective, either a drugresistant strain or another etiologic agent should be suspected. In some instances, severe shigellosis with pancolitis has been misdiagnosed as inflammatory bowel disease. This can be a disaster if the patient is treated with steroids. Whereas early and effective treatment of S. dysenteriae type 1 reduces the risk of complications such as HUS and probably megacolon and bowel perforation, about 25% of those with HUS will have some permanent renal impairment, and a small percentage may progress to end-stage renal failure. Because of the autoimmune nature of reactive arthritis, usually associated with S. flexneri, early treatment might not prevent its occurrence but could mitigate its severity. Infection with one serotype of Shigella generally provides durable immunity to reinfection with the same strain but leaves the individual susceptible to other antigenically distinct strains and serotypes.

  Grade A Reference A1. Christopher PR, David KV, John SM, et al. Antibiotic therapy for Shigella dysentery. Cochrane Database Syst Rev. 2010;8:CD006784.



PREVENTION

Personal hygiene (in particular hand washing after handling the diapers of infected children and before food preparation), the sanitary disposal of feces, and protection of food and water sources from microbial contamination are essential to limit the spread of shigellosis. Preventing spread in daycare settings is a particular problem because it is so difficult to stop young children from constantly exploring their world, picking up bacteria on their hands, and bringing their fingers or contaminated fomites to their mouths. In such settings, it is particularly important for the adult caretakers to observe good personal hygiene and to supervise children in handwashing. Soap and water are sufficient, but hand sanitizers do work and may be more convenient. In environments where soap is unavailable, water used with sand or ash for scrubbing is helpful. Keeping infected children away from daycare until their stool is negative, if indeed cultures were done, or separating recently ill children from the susceptible has been recommended but is costly and not easy to accomplish. Household hygiene in the setting of an index case, including frequent handwashing, caution in the disposal of soiled diapers or underwear, regularly wiping down the area where these are collected with a disinfectant such as Lysol, and precautions in food preparation, can help limit intrahousehold spread. All of these measures are more difficult to implement in resourcelimited environments without access to water for hygiene, let alone safe water for drinking or cooking, toilets or properly maintained latrines, ability to purchase disinfectants, or refrigeration to store food. Vaccines, particularly for the more virulent species, would be useful. However, despite much effort and progress, a safe and effective vaccine has not been developed or approved for general use.21 Vaccines against Shigella have been challenging to design because of the multiple species and serotypes, insufficient understanding of immunity and cross-protection, lack of a faithful animal model, and low commercial interest. In the United States populations for whom a Shigella vaccine could be recommended include travelers or military personnel who are deployed to high-risk locations. In low-income countries, a multivalent broadly protective vaccine against S. flexneri and S. sonnei, the two most common Shigella species in the last 20 years, would be extremely useful because shigellosis continues to account for a significant portion of the annual mortality due to diarrheal diseases. Our current understanding of immunity in shigellosis indicates that serotype-specific immunity is most important. Attenuated live oral vaccine candidates are probably most useful for travelers and military personnel deployed to high-risk areas; unfortunately, vaccines most effective in inducing immunity have also been the most reactogenic, causing fever and often mild diarrhea in recipients. For children in endemic areas current vaccines under clinical development are killed whole cells,22 lipopolysaccharide (LPS) O antigen–protein conjugates, and subunit candidates. Newer-generation conjugate vaccines against LPS seem the most promising. Unless common protective antigens are identified, the most likely vaccine strategy will be based on a combination of serotype-specific antigens from the most prevalent and cross-protective Shigella isolates. Refinement of Shigella human challenge models offer promise for accelerating Shigella vaccine development.  

PROGNOSIS

S. sonnei infection and most cases of S. boydii infection are mild and self-limited, with no sequelae. Infection with S. flexneri or S. dysenteriae responds to proper

  GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 293 Shigellosis  

GENERAL REFERENCES 1. Kotloff KL, Riddle MS, Platts-Mills JA, et al. Shigellosis. Lancet. 2018;391:801-812. 2. Anderson M, Sansonetti PJ, Marteyn BS. Shigella diversity and changing landscape: insights for the twenty-first century. Front Cell Infect Microbiol. 2016;6:1-9. 3. Thompson CN, Duy PT, Baker S. The rising dominance of Shigella sonnei: an intercontinental shift in the etiology of bacillary dysentery. PLoS Negl Trop Dis. 2015;9:1-13. 4. Muthuirulandi Sethuvel DP, Devanga Ragupathi NK, Anandan S, et al. Update on: Shigella new serogroups/serotypes and their antimicrobial resistance. Lett Appl Microbiol. 2017;64:8-18. 5. Wikswo ME, Kambhampati A, Shioda K, et al. Outbreaks of acute gastroenteritis transmitted by person-to-person contact, environmental contamination, and unknown modes of transmission–United States, 2009-2013. MMWR Surveill Summ. 2015;64:1-16. 6. Baker KS, Dallman TJ, Ashton PM, et al. Intercontinental dissemination of azithromycinresistant shigellosis through sexual transmission: a cross-sectional study. Lancet Infect Dis. 2015;15: 913-921. 7. Hines JZ, Pinsent T, Rees K, et al. Notes from the field: shigellosis outbreak among men who have sex with men and homeless persons—Oregon, 2015–2016. MMWR Morb Mortal Wkly Rep. 2016;65:812-813. 8. Marder EP, Cieslak PR, Cronquist AB, et al. Incidence and trends of infections with pathogens transmitted commonly through food and the effect of increasing use of culture-independent diagnostic tests on surveillance—Foodborne Diseases Active Surveillance Network, 10 U.S. Sites, 2013-2016. MMWR Morb Mortal Wkly Rep. 2017;66:397-403. 9. George CM, Ahmed S, Talukder KA, et al. Shigella infections in household contacts of pediatric shigellosis patients in rural Bangladesh. Emerg Infect Dis. 2015;21:2006-2013. 10. Pires SM, Fischer-Walker CL, Lanata CF, et al. Aetiology-specific estimates of the global and regional incidence and mortality of diarrhoeal diseases commonly transmitted through food. PLoS ONE. 2015;10:1-17. 11. GBD Diarrhoeal Diseases Collaborators. Estimates of global, regional, and national morbidity, mortality, and aetiologies of diarrhoeal diseases: a systematic analysis for the Global Burden of Disease Study 2015. Lancet Infect Dis. 2017;17:909-948.

1948.e1

12. Lederer I, Taus K, Allerberger F, et al. Shigellosis in refugees, Austria, July to November 2015. Euro Surveill. 2015;20:1-4. 13. McCrickard LS, Crim SM, Bowen A. Disparities in severe shigellosis among adults—Foodborne Diseases Active Surveillance Network, 2002-2014. BMC Public Health. 2018;18:1-8. 14. Mattock E, Blocker AJ. How do the virulence factors of Shigella work together to cause disease? Front Cell Infect Microbiol. 2017;7:1-24. 15. Bonnet M, Tran Van Nhieu G. How Shigella utilizes Ca(2+) jagged edge signals during invasion of epithelial cells. Front Cell Infect Microbiol. 2016;6:1-8. 16. Liu J, Platts-Mills JA, Juma J, et al. Use of quantitative molecular diagnostic methods to identify causes of diarrhoea in children: a reanalysis of the GEMS case-control study. Lancet. 2016;388:1291-1301. 17. Pavlinac PB, Denno DM, John-Stewart GC, et al. Failure of syndrome-based diarrhea management guidelines to detect Shigella infections in Kenyan children. J Pediatric Infect Dis Soc. 2016;5: 366-374. 18. Puzari M, Sharma M, Chetia P. Emergence of antibiotic resistant Shigella species: a matter of concern. J Infect Public Health. 2018;11:451-454. 19. CDC recommendations for diagnosing and managing Shigella strains with possible reduced susceptibility to ciprofloxacin. CDC Health Alert Network; 2017. https://emergency.cdc.gov/han/ han00401.asp. Accessed April 30, 2019. 20. The selection and use of essential medicines: report of the WHO Expert Committee, 2017 (including the 20th WHO Model List of Essential Medicines and the 6th WHO Model List of Essential Medicines for Children). Geneva: World Health Organization; 2017 (WHO technical report series; no. 1006). https://www.who.int/medicines/publications/essentialmedicines/EML_2017_EC21 _Unedited_Full_Report.pdf. Accessed April 30, 2019. 21. Walker RI. An assessment of enterotoxigenic Escherichia coli and Shigella vaccine candidates for infants and children. Vaccine. 2015;33:954-965. 22. Raqib R, Sarker P, Zaman K, et al. A phase I trial of WRSS1, a Shigella sonnei live oral vaccine in Bangladeshi adults and children. Hum Vaccin Immunother. 2019;1-12.

1948.e2

CHAPTER 293 Shigellosis  

REVIEW QUESTIONS 1. Spread of shigellosis can be difficult to control because the major route of transmission of infection is: A . Food B. Water C. Person to person D. Aerosol E. Dirty toilet seats Answer: C Because Shigella resist the antibacterial effects of gastric acid and invade intestinal epithelial cells, which protect them from immune mechanisms, the infectious inoculum can be very small. This eliminates the need for organisms to multiply before ingestion and facilitates direct (skin-to-skin) or indirect (by contaminated fomites) person-to-person spread. Although Shigella can be transferred to food or water and subsequently cause outbreaks, transmission by these routes can be diminished by commonsense care around food preparation, refrigeration, and chlorination of the water supply. Shigella are not respiratory pathogens and aerosol transmission does not occur. Transmission by contact with contaminated toilet seats is possible; however, acquisition of infection by this route has not been described. 2. The most common clinical presentation of shigellosis in the United States is: A Watery diarrhea associated with fever B. Dysentery C. Hemolytic-uremic syndrome D. Leukemoid reaction E. Reactive arthritis Answer: A  Because 85% of Shigella infection in the United States is due to S. sonnei, which only rarely proceeds beyond the watery diarrhea stage to bloody diarrhea or dysentery, the vast majority of patients develop acute selflimited water diarrhea. However, even S. sonnei induces an inflammatory enteritis with the production of pyrogenic cytokines such as interleukins 1, 6, and 8, and therefore fever is commonly present, and peak fever correlates with the severity of other clinical manifestations of shigellosis. The other major species found in the United States is Shigella flexneri, often involving men who have sex with men, which more commonly causes bloody diarrhea or even dysentery. 3. The pathobiology of Shigella infection is a consequence of: A . Activation of intestinal secretion pathways B. Lactose malabsorption C. Hyperperistalsis D. Mucosal edema E. Mucosal invasion and inflammation Answer: E  A hallmark of pathogenic Shigella is the ability to invade cells, including phagocytic leukocytes and nonphagocytic intestinal epithelial cells. The release of pro-inflammatory cytokines during this process results in fever, recruitment of leukocytes to the intestinal mucosa, epithelial cell death and mucosal ulceration, and the characteristic finding of white and red blood cells in the stool. Experimentally reducing the inflammatory response attenuates the severity of clinical manifestations. Clinically significant dehydration due to fluid losses in stool is not characteristic of shigellosis. Lactose malabsorption is a consequence of damage to intestinal epithelial brush border membranes, where the enzyme lactase is localized, and may develop during and persist after clinical shigellosis, but it is not the cause of symptoms during infection. Inflammation of the colonic mucosa is common in shigellosis, manifested as cramps or, in severe cases, with bloody diarrhea or dysentery and tenesmus, but again, this is a consequence of the severity of the underlying inflammatory reaction. Rather than mucosal edema resulting from the myriad inflammatory pathways being activated, the histology of the affected intestine reveals leukocytic infiltration, blood, organisms in the submucosa and within epithelial cells, and mucosal ulcerations where dead epithelial cells have been sloughed off the mucosal surface.

4. The primary treatment of bloody diarrhea due to Shigella species is: A . Oral rehydration fluids B. Intravenous rehydration fluids C. H2 blockers D. Antibiotics such as ciprofloxacin (Cipro) E. Loperamide (Lomotil) Answer: D  Mild to moderate shigellosis presenting as watery diarrhea, and typically due to S. sonnei, is a self-limited illness, usually lasting only a couple of days and does not require antibiotic or other treatment. More severe clinical shigellosis, associated with bloody diarrhea, is uncomfortable because of the high frequency of bowel movements and associated cramps. Treatment of these patients with an antibiotic to which the infecting organism is susceptible is well documented to resolve clinical symptoms more rapidly and, if administered early, to prevent late complications due to S. flexneri and S. dysenteriae type 1. Because of increasing antibiotic resistance, the problem is to identify a proven effective antibiotic to which the causative organism is susceptible. At present, a quinolone such as ciprofloxacin (Cipro) or a macrolide such as azithromycin (Zithromax) is the most reliable choice, as the prevalence of resistance is still relatively low in most parts of the world. Rehydration, by either oral rehydration solutions or intravenously administered fluids, is usually not necessary, however not contraindicated, and rehydration will not reverse the clinical manifestations due to inflammation. H2 blockers are known to increase the susceptibility to acid-sensitive enteric pathogens such as V. cholerae but not acid-resistant organisms like Shigella and have no effect on already established infections. Lomotil or other gut motility–inhibiting drugs such as Imodium that reduce peristalsis are, in fact, potentially dangerous in that they can prolong the contact between invasive Shigella and the mucosa and worsen the clinical severity of the infection. Where S. flexneri or S. dysenteriae type 1 is common, these drugs are contraindicated, especially in children. 5. The best available way to prevent shigellosis is: A . Immunization with Shigella vaccine B. High standards of personal and household hygiene C. Provision of prophylactic antibiotics to contacts of patients D. Quarantine E. Anti-inflammatory drug Answer: B Shigellosis is primarily contracted directly through fecal-oral transmission by contact between an infected individual or indirectly by contaminated fomites to a susceptible individual. To break the cycle, improved hygiene such as handwashing when taking care of a patient with shigellosis (e.g., after handling soiled diapers or other clothes) or before preparing and serving food in the household is most useful. Additional measures, such as disinfecting the areas where soiled clothes are gathered with a simple disinfectant such as Lysol or diluted bleach, are helpful but not necessary to reduce the transmission of infection to others in the household. In developing countries with a higher incidence of more virulenct Shigella species and limited supplies of safe drinking water or facilities for the sanitary disposal of feces, the use of hand sanitizers would be useful but relatively expensive. There is no available licensed vaccine for shigellosis, although considerable effort has been expended to develop one that is both effective and safe. Prophylactic administration of antibiotics is a bad idea because antibiotic exposure only increases the selective pressure for drug resistance to emerge. Quarantine is no longer a primary strategy for control of most infectious diseases because diagnosis and more specific measures can usually be implemented. Quarantine usually fails to prevent the spread of contagious diseases in the community because it is difficult to isolate all of the potential spreaders of infection. Whereas certain anti-inflammatory drugs (e.g., cytokine antagonists) have been shown to be effective in experimental models of shigellosis, none are currently available for clinical use.

1948

CHAPTER 294 Brucellosis  

294  BRUCELLOSIS EDSEL MAURICE T. SALVANA AND ROBERT A. SALATA



DEFINITION

Brucellosis, previously known as Malta fever or undulant fever, is a zoonotic disease caused by bacteria of the genus Brucella. Human infection, which has protean manifestations, is acquired via direct contact, ingestion, or inhalation. Most disease is acquired by eating unpasteurized dairy products or undercooked meat. Occupational infection is typically inhalational or through contamination of exposed wounds and mucous membranes. Human-to-human transmission occurs but is rare. Brucellosis remains a significant public health and economic burden in many countries despite advances in detection, treatment, and prevention. In regions where the disease is endemic, brucellosis continues to have far-reaching deleterious effects on humans and animals alike.1

The Pathogen

Brucella is a slow-growing, small, aerobic, nonmotile, nonencapsulated, non– spore-forming, gram-negative coccobacillus. Brucella abortus (from cattle, bison, elk), B. suis (from pigs and feral swine), B. melitensis (from sheep, goats, and camels), and B. canis (from dogs) are the species that most commonly infect humans. Genetic analysis shows a high degree of homology among different species despite disparate preferred hosts, and virulence factors can vary between and within species. Whole genome sequencing can distinguish between genotypes at a higher resolution, and can provide clues to the organism’s geographic origin.2  

EPIDEMIOLOGY

Etiology

Brucella abortus is usually associated with mild to moderate sporadic disease. B. suis and B. melitensis infections are associated with suppurative or disabling complications and can have a prolonged course. Infection with B. canis has an insidious onset, relapses frequently, and has a chronic but relatively mild course. Two marine species, B. pinnipedialis (from seals) and B. ceti (from porpoises and dolphins) can cause neurobrucellosis in humans. Brucella microti (from common vole, red fox) has a high potential for pathogenicity, but no instances of human infection have been reported. Other Brucella species with known or potential human pathogenicity include B. inopinata (breast implant infection), B. ovis (sheep infection; no human cases), and B. neotomae (rodent infection; no human cases reported). BO2, a proposed species closely related to B. inopinata, has caused one human case of chronic destructive pneumonia.

CHAPTER 294 Brucellosis  

ABSTRACT

Brucellosis, previously known as Malta fever or undulant fever, is a zoonotic disease caused by intracellular bacteria of the genus Brucella. Human infection, which has protean manifestations, is acquired via direct contact, ingestion, or inhalation. Most disease is acquired by eating unpasteurized dairy products or undercooked meat. Occupational infection is typically inhalational or through contamination of exposed wounds and mucous membranes. Human-to-human transmission occurs but is rare. Brucellosis remains a significant public health and economic burden in many countries, and is proving a resilient disease despite advances in detection, treatment, and prevention. Brucella is a potential agent of bioterrorism due to its pathogenecity.

KEYWORDS

Brucella zoonosis foodborne illness occupational exposure sacroiliitis bioterrorism abortion

1948.e3

CHAPTER 294 Brucellosis  

Because of its pathogenicity and ability to remain viable in storage for long periods, Brucella spp. are potential agents for bioterrorism.3

Incidence and Prevalence

More than 500,000 cases of brucellosis are reported yearly to the World Health Organization (WHO) from 100 countries. B. melitensis infection accounts for most diagnosed cases, primarily from the Mediterranean region, Latin America, the Arabian Gulf, China, and the Indian subcontinent. B. abortus infection occurs worldwide but has been effectively eradicated in most northern and western European countries, Japan, and Israel. B. suis is found in the United States, South America, and Southeast Asia. B. canis infection occurs in North America, South America, Japan, and central Europe. High rates of Brucella seropositivity in African communities that keep livestock and consume unpasteurized milk have been reported, and undiagnosed African brucellosis likely represents a significant burden of infection. Returning travelers and immigrants from endemic countries continue to be diagnosed in otherwise low prevalence or nonendemic countries. Brucellosis in animals is a chronic infection that can persist throughout life. Effective control programs in animals have decreased human brucellosis dramatically in the United States from more than 6000 cases in 1947 to fewer than 140 cases per year since 1993. Texas, California, Arizona, and Florida reported the most cases in 2010 (https://www.cdc.gov/brucellosis/resources/ surveillance.html). Brucella infection in the United States occurs mostly through direct contact with animals or animal secretions in high-risk groups, including slaughterhouse workers, farmers, dairy workers, and veterinarians. Laboratory workers handling infected animals or Brucella cultures are also at risk. More than one half of reported cases are associated with meat processing, particularly from “kill areas.” Many cases of B. abortus infection in veterinarians are from accidental exposure to live vaccines (which are pathogenic to humans) used to immunize livestock. Human infections in hunters of feral swine occur sporadically. Imported goat milk and cheese from Mexico (unpasteurized)4 is an important source of B. melitensis infection. Brucellosis contracted abroad may not become symptomatic until the traveler returns. Human-to-human transmission has been documented in 45 cases, and has occurred through sexual transmission, breastfeeding, blood transfusion, bone marrow transplantation, nosocomial exposure, and perinatally.5 Brucellosis in pregnancy has been associated with spontaneous abortions, congenital abnormalities, and neonatal infections. Childhood brucellosis, which occurs mostly in school-aged children, accounts for 3 to 10% of all reported cases worldwide. Human brucellosis incidence is epidemiologically linked to serious pregnancy complications.6  

PATHOBIOLOGY

Pathogenesis

After penetrating the epithelial cells of human skin, conjunctiva, pharynx, intestine, or lung, Brucella organisms in naïve individuals induce a delayed inflammatory response (up to 48 hours) with polymorphonuclear leukocyte infiltration at the infection site. Brucella organisms are then ingested by dendritic cells, neutrophils, and tissue macrophages, and these subsequently spread to regional lymph nodes. If host defenses within the lymph nodes are overwhelmed or the burden of infection is high, bacteremia follows. The usual incubation period from infection to bacteremia is 2 to 4 weeks. Bacteremia is accompanied by phagocytosis of free Brucella organisms by macrophages, with localization primarily to the spleen, liver, and bone marrow. These form small, noncaseating granulomas which can serve as persistent sources of infection. As an intracellular organism, Brucella has to avoid detection by the immune system and survive a hostile intracellular environment. Brucella organisms avoid initial detection by the host through multiple mechanisms. Its cell wall lipopolysaccharide (smooth LPS) differs significantly from regular bacterial LPS in two important ways: it has very little effect on Toll-like receptor type 4 (TLR4) activation, and it is resistant to complement activation. In addition, Brucella organisms deploy a protein that interferes with TLR signaling. Upon successful entry into the host, Brucella organisms in phagosomes are able to survive acidification and lysosome fusion through the induction of specific virulence factors such as the VirB type IV secretion system (T4SS) and urease (ure). To replicate, Brucella organisms intercept traffic between the endoplasmic reticulum and the Golgi apparatus. They also seem to inhibit apoptosis of infected cells, such as synoviocytes, thereby maintaining a persistent presence. A major role for the family of outer membrane proteins Omp25/Omp31 in pathogenicity has been elucidated. Omp22, omp25, and omp31 knockout mutants show substantially decreased virulence. Omp25 seems to inhibit TNF-α production by macrophages, and facilitates secretion of periplasmic proteins in acidic medium by modulating the permeability of the Brucella

1949

membrane. Omp22 and omp25 are necessary for invasion and intracellular survival in host cells. Other known virulence factors that occur at high frequencies in pathogenic Brucella include integral membrane-bound protein (MviN), mannose-6-phosphateisomerase (ManA), mannosyl-transferase (WbkA), perosamine-synthetase (PerA), and outer membrane protein 19 (omp19).7 Inflammation from the innate host immune response is likely the main driver of pathology, because no secreted proteolytic enzymes or bacterial toxins have been found. Brucella induces the production of host proinflammatory cytokines and metalloproteinases, which cause tissue damage.8

Immunity

Humoral factors play an important role in host defense against Brucella spp. Even in the absence of specific agglutinating antibody, normal human serum is bactericidal for Brucella organisms. The intracellular location of Brucella spp. within macrophages enables it to escape the lethal effects of serum to a certain extent. Specific serum agglutinating antibody has opsonic activity but does not correlate with the development of protective immunity. A role for mononuclear phagocytes and cell-mediated immunity in brucellosis has been demonstrated. Prior infection with Listeria monocytogenes or Mycobacterium tuberculosis, both of which stimulate cell-mediated immune mechanisms, is protective against Brucella infection in animals. Skin testing with Brucella proteins elicits a typical delayed hypersensitivity response in infected individuals. Macrophages, activated with T helper 1 (TH1)–type cytokines (including interferon-γ, tumor necrosis factor-α, interleukin-1, interleukin-12), kill Brucella. Despite high levels of TH1 cytokine production, deficient effector phagocytic activity persists. Later in the course of infection, there is evidence of an unexpected inhibitory effect from neutrophils. Animal models have demonstrated more efficient killing of Brucella organisms in the absence of polymorphonuclear cells, which somehow dampen the immune response to this pathogen.  

CLINICAL MANIFESTATIONS

Clinically, human brucellosis can be divided into subclinical illness, acute or subacute disease, localized disease, relapsing infection, and chronic disease (Table 294-1).

Subclinical Illness

Asymptomatic or clinically unrecognized human brucellosis often occurs in high-risk groups, including slaughterhouse workers, farmers, and veterinarians. The diagnosis is usually made through serologic means. More than 50% of abattoir workers and up to 33% of veterinarians have high anti-Brucella antibody titers but no history of recognized clinical infection. Children in endemic areas frequently have subclinical illness.

Acute and Subacute Disease

After an incubation period of several weeks or months, acute brucellosis may occur as a mild, transient illness (B. abortus or B. canis) or as an explosive, toxic illness with the potential for multiple complications (B. melitensis or B. suis). Approximately 50% of patients have an abrupt onset over days, but the remainder have an insidious onset over weeks. Symptoms in brucellosis are protean and nonspecific. More than 90% of patients experience malaise, chills, sweats, fatigue, and weakness. More than 50% of patients have myalgias, anorexia, and weight loss. Some patients complain of arthralgias, cough, testicular pain, dysuria, ocular pain, or blurring of vision. Few localizing physical signs are apparent. Fever, with temperatures often greater than 39.4° C (103° F), occurs in 95% of patients. An undulating or intermittent fever pattern is not unusual. A pulse–temperature deficit (i.e., relative bradycardia) may occur. Splenomegaly is present in 10 to 15% of cases, and lymphadenopathy occurs in about 14% of patients. Axillary, cervical, and supraclavicular lymphadenopathy are most frequent and may be related to hand wounds or oropharyngeal routes of infection. Hepatomegaly is less frequent. Other laboratory findings in acute or subacute disease may include mild anemia; lymphopenia; or neutropenia (especially with bacteremia); lymphocytosis; thrombocytopenia; or, in rare cases, pancytopenia. The majority of infected individuals recover completely without sequelae if diagnosed early with prompt initiation of therapy.

Localized Disease and Complications

Brucella organisms can localize to almost any organ but usually target the bones, joints, central nervous system, heart, lung, spleen, testis, liver, gallbladder, kidney, prostate, pancreas, thyroid, and skin. Disease may occur at multiple sites. About 47% of Brucella patients manifest osteoarticular complications. The most common manifestation is peripheral arthritis, particularly in the

1950

CHAPTER 294 Brucellosis  

TABLE 294-1 CLINICAL CLASSIFICATION OF HUMAN BRUCELLOSIS CLASSIFICATION

DURATION OF SYMPTOMS BEFORE DIAGNOSIS

Subclinical

MAJOR SYMPTOMS AND SIGNS

DIAGNOSIS

COMMENTS

Asymptomatic

Positive (low-titer) serology, negative cultures

Occurs in abattoir workers, farmers, and veterinarians

Acute and subacute

Up to 2-3 mo and 3 mo-1 yr, respectively

Malaise, chills, sweats, fatigue, headache, anorexia, arthralgias, fever, splenomegaly, lymphadenopathy, hepatomegaly

Positive serology, positive blood or bone marrow cultures

Presentation can be mild, self-limited (B. abortus) or fulminant with severe complications (B. melitensis)

Localized

Occurs with acute or chronic untreated disease

Related to involved organs

Positive serology, positive cultures in specific tissues

Bone or joint, genitourinary, hepatosplenic involvement most common

Relapsing

2-3 mo after initial episode

Same as acute illness but may have higher fever and more fatigue, weakness, chills, and sweats

Positive serology, positive cultures

May be extremely difficult to distinguish relapse from reinfection

Chronic

>1 yr

Nonspecific presentation, but neuropsychiatric symptoms and low-grade fever most common

Low titer or negative serology, negative cultures

Most controversial classification; localized disease may be associated

acute setting, which typically affects wrists, knees, hips, and ankles. Sacroiliitis typically occurs in the acute setting and is a classical presentation that should prompt workup for brucellosis. Spondylitis and vertebral osteomyelitis, particularly in the lumbar area, are less common but also well-recognized complications and can be associated with paravertebral, epidural, and psoas abscesses. Prosthetic joint infections have been observed, with 24 reported cases mostly due to B. melitensis followed by B. abortus. These typically involve a hip or knee replacement, and one fourth responded to combination antimicrobial therapy while the rest required débridement or revision.9

Relapsing Infection

Up to 10% of patients with brucellosis relapse after antimicrobial therapy. The intracellular location of Brucella organisms predisposes to recurrence because these are relatively protected from host defense mechanisms, and antimicrobial agents may be unable to penetrate efficiently enough to kill all the bacteria. Intrinsic or acquired resistance to antibiotics is another factor that can lead to treatment failure. Relapses usually occur 3 to 6 months after completion of therapy but may be seen up to 2 years after initial treatment. Relapsing infection is difficult to distinguish from reinfection in high-risk groups with continued exposure. Relapses are associated with inappropriate or insufficient antimicrobial therapy, growth on blood cultures during the initial presentation, and an acute onset of disease.

Chronic Disease

Disease lasting more than 1 year is referred to as chronic brucellosis. A majority of patients classified as having chronic brucellosis really have persistent disease caused by inadequate treatment of the initial episode, or they have focal disease in bone, liver, or spleen. About 20% of patients diagnosed with chronic brucellosis complain of persistent fatigue, malaise, and depression; in many respects, this condition resembles chronic fatigue syndrome.  

DIAGNOSIS

Culture

Many common illnesses mimic brucellosis, and a thorough history is essential, including occupation, travel to endemic areas, avocations, and ingestion of at-risk food and beverages. The most conclusive means of establishing the diagnosis of brucellosis is the recovery of the organism from a culture from normally sterile body fluid or tissue. Sensitivity of cultures range from 15 to 90%, depending on the methods used and the specimen type. Handling of Brucella cultures is potentially hazardous to laboratory personnel, and the laboratory should be informed if this organism is suspected, in addition to requesting them to extend the length of incubation because it may take more than 5 days for Brucella organisms to grow. Blood cultures are positive in 10 to 30% of cases of acute brucellosis, although this can be as high as 85% with B. melitensis. The sensitivity of blood cultures decreases with increasing duration of illness. In B. melitensis infection, bone marrow cultures are more sensitive than blood cultures. In localized brucellosis (e.g., lymph nodes, spleen, liver, skeletal system), cultures of purulent material or tissues usually yield Brucella organisms. Culture of cerebrospinal fluid is positive in 45% of patients with meningitis. Antibodies against Brucella may be demonstrated in cerebrospinal fluid by enzyme-linked immunosorbent assay (ELISA).

Standard Tube Agglutination and Other Antibody Tests

In the absence of microbiologic confirmation, a presumptive diagnosis can be made by history and serology. The most frequently used test is the standard tube agglutination test (SAT), measuring antibody titers against B. abortus antigen. A fourfold or greater rise in titer over 2 weeks is considered significant. A presumptive case is one in which the agglutination titer is positive (1 : 160 in endemic areas; 1 : 80 in nonendemic areas) in single or serial specimens, with symptoms consistent with brucellosis. By 3 weeks of illness, more than 97% of patients demonstrate serologic evidence of infection. SAT detects antibodies to B. abortus, B. suis, and B. melitensis but not to B. canis. Serologic confirmation of B. canis infection requires B. canis or B. ovis antigen. After adequate antibiotic treatment, significant SAT titers can persist for up to 2 years in 5 to 7% of cases. Because of this, SAT titers are not useful in differentiating relapsing infection from other febrile illnesses in patients with a history of past Brucella infections. Individuals with subclinical infection may demonstrate significant SAT titers. In chronic brucellosis, antibodies which do not agglutinate at neutral pH increase and replace agglutinating antibodies, and so a complementary test that detects antibodies to smooth LPS, such as the Brucella Coombs test is recommended. A lateral flow immunochromatographic assay (LFA) using the purified antigen is another non–agglutination-based test. The Center for Disease Control uses a modified SAT known as the Brucella microagglutination test (BMAT). Falsepositive SAT titers have been associated with Brucella skin testing, cholera vaccination, or infection with Vibrio cholerae, Francisella tularensis, or Yersinia enterocolitica.10

Other Tests

Polymerase chain reaction with DNA sequencing offers highly accurate diagnosis of brucellosis in blood and other fluids or tissue samples. However, protocols still need to be standardized on a wider scale, and access to expertise and adequate laboratory facilities remains a significant limiting factor.  

PREVENTION

The control of human brucellosis is directly related to prevention programs in domestic animals and the avoidance of unpasteurized milk and milk products. In slaughterhouses, important means of prevention include careful wound dressing, the use of protective glasses and clothing, the prohibition of raw meat ingestion, and the use of previously infected (immune) individuals in high-risk areas. Work is ongoing to find an effective vaccine for humans. Postexposure antimicrobial prophylaxis is controversial.  

PROGNOSIS

Brucellosis treated appropriately within the first month of symptom onset is curable. Acute brucellosis often produces severe weakness and fatigue, and patients are frequently unable to work for up to 2 months. Immunity to reinfection follows initial Brucella infection in the majority of individuals. With early antimicrobial therapy, cases of chronic brucellosis or localized disease and complications are rare. Of the patients who die of brucellosis, 84% have endocarditis involving a previously abnormal aortic valve, often associated with severe congestive heart failure. Medical treatment alone is associated with a much higher risk of death compared to a combined medical and surgical approach for Brucella endocarditis.

TABLE 294-2 TREATMENT FOR BRUCELLOSIS TREATMENT Acute, with no endocarditis or CNS involvement

Doxycycline (200 mg/day) for 6 weeks plus streptomycin (1 g/day) for 3 wk

Alternative agents: chloramphenicol, fluoroquinolones, TMP-SMX, imipenem

Combination therapy still preferred; ofloxacin (800 mg/day) plus rifampin (15 mg/kg/day) is the preferred alternative

In children

TMP-SMX plus rifampin

CNS

Doxycycline plus rifampin and TMP-SMX

Localized

Surgically drain abscesses plus antimicrobial therapy for ≥6 wk

Brucella endocarditis

Bactericidal drugs; early valve replacement may be necessary

COMMENTS Treatment of choice by WHO; widely used; low rate of relapse; IM administration of streptomycin may be difficult

Third-generation cephalosporin can be substituted if susceptible in vitro Possible aortic valve destruction and/or major arterial emboli

CNS = central nervous system; IM = intramuscular; TMP-SMX = trimethoprim–sulfamethoxazole; WHO = World Health Organization.

TREATMENT  Effective treatment of Brucella infection requires antibiotics that can penetrate the intracellular compartment, have little or no toxicity even with prolonged use for preventing relapse, and are bactericidal for adequate treatment of central nervous system infection and endocarditis. There remains considerable debate over which antibiotic regimen is best. In adults, the combination of oral doxycycline at 100 mg orally twice a day for 6 weeks plus streptomycin 1 g intramuscularly daily for 3 weeks has shown a higher cure rate and lower relapse than ofloxacin plus rifampin and doxycycline plus rifampin. A1  A quinolone plus rifampicin may be better tolerated than doxycyline plus rifampicin as alternative therapy. A2  These regimens are less effective for cases of spondylitis, which may require up to 3 months of treatment with any of the above regimens. Triple combination doxycycline, streptomycin, and rifampicin has been shown to clear Brucella DNA better than dual combination, and may be useful for treatment in chronic brucellosis. Recommendations are summarized in Table 294-2.

  Grade A References A1. Hashemi SH, Gachkar L, Keramat F, et al. Comparison of doxycycline-streptomycin, doxycyclinerifampin, and ofloxacin-rifampin in the treatment of brucellosis: a randomized clinical trial. Int J Infect Dis. 2012;16:e247-e251. A2. Yousefi-Nooraie R, Mortaz-Hejri S, Mehrani M, et al. Antibiotics for treating human brucellosis. Cochrane Database Syst Rev. 2012;10:CD007179.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 294 Brucellosis  

GENERAL REFERENCES 1. Franc KA, Krecek RC, Häsler BN, et al. Brucellosis remains a neglected disease in the developing world: a call for interdisciplinary action. BMC Public Health. 2018;18:1-9. 2. Georgi E, Walter MC, Pfalzgraf MT, et al. Whole genome sequencing of Brucella melitensis isolated from 57 patients in Germany reveals high diversity in strains from Middle East. PLoS ONE. 2017;12:1-15. 3. Franco-Paredes C, Chastain D, Taylor P, et al. Boar hunting and brucellosis caused by Brucella suis. Travel Med Infect Dis. 2017;16:18-22. 4. Dadar M, Shahali Y, Whatmore AM. Human brucellosis caused by raw dairy products: a review on the occurrence, major risk factors and prevention. Int J Food Microbiol. 2019;292:39-47. 5. Tuon FF, Gondolfo RB, Cerchiari N. Human-to-human transmission of Brucella—a systematic review. Trop Med Int Health. 2017;22:539-546.

1951.e1

6. Ghanem-Zoubi N, Eljay SP, Anis E, et al. Association between human brucellosis and adverse pregnancy outcome: a cross-sectional population-based study. Eur J Clin Microbiol Infect Dis. 2018;37:883-888. 7. Mirnejad R, Jazi FM, Mostafaei S, et al. Molecular investigation of virulence factors of Brucella melitensis and Brucella abortus strains isolated from clinical and non-clinical samples. Microb Pathog. 2017;109:8-14. 8. Giambartolomei GH, Arriola Benitez PC, Delpino MV. Brucella and osteoarticular cell activation: partners in crime. Front Microbiol. 2017;8:1-9. 9. Lewis JM, Folb J, Kalra S, et al. Brucella melitensis prosthetic joint infection in a traveller returning to the UK from Thailand: case report and review of the literature. Travel Med Infect Dis. 2016;14:444-450. 10. de Glanville WA, Conde-Álvarez R, Moriyón I, et al. Poor performance of the rapid test for human brucellosis in health facilities in Kenya. PLoS Negl Trop Dis. 2017;11:1-15.

1951.e2

CHAPTER 294 Brucellosis  

REVIEW QUESTIONS 1. A 38-year-old Iran-born male veterinarian from Milwaukee, Wisconsin develops fever and left sacroiliac pain over the past week. He does administer Brucella vaccines to cattle but does not recall any accidental exposure, and he always uses thick gloves when giving the vaccine. He hasn’t travelled back to the Middle East in the last 10 years. He has never had any similar symptoms before and he does not hunt. He recently came back from a trip from Mexico where he did eat cheese and drank milk on a farm. What is the most likely source of this patient’s acute illness? A . Relapse from a remote infection in childhood B. Unpasteurized dairy products from Mexico C. Accidental inoculation with strain 19 vaccine D. Feral pigs E. Cats he saw in clinic Answer: B  Unpasteurized dairy products from Mexico remain a major source of Brucella infection in the United States. A relapse of more than 10 years is highly unusual, although Iran is endemic for Brucella. Brucella vaccine is pathogenic in humans, but as long as proper protective equipment is used, infection is rare. Feral pigs are a source of infection in hunters who handle and dress infected animals. Although B. canis from dogs does occur in the United States, symptoms are typically mild, and are only very rarely transmitted by cats. 2. A 38-year-old male rancher from Texas is admitted with a 10-day history of dyspnea and fever. He had assisted in the stillbirth of a calf on his ranch. Chest radiography is consistent with congestive changes with small bilateral pleural effusions. Examination is significant for neck vein engorgement, fine bibasilar crackles, and a grade 3/6 systolic murmur. A transesophageal echocardiogram shows an 8-mm vegetation on the aortic valve. Vancomycin 1 g intravenously every 12 hours, ceftriaxone 2 g intravenously every 24 hours, and diuretics are started with some improvement. Brucella microagglutination test (BMAT) is sent to the CDC and comes back positive. Appropriate treatment for this patient is: A . Doxycycline alone B. Doxycycline in combination with gentamicin C. Doxycyline plus streptomycin plus early valve replacement D. Doxycycline plus trimethoprim-sulfamethoxazole plus early valve replacement E. Doxycyline alone plus delayed valve replacement Answer: C  Brucella endocarditis is a life-threatening illness and in the presence of congestive heart failure symptoms, early valve replacement substantially improves survival. Choice of antimicrobials should include bactericidal drugs. Doxycyline and trimethoprim are bacteriostatic, but gentamicin and streptomycin are bactericidal. Some advocate addition of another bactericidal drug such as imipenem on top of doxycyline and streptomycin.

3. Infection of the patient in question 2 could have been prevented through (multiple answers possible): A . Proper personal protective equipment B. Vaccination of cattle C. Prophylaxis with doxycyline D. Vaccination of humans E. Antibacterial baths weekly for all cattle Answer: A, B  Brucellosis can be prevented with proper use of personal protective equipment and vaccination of cattle. Prophylaxis is not a usual practice and its efficacy is unknown. There is no approved human vaccine, and animal vaccines are pathogenic to humans. Use of antibacterial baths has not been proven to prevent transmission of Brucella. 4. Brucella can produce debilitating symptoms especially in joints and can persist for months. Which of the following statements describes the pathogenesis of joint destruction by Brucella? A . Brucella has a modified lipopolysaccharide (LPS) that induces a strong immune response in the joint. B. Brucella induces rapid apoptosis of synoviocytes leading to extensive joint capsule necrosis. C. Brucella produces highly active toxins which lyse host cell walls and structures. D. Brucella through Omp25 induces robust secretion of TNF-α by macrophages during initial infection. E. Brucella is an intracellular organism that is able to persist in the joints, and inflammatory mediators from the innate immune response are mostly responsible for tissue destruction. Answer: E  Brucella does not secrete toxins and inhibits synoviocyte apoptosis leading to persistence of infection. The innate immune response produces host proinflammatory cytokines and metalloproteinases, eventually damaging the joints. Unlike regular bacterial LPS, modified Brucella LPS has little activity against TLR4 and is resistant to complement activation. Brucella inhibits TNF-α production in macrophages, and modulates permeability of the Brucella cell membrane. 5. A 28-year-old hunter develops fever and severe hip pain over the past 3 weeks. He has recently gone on a hunting tour around the world and has shot deer in Japan, hunted feral pigs in South Carolina, and gone fox hunting in Britain. He denies eating any unpasteurized milk and milk products. He does not have any pets. Magnetic resonance imaging shows sacroiliitis, and joint aspiration grows Brucella spp. The likely species of Brucella that is isolated is: A . B. melitensis B. B. suis C. B. abortus D. B. microti E. B. canis Answer: B  His likely exposure is from feral pigs in South Carolina which is known to have seropositive feral pigs. There are no reports of Brucella infection from deer in Japan, and no human cases of B. microti from foxes. There is no epidemiologic risk factor for B. abortus (exposure to placentas/still births). B. melitensis can be acquired from sheep and goats, or eating unpasteurized dairy products, none of which were in the patient’s history.

CHAPTER 295  TULAREMIA AND OTHER FRANCISELLA INFECTIONS  

1951

flies can transmit the infection from animals to humans and provided detailed descriptions of its clinical manifestations. Colloquially, the disease is often referred to as rabbit fever or deer fly fever.

The Pathogen

The organism occurs in two major subspecies (biovars). F. tularensis biovar tularensis (type A) is more virulent in animals and humans, has distinctive biochemical reactions (it produces acid from glycerol and has citrulline ureidase activity), and is the common North American biovar. In contrast, F. tularensis biovar holarctica (type B) is less virulent and occurs commonly in Europe and Asia.2 Type B is most frequently isolated from rodent species, including muskrats (Ondatra zibethicus), mice (Mus musculus), beavers (Castor canadensis), voles (Microtus spp), and water voles (Arvicola terrestris), and has been associated with an outbreak of infection in wild-caught prairie dogs. Specific virulence factors for F. tularensis have not been identified.  

295  TULAREMIA AND OTHER FRANCISELLA INFECTIONS KAREN C. BLOCH AND WILLIAM SCHAFFNER



DEFINITION

Tularemia is an infectious zoonosis caused by Francisella tularensis, a small aerobic, pleomorphic, gram-negative bacillus. Many animal species harbor the organism, most prominently rabbits, squirrels, and muskrats. Humans acquire the infection through various means, including direct contact with infected animal tissues, ingestion of contaminated water or meat, the bite of an infected tick or deer fly, or breathing an aerosol of bacteria.1 F. tularensis is highly infectious and is a well-recognized risk to laboratory personnel manipulating culture plates of the organism; paradoxically, the illness is not communicable from person to person. Edward Francis established that deer

EPIDEMIOLOGY

Tularemia has been reported in the United States, Canada, Mexico, Japan, and Europe (particularly Scandinavia).3 In Iran, seroprevalence ranges from 6 to 14%.4 It has not been reported in the United Kingdom or the Southern Hemisphere. In the United States, reported cases diminished during the second half of the 20th century from a high of 2291 cases in 1939 to the approximately 125 cases reported annually (E-Fig. 295-1). In 2015 the Centers for Disease Control and Prevention (CDC) reported 314 cases, driven by a 975% increase in the incidence of tularemia in Colorado, Nebraska, South Dakota, and Wyoming.5 The increased incidence of tularemia in these states is notable, because although the disease has occurred in all the continental states, historically more than half of reported cases have been from Arkansas, Missouri, Oklahoma, and South Dakota. The island of Martha’s Vineyard off the coast of Massachusetts is also a focus of tularemia. In the United States, tularemia is usually acquired from tick bites or from contact with infected animals, especially rabbits. Tick-associated cases now constitute the most common route of transmission, and primarily occur during the summer.6 The most common vectors in the United States are the wood tick (Dermacentor andersoni), the dog tick (Dermacentor variabilis), and the Lone Star tick (Amblyomma americanum). A smaller peak of autumn and winter cases is a consequence of rabbit hunters skinning and eviscerating their game. Public health education materials aimed at decreasing the hazards of handling wild animals have contributed to the reduction of tularemia in hunters. Mosquitoes are the common vectors in northern Europe. Occasional individuals acquire infection from the bite of an infected animal or, more likely, from the bite of an animal whose mouth was contaminated from recently eating a diseased animal. The latter likely explains most instances of cat-bite tularemia. Males experience a higher incidence of disease than females in all age groups, probably as a consequence of their greater exposure to the outdoors and animal-related activities and less use of protective measures against tick bites. Persons in all age groups are affected, with children 5 to 14 years of age and older adults most prominently represented. In the United States, American Indians and Alaska natives experience the highest annual incidence (0.5 per 100,000); whites have a lower risk (0.04 per 100,000), and African Americans and Asians/Pacific Islanders have the lowest occurrence of tularemia (≤0.01 per 100,000). Although tularemia is usually a sporadic infection, outbreaks of disease have been traced to laboratory exposure, contaminated groundwater, muskrat

CHAPTER 295  TULAREMIA AND OTHER FRANCISELLA INFECTIONS  

ABSTRACT

Tularemia is a zoonotic infection that causes human disease throughout the world. Infection is spread through direct contact with animals, ingestion of contaminated food or water, arthropod bite, or inhalation of aerosolized bacteria. This latter mechanism of transmission makes tularemia a potential agent of bioterrorism. Classically, six distinct clinical syndromes have been described, although there is significant overlap among these presentations. Symptoms develop within a week of exposure, and fever is almost universally present. Clinical presentations include ulceroglandular disease characterized by skin lesions and lymphadenopathy, typhoidal disease characterized by an undifferentiated febrile illness, oculoglandular disease characterized by ocular inflammation and regional lymphadenopathy, oropharyngeal disease characterized by pharyngitis, and pneumonic disease. Regardless of the syndrome, the recommended treatment is streptomycin or gentamicin, but other agents including quinolones have activity against this organism. Currently there is no vaccine commercially available, and prevention entails reducing exposure to vectors.

KEYWORDS

tularemia zoonoses vector-borne infections bioterrorism agents pneumonia lymphadenopathy skin ulcers pulse-temperature dissociation ulceroglandular oculoglandular

1951.e3

1951.e4

CHAPTER 295  TULAREMIA AND OTHER FRANCISELLA INFECTIONS  

1000 900 800

Cases

700 600 500 400 300 200 100 2015

2010

2005

2000

1995

1990

1985

1980

1975

1970

1965

1960

1955

1950

0

Year E-FIGURE 295-1.  Tularemia cases reported annually in the United States, 1950–2015. (Source: https://www.cdc.gov/tularemia/statistics/index.html.)

1952

CHAPTER 295  TULAREMIA AND OTHER FRANCISELLA INFECTIONS  

handling, dressing a dead hare, lawn mowing, and brush cutting. In the latter two cases, primary pneumonic tularemia apparently occurred when the affected individuals created an environmental aerosol by mowing grass and cutting brush that had been contaminated with F. tularensis excreted in the urine and feces of infected rodents. An outbreak of oropharyngeal tularemia in Turkey affecting 55 patients was ultimately traced to contaminated tap water. The organism can survive in water, mud, and straw for weeks to months. Interest in tularemia has been enhanced because of its potential use as a bioterrorism agent.7 Its high infectivity (as few as 10 organisms induces pneumonic disease), its ease of dissemination, and the difficulty of rapidly diagnosing acute illness are characteristics that merit its inclusion among threat agents. Thus tularemia must be reported immediately to local public health authorities. Unusual patterns of disease will be investigated for both conventional and bioterrorist sources.  

PATHOBIOLOGY

F. tularensis can infect humans through several portals of entry, including the skin, mucous membranes, and gastrointestinal and respiratory tracts. It requires intracellular residence and can multiply within macrophages and other cells. A large, not completely understood secretion apparatus is found in Francisella, known as a type VI secretion system (T6SS). It is essential for Francisella to escape from their phagosomes and multiply within host macrophages to cause disease.8 After inoculation into the skin and subcutaneous tissue, local bacterial multiplication occurs and evokes a suppurative necrotic reaction characterized by an initial polymorphonuclear response followed by an influx of macrophages and lymphocytes. These suppurative lesions evolve into granulomas. Bacteremia can occur both early and late during this process. The infection can disseminate to the lymph nodes, liver, spleen, lungs, and pleura. Viable F. tularensis can persist in tissues for long periods, contributing to the tendency to relapse after treatment.  

CLINICAL MANIFESTATIONS

Classically, the clinical manifestations of tularemia have been separated into six categories: ulceroglandular, glandular, oculoglandular, typhoidal, oropharyngeal, and pneumonic.9 Although this classification has historic roots, it should not be used rigidly because many patients have features of several types. The course of illness is determined by the portal of entry, the degree of systemic involvement, and the dose and virulence of the infecting strain of F. tularensis.

The general features of tularemia are similar regardless of the portal of entry. After exposure, the usual incubation period is 3 to 5 days (range, 1 to 21 days). The disease begins abruptly with the onset of fever (≥101° F), chills, malaise, and headache. Myalgia, vomiting, sore throat, and abdominal pain can also occur. Almost half the patients have a pulse rate that is substantially slower than would be anticipated based on the degree of fever (pulse-temperature dissociation). The fever may abate somewhat after 1 to 3 days, only to recur and continue along with other symptoms for 2 to 3 weeks. Untreated, weight loss, easy fatigability, and lymphadenopathy may persist for weeks longer.

Ulceroglandular Disease

Ulceroglandular disease is the form of infection most readily recognized by physicians. Along with fever and other constitutional symptoms, the patient calls attention to tender, swollen lymph nodes that drain an inoculation site. The nodes are usually axillary or inguinal, and a local lesion appears concurrently or 1 or 2 days before or after the lymphadenopathy. The lesions at the site of inoculation begin as small, red, tender, or painful papules that progress to pustules and then undergo necrosis to produce an ulcer with sharp, somewhat elevated edges and a flat base that becomes black (Fig. 295-1). Untreated, the ulcers heal over a period of weeks and leave scars.10 Tick-induced infections produce lesions on the trunk, about the waist, and in the perineum, along with the expected local adenopathy. Children typically have occipital and cervical adenopathy from tick bites on the neck and in the hair. Animal exposure often produces lesions on the hands and forearms. Lesions may be multiple. Because the organisms evoke a localized granulomatous response, frank lymphangitis does not occur in uncomplicated tularemia, but an occasional patient manifests a chain of nodules in “sporotrichoid” fashion along the lymphatic drainage. Patients with such apparently “localized” disease often have symptoms and findings indicating a more widespread infection. Sore throat with or without an erythematous pharynx occurs, as well as chest radiographic findings of patchy infiltrates in the lower lobes, pleural effusions, and hilar adenopathy.

Glandular and Typhoidal Disease

Glandular disease is essentially the same clinical syndrome as ulceroglandular disease but without the local lesion. Thus the patient has fever, constitutional symptoms, and lymphadenopathy. The local lesion may have been on a part of the body where it was not seen, or it may have been small and already healed

A

B

C

D

FIGURE 295-1.  Ulcerative tularemia. (Source: Holland SD, Michelow IC. Tularemia masquerading as ecthyma. J Pediatr. 2016 Nov; 178.299. doi:10.1016/j.jpeds.2016.07.023. [Epub

2016 Aug 22.])

by the time the patient sought medical care. Glandular disease accounts for only 3 to 20% of cases. Typhoidal disease does not show evidence of lymphadenopathy and is essentially characterized by fever of unknown cause. These illnesses evade diagnosis unless the physician specifically considers the possibility of tularemia and inquires about tick or animal exposure. Occasionally, the diagnosis is made fortuitously when a positive blood culture is reported.

Oculoglandular Disease

Oculoglandular disease is rare (1 —

ENVIRONMENTAL FACTORS Travel Recent plumbing work in home or at work Hospitalization Exposure to contaminated water sources—cooling towers, hot tubs, decorative fountains Exposure to potting soil (Australia) for Legionella longbeachae

2 2 — — —

The most common risk factors for the acquisition of legionnaires disease are listed in Table 298-1.5,6 Legionella species account for 1 to 5% of communityacquired pneumonia requiring admission to a hospital. In some areas, Legionella accounts for about the same percentage of pneumonia treated on an ambulatory basis. Legionella infections can be sporadic or occur in outbreaks. Outbreaks have been associated with exposure to a variety of aerosol-producing devices, including showers, a grocery store mist machine, cooling towers, whirlpool spas, decorative fountains, and evaporative condensers.7 Other water sources implicated in transmission of legionnaires disease include water on trains, birthing pools, dental units, asphalt paving machines, and windscreen wiper fluid without added screen wash. Legionnaires disease can be acquired up to 10 to 11 km away from contaminated cooling towers. Aspiration of contaminated potable water by immunosuppressed patients is another mechanism by which Legionella is acquired. The U.S. Centers for Disease Control and

Prevention (CDC) investigated 38 outbreaks of legionnaires disease from 2000 to 2014, 27 of which were land based. Median number of cases per outbreak was 10, ranging from 3 to 82. In 85% there was enough information to evaluate maintenance deficiencies.8 These included process failures, human errors, equipment failures, and unmanaged external changes.9 It was also noted that health care–associated outbreaks accounted for 57% of the cases and 85% of the deaths. Seven of the health care–associated outbreaks were in long-term care facilities. Hotel travel–associated cases tend to be due to contaminated cooling towers and potable water, while ship travel–associated cases are usually due to contaminated hot tubs. It is important to note that outbreak-associated cases account for the minority, only 5 to 10%, of cases of legionnaires disease. Legionnaires disease in transplant recipients is frequently due to nonPneumophila species, and hence urinary antigen cannot be relied on as a diagnostic tool in this setting. Only about 20% of the Legionella infections in this setting occur within the first 3 months after transplantation. A number of biologic modifying agents used to treat cancer, autoimmune disease, multiple sclerosis, and other factors that compromise hosts seem to increase the risk for legionnaires disease. For example, treatment with infliximab results in a 15-fold increase in risk, and adalimumab with a 38-fold increase.10 Exposure to contaminated potting soil is a risk factor for L. longbeachae infection in Australia and New Zealand. Pontiac fever occurs predominantly in outbreaks with very high attack rates. L. pneumophila, L. micdadei, and L. anisa have been implicated in outbreaks of Pontiac fever. Among residents of nursing homes, Pontiac fever has been associated with L. pneumophila concentrations of greater than 104 colonyforming units/L in shower water. Those receiving corticosteroid therapy have a six-fold higher risk for development of Pontiac fever.  

PATHOBIOLOGY

After being inhaled, legionellae are phagocytosed in the lungs by alveolar macrophages. Only virulent strains of Legionella are capable of initiating organism-directed endocytosis when attachment to the alveolar macrophage through E-cadherin and β1-integrin receptors occurs.11 Legionellae abrogate phagosome-lysosome fusion and replicate in an endosome surrounded by the endoplasmic reticulum. Once it is intracellular, the bacteria-laden endosome recruits small vesicles, mitochondria, and ribosomes, and within 4 to 6 hours it becomes enveloped by the endoplasmic reticulum, thereby establishing the replicative endosome. After a latent period of about 12 hours, the bacteria start dividing. During this time, there is synthesis of up to 35 proteins and repression of 32 proteins. Iron must be available in the phagosome for growth. Growth continues in the macrophages for approximately 24 hours, at which time the macrophage dies and the bacteria are released. The released bacteria

CHAPTER 298  LEGIONELLA INFECTIONS  

are often phagocytosed by other macrophages, dendritic cells, and epithelial cells, perpetuating the infection. Cell-mediated immunity is necessary for recovery from Legionella infection. Production of type 1 interferons has a protective effect by promoting the activation of macrophages. Activated macrophages limit the intracellular replication of legionellae by downregulating the expression of their transferrin receptors and limiting the availability of iron to the bacteria. The pathogenesis of Pontiac fever is unclear. The onset of illness occurs within 12 to 36 hours after the inhalation of, presumably, endotoxin. This period is too short for bacterial multiplication to cause the symptoms.  

CLINICAL MANIFESTATIONS

Most of our knowledge of the clinical features of legionnaires disease comes from studying patients who have been hospitalized with this illness, that is, those with the most severe manifestations. Fever (often high), malaise, and cough are present in most patients. Chills occur in about 75%, and dyspnea in just more than half the patients. Other features include myalgias, headache, chest pain, and diarrhea. The cough is nonproductive in 50% of patients; others have scant sputum production that is usually mucoid, rarely purulent, and very rarely bloody.12 There are no clinical features that distinguish individual patients with legionnaires disease from those with pneumonia caused by other pathogens.13 However, when patients with legionnaires disease are compared with those with community-acquired pneumonia due to other agents, the patients with legionnaires disease are more likely to have myalgias, headache, diarrhea, and a higher mean oral temperature at the time of presentation. They also present to the hospital sooner after the onset of symptoms, 4.7 days versus 7.7 days. When patients with legionnaires disease were compared with patients with bacteremic pneumococcal pneumonia, the following features were more associated with Legionella pneumonia: male sex, heavy drinking of alcohol, previous β-lactam therapy, temperature higher than 39° C, myalgias, and gastrointestinal symptoms. Pleuritic chest pain and purulent sputum were less likely to be present. In a young, otherwise healthy person with rapidly progressive pneumonia (especially if the progression occurs in the setting of β-lactam therapy), legionnaires disease should be strongly suspected. Mental confusion is common, and on occasion, the presentation is dominated by extrapulmonary manifestations such as reactive arthritis, cerebellar ataxia, seizures, myoclonus, or encephalitis. Rarely, extrapulmonary infection, such as prosthetic valve endocarditis, sinusitis, dialysis shunt infection, or abscess formation, occurs. Physical findings include fever, tachypnea, relative bradycardia, and initially only a few crackles on chest examination. Later, the findings of pulmonary consolidation are not uncommon. Abdominal examination is usually unremarkable. Rash as a manifestation of legionnaires disease is very rare. Progression of the illness is not uncommon, even after the institution of antibiotic therapy. About half the patients with legionnaires disease who require hospitalization have a complicated course. The clinical presentations and outcomes of Legionella pneumonia in HIV-infected patients have been noted to be comparable to those without HIV infection.14 Survivors of legionnaires disease in one study had major sequelae persisting 17 months after diagnosis. Pontiac fever has an incubation period of about 36 hours. Fever, severe myalgia, headache, and extreme fatigue are the dominant manifestations. The illness is of short duration, lasting, on average, 3 days.  

1965

The sensitivity of the urinary antigen test in a review of published data was 0.74 (0.68 to 0.81), and the specificity was 0.991 (0.984 to 0.997). Rarely, the urinary antigen test result can remain positive for up to 1 year. Use of this test has allowed early diagnosis of legionnaires disease because of the very short time required to do the test. This may be a factor in the lower mortality rates from legionnaires disease compared with historical rates. If the urinary antigen test is positive for Legionella, it should be repeated after boiling the urine for 5 minutes and centrifuging it at 12,000 g for 5 minutes to avoid false-positive results. In hospitals with high numbers of immunosuppressed patients where legionellae other than serogroup 1 are more common, urinary antigen should not be relied on to make a diagnosis of legionnaires disease. Sputum culture has a low sensitivity but is 100% specific. It should be performed on all patients suspected of having legionnaires disease. Serologic tests are not useful in the immediate management of a patient because of the long time (6 to 12 weeks) required to seroconvert; however, they do have a role in the work-up of outbreaks of legionnaires disease. False-negative and falsepositive serologic results do occur. A four-fold or greater increase in antibody titer between the acute and convalescent phase serum samples is diagnostic. A former criterion of a stable antibody titer of 1 : 256 or higher is no longer considered diagnostic. PCR can be used to amplify Legionella DNA in sputum, bronchoalveolar lavage fluid, pleural fluid, pulmonary tissue, or serum. PCR can detect 1 fg of Legionella DNA, equivalent to one microorganism. These tests have not yet gained widespread use clinically. In everyday use, PCR has a higher sensitivity than culture and is about 30% more sensitive than legionella urinary antigen.17 Legionella can be isolated from the blood with special media or by subculturing onto BCYE (buffered charcoal–yeast extract) agar plates, but this is not used in practice. A chest radiograph is necessary to establish a diagnosis of pneumonia. About half the patients with legionnaires disease have unilateral pulmonary involvement. The lower lobes are involved most commonly. About one third of patients have a pleural effusion. Dense opacification is common, but interstitial and nodular opacities also occur. Cavitation is uncommon; 70% of the 79 patients reported to date with lung abscess due to Legionella were receiving corticosteroids. Figures 298-3 to 298-6 illustrate some of the radiographic findings in legionnaires disease. The diagnosis of Pontiac fever is based on demonstration of Legionella in water to which the patient was exposed, seroconversion to Legionella, and a compatible clinical course. In patients who have died of legionnaires disease, the gross pathology examination shows focal or patchy lesions in about one third of cases, lobar pneumonia in about half, and focal hemorrhages in about one fourth. On microscopic examination, there is bronchopneumonia with diffuse alveolar damage and heavy infiltration of neutrophils, macrophages, desquamation of alveolar epithelial cells, and fibrin proteinaceous debris. On occasion, there is inflammation of blood vessels mimicking a vasculitis. Organisms can be visualized with Dieterle silver impregnation stain or by direct immunofluorescent staining (E-Fig. 298-1).

DIAGNOSIS

It is most important to have a high index of clinical suspicion that a patient might have legionnaires disease.15 Routine laboratory test results are nonspecifically abnormal. Leukocytosis is common; leukopenia, thrombocytopenia, and disseminated intravascular coagulation also occur. Other laboratory abnormalities may include hyponatremia (in about half the patients, sometimes severe), hypophosphatemia (also common, occurring early and resolving within a few days of the initiation of treatment), mild liver function test abnormalities (except for alkaline phosphatase, which is occasionally very elevated), elevated creatine kinase (occasionally with rhabdomyolysis), microscopic hematuria, and mild proteinuria. High procalcitonin levels exceeding 1.5 are associated with a higher rate of admission to intensive care units and death. Combinations of findings may be suggestive of legionnaires disease. These include high temperature, absence of sputum production, high lactate levels, increased C-reactive protein level, and low platelet counts. There are a number of specific tests for the diagnosis of legionnaires disease. Tests to detect L. pneumophila SG 1 antigen in urine are available commercially.16 These are easy to use, but there is a false-negative rate of up to 26%.

FIGURE 298-3.  Posteroanterior chest radiograph of a patient with community-acquired pneumonia due to Legionella pneumophila. Note the dense consolidation of the right upper lobe, with bulging of the fissure. Such dense consolidation is a common radiographic appearance of legionnaires disease.

CHAPTER 298  LEGIONELLA INFECTIONS  

Toll-like receptors 2, 4, 5, and 9 are activated during infection with L. pneumophila. Interleukin-1α, -1β, -4, -6, -12, and -18 are detected during Legionella infection. Vaccination of guinea pigs with the purified major outer membrane protein OmpS protects against an LD100 lethal challenge, whereas immunization with purified heat shock protein 60 provides little protection. Polymorphonuclear leukocytes are present in abundance in the infected lung, but their role in clearing the infection is unclear. Bacteria can spread beyond the lung and cause metastatic infection. However, many extrapulmonary effects of legionnaires disease, such as cerebellar ataxia and confusion, are not due to metastatic infection; they are presumably due to as yet unidentified toxins. The infected lung is consolidated, and there is usually no parenchymal damage after recovery occurs. Abscess formation and bronchiolitis obliterans or fibrosing alveolitis are occasionally seen.

1965.e1

1965.e2

CHAPTER 298  LEGIONELLA INFECTIONS  

A

B

C

D

E-FIGURE 298-1.  A, Gross pathology specimen of formalin-fixed lung from a patient with nosocomial legionnaires disease. Note the multiple nodular consolidated areas. B, Gross appearance of fresh lung tissue from a patient with legionnaires disease. Note the “hepatization of the lung” representing consolidation. C, Dieterle stain on lung tissue from a patient with legionnaires disease showing many microorganisms (magnification ×1000). D, Histologic appearance of pulmonary tissue from a patient with legionnaires disease. Note the dense inflammatory infiltrate with evolving microabscess (magnification ×200).

1966

CHAPTER 298  LEGIONELLA INFECTIONS  

FIGURE 298-4.  Posteroanterior chest radiograph of a patient with community-acquired legionnaires disease (Legionella pneumophila) manifesting as a right lower lobe nodular opacity.

FIGURE 298-6.  Posteroanterior chest radiograph of a patient with community-acquired pneumonia due to Legionella pneumophila. There is patchy consolidation at the right base, with subsegmental atelectasis and elevation of the right hemidiaphragm.

TABLE 298-2 TREATMENT FOR LEGIONNAIRES DISEASE SEVERITY OF LEGIONELLA PNEUMONIA Mild pneumonia in a nonimmunocompromised person treated at home

DRUG Azithromycin Clarithromycin Doxycycline Levofloxacin Moxifloxacin

Pneumonia requiring hospitalization

Levofloxacin Azithromycin Moxifloxacin

FIGURE 298-5.  Posteroanterior chest radiograph of a patient with community-acquired Legionella feeleii pneumonia. There is patchy consolidation of the right upper lobe.

Differential Diagnosis

Legionnaires disease should be considered in any patient with pneumonia who is admitted to the hospital, especially those who require treatment in an intensive care unit. If Legionella is present in a hospital’s water supply, legionnaires disease should be considered in all patients with nosocomial pneumonia.

TREATMENT AND PROGNOSIS  The absence of data from randomized clinical trials limits the availability of high-grade evidence from which to make recommendations for the treatment of Legionella infections. However, the convergence of the results of data from animal experiments, large observational studies, and meta-analysis does provide guidance to current recommendations (Table 298-2). Before reviewing these data, it is important to understand that the mortality from Legionella infections in the United States had declined from 1980 through 1998. During that time, 6757 cases were reported to the CDC, and use of urinary antigen as a method of diagnosis increased from 0 to 69%. The case-fatality rate for communityacquired legionellosis declined from 26 to 10% and for nosocomial infections with Legionella from 46 to 14%. Although early diagnosis with Legionella urinary

Pneumonia in immunocompromised host

Levofloxacin Azithromycin Moxifloxacin

DOSAGE 500 mg once daily for 3 days PO 500 mg bid for 5-7 days 200 mg loading dose, then 100 mg bid for 5-7 days 500 mg once daily for 5 days 400 mg once daily for 5 days 750 mg once daily (IV initially) for 10 days 500 mg once daily (IV initially) for 10 days 400 mg once daily (IV initially) for 10 days 750 mg once daily (IV initially) for 21 days 500 mg once daily (IV initially) for 21 days 400 mg once daily (IV initially) for 21 days

antigen may have played a role in this decrease, it is important to note that new therapeutic options such as levofloxacin were introduced in 1998 and may also have played a role in this reduction in mortality. In terms of recommending therapy, there are really three categories of Legionella infection: mild and moderate-severe community-acquired legionnaires disease and nosocomial legionnaires disease/immunocompromised host. Review of the results of treatment of 446 patients with legionnaires disease, of whom 175 were treated with levofloxacin, 177 with azithromycin, and 58 with clarithromycin, showed that there were no significant differences in time to defervescence or time to clinical stability between those treated with levofloxacin or azithromycin.18 Clarithromycin-treated patients had longer duration of intravenous antibiotic therapy and longer length of stay compared with levofloxacin-treated patients. The overall mortality rate was 4.3%. The immunocompromised patient with legionnaires disease is a special case. In one study of 49 patients with legionnaires disease and cancer or hematologic malignancy, the case-fatality rate was 31%. The median time to a clinical response was 8 days. Thirty-five percent required prolonged treatment of 25 days, and

two patients relapsed despite appropriate therapy. Twenty-seven percent who received combination therapy failed, and 34% who received monotherapy failed. None of the seven patients treated with clarithromycin or azithromycin failed. The addition of rifampin to a macrolide or a fluoroquinolone has to be carefully used in transplant patients because of interaction with immunosuppressive drugs. Thus, in this setting, azithromycin or a fluoroquinolone is considered first-line therapy.

The incidence of legionnaires disease is rising, and the mortality rate remains high, particularly for immunocompromised patients.19 Cases of legionnaires disease should be reported to local health officials. An investigation is often required to determine the source of the Legionella.   GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 298  LEGIONELLA INFECTIONS  

GENERAL REFERENCES 1. Massis LM, Assis-Marques MA, Castanheira FVS, et al. Legionella longbeachae is immunologically silent and highly virulent in vivo. J Infect Dis. 2017;215:440-451. 2. Cameron RL, Pollock KG, Lindsay DS, et al. Comparison of Legionella longbeachae and Legionella pneumophila cases in Scotland. Implications for diagnosis, treatment and public health response. J Med Microbiol. 2016;65:142-146. 3. Garrison LE, Kunz JM, Cooley LA, et al. Vital signs: deficiencies in environmental control identified in outbreaks of legionnaires’ disease—North America, 2000-2014. MMWR Morb Mortal Wkly Rep. 2016;65:576-584. 4. Rucinski SL, Murphy MP, Kies KD, et al. Eight years of clinical Legionella PCR testing illustrate seasonal pattern. J Infect Dis. 2018;218;669-670. 5. Orkis LT, Harrison LH, Mertz KJ, et al. Environmental sources of community-acquired legionnaires’ disease: a review. Int J Hyg Environ Health. 2018;221:764-774. 6. Burillo A, Pedro-Botet ML, Bouza E. Microbiology and epidemiology of Legionnaire’s disease. Infect Dis Clin North Am. 2017;31:7-27. 7. Hlavsa MC, Cikesh BL, Roberts VA, et al. Outbreaks associated with treated recreational water— United States, 2000-2014. MMWR Morb Mortal Wkly Rep. 2018;67:547-551. 8. Dooling KL, Toews KA, Hicks LA, et al. Active bacterial core surveillance for legionellosis, United States, 2011-2013. MMWR. 2015;64:1190-1192. 9. Hamilton KA, Prussin AJ 2nd, Ahmed W, et al. Outbreaks of Legionnaire’s disease and Pontiac Fever 2006-2017. Curr Environ Health Rep. 2018;5:263-271.

1967.e1

10. Lanternier F, Ader F, Pilmis B, et al. Legionnaire’s disease in compromised hosts. Infect Dis Clin North Am. 2017;31:125-135. 11. Oliva G, Sahr T, Buchrieser C. The life cycle of L. pneumophila: cellular differentiation is linked to virulence and metabolism. Front Cell Infect Microbiol. 2018;8:1-120. 12. Casey JD, Engelert JA, Katz JT, et al. An unexpected expectoration. N Engl J Med. 2018;378: 853-858. 13. Cunha BA, Burillo A, Bouza E. Legionnaires’ disease. Lancet. 2016;387:376-385. 14. Cillóniz C, Miguel-Escuder L, Pedro-Bonet ML, et al. Community-acquired Legionella pneumonia in HIV-infected adult patients: a matched case-control study. Clin Infect Dis. 2018;67:958-961. 15. Cunha BA, Cunha CB. Legionnaire’s disease: a clinical diagnostic approach. Infect Dis Clin North Am. 2017;31:81-93. 16. Viasus D, Calatayud L, McBrown MV, et al. Urinary antigen testing in community-acquired pneumonia in adults: an update. Expert Rev Anti Infect Ther. 2019;17:107-115. 17. Cristovam E, Almeida D, Caldeira D, et al. Accuracy of diagnostic tests for Legionnaires’ disease: a systematic review. J Med Microbiol. 2017;66:485-489. 18. Garcia Vidal C, Sanchez-Rodriguez I, Simonetti AF, et al. Levofloxacin vs azithromycin for treating legionella pneumonia a propensity score analysis. Clin Microbiol Infect. 2017;23: 653-658. 19. Herwaldt LA, Marra AR. Legionella: a reemerging pathogen. Curr Opin Infect Dis. 2018;31: 325-333.

1967.e2

CHAPTER 298  LEGIONELLA INFECTIONS  

REVIEW QUESTIONS 1. All of the following are risk factors for legionnaires disease, except: A . Cigarette smoking B. Travel to Europe C. Treatment with anti–tumor necrosis factor D. Using potting soil for your garden in Australia E. Being a member of the American Legion Answer: E  The disease was first recognized during a convention by the American Legion and was caused by water contamination with the organism at the hotel in Philadelphia that served as the meeting headquarters, not by any host factors. All of the other factors have been shown in one or more studies to increase one’s risk for acquiring legionnaires disease. 2. You have just seen a 55-year-old man with right lower lobe pneumonia. He has been receiving dialysis three times per week at an outpatient facility. He is quite ill, and you are concerned that he might have legionnaires disease. The result of the Legionella urinary antigen test that you ordered is negative. Which of the following is the most correct? A . Your patient does not have legionnaires disease. B. Your patient may still have legionnaires disease. C. Serologic testing will confirm the diagnosis. D. Blood cultures will be positive for Legionella. Answer: B  The urine antigen test detects only infection due to L. pneumophila serogroup 1, and even then it is only 74% sensitive. So your patient could still have legionnaires disease, either serogroup 1 or another Legionella species. Sputum culture, if positive, is 100% specific. Unfortunately, most patients with legionnaires disease do not produce sputum, and even more problematic is the fact that many laboratories are unable to grow this microorganism. Serologic diagnosis takes 4 to 6 weeks, so it is not of any use in therapeutic decision making. Also, the sensitivity can be low, especially in patients with an impaired immune response. 3. A patient has developed legionnaires disease while hospitalized for treatment of acute pancreatitis. He spent 3 weeks in intensive care and had a nasogastric tube in place for most of that time. What was the most likely source of the patient’s Legionella? A . He acquired it before admission. B. His daughter, who lives in Athens, came to visit him. She was coughing frequently during her many visits, and it is likely he acquired it from her. C. The air-conditioning system D. The potable water

Answer: D  In all likelihood, the potable water is the source. You should check the protocol for nasogastric tubes in your hospital. Not uncommonly, potable water is used to initiate feeding. If the potable water is contaminated with Legionella, it is readily aspirated into the lungs in patients who are in a recumbent position in an intensive care unit. Person-to-person transmission of Legionella has been documented. If the air-conditioning system was contaminated, it would likely be through a cooling tower, and there would have likely been an outbreak of many cases of Legionella. Whereas it is always possible that Legionella was acquired before admission, the incubation period for legionnaires disease is 2 to 10 days. It is possible to have a much longer incubation period. In the event of other cases in the hospital, molecular biology typing techniques can be used to identify the organism as coming from the potable water. 4. The following are extrapulmonary manifestations of legionnaires disease. A . Prosthetic valve endocarditis B. Cerebellar ataxia C. Reactive arthritis D. Oral ulcers E. A, B, and C Answer: E  Prosthetic valve endocarditis is due to direct infection of the valve by circulating Legionella bacteria. Cerebellar ataxia and reactive arthritis represent immunologic reactions in these tissues and are not due to direct infection. 5. The treatment of choice for moderate to severe legionnaires disease is: A . Erythromycin B. Amoxicillin C. Levofloxacin D. Vancomycin E. Ceftriaxone Answer: C  Data from multiple studies suggest that a respiratory fluoroquinolone such as levofloxacin or azithromycin is effective in the treatment of moderate to severe legionnaires disease. We do know that β-lactams are ineffective and that erythromycin, although initially used to treat legionnaires disease, takes 4 to 5 days to show improvement and is no longer recommended. Indeed, most patients get worse before they get better after treatment with erythromycin. Vancomycin would not be expected to work on a microorganism with a gram-negative cell wall.

CHAPTER 299  BARTONELLA INFECTIONS  

koehlerae, B. mayotimonensis, and B. ancashensis (Table 299-1). The other Bartonella species have been isolated only from the blood of animals, including rodents, felids, canids, dolphins, bats, and ruminants. The route of transmission of Bartonella species in mammals and humans is by fleas, ticks, mites, and lice (see Table 299-1). Bartonella infections are emerging infectious diseases that lead to a wide spectrum of either acute or chronic diseases. The status of the host immune response plays an important role in the development of the different manifestations. Four different clinical syndromes may occur with Bartonella infections: (1) infection of red blood cells and erythrophagocytosis, (2) granulomatous disease controlled by the immune response, (3) blood culture–negative endocarditis and bacteremia, and (4) vasculoproliferative diseases.1 A single Bartonella species can cause either acute or chronic infections and either vasculoproliferative or suppurative manifestations, but with different pathogenetic mechanisms that mainly depend on the patient’s immune status. For example, B. quintana is responsible for trench fever as well as for endocarditis, bacteremia in the homeless population, and vasculoproliferative diseases, whereas B. bacilliformis is the agent of Carrión disease, which corresponds to either an acute intraerythrocytic bacteremic disease (Oroya fever) or a chronic vasculoproliferative disease (verruga peruana). Infection of red blood cells has been well established for B. bacilliformis (Oroya fever) and B. quintana (trench fever and bacteremia in the homeless), whereas B. henselae and B. koehlerae have been seen in erythrocytes of infected cats. B. henselae can cause granulomatous disease, that is, cat-scratch disease, which affects lymph nodes, but can also be responsible for other clinical manifestations or complications, such as endocarditis. Vasculoproliferative diseases include bacillary angiomatosis caused by B. henselae and B. quintana, peliosis hepatis caused by B. henselae, and verruga peruana caused by B. bacilliformis. The immune status of the host plays a critical role in the development of these different forms of the disease. B. henselae usually causes cat-scratch disease (a self-limited disease) in immunocompetent hosts, whereas it is responsible for bacillary angiomatosis in immunocompromised patients. In patients with a previous valvulopathy, any Bartonella infection may lead to endocarditis.

299  BARTONELLA INFECTIONS JEAN-MARC ROLAIN AND DIDIER RAOULT



1967

DEFINITION

Bartonella species belong to the alpha-2 subgroup of Proteobacteria and are closely related to the genera Brucella, Agrobacterium, and Rhizobium. Since 1993, the genus Bartonella has been reorganized by addition of the genera Rochalimaea and Grahamella to the family Bartonellaceae. Currently more than 40 known Bartonella species have been isolated from both animals and humans. These bacteria are considered emerging pathogens that are associated with zoonosis and human infections. Among them, 15 validated species have been implicated in human diseases: B. henselae, B. quintana, B. bacilliformis, B. elizabethae, B. clarridgeiae, B. vinsonii subsp. arupensis, B. vinsonii subsp. berkhoffii, B. alsatica, B. tamiae, B. grahamii, B. washoensis, B. rochalimae, B.

The Pathogen

Bartonella species are small, gram-negative, fastidious, pleomorphic coccobacilli or slightly curved rods (0.5 by 1 to 2 µm). Because of the slow growth of these bacteria and the lack of reproducible biochemical methods for their identification, they are usually identified by molecular methods. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry has emerged as a new technique for species identification and is an accurate and reproducible method for the rapid and inexpensive identification of Bartonella species. The

TABLE 299-1 BARTONELLA SPECIES CAUSING HUMAN DISEASE FIRST CULTIVATION

BARTONELLA SPECIES

MAMMAL

COUNTRY

YEAR OF DESCRIPTION

RESERVOIR HOST/VECTOR

HUMAN DISEASE

B. alsatica

Wild rabbit (Oryctolagus cuniculus)

France

1999

B. ancashensis

Human

Peru

2013

B. bacilliformis

Human

1909

Human/sandfly

Carrión disease, Oroya fever, verruga peruana

B. clarridgeiae

Cat

1996

Cat/cat flea

Cat-scratch disease

B. elizabethae

Endocarditis patient

United States

1993

Rat

Endocarditis, neuroretinitis

B. grahamii

Woodland mammal (Clethrionomys glareolus)

United Kingdom

1995

Rat, insectivore

Neuroretinitis

B. henselae

Cat

1990

Cat/cat flea

Cat-scratch disease, endocarditis, bacillary angiomatosis, bacillary peliosis, Parinaud oculoglandular syndrome, neuroretinitis, osteomyelitis, arthropathy, bacteremia with fever

B. koehlerae

Domestic cat

United States

1999

Cat

Endocarditis

B. mayotimonensis

Endocarditis patient

United States

2009

Unknown

Endocarditis

B. quintana

Human

1920

Human/body louse

Trench fever, endocarditis, bacillary angiomatosis

B. rochalimae

Human

United States

2007

B. tamiae

Human

Thailand

2008

B. vinsonii arupensis

Cattle rancher

United States

1999

Dog, rodent/ticks

Bacteremia with fever

B. vinsonii berkhoffii

Dog

United Kingdom

1998

Dog

Endocarditis

2000

Ground squirrel

Myocarditis

B. washoensis

Rabbit

Endocarditis, lymphadenopathy Verruga peruana

Bacteremia, fever, splenomegaly Febrile illness

CHAPTER 299  BARTONELLA INFECTIONS  

ABSTRACT

Bartonella infections are vector-borne bacteria responsible for a wide spectrum of human acute and chronic diseases. More than 40 different bacterial species from this genus have been isolated from animals and humans. This chapter provides an overview of the nature of these bacteria, the epidemiology of human infections linked to these different bacteria, the clinical manifestations of the different diseases (both acute and chronic diseases including cat scratch disease, Oroya fever and verruga peruana, trench fever, endocarditis, bacillary angiomatosis, peliosis hepatis), the current methods used for their diagnosis, and finally the different therapeutic strategies to treat these diseases.

KEYWORDS

Bartonella cat scratch disease endocarditis trench fever Carrión disease bacillary angiomatosis peliosis hepatis

1967.e3

1968

CHAPTER 299  BARTONELLA INFECTIONS  

bacteria can grow on enriched blood-containing media with a 5% carbon dioxide atmosphere after 5 to 15 days to up to 45 days on primary culture. The optimal growth temperature ranges from 28° C for B. bacilliformis to 35° to 37° C for the other species. Bartonella species can also be cocultured with endothelial cells. Bartonella species are either flagellated or nonflagellated cells. B. bacilliformis uses flagella for binding and deforming into the surface of erythrocytes. Bacteria can either persist in the blood stream of the host as intraerythrocytic parasites or colonize human endothelial cells.  

without antibiotic treatment, it causes death in up to 85% of infected humans by hemolysis or when complicated by opportunistic infections such as salmonellosis. The onset is usually abrupt, with high fever, chills, headache, and anorexia. Patients have intense myalgias and arthralgias, abdominal pain, and jaundice. Complications are frequent, including meningoencephalitis, dyspnea, delirium, and superinfection leading to death. Asymptomatic persistent bacteremia may serve as the reservoir of the organism. Trench fever is transmitted by lice and is the clinical manifestation of B. quintana. Trench fever affected more than 1 million people during World War I; more recently, B. quintana has been recognized in immunocompromised hosts, homeless people, and chronic alcoholics. Clinical manifestations of trench fever may range from asymptomatic infection to severe, life-threatening illness. After an incubation period of 2 to 3 weeks, there is a sudden onset of fever that lasts 1 to 3 days associated with headache, shin pain, and dizziness. Although fatal cases have not been reported, the disease may persist for 4 to 6 weeks and result in prolonged disability. Relapses may occur years later, and in some cases there may be bacteremia with no clinical signs.

EPIDEMIOLOGY

Almost all Bartonella species are vector-borne bacteria (see Table 299-1). Some are limited geographically, such as B. bacilliformis, which is found only in the Andes Mountains in South America at high altitudes, where its principal vector, Lutzomyia verrucarum, is distributed; others have a worldwide distribution, such as B. henselae and B. quintana. Each Bartonella species is highly adapted to its mammalian reservoir, in which bacteria usually cause a long-lasting intraerythrocytic bacteremia that may be asymptomatic. Humans are the hosts and reservoirs for B. bacilliformis and B. quintana. B. quintana is transmitted by the human body louse by inoculation of arthropod feces through broken skin. Cats represent the main reservoir hosts for B. henselae infection; this pathogen is the agent of cat-scratch disease in humans, caused by cat bites or scratches. In the United States, about 12,000 outpatients and 500 inpatients are diagnosed annually.2 B. henselae infection is transmitted from cat to cat by the cat flea. Cat fleas may also be infected by B. quintana. The role of dogs as reservoir hosts has been documented for several species, including B. vinsonii subsp. arupensis, B. vinsonii subsp. berkhoffii, and B. henselae. Wild rabbits are the reservoir hosts for B. alsatica, which is an agent of endocarditis and lymphadenopathy in humans in close contact with rabbits. For other Bartonella species known to cause diseases in humans, their pathogenic role and mode of transmission are not fully understood.

  CAT-SCRATCH DISEASE PATHOBIOLOGY  

Little is known about the pathogenesis of the long-lasting lymphadenopathy in cat-scratch disease. Immunopathogenesis is assumed to play an important role in cat-scratch disease because bacteria have only rarely been isolated from affected lymph nodes. Thus, the disease is usually controlled by the host immune response, and there are few or no viable bacteria when lymph node biopsy specimens are analyzed; they are necrotic by pathologic examination.  

INFECTION OF RED BLOOD CELLS: OROYA FEVER AND TRENCH FEVER  

CLINICAL MANIFESTATIONS

Typical cat-scratch disease is the most common manifestation of infection with B. henselae and usually is manifested as a self-limited regional lymphadenitis. Transmission from cat to human occurs directly by a cat scratch or cat bite or possibly by a cat flea or tick bite. A typical papule or pustule may be seen

PATHOBIOLOGY

In Oroya fever, B. bacilliformis invades up to 80% of erythrocytes and produces their massive lysis, which results in severe hemolytic anemia, the major symptom of the disease. Similarly, trench fever is characterized by intracellular erythrocyte parasitism by B. quintana, with the percentage of infected red blood cells ranging from 0.001 to 0.005% (Fig. 299-1). Bacteria can also be seen extracellularly and in erythroblasts. This intracellular erythrocyte parasitism can presumably preserve the pathogens for efficient transmission by body lice, protect B. quintana from the host immune response, and contribute to decreased antimicrobial efficacy. During bacteremia in the homeless, B. quintana can also be seen in red blood cells.  

CLINICAL MANIFESTATIONS

The main clinical manifestations of infection by Bartonella species are summarized in Table 299-2. Oroya fever is the acute or hemolytic phase of Carrión disease, caused by B. bacilliformis3; it usually develops 3 to 12 weeks after inoculation. Oroya fever results from the massive invasion of erythrocytes by B. bacilliformis, and

FIGURE 299-1.  Section of human red blood cell infected with Bartonella quintana as viewed by confocal microscopy.

TABLE 299-2 CLINICAL MANIFESTATIONS ASSOCIATED WITH BARTONELLA SPECIES CLINICAL MANIFESTATION

B. BACILLIFORMIS

B. QUINTANA

Intraerythrocytic bacteremia

+

+

Chronic bacteremia

+

Infective endocarditis Verruga peruana

B. HENSELAE

+

+

+

+

+

+

B. ALSATICA

OTHERS

+

+

+

+

+

+

Bacillary angiomatosis

+

Peliosis hepatis Lymphadenopathy

+

+

Meningoencephalitis Uveitis-retinitis CSD = cat-scratch disease; SENLAT = scalp eschar and neck lymphadenopathy.

+ (CSD) +

SENLAT with skin lesion +

+

+

CHAPTER 299  BARTONELLA INFECTIONS  

3 to 10 days after the scratch or bite at the site of inoculation and may last for 1 to 3 weeks. The subsequent lymphadenopathy is localized mainly to the axillary, cervical, or submaxillary nodes that drain the area where the cat scratch occurred. The enlarged lymph node is often painful and tender. Lymphadenopathy sometimes lasts for months, and in a few cases it can persist for as long as 1 to 2 years. In some cases, the lymph node may suppurate if it is not drained. Most patients are not febrile during the course of typical catscratch disease. Systemic or severe disease may occur in about 5 to 14% of patients, with most of them suffering severe systemic symptoms due to disseminated infection.4 After exposure to B. henselae, patients may develop bacteremia with or without clinical signs of typical cat-scratch disease, and in patients with valvular lesions, this may result in infective endocarditis.5,6 Thus, cat-scratch disease represents the primary infection of B. henselae, and endocarditis may follow in patients with heart valve lesions. It is a potential cause of “culture-negative endocarditis.” Approximately 10% of patients with cat-scratch disease have atypical clinical manifestations, including prolonged fever (>2 weeks), malaise, neuroretinitis, encephalitis, erythema nodosum, hepatitis, fatigue, weight loss, and splenomegaly. A recent clinical study showed that musculoskeletal manifestations (myalgia, arthritis, arthralgia, tendinitis, osteomyelitis, neuralgia) were present in more than 10% of patients with cat-scratch disease, demonstrating that these clinical manifestations are not as rare as might be expected from the cases reported in the past. In this series of 913 patients, myalgia and arthropathy were the most common manifestations, with an incidence of 5.8% and 5.5%, respectively. Moreover, these manifestations occurred primarily in adults whose ages ranged from 20 to 59 years. Myalgia had a mean duration of 4 weeks and was often severe. Arthropathy had a mean duration of 5.5 weeks, was more common in female patients older than 20 years, affected large and medium joints (half of those involved being weight-bearing joints), and was associated with symmetrical erythema nodosum early in the course of cat-scratch disease. These musculoskeletal manifestations are often severe and may evolve into chronic forms that persist for more than a year. Tendinitis, neuralgia, and osteomyelitis are less common, with incidences lower than 1%.

Ocular and Neurologic Manifestations

Parinaud oculoglandular syndrome is a self-limited conjunctivitis associated with preauricular lymphadenopathy. Other atypical manifestations include neurologic syndromes (meningoencephalitis, meningitis, neuroretinitis). Encephalopathy may occur in 2 to 4% of cat-scratch disease patients, mainly7 adolescents and adults. Patients usually have persistent headaches with or without fever and may develop seizures. Acute neurologic disorders range from self-limited nuchal rigidity to pupillary dilation or aphasia and hemiplegia; they may last for several weeks to months. Neuroretinitis has been associated with cat-scratch disease in patients experiencing a sudden unilateral loss of visual acuity. The most common picture remains papilledema associated with macular exudates causing stellar retinitis. A few reports have now established that B. henselae can be responsible for uveitis, along with B. grahamii and B. quintana. Patients present with either nongranulomatous or granulomatous uveitis.8,9 Finally, tick-borne B. henselae infection has been described, including scalp eschar and neck lymphadenopathy after tick bites in three patients during the colder months in France. B. henselae was detected by molecular tools both in skin biopsy specimens (cervical and occipital) and in a Dermacentor marginatus tick removed from the scalp of one patient. All three patients had asthenia, but none had alopecia.



CLINICAL MANIFESTATIONS

The most commonly identified agents of Bartonella endocarditis are B. quintana, followed by B. henselae and other Bartonella species. Patients appear to have chronic, blood culture–negative endocarditis, usually with fever (90%). Echocardiography reveals vegetations (in 90%).10 Infections with B. henselae are epidemiologically linked to close contact with cats or cat fleas and previous valvular heart disease, whereas B. quintana endocarditis is frequently described in homeless and alcoholic patients with body lice infection and can be observed in patients without previous valve lesions. The onset is usually subacute, with some patients being afebrile at the time of admission. About half the patients have embolic phenomena. Interestingly, there is a north (Europe) to south (North Africa) gradient for the proportion of Bartonella endocarditis in humans; thus Bartonella is apparently a common cause of endocarditis in North Africa. Sporadic cases of endocarditis have also been associated with B. koehlerae, B. vinsonii subsp. berkhoffii, B. vinsonii subsp. arupensis, B. elizabethae, B. alsatica, and “Candidatus Bartonella mayotimonensis.”

VASCULOPROLIFERATIVE DISEASE: VERRUGA PERUANA, BACILLARY ANGIOMATOSIS, AND PELIOSIS HEPATIS  

PATHOBIOLOGY

Bartonella species have the ability to cause vasculoproliferative lesions through a process of pathologic angiogenesis resulting in the formation of new capillaries from preexisting ones.11 These typical vasoproliferations can be expressed as skin lesions called bacillary angiomatosis, which are caused by B. quintana and B. henselae; there is also a cystic form in the liver and spleen called peliosis hepatis, which is caused only by B. henselae. Skin lesions are similar to those reported for verruga peruana, the chronic form of Carrión disease. Bacillary angiomatosis is a neovascular proliferation that has been reported most commonly in AIDS patients and involves the skin (Fig. 299-2) and lymph nodes; it occurs less frequently in patients with other causes of immunosuppression and only exceptionally in immunocompetent patients. In bacillary angiomatosis, lesions comprise proliferating endothelial cells, bacteria, and mixed infiltrates of macrophages/monocytes and polymorphonuclear neutrophils, leading to chronic inflammation. Bacteria are clustered as aggregates both surrounding and within endothelial cells, indicating that the vascular endothelium represents a target tissue for intracellular and extracellular colonization in vivo. On histologic evaluation, bacillary angiomatosis is a lobular proliferation of small blood vessels containing endothelial cells and bacteria, usually seen in clusters when stained with Warthin-Starry. As in bacillary angiomatosis, lesions of verruga peruana are characterized by lobular proliferations and atypical endothelial cells forming both relatively solid sheets and small, wellformed vessels with patent lumens. Lesions are typically infiltrated, indicating a chronic inflammatory process.  

CLINICAL MANIFESTATIONS

As already noted, bacillary angiomatosis is seen most often in AIDS patients. Cutaneous lesions often arise in crops and can be subcutaneous or dermal nodules with red or purple papules millimeters to centimeters in diameter. When cutaneous lesions are absent, the diagnosis is often difficult and delayed because signs of visceral involvement are usually nonspecific. The potentially systemic nature of bacillary angiomatosis is reflected by the involvement of brain, bone, lymph node, bone marrow, skeletal muscle, conjunctiva, and

 ENDOCARDITIS PATHOBIOLOGY  

B. quintana, B. henselae, B. alsatica, B. vinsonii subsp. berkhoffii, B. elizabethae, and “Candidatus Bartonella mayotimonensis” are common causes of blood culture–negative endocarditis, whereas B. vinsonii subsp. arupensis has been detected in a patient with fever and bacteremia; B. washoensis has been identified in one patient with myocarditis; B. rochalimae was reported in a patient with fever, bacteremia, and splenomegaly; and B. tamiae has been isolated in a patient with a febrile illness. Patients with endocarditis usually have preexisting heart valve disease that promotes the development of infective endocarditis and, in some cases, a definite risk factor for infection specifically with Bartonella. Endocarditis caused by Bartonella species exhibits slight inflammation, with a few inflammatory mononuclear cells and small vegetations; the bacteria are seen extracellularly in dense immunopositive clusters that are mainly included in vegetations and in neutrophil and macrophage cytoplasm.

1969

FIGURE 299-2.  Skin lesion of bacillary angiomatosis.

1970

CHAPTER 299  BARTONELLA INFECTIONS  

TABLE 299-3  LABORATORY METHODS FOR THE DIAGNOSIS OF BARTONELLA INFECTIONS CLINICAL MANIFESTATION

SEROLOGY

CULTURE

MOLECULAR METHODS

IMMUNOHISTOCHEMISTRY (WARTHIN-STARRY STAIN)

MICROSCOPY FOR INTRAERYTHROCYTIC ORGANISM (GIEMSA OR IF)

Oroya fever



+

+



+++

Verruga peruana





+

+



+/−

+

+



+

Chronic bacteremia

+/−

+++

+





Infective endocarditis

+++

+++

+++

+++



Bacillary angiomatosis

+/−

++

+++

+++



Peliosis hepatis

+/−

++

+++

+++



+



+++

+/−

+/−

Trench fever

Cat-scratch disease SENLAT with skin lesion

+

+

+++

+



Meningoencephalitis

++



++





Uveitis-retinitis

++



++

+



The usefulness of these techniques varies according to the disease being tested. IF = immunofluorescence; SENLAT = scalp eschar and neck lymphadenopathy.

mucosal surfaces of the gastrointestinal and respiratory tracts. Peliosis hepatis affects solid internal organs, primarily the liver, with reticuloendothelial elements; in the liver, it is defined as a vascular proliferation of sinusoidal hepatic capillaries resulting in blood-filled spaces. The spleen, abdominal lymph nodes, and bone marrow may also be involved. Following acute Oroya fever, patients usually develop angioproliferative cutaneous tumors called verruga peruana after a latent period ranging from weeks to months. The infection is characterized by benign cutaneous vascular lesions typically consisting of round papules that are frequently pruritic and bleeding. The infection is accompanied by malaise and arthralgias. Skin lesions may change over time from miliary to nodular subcutaneous lesions to large mulaire lesions. These large lesions are often engorged with blood and prone to ulceration and bleeding. This eruptive phase clinically resembles Kaposi sarcoma or bacillary angiomatosis. However, it has a low morbidity, and there are no reports of mortality.  

DIAGNOSIS OF BARTONELLA INFECTION

Methods used for the diagnosis of Bartonella infection include serology, microscopy, culture, molecular amplification of Bartonella species genes, direct immunofluorescence, and immunohistochemistry. The usefulness of these techniques may vary according to the disease involved (Table 299-3).

Serologic Tests

Serology remains the most widely used method for the diagnosis of cat-scratch disease and Bartonella endocarditis because culture and isolation are difficult and time-consuming, and molecular methods are not available in all laboratories. There are currently two classic serologic methods for the diagnosis of Bartonella infections: enzyme-linked immunosorbent assay and immunofluorescence assay. By immunofluorescence assay, an immunoglobulin G titer of 1 : 64 or greater should be considered positive for cat-scratch disease, whereas patients with endocarditis usually have higher antibody titers (≥1 : 800). In homeless patients, bacteremia has been associated with serologic tests that were positive for B. quintana. However, reported sensitivities of immunofluorescence assay vary considerably, from nearly 100% to less than 30%, depending on the nature of the antigens used and the selected patients.12 Moreover, owing to cross-reactive antibodies between Bartonella species, the diagnosis of Bartonella infection at the species level is usually not possible. More sophisticated methods should be used, especially Western blot with cross-adsorption analysis. Western blot is also useful for the differential diagnosis of endocarditis because the profile obtained for endocarditis is specific compared with that for cat-scratch disease or chronic bacteremia.

Microscopy, Immunofluorescence, and Immunohistochemistry

The diagnosis of Oroya fever is based on examination of a peripheral blood smear stained by Giemsa; the percentage of infected red blood cells is sufficiently high so that bacteria are visible. Similarly, B. quintana can be seen within red blood cells with use of specific monoclonal antibody by immunofluorescence and confocal microscopy (see Fig. 299-1). Microscopic

examination after Warthin-Starry silver staining or immunohistochemistry of a cardiac valve or skin biopsy specimen is also useful for the detection of Bartonella organisms in patients with endocarditis and bacillary angiomatosis.

Culture

Bartonella species can be recovered from the blood of bacteremic patients as well as from cardiac valves, skin and liver biopsy specimens, and, rarely, lymph nodes. Bacteria can be isolated directly from these specimens after plating onto blood agar solid media, blood culture in broth, and cocultivation in endothelial cells. Cultures are usually positive after 2 weeks of incubation, but up to 45 days may be necessary for primary isolation.

Molecular Detection Methods

Direct detection and final identification of Bartonella species from blood and tissue specimens, including lymph node, cardiac valve, skin, and liver, can be achieved by polymerase chain reaction (PCR) amplification and sequencing of various housekeeping genes.13,14 PCR can also detect Bartonella at the species level from lymph nodes in cat-scratch disease, directly from cardiac valves in patients with endocarditis, and in patients with bacillary angiomatosis or peliosis hepatis. In one study, all cases of blood culture–negative Bartonella were diagnosed by real-time PCR.15

Differential Diagnosis

The differential diagnosis of bacillary angiomatosis is Kaposi sarcoma, but visualization of bacteria in the tissue specimen can help distinguish between these two entities. Trench fever may be confused mainly with typhus group rickettsiosis, relapsing fever, and malaria. Cat-scratch disease may be confused with tularemia, pyogenic lymphadenitis, mycobacterial infection, and neoplasia.

TREATMENT  Trench Fever

Before the advent of antibiotics, acetylsalicylic acid was the most effective drug for the pain caused by trench fever. However, a randomized clinical trial in homeless persons with episodes of bacteremia demonstrated that the combination of gentamicin and doxycycline is more effective in stopping bacteremia compared with no treatment or the use of β-lactams or doxycycline alone. Because the intraerythrocytic presence of B. quintana decreases antimicrobial efficacy, the duration of treatment is critical in trench fever. Patients with B. quintana bacteremia should be treated with gentamicin (3 mg/kg body weight/day IV for 14 days), in combination with doxycycline (200 mg/day PO) for 28 days.

Oroya Fever and Verruga Peruana

In Oroya fever, treatment with penicillin, streptomycin, fluoroquinolones, tetracycline, or erythromycin produces rapid defervescence and disappearance of the organisms from the blood, usually within 24 hours. As an alternative treatment, chloramphenicol can be used alone or in combination with a

β-lactam. Treatment with chloramphenicol may also have the advantage of covering commonly associated Salmonella species. Patients with Oroya fever should be treated with chloramphenicol (500 mg PO four times a day) for 14 days in combination with another antibiotic (especially a β-lactam compound). Streptomycin (15 to 20 mg/kg/day for 10 days) was the traditional treatment for verruga peruana. However, its use is problematic, especially in children, and rifampin has become the drug of choice for the treatment of eruptive-phase bartonellosis. Failures of rifampin treatment have been reported in verruga peruana. Finally, ciprofloxacin (500 mg PO twice daily for 7 to 10 days) has been used successfully for the treatment of multiple eruptive-phase lesions in adults and has been proposed as an appropriate alternative. Doxycycline in association with gentamicin may be used to treat the eruptive phase of Carrión disease.

Cat-Scratch Disease

Typical cat-scratch disease is a self-limited illness that resolves within 2 to 6 months and usually does not respond to therapy because the bacteria within necrotic lymph nodes are not alive. In cases of long-lasting lymphadenopathy, patients should be reassured that it is benign and will probably subside spontaneously within 2 to 4 months. Management consists of analgesics for pain and prudent follow-up. However, azithromycin (500 mg PO on day 1 and 250 mg PO on days 2 to 5 as single daily doses) is an alternative for patients with large, bulky lymphadenopathy. If lymphadenopathy does not resolve, lymph nodes can be removed surgically. For atypical presentations of cat-scratch disease, there are no data regarding the benefit of specific antimicrobial therapy for immunocompetent patients. For neuroretinitis, the combination of doxycycline (100 mg PO or IV twice daily) with rifampin (300 mg PO twice daily) seems to promote disease resolution, to improve visual acuity, to reduce optic disc edema, and to decrease disease duration.

Endocarditis

Patients with Bartonella endocarditis have a higher death rate and undergo valvular surgery more frequently than patients with endocarditis caused by many other pathogens. Patients with suspected (but culture-negative) Bartonella endocarditis or proved B. quintana endocarditis should be treated with oral doxycycline 100 mg twice a day for 6 weeks in combination with gentamicin 3 mg/kg/day in one intravenous daily dose for 14 days.16 The American Heart Association consensus on treating infective endocarditis is ceftriaxone plus gentamicin, with or without doxycycline, when Bartonella is suspected, and doxycycline plus gentamicin when Bartonella endocarditis is confirmed. However, there is direct evidence that patients receiving an aminoglycoside are more likely to fully recover, and those treated with aminoglycosides for at least 14 days are more likely to survive than those receiving a shorter duration of therapy. In the absence of any prospective study for the treatment of Bartonella endocarditis, the same regimen as for B. henselae and B. quintana should be used for endocarditis when another Bartonella species has been identified as the causative agent.

Bacillary Angiomatosis and Peliosis Hepatis

Without appropriate therapy, infection spreads systemically and can involve virtually any organ, and the outcome is sometimes fatal. Thus, antibiotic treatment is warranted in all cases of Bartonella-associated vasculoproliferative disease. On the basis of empirical clinical reports, erythromycin remains the treatment of choice and has been used successfully to treat many patients with bacillary angiomatosis. The response to treatment in bacillary angiomatosis can be dramatic in immunocompromised patients, with resolution of palpable subcutaneous lesions within hours. Erythromycin also has an antiangiogenic effect on microvascular endothelial cells that could explain this quick disappearance of lesions. Erythromycin (500 mg four times daily) for 3 months is first-line therapy. Doxycycline (100 mg PO or IV twice daily) is currently proposed as an appropriate alternative. In patients with serious infections, erythromycin or doxycycline can be used in combination with rifampin (300 mg PO twice daily). The duration of therapy is critical. We recommend that treatment be given for at least 3 months for bacillary angiomatosis and 4 months for peliosis hepatis. Peliosis hepatis responds slowly, and hepatic lesions continue to improve after several months of treatment, whereas cutaneous bacillary angiomatosis demonstrates improvement after 4 to 7 days of treatment and resolves after about 1 month of treatment. Relapses of peliosis hepatis and bacillary angiomatosis lesions after antibiotic treatment have been reported frequently, especially in immunocompromised patients with a shorter duration of therapy. Patients who relapse after the recommended treatment should probably be re-treated for 4 to 6 months with erythromycin (500 mg PO four times daily) or doxycycline (100 mg PO twice daily), and those with repeated relapses should receive antibiotic therapy as long as they are immunocompromised.



PREVENTION

Because arthropods play an important role in the transmission of feline Bartonella species to humans, rigorous arthropod control should be recommended by health care workers, particularly when advising immunocompromised

individuals on the risks related to pet ownership. Cat fleas live on both cats and dogs and are best controlled by fumigating areas where these animals live.  

PROGNOSIS

Mortality in those with Oroya fever was as high as 50% before the antibiotic era but is now limited by the use of antibiotics. Trench fever should be treated with antibiotics to avoid more severe disease, especially in patients with valvulopathy who are at risk for development of endocarditis. Cat-scratch disease usually resolves spontaneously without any treatment in immunocompetent patients. In the case of complications or in immunocompromised patients, antibiotic therapy should be given, and patients usually respond well to treatment. Finally, for vasculoproliferative diseases, antibiotics are usually effective if they are given for a long time, but cutaneous lesions in verruga peruana may benefit from surgery. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 299  BARTONELLA INFECTIONS  

GENERAL REFERENCES 1. Ruiz J. Bartonella quintana, past, present, and future of the scourge of World War I. APMIS. 2018;126:831-837. 2. Nelson CA, Moore AR, Perea AE, et al. Cat scratch disease: U.S. clinicians’ experience and knowledge. Zoonoses Public Health. 2018;65:67-73. 3. Gomes C, Ruiz J. Carrion’s disease: the sound of silence. Clin Microbiol Rev. 2017;31:1-51. 4. Barros S, De Andrade GC, Cavalcanti C, Nascimento H. Cat scratch disease: not a benign condition. Ocul Immunol Inflamm. 2018;26:1115-1122. 5. Okaru U, Addisu A, Casanas B, et al. Bartonella species, an emerging cause of blood-culture-negative endocarditis. Clin Microbiol Rev. 2017;30:709-746. 6. Nakasu A, Ishimine T, Yasumoto H, et al. Infective endocarditis associated with Bartonella henselae: a case series. IDCases. 2018;12:127-129. 7. Canneti B, Cabo-López I, Puy-Núñez A, et al. Neurological presentations of Bartonella henselae infection. Neurol Sci. 2019;40:261-268. 8. Mabra D, Yeh S, Shantha JG. Ocular manifestations of bartonellosis. Curr Opin Ophthalmol. 2018;29:582-587.

1971.e1

9. Amer R, Tugal-Tutkun I. Ophthalmic manifestations of bartonella infection. Curr Opin Ophthalmol. 2017;28:607-612. 10. Pachirat O, Prathanee S, Watt G. Echocardiographic features in Bartonella endocarditis: a case series. Cardiol Res. 2018;9:116-119. 11. Urbanowicz M, Kutzner H, Riveiro-Falkenbach E, et al. Infectious angiogenesis—different pathways, the same goal. Am J Dermatopathol. 2016;38:793-801. 12. Allizond V, Costa C, Sidoti F, et al. Serological and molecular detection of Bartonella henselae in specimens from patients with suspected cat scratch disease in Italy: a comparative study. PLoS ONE. 2019;14:1-11. 13. Hobson C, Le Brun C, Beaurelle C, et al. Detection of Bartonella in cat scratch disease using a singlestep PCR assay kit. J Med Microbiol. 2017;66:1596-1601. 14. Yanagihara M, Tsuneoka H, Tanimoto A, et al. Bartonella henselae DNA in seronegative patients with cat-scratch disease. Emerg Infect Dis. 2018;24:924-925. 15. Edouard S, Nabet C, Lepidi H, et al. Bartonella, a common cause of endocarditis: a report on 106 cases and review. J Clin Microbiol. 2015;53:824-829. 16. Okaro U, Addisu A, Casanas B, et al. Bartonella species, an emerging cause of blood-culture-negative endocarditis. Clin Microbiol Rev. 2017;30:709-746.

1971.e2

CHAPTER 299  BARTONELLA INFECTIONS  

REVIEW QUESTIONS 1. Which of the following is the main vector of Bartonella quintana infections in humans? A . Mites B. Lice C. Ticks D. Bed bugs E. Fleas Answer: B  The human body louse is the main vector (see Table 315-1), but B. quintana can be detected in fleas and ticks.

4. Which Bartonella species is the causative agent of peliosis hepatis? A . B. quintana B. B. henselae C. B. bacilliformis D. B. grahamii E. B. alsatica Answer: B  Peliosis hepatis is characterized by the development of cystic, blood-filled cavities throughout the liver parenchyma and can be caused by a variety of drugs, malignant neoplasms, post–renal transplantation, and different infections including human immunodeficiency virus infection, tuberculosis, and Bartonella henselae.

2. Which treatment is the most appropriate recommendation for Bartonella endocarditis? A . Erythromycin B. Doxycycline plus rifampin C. Doxycycline D. Doxycycline plus gentamicin E. Streptomycin Answer: D  Bartonella endocarditis should be treated with doxycycline for 6 weeks plus gentamicin for 14 days. (See Treatment section.)

5. Which treatment is the most appropriate recommendation for bacillary angiomatosis? A . Erythromycin B. Doxycycline plus rifampin C. Doxycycline D. Doxycycline plus gentamicin E. Streptomycin Answer: A  Erythromycin (500 mg four times daily) for 3 months is first-line therapy. (See Treatment section.)

3. Which laboratory method is the most appropriate for the diagnosis of cat-scratch disease (CSD)? A . Real-time polymerase chain reaction B. Serology C. Gram staining D. Cell culture E. Immunohistochemistry Answer: A  Serology remains a widely used method where molecular methods are not available for the diagnosis of CSD and Bartonella endocarditis because culture and isolation are difficult and time-consuming. However, the reported sensitivities of immunofluorescence assays vary considerably, depending on the nature of the antigen used and the selected patients. Furthermore, because of cross-reactivity of antibodies between Bartonella species, serologic methods usually cannot make the diagnosis of Bartonella infection at the species level. Direct detection and definitive Bartonella species identification are possible from blood and tissue (e.g., lymph node) specimens. See Table 315-3.

CHAPTER 300  Granuloma Inguinale (Donovanosis)  

1971

300  GRANULOMA INGUINALE (DONOVANOSIS) KHALIL G. GHANEM AND EDWARD W. HOOK, III



DEFINITION

Granuloma inguinale, also known as donovanosis, is a slowly progressive ulcerative disease that involves principally the skin and subcutaneous tissues of the genital, inguinal, and anal regions.

The Pathogen

The causative organism is Klebsiella granulomatis (formerly Calymmatobacterium granulomatis), a gram-negative facultative intracellular parasite. The organism is challenging to cultivate but can sometimes be grown in yolk sacs, and successful cell culture has been reported from South Africa and Australia. Successful culture, in turn, has permitted the development of polymerase chain reaction assays, currently for research purposes.  

EPIDEMIOLOGY

The organism is primarily transmitted sexually, but it can probably be transmitted by nonsexual contact as well. Transmission efficiency is relatively low, and multiple sexual contacts with an infected partner seem to be necessary for the transmission of infection. The disease is rarely encountered in the United States. Although still relatively rare, sporadic cases occur in India, Papua New Guinea, the Caribbean, Brazil, southern Africa, and parts of Australia.  

CLINICAL MANIFESTATIONS

After an incubation period of up to 50 days, the initial lesion usually appears as a subcutaneous nodule that erodes through the surface and develops into a beefy, elevated granulomatous lesion (Fig. 300-1). The lesion is painless but tends to bleed easily, and it is not associated with systemic symptoms. Lesion exudate is often described as foul smelling. Secondary bacterial infection may cause a necrotic, painful, ulcerative lesion that may be rapidly destructive. A cicatricial form may also occur, with a depigmented elevated area of keloidlike scar containing scattered islands of granulomatous tissue. About 90% of lesions occur on the genitals and are commonly associated with pseudobuboes; these swellings usually are not due to involvement of the inguinal lymph nodes but rather are due to granulomatous involvement of subcutaneous tissue. Metastatic infection of bones or viscera (e.g., liver) is occasionally seen. Clinical experience suggests that secondary carcinomas (typically squamous cell carcinoma in patients with longstanding, chronic infection) may be a rare complication of granuloma inguinale.1

CHAPTER 300  Granuloma Inguinale (Donovanosis)  

ABSTRACT

Granuloma inguinale, also known as donovanosis, is a slowly progressive ulcerative disease that involves principally the skin and subcutaneous tissues of the genital, inguinal, and anal regions. The causative organism is Klebsiella granulomatis (formerly Calymmatobacterium granulomatis), a gram-negative facultative intracellular bacterium.

KEYWORDS

STD sexually transmitted infection donovanosis Klebsiella granulomatis Calymmatobacterium granulomatis

1971.e3

FIGURE 300-1.  Typical primary lesion of granuloma inguinale. (Reproduced with permission from Herpes-Coldsores.com. http://www.herpes-coldsores.com/std/lymphogranuloma-pictures.htm.)



DIAGNOSIS

The diagnosis is made by demonstrating intracellular Donovan bodies in histiocytes or other mononuclear cells from lesion scrapings or biopsy samples. Wright stain and Giemsa stain of fresh impression smears or unfixed biopsy specimens usually demonstrate the bacilli relatively easily, although multiple biopsies may be necessary in chronic cases. Histologic examination of biopsy specimens shows plasma cells with some infiltration by polymorphonuclear leukocytes but no giant cells. In infected lesions, K. granulomatis is found primarily in histiocytes or other mononuclear cells. Cell culture and polymerase chain reaction methods are currently primarily research tools. No serologic test is clinically available.

Differential Diagnosis

The differential diagnosis includes squamous cell carcinoma, chancroid (Chapter 285), lymphogranuloma venereum (Chapter 302), syphilis (Chapter 303), and other ulcerative granulomatous diseases.2 In the absence of therapy, patients may not present until lesions have been present for months, long after the lesions of syphilis and other ulcerative sexually transmitted diseases have resolved. Chancroid is usually differentiated by its irregular undermined borders, which are not seen in typical cases of granuloma inguinale. Dark-field examination and serologic tests should help distinguish syphilis. Biopsy of lesions may be necessary to distinguish granuloma inguinale from certain tumors.3

TREATMENT  There are no high quality clinical trials of therapy for granuloma inguinale. Recommended treatment consists of azithromycin administered either as 1.0 g weekly or 500 mg daily or doxycycline 100 mg twice daily.4 One double-strength trimethoprim-sulfamethoxazole tablet twice daily or erythromycin base 500 mg four times daily is recommended as alternative therapy. An aminoglycoside (e.g., gentamicin 1 mg/kg intravenously every 8 hours) may be added if these regimens do not result in clinical improvement within a few days. Treatment should be administered for at least 3 weeks and until lesions are completely healed.5 Patients should be monitored for at least several weeks after treatment is discontinued because of the possibility of relapse. Although the risk of communicability appears to be low, sexual contacts should also be examined; at present, treatment of contacts is not indicated in the absence of clinically evident disease.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 300  Granuloma Inguinale (Donovanosis)  

GENERAL REFERENCES 1. Arora AK, Kumaran MS, Narang T, et al. Donovanosis and squamous cell carcinoma: the relationship conundrum! Int J STD AIDS. 2017;28:411-414. 2. Pfennig CL. Sexually transmitted diseases in the emergency department. Emerg Med Clin North Am. 2019;37:165-192.

1972.e1

3. O’Farrell N, Hoosen A, Kingston M. 2018 UK national guidelines for the management of donovanosis. Int J STD AIDS. 2018;29:946-948. 4. O’Farrell N, Moi H. 2016 European guideline on donovanosis. Int J STD AIDS. 2016;27:605-607. 5. Copeland NK, Decker CF. Other sexually transmitted diseases chancroid and donovanosis. Dis Mon. 2016;62:306-311.

1972.e2

CHAPTER 300  Granuloma Inguinale (Donovanosis)  

REVIEW QUESTION 1. Which of the following statements regarding granuloma inguinale is correct? A . It is transmitted efficiently and only by sexual contact. B. The initial lesion is typically painful and associated with fever. C. Lesions are associated with buboes. D. Diagnosis requires demonstration of intracellular Donovan bodies. E. Diagnosis requires serologic confirmation. Answer: D  No serologic test is clinically available for granuloma inguinale, and definitive diagnosis depends on demonstration of Donovan bodies in mononuclear cells from lesion scrapings or biopsy samples. Although it is primarily transmitted through sexual contact, the organism can probably be transmitted by nonsexual contact as well; transmission efficiency is low. The lesion is painless, and it is not accompanied by systemic symptoms. It is associated with a so-called pseudobubo, a swelling in the inguinal area caused by granulomatous infiltration of subcutaneous tissue—not actual inguinal lymph node involvement that would constitute a bubo.

1972

CHAPTER 301  MYCOPLASMA INFECTIONS  

301  MYCOPLASMA INFECTIONS STEPHEN G. BAUM AND DAVID L. GOLDMAN   MYCOPLASMA PNEUMONIAE  

DEFINITION

Mycoplasma organisms of the class Mollicutes are ubiquitous as pathogens and colonizing agents in the plant, animal, and insect kingdoms. They represent the smallest known free-living forms, but because they have fastidious growth requirements, they are difficult to culture. The presence of several species of Mycoplasma as commensals in animals and on human oral and genital mucosa frequently results in contamination of cell cultures. Such contamination led to the false implication of mycoplasmas as causative agents in many human diseases, both trivial and life-threatening. Of the human diseases that have proven to be due to mycoplasmas, pneumonia caused by Mycoplasma pneumoniae is by far the most clinically important.1 This infection constitutes a significant proportion of cases previously classified as atypical pneumonia, a nonspecific term for patchy pneumonias that generally do not respond to β-lactam antibiotics and have etiologic agents that are not easily cultured or visible on Gram stain. The term atypical pneumonia persists despite our increasing ability to identify specific etiologic agents, such as viruses, Legionella, and Chlamydophila.

The Pathogen

Mycoplasmas are short rods (10 × 200 nm) that have no cell wall and are bounded by a sterol-containing membrane; thus, they are unaffected by cell wall–inhibiting antimicrobials such as β-lactams. In tissue culture, mycoplasmas are intracellular; but in vivo, infection is primarily extracellular and affects epithelial cells and their organelles, such as cilia. Attachment to respiratory epithelium is by way of terminal adhesin proteins in specialized tip organelles.  

EPIDEMIOLOGY

M. pneumoniae infection is spread person to person by respiratory droplets produced by coughing. Relatively close association with the index case appears to be required. The disease is usually introduced into families by a young child; in some studies, most of the infected adults were the parents of young children. As opposed to most viral respiratory infections, which are manifested 1 to 3 days after infection, Mycoplasma has an incubation period of 2 to 3 weeks. Therefore, a careful history showing several weeks between cases within a family may be an important clue to the mycoplasmal etiology. Organisms can be cultured from the sputum of infected individuals for weeks to months after clinically efficacious treatment. Most cases of Mycoplasma respiratory infection occur singly or as family outbreaks. However, in closed populations, such as military recruit camps and boarding schools, Mycoplasma can cause mini-epidemics and may be responsible for 25 to 75% of cases of pneumonia in such settings. Serologically based epidemiologic studies have documented the high incidence of Mycoplasma respiratory infection throughout the world. In the United States, it is estimated that each year at least one case of Mycoplasma pneumonia occurs for every 1000 persons, or more than 2 million cases annually. The incidence of Mycoplasma nonpneumonic respiratory infection may be 10 to 20 times greater. The highest attack rates are in individuals 5 to 20 years old, but M. pneumoniae infection can occur at any age and may cause particularly severe disease in neonates. As opposed to viral respiratory infections that peak in winter in temperate climates, a few studies have reported a peak incidence of M. pneumoniae outbreaks in the fall. Most surveys, however, show little or no seasonal predominance in sporadic cases. Despite this, recurrent epidemic 4-year cycles have been described in several countries. There is an age-related relationship of upper versus lower respiratory tract infection caused by M. pneumoniae. In children younger than 3 years, primarily upper respiratory tract infection develops, whereas in those 5 to 20 years old, bronchitis and pneumonia tend to occur. In older adults, pneumonia predominates.  

PATHOBIOLOGY

Because of the low fatality of most Mycoplasma infections, there is little human pathologic material. Inoculation onto animal tracheal organ cultures is

CHAPTER 301  MYCOPLASMA INFECTIONS  

ABSTRACT

Mycoplasma pneumoniae is a leading cause of community-acquired respiratory infections in children and young adults. It can cause disease sporadically and as outbreaks in closed communities such as boarding schools and military recruit camps. Pneumonia secondary to this organism is remarkable for a paucity of auscultatory chest findings, and radiographic findings are nonpathognomonic. A host of extrapulmonary complications have been reported, but many of these may be coincidental because the underlying disease is very common. Diagnosis is usually made on clinical grounds, but new commercially available nucleic acid–based assays provide excellent sensitivity and specificity. Disease is usually self-limited and rarely fatal. Because of the absence of a cell wall, β-lactam antibiotics are ineffective. Treatment with either macrolides or doxycycline shortens the clinical course. Other mycoplasmas causing human disease are also briefly discussed.

KEYWORDS

Mycoplasma pneumoniae pneumonia community-acquired pneumonia macrolides Mycoplasma hominis urethritis

1972.e3

CHAPTER 301  MYCOPLASMA INFECTIONS  

1973

Headache and malaise

Temp.

104 102 100

With proper Rx

Cough

Chest soreness

Chest radiograph

Cold agglutinins

+

Rapid test 1:8

1:32

1:64

+

+

+

+ + Remains even with proper Rx

+



+

+

+

+

+

5

10

15

20

25

40

1:256

Culture ELISA and complement fixation tests

1:16

Days of Illness FIGURE 301-1.  Major clinical and laboratory manifestations of mycoplasmal pneumonia. ELISA = enzyme-linked immunosorbent assay; Rx = treatment; Temp. = temperature. (Modified from Baum SG. Mycoplasmal infections. In: Wyngaarden JB, Smith LH Jr, eds. Cecil Textbook of Medicine.17th ed. Philadelphia: Saunders; 1985:1506.)

followed by ciliary damage and desquamation of surface epithelium. This latter effect is probably responsible for the hacking cough in Mycoplasma respiratory infection. Several characteristics of M. pneumoniae probably play a direct role in the respiratory pathogenicity of this organism. The first is the affinity of M. pneumoniae for respiratory epithelial cells. Attachment appears to be between a terminal organelle at one end of the filamentous organism and a sialated glycoprotein (I-FI) on the surface of both respiratory epithelium and erythrocytes that acts as a receptor. M. pneumoniae attaches to ciliated epithelial cells at the base of cilia and appears to produce most of its physiologic and cytolytic changes while remaining extracellular. Hydrogen peroxide produced by M. pneumoniae (the only human mycoplasma to do so) may be responsible for some in vivo cell damage, as it is for the hemolysis seen when the organisms are grown on blood agar plates. Mycoplasma infection activates T and B cells and induces many pro-inflammatory and anti-inflammatory cytokines and chemokines that may play a role in inflammation-related cell destruction. M. pneumoniae also secretes a community-acquired respiratory distress syndrome (CARDS) toxin, which binds surfactant protein and dysregulates host adenosine diphosphate (ADP) ribosylation. High levels of CARDS toxin are produced during infection, and inoculation of this toxin in an animal model induces many of the pathologic features of M. pneumoniae infection, including vacuolization, cilostasis, and inflammatory changes.2 Strain-specific elaboration of biofilms probably plays a role in protecting the organism from host immune cells and may decrease antimicrobial penetration. In the course of M. pneumoniae infection, some patients produce cold agglutinins. These oligoclonal M-type immunoglobulins (IgM) cross-react with I antigens, one of the blood group antigens common to almost all mature human erythrocytes. High titers of the cold reactive antibody may cause hemolysis (presumably as a result of complement-activated, Coombs-positive erythrocyte destruction) and lead to some of the complications described in the Clinical Manifestations section. Like other IgM antibodies, the Mycoplasma-induced cold agglutinins (Chapter 151) develop early in the disease (7 to 10 days) and are often present by the time the patient seeks medical attention. The

titer of these agglutinins peaks at 2 to 3 weeks, and they persist for 2 to 3 months (Fig. 301-1).  

CLINICAL MANIFESTATIONS

In view of the very high incidence of Mycoplasma respiratory infection when it is studied epidemiologically in large populations, versus the rarity of individual sporadic diagnoses, it appears that a specific, laboratory-confirmed diagnosis of this entity is seldom accomplished in routine clinical practice.3 There are probably four reasons for this. First, Mycoplasma pneumonia is usually self-limited and rarely fatal. This fact dampens the zeal to establish the cause of infection. Second, mycoplasmas are relatively fastidious and slow growing; therefore, culture results, if obtained at all, often return only after the patient has recovered. Third, M. pneumoniae responds to the empirical antimicrobial therapy suggested for community-acquired pneumonia. Finally, knowledge of the epidemiology and clinical manifestations of infection is deficient, so the diagnosis is often not considered outside the classic age group.

Respiratory Infection

The majority of M. pneumoniae infections involve only the upper respiratory tract. After a 2- to 3-week incubation period, the disease has an insidious onset consisting of fever, malaise, headache, and cough (see Fig. 301-1). Cough is the clinical hallmark of M. pneumoniae infection. The frequency and severity of the cough increase during the next 1 to 2 days, and it may become debilitating. The gradual onset of symptoms is in contradistinction to the often fulminant manifestation of respiratory infection caused by influenza virus or adenovirus.4 In 5 to 10% of patients, depending somewhat on age, the infection progresses to tracheobronchitis or pneumonia. In these cases, the initial manifestations persist, and the cough becomes more severe. It is usually relatively nonproductive but may yield white or occasionally blood-flecked sputum. Gram staining of this sputum reveals inflammatory cells but no predominant bacterial species (part of the definition of atypical pneumonia). With continued cough, parasternal chest soreness may develop as a result of muscle strain, but true pleuritic

CHAPTER 301  MYCOPLASMA INFECTIONS  

Several theories about the factors triggering the formation of cold agglutinins in Mycoplasma pneumonia have been proposed. One is that the organism alters the I antigen such that it becomes antigenic to the patient. The hydrogen peroxide elaborated by M. pneumoniae could be responsible for this alteration. One study indicated that the I antigen in a sialated state may serve as a receptor for M. pneumoniae and that the cold agglutinins are directed at the modified receptor. Other studies indicate that the cold agglutinins are directed at mycoplasmal substructures themselves and merely cross-react with the I antigen (altered or native) on red cells. Given their apparent target, these antibodies could either contribute to cytolysis and exacerbate infection or interfere with cell-to-cell spread by blocking or disrupting the cell receptor for the mycoplasma. High titers of cold agglutinins have also been associated with hemolysis, presumably as a result of the activation of complement-mediated erythrocyte destruction (Chapter 151). Although clinically significant hemolysis is uncommon, subclinical levels of red cell destruction are common.

1973.e1

Sickle cell disease, sickle-related hemoglobinopathies, and hypogammaglobulinemia predispose to increased severity and mortality. Therefore, some of the available pathologic data may be influenced by the pathophysiologic mechanisms of these underlying conditions. Deaths have occurred in patients with diffuse pneumonia, acute respiratory distress syndrome, thromboembolism, and disseminated intravascular coagulation. In nonfatal cases in which lung biopsy was performed, the inflammatory process involved primarily the trachea, bronchioles, and peribronchial tissue. The lumen of the respiratory tree was filled with a purulent exudate rich in polymorphonuclear leukocytes. The lining of the bronchial and bronchiolar walls showed metaplastic cells, and the walls themselves were infiltrated with monocytic elements, especially plasma cells. There was widening of the peribronchial septa and hyperplasia of type II pneumocytes.

1974

CHAPTER 301  MYCOPLASMA INFECTIONS  

FIGURE 301-2.  Chest radiograph showing moderate interstitial pneumonia due to

Mycoplasma pneumoniae infection in a patient with a paucity of findings on chest auscultation.

pain is unusual. Fever is usually at the level of 101° to 102° F and may be associated with chills. As opposed to pneumonia caused by Streptococcus pneumoniae (Chapter 273), that caused by M. pneumoniae rarely produces true shaking chills. In comparison to influenza, which can also be manifested as an atypical pneumonia syndrome, myalgias and gastrointestinal complaints of nausea and vomiting are unusual. Diarrhea, sometimes a concomitant of adenoviral or Legionella pneumonia, is uncommon in Mycoplasma infection. Unfortunately, no clinical signs or symptoms can reliably differentiate M. pneumoniae infections from other community-acquired pneumonias. On physical examination, the patient does not appear to be severely ill. In fact, this disease is the paradigm of the term walking pneumonia. The pharynx may be injected and erythematous, usually without the marked cervical adenopathy seen with group A streptococcal pharyngitis. M. pneumoniae is not a common cause of isolated pharyngitis in the pediatric or adult population. Much has been made of the finding of bullous myringitis in this disease. Although this abnormality was present in about 20% of volunteers with experimentally induced M. pneumoniae infection, true bullous myringitis in naturally occurring Mycoplasma disease is rare. In a study involving children, otitis was rarely associated with isolation of Mycoplasma and, on the contrary, was often associated with bacterial and viral upper respiratory tract pathogens. Thus, the absence of myringitis, bullous or otherwise, should not dissuade one from a diagnosis of Mycoplasma pneumonia. Examination of the chest in patients with Mycoplasma pneumonia is often unrevealing, even in those with severe, productive cough. There may be no auscultative or percussive findings, or only minimal rales (crackles) may be present. The disparity between physical findings and radiographic evidence of pneumonia in this condition may be the greatest of any of the atypical pneumonia syndromes (Fig. 301-2). Although wheezing can occur in this disease, in one study of asthmatic patients, the presence of wheezing had a negative correlation with isolation of M. pneumoniae compared with isolation of viral respiratory pathogens such as respiratory syncytial virus (Chapter 338). M. pneumoniae does not seem to be a common pathogen in patients with preexisting chronic obstructive lung disease, either. Bacterial superinfection after M. pneumoniae respiratory infection is rare. The radiographic finding of interstitial or patchy alveolar pneumonia does not allow differentiation from any of the other causes of the atypical pneumonia syndrome. Pleural effusion (usually small) occurs in 5 to 20% of patients with M. pneumoniae pneumonia. This low incidence of pleural inflammation is consistent with the rarity of pleuritic pain. If effusion is present, thoracentesis reveals serous fluid that is exudative, with minimal inflammatory reaction. The cell differential count in the fluid is variable, and bloody effusions are rare. It is unusual to isolate M. pneumoniae from effusions when they do occur. Although the pneumonia is generally mild and self-limited, fulminant, severe, and lethal cases have been reported in normal young adults and may be underdiagnosed.

Extrapulmonary Involvement

Abnormalities in almost every organ system have been described as examples of the extrapulmonary manifestations of M. pneumoniae infection. The

FIGURE 301-3.  Stevens-Johnson syndrome in a child with Mycoplasma pneumonia. (From Baum SG. Mycoplasma pneumoniae and atypical pneumonia. In: Mandell GL, Bennett JE, Dolin R, eds. Mandell, Douglas, and Bennett’s Principles and Practice of Infectious Diseases. 6th ed. Philadelphia: Churchill Livingstone; 2005:2274.)

frequency of these extrapulmonary manifestations varies greatly from one report to another, and they are much less common when viewed as part of a prospective epidemiologic study rather than as the sum of isolated case reports. The conclusion appears to be that the high prevalence of Mycoplasma infection predisposes to the reporting of many concurrent but perhaps unrelated events as though they were part of the mycoplasmal disease. Several clinical syndromes have been reported with sufficient frequency to provide some support for a causal relationship.

Dermatologic Involvement

A wide variety of transient dermatologic conditions have been reported in conjunction with Mycoplasma pneumonia, including macular, morbilliform, and papulovesicular eruptions, as well as erythema nodosum and urticaria. Again, the variety and high incidence of these rashes in the absence of Mycoplasma infection make it difficult to define the relationships, if any, among these occurrences. Furthermore, the role of concurrent antibiotic therapy in the development of exanthems during M. pneumoniae infection is unknown. One skin condition that occurs often enough in concert with M. pneumoniae infection to provide some basis for relatedness is erythema multiforme majus, or Stevens-Johnson syndrome (Fig. 301-3). This rash has been reported in up to 7% of patients with Mycoplasma pneumonia and consists of erythematous vesicles, plaques, and bullae involving the skin, with particular localization at mucocutaneous junctions. The conjunctivae as well as organs of the gastrointestinal and genitourinary tracts and the joints may also be involved. StevensJohnson syndrome has been associated with isolated cases of many other infections, including some that can be manifested as atypical pneumonia syndrome, such as Legionnaires pneumonia, adenovirus respiratoryconjunctivitis syndrome, and influenza B infection. However, the association with M. pneumoniae infection is by far the most common. Stevens-Johnson syndrome tends to occur in younger patients with Mycoplasma pneumonia and has a definite male preponderance (2 : 1 to 4 : 1). The pathogenesis of Stevens-Johnson syndrome is unclear. It has long been supposed that immunity plays a major role, but several reports have noted the culture of M. pneumoniae from the lesions. The relationship to the level of cold agglutinins in this disease is variable. It has been suggested that the development of Stevens-Johnson syndrome may be the result of augmented sensitivity to antibiotics in the presence of M. pneumoniae infection, but the syndrome develops in some patients even in the absence of previous or concurrent antibiotic therapy. Most patients clear the lesions in 1 to 2 weeks without scarring unless impetiginization supervenes. A debilitating form of M. pneumoniae–associated mucositis without skin involvement has been described as an atypical Stevens-Johnson syndrome, but it is now increasingly recognized as a separate entity, termed M. pneumoniae– associated mucositis.5,6 Raynaud phenomenon (Chapter 72), a transient, reversible vasospasm of the digits that develops on exposure to cold, is not technically a dermatologic syndrome; however, it is manifested in the skin. This phenomenon occurs in many people, usually women, without any association with infection. Although the pathophysiologic mechanism of this condition in M. pneumoniae infection

CHAPTER 301  MYCOPLASMA INFECTIONS  

1975

is unclear, it may be related to the in vivo action of cold agglutinins (see Diagnosis, later). Other vascular complications reported in M. pneumoniae infection include internal carotid artery occlusion and cerebral infarction.

Unusually severe but nonlethal M. pneumoniae infection has also been reported in children with Down syndrome.

Cardiac Complications

The diagnosis of Mycoplasma pneumonia is made primarily on clinical grounds. The organism can be grown in cell-free media, but most hospital diagnostic laboratories are not prepared to culture mycoplasmas. Because of this, there is considerable interest in developing rapid diagnostic tests with high sensitivity and specificity. These assays fall into three categories: detection of M. pneumoniae–specific immunoglobulins in serum and detection of M. pneumoniae–specific antigens or nucleotide sequences directly in clinical specimens. Diagnostically, the most useful M. pneumoniae–specific immunoglobulin to detect is IgM because it is most likely to indicate recent infection. Enzyme-linked immunoassays have been developed to detect IgM and IgG directed against M. pneumoniae. When used in patients with positive assays for complement-fixing antibodies, the enzyme immunoassay had a specificity of more than 99% and a sensitivity of 98%. Specificity was retained but sensitivity dropped to only 46% when IgG alone was the target. Variations on this theme detect IgM antibodies directed at specific M. pneumoniae antigens. The tests are simple to perform and have high sensitivity and specificity, but because all these tests are designed to detect IgM antibody, results may be negative early in the course of infection ( M) Conjunctivae (neonate) Blood (peripartum) Surgical wounds, joints

DISEASES Cervicitis, vaginitis, ?prostatitis Conjunctivitis

OCCURRENCE Common

Peripartum sepsis Sternotomy infection, arthritis

M. orale

Oropharynx

?

Common

M. salivarium

Oropharynx, gingiva

?Periodontal disease

Common

M. fermentans

GU tract, blood, tissues

Multisystem disease in AIDS

Uncommon

M. genitalium

GU tract

Urethritis, cervicitis, PID

Uncommon

Ureaplasma spp

GU tract

Urethritis, upper GU infection

Common

AIDS = acquired immunodeficiency syndrome; F = female; GU = genitourinary; M = male; PID = pelvic inflammatory disease; URI = upper respiratory tract infection.

immunocompromised patients caused by Mycoplasma fermentans (incognitus strain).  

EPIDEMIOLOGY

Ureaplasma urealyticum, Ureaplasma parvum, M. hominis, and M. genitalium are spread venereally, are not well documented (Table 301-1), and are difficult to culture but can be better identified by nucleic acid amplification testing. Their true incidence and prevalence are uncertain, as is their public health impact.  

PATHOBIOLOGY

The genital mycoplasmas attach to cells using an adhesion protein as well as lipid-associated membrane proteins that adhere by way of toll-like receptors. Once within the cell, M. genitalium has been shown to trigger elaboration of numerous cytokines that cause inflammation.  

Outbreaks of M. pneumoniae respiratory infection among military recruits have led to interest in producing a vaccine to protect against this organism. The resulting vaccines induced specific antibody responses, but protection was limited to no more than 50% of vaccine recipients. Live vaccines using attenuated wild-type and temperature-sensitive mutant mycoplasmas have proved no more effective. In one study, volunteers who received vaccine but did not mount an antibody response had more severe disease when rechallenged with wild-type Mycoplasma than did nonvaccinated personnel. Although M. pneumoniae is perhaps the leading cause of atypical pneumonia syndrome in closed populations, the enthusiasm for developing a vaccine for this disease has waned. New technology involving DNA expression library immunization has proved successful in animal studies with nonhuman mycoplasmas, and these methods may breathe new life into M. pneumoniae vaccine development.

SITES OF ISOLATION

SUBGROUP

CLINICAL MANIFESTATIONS

Diseases attributed to Ureaplasma species include urinary tract infection with and without calculus formation. The organism has been implicated as a cause of low birthweight in neonates. There is increasing evidence that chronic M. genitalium infection causes nongonococcal urethritis (Chapter 269) in males and cervicitis and pelvic inflammatory disease (Chapters 269, 283, and 302) in females, in whom it leads in some cases to tubal factor infertility (Chapter 223), preterm birth, and spontaneous abortion.12,13 Chronic M. genitalium infection may enhance the acquisition of human immunodeficiency virus (HIV) through increased shedding, and prevalence of M. genitalium appears higher in patients with HIV than in the general population. M. hominis is a common genitourinary and oral commensal as well and has been documented as a cause of endometritis and postpartum fever. M. hominis can also cause sternal wound infection after cardiothoracic surgery and has been implicated in arthritis in immunocompromised patients. M. salivarium may be involved in periodontal disease. M. fermentans, incognitus strain, was identified about four decades ago as an infectious agent in immunocompromised patients, in whom it causes overwhelming multisystem involvement.  

DIAGNOSIS

The development of nucleic acid amplification tests such as PCR has enhanced the ability to detect M. genitalium and other genital mycoplasmas in secretions. However, the sensitivity and specificity of these tests appear to vary. Some commercially available tests can detect several mycoplasmas and chlamydial organisms in a single panel, and the refinement and standardization of these tests should lead to additional clarity of the pathogenic role of these organisms.

TREATMENT  M. hominis is not sensitive to macrolides, but it is sensitive to the other antimicrobials recommended for M. pneumoniae infection. Azithromycin (1-g single dose) achieves a significantly higher cure rate for M. genitalium–associated urethritis than does multidose doxycycline. The success rate for a single dose of 500 mg azithromycin is only about 60%, whereas moxifloxacin may be successful in close to 90% of cases.14



PROGNOSIS

Prognosis of early disease without scarring of the cervix and fallopian tubes is relatively good. However, treatment of all sexual partners of the index case is paramount to prevent reinfection. Prognosis also depends on the susceptibility of the offending organisms. It is highly likely that antimicrobial resistance will increase in the future, just as it has for azithromycin in the past.

Grade A References A1. Mulholland S, Gavranich JB, Gillies MB, et al. Antibiotics for community-acquired lower respiratory tract infections secondary to Mycoplasma pneumoniae in children. Cochrane Database Syst Rev. 2012;9:CD004875. A2. Huang L, Gao X, Chen M. Early treatment with corticosteroids in patients with Mycoplasma pneumoniae pneumonia: a randomized clinical trial. J Trop Pediatr. 2014;60:338-342.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 301  MYCOPLASMA INFECTIONS  

GENERAL REFERENCES 1. Bajantri B, Venkatram S, Diaz-Fuentes G. Mycoplasma pneumoniae: a potentially severe infection. J Clin Med Res. 2018;10:535-544. 2. Maselli DJ, Medina JL, Brooks EG, et al. The immunopathologic effects of Mycoplasma pneumoniae and community-acquired respiratory distress syndrome toxin. A primate model. Am J Respir Cell Mol Biol. 2018;58:253-260. 3. Waites KB, Xiao L, Liu Y, et al. Mycoplasma pneumoniae from the respiratory tract and beyond. Clin Microbiol Rev. 2017;30:747-809. 4. Clemmons NS, Jordan NN, Brown AD, et al. Outbreak of Chlamydia pneumoniae infections and X-ray-confirmed pneumonia in army trainees at Fort Leonard Wood, Missouri, 2014. Mil Med. 2019. [Epub ahead of print.] 5. Vujic I, Shroff A, Grzelka M, et al. Mycoplasma pneumoniae–associated mucositis—case report and systematic review of literature. J Eur Acad Dermatol Venereol. 2015;29:595-598. 6. Canavan TN, Mathes EF, Frieden I, et al. Mycoplasma pneumoniae-induced rash and mucositis as a syndrome distinct from Stevens-Johnson syndrome and erythema multiforme: a systematic review. J Am Acad Dermatol. 2015;72:239-245. 7. Kammer J, Ziesing S, Davila LA, et al. Neurological manifestations of Mycoplasma pneumoniae infection in hospitalized children and their long-term follow-up. Neuropediatrics. 2016;47:308-317.

1977.e1

8. Huang C, Huang PT, Yao JY, et al. Pooled analysis of nuclear acid sequence-based amplification for rapid diagnosis of Mycoplasma pneumoniae infection. J Clin Lab Anal. 2019. [Epub ahead of print.] 9. Tashiro M, Fushimi K, Kawano K, et al. Comparison of efficacy of antimicrobial agents among hospitalized patients with Mycoplasma pneumoniae pneumonia in Japan during large epidemics of macrolide-resistant M. pneumoniae infections: a nationwide observational study. Clin Infect Dis. 2017;65:1837-1842. 10. Diaz MH, Benitez AJ, Winchell JM. Investigations of Mycoplasma pneumoniae infections in the United States: trends in molecular typing and macrolide resistance from 2006 to 2013. J Clin Microbiol. 2015;53:124-130. 11. Miyashita N, Kawai Y, Inamura N, et al. Setting a standard for the initiation of steroid therapy in refractory or severe Mycoplasma pneumoniae pneumonia in adolescents and adults. J Infect Chemother. 2015;21:153-160. 12. Lis R, Rowhani-Rahbar A, Manhart LE. Mycoplasma genitalium infection and female reproductive tract disease: a meta-analysis. Clin Infect Dis. 2015;61:418-426. 13. Jensen JS. Mycoplasma genitalium: yet another challenging STI. Lancet. 2017;17:795-796. 14. Bissessor M, Tabrizi SN, Twin J, et al. Macrolide resistance and azithromycin failure in a Mycoplasma genitalium-infected cohort and response of azithromycin failures to alternative antibiotic regimens. Clin Infect Dis. 2015;60:1228-1236.

1977.e2

CHAPTER 301  MYCOPLASMA INFECTIONS  

REVIEW QUESTIONS 1. Mycoplasma pneumoniae is one of the most common causes of the atypical pneumonia syndrome. Another organism that shares many of the characteristics of organisms causing this syndrome is: A . Streptococcus pneumoniae B. Chlamydophila pneumoniae C. Staphylococcus aureus D. Klebsiella pneumoniae E. Acinetobacter baumannii Answer: B  Chlamydophila does not show up on Gram stain or grow in common bacterial culture, two of the common characteristics of these “atypical” organisms. (See Definition section.) 2. Which of the following classes of antimicrobials would not be effective in treating pneumonia in a young adult caused by M. pneumoniae? A . Fluoroquinolones B. β-Lactamase–resistant penicillins C. Extended-spectrum macrolides D. Tetracyclines E. MLSK class Answer: B  Mycoplasmas do not have cell walls, and the penicillins work by inhibiting cell wall synthesis. The other drugs all are effective in treating M. pneumoniae. (See section on treatment.) 3. Which of the following is true of cold agglutinins that occur in about 50 to 70% of patients with M. pneumoniae pneumonias? A . They are IgG antibodies directed against the nucleic acid of the organism. B. They are secreted by the organism as part of the infectious process. C. They help to neutralize the organism and thereby protect the host. D. They are IgM class antibodies directed at antigens on the host erythrocyte. E. When present, they are pathognomonic of Mycoplasma infection. Answer: D  Mycoplasma-induced cold agglutinins are IgM antibodies directed at the I antigens of red cell membranes. They are detected by a positive direct antiglobulin test (DAT) result, also known as the direct Coombs’ test. Although clinically significant hemolysis is rare, subclinical positivity for DAT is common in Mycoplasma pneumonia. (See Pathobiology section.)

4. Which of the following is true about Mycoplasma hominis? A . Causes puerperal sepsis B. Is sensitive to macrolides, as are the other mycoplasmas C. Is a common cause of infectious carditis D. Is the most common cause of nonbacterial urinary tract infection E. Has, as its primary reservoir in nature, the feral cat Answer: A  M. hominis is a common genitourinary and oral commensal, but it has also been documented as a cause of endometritis and postpartum fever. (See section on epidemiology.) 5. Which of the following chest radiographic patterns would be most characteristic of M. pneumoniae pneumonia? A . Pneumothorax B. Dense, multilobar alveolar infiltrates C. Dense, unilobar infiltrate with ipsilateral pleural effusion D. No infiltrate seen E. Multifocal interstitial pattern Answer: E  The disparity between the paucity of physical findings and the radiographic appearance of pneumonia in this condition can be striking (see Fig. 301-2).

CHAPTER 302  Diseases Caused by Chlamydiae  

1977

cytokines, restriction of select nutrients) in which they have reduced metabolic activity and may be more refractory to antibiotic treatment; whether this occurs in vivo is unclear. Chlamydiae are unable to synthesize adenosine triphosphate and therefore depend on the host cell for nutrients to meet their energy requirements.  

PATHOBIOLOGY

Macrophages are the principal host cells for C. psittaci and C. trachomatis LGV biovar, whereas the principal host cells for C. trachomatis trachoma biovar and C. pneumoniae strains are columnar epithelial cells at mucosal sites. Host cell tropism correlates with the type of inflammation elicited by chlamydiae. The LGV biovar and C. psittaci produce granulomatous inflammation, characteristic of delayed hypersensitivity reactions. The trachoma biovar produces neutrophilic exudate during acute infection and submucosal mononuclear infiltration with lymphoid follicle formation during later stages of infection. Chlamydiae elicit both humoral and cellular immune responses. Infection can persist or recur even after an adaptive immune response develops, suggesting that the organism has evolved strategies for immune evasion. Persistent or recurrent infections can elicit inflammatory cellular immune responses that cause tissue injury.

  CHLAMYDIAL DISEASES

Table 302-1 summarizes the diseases caused by chlamydiae in humans and the associated clinical and laboratory characteristics.

Chlamydia trachomatis

302  DISEASES CAUSED BY CHLAMYDIAE WILLIAM M. GEISLER



DEFINITION

Chlamydiae are obligate intracellular bacteria that cause a variety of human and animal diseases and much morbidity. Chlamydiae were originally classified taxonomically into one genus, Chlamydia. On the basis of sequence analysis of 16S rRNA genes, it had been proposed that chlamydial taxonomy be revised to contain two genera: Chlamydia and Chlamydophila. However, on the basis of additional data on chlamydia genome sequences and meetings within the Chlamydia scientific community, it has been agreed that the family Chlamydiaceae will contain a single genus, Chlamydia. Within the family are now nine recognized species: C. trachomatis, C. pneumoniae, C. psittaci, C. pecorum, C. muridarum, C. felis, C. abortus, C. suis, and C. caviae. C. trachomatis is classified into a trachoma biovar and lymphogranuloma venereum (LGV) biovar. This chapter is limited to human diseases caused by chlamydiae.

The Pathogen

Chlamydiae have a gram-negative cell wall structure consisting of an outer membrane that contains lipopolysaccharide and an inner cytoplasmic membrane. The outer membrane contains a single 40-kD major outer membrane protein, OmpA (also known as MOMP), and two cysteine-rich minor outer membrane–associated proteins (OmcA and OmcB); through intermolecular and intramolecular disulfide bonding, these proteins form a complex that provides structural rigidity. Chlamydiae grow only within intracellular membrane-bound vacuoles, termed inclusions,1 that seclude the organism from extracellular and cytoplasmic environments. They share a distinct biphasic developmental cycle (Fig. 302-1) that includes an extracellular, metabolically inactive, infectious form (an elementary body) and an intracellular replicative form (a reticulate body). In vitro studies have shown that chlamydiae may enter a persistent state under certain conditions (penicillin treatment, challenge with certain

C. trachomatis infection is a common bacterial infection in humans and accounts for significant morbidity worldwide. C. trachomatis isolates have been differentiated into 18 major serovars (i.e., OmpA types) based on variations in OmpA that are identified on antigen cross-reactivity in the microimmunofluorescence test. The major diseases caused by C. trachomatis are trachoma, caused by serovars A, B, Ba, and C; sexually and perinatally transmitted diseases, caused by serovars D through K (and, rarely, serovars B and Ba); and sexually transmitted LGV, caused by serovars L1, L2, L2a, and L3. Sequencing of the ompA gene has led to the recognition of more OmpA variants, including L2b. Multilocus sequencing typing is a newer tool that has been used to further discriminate between C. trachomatis strains of the same OmpA genotype. Trachoma and LGV are endemic in developing areas of the world (although LGV outbreaks have also occurred in populations of men who have sex with men in developed countries), whereas sexually and perinatally transmitted non-LGV chlamydial infections occur worldwide.

TRACHOMA EPIDEMIOLOGY  

Trachoma is a chronic follicular conjunctivitis. The overall incidence is unknown, but it has been estimated by the World Health Organization (WHO) that 21.4 million people have active trachoma. Trachoma is endemic in more than 41 countries and is especially common in poor areas of sub-Saharan Africa, where the disease prevalence in children may exceed 40%. According to the WHO, there are 190.2 million people living in trachoma-endemic areas. Trachoma is a major public health problem because scarring from trachoma causes blindness, affecting 7.2 million people by WHO estimates. Trachoma is the most common preventable cause of blindness worldwide. Active trachoma often occurs in the first few years of life. The inflammation from recurrent or persistent trachoma can lead to conjunctival scarring, which can ultimately cause corneal damage and blindness later in life. The C. trachomatis serovars that produce trachoma are spread by direct contact with fingers or fomites (e.g., washcloths, handkerchiefs) contaminated with eye discharge from an infected person or by eye-seeking flies. Because of this mode of transmission, trachoma often clusters in households. Risk factors for trachoma include poor facial hygiene, limited access to water, poor sanitation, and proximity to other infected persons or to a heavy density of eye-seeking flies.  

CLINICAL MANIFESTATIONS

There are two stages of trachoma disease, and they can overlap. Initially, trachoma begins as an inflammatory follicular conjunctivitis (i.e., active trachoma). On eversion of the upper eyelid, white to pale yellow lymphoid follicles can be visualized on the superior tarsal conjunctival surface, and papillae may be noted between follicles. Minimal watery or mucoid ocular discharge may also be seen. In more severe active trachoma, the conjunctiva can be thickened and edematous. Subsequently, conjunctival inflammation can progress to cause

CHAPTER 302  Diseases Caused by Chlamydiae  

ABSTRACT

Chlamydiae are gram-negative obligate intracellular bacteria that grow only within intracellular vacuoles, called inclusions that seclude the microorganism from extracellular and cytoplasmic environments. Major diseases caused by C. trachomatis are trachoma, the most common preventable cause of blindness worldwide; the most prevalent bacterial sexually transmitted infection in the United States; and lymphogranuloma venereum. Infections are difficult to diagnose clinically because the majority of patients are asymptomatic, and even when symptoms or signs are present, they are nonspecific. With the development of nonculture tests, nucleic acid amplification testing (NAAT) is now the recommended diagnostic method for clinical testing chlamydia. C. trachomatis is susceptible to tetracyclines, macrolides, and select quionolones (ofloxacin and levofloxacin but not ciprofloxacin). Regimens vary depending on clinical manifestations.

KEYWORDS

chlamydia trachomatis pneumoniae psittaci lymphogranuloma venereum

1977.e3

1977.e4

CHAPTER 302  Diseases Caused by Chlamydiae  

Additional knowledge of chlamydial cell biology emerged after several chlamydial genomes were sequenced. Polymorphic outer membrane proteins were identified, but their role in pathogenesis is unclear. Chlamydiae were also found to express proteins that localize to the cytoplasmic surface of the inclusion membrane (e.g., IncA). C. trachomatis was demonstrated to have type III secretion genes, which may encode a secretion apparatus providing a means for protein transport. Chlamydiae have small genomes (C. trachomatis contains 894 protein-coding genes, and C. pneumoniae contains 1052 genes). Most strains of C. trachomatis and some strains of C. psittaci contain a 7-kilobase cryptic plasmid.

1978

CHAPTER 302  Diseases Caused by Chlamydiae  

8 hrs

Attachment

Endocytosis

40-60 hrs

Differentiation EB

Binary fission

Differentiation RB

RB

Release

EB

EB RB

EB

2.5m

1m

1m

0.5m

FIGURE 302-1.  Developmental cycle of chlamydiae. The top panel shows the developmental cycle common to all chlamydiae. The red dots represent elementary bodies (EBs), and the blue dots represent reticulate bodies (RBs). Chlamydiae infect eukaryotic cells through multiple attachment mechanisms. After attachment, EBs enter the cell within a membranebound vacuole that remains unfused with lysosomes. EBs reorganize into RBs and asynchronously replicate 8 to 12 times, with a doubling time of 2 to 3 hours. At the conclusion of the growth cycle, RBs differentiate back to EBs, and each inclusion yields 100 to 1000 new infectious EBs. The bottom left panel, a transmission electron micrograph taken 40 hours after infection, shows the large RBs and the smaller EBs, which have a condensed nucleoid structure within their cytoplasm. The bottom right panel, a scanning electron micrograph taken 60 hours after infection, shows a membrane-bound vacuole containing many EBs exiting from an infected HeLa cell.

TABLE 302-1 MAJOR DISEASES CAUSED BY CHLAMYDIAE IN HUMANS AND ASSOCIATED CLINICAL AND LABORATORY FEATURES SPECIES

SEROVAR

C. trachomatis

A–C D–K D–K L1–L3

DIAGNOSIS

PREVENTION

Trachoma Urethritis, cervicitis, proctitis, epididymitis, PID Inclusion conjunctivitis, infant pneumonia Lymphogranuloma venereum

DISEASE

Fomites, eye-seeking flies Sexual contact

TRANSMISSION ROUTE

Clinical criteria or culture/NAAT NAAT

Perinatal contact

Culture, DFA, NAAT, or serology (for pneumonia) Serology or culture/NAAT plus OmpA typing

SAFE strategy Abstinence or monogamy, education, condoms, partner treatment Prenatal chlamydia screening: treat infected mothers Abstinence or monogamy, education, condoms, partner treatment

Sexual contact

C. pneumoniae

One

Upper respiratory infections, atypical pneumonia, asthma exacerbations

Respiratory droplet

Serology or culture/PCR

None

C. psittaci

Multiple

Psittacosis, atypical pneumonia, febrile illness

Aerosolized bird secretions, dust

Serology

Quarantine and chlortetracycline for imported birds, avoidance or precautions for at-risk subjects

DFA = direct fluorescent antibody test; NAAT = nucleic acid amplification test; PCR = polymerase chain reaction; PID = pelvic inflammatory disease; SAFE = World Health Organization’s recommended strategy acronym for: surgery (for trichiasis), antimicrobials (periodic community-wide treatment), facial cleanliness, and environmental improvement.

scarring of the upper tarsal conjunctiva (the cicatricial stage of disease). Scarring deforms the eyelid and can lead to an inward turning of the eyelashes, which can result in corneal abrasion (trichiasis). Over time, trichiasis causes corneal edema, ulceration, vascularization (pannus), scarring, and opacification. The corneal damage leads to decreased vision or blindness, occurring mostly in young adults and middle-aged persons. Viral conjunctivitis (e.g., adenovirus) is manifested in a clinical fashion similar to active trachoma, but it is self-limited and usually resolves within a week. Trachoma can be complicated by superinfection with other bacterial pathogens (e.g., Haemophilus influenzae), which should be considered when purulent ocular discharge or significant inflammation of the bulbar conjunctiva is present.  

DIAGNOSIS

Because the majority of trachoma cases occur in developing countries without access to laboratory testing or the necessary resources, trachoma is often diagnosed clinically on the basis of findings of active trachoma (follicles on the upper tarsal conjunctiva or pronounced inflammatory thickening of the tarsal conjunctiva) or cicatricial disease. When laboratories are available, detection of C. trachomatis provides definitive evidence of active trachoma and may identify infection in subjects with minimal clinical evidence of active trachoma. Isolation of the organism in cell culture is one means to detect C. trachomatis,

but the test’s sensitivity is less than 50%, and the methods are labor-intensive. Nonculture tests have higher sensitivities in active trachoma. For instance, nucleic acid amplification tests (NAATs) are the most sensitive diagnostic tests, but they are not widely available in many trachoma-endemic areas. It is uncommon for adults with late scarring to have C. trachomatis detected by any of these assays.

TREATMENT  Active trachoma can be treated with a tetracycline eye ointment twice daily for 6 weeks or oral macrolide therapy.2 The latter is preferred in part because extraocular sites such as the nasopharynx may be infected in young children. Single-dose oral azithromycin (20 mg/kg; maximum of 1 g) is as effective as tetracycline ointment and is more advantageous in terms of compliance and side-effect profile; tetracycline application can irritate the ocular surface. Surgical intervention is the only effective management for trichiasis. Eyelid rotation surgery prevents the eyelashes from abrading the cornea, preventing blindness and other nonvisual symptoms.3 Trichiasis recurrence after surgery is a major concern, and recurrence rates are highly variable across studies. Other concerns include accessibility to surgery and the patient’s acceptance.

CHAPTER 302  Diseases Caused by Chlamydiae  



PREVENTION

The WHO is committed to eliminating blinding trachoma by 2020 and recommends that all countries with endemic trachoma adopt the SAFE strategy: surgery (for trichiasis), antimicrobials (periodic community-wide treatment), facial cleanliness, and environmental improvement. Mass treatment of a community with single-dose oral azithromycin is safe and has dramatically reduced the prevalence of infection for up to 1 year after treatment; it has also been shown to reduce mortality in children. A1  Annual azithromycin treatment is recommended for trachoma-endemic areas. Reintroduction of trachoma after mass treatment has been demonstrated, which may be due in part to decreased herd immunity. Repeated mass treatment (every few months) provides herd protection to the entire community. Mass antibiotic treatment as the sole intervention for eliminating trachoma is unlikely to be successful if other factors that facilitate transmission are not addressed. Face-washing and good hygiene help reduce the risk for transmission through contact with fingers and flies. Although the promotion of facial cleanliness through educational campaigns may be one of the most important interventions, sustaining such behavioral changes can be challenging. Achieving better environmental conditions through measures that reduce household and community fly density and improve waste management and access to clean water can also limit transmission. Improvement in socioeconomic conditions in a community correlates with a decline in trachoma prevalence.

SEXUALLY AND PERINATALLY TRANSMITTED CHLAMYDIAL INFECTIONS EPIDEMIOLOGY  

Chlamydia is the most prevalent bacterial sexually transmitted infection in the United States. More than 1.7 million infections were reported to the Centers for Disease Control and Prevention (CDC) in 2017.4 The number of reported

1979

cases is increasing each year; this could reflect increased chlamydia screening efforts rather than a true increase in infection burden,5 but there is no strong evidence to support that screening rates are increasing, and thus it may be a true increase in the number of chlamydia cases. Taking into account underreporting and underscreening, it is estimated by the CDC that more than 2.8 million new chlamydial infections occur annually in the United States. Higher chlamydia prevalence rates have been associated with younger age (sexually active adolescents and young adults), select minority ethnic groups (especially African Americans), and new or multiple sexual partners. Chlamydia prevalence is higher in women than in men; it is unclear whether this is due to higher screening rates in women or whether women may be more susceptible to infection acquisition or persistence. Chlamydia prevalence is highest in the southeastern United States. The estimated total cost attributable to chlamydial disease in the United States exceeds $2.4 billion annually. From a global perspective, WHO estimates that there are approximately 131 million new cases of chlamydia annually. In addition to the adverse effects on the reproductive health of women, chlamydia has a substantial impact on prenatal and perinatal outcomes and facilitates the transmission of human immunodeficiency virus (HIV).  

CLINICAL MANIFESTATIONS

Urethritis

C. trachomatis is the most common cause of nongonococcal urethritis in men,6 being responsible for 20 to 50% of cases. Although the majority of men with chlamydial urethritis do not have symptoms, studies in high-prevalence venues (e.g., sexually transmitted disease clinics) have reported that 40 to 60% of men with chlamydial urethritis have symptoms. The most frequent symptoms are urethral discomfort (itching or pain) with urination and urethral discharge. On examination, a mild to moderate amount of clear or cloudy/mucoid (rarely purulent) urethral discharge may be visualized (Fig. 302-2A). Urethral discharge

A

C

B FIGURE 302-2.  Clinical manifestations of genital Chlamydia trachomatis infection. A, Cloudy urethral discharge of urethritis. B, Urethral specimen Gram stain revealing nongonococcal urethritis findings: two or more polymorphonuclear cells per oil field (1000×) and the absence of intracellular gram-negative diplococci. C, Purulent endocervical discharge of mucopurulent cervicitis. (A, Courtesy James Sizemore, MD. B and C, From Practitioner’s Handbook for the Management of Sexually Transmitted Disease. Retrieved from http://depts. washington.edu/handbook.)

1980

CHAPTER 302  Diseases Caused by Chlamydiae  

sometimes becomes apparent only after stripping the urethra from the base of the penis to the glans; this should be considered in men reporting urethral symptoms but without urethral discharge on initial inspection. Gram stain from a urethral swab specimen demonstrates five or more polymorphonuclear leukocytes (PMNs) per oil field (1000×) in the majority of chlamydial infections (Fig. 302-2B). Chlamydial organisms cannot be visualized on Gram stain. Up to 20% of men with chlamydial urethritis have fewer than five PMNs per oil field on urethral Gram stain, reflecting the fact that minimal inflammation may be elicited in some cases of chlamydial urethritis. Urethral inflammation can also be detected by a positive urinary leukocyte esterase test result on unspun first-void urine. C. trachomatis urethritis also occurs in women, who may have no symptoms or present with an acute urethral syndrome characterized by dysuria, urinary frequency, and/or pyuria. This acute urethral syndrome mimics a urinary tract infection, and chlamydia should be suspected in women with pyuria but negative urine nitrite and/or negative urine culture, especially in sexually active adolescents and young adults; many urinary tract infection treatments are not effective against chlamydia. Mild urethral discharge may be seen. Pelvic examination should be performed in women with suspected chlamydia urethritis to search for other clinical findings of chlamydia (e.g., cervicitis).

Epididymitis

Chlamydia can spread from the urethra to the epididymis, causing epididymitis in up to 1% of infected men.7 Symptoms include testicular pain and scrotal erythema and swelling that are typically unilateral. On examination, there is palpable swelling and tenderness of the epididymis; accompanying findings may include testicular tenderness, scrotal erythema and swelling, urethral discharge, or hydrocele. In men younger than 35 years, C. trachomatis is the principal cause of epididymitis, whereas in men older than 35 years, complicated urinary tract infection with uropathogens is a more common cause, especially in those with prostate disorders. Up to 15% of epididymitis cases are complicated by chronic pain that is usually idiopathic and often unresponsive to antibiotics. Other complications include decreased fertility and, rarely, testicular abscess.

Cervicitis

C. trachomatis is the most common cause of cervicitis, being responsible for up to 50% of cases. The majority of women with endocervical chlamydial infection are asymptomatic. Symptoms, when present, are often mild and nonspecific for chlamydia and include the following: vaginal discharge, intermenstrual vaginal bleeding, dysuria, and pain during intercourse (dyspareunia). Up to 13% of women with asymptomatic cervical chlamydial infection have mucopurulent cervicitis detected on pelvic examination (Fig. 302-2C); this is characterized by a purulent or cloudy endocervical discharge visible in the endocervical canal or on the tip of an endocervical swab. A similar proportion may have endocervical bleeding that is easily induced with passage of a swab through the cervical os. Nonspecific findings may include vaginal discharge and edematous ectopy (a darker red area of columnar epithelium visible on the face of the cervix). A vaginal wet mount often shows more than 5 to 10 PMNs per 400× field.

Pelvic Inflammatory Disease

Chlamydia can spread from the cervix to the endometrium (causing endometritis), fallopian tubes (causing salpingitis), and peritoneum (causing peritonitis or perihepatitis). These upper genital tract infections are collectively referred to as pelvic inflammatory disease (PID; Chapters 269 and 283). Estimates of the proportion of cervical chlamydial infections that progress to PID vary greatly but are most commonly 10 to 35%.9 The majority of PID cases are subclinical or silent. PID symptoms include pelvic or lower abdominal pain (especially during menses or the first 2 weeks of the menstrual cycle) and nausea. Fever is less common. Examination findings include cervical motion tenderness and tenderness of the uterus or adnexa. Although most cases of chlamydial PID are due to the natural progression of infection, they may also occur postpartum or after pregnancy termination. The long-term consequences of PID include infertility, ectopic pregnancy, and chronic pelvic pain.

Complications during Pregnancy

There is some evidence that genital chlamydial infection during pregnancy can lead to adverse outcomes, including preterm labor, low birthweight, miscarriage, and stillbirth.10

Reactive Arthritis

Reactive arthritis, characterized by the classic triad of trigger infection (e.g., chlamydia), conjunctivitis, and arthritis, can complicate chlamydial infection (Chapter 249). There is a male predominance in reactive arthritis cases triggered by chlamydia, and it has been estimated that reactive arthritis occurs in up to 1% of men presenting with chlamydial urethritis.

Proctitis

Proctitis caused by non-LGV C. trachomatis OmpA types is usually asymptomatic. Subjects with acute symptomatic proctitis may report rectal pain or bleeding, tenesmus, pruritus, rectal discharge, or diarrhea. Anoscopy or sigmoidoscopy may reveal friable mucosa and a mucoid or mucopurulent discharge. A rectal swab Gram stain often reveals many PMNs per oil field.

Conjunctivitis

An acute follicular conjunctivitis may rarely occur in adolescents or adults with genital chlamydial infection. The presumed mode of acquisition is autoinoculation with infected genital secretions. The typical clinical presentation is a subacute or indolent infection characterized by unilateral conjunctival redness, mucoid or mucopurulent ocular discharge, a foreign body sensation, and preauricular adenopathy.

Oropharyngeal Infection

C. trachomatis has been detected in the oropharynx of sexually active subjects and is asymptomatic in most instances. Recent evidence suggests that C. trachomatis can be transmitted from oropharyngeal sites to the genital tract, providing rationale for treatment of oropharyngeal chlamydia. However, because oropharyngeal chlamydia prevalence is very low in most populations and the clinical significance of C. trachomatis detected in the oropharynx is unclear, routine oropharyngeal chlamydia screening is not recommended.

Infant Inclusion Conjunctivitis and Pneumonia

Because the prevalence of chlamydia in pregnant adolescents and young adults in the United States can be high (>5%), the morbidity associated with perinatally transmitted chlamydia is considerable. Neonates exposed to C. trachomatis during passage through the birth canal may develop inclusion conjunctivitis (≈20 to 40%) or pneumonia (≈10 to 20%). The conjunctivitis, termed inclusion conjunctivitis because the cytoplasmic chlamydia inclusion bodies demonstrated in neonatal conjunctival scrapings are the same as those seen in genital scrapings from adults with genital chlamydia, usually develops 5 to 12 days after birth but may occur as long as 4 to 6 weeks after birth. Clinical manifestations include conjunctival injection and thickening, a clear or mucopurulent ocular discharge, and eyelid swelling. Chlamydia pneumonia in infants usually occurs subacutely between 1 and 3 months of age. Characteristic clinical features include a repetitive staccato cough and absence of fever. Other clinical findings may include tachypnea, crackles on auscultation of the lungs, nasal discharge, and eosinophilia. Chest radiographs may reveal bilateral diffuse infiltrates.

Lymphogranuloma Venereum

LGV is a sexually transmitted infection caused by invasive C. trachomatis LGV OmpA serovars. In contrast to infection with non-LGV strains, LGV is a more invasive systemic infection that involves lymphoid tissue (causing lymphadenitis) and can be ulcerative. LGV is endemic in Africa, India, Southeast Asia, South America, and the Caribbean. LGV had been uncommon in the United States. However, in recent years, developed countries have experienced a shift in LGV epidemiology and clinical presentation, and LGV has emerged in Europe and North America as a leading cause of proctitis and proctocolitis in men who have sex with men.11 Classically, the clinical manifestations of LGV differ in early versus later stages of infection. In the early stage (3 to 30 days after acquisition) of genital LGV, a primary skin lesion or lesions may develop on the genital mucosa or adjacent skin in the form of a papule, ulcer, or herpetiform lesion. The lesion is usually asymptomatic and goes unnoticed, but it may be erosive; it heals quickly without scarring. Early genital LGV may also be manifested as a nonspecific inflammatory syndrome (e.g., urethritis and cervicitis), similar to infection with non-LGV strains. Genital LGV can progress to an inguinal syndrome 2 to 4 weeks later, characterized by painful, erythematous inguinal lymphadenopathy (buboes) and systemic manifestations including fever, headache, arthralgias, myalgias, and leukocytosis. The buboes are commonly unilateral, and about one third spontaneously rupture and

CHAPTER 302  Diseases Caused by Chlamydiae  

drain pus, which can be complicated by fistulas or sinus tracks; unruptured buboes usually heal. In later stages, genital tract fibrosis can lead to complications such as infertility, elephantiasis, strictures, fistulas, and subcutaneous sclerosis. LGV may also be manifested as an anogenital syndrome with invasive proctitis. Symptoms include fever, tenesmus, anal pruritus, and a rectal discharge that may be mucoid or, less commonly, mucopurulent or bloody. Up to 20 to 30% of patients do not have symptoms. The rectal mucosa is friable, with multiple superficial ulcerations, and biopsy may reveal submucosal granulomas and crypt abscesses; these clinical and histopathologic findings resemble Crohn disease (Chapter 132). Late complications include rectal strictures, anal fistulas, and lymphorrhoids (perianal growths of lymphatic tissue).

Natural History

The natural history of untreated genital C. trachomatis infection has not been fully elucidated. Outcomes of untreated genital chlamydia include spontaneous resolution (i.e., immune-mediated clearance) and persisting infection; the latter may be subclinical or may progress to clinical manifestations (e.g., urethritis, cervicitis), which can remain uncomplicated or lead to a complication (e.g., PID). On the basis of limited evidence in subjects with chlamydia detected initially by a screening test who returned within weeks to months for treatment and were retested, anywhere from 10 to 40% of infections resolve spontaneously before treatment. Some patients with persisting infection develop clinical manifestations before returning for treatment, including a small proportion of females (up to 4%) who develop symptomatic PID. Other sparse data in females suggest that up to 50% of genital chlamydial infections resolve after a year, but a small percentage (1 : 16) or IgG (>1 : 512) titer. However, serology is limited by its specificity, reproducibility, and clinical correlation. C. pneumoniae can be isolated in cell culture, but culture is technically challenging and time-consuming. Antigen detection with use of fluorescent monoclonal antibodies has a lower sensitivity than culture and is also technically challenging. C. pneumoniae PCR is more sensitive than culture, and real-time PCR appears to have advantages over conventional PCR. Although there are still issues involving the standardization of PCR methods for C. pneumoniae detection, PCR is promising and will likely become the test of choice.

mild symptomatic illness manifested by ruffled feathers, anorexia, shivering, dyspnea, diarrhea, or depression. Infected birds shed C. psittaci in urine, feces, or secretions from their beaks or eyes. Their feathers and surrounding environment become contaminated. Transmission to humans is primarily by inhalation of aerosolized bird secretions or dust. Infected birds may shed the organism for months. Person-to-person transmission rarely occurs. Human psittacosis is a rare infection, due in part to antibiotic-laced bird feed and a mandated quarantine for imported birds. The number of cases of psittacosis in the United States has been stable for the past 10 years, with fewer than 50 confirmed cases reported annually; a larger number of cases are reported but not confirmed. An outbreak of psittacosis occurred in 2014 in New South Wales from a novel source, exposure to abnormal equine fetal membranes. This suggests that horse exposure may be a potential risk factor for psittacosis and highlights the potential for humans to acquire the infection from other infected mammalian hosts.  



TREATMENT  C. pneumoniae is susceptible to tetracyclines, macrolides, and fluoroquinolones. Treatment trials using culture have demonstrated that select macrolide and fluoroquinolone regimens eradicate C. pneumoniae in approximately 70 to 85% of subjects with pneumonia. The clinical response to treatment may be slow, and some patients may need retreatment. The suggested treatment duration for most regimens is typically 10 to 14 days, except that shorter courses may be effective for azithromycin (10 mg/kg on day 1, followed by 5 mg/kg during the next 4 days; up to 1.5 g orally during 5 days). Chronic C. pneumoniae infections may require even longer courses of treatment (e.g., 6 weeks), and macrolides are suggested in this setting. Protective immunity after C. pneumoniae infection may not occur, and therefore reinfection is common.

Chlamydia psittaci  

EPIDEMIOLOGY

C. psittaci naturally infects a variety of mammals and birds. C. psittaci strains appear to be host specific, and most human infections are linked to contact with an infected bird.17 C. psittaci infection in humans is termed psittacosis, in part because exposure to psittacine birds (parrots, parakeets, and budgerigars) is commonly implicated in infections. However, because human cases have been linked to exposure to finches, pigeons, pheasants, ducks, turkeys, chickens, seagulls, and other birds, ornithosis may be a more appropriate term. Psittacosis disease in birds ranges from an asymptomatic carriage state to a

CLINICAL MANIFESTATIONS

Psittacosis initially involves the lungs and then spreads to the reticuloendothelial system. The clinical spectrum of infection ranges from asymptomatic to fulminant, and clinical manifestations may resemble several other nonspecific febrile systemic illnesses, including Q fever, typhoid fever, and legionnaires disease.18 After an incubation period of 5 to 14 days, some patients may present with a nonspecific virus-like illness or a mononucleosis-like syndrome. The presentation most suggestive of psittacosis is an acute febrile atypical pneumonia.19 Patients initially have an abrupt onset of shaking chills and a temperature as high as 40.5° C. Temperature-pulse dissociation (i.e., elevated temperature with a normal pulse) may occur. Constitutional symptoms, including headache, myalgias, and arthralgias, are prominent. A cough, usually nonproductive, appears early in the illness and may accompany chest pain, which is usually nonpleuritic. Auscultation may be normal or reveal bilateral crackles. Chest radiograph findings are usually more dramatic than lung examination findings; the most common finding is single lower lobe consolidation, but multiple localized bronchopneumonic patches, diffuse ground-glass changes, and a miliary pattern have been described. Small pleural effusions may be seen. In contrast to C. pneumoniae pneumonia, psittacosis is more severe, with high fever and absent or minimal upper respiratory complaints. Extrapulmonary findings frequently occur in psittacosis. Splenomegaly is common. A faint erythematous, blanching, maculopapular rash (Horder spots), resembling the rose spots of typhoid fever, can occur, as can erythema nodosum. Signs of hepatitis, endocarditis (culture negative), pericarditis, myocarditis, meningoencephalitis, hemolytic anemia, or disseminated intravascular coagulation may be noted. C. psittaci infection has also been associated with nongastrointestinal extranodal marginal zone lymphomas of mucosa-associated lymphoid tissue, including ocular and central nervous system sites.

DIAGNOSIS

Psittacosis should be suspected in patients with a febrile illness (especially atypical pneumonia) who report exposure to sick or imported birds or who have regular exposure to birds, including bird owners, pet shop workers, veterinarians, zookeepers, and poultry processing plant workers. The diagnosis can be made with serology or by isolating the organism in cell culture. C. psittaci is a biocontainment level 3 agent because of its stability in the environment and aerosol transmission. Because laboratory-acquired C. psittaci infections are well documented, culture is discouraged and serology is preferred. If culture is attempted, laboratory staff should be notified in advance so that appropriate precautions can be taken. A serologic diagnosis of psittacosis is made by demonstrating (1) a four-fold or greater rise in CF or MIF antibody against C. psittaci to a titer of at least 1 : 32 from acute to convalescent sera collected at least 2 weeks apart (3 to 6 weeks is recommended) or (2) an IgM titer of 1 : 16 or greater against C. psittaci by MIF.

TREATMENT  Untreated psittacosis can be fatal, but mortality is rare with prompt antimicrobial treatment. Because of the delay in laboratory diagnosis of psittacosis, empirical therapy should be provided on the basis of clinical suspicion. C. psittaci susceptibility has been demonstrated to tetracyclines, macrolides, and newer generation fluoroquinolones. The recommended treatment regimen, based on clinical experience, is either tetracycline 500 mg four times a day or doxycycline 100 mg twice a day for 10 to 21 days. Further studies on the clinical efficacy of azithromycin and fluoroquinolones are needed. The initial treatment response

CHAPTER 302  Diseases Caused by Chlamydiae  

Association with Atherosclerosis

There is evidence that C. pneumoniae may cause or contribute to atherosclerosis and cardiovascular disease. Seroepidemiologic studies have demonstrated a higher prevalence of C. pneumoniae antibodies in patients with atherosclerosis, although the serologic tests used and seropositivity criteria varied across studies. In addition, C. pneumoniae has been identified in atherosclerotic plaques by culture, immunohistochemical staining, and PCR. A recent meta-analysis reported that C. pneumoniae IgA seropositivity, along with high-sensitivity C-reactive protein, fibrinogen, and IL-6, may be predictors of C. pneumoniae– associated atherosclerotic cardiovascular risk. Animal studies have demonstrated that C. pneumoniae infection can initiate or enhance the progression of atherosclerosis, and sparse animal studies have demonstrated that antibiotics prevent atherosclerosis. However, large-scale randomized treatment trials have failed to show that C. pneumoniae treatment prevents cardiovascular events or mortality; some of these studies differed in terms of study populations, macrolide used, and length of treatment. Currently, antibiotics are not recommended for the prevention of adverse atherosclerotic cardiovascular events.

Association with Asthma

Epidemiologic and clinical studies have demonstrated an association between C. pneumoniae infection and asthma, both acute exacerbations and more severe

1982.e1

chronic asthma. It is thought that persistent C. pneumoniae infection contributes to airway inflammation and hyperresponsiveness. Select randomized clinical trials have demonstrated improvement in asthma disease (symptoms, inflammatory markers, or peak expiratory flow) after antimicrobial treatment of presumed or proven C. pneumoniae infection, especially during acute exacerbations. However, it is difficult to determine how much of the improvement may have been due to the immunomodulatory effects of the antibiotics rather than their antimicrobial activity. Although the role of antibiotic treatment for C. pneumoniae infection in patients with asthma remains somewhat controversial, treatment for C. pneumoniae infection is currently recommended for patients with acute exacerbations of asthma and clinical evidence of infection.

Association with Chronic Neurologic Diseases

C. pneumoniae infection has been implicated in the pathogenesis of multiple sclerosis and Alzheimer disease on the basis of some studies demonstrating detection of the organism at a higher frequency in affected neurologic tissue. However, to date, the studies are contradictory, and a causal relationship between C. pneumoniae infection and either multiple sclerosis or Alzheimer disease has not been established.

can be dramatic, with defervescence and marked clinical improvement within 24 to 48 hours. Full recovery may take several weeks, and relapse or reinfection can occur. Endocarditis treatment includes prolonged antibiotic therapy and consideration of valve replacement.



PREVENTION

Epidemic psittacosis is preventable by a 30-day period of quarantine for all imported psittacine birds and their treatment with feed containing chlortetracycline. The U.S. Department of Agriculture recommends extending treatment for an additional 15 days after quarantine. Prevention of epidemic and endemic psittacosis also relies on avoidance of or protection from exposure to dust or body secretions from birds or their living areas as well as avoidance of the handling of sick birds. Environmental sanitation is another important preventive measure, considering the organism’s resistance to drying.

  Grade A References A1. Evans JR, Solomon AW. Antibiotics for trachoma. Cochrane Database Syst Rev. 2011;3:CD001860. A2. Páez-Canro C, Alzate JP, Gonzalez LM, et al. Antibiotics for treating urogenital Chlamydia trachomatis infection in men and non-pregnant women. Cochrane Database Syst Rev. 2019;1:CD010871. A3. Geisler WM, Uniyal A, Lee JY, et al. Azithromycin versus doxycycline for urogenital Chlamydia trachomatis infection. N Engl J Med. 2015;373:2512-2521.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 302  Diseases Caused by Chlamydiae  

GENERAL REFERENCES 1. Pettengill MA, Abdul-Sater A, Coutinho-Silva R, et al. Danger signals, inflammasomes, and the intricate intracellular lives of chlamydiae. Biomed J. 2016;39:306-315. 2. Lansingh VC. Trachoma. BMJ Clin Evid. 2016;2016:1-14. 3. Habtamu E, Wondie T, Aweke S, et al. Predictors of trachomatous trichiasis surgery outcome. Ophthalmology. 2017;124:1143-1155. 4. Centers for Disease Control and Prevention. Sexually transmitted disease surveillance 2017. https:// www.cdc.gov/std/stats17/2017-STD-Surveillance-Report_CDC-clearance-9.10.18.pdf. Accessed May 1, 2019. 5. Rönn MM, Tuite AR, Menzies NA, et al. The impact of screening and partner notification on chlamydia prevalence and numbers of infections averted in the United States, 2000-2015: evaluation of epidemiologic trends using a pair-formation transmission model. Am J Epidemiol. 2019;188: 545-554. 6. Moi H, Blee K, Horner PJ. Management of non-gonococcal urethritis. BMC Infect Dis. 2015;15: 1-7. 7. Taylor SN. Epididymitis. Clin Infect Dis. 2015;61(suppl 8):S770-S773. 8. Hoenderboom BM, van Benthem BHB, van Bergen J, et al. Relation between Chlamydia trachomatis infection and pelvic inflammatory disease, ectopic pregnancy and tubal factor infertility in a Dutch cohort of women previously tested for chlamydia in a chlamydia screening trial. Sex Transm Infect. 2019. [Epub ahead of print.] 9. Price MJ, Ades AE, Welton NJ, et al. Proportion of pelvic inflammatory disease cases caused by Chlamydia trachomatis: consistent picture from different methods. J Infect Dis. 2016;214:617-624.

1983.e1

10. Olson-Chen C, Balaram K, Hackney DN. Chlamydia trachomatis and adverse pregnancy outcomes: meta-analysis of patients with and without infection. Matern Child Health J. 2018;22:812-821. 11. Stoner BP, Cohen SE. Lymphogranuloma venereum 2015: clinical presentation, diagnosis, and treatment. Clin Infect Dis. 2015;61(suppl 8):S865-S873. 12. Wiesenfeld HC. Screening for Chlamydia trachomatis infection in women. N Engl J Med. 2017;376: 765-773. 13. Workowski KA, Bolan GA. Sexually transmitted diseases treatment guidelines, 2015. MMWR Recomm Rep. 2015;64:1-137. 14. Kong FY, Tabrizi SN, Fairley CK, et al. The efficacy of azithromycin and doxycycline for the treatment of rectal chlamydia infection: a systematic review and meta-analysis. J Antimicrob Chemother. 2015;70:1290-1297. 15. de Vries HJ, Zingoni A, Kreuter A, et al. 2013 European guideline on the management of lymphogranuloma venereum. J Eur Acad Dermatol Venereol. 2015;29:1-6. 16. Clemmons NS, Jordan NN, Brown AD, et al. Outbreak of Chlamydia pneumoniae infections and X-ray-confirmed pneumonia in army trainees at Fort Leonard Wood, Missouri, 2014. Mil Med. 2019. [Epub ahead of print.] 17. de Gier B, Hogerwerf L, Dijkstra F, et al. Disease burden of psittacosis in the Netherlands. Epidemiol Infect. 2018;146:303-305. 18. Cunha BA, Cunha CB. Legionnaire’s disease and its mimics: a clinical perspective. Infect Dis Clin North Am. 2017;31:95-109. 19. Hogerwerf L, DE Gier B, Baan B, et al. Chlamydia psittaci (psittacosis) as a cause of community-acquired pneumonia: a systematic review and meta-analysis. Epidemiol Infect. 2017;145: 3096-3105.

1983.e2

CHAPTER 302  Diseases Caused by Chlamydiae  

REVIEW QUESTIONS 1. Which of the following is not a recommended strategy for the elimination of trachoma and its complications? A . Surgery to correct trichiasis B. Mass antibiotic treatments within a trachoma-endemic community C. Face-washing and good hygiene D. Achieving better sanitary conditions in the environment E. Education to promote safer sexual practices Answer: E  The World Health Organization is committed to the elimination of trachoma and its complications by 2020 and recommends that endemic countries adopt the SAFE strategy: surgery (for trichiasis), antimicrobials (periodic community-wide treatment), facial cleanliness, and environmental sanitation improvement. Trachoma is spread by direct contact with fingers or fomites (e.g., washcloths, handkerchiefs) contaminated with eye discharge from an infected person or by eye-seeking flies, not through sexual transmission. 2. Routine annual screening for genital chlamydial infection is recommended by the Centers for Disease Control and Prevention (CDC) for the following patient group: A . All sexually active women B. All sexually active men C. Sexually active women younger than 25 years D. Sexually active men 25 years and younger E. None of the above Answer: C  Routine genital chlamydia screening (i.e., testing of asymptomatic persons for chlamydia) in women has been shown to reduce the rate of pelvic inflammatory disease. Surveillance data from the CDC have shown that the majority of chlamydia cases reported each year are in women younger than 25 years. Therefore, the CDC recommends routine annual screening for genital chlamydial infection in women within this age group. Screening in women 25 years or older is recommended only if risk factors (e.g., new sexual partner, multiple sexual partners) are present. Routine chlamydia screening is not recommended for men; however, screening should be considered for men in venues with a high prevalence of chlamydia (e.g., STD clinics). 3. The recommended treatment for lymphogranuloma venereum (LGV) infection is: A . Doxycycline 100 mg twice daily orally for 7 days B. Doxycycline 100 mg twice daily orally for 21 days C. Azithromycin 1 g orally single-dose therapy D. Erythromycin 500 mg four times a day for 21 days E. Ciprofloxacin 500 mg twice daily for 7 days Answer: B  Chlamydia trachomatis strains of the LGV serovar/genotype cause more invasive disease than the non-LGV strains do, and therefore a course of doxycycline 100 mg twice daily for 21 days is recommended. There is insufficient clinical experience with macrolides to recommend them as a first-line agent for LGV. Some experts do recommend azithromycin 1 g weekly for 3 weeks as an alternative to the 3-week doxycycline regimen, but a single dose of azithromycin is believed to be insufficient to reliably cure invasive LGV infections. Ciprofloxacin has poor in vitro activity against C. trachomatis and is not recommended for treatment of LGV or non-LGV C. trachomatis infections.

4. Which of the following statements regarding Chlamydia pneumoniae is most accurate? A . The seroprevalence of C. pneumoniae in adults in the United States is estimated to be about 15%. B. Pneumonia caused by C. pneumoniae usually is manifested as an acuteonset illness with cough productive of purulent sputum. C. C. pneumoniae causes upper respiratory tract infections, including pharyngitis and bronchitis. D. Penicillin is a highly effective treatment for C. pneumoniae respiratory infections. E. Large clinical trials have consistently demonstrated a reduction in adverse cardiovascular events in patients receiving antibiotic treatment directed against C. pneumoniae. Answer: C  The majority of adults in the United States and other developed nations are seropositive for C. pneumoniae. It is well known that C. pneumoniae can cause upper and lower respiratory tract infections. Pneumonia caused by C. pneumoniae usually develops gradually with a nonproductive cough, referred to as a walking pneumonia, in contrast to pneumococcal pneumonia, which is more likely to be acute in onset with a productive cough. C. pneumoniae respiratory infections are most effectively treated with an antibiotic from the tetracycline, macrolide, or respiratory fluoroquinolone (levofloxacin, moxifloxacin) class and are not adequately treated by a β-lactam antibiotic such as penicillin. Even though C. pneumoniae has been associated with atherosclerosis, large C. pneumoniae treatment trials have not demonstrated benefit in preventing adverse cardiovascular events. 5. Which of the following statements regarding Chlamydia psittaci is not true? A . C. psittaci infection has been associated with exposure to psittacine birds (parrots, parakeets, and budgerigars). B. C. psittaci infection has been associated with exposure to turkeys. C. Birds may be carriers of C. psittaci but do not develop clinical disease. D. C. psittaci infection can be manifested as an abrupt-onset febrile illness in humans. E. C. psittaci has been classified as a biocontainment level 3 agent. Answer: C  C. psittaci infection has been associated with exposure to psittacine birds and several other nonpsittacine birds, including turkeys. Whereas birds may be asymptomatic carriers of C. psittaci, they may also develop a mild symptomatic illness manifested by ruffled feathers, anorexia, shivering, dyspnea, diarrhea, or depression. C. psittaci infection can be manifested as an abruptonset febrile illness in humans with pulmonary and extrapulmonary manifestations. C. psittaci has been classified as a biocontainment level 3 agent because of its stability in the environment and aerosol transmission. Because laboratoryacquired C. psittaci infections are well documented, culture is discouraged and serology is preferred for diagnosis.

CHAPTER 303 Syphilis  

1983

structurally and biochemically similar to flagella and are in part responsible for the organism’s motility. It is possible to culture T. pallidum, but sustained in vitro cultivation is limited, and yields are very low. Culture is of no use in clinical practice. All isolates studied have been susceptible to penicillin and are antigenically similar. The only known natural hosts for T. pallidum are humans and certain monkeys and higher apes.  

303  SYPHILIS KHALIL G. GHANEM AND EDWARD W. HOOK, III



DEFINITION

Syphilis, which is a chronic infectious disease caused by the bacterium Treponema pallidum subspecies pallidum, is usually acquired by sexual contact with another infected individual. Syphilis is remarkable among infectious diseases for its large variety of clinical manifestations. If untreated, it progresses through primary, secondary, and tertiary stages. The early stages (i.e., primary and secondary), when lesions are present, are infectious. Spontaneous healing of early lesions occurs, followed by a long latent period. In about 30% of untreated patients, late disease of the heart, central nervous system (CNS), or other organs may develop years after the initial infection. Although the disease is less common now than previously, it remains a challenge to clinicians because of its protean manifestations, and it is of interest to biologists because of the prolonged, tenuous balance between the host and the invading spirochete.1

The Pathogen

The causative agent of syphilis, T. pallidum subspecies pallidum, is closely related to other pathogenic spirochetes (Chapter 304), including those causing yaws (T. pallidum subspecies pertenue) and pinta (Treponema carateum). T. pallidum is a thin, helical bacterium approximately 0.15 µm wide and 6 to 15 µm long. The organism has 6 to 14 spirals and is tapered on either end. It is too thin to be seen by ordinary Gram stain microscopy but can be visualized in wet mounts by dark-field microscopy or in fixed specimens by silver stain or fluorescent antibody methods. Unlike most bacteria, which have protein-rich outer membranes, the T. pallidum outer membrane appears to be composed of predominantly phospholipids, with few surface-exposed proteins.2 It has been hypothesized that because of this structure, syphilis can progress despite the brisk antibody response to non–surface-exposed internal antigens, which is the basis for serologic tests for the diagnosis and management of syphilis. Between the outer membrane and the peptidoglycan cell wall are six axial fibrils; three are attached at each end, and they overlap in the center of the organism. They are

EPIDEMIOLOGY

With the exception of congenital syphilis, syphilis is acquired almost exclusively by intimate contact with the infectious lesions of primary or secondary syphilis (e.g., chancres, mucous patches, condylomata lata). Disease is usually acquired through sexual intercourse, including anogenital and orogenital intercourse. Health care workers are sometimes unexpectedly infected during examination of patients with infectious lesions. Infection by contact with fomites is extremely uncommon. Before the advent of modern blood banking techniques, syphilis was occasionally transmitted through the transfusion of blood from persons with T. pallidum bacteremia, and occasional parenteral transmission still occurs as a result of the sharing of contaminated needles. Syphilis is most common in large cities among sexually active individuals. The highest rate is found in men between the ages of 20 and 29 years. In the United States, it is most prevalent in the West and in the South.3 Syphilis spares no class, race, or group. U.S. syphilis rates are about five-fold greater in African Americans than in non-Hispanic whites. In 2017, more than 85% of reported early syphilis occurred in men, and nearly 60% of total early cases occurred among men who acknowledged sex with other men. A traditional cornerstone of syphilis control has been the epidemiologic investigation and treatment of sexual contacts of patients with primary or secondary lesions and patients with early latent disease. Patients with primary and secondary syphilis name, on average, nearly three different sexual contacts within the previous 90 days. As syphilis has become associated with drug use and anonymous sex, epidemiologic investigations have become less efficacious. The incidence of syphilis has generally declined worldwide for more than 100 years, with the exception of periods of war or social upheaval. With the introduction of penicillin, there was a rapid decline in primary and secondary syphilis, to approximately 4 cases per 100,000 people in 1957. This decline was followed by reductions in federal expenditures for syphilis control, which resulted in resurgence of infectious primary and secondary syphilis in the United States; peaks of more than 12 cases per 100,000 people were attained several times from 1965 through the mid-1990s. During the past 40 years, syphilis epidemics have occurred serially in at least three U.S. population subgroups. In the 1970s and 1980s, men who had sex with other men accounted for a disproportionate number of the total cases of infectious syphilis. Similar trends occurred in other countries. Then, after a period of decline, U.S. syphilis rates nearly doubled from 1986 to 1990, with 50,578 cases reported in 1990 in an epidemic disproportionately affecting multiracial heterosexual men and women and occurring contemporaneously with an epidemic of crack cocaine use. After 1990, syphilis rates again declined; in 2001, there were 6103 cases of primary and secondary syphilis reported, one of the lowest numbers since 1959. Since 2001, syphilis rates have again begun to increase in men, and now especially men infected with HIV. Likewise, in the United Kingdom syphilis has been a resurgent sexually transmitted infection that is disproportionately diagnosed in patients with HIV, particularly in men who have sex with men.4 In 2017, the rate of reported primary and secondary syphilis in the United States was 9.5 cases per 100,000 population, more than triple the lowest-ever rate of 2.1 in 2000. During 2005 to 2017, primary and secondary syphilis rates increased among men of all ages, races, and ethnicities across all regions. Among women, rates increased during 2005 to 2008 and decreased during 2009 to 2013. From 2013 to 2017, rates among women increased 155.6% with a concomitant increase in cases of congenital syphilis. Patients with clinically evident late syphilis, particularly those with cardiovascular or gummatous syphilis, are becoming less common, perhaps as a result of the effectiveness of penicillin therapy for early syphilis. However, surveys indicate that there are still significant numbers of patients with untreated neurologic syphilis, especially in older age groups.

Natural Course of Untreated Syphilis

The incubation period from the time of exposure to development of the primary lesion averages approximately 21 days (range, 10 to 90 days). Initially, a painless papule develops at the site of inoculation and soon breaks down to form a clean-based ulcer—the chancre—with raised, indurated margins (Fig. 303-1A).

CHAPTER 303 Syphilis  

ABSTRACT

Syphilis, which is a chronic infectious disease caused by the bacterium Treponema pallidum subspecies pallidum, is usually acquired by sexual contact with another infected individual. Syphilis is remarkable among infectious diseases for its large variety of clinical manifestations. If untreated, it progresses through primary, secondary, and tertiary stages. The early stages (i.e., primary and secondary), when lesions are present, may be transmitted sexually. Spontaneous healing of early lesions occurs, followed by a long latent period. In about 30% of untreated patients, late disease of the heart, central nervous system (CNS), or other organs may develop years after the initial infection. Although the disease is less common now than previously, it remains a challenge to clinicians because of its protean manifestations.

KEYWORDS

syphilis neurosyphilis STD Treponema pallidum sexually transmitted infection aortitis tabes dorsalis chancre

1983.e3

1984

CHAPTER 303 Syphilis  

The chancre persists for 2 to 6 weeks and then heals spontaneously. Several weeks later, a secondary stage characterized by low-grade fever, headache, malaise, generalized lymphadenopathy, and a mucocutaneous rash typically develops. There may be involvement of visceral organs. The secondary eruption may occur while the primary chancre is still healing or up to several months after disappearance of the chancre. Secondary lesions also heal spontaneously within 2 to 6 weeks, and the infection then becomes latent. In more than 20% of patients with untreated latent syphilis, relapsing lesions later develop, similar to those of the secondary stage; rarely, the relapse takes the form of recurrence of the primary chancre. In the era before antibiotics, late, destructive tertiary lesions involving the eyes, the CNS, the heart, and other organs, including the skin, eventually developed in about a third of untreated patients. These lesions may occur a few years to as long as 25 years after infection. The incidence of late complications of untreated syphilis is currently unknown, but it seems to be less than that seen previously. Cases of gumma are now so rare as to be reportable.  

C

PATHOBIOLOGY

T. pallidum may penetrate through normal mucosal membranes and minor abrasions on epithelial surfaces. The first lesions appear at the site of direct, primary inoculation. The minimal number of treponemes needed to establish infection is not known but may be as low as one. Multiplication of organisms is slow, with a division time in rabbits of approximately 33 hours. The slow growth of treponemes in humans probably accounts in part for the protracted nature of the illness, the relatively long incubation period, and the need for relatively long duration of therapy. Syphilis is a systemic disease from the onset. Treponemes are capable of specific attachment to host cells, but it is not known whether attachment results in damage to the host cells. Most treponemes are found in the intercellular spaces, but they are occasionally seen within phagocytic cells. However, there is no evidence of prolonged intracellular survival of treponemes. T. pallidum is not known to produce toxins. The primary pathologic lesion of syphilis is a focal endarteritis with an increase in adventitial cells, endothelial proliferation, and the presence of an inflammatory cuff around affected vessels. Lymphocytes, plasma cells, and monocytes predominate in the inflammatory lesion, and polymorphonuclear cells are seen in some cases. The vessel lumen is frequently obliterated. With healing, there is considerable fibrosis. Treponemes may be seen in most early lesions of syphilis and in some of the late lesions, such as the meningoencephalitis of general paresis. Granulomatous reaction is common in secondary and late syphilis. The granulomas are histologically nonspecific, and cases of syphilis have been incorrectly diagnosed as sarcoidosis or other granulomatous diseases. Human inoculation studies suggest that the pathogenesis of the gumma, which is a granulomatous lesion, involves hypersensitivity to small numbers of virulent treponemes introduced into a previously sensitized host. Intracutaneous inoculation of partially purified antigens of T. pallidum into patients with syphilis in various stages has shown that delayed cellular hypersensitivity develops only late in secondary syphilis but is uniformly present in latent syphilis. There may be temporary hyporesponsiveness of lymphocytes to treponemal antigens in patients with primary and secondary syphilis. It is possible that the waxing and waning of lesions in early syphilis depend on the balance between the development of effective cellular immunity and the suppression of thymus-derived lymphocyte function. The host responds to infection by producing numerous antibodies; in some instances, circulating immune complexes may be formed as well. For example, nephrotic syndrome has occasionally been recognized in secondary syphilis, and renal biopsy specimens from such patients have shown membranous glomerulonephritis characterized by focal subepithelial basement membrane deposits containing immunoglobulin G, C3, and treponemal antibody.  

A

CLINICAL MANIFESTATIONS

Primary Syphilis

The typical lesion of primary syphilis, the chancre, is a painless, clean-based, indurated ulcer (Fig. 303-1A). The chancre starts as a papule, but then superficial erosion results in ulceration. The borders of the ulcer are raised, firm, and indurated. On occasion, secondary infections change the appearance and cause a painful lesion. Most chancres are single, but multiple ulcers are sometimes seen, particularly when skinfolds are apposed (i.e., kissing chancres). An untreated chancre heals in several weeks and leaves a faint scar. The chancre is usually associated with regional adenopathy, which may be unilateral or bilateral. The regional nodes are movable, discrete, and rubbery. If the chancre

B

D

FIGURE 303-1.  Syphilis lesions. A, Chancre in primary syphilis. B, Palmar lesions of a coppery color in secondary syphilis. C, Mucous patch in secondary syphilis. D, Condylomata lata in secondary syphilis. (A, C, and D from Forbes CD, Jackson WF. Color Atlas and Text of Clinical Medicine. 3rd ed. London: Mosby; 2003. B from Habif TP, Cambell JI, Quitadamo MJ, et al. Skin Disease: Diagnosis and Treatment. St. Louis: Mosby; 2001.)

occurs in the cervix or the rectum, the affected regional iliac nodes are not palpable. Chancres can occur at any site of potential inoculation by direct contact, with most occurring in anogenital locations. Chancres may also be seen in the pharynx, on the tongue, around the lips, on the fingers, on the nipples, and in other diverse areas. The morphology depends in part on the area of the body where they occur and on the host’s immune response. Chancres in previously infected individuals may be small and remain papular. Chancres of the finger may appear more erosive and can be quite painful. Chancres of the anal canal may be missed in men who have sex with men unless a careful examination is undertaken.

Secondary Syphilis

Between 4 and 8 weeks after the appearance of the primary chancre, signs and symptoms of secondary syphilis typically develop. Symptoms may include malaise, fever, headache, sore throat, and other systemic complaints. Most patients have generalized lymphadenopathy, including involvement of the epitrochlear nodes. Approximately 30% of patients have evidence of a healing chancre, although many patients (including a disproportionate number of women and of men who have sex with men) give no history of a primary lesion. At least 80% of patients with secondary syphilis have cutaneous or mucocutaneous lesions at some point in their illness. The diagnosis is frequently first suspected on the basis of the cutaneous eruption. The rash is often minimally symptomatic, and many patients with late syphilis do not recall primary or secondary lesions. The rashes are varied in appearance but have certain characteristic features. The lesions are usually widespread, are symmetrical in distribution, and are frequently pink, coppery, or dusky red (particularly the earliest macular lesions). They are generally nonpruritic, although occasional exceptions have been reported, and they are rarely vesicular or bullous in adults. They are indurated, except for the very earliest macular lesions, and frequently have a superficial scale (i.e., papulosquamous lesions). The lesions tend to be polymorphic and rounded, and on healing, they may leave residual pigmentation or depigmentation. They may be faint and difficult to visualize, particularly on dark-skinned individuals.

1985

CHAPTER 303 Syphilis  

The earliest pink macular lesions are typically seen on the trunk, with later spread to the rest of the body. The face is often spared, except around the mouth. Subsequently, a papular rash appears that is usually generalized but is marked on the palms and soles (Fig. 303-1B). These rashes are often associated with a superficial scale and may be hyperpigmented. When the rash occurs on the face, it may be pustular and resemble acne vulgaris. On occasion, the scaling may be so prominent that it resembles psoriasis. Ulceration may occur and produce lesions resembling ecthyma. In malnourished or debilitated patients, extensive and destructive ulcerative lesions with a heaped-up crust may occur, the so-called rupial lesions. Lesions around the hair follicles may result in patchy alopecia of the beard or scalp. Ringed or annular lesions may occur, especially around the face and particularly on dark-skinned individuals. A lesion at the angle of the mouth or the corner of the nose may have a central linear erosion, the so-called split papule. The palate and pharynx may be inflamed. In approximately 30% of secondary syphilis patients, so-called mucous patches (Fig. 303-1C) develop; these slightly raised, oval areas are covered by a grayish white membrane that, when raised, reveals a pink base that does not bleed. These lesions may be seen on the genitalia, in the mouth, or on the tongue. In warm, moist areas such as the perineum, large, pale, flat-topped papules may coalesce to form condylomata lata (Fig. 303-1D). Papules may also be seen in the axilla and rarely occur in a generalized form. These papules are not to be confused with the common venereal warts (i.e., condylomata acuminata), which are small, often multiple, and more sharply raised than condylomata lata. Like mucous patches, condylomata lata are highly infectious. Other manifestations of secondary syphilis include hepatitis, which has been reported in up to 10% of patients in some series. Jaundice is rare, but an elevated alkaline phosphatase level is common. Liver biopsy reveals small areas of focal necrosis and mononuclear infiltrate or periportal vasculitis. Spirochetes can often be visualized with silver stains. Periostitis with widespread lytic lesions of bone has been reported occasionally; bone scanning appears to be a sensitive test for early syphilitic osteitis. An immune complex type of nephropathy with transient nephrotic syndrome has been documented rarely. There may be iritis or an anterior uveitis. Between 10 and 30% of patients have pleocytosis in cerebrospinal fluid (CSF), but symptomatic meningitis is seen in less than 1% of patients. Symptomatic gastritis may occur.

Relapsing Syphilis

After resolution of the primary or secondary skin lesions, 20 to 30% of patients experience cutaneous recurrences. Recurrent lesions may be fewer or more firmly indurated than the initial lesions. Like the typical lesions of primary or secondary syphilis, they are infectious for exposed sexual partners.

Latent Syphilis

By definition, latent syphilis is the stage at which there are no clinical signs of syphilis. Latency, which begins when the first attack of secondary syphilis has passed and may last for a lifetime, is usually detected by reactive serologic tests for syphilis (see Diagnosis). Congenital syphilis must also be excluded before the diagnosis of latent syphilis can be made. Patients may or may not have a clinical history of earlier primary or secondary syphilis manifestations. Latency has been divided into two stages: early and late. Most infectious relapses occur in the first year, and epidemiologic evidence shows that the most infectious period is during the first year of infection. Early latency is therefore defined as the first year after resolution of the primary or secondary lesions or as a newly reactive serologic test response for syphilis in an otherwise asymptomatic individual who has had a negative serologic test result within the preceding year. Late latent syphilis, or, more accurately, latent syphilis of unknown duration, is ordinarily not infectious, except for pregnant women, who can transmit infection to the fetus despite long-standing infection.

Late Syphilis

Late or tertiary syphilis (Table 303-1) is usually slowly progressive, although certain neurologic syndromes may have a sudden onset because of endarteritis and CNS thrombosis. Late syphilis is not infectious through sexual contact. Any organ of the body may be involved, but three main types of disease can be distinguished: late benign (gummatous), cardiovascular, and neurosyphilis.

Late Benign Syphilis

In the penicillin era, gummas are rare. They typically develop 1 to 10 years after initial infection and may involve any part of the body. Although gummas may be destructive, they respond rapidly to treatment and are therefore relatively benign. On histologic evaluation, the gumma is a granuloma.

TABLE 303-1 NEWLY DIAGNOSED TERTIARY SYPHILIS IN 105 PATIENTS IN DENMARK, 1961-1970 TYPE OF TERTIARY SYPHILIS

NO. OBSERVED*

Neurosyphilis Asymptomatic Tabes dorsalis General paresis Meningovascular Optic atrophy

72 45 11 13 1 2

Cardiovascular syphilis Aortic insufficiency Aortic aneurysm Uncomplicated aortitis†

44 16 13 15

Late benign syphilis (gumma)

4

*Some patients had more than one form of late syphilis. † Autopsy diagnoses only.

Gummas may be solitary or multiple and most often come to medical attention as space-occupying lesions. They are usually asymmetrical and are often grouped. Gummas may start as a superficial nodule or as a deeper lesion that breaks down to form punched-out ulcers. They are ordinarily indolent, slowly progressive, and indurated on palpation. Cutaneous gummas may resemble other chronic granulomatous ulcerative lesions caused by tuberculosis, sarcoidosis, leprosy, and other deep fungal infections. Precise histologic diagnosis may not be possible. However, syphilitic gummas are the only such lesions to heal dramatically with penicillin therapy. Gummas may also involve deep visceral organs, particularly the respiratory tract, gastrointestinal tract, and bones. In addition, they may involve the larynx or the pulmonary parenchyma. Gummas of the stomach may masquerade as carcinoma of the stomach or lymphoma. Gummas of the liver were once the most common form of visceral syphilis and often manifested as hepatosplenomegaly and anemia and occasionally as fever and jaundice. Skeletal gummas typically produce lesions in the long bones, skull, and clavicle; a characteristic symptom is nocturnal pain. Radiologic abnormalities, when present, include periostitis and lytic or sclerotic destructive osteitis.

Cardiovascular Syphilis

The primary cardiovascular complications of syphilis are aortic insufficiency (Chapter 66) and aortic aneurysm (Chapter 69), usually of the ascending aorta. Less commonly, other large arteries may be affected, and involvement of the coronary ostia rarely results in coronary insufficiency. All these complications are caused by obliterative endarteritis of the vasa vasorum, with resultant damage to the intima and media of the great vessels. This damage results in dilation of the ascending aorta, but the valve cusps remain normal. An aneurysm occasionally is manifested as a pulsating mass bulging through the anterior chest wall. Syphilitic aortitis may also involve the descending aorta proximal to the renal arteries. Cardiovascular syphilis usually begins within 5 to 10 years of the initial infection but may not be manifested clinically until 20 to 30 years later. Cardiovascular syphilis does not occur after congenital infection, a phenomenon that remains unexplained. Asymptomatic aortitis is best diagnosed by visualizing linear calcifications in the wall of the ascending aorta by radiography. The signs of syphilitic aortic insufficiency are the same as for aortic insufficiency of other causes. In aortic insufficiency resulting from dilation of the aortic ring, the decrescendo murmur is often loudest along the right sternal margin. Syphilitic aneurysms may be fusiform but are more typically saccular and do not lead to aortic dissection. Between 10 and 20% of patients with cardiovascular syphilis have coexistent neurosyphilis.

Neurosyphilis

CNS involvement occurs throughout the natural history of syphilis. Neurosyphilis5 can be divided into five groups: asymptomatic, syphilitic meningitis, meningovascular syphilis, tabes dorsalis, and general paresis. Asymptomatic neurosyphilis can occur at any time, whereas syphilitic meningitis is most common during the secondary stage of infection. Meningovascular syphilis, tabes dorsalis, and general paresis are typically manifestations of late syphilis. The divisions are not absolute, and overlap between syndromes is typical.

1986

CHAPTER 303 Syphilis  

Syphilitic Meningitis

Acute to subacute aseptic meningitis can occur at any time after the primary stage, but it usually occurs within the first year of infection. It frequently involves the base of the brain and may result in unilateral or bilateral cranial nerve palsies. Mild aseptic meningitis may be relatively common in patients with early syphilis, but severe disease occurs in only about 1.5% of untreated patients. Syphilitic meningitis typically resolves without treatment. Meningovascular Syphilis

Some patients have sufficient endarteritis and perivascular inflammation to result in cerebrovascular thrombosis and infarction, generally 5 to 10 years after the initial infection. However, case reports suggest that in syphilis patients with coexistent HIV infection, meningovascular syphilis may be manifested earlier or may be a manifestation of treatment failure. Patients frequently have associated aseptic meningitis. Most cerebrovascular accidents are not caused by syphilitic arteritis, even in patients with a reactive serologic test result for syphilis. However, syphilis should be considered a potential cause in relatively young patients with a history of syphilis and without other risk factors for cerebrovascular disease. Tabes Dorsalis

Tabes dorsalis, which appears to be far less common than in the pre-penicillin era, is a slowly progressive, degenerative disease that involves the posterior columns and posterior roots of the spinal cord and results in progressive loss of peripheral reflexes, impairment of vibration and position sense, and progressive ataxia. Sensory changes may lead to chronic destructive changes in the large joints of the affected limbs in advanced cases (i.e., Charcot joints). Urinary incontinence and impotence are common. Sudden and severely painful crises of uncertain origin are a characteristic part of the syndrome. Severe, sharp abdominal pain may lead to exploratory surgery. Optic atrophy is seen in 20% of cases. In 90% of patients, the pupils are bilaterally small and fail to constrict further in response to light, but they do respond normally to accommodation (i.e., Argyll Robertson pupils). The onset of tabes dorsalis is usually first noticed 20 to 30 years after the initial infection. Its cause is unclear, and spirochetes cannot be demonstrated in the posterior column or dorsal root. General Paresis

This form of neurosyphilis is a chronic meningoencephalitis resulting in the gradual and progressive loss of cortical function. It typically occurs 10 to 20 years after the initial infection. On pathologic examination, there is a perivascular and meningeal chronic inflammatory reaction, with thickening of the meninges, granular ependymitis, degeneration of the cortical parenchyma, and abundant spirochetes in tissues. In its early stages, general paresis results in nonspecific symptoms of early dementia, such as irritability, fatigue, headaches, forgetfulness, and personality changes. Later, there is impaired memory, defective judgment, lack of insight, confusion, and often depression or marked elation. Patients may be delusional, and seizures sometimes occur. There may also be loss of other cortical functions, including paralysis or aphasia. Physical signs are primarily those of the altered mental status. Cranial nerve palsies are uncommon, and optic atrophy is rare. The complete Argyll Robertson pupil is also uncommon, but irregular or otherwise abnormal pupils are not infrequent. Peripheral reflexes are often somewhat increased.

Ocular and Otic Syphilis

The eyes and ears may be affected during any stage of syphilis. Panuveitis is the most common ocular manifestation reported, but any portion of the eye may be affected.6 Consequently, the clinical manifestations of ocular syphilis are broad. In half the cases, bilateral ocular involvement is noted. In 70% of cases, concomitant cerebrospinal fluid abnormalities consistent with neurosyphilis are detected. Otosyphilis presents with sudden fluctuating or persistent sensorineural hearing loss or vestibular symptoms. Common complaints, in addition to hearing loss, include tinnitus and vertigo. In half the cases, both ears are affected. Unlike ocular syphilis, over 90% of persons with otosyphilis will have a normal cerebrospinal fluid examination.

Syphilis-HIV Interactions

Syphilis, like other genital ulcer diseases, is associated with a three- to five-fold increased risk for acquisition of HIV infection. Presumably, genital ulcers act as portals of entry through which HIV may more readily infect exposed individuals. As a result, HIV serologic testing 3 months after a diagnosis of syphilis

is recommended for all patients. Conversely, in individuals with HIV infection who acquire syphilis, the natural history of the infection may be modified.7 HIV-infected syphilis patients are somewhat more likely than non–HIV-infected patients to present initially with secondary syphilis. HIV-infected secondary syphilis patients are also more likely than HIV-negative secondary syphilis patients to have coexistent chancres, suggesting that the healing of chancres is delayed or the appearance of secondary manifestations is accelerated in the presence of HIV coinfection. Coinfected patients may be at higher risk of developing neurologic complications, particularly early neurosyphilis.

Congenital Syphilis

Congenital syphilis results from the transplacental, hematogenous spread of syphilis from the mother to the fetus.8 In 2017, 918 cases of congenital syphilis were reported in the United States with an estimated 660,000 annual cases worldwide.9 A serologic test for syphilis should be performed in all expectant mothers at the beginning of pregnancy and should be repeated during the third trimester in women living in areas where syphilis is relatively common.10,11 The risk for fetal infection is greatest in the early stages of untreated maternal syphilis and declines slowly thereafter, but the untreated mother can infect her fetus during at least the first 5 years of her infection. Adequate treatment of the mother before the 16th week of pregnancy usually prevents clinical illness in the neonate. Later treatment may not prevent late sequelae of the disease in the child. Untreated maternal infection may result in stillbirth, neonatal death, prematurity, or syndromes of early or late congenital syphilis in surviving infants. Manifestations of early congenital syphilis are often seen in the perinatal period but may not develop until the infant has been discharged from the hospital. The disease resembles secondary syphilis in adults, except that the rash may be vesicular or bullous. The child often has rhinitis, hepatosplenomegaly, hemolytic anemia, jaundice, and pseudoparalysis (i.e., immobility of one or more extremities) as a result of painful osteochondritis. Late congenital syphilis is defined as congenital syphilis diagnosed more than 2 years after birth. The disease may remain latent, with no manifestations of late damage. Cardiovascular alterations have not been observed in patients with congenital syphilis. Neurologic manifestations are common and may include eighth cranial nerve deafness and interstitial keratitis. Periostitis may result in prominent frontal bones of the skull, depression of the bridge of the nose (saddle nose), poor development of the maxilla, and anterior bowing of the tibia (saber shins). There may be late-onset arthritis of the knees (Clutton joints). The permanent dentition may show characteristic abnormalities known as Hutchinson teeth; the upper central incisors are widely spaced, centrally notched, and tapered in the manner of a screwdriver. The molars may show multiple poorly developed cusps (mulberry molars).  

DIAGNOSIS

Dark-Field Examination

The most definitive means of syphilis diagnosis is finding typical spirochetes in lesions of early acquired or congenital syphilis. Dark-field examination is often positive in cases of primary syphilis and in patients with the moist mucosal lesions of secondary and congenital syphilis. The result may occasionally be positive for aspirates of lymph nodes in secondary syphilis. False-negative results may occur in primary syphilis because of the application of soaps, antiseptics, or other compounds toxic to T. pallidum to the lesions. A single negative result is therefore insufficient to exclude syphilis. For high-risk individuals (e.g., drug users, men who have sex with men), it is appropriate to treat presumptively on the basis of suspicious lesions after performing serologic tests. Confusion may also arise because of the presence of spirochetes that are morphologically indistinguishable from T. pallidum organisms in the mouth, particularly around the gingival margins. Living T. pallidum organisms demonstrate gradual motion to and fro, rotational movement around the long axis, and rather sudden 90-degree flexing near the center of the organism.

Serologic Tests

Two basic types of serologic tests (Table 303-2) are widely used to diagnose infection with T. pallidum: (1) nontreponemal tests that detect antibodies reactive with diphosphatidylglycerol (cardiolipin), which is a normal component of many tissues; and (2) tests that detect antibodies to specific treponemal antigens.

Nontreponemal Tests

The standard tests to detect anticardiolipin antibody are the rapid plasma reagin (RPR) and Venereal Disease Research Laboratory (VDRL) tests, which

CHAPTER 303 Syphilis  

TABLE 303-2 SEROLOGIC TESTS FOR SYPHILIS TYPE

USE

NONTREPONEMAL (ANTICARDIOLIPIN) ANTIBODIES VDRL (slide flocculation) RPR (circle card) (agglutination)

Screening, quantitation of response to treatment Screening, quantitation of response to treatment

SPECIFIC TREPONEMAL ANTIBODIES FTA-ABS (immunofluorescence with absorbed serum) TP-PA (microhemagglutination) EIA and CIA

Confirmatory, diagnostic; not for routine screening Similar to FTA-ABS but can be quantified and automated Confirmatory and increasingly used for screening; automated

EIA = enzyme immunoassay; CIA= chemiluminescence immunoassay; FTA-ABS = fluorescent treponemal antibody absorption test; RPR = rapid plasma regain test; TP-PA = Treponema pallidum particle agglutination; VDRL = Venereal Disease Research Laboratory.

TABLE 303-3 FREQUENCY OF POSITIVE SEROLOGIC TEST RESULTS IN UNTREATED SYPHILIS STAGE

VDRL (%)

FTA-ABS (%)

TP-PA (%)

Primary Secondary

70

85

50-60

99

100

Latent or late

100

70

98

98

FTA-ABS = fluorescent treponemal antibody absorption test; TP-PA = Treponema pallidum particle agglutination; VDRL = Venereal Disease Research Laboratory.

are slide flocculation tests. The RPR and VDRL are readily quantified, so they are the tests of choice for monitoring patients’ responses to treatment. The relative proportion of patients with a false-positive RPR result depends on the prevalence of syphilis in the community; the lower the prevalence of syphilis, the higher the proportion of reactive RPR test results from nonsyphilitic causes. The RPR test result begins to turn positive less than 1 week after onset of the chancre; thus, a nonreactive RPR test result does not exclude primary syphilis, particularly if the lesion is less than 1 week old. The RPR test result is positive in 99% of patients with secondary syphilis (Table 303-3). Patients with advanced HIV infection may have negative test results, and some patients have such high titers of antibody that they are in antibody excess; dilution of their serum paradoxically results in conversion of a negative test result to a positive one, the so-called prozone reaction. RPR reactivity tends to diminish in later stages of syphilis, and only about 70% of patients with cardiovascular syphilis or late neurosyphilis have positive RPR test results. The quantitative titer of the RPR or VDRL test is somewhat useful in diagnosis and is quite useful for monitoring of the therapeutic response. Most patients with secondary syphilis have titers of at least 1 : 16. Most patients with false-positive RPR test results have titers of less than 1 : 8. No single titer is diagnostic by itself. Significant rises (four-fold or greater) in paired sera, however, strongly indicate acute syphilis.

Treponemal Tests

Several types of treponemal tests are widely used. Treponemal enzyme immunoassays (EIAs) and chemiluminescence immunoassays (CIAs) using cloned treponemal antigens for treponemal antibody detection have become available from several manufacturers and have gained favor because of their low cost and ease of use. In addition to EIA and CIA tests, agglutination of particles to which T. pallidum antigens have been fixed is the basis of the widely used T. pallidum particle agglutination (TP-PA) test. The fluorescent treponemal antibody absorption (FTA-ABS) test has been widely used as well and is reported in terms of relative brilliance of fluorescence, from borderline to 4 plus; most laboratories report only test results with 2 plus or greater reactivity as positive. For patients lacking historical or clinical evidence of syphilis but with a reactive FTA-ABS test result, the test should be repeated. Use of another treponemal test may be helpful in problem cases. The TP-PA test is slightly

1987

less sensitive than the RPR or FTA-ABS test in primary syphilis. Its sensitivity and specificity are otherwise nearly identical to those of the FTA-ABS test. Because EIA and CIA serologic tests permit the screening of large numbers of sera and have performance characteristics (sensitivity, specificity, predictive values) similar to those of other treponemal tests, they have been increasingly used for syphilis screening. Persons with reactive treponemal antigen EIAs or CIAs should be tested with a quantitative nontreponemal test, such as the RPR or VDRL test, for confirmation and to permit the use of that test to evaluate the subsequent response to therapy. It is not unusual for patients to have a reactive EIA or CIA test result for syphilis and a nonreactive RPR or VDRL test result. A substantial proportion of these EIA or CIA-only positive test results are falsely positive or are detecting long-standing, often previously treated syphilis, but occasionally they may detect very recent infection before RPR or VDRL test results become positive. When nontreponemal tests such as the RPR and VDRL are used for screening, treponemal tests are used to confirm that persons with reactive nontreponemal test results have antibodies to T. pallidum. Results of treponemal tests are not reliably quantified. They are sensitive and have a high degree of specificity, in that only approximately 1% of normal individuals have reactive treponemal test results. They are reactive in 85% of patients with primary syphilis, 99% with secondary syphilis, and at least 95% with late syphilis. They may therefore be the only test with a positive result in patients with cardiovascular or neurologic syphilis. For patients with late syphilis, treponemal test results often remain reactive for life, despite adequate therapy. Point of care serologic tests for syphilis are currently available. These tests can detect treponemal antibodies, nontreponemal antibodies, or both. The performance characteristics vary, but in general, they are reasonably good.12

Differential Diagnosis

The differential diagnosis of a genital ulcer (Chapter 269) includes genital herpes (Chapter 350), chancroid (Chapter 285), lymphogranuloma venereum (Chapter 302), and a number of other ulcerative processes. Classically, herpetic ulcers are multiple, painful, superficial, and, if seen early, vesicular. However, atypical manifestations may be indistinguishable from a syphilitic chancre. Genital herpes is much more common than syphilis and is now the most common cause of a “typical chancre” in North America. Syphilitic chancres may also be coinfected with herpes simplex virus in about 15% of cases. The ulcers of chancroid are usually painful, often multiple, and frequently exudative and nonindurated. Lymphogranuloma venereum may produce a small, papular lesion associated with regional adenopathy. Other conditions that must be distinguished include granuloma inguinale (Chapter 300), drug eruptions, carcinoma, superficial fungal infections (Chapter 409), traumatic lesions, and lichen planus (Chapter 409). In most cases, the final distinction is based on dark-field examination, which is positive only in syphilis, and on serologic test results. The differential diagnosis of the skin lesions of secondary syphilis includes pityriasis rosea (Chapter 409), which can be differentiated by the occurrence of lesions along lines of skin cleavage and frequently by the presence of a herald patch. Drug eruptions, acute febrile exanthems, psoriasis, lichen planus, scabies, and other diseases must also be considered in some cases. A mucous patch may superficially resemble oral candidiasis (i.e., thrush). Infectious mononucleosis (Chapter 353) may appear similar to secondary syphilis, with sore throat, generalized adenopathy, hepatitis, and a generalized rash. Hepatitis (Chapter 139) may also cause confusion.

False-Positive Serologic Test Results for Syphilis

The RPR or VDRL test result is reactive in patients with other treponemal diseases, such as pinta, yaws, and endemic syphilis (Chapter 304). These test results may also be falsely reactive in persons who do not have treponemal infections based on a negative clinical history or negative results of serum treponemal tests. The origins of false-positive results are better studied for nontreponemal tests than for treponemal tests. Acute (6 months) are relatively common in patients with autoimmune disorders such as systemic lupus erythematosus (Chapter 250), parenteral drug users, HIV-infected patients, patients with leprosy, and the aged. Between 8 and 20% of patients with systemic lupus erythematosus have false-positive RPR test results. Chronic false-positive RPR test results in female patients 20 years or younger indicate a significant risk for the future development of systemic

1988

CHAPTER 303 Syphilis  

TABLE 303-4 PENICILLIN TREATMENT FOR SYPHILIS AS RECOMMENDED BY THE U.S. PUBLIC HEALTH SERVICE DOSAGE AND ADMINISTRATION† INDICATIONS FOR SYPHILIS THERAPY*

BENZATHINE PENICILLIN G

AQUEOUS BENZYLPENICILLIN G OR PROCAINE PENICILLIN G

Primary, secondary, and early latent syphilis (1 year); cardiovascular syphilis, late benign (cutaneous, osseous, visceral gumma)

Total of 7.2 million units IM in doses of 2.4 million units at 7-day intervals during a 21-day period

Total of 9 million units IM in doses of 600,000 units/day during a 15-day period

Symptomatic or asymptomatic neurosyphilis

2-4 million units aqueous (crystalline) penicillin G IV q4h for at least 10 days

2-4 million units procaine penicillin IM daily and probenecid 500 mg orally four times daily, for 10-14 days

CSF normal: total of 50,000 units/kg IM in a single or divided dose at one session CSF normal: same as for early congenital syphilis, up to 2.4 million units

CSF abnormal: total of 50,000 units/kg/day IM for 10 consecutive days‡ CSF abnormal: 200,000-300,000 units/kg/day aqueous crystalline penicillin IV for 10-14 days

Congenital Infants Older children

CSF = cerebrospinal fluid. *In pregnancy, treatment depends on the stage of syphilis. † Individual doses can be divided for injection in each buttock to minimize discomfort. ‡ For aqueous penicillin, give in two divided intravenous doses per day; for procaine penicillin, give as one daily dose intramuscularly. Data from Workowski KA, Bolan GA. Sexually transmitted diseases treatment guidelines, 2015. MMWR Recomm Rep. 2015;64:1-137.

lupus erythematosus, thyroiditis, or other autoimmune disorders. As many as a third of parenteral drug users have false-positive RPR test results. More than 1% of persons 70 years old and 10% of those older than 80 years also have low-titer, false-positive RPR test results. In most cases of false-positive RPR test results, the titer is less than 1 : 8, although a few patients with lymphoma and other diseases have very high-titer, false-positive results.

Neurosyphilis

Asymptomatic neurosyphilis is diagnosed when there are CSF abnormalities, such as lymphocytic pleocytosis, protein elevation, or a reactive VDRL test result, in a syphilis patient in the absence of signs and symptoms of neurologic disease. Unlike serologic tests, the VDRL and RPR tests do not perform equally for CSF, and only the VDRL is recommended. Although numerous other processes can cause CSF pleocytosis or protein elevations, false-positive CSF VDRL test results are rare in the absence of a traumatic tap. If the CSF is normal 2 years or longer after the initial infection, a positive CSF finding is not likely to develop later. Routine lumbar punctures to examine CSF are not indicated in neurologically asymptomatic, immunocompetent patients with early syphilis. Lumbar puncture in HIV-infected persons with early syphilis is the subject of controversy.13 Although HIV-infected persons, particularly those with a CD4 count less than 350 cells/µL or an RPR titer more than or equal to 1 : 32, may be more likely to have cerebrospinal fluid abnormalities consistent with neurosyphilis, there is no evidence that lumbar punctures in these asymptomatic patients leads to improved clinical outcomes. Although a nonreactive CSF FTA-ABS result may be useful to rule out the diagnosis, no diagnosis of neurosyphilis should be based solely on the CSF FTA-ABS test. In syphilitic meningitis, the CSF shows a lymphocytic pleocytosis, with increased protein and usually normal glucose concentrations. The CSF VDRL test is often reactive. Rarely, the CSF glucose concentration is decreased. Without treatment, syphilitic meningitis generally resolves, similar to the course of other manifestations of early syphilis. This syndrome can mimic tuberculous or fungal meningitis or nonpurulent meningitis of various causes. In tabes dorsalis, the VDRL test for serum is nonreactive in as many as 30 to 40% of patients, and 10 to 20% of patients (even before the advent of penicillin) have normal CSF VDRL results. The FTA-ABS test for serum is nearly always reactive. In general paresis, the CSF is nearly always abnormal, with lymphocytic pleocytosis and an increased total protein concentration. The VDRL test is usually reactive for CSF and serum.

Congenital Syphilis

Because many infants with congenital syphilis may be clinically normal at birth but develop serious, symptomatic disease some weeks later, it is important to determine whether a newborn with a reactive serologic test result for syphilis has passively transferred maternal antibody or is actively infected. If the mother has been adequately treated for syphilis during pregnancy and the infant is clinically normal at birth, one option is to monitor the infant carefully by serial examinations and RPR titers. If the reactive RPR result for the infant is caused by passively transferred maternal antibody, the titer will fall markedly

in the first 2 months of life; a rising titer indicates active disease and the need for treatment. However, the risk of improper follow-up of RPR-positive but clinically normal neonates makes the immediate empirical administration of effective therapy an attractive alternative.

TREATMENT  T. pallidum is inhibited by less than 0.01 µg/mL of penicillin G. Because treponemes divide slowly and penicillin acts only on dividing cells, it is necessary to maintain serum levels of penicillin for many days (Table 303-4).14

Early Infectious Syphilis

Early syphilis (39° C in most patients), headache, myalgia, and shaking chills characterizes the onset of illness. Cough, nausea and vomiting, and fatigue are less frequent complaints. Signs include fever, tachycardia, lethargy or confusion, conjunctival injection, and epistaxis. Hepatosplenomegaly, jaundice, and often a truncal petechial rash are common signs in louse-borne relapsing fever. Neurologic findings may occur, including meningitis, meningoencephalitis, and facial palsy, although these entities are more common with tick-borne relapsing fever. Untreated louse-borne disease lasts 6 days, and relapses occur once after an afebrile period of 9 days. The initial illness of tick-borne relapsing fever lasts about 4 days without antibiotic treatment, with an average of two relapses (each after an average 10-day intervening afebrile period) before the diagnosis is made.5 Borrelia miyamotoi presents with high fever, chills, marked headache, and myalgia or arthralgia. Elevated liver enzyme levels, neutropenia, and thrombocytopenia are common.6 Relapsing fever in pregnancy can cause placental damage and intrauterine growth retardation and results in miscarriage in a third of patients. Neonatal infection by both the tick- and louse-borne forms is accompanied by jaundice, hepatosplenomegaly, and often sepsis and hemorrhage. Fever and hepatosplenomegaly are also common signs in children. It has been also recognized in immunocompromised patients.  

DIAGNOSIS

The diagnosis should be considered in patients with fever who are returning from a stay in cabins in the mountainous and high-elevation areas of the western United States.7 Only a few patients will remember tick exposure, because O. hermsii is a night feeder, has a painless bite, and remains attached for only 15 minutes. Internationally, relapsing fever can occur sporadically wherever dwellings are infected with Ornithodoros ticks, as well as in epidemics with louseborne disease. Because the number of organisms in blood is extremely high, the diagnosis is most often made by direct visualization of the organism in a blood smear (Fig. 306-1), although the diagnosis can also be made with polymerase chain reaction and serodiagnostic tests. For B. miyamotoi, only 16% of patients are seropositive, but polymerase chain reaction testing is highly sensitive and specific.8 Spirochetes can be demonstrated in peripheral blood smears taken during febrile episodes in 70% of patients. Additional sensitivity may be gained by examination of a buffy coat preparation of peripheral blood. Because of

FIGURE 306-1.  A single spirochete is seen in a Wright-stained thin blood smear from a patient with relapsing fever.

their characteristic locomotion, spirochetes can be readily detected by direct visualization of thick blood films under low-power microscopy. Culture of the organism requires a special medium and is not practical in a clinical laboratory setting. The white blood cell count is generally normal, but platelet counts of less than 50,000/µL occur in up to 90% of cases of louse-borne disease. Prothrombin and activated partial thromboplastin times are often prolonged. In louse-borne disease, elevations in liver function test results (serum aminotransferase and bilirubin levels) and blood urea nitrogen are common. Urinalysis may reveal proteinuria and microscopic hematuria. Examination of cerebrospinal fluid may show a lymphocytic pleocytosis, and spirochetes may be directly visualized.

TREATMENT  Borrelia is generally quite sensitive to antibiotics, which has led to recommendations for single-dose treatments. Although this may be sufficient, especially for louse-borne disease, recent reports suggest that silent residual infections occur and may best be addressed by longer treatments. For tick-borne relapsing infection, treatment should extend for 7 days to reduce the risk for persistent infection. Tetracycline, doxycycline, ceftriaxone, and erythromycin are all effective antibiotics. Erythromycin should not be used in pregnant women and children younger than 7 years (in whom tetracyclines can stain the permanent teeth). Penicillin treatment has been reported to clear the spirochetemia more slowly than tetracycline does. The Jarisch-Herxheimer reaction (typically characterized by a rise in body temperature of 1° C, rigors, a rise in blood pressure followed by a fall, and transient leukopenia) occurs 2 to 3 hours after treatment in many patients with louse-borne disease, less commonly in tick-borne disease, and should be anticipated and managed supportively. Death as a result of shock from the JarischHerxheimer reaction occurs rarely. The Jarisch-Herxheimer reaction has been associated with accelerated phagocytosis of spirochetes by neutrophils and transient elevations in tumor necrosis factor–α (TNF-α), interleukin (IL)–6, IL-8, and IL-10. In small numbers of patients with louse-borne relapsing fever, anti– TNF-α antibodies have been effective in prevention.



PREVENTION

Prevention of louse-borne relapsing fever hinges on improving hygienic conditions, delousing affected areas, and antibiotic treatment of patients and close contacts. Tick-borne relapsing fever can be prevented by reducing the risk of contact with rodents and ticks, including repair of structural flaws in cabins and other residences so rodents cannot nest in or around them, as well as spraying infested indoor environments. Tick-bite screening and prophylactic treatment with doxycycline in highly endemic areas has been reported to be a practical, safe, and effective policy in preventing tick-borne relapsing fever.  

PROGNOSIS

Epidemics of louse-borne relapsing fever have been reported, with mortality rates approaching 40%; as much as 5% of the mortality is related to

Jarisch-Herxheimer reactions with treatment. Mortality from tick-borne disease is less than 5%. Autopsies of patients with louse-borne disease have documented intracranial hemorrhage, brain edema, bronchopneumonia, hepatic necrosis, and splenic infarcts. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 306  RELAPSING FEVER AND OTHER BORRELIA INFECTIONS  

GENERAL REFERENCES 1. Talagrand-Reboul E, Boyer PH, Bergstrom S, et al. Relapsing fevers: neglected tick-borne diseases. Front Cell Infect Microbiol. 2018;8:1-21. 2. Kingry LC, Anacker M, Pritt B, et al. Surveillance for and discovery of Borrelia species in US patients suspected of tickborne illness. Clin Infect Dis. 2018;66:1864-1871. 3. Osthoff M, Schibli A, Fadini D, et al. Louse-borne relapsing fever—report of four cases in Switzerland, June-December 2015. BMC Infect Dis. 2016;16:1-5. 4. Stone BL, Brissette CA. Host immune evasion by Lyme and relapsing fever Borrelliae: findings to lead future studies for Borrelia miyamotoi. Front Immunol. 2017;8:1-16.

1998.e1

5. Warrell DA. Louse-borne relapsing fever (Borrelia recurrentis infection). Epidemiol Infect. 2019; 147:1-8. 6. Molloy PJ, Telford SR 3rd, Chowdri HR, et al. Borrelia miyamotoi disease in the northeastern United States: a case series. Ann Intern Med. 2015;163:91-98. 7. Eldin C, Parola P. Update on tick-borne bacterial diseases in travelers. Curr Infect Dis Rep. 2018;20:1-9. 8. Modarelli JJ, Piccione J, Ferro PJ, et al. Novel real-time PCR assays for genomic group identification of tick-borne relapsing fever species Borrelia hermsii. Diagn Microbiol Infect Dis. 2019;93:24-29.

1998.e2

CHAPTER 306  RELAPSING FEVER AND OTHER BORRELIA INFECTIONS  

REVIEW QUESTIONS 1. The differential diagnosis for a patient presenting with an undifferentiated febrile illness, chills, headache, and recent tick exposure include all but one of the following: A . Lyme disease B. Rocky Mountain spotted fever (RMSF) C. Ehrlichiosis D. Tick paralysis E. Anaplasmosis Answer: D  Not tick paralysis, which is not an infection but in fact an intoxication due to a tick neurotoxin. An undifferentiated febrile illness with recent tick exposure could include Lyme disease, RMSF, ehrlichiosis, and anaplasmosis. 2. Causes of a peripheral facial nerve palsy include all but one of the following: A . Lyme disease B. Herpes simplex virus (HSV) C. Sarcoidosis D. Babesiosis E. Relapsing fever Answer: D  Not babesiosis. HSV is the most likely cause of Bell palsy, with the borrelial infections, Lyme disease, and relapsing fever not uncommon, and sarcoidosis rare. 3. Diagnosis of relapsing fever can include all but one of the following: A . Serology (acute and convalescent) B. Positive blood smear for a spirochete C. Polymerase chain reaction (PCR) assay of blood D. Sleeping in rustic cabins in the U.S. West E. Blood cultures Answer: E  Blood cultures cannot culture the Borrelia bacteria that cause relapsing fever; however, the number of organisms per milliliter of blood is so high they can be seen on a blood smear and detected by PCR.

4. Treatment of relapsing fever is: A . Tetracycline B. Ceftriaxone C. Complicated by Jarisch-Herxheimer reaction D. A and B E. A, B, C Answer: E  Tetracycline and ceftriaxone are equally effective, and either can cause the Jarisch-Herxheimer reaction. 5. Complications of relapsing fever include: A . Jarisch-Herxheimer reaction B. Meningitis C. Jaundice D. A and C E. All of the above Answer: E  Relapsing fever can be complicated by meningitis, jaundice (especially the louse-borne form), and the Jarisch-Herxheimer reaction, which is treatment associated and probably due to accelerated phagocytosis of spirochetes by neutrophils and the consequent elevation in levels of tumor necrosis factor–α (TNF-α) and interleukin (IL)–6, IL-8, and IL-10.

1998

CHAPTER 307 Leptospirosis  

seven nonpathogenic species. Pathogenic leptospires are further classified into over 25 serogroups and 250 serovars that differ by geographic distribution and host specificity, which is useful information for outbreak and other epidemiologic investigations.  

307  LEPTOSPIROSIS SHERIF ZAKI AND WUN-JU SHIEH



DEFINITION

Leptospirosis is a zoonotic disease caused by pathogenic Leptospira species spirochetes. Leptospirosis is distributed worldwide and is most prevalent in tropical developing countries. Leptospires frequently infect wild and domestic mammals. Humans are infected directly by contact with infected animals or indirectly through contact with soil or water contaminated by urine from infected animals. Disease severity ranges from mild and self-limiting to severe with life-threatening manifestations, including massive pulmonary hemorrhage and Weil disease (the triad of jaundice, acute renal failure, and bleeding).1

The Pathogen

Leptospires are thin, coiled, highly motile spirochetes in the Leptospira genus of the Leptospiraceae family. They measure 6 to 20 microns in length and 0.1 µm in diameter. One or both ends of the spirochete are usually hooked. They are obligate aerobes that can survive for several weeks in the environment. Leptospira is currently genetically classified into ten pathogenic species (L. interrogans, L. kirschneri, L. borgpetersenii, L. santarosai, L. noguchii, L. weilii, L. alexanderi, L. alstoni, L. kmetyi, L. mayottensis), five intermediate or opportunistic species, (L. inadai, L. broomii, L. fainei, L. wolffii, L. licerasiae), and

A

EPIDEMIOLOGY

Over 350,000 cases of leptospirosis are estimated to occur each year worldwide and are generally underreported. In the United States, leptospirosis is a nationally notifiable disease. The National Notifiable Diseases Surveillance System (NNDSS) began receiving case notifications for leptospirosis in 2014. In the first two years of notifications combined (2014 to 2015), the total case count for leptospirosis was 203 cases reported from 17 states, jurisdictions, and territories, with 114 total cases reported from Puerto Rico, 45 from Hawaii, and 11 from Guam.2 The majority of infections are mild and self-limiting, but case fatality in reported cases may be as high as 10%. In endemic areas, up to 20 to 30% of cases with acute undiagnosed febrile illness may be due to leptospirosis,3 and seroprevalence can range from 5 to 15%. Leptospirosis most often affects people who work outdoors or with animals, or those who take part in recreational activities involving water or soil, like swimming, boating, and gardening. After floods, heavy rains, or other natural disasters, anyone who has been in contact with floodwater, contaminated freshwater, or soil could be at risk for infection. The major groups at risk are slum dwellers, subsistence farmers, and animal workers, owing to exposure to rodent, domestic, and wild animal reservoirs. In both tropical and temperate climates, the urban poor are an underrecognized population at risk. Humans are considered accidental hosts; rare human-to-human transmission by transplacental infection and breast-feeding has been reported.  

PATHOBIOLOGY

Leptospires can directly penetrate abraded skin and mucous membranes and spread hematogenously to target organs. The classic illness is biphasic, with the first phase characterized by leptospiremia and the second phase with organism clearance by agglutinating antibodies and an associated host response that can be immunopathogenic.4 Leptospires can persist for a period of time in target organs. In asymptomatic reservoir animals, leptospires can reach massive densities within the renal tubules, resulting in continuous urinary excretion. Pathologic findings may include pulmonary hemorrhage, diffuse alveolar damage, mild to marked hepatocellular dissociation, mild portal hepatitis, lymphohistiocytic interstitial nephritis, renal tubular necrosis, and mild renal glomerular mesangial hyperplasia (Fig. 307-1).5 Hemorrhage in other

B

C FIGURE 307-1.  Pathologic features of severe leptospirosis. A, Pulmonary hemorrhage. B, Hepatocellular dissociation. C, Interstitial nephritis.

CHAPTER 307 Leptospirosis  

ABSTRACT

Leptospirosis is a zoonotic disease caused by spirochetes of the genus Leptospira. Leptospirosis is distributed worldwide and is a common infection affecting many species of wild and domestic animals. Human infection can occur either through direct contact with infected animals or more commonly through indirect contact with water or soil contaminated with the urine of infected animals. There are few reliable data on the global incidence of leptospirosis because of its variable clinical presentations and lack of diagnostic capabilities in many areas. The infection can be severe or even fatal if the patient is not diagnosed and treated properly.

KEYWORDS

leptospirosis zoonosis hepatorenal failure pulmonary hemorrhage conjunctival suffusion contaminated water

1998.e3

CHAPTER 307 Leptospirosis  

A

1999

B

FIGURE 307-2.  Leptospira in liver. Note the coiled nature of the spirochetes. A, Immunohistochemistry showing spirochetal and granular antigen staining. B, Warthin-Starry silver stain showing spirochetes.

organs, multifocal myocarditis, myositis, and hemophagocytosis may also be present. By immunohistochemistry or silver stains (Fig. 307-2), leptospires can be seen within the renal interstitium, hepatic parenchyma and sinusoids, and within the walls of small, medium, and large pulmonary blood vessels. Leptospire tissue penetration may be mediated by a burrowing motion and secreted enzymes including collagenase and sphingomyelinase. Leptospiral proteins interact directly with host extracellular matrix components such as collagen, fibronectin, and laminins. Leptospires are resistant to the alternative pathway of complement-mediated lysis and can bind inhibitory complement factor H. Leptospiral lipopolysaccharides and lipopeptides have low endotoxic potency but can activate innate immune response through toll-like receptor (TLR)-2 signaling and are thought to generate a cytokine response.6 Immunity is considered to be primarily humoral and serotype specific. Circulating immune complexes may contribute to renal damage and endothelial dysfunction. An expansion of γδ T cells occurs during infection. Leptospires may also directly activate plasminogen to plasmin, the main enzyme of the fibrinolytic system, which could promote hemorrhage. Genomic studies of pathogenic intermediate and saprophytic Leptospira species have revealed a relatively large genome that contains genes involved in environmental survival, chemotaxis, and motility that may be involved in pathogenesis. Little is known about host genetic risk factors, although the HLA-DQ6 allele has been associated with increased susceptibility to infection.  

CLINICAL MANIFESTATIONS

The incubation period is typically 7 to 12 days (range, 2 to 30 days). During the early phase of illness (first 3 to 7 days), the majority of patients present with high fever (38° to 40° C) and myalgia. Cough, nausea and vomiting, diarrhea, headache, photophobia, and rash may be seen. Conjunctival suffusion is a characteristic finding (Fig. 307-3), but it is only seen in a third of patients near the end of early-phase illness. Myalgia may be pronounced and most frequently involves the calves and lumbar musculature, with creatine phosphokinase elevation. Severe cervical and abdominal myalgia may mimic nuchal rigidity or an acute abdomen, respectively. Rash occurs in 10 to 20% of patients and may be urticarial, maculopapular, or purpuric in a typically pretibial and truncal distribution. Hepatosplenomegaly and lymphadenopathy may also be present. Resolution of symptoms coincides with the presence of agglutinating immunoglobulin (Ig)M antibodies and reduction in leptospiremia. As a classic biphasic disease, fever may recur 3 to 4 days after remission. In this later immune phase, severe headache is often present and can be associated with photophobia, meningeal signs, and cerebrospinal fluid (CSF) pleocytosis. Severe disease may occur progressively at initial presentation or during late-phase leptospirosis and result in 10 to 50% mortality. Life-threatening manifestations are renal failure, hypotension, hemorrhage, and respiratory failure. Jaundice occurs in 5 to 10% of patients, and serum bilirubin levels can be elevated up to 40 to 80 mg/dL, with only moderate and minor elevations in transaminase and alkaline phosphatase levels, respectively. Leptospirosis may present as acute cholecystitis. Long-term hepatic sequelae are typically not seen. Renal findings are typically of nonoliguric hypokalemic renal insufficiency and impaired tubular sodium reabsorption. Volume loss may result in oliguric renal insufficiency and acute tubular necrosis, and renal failure occurs in about half of severe cases. Common urinalysis findings are proteinuria, white blood cells, hematuria, and hyaline and granular casts.

FIGURE 307-3.  Conjunctival suffusion in a patient with severe leptospirosis. Conjunctival suffusion is a characteristic sign of leptospirosis and best seen along the right upper palpebral border above the subconjunctival hemorrhage. (Courtesy Antonio Seguro and Paulo Marotto, Hospital Emílio Ribas and Universidade de São Paulo.)

Pulmonary findings may include cough, dyspnea, and hemoptysis. Leptospirosis-associated pulmonary hemorrhage and acute respiratory distress syndrome is now recognized as a common clinical presentation.7 The most frequent radiographic findings in patients with leptospirosis causing diffuse alveolar hemorrhage are ground-glass opacities, airspace nodules, ground-glass nodules, and consolidations.8 Cardiac conduction abnormalities can be seen and tend to be nonspecific in mild disease. First-degree atrioventricular block and features of pericarditis are the most common findings in severe disease. Arrhythmias including ventricular fibrillation may also occur. Thrombocytopenia is a frequent complication in leptospirosis, and this condition has been reported to be present in more than half of patients at the time of hospital admission. Lengthy disease and acute kidney injury are risk factors for thrombocytopenia. Prothrombin and partial thromboplastin times are typically normal or only mildly elevated. Petechiae, conjunctival hemorrhage, and purpura can be seen in addition to the more severe hemorrhagic manifestations. Aseptic meningitis is the most frequent neurologic manifestation. Cerebrospinal fluid (CSF) findings include pleocytosis with neutrophil predominance in early disease, followed by lymphocyte predominance in later disease. Glucose is generally normal, and CSF pressure may be elevated. Less common are intracerebral hemorrhage, encephalitis, myelitis, and peripheral neuropathy. Ocular leptospirosis, including chronic uveitis, is thought to be primarily caused by an immunopathogenic mechanism and occurs late in the disease process.  

DIAGNOSIS

Few clinical signs differentiate early-stage leptospirosis, and severe leptospirosis may be recognized in the form of Weil disease that is characterized by hepatorenal failure. Reference tests require specialized laboratories, and there is a need for point-of-care testing in tropical areas. Culture and serologic assays may provide false negatives. Culture of the organism requires specialized media (EllinghausenMcCullough-Johnson-Harris [EMJH] or Fletcher media), with often more than 4 weeks needed with observation under darkfield microscopy. Cultures generally have low sensitivity, with the best yield obtained from peripheral blood during days 1 through 4 of the acute illness; urine may be positive for up to day 10. Direct darkfield microscopy of clinical specimens can also be attempted but has low sensitivity and specificity.

The gold standard serologic test is the microagglutination test (MAT), available at the U.S. Centers for Disease Control and Prevention. MAT is considered positive if there is a four-fold rise in titer between acute and convalescent sera. This assay requires cultured Leptospira; it provides serotype data and has a very high specificity but lower sensitivity. U.S. Food and Drug Administration–approved IgM enzyme-linked immunosorbent assay (ELISA) and indirect hemagglutination assays are also available and appear to be as good as MAT. Multiple lateral flow rapid diagnostic “dipstick” tests are manufactured and used in other countries; however, their performance may not be well validated. Diagnostic polymerase chain reaction (PCR) assays are the most sensitive test overall, but combined PCR and MAT, supplemented by regionally validated ELISA testing, may be needed to maximize sensitivity.9,10 Organisms within tissues can be detected by Warthin-Starry silver stain or immunohistochemistry.

Differential Diagnosis

As previously discussed, few clinical or laboratory findings differentiate leptospirosis from other causes of acute fever. Differential diagnosis depends on other concurrent diseases in the geographic area, most frequently malaria, dengue, typhoid fever, melioidosis, scrub typhus, and rickettsial diseases. Other diseases in the differential include influenza, acute viral hepatitis, yellow fever, bacterial and viral meningitis, bacterial sepsis, Zika virus, chikungunya virus, and hantavirus infection. A high index of suspicion is needed in endemic areas. Useful diagnostic clues are conjunctival suffusion, muscle tenderness, and pulmonary bleeding. Hypokalemia, elevated creatinine, elevated creatine phosphokinase, and thrombocytopenia are nonspecific but may also suggest leptospirosis.

TREATMENT  The World Health Organization guidelines and widespread clinical practice are to treat patients early with antibiotics for leptospirosis. Penicillin, doxycycline, or a cephalosporin appear to be equally efficacious, and doxycycline has the advantage of also treating rickettsial infections. Of note, a recent meta-analysis of five randomized trials concluded that the role of various antibiotics in treatment of leptospirosis is uncertain, and there is insufficient “grade A” evidence to recommend for or against the use of antibiotics to treat leptospirosis. A1  Additional randomized controlled trials are needed. Treatment regimens for mild leptospirosis include doxycycline (100 mg PO bid), ampicillin (500 mg PO q6h), amoxicillin (500 mg PO q8h), or azithromycin (1 g followed by 500 mg daily for 2 days); and for severe disease, intravenous (IV) penicillin (1.5 million units q6h), ceftriaxone (1 g daily), cefotaxime (1 g q6h), or doxycycline (100 mg IV q12h). Treatment duration is typically 7 days. The Jarisch-Herxheimer reaction (JHR), a febrile inflammatory reaction that occurs with initiation of treatment and results from clearance of the organism from the circulation, can occur in 20% of patients who receive antibiotics treatment.11 Two risk factors have been independently associated with JHR occurrence: (1) Leptospira interrogans serogroup Australis as the infecting strain; and (2) starting antibiotics earlier than 3 days after symptom onset. Therefore, patients require monitoring at initiation of antibiotics. Prompt triage of high-risk patients and aggressive supportive care are essential. Hypotension should be treated, and volume repletion is useful in limiting renal damage. Patients with nonoliguric hypokalemic renal insufficiency may be treated by volume and potassium repletion. Prompt dialysis is indicated for oliguric renal insufficiency, either by continuous hemofiltration or by peritoneal dialysis. Serial electrocardiograms are helpful to monitor for arrhythmia. Aggressive therapy may be needed to treat hemorrhage and respiratory failure.



PREVENTION

Preventive measures include sanitation to prevent population exposure to contaminated water, and limiting water contamination by animal reservoirs such as dogs, pigs, and cattle. Vaccination is available for domestic and livestock animals. Rodent control is important. Workers with occupational exposure to animals or contaminated water or soil are encouraged to use protective equipment such as gloves and boots. Travelers to endemic areas should be counseled on fresh water exposure. A vaccine is not available for human use within the United States. Vaccine trials have been performed in Cuba, Russia, and China; however, vaccine safety and efficacy are uncertain. Prophylaxis with doxycycline (200 mg PO weekly) is widely used for persons with exposure to contaminated water or at high risk for leptospirosis. Prophylaxis may work better in reducing risk of clinical disease for short-term travelers with high risk, rather than residents in an endemic area. A single dosage of

200 mg doxycycline for prophylaxis might be effective for preventing leptospirosis among flood victims with laceration wounds after recent flood exposure A2 ; additional randomized controlled trials are needed.  

PROGNOSIS

The majority of infections are self-limiting, and the median case fatality among patients who seek medical attention is about 2%. Severe leptospirosis is associated with a higher mortality: about 12% in patients with renal failure, about 20% in patients with jaundice, and about 60% in patients over age 60 years.12 Death is more frequent during infection if there is oliguria, pulmonary hemorrhage, or respiratory insufficiency. Ocular disease, including chronic uveitis, may result in severe visual impairment.

  Grade A References A1. Charan J, Saxena D, Mulla S, et al. Antibiotics for the treatment of leptospirosis: systematic review and meta-analysis of controlled trials. Int J Prev Med. 2013;4:501-510. A2. Schneider MC, Velasco-Hernandez J, Min KD, et al. The use of chemoprophylaxis after floods to reduce the occurrence and impact of leptospirosis outbreaks. Int J Environ Res Public Health. 2017;14:1-18.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 307 Leptospirosis  

GENERAL REFERENCES 1. Jimenez JIS, Marroquin JLH, Richards GA, et al. Leptospirosis: report from the task force on tropical diseases by the World Federation of Societies of Intensive and Critical Care Medicine. J Crit Care. 2018;43:361-365. 2. Adams DA, Thomas KR, Jajosky RA, et al. Summary of notifiable infectious diseases and conditions—United States, 2015. MMWR Morb Mortal Wkly Rep. 2017;64:1-143. 3. Abdul Mutalip MH, Mahmud MAF, Lodz NA, et al. Environmental risk factors of leptospirosis in urban settings: a systematic review protocol. BMJ Open. 2019;9:1-5. 4. Priya SP, Sakinah S, Sharmilah K, et al. Leptospirosis: molecular trial path and immunopathogenesis correlated with dengue, malaria and mimetic hemorrhagic infections. Acta Trop. 2017;176:206-223. 5. Chin VK, Basir R, Nordin SA, et al. Pathology and host immune evasion during human leptospirosis: a review. Int Microbiol. 2019. [Epub ahead of print.] 6. Chin VK, Lee TY, Lim WF, et al. Leptospirosis in human: biomarkers in host immune responses. Microbiol Res. 2018;207:108-115.

2000.e1

7. Vandroux D, Chanareille P, Delmas B, et al. Acute respiratory distress syndrome in leptospirosis. J Crit Care. 2019;51:165-169. 8. von Ranke FM, Zanetti G, Escuissato DL, et al. Pulmonary leptospirosis with diffuse alveolar hemorrhage: high-resolution computed tomographic findings in 16 patients. J Comput Assist Tomogr. 2016;40:91-95. 9. Galloway RL, Hoffmaster AR. Optimization of LipL32 PCR assay for increased sensitivity in diagnosing leptospirosis. Diagn Microbiol Infect Dis. 2015;82:199-200. 10. Waggoner JJ, Pinsky BA. Molecular diagnostics for human leptospirosis. Curr Opin Infect Dis. 2016;29:440-445. 11. Guerrier G, Lefevre P, Chouvin C, et al. Jarisch-Herxheimer reaction among patients with leptospirosis: incidence and risk factors. Am J Trop Med Hyg. 2017;96:791-794. 12. Taylor AJ, Paris DH, Newton PN. A systematic review of the mortality from untreated leptospirosis. PLoS Negl Trop Dis. 2015;9:1-13.

2000.e2

CHAPTER 307 Leptospirosis  

REVIEW QUESTIONS 1. Which group is least at risk for leptospirosis? A . Veterinarians B. Hospital workers C. Veterinarians D. Farmers E. Sewage workers F. Urban homeless Answer: B  Health care workers are at low risk for leptospirosis because humanto-human transmission is rare and primarily involves vertical transmission from mother to infant. Because Leptospira species are zoonoses that can persist in the environment, persons with exposure to domestic or wild animals, soil, or fresh water are at greatest risk. The urban poor of both temperate and tropical climates are an underappreciated population at risk. 2. Which is the most likely presentation of leptospirosis? A . High fever, myalgia, and headache B. Sore throat C. Diarrhea D. Localized seizure E. Nasal congestion and rhinorrhea Answer: A  High fever, myalgia, and headache are common clinical manifestations of leptospirosis. Leptospirosis can be associated with acute abdominal pain, sometimes mimicking acute cholecystitis; however, diarrhea rarely occurs. Localized seizure would be an atypical presentation of leptospirosis, which is generally not considered to be associated with focal neurologic findings. 3. Of the following, which are the most likely pathologic findings in leptospirosis? A . Massive hepatic necrosis and acute renal tubular injury B. Pulmonary hemorrhage, interstitial nephritis, and hepatocellular dissociation C. Neutrophilic bronchopneumonia and interstitial nephritis D. Vasculitis with abundant intravascular spirochetes E. Pulmonary and renal infarction Answer: B  Massive pulmonary hemorrhage, interstitial nephritis, and hepatocellular dissociation are the typical findings seen at autopsy. Massive hepatic necrosis is generally not seen and would suggest dengue or yellow fever virus infection (depending on the geographic location) or other causes of acute fulminant liver failure. Leptospiral organisms do not cause neutrophilic bronchopneumonia. Vasculitis is generally not a prominent feature of leptospirosis, although a degree of inflammation has been described within pulmonary vessel walls. Leptospira species spirochetes are not intravascular but associated with the interstitium; these organisms have several well-defined mechanisms to bind collagen and other components of the extracellular matrix. Infarcts are not a typical feature and would suggest a thromboembolic process.

4. What is the most likely renal finding in leptospirosis? A . Massive proteinuria B. Cortical necrosis C. Pyelonephritis D. Hyperkalemic nonoliguric renal failure E. Allergic nephritis Answer: D  Hyperkalemic nonoliguric renal failure is the typical renal presentation of leptospirosis, which can progress to oliguric renal failure. Hypokalemia is an important laboratory value that can support the diagnosis of leptospirosis. Treatment involves volume resuscitation to prevent oliguric renal failure, electrolyte replacement, and prompt consideration of dialysis. Massive proteinuria would be atypical and suggests a different etiology. The pathologic findings are of interstitial nephritis; cortical necrosis, pyelonephritis, and eosinophilic infiltrates are typically not seen. 5. What is the most likely pulmonary finding in leptospirosis? A . Chest pain B. Bronchopneumonia C. Cough, dyspnea, and hemoptysis D. Pneumothorax E. Lung abscess Answer: C  Pulmonary findings in leptospirosis may include cough, dyspnea, and hemoptysis. Leptospirosis-associated pulmonary hemorrhage and acute respiratory distress syndrome is now recognized as a common clinical presentation. Radiographic findings may show a patchy alveolar infiltrate to large areas of consolidation due to hemorrhage. Bronchopneumonia and lung abscess are not commonly associated with leptospirosis.

2000

CHAPTER 308 Tuberculosis  

308  TUBERCULOSIS JERROLD J. ELLNER AND KAREN R. JACOBSON



DEFINITION

Tuberculosis (TB) is a chronic granulomatous disease with a unique latent stage caused by the acid-fast bacillus (AFB) Mycobacterium tuberculosis. The lung is a site of disease in 75 to 80% of cases; frequent extrapulmonary sites are the lymph nodes, pleura, bones, and joints. TB is spread from person to person primarily by inhalation of infectious droplet nuclei aerosolized by patients with active pulmonary TB. TB is the leading infectious disease killer globally, with over 95% of cases and 99% of deaths occurring in resourcelimited settings. The human immunodeficiency virus (HIV) pandemic led to a resurgence of TB worldwide and promoted explosive nosocomial outbreaks of multiple-drug resistant TB. The result was increased attention to TB as a global public health emergency and increased funding for TB control and research. The problems posed for TB control are compounded by increasing prevalence of drug-resistant disease that is expensive to treat and may be refractory to available drugs.1

The Pathogen

TB is caused by infection with one of four members of the Mycobacterium tuberculosis complex: M. tuberculosis, M. africanum, M. orygis or M. bovis. The causative organism is a slender, non-motile, non–spore-forming, non–toxinproducing bacillus that may be beaded in appearance and is approximately 2 to 4 µm in length. It is a slow-growing (doubling time of 18 to 24 hours) facultative aerobe that can persist intracellularly for prolonged periods. The organism is identified in clinical specimens as an acid-fast bacillus (AFB). M. tuberculosis can be stained with carbol fuchsin by either alkalinization (Kinyoun) or heat (Ziehl-Neelsen) methods. The waxy coat of M. tuberculosis, composed of mycolic acid and other complex lipopolysaccharides, precludes decolorization of the stain with a mixture of acid and alcohol. DNA sequencing of M. tuberculosis and genetic manipulations have promoted basic understanding of the metabolism and virulence of the organism, its immunodominant antigens, and capacity to survive adverse conditions and persist intracellularly. Clinical isolates of M. tuberculosis differ in their virulence, potential for transmission in humans, and interaction with the host (immunopathology, induction of host cytokines, delayed-type hypersensitivity). For example,

CHAPTER 308 Tuberculosis  

ABSTRACT

Tuberculosis (TB) is a chronic infectious disease with a unique latent stage caused by the acid-fast bacillus (AFB) Mycobacterium tuberculosis. TB has had a worldwide resurgence in incidence in contemporary times, complicated by underlying HIV infection and drug resistance in many individuals. With treatment, the prognosis of patients with TB depends on the extent of pulmonary TB, the sites of extrapulmonary TB, drug susceptibility of the isolate, and the presence of HIV infection and other comorbid conditions. Miliary TB is associated with a high case-fatality rate, in part related to delays in diagnosis. TB meningitis is associated with serious neurologic residua, as well as high mortality. Multidrug-resistant (MDR) TB and extensively drug-resistant TB are accompanied by high rates of treatment failure, morbidity, and mortality. TB in HIV-infected persons is associated with high early mortality. The addition of ART can result in morbidities associated with concurrent administration of TB and HIV drugs and with TB-immune reconstitution inflammatory syndrome.

KEYWORDS

Mycobacterium tuberculosis primary TB reactivation TB reinfection TB pleural TB military TB tuberculous meningitis tuberculous lymphadenitis tuberculous pericarditis tuberculous peritonitis gastrointestinal TB renal TB vertebral osteomyelitis multidrug-resistant (MDR) TB extensively drug-resistant TB HIV coinfection

2000.e3

CHAPTER 308 Tuberculosis  

2001

TABLE 308-1 RISK FACTORS FOR TUBERCULOSIS RISK FACTOR

INCREASED RISK OF RECENT INFECTION*

Household contact of PTB

INCREASED RISK OF PROGRESSION FROM INFECTION TO DISEASE

>5 mm

X

Solid organ transplant recipients, immunosuppressive treatment (TNF inhibitors, prednisone >15 mg/day for >1 month), fibrotic lesions on chest radiograph consistent with prior TB

TST CUT POINT

X

>5 mm

X

>5 mm

HIV infection

X

Foreign-born, injecting drug users, TST-positive children, adolescents, young adults

X

>10 mm

Residents or workers in hospitals, homeless shelters, correctional facilities, nursing homes, residences for the HIV-infected

X

>10 mm

Underweight (>15%), silicosis, diabetes mellitus (particularly insulin-dependent or poorly controlled), renal failure, hemodialysis, gastrectomy, jejuno-ileal bypass, carcinoma of the head and neck, lung cancer, lymphoma, leukemia

X

>10 mm

>15 mm

None *Recent infection per se increases risk of progressing from infection to disease (12.9 cases per 1000 person-years in the first year compared to 1.6 per 1000 person-years in the subsequent 7 years). HIV = human immunodeficiency virus; PTB = pulmonary TB; TNF = tumor necrosis factor; TST = tuberculin skin test.

the hypervirulent Beijing strain family overexpresses a phenolic glycolipid that inhibits innate immunity and may thereby contribute to its pathogenicity. There are six main phylogeographic lineages of M. tuberculosis, each associated with a specific human population. The families differ in geographic distribution and in some cases the potential for transmission and pathogenesis. Strain typing is particularly useful in outbreak investigations and can be performed by several techniques, including restriction fragment length polymorphism of the insertional element IS 6110 or spoligotyping, and increasingly by deep DNA sequencing. The finding that multiple cases of TB are caused by the same strain and constitute a “cluster” suggests that they are epidemiologically linked although transmission may be recent or remote. Whole-genomic sequencing has emerged as a powerful tool to establish transmission even absent strong epidemiologic links. TB caused by M. africanum is clinically identical to that caused by M. tuberculosis. M. bovis has greater than 95% DNA homology with M. tuberculosis and causes disease in humans, cattle, deer, badgers, and other animals. The main route of transmission of M. tuberculosis is person-to-person through respiratory aerosols generated by coughing. Bacilli in small droplet nuclei (1 to 5 µm in diameter) remain suspended in air for long periods and once inhaled can reach the airways, where only 1 to 5 organisms are sufficient to cause infection. Laryngeal involvement renders the patient highly infectious. Direct cutaneous inoculation (“prosector’s wart”) does occur. M. bovis can be transmitted by the gastrointestinal route, usually through ingestion of contaminated dairy products.  

EPIDEMIOLOGY

The World Health Organization (WHO) estimates that in 2016 there were 10.4 million new cases of TB (10% in HIV-infected persons) and 1.7 million deaths (including 374,000 among people with HIV infection).2 An estimated 40% of deaths in persons with HIV were due to TB. Fifty-six percent of TB cases were in five countries: India, Indonesia, China, the Philippines, and Pakistan. Worldwide, 65% of TB cases occur in males; the prevalence of disease peaks in young adults, with major economic consequence. Globally, both TB incidence (declining 2% per year) and case-fatality rates have been falling. However, TB incidence decline needs to accelerate to 4 to 5% annually to reach the WHO’s 2020 End TB Strategy Milestones of 20% incidence reduction compared to 2015. HIV infection has a profound effect on the epidemiology of TB, promoting and accelerating progression from infection to active TB, and on both reactivation and reinfection disease. Seventy-four percent of HIV-infected TB cases are in sub-Saharan Africa, resulting in TB case rates as high as 1% in South Africa and Swaziland. In the United States in 2017 the incidence of TB was 2.8 per 100,000, with about 9000 new cases reported (about 6% were HIV-infected, 5% homeless, and 4.0% incarcerated).3 The incidence rate in foreign-born individuals was 13-fold higher than in those born in the United States and foreign-born individuals accounted for 69% of new cases. Rates were 30-fold higher in nonHispanic Asians than non-Hispanic whites. Of genotyped cases, 14% appear due to recent transmission versus 86% that appear to be reactivation of previous infection, which in foreign-born was due to acquisition in their country of

origin whereas U.S.-born more likely acquired when TB was more common domestically. There is a corresponding shift of age-specific prevalence of TB in the United States towards older adults. Country of origin is a large determinant of both the risk of latent TB infection and of TB disease. In a low-prevalence setting such as the United States, the prevalence of latent TB infection (defined as a positive tuberculin skin test or interferon-γ release assay [IGRA] but no active disease) is approximately 4%. An estimated 13 million people have latent TB in the United States.4 Those infected are at markedly increased risk of disease compared to uninfected, a risk that is further increased by medical comorbidities and other factors shown in Table 308-1. The risk is not homogeneous within groups affected, for example, by extent of immunosuppression in HIV or duration, severity, and control of diabetes. Smoking and alcoholism also confer an increased risk for TB, although smaller than the conditions listed. In the United States in 2016, 16.4% of persons with TB had diabetes, 1.3% reported injecting drugs, 6.8% reported using noninjectable drugs, and 10% reported excessive alcohol use. TB caused by drug-resistant organisms is a continuing threat.5 Multidrugresistant (MDR) TB (resistant to isoniazid [INH] and rifampin) and extensively drug-resistant TB (MDR plus resistance to fluoroquinolones [FQs] and to a second-line injectable [kanamycin, amikacin, or capreomycin]) are much more difficult and expensive to treat and in some cases may be incurable. In 2016, there were 600,000 new TB cases with rifampin resistance, the most effective first-line drug, of which 490,000 had MDR TB. Almost one-half (47%) of these cases were in India, China, and the Russian Federation. About 6.2% of MDR TB cases are extensively drug-resistant. Ninety-seven U.S. cases of MDR TB and one case of extensively drug-resistant TB were reported in 2016. Outcomes for drug-resistant TB remain poor, with only 54% of MDR TB patients and 30% of XDR TB patients successfully treated. This should improve with the introduction of more active drugs and drug regimens. The majority of cases of XDR TB in an area with a high TB burden were found to be probably due to transmission rather than to inadequate treatment, suggesting that control of the epidemic requires an increased focus on interruption of transmission.6  

PATHOBIOLOGY

Typically the chain of transmission of TB begins with an infectious case of pulmonary TB (Fig. 308-1). Infectiousness of a patient is determined by sputum smear status (3 to 4+ AFB), cough strength and frequency, the presence of cavitary lung disease, and the characteristics of the physical space shared with the source (ventilation and air recirculation). However, not all strongly AFB smear–positive patients with pulmonary TB are equally infectious, and there may be high transmitters, owing to host or bacterial factors or both. Only about 50 to 60% of strongly sputum smear–positive persons with pulmonary TB generate aerosols that contain viable organisms. Certain strains of M. tuberculosis also may be transmission-prone. In both low- and high-prevalence countries, exposure/infection may occur in the household. In this setting, where exposure may be intense and protracted, 50 to 75% of contacts become infected. The higher numbers result from studies in which repeated testing identifies all tuberculin skin test convertors. In

2002

CHAPTER 308 Tuberculosis  

Primary progressive TB (5%; 40%)

Treatment

Primary infection (50–75%)

No infection

Cured TB

Exposure

Reactivation TB (5%; 5–10%/year)

Latency Exposure

Exposure

Reinfection TB

Disease

Bacterial replication maintained at a subclinical level by the immune system

Active infection

Infection controlled with some bacteria persisting in non-replicating form

Quiescent infection

Infection eliminated in association with T-cell priming

Acquired immune response

Infection eliminated without priming antigen-specific T-cells

Innate immune response

Bacterial load?

Clinical disease

Effect of HIV infection

outbreaks occurring in residential shelters, hospitals, and prisons, M. tuberculosis infection or disease also has been documented after brief exposure. Recent data suggest that in high prevalence settings, community transmission may be more common than household transmission, although the sites in the community may not be known. Important variables that may explain differences in transmission include virulence of the organism, innate immunity, and susceptibility of the exposed populations (e.g., HIV infected). Human genetic factors such as polymorphisms in expression or regulation of toll-like receptors (TLR), pattern recognition receptors important to innate immunity, may modulate risk and expression of infection (interferon gamma release assay [IGRA] or tuberculin skin test) as well as risk and expression of disease. Predisposition to disease is seen with defects in interferon (IFN)-γ and interleukin (IL)-12 receptors, consistent with their role in adaptive immunity. Polymorphisms that modify inflammation (e.g., by affecting leukotriene A4 hydrolase) also may affect disease manifestations and response to therapy. Two models of the natural history of TB are shown in Figures 308-1 and 308-2. There is increasing evidence that the natural history represents a continuum rather than distinct entities of latent and active TB. Diagnostic biomarkers that stratify risk of progression from latent TB infection to active TB would be of enormous value for targeting public health interventions. Host blood RNA signatures are currently being studied as predictors of those most at risk of developing active tuberculosis and therefore preferentially receive prophylaxis. The lung is the site of most cases of reactivation TB. The hallmark of the pathology is granuloma formation with caseation necrosis and multinucleated Langerhans giant cells. The caseous material found in necrotic cavities contains AFB. M. tuberculosis multiplies exuberantly in the liquid caseum. Immunologically, expectorated sputum contains cytokines and both upregulators and downregulators of the immune and inflammatory response, the downregulation being dominant. Bronchoalveolar lavage shows a lymphocytic alveolitis, with an influx of immature macrophages representing monocytes attracted from blood. In sum, there is an active but well-regulated immune and inflammatory response concomitant with bacterial replication. As a consequence of the inflammation, extensive apoptosis occurs that might lead to the deletion of M. tuberculosis-responsive T cells, that may play a role in the requirement for a long duration of therapy. This is consistent with the recent finding of persistent areas of inflammation (by PET-CT scan) at the end of TB treatment, and in the susceptibility to reinfection TB. In HIV-infected persons with advanced immunosuppression, granulomas may be poorly formed or absent. Lung tissue is infiltrated with foamy epithelioid cells that are macrophages laden with AFBs. Caseation may or may not be present, but there is extensive inflammation and necrosis. Extrapulmonary TB can involve any organ. Persistence of organisms in areas that are relatively well oxygenated may explain the more frequent sites of reactivation, such as the apices of the lung, cortices of the kidney, and vertebral bodies.

Symptoms

FIGURE 308-1.  Natural history of TB. The proportion of individuals affected is shown in parentheses. Bolded figures are for HIV infection with severe immunosuppression. A number of medical risk factors besides HIV promote progression from Mycobacterium tuberculosis infection to disease (see Table 308-1).

FIGURE 308-2.  Tuberculosis infection as a spectrum. The outcome of infection by Mycobacterium tuberculosis is generally represented as a bimodal distribution between active tuberculosis (TB) and latent TB on the basis of the presence or absence of clinical symptoms. It is proposed that latent TB is usefully represented as part of a spectrum of responses to infection. One consequence of this model is that there may be a subpopulation within the group that is currently defined as having latent TB that should be preferentially targeted for preventive therapy. A second consequence is that efforts to develop drugs for effective treatment of latent TB would overlap the search for drugs that shorten treatment times for active TB. (From Barry CE 3rd, Boshoff HI, Dartois V, et al. The spectrum of latent tuberculosis. Rethinking the biology and intervention strategies. Nat Rev Microbiol. 2009;7:845-855, Figure 1.)

Several forms of extrapulmonary TB have a shared pathogenesis: discharge of a contiguous tuberculous focus into a serosal cavity, a brisk inflammatory reaction based on preexisting delayed-type hypersensitivity, fever, frequently negative smears of exudative fluid for AFBs, and sometimes a transiently negative tuberculin skin test. The site of infection that discharges may be a longstanding focus or one that was seeded during recent dissemination associated with primary infection. This basic scenario occurs in pleural TB, TB pericarditis, TB meningitis (the parameningeal focus is called a “Rich focus”), TB peritonitis, and TB arthritis. Pleural TB represents an in situ delayed-type hypersensitivity reaction with activation of TH1 helper T cells, abundant cytokines, including IFN-γ and tumor necrosis factor (TNF)-α, and apoptosis. In non–HIV-infected individuals,

CHAPTER 308 Tuberculosis  

2003

A

B

C

D FIGURE 308-3.  A, Ghon complex. B, Moderately advanced pulmonary tuberculosis (TB). C, Far advanced pulmonary TB. D, Pulmonary (left) and extrapulmonary (right) TB. (Radiographs courtesy Thomas M. Daniel, MD.)

organisms are sparse, which may be why self-cure of pleural TB can take place. However, in the absence of chemotherapy for TB, there is a high risk for TB recurrence, usually as pulmonary TB on the side contralateral to the effusion. In addition to reactivation of a latent focus, reinfection with M. tuberculosis may occur and progress to disease. Reinfection is more likely if the host is immunosuppressed or if there is repeated or intense exposure. Treated cases of pulmonary TB also are predisposed to reinfection disease as discussed earlier. Latent TB infection is over 70% protective against reinfection TB.  

CLINICAL MANIFESTATIONS

Primary Tuberculosis

Most cases of primary TB are unrecognized clinically except by conversion of the tuberculin skin test or IGRA. There may be fever, shortness of breath, nonproductive cough, and rarely erythema nodosum. Crepitations and focal wheezes may be present. Chest radiographs show small patchy opacities in the mid-lung fields, often with unilateral hilar lymphadenopathy. Upper or middle lobe collapse may also be seen as a result of bronchial compression by enlarged nodes or transient pleural effusion. Studies with positron emission tomographic computed tomography (PET-CT) scan show that most household contacts of infectious TB cases with a positive tuberculin skin test have mediastinal adenopathy that resolves with INH preventive therapy. HIVinfected and -uninfected persons with latent TB infection studied with this modality may show parenchymal uptake suggestive of subclinical disease. In most individuals (immunosuppression being the exception), the manifestations

of primary TB resolve without treatment, concurrent with the development of an adaptive immune response. During the subsequent period of clinical latency, evidence of the primary infection may be found as a small calcified parenchymal scar in the mid-lung fields (Ghon complex), sometimes associated with similar findings in the draining hilar nodes (Ranke complex) (Fig. 308-3A). A small scar caused by an arrested lesion in the apices of the lung is called a Simon focus.

Progressive Primary Tuberculosis

Failure to develop adaptive immunity is most common in young children, the elderly, and the immunocompromised. Progressive primary TB manifested as TB meningitis, miliary TB, or disseminated TB may develop in this setting. Primary infection also may progress to pulmonary TB within the first 1 to 2 years. In this case pulmonary TB usually is in the upper lobe and cavitary, distant from the site of primary infection. Recent data indicate that in highprevalence areas most pulmonary TB cases represent progressive primary disease. Clinically, progressive primary and “post-primary” or reactivation TB are indistinguishable.

Reactivation (Post-primary) Tuberculosis

The terms reactivation TB and post-primary TB are used interchangeably to connote that primary TB is followed by a variable period of at least 2 years of clinical latency, after which TB develops in the setting of existing delayed-type hypersensitivity/adaptive immunity: existing sensitization to

2004

CHAPTER 308 Tuberculosis  

mycobacterial antigens contributes importantly to the pathogenesis and clinical manifestations. Pulmonary TB is the most common form of reactivation TB. Typical clinical findings in pulmonary TB consist of the insidious onset of a productive cough, night sweats, anorexia, and weight loss. Fever is present in approximately one half of those affected. Patients may be asymptomatic and the diagnosis suggested only by a chest radiograph obtained for other reasons (subclinical TB). The sputum may be purulent, blood streaked, or frankly bloody. Pleuritic chest pain may occur when there is subpleural inflammation. Dyspnea is not a hallmark of pulmonary TB, in part because thrombosis of vessels limits the perfusion of inflamed areas, so that hypoxemia is not a prominent clinical feature. Physical examination may show dullness to percussion, low-pitched amphoric (hollow-sounding) breath sounds, and occasionally crepitations that may be post-tussive. Chest radiographs often reveal more disease than suggested by physical examination (see Fig. 308-3B and C). Typically (>95% of cases), lesions are found in the apical and posterior segments of the upper lobes and the superior (dorsal) segment of the lower lobe. There is a progression from patchy opacities and consolidation to cavitation reflective of liquefaction and caseation. Advanced imaging with PET-CT shows a heterogeneity in metabolic activity of different lesions within the same patient (Fig. 308-4), which may be associated with variable response to treatment and may in some cases persist even after sterilization has been achieved. Rupture and discharge into bronchi and intrabronchial spread may lead to disease in multiple areas, including the other lung (so-called TB bronchopneumonia). There may be involvement of the larynx and middle ear. Early cavities are thin walled and evolve into characteristic chronic thick-walled cavities. Ten percent of all cavities have an air-fluid level. There may be an associated pleural effusion or rarely, with rupture of cavities into the air space, pyopneumothorax. If the disease is minimal, it may best be seen on apical lordotic chest radiographs or on CT scan. Rarely, chest radiographs are normal, and the accompanying symptoms and positive sputum smears may be the result of endobronchial lesions or rupture of a tuberculous node into the bronchi. Healing, fibrosis, and contraction obliterate small cavities, although large cavities may persist and even become the eventual nidus for an aspergilloma or a “scar” carcinoma. In immunocompromised persons, the opacities may be located in the midand lower lung fields and be manifested as poorly resolving lobar or segmental pneumonitis, atelectasis, nodules, and cavities. In individuals with HIV whose CD4+ count exceeds 200/µL, pulmonary TB may be typical in its manifestation. At lower CD4+ counts, mid- and lower lung abnormalities are more common. At a CD4+ count below 100/µL, the findings may be quite atypical, with prominent hilar and mediastinal adenopathy, pleural disease, interstitial or miliary opacities, or any combination of these manifestations. This picture resembles primary TB and, in fact, may represent progressive primary TB or reinfection disease. Chest radiographs are normal in up to 20% of persons with culture-confirmed TB, sometimes in the presence of a smear that contains AFB. In this CD4 strata, disseminated and extrapulmonary TB are the rule, with or without concurrent pulmonary TB.

Reinfection Tuberculosis

Reinfection TB is clinically indistinguishable from other forms and an important pathogenetic mechanism in high transmission settings. A documented change in the DNA fingerprint or sequence or occurrence in an outbreak setting, including among patients hospitalized in TB wards, may be the only evidence supporting the diagnosis of reinfection TB. In high TB burden settings, individuals previously successfully treated for TB may have as high as a five-fold risk compared to those with no history of developing TB disease, reflecting the high frequency of reinfection events that make up a high proportion of active cases.

Extrapulmonary Tuberculosis

Approximately 20% of cases of TB in non–HIV-infected populations are extrapulmonary (see Fig. 308-3D). In areas endemic for TB, extrapulmonary TB often occurs concurrently with pulmonary TB and is more common in children and young adults, in whom it represents progressive primary infection. By contrast, in low-prevalence areas, isolated extrapulmonary TB is more common, and there is a shift to the elderly that represents reactivation TB. HIV infection is associated with a higher frequency of extrapulmonary disease, including the more serious forms, disseminated TB and TB meningitis.

Pleural Tuberculosis

Pleural TB occurs by direct extension when a subpleural caseous focus discharges into the pleural space or through hematogenous seeding.7 There may

FIGURE 308-4.  Positron emission tomographic computed tomography (PET-CT) imaging. An 18F-fluorodeoxyglucose (FDG) PET-CT scan of a patient with tuberculosis with extensive bilateral disease and a complete collapse of the left lung. The right lung also shows extensive disease throughout and illustrates the variability of FDG-PET uptake among lesions within even a single infected patient. The yellow star illustrates one lesion that fails to take up FDG that lies immediately adjacent to a string of three lesions that take up label avidly (red star). These different types of lesions respond to chemotherapy with different kinetics, indicating that they represent distinct bacterial subpopulations in different microenvironments. (From Barry CE 3rd, Boshoff HI, Dartois V, et al. The spectrum of latent tuberculosis. Rethinking the biology and intervention strategies. Nat Rev Microbiol. 2009;7:845-855, Figure 2.)

be concurrent pulmonary TB. Its peak occurrence is 3 to 6 months after primary infection. The typical manifestation is abrupt onset of fever, pleuritic chest pain, and cough. Occasionally there is an insidious presentation consisting of fever, weight loss, and malaise. If the pleural effusion is large enough, shortness of breath may be seen. Physical examination shows dullness to percussion and decreased breath sounds. Above the area of dullness there may be true egophony. Chest radiographs typically show unilateral pleural effusion, more frequently in the right hemithorax. Bilateral disease occurs in 10% of cases. Pleural effusions may be medium-sized, large, or, uncommonly, massive.

Miliary Tuberculosis

Miliary TB usually has an insidious manifestation consisting of fever, weight loss, night sweats, and little in the way of localizing symptoms or signs. There may be concurrent TB meningitis with associated symptoms. Physical examination may show choroidal tubercles (raised white-yellow plaques on funduscopic examination, present in 15% of cases), lymphadenopathy, and hepatomegaly. Chest radiographs may show multiple bilateral small opacities termed miliary infiltrates because of their resemblance to millet seeds. The findings on initial chest radiographs are often subtle and may be clear-cut only in retrospect after 3 months of follow-up. Performance of CT or high-resolution CT is useful because of its increased sensitivity (Fig. 308-5). A variant of miliary TB is disseminated areactive TB, as may occur in HIV-infected patients or those treated with TNF inhibitors. In this entity, chest radiographic findings may be even more minimal or absent. In the HIV-infected individual with advanced immunodeficiency, blood cultures are positive for M. tuberculosis in 20 to 40% of patients and may be the only laboratory manifestation of TB.

Tuberculous Meningitis

TB meningitis is usually characterized by less than 2 weeks of fever, headache, and meningismus.8 There may be depressed levels of consciousness, diplopia, and (rarely) hemiparesis. Physical examination shows a stiff neck and occasionally cranial neuropathy (VI, III, IV, VII in order of frequency) and longtract signs. Chest radiographs may be consistent with pulmonary TB or miliary TB. CT of the head may show contrast enhancement over the basilar meninges, hypodense areas consistent with infarcts, hydrocephalus, and sometimes focal inflammatory lesions (tuberculomas). CT angiography may show entrapment of vessels or vasculitis.

CHAPTER 308 Tuberculosis  

A

2005

B

FIGURE 308-5.  Miliary tuberculosis in a 70-year-old man. A, Posteroanterior chest radiograph shows evenly distributed, discrete, uniformly millet-sized nodular opacities in both lungs. B, High-resolution computed tomography (1.0-mm section thickness) at the level of the right upper lobar bronchus shows uniformly sized small nodules randomly distributed throughout both lungs. Note the subpleural and subfissural nodules (arrows). (From Jeong YJ, Lee KS. Pulmonary tuberculosis: up-to-date imaging and management. AJR Am J Roentgenol. 2008;191:834-844.)

Tuberculous Lymphadenitis

Lymphadenitis may be the sole manifestation of TB or, more frequently, particularly in the HIV infected, may accompany pulmonary TB. Patients with isolated lymph node disease may be afebrile. The supraclavicular and posterior cervical lymph nodes are most frequently involved, referred to as scrofula. This is in contrast to scrofula caused by atypical mycobacteria or M. bovis and often seen in children, in whom submandibular and high anterior cervical adenopathy predominates. The lymphadenitis is not usually painful although there may be tenderness and fluctuance, and aspiration of the lymph node with the finding of acid-fast bacilli (AFB) is an excellent approach to establish the diagnosis.

Tuberculous Pericarditis

The usual manifestation of TB pericarditis is chronic but may occasionally be subacute with fever, night sweats, chest pain, shortness of breath, pedal edema, and other signs of right heart failure.9 Physical examination shows signs of pericardial disease, right-sided heart failure, and tamponade (in ≈10%). Pericardial aspiration and biopsy are the diagnostic procedures of choice. When the pericardial effusion is large or tamponade is present, a pericardial window can be both diagnostic and therapeutic.

Tuberculous Peritonitis

TB peritonitis may be accompanied by abdominal pain and fever, at times mimicking an acute abdomen. Alternatively, there may be an insidious presentation consisting of abdominal pain, swelling, night sweats, and weight loss. The clinical syndrome is caused by discharge of tuberculous lymph nodes into the peritoneal space. Exudative ascites is usually present unless TB is superimposed on preexisting transudative ascites, as in alcoholic liver disease. On physical examination, the abdomen has been described as “doughy,” because matted loops of bowel may be palpable. A variant of this syndrome is perhaps best termed abdominal TB. In this case, the abdominal pain is subacute, the associated findings on physical examination less striking, and ascites less prominent or absent. The best method for diagnosis when ascites is present is laparoscopically guided peritoneal biopsy. In areas endemic for TB and HIV, the finding of intra-abdominal lymphadenopathy on abdominal ultrasound or CT is often used to support the diagnosis of abdominal TB.

Gastrointestinal Tuberculosis

Patients with gastrointestinal TB have fever, abdominal pain, diarrhea, and gastrointestinal bleeding or obstruction. Roentgenograms of the small bowel and abdominal CT show involvement of the terminal ileum, similar to Crohn disease. The diagnosis is made on clinical suspicion in areas endemic for TB and HIV or by the finding of TB elsewhere. Occasionally, intraluminal biopsy of the terminal ileum or other involved sites is used to establish the diagnosis.

Renal Tuberculosis

There may be few symptoms and signs associated with renal TB, although occasionally dysuria, hematuria, and flank pain are present. The diagnosis is

often suggested by the finding of sterile pyuria or hematuria as initial abnormalities that trigger evaluation. Physical examination is usually unremarkable. CT shows renal cortical scarring, occasionally with mass or cavitary lesions, papillary necrosis with calyceal and ureteral dilation, or “beading” of the ureter because of ureteral strictures.

Vertebral Osteomyelitis

The initial site of disease is the subchondral region of the anterior portion of the vertebral body.10 The lower thoracic and lumbar vertebrae are involved most commonly. The disc space is initially spared but becomes involved late with spread to adjacent vertebrae. Paravertebral “cold abscesses” may dissect through tissue planes. Patients have back and sometimes radicular pain. Occasionally and more often with cervical disease, there may be weakness of the legs and incontinence of stool and urine. Physical examination may show a gibbus deformity caused by anterior compression fractures or paraparesis. Radiographs of the spine, as well as CT and magnetic resonance imaging, may show abnormalities in adjacent vertebrae, with anterior compression (see Fig. 308-3D). Cold abscesses may be appreciated as well.

Other Forms of Extrapulmonary Tuberculosis

TB of the bone or joints may be manifested subacutely as a combination of synovitis and osteomyelitis. The joints involved may have sustained previous trauma. TB of the female genital tract may result in pelvic pain, menorrhagia, vaginal discharge, or infertility. Males may have an epididymal mass, sometimes seen in patients with miliary TB. TB also can cause granulomatous uveitis as well as phlyctenular keratitis.  

DIAGNOSIS

Infection with Mycobacterium Tuberculosis

The diagnosis of latent TB infection (we will use the term LTBI because it still is standard, although a better term is MTB infection) is based on the finding of delayed-type hypersensitivity to mycobacterial antigens and the absence of clinically active TB. The tuberculin skin test has been used widely, and there is strong epidemiologic evidence supporting its interpretation. Tuberculin purified protein derivative (PPD) derived from autoclaved culture filtrates of M. tuberculosis is used to elicit delayed-type hypersensitivity. The response elicited by PPD is nonspecific because of broad cross-reactivity among tuberculous and nontuberculous mycobacteria and other organisms as well. The tuberculin skin test is performed by injecting 5 tuberculin units of PPD in 0.1 mL intradermally. The reaction is assessed as induration after 48 to 72 hours. Problems with the tuberculin skin test are legion. It is the only bioassay used in clinical medicine, and its accuracy depends strongly on correct user application and interpretation The sensitivity of the tuberculin skin test is less in immunosuppressed patients, such as those with HIV infection, and also less in the presence of active TB. The tuberculin skin test may revert to negative over time, and it may be boosted by the repeat application of PPD. On repeat testing the result is a “pseudoconversion” that does not represent new infection with M. tuberculosis. The greatest limitation of the tuberculin skin test is its nonspecificity. There has been uncertainty in its

2006

CHAPTER 308 Tuberculosis  

interpretation, particularly in the setting of previous vaccination with M. bovis bacille Calmette-Guérin (BCG). In fact, BCG administered once at birth has little effect on the tuberculin skin test beyond the first year. By 10 years of age, only 1% of positive tuberculin skin tests can be ascribed to previous BCG administration. The tuberculin skin test remains of value in the diagnosis of latent TB infection in at-risk individuals who are candidates for treatment (preventive therapy). Interpretation of the tuberculin skin test is based on a “sliding scale” that takes into account an individual’s a priori risk for M. tuberculosis infection (see Table 308-1). Changing the cut point for positives, in effect, modifies the sensitivity and specificity of the tuberculin skin test. Routine testing is not recommended for low-risk populations; however, testing may be performed because of employment. Certain populations should undergo annual testing, including staff or individuals living or working in congregate settings (hospital staff, incarcerated, homeless, HIV-infected, correctional facility staff), injection drug users, and others at risk because of sociodemographic factors. For individuals who will undergo an annual tuberculin skin test, a true baseline should be established by two-step skin testing. After the initial negative tuberculin skin test, the test is repeated in 1 to 3 weeks (there is no need to repeat the test if the first tuberculin skin test is positive). An increased reaction size on the second tuberculin skin test is known as “boosting” and may be due to previous infection with non-tuberculous mycobacteria or M. tuberculosis or vaccination with BCG. Tuberculin skin test conversion from negative to positive is the best indicator of intervening new infection with M. tuberculosis and is defined as an increase in reaction (induration) size of 6 mm or greater from less than 10 mm to 10 mm or greater. It also may be of value to perform a two-step tuberculin skin test on individuals older than 60 years and therefore at risk for tuberculin skin test reversion. In HIV infection, the tuberculin skin test may be negative before administration of antiretroviral therapy (ART) and convert to positive with treatment. For this reason, it is recommended that the tuberculin skin test be repeated in tuberculin skin test–negative HIV-infected persons once their CD4+ count reaches 200/µL and annually thereafter. The sensitivity of the tuberculin skin test is decreased in individuals with active TB, more so in certain forms of extrapulmonary TB. Though insensitive for the diagnosis of active TB, the tuberculin skin test has another application in low-prevalence areas. If the differential diagnosis of a clinical condition includes TB, establishment of previous M. tuberculosis infection increases the likelihood that the clinical findings represent TB. The tuberculin skin test is of particular value in the evaluation of smear-negative patients with pulmonary disease suggestive of pulmonary TB and in patients suspected of having extrapulmonary TB. Confirmatory findings such as positive PCR, culture, or histology, response to therapy, and lack of an alternative diagnosis are necessary to establish the diagnosis of TB. Interferon gamma releasing assays (IGRAs) (QuantiFERON-TB PLUS, T-test) have also been approved for the diagnosis of latent TB infection, and the CDC considers them interchangeable with the tuberculin skin test for individuals over 2 years of age. The assays represent in vitro cell culture in which blood cells are stimulated with a mix of antigens present in M. tuberculosis but not in BCG and most nontuberculous mycobacteria. The main advantage of IGRAs is specificity and ability to have interpretable results with only one patient visit. The disadvantages of IGRAs are expense, technical requirements, and controversy over test sensitivity in certain situations such as HIV infection and in household contacts of patients with pulmonary TB. Another issue has been instability of the result in individuals undergoing annual testing with IGRAs close to the cut point. That said, IGRAs have come to replace TST in many settings due to the single visit requirement, lack of need for operator experience, and higher specificity. Another factor is the current worldwide shortage of PPD. Because treatment of latent TB infection is effective in tuberculin skin test–positive HIV-infected persons, and the risk for progression of infection is inordinately high in such individuals, it is important to perform tests with high negative predictive value. Therefore, both a tuberculin skin test and IGRA should be performed in HIV-infected persons and others at high risk of progression from M. tuberculosis infection to disease or of a poor outcome. Ideally the blood for IGRA should be drawn prior to TST placement to avoid falsepositive IGRA from the PPD. If either the tuberculin skin test or the IGRA is positive, the individual is a candidate for treatment of latent TB infection. Recent studies indicate that IGRAs may be less sensitive than tuberculin skin test in household contacts recently infected with M. tuberculosis, a particularly high-risk group. This may be due to delayed conversion of IGRAs relative to tuberculin skin test. Mathematical modeling suggests IGRAs should replace

tuberculin skin test in immunocompromised and perhaps all individuals, and that despite their increased cost, they are cost-effective in the United States. Unfortunately, modeling currently is based on data that are quite variable between studies. A fourth-generation test, Quantiferon-Gold PLUS also contains CD8 stimulatory antigens and is supplanting earlier generations of this IGRA. Initial studies indicate similar performance to the Quantiferon-GOLD assay. In high TB–prevalence areas, the World Health Organization (WHO) recommends preventive therapy for all HIV-infected persons because of the difficulty in implementing a tuberculin skin test program that would identify those who will achieve the greatest benefit of INH preventive therapy.

Active Tuberculosis

In countries with a high TB burden, the diagnosis of TB is often based on clinical symptoms and sputum microscopy. Clinical diagnosis without the benefit of culture confirmation or radiography is the norm in endemic countries where access to diagnostics is limited. The diagnosis is also made on clinical grounds alone when smears are negative and suspicion for TB is high. Clinical diagnosis is particularly important in HIV-infected persons because of the risk for rapid progression of the TB if left untreated, the more frequent occurrence of AFB smear–negative pulmonary TB, and in those with forms of TB that are “paucibacillary” (pediatric, meningeal, miliary, abdominal, pleural, pericardial), in which bacteria are few and AFB smears typically negative. The diagnosis of miliary, abdominal, pleural, and pericardial TB may be confirmed by the finding of AFBs in biopsied tissue or by culture. In the absence of bacteriologic confirmation, either because cultures (and GenXpert MTB/RIF) are unavailable or because they are negative, the final diagnosis often relies on response to therapy or establishment of an alternative diagnosis. It should be noted that the empirical approach, taken of necessity in resource-limited settings, leads to overdiagnosis and overtreatment of TB, which expends TB program resources and delays treatment of other infections. It is therefore preferable to attempt to establish a definite diagnosis based on the demonstration of M. tuberculosis by smears, cultures, or nucleic acid amplification tests in infected secretions or tissue specimens. Sputum microscopy is the standard approach to the diagnosis of pulmonary TB. A smear requires 1000 to 10,000 bacilli/mL to be read as positive. Both hot and cold carbol fuchsin methods (Ziehl-Neelsen and Kinyoun) are used extensively. The use of fluorochrome stains such as auramine-rhodamine allows more rapid screening of sputum smears and improves sensitivity by about 10%. Three specimens, preferably early morning samples, should be examined to establish the diagnosis. Yield is higher in the presence of cavitary lung disease. Approximately one half of individuals with pulmonary TB are AFB sputum smear negative, and this proportion is higher in the HIV infected. The quantity of AFBs present in the sputum smear is a rough measure of the infectiousness of patients with pulmonary TB, and it is a convenient way to monitor response to treatment. On this basis, additional roles have evolved for the sputum smear as a tool to monitor the potential for transmission and response to therapy. HIV-infected persons with pulmonary TB carry a particular risk for transmission to health care workers and have been documented as sources of nosocomial outbreaks. Therefore, in the United States, in areas of high prevalence of TB, HIV-infected persons with pulmonary symptoms should be placed into respiratory isolation until infectious TB can be reasonably excluded by three negative sputum smears for AFB on specimens separated by at least 8 hours. The diagnosis of pediatric TB has always been problematic. Children do not produce sputum readily, and TB is often noncavitary, extrapulmonary, or both. Sputum samples may not be readily obtained from infants and children. Options in this case include sputum induction and gastric aspiration. The sensitivity of AFB smears of gastric aspirates and induced sputum is 25 to 30%. Recent studies indicate that induced sputum, culture of nasopharyngeal swabs, and GenXpert on stools have high enough yield so that gastric aspirates rarely are necessary. Nonetheless, the diagnosis often is based on clinical and epidemiologic features as well as response to therapy. Bronchoscopy with bronchoalveolar lavage or transbronchial biopsy is another option for the diagnosis of TB and is useful in all severely ill individuals and the immunocompromised, in whom the diagnosis of TB or an alternative infection must be made quickly if treatment is to have an impact on patient outcome. Additionally, a post-bronchoscopy induced sputum has been shown to have higher yield. AFB smears should be performed on normally sterile fluid obtained from all patients suspected of having TB. The yield of smears from pleural fluid, pericardial fluid, ascitic fluid, and cerebrospinal fluid (CSF) is low in patients

CHAPTER 308 Tuberculosis  

with TB but may be higher in those with HIV coinfection, particularly if immunodeficiency is advanced. In TB meningitis, CSF may clot spontaneously, and AFB stains of the clot have increased yield. Rapid diagnosis of some forms of extrapulmonary TB may be made by biopsy of tissue (pleural, pericardial, peritoneal, synovial, terminal ileum); the presence of granulomas, particularly if necrotizing, virtually confirms the diagnosis. Necrotizing granulomas are seen in TB and fungal diseases (particularly histoplasmosis, blastomycosis, coccidioidomycosis, and sporotrichosis). AFBs may also be seen in tissue histology, and M. tuberculosis cultured from the specimen. The diagnosis of miliary TB can be suggested by CT of the chest and confirmed by transbronchial lung biopsy (highest yield), as well as biopsy of the liver, bone marrow, or abnormal lymph nodes. If TB meningitis is suspected and the patient is immunosuppressed, it is particularly important to exclude cryptococcal meningitis by performing a cryptococcal polysaccharide antigen test, as well as an India ink preparation on CSF sediment. The diagnosis of TB from specimens of normally sterile fluid can be difficult and is increased by culturing relatively large volumes. In addition, the yield of biopsy and culture of tissue (pleura, pericardium) is additive. There are particular diagnostic features for various forms of extrapulmonary TB. In TB meningitis, the initial CSF examination may show neutrophil predominance, but this evolves into a lymphocytic meningitis (100 to 500 cells/µL) with high protein and depressed glucose. TB of the pleura, pericardium, and peritoneum is associated with an exudative effusion, often with a lymphocyte predominance. Low glucose may be found in 20% of TB effusions but limits the differential diagnosis considerably. For example, malignancy, empyema, and rheumatoid arthritis are the other causes of pleural effusion with low glucose. Pericardial fluid in patients with TB pericarditis may be bloody. Eosinophilic meningitis and chylous pleural effusions or ascites may also be seen in TB. Currently the gold standard for the diagnosis of TB is culture on solid (Löwenstein-Jensen) or in liquid (BACTEC MGIT 960 system) media. The mycobacteria growth indicator tube (MGIT) system is non-radiometric and based on oxygen quenching in the presence of replicating mycobacteria. When compared with solid media, culture with liquid media is more sensitive and growth is more rapid (1 to 3 weeks vs. 3 to 8 weeks for solid media). Once an isolate is available, drug susceptibility testing should be performed to guide therapy. This takes an additional 2 to 4 weeks on solid media, although isoniazid (INH) and rifampin susceptibility results are available in several days when the molecular line probe assay (described later) is used. Liquid medium can be inoculated with smear-positive specimens for direct drug susceptibility testing, which also accelerates the process. Once mycobacterial growth occurs, speciation is possible within hours with commercially available DNA probes. Nuclear acid amplification tests are approved and commercially available for use in TB diagnostics. Their sensitivity is somewhat higher than that of AFB smears, and their specificity is excellent. Expense precludes routine use of such tests, however. The Xpert MTB/RIF has transformed TB diagnosis globally. Through in situ DNA amplification reaction, this test allows a specific diagnosis of TB and determination of susceptibility to rifampin within 90 minutes. After minimal processing, sputum is added to a cartridge. Gene amplification is done with primers based on the rpoB gene, which encodes the target of rifampin, and resistance-conferring mutations are detected. This method is capable of establishing the diagnosis of TB in 97% of patients with pulmonary TB, including 98% of AFB sputum smear–positive and 73% of smear-negative individuals, thus rivaling the sensitivity of solid culture. It does not require molecular expertise by the technician and is not subject to amplicon (DNA) contamination, because it is a closed system. The uptake of GenXpert has been remarkable. The government of South Africa has replaced sputum smear analysis with GenXpert for TB diagnosis. The government of Brazil is developing a similar policy. In Uganda, there will be a single reference laboratory for culture and drug susceptibility testing (DST). GenXpert will be available regionally and used mainly for the diagnosis of smear-negative cases. GenXpert is priced differently for low-income countries, but cost still may be prohibitive in some settings. The test allows more rapid diagnosis and initiation of treatment but has not led to more treated cases in high burden settings because many smearnegative patients were receiving empiric treatment if clinical suspicion was present. A new Xpert MTB/RIF Ultra cartridge was endorsed by the WHO as a replacement for the Xpert MTB/RIF cartridge in March, 2017. The cartridge has increased sensitivity, particularly among paucibacillary cases, but a slight decrease in specificity (from 98% to 96%). Encouraging is the test’s reported increased capacity to detect TB meningitis, in which sensitivity for probable or definite infection was 70% (95% CI 47 to 87) for Xpert Ultra,

2007

compared with 43% (23 to 66) for Xpert MTB/RIF, and 43% (23 to 66) for culture. There are other candidate nucleic acid amplification tests (NAATs) about to undergo evaluation that are less expensive than GenXpert and may be truly point of care. A1  For example, elevated blood levels of BATF2 (basic leucine zipper transcription factor 2) are a sensitive biomarker for active TB.11 Diagnostics for pediatric and extrapulmonary TB more sensitive than NAATs are likely to be based on host responses. As regards host-based diagnostics, there are promising data based on transcriptomics, proteomics, and metabolomics. For example, a three-gene transcriptional signature proposed for the diagnosis of TB is likely to be commercialized, and a 16-gene transcriptional signature of risk indicates increased risk of developing TB within a 2-year time frame. As host-based diagnostics become available, a new category of subclinical TB may become easier to diagnose, a category which has been challenging because by definition there are no symptoms and bacterial-based diagnostics are usually negative. The appropriate management of such patients is not clear because they may spontaneously resolve their minimal disease. If they are HIV infected there will be the risk of “unmasking TB” with the start of ART. Even HIVuninfected patients they represent a high risk group with LTBI and probably should be treated as such. TB in the HIV-infected patient poses particular diagnostic issues because of the increased likelihood of smear-negative pulmonary TB, and in advanced HIV, atypical presentation and extrapulmonary disease. The Alere Determine LAM TB test is a low-cost, rapid, lateral flow device (dipstick) that detects lipoarabinomannan, the major cell wall glycolipid of M. tuberculosis. Its use as a TB diagnostic test improved survival of HIV-infected patients admitted to hospital with low CD4 counts (15 mm) more commonly signify tuberculosis. With the progressive decline in active tuberculosis in the United States, nontuberculous mycobacteria are likely to account for significant proportions of PPD reactivity. Newer IFN-γ release assays (IGRAs) incubate blood with relatively tuberculosis-specific recombinant proteins and elicit T-cell secretion of IFN-γ, thereby helping to clarify whether PPD reactivity is due to tuberculosis; however, M. kansasii, M. szulgai, and M. marinum cross-react in some IGRAs. Isolation of nontuberculous mycobacteria from blood specimens is clear evidence of disease. However, because the slow-growing nontuberculous mycobacteria typically do not grow well in routine blood culture media, the diagnosis must be suspected to be made. Isolation of nontuberculous mycobacteria from a biopsy specimen is strong evidence of infection, but cases of laboratory contamination do occur. Identification of organisms on stained sections of biopsy material confirms the authenticity of the culture. Some unusual nontuberculous mycobacteria require lower incubation temperatures or special additives for growth (e.g., M. haemophilum). The radiographic appearance of nontuberculous mycobacterial disease in the lung ranges from normal to nodules, bronchiectasis, air space disease, and extensive cavity formation, similar to that seen in tuberculosis (Fig. 309-2). Isolation of nontuberculous mycobacteria from respiratory samples presents special problems in both sensitivity and specificity. M. gordonae is often recovered from respiratory samples and is almost never thought to be a real pathogen. Many patients, especially those with bronchiectasis, will occasionally have nontuberculous mycobacteria recovered from sputum culture without such mycobacteria being seen on smear. Specific criteria for definitive diagnosis of

nontuberculous mycobacterial lung disease exist for MAC, M. abscessus, and M. kansasii, but they are probably good guidelines for other nontuberculous mycobacteria as well. A positive diagnosis requires that two of three sputum samples grow nontuberculous mycobacteria, regardless of smear findings; a positive bronchoscopic alveolar sample, regardless of smear findings; or a biopsy specimen of pulmonary parenchyma with granulomatous inflammation or mycobacteria found on section and nontuberculous mycobacteria on culture. Once isolated, identification of nontuberculous mycobacteria is important because it will determine the broad class of antimycobacterial therapy to be used. Many laboratories now use DNA probes to identify MAC, M. gordonae, and M. kansasii.9 Drug susceptibility testing is of limited and largely unproven value, although clarithromycin susceptibility testing for MAC and rifampin susceptibility testing for M. kansasii are indicated. Initial isolates of MAC that have not been exposed to macrolides are almost always susceptible to macrolides. Any nontuberculous mycobacteria that have resisted a course of antimicrobials should probably be tested for antibiotic susceptibility as well.  

TREATMENT  It is rarely an emergency to initiate treatment of nontuberculous mycobacterial infections, which are relatively slow-growing chronic infections that evolve over a period of weeks to years, not hours to days. Therefore, empirical therapy is not usually needed, and identification of the species is advisable before starting complex, often poorly tolerated and potentially toxic regimens.10 Similar to the case with tuberculosis, single-drug therapy is almost always associated with the emergence of antimicrobial resistance and is strongly discouraged. MAC infection frequently requires complex multidrug therapy, the foundation of which is a macrolide (clarithromycin or azithromycin), ethambutol, and a rifamycin (rifampin or rifabutin). For disseminated nontuberculous mycobacterial disease in HIV-infected patients, the use of rifamycins poses special problems of drug interactions with protease inhibitors. For pulmonary MAC disease, three-times-weekly administration of drugs has been used successfully. The duration of therapy is prolonged, generally for 12 months after culture conversion and typically for a total of at least 18 months. Other drugs with activity against MAC include aminoglycosides, fluoroquinolones, and clofazimine. Adjuvant pulmonary resection is associated with a relatively high complication rate but also can provide a high level of success for selected patients who respond poorly to antibiotic treatment alone.11 M. kansasii lung disease is similar to tuberculosis in many ways and is also effectively treated with isoniazid (300 mg/day), rifampin (600 mg/day), and ethambutol (15 mg/kg/day). Treatment should continue until cultures have been negative for at least 1 year. Other drugs with very high activity against M. kansasii include clarithromycin, fluoroquinolones, and aminoglycosides. For skin ulcers caused by M. ulcerans, which is sensitive to temperatures above 37° C, local thermotherapy is highly effective and safe.12 Rapidly growing mycobacteria pose special therapeutic problems. Extrapulmonary disease in an immunocompetent host is usually due to inoculation (e.g., surgery, injection, trauma) or line infection and is often treated successfully with a macrolide and another drug (based on in vitro susceptibility), along with removal of the offending focus. By comparison, pulmonary disease, especially that caused by M. abscessus, is extremely difficult to eradicate, although repeated courses of treatment are usually effective in reducing the infectious burden and symptoms. Therapy generally includes a macrolide along with an intravenous agent such as amikacin, a carbapenem, cefoxitin, or tigecycline. Other oral agents used according to in vitro susceptibility testing and tolerance include fluoroquinolones, doxycycline, and linezolid. Inhaled amikacin may be an option for treatment-refractory pulmonary infections.13 Treatment of the other nontuberculous mycobacteria is less well defined, but macrolides and aminoglycosides are usually effective, with other agents added as indicated. Expert consultation is strongly encouraged for difficult or unusual nontuberculous mycobacterial infections.



FIGURE 309-2.  Chest computed tomography in a patient with severe pulmonary Mycobacterium abscessus infection. Arrows indicate bronchiectasis. Note the extensive left upper lobe destruction and diffuse pleural reaction. In addition, the left lung is smaller than the right as a result of extensive loss of lung parenchyma.

PREVENTION

Prophylaxis of MAC disease in patients infected with HIV is started when the CD4+ T-lymphocyte count is less than 50 cells/µL. Azithromycin 1200 mg weekly, clarithromycin 1000 mg daily, and rifabutin 300 mg daily are effective.

PROGNOSIS

The effect of nontuberculous mycobacterial infection on longevity is closely tied to the underlying condition (e.g., IFN-γ/IL-12 pathway defect, cystic fibrosis). With no or inadequate treatment, symptoms are intrusive, and the infections can lead to fatal complications, including overwhelming infection or severe lung destruction.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 309  The Nontuberculous Mycobacteria  

GENERAL REFERENCES 1. Spaulding AB, Lai YL, Zelazny AM, et al. Geographic distribution of nontuberculous mycobacterial species identified among clinical isolates in the United States, 2009-2013. Ann Am Thorac Soc. 2017;14:1655-1661. 2. Busatto C, Vianna JS, da Silva LVJ, et al. Mycobacterium avium: an overview. Tuberculosis (Edinb). 2019;114:127-134. 3. Wu UI, Holland SM. Host susceptibility to non-tuberculous mycobacterial infections. Lancet Infect Dis. 2015;15:968-980. 4. Park IK, Olivier KN. Nontuberculous mycobacteria in cystic fibrosis and non-cystic fibrosis bronchiectasis. Semin Respir Crit Care Med. 2015;36:217-224. 5. Flume PA, Chalmers JD, Olivier KN. Advances in bronchiectasis: endotyping, genetics, microbiome, and disease heterogeneity. Lancet. 2018;392:880-890. 6. Gonzalez-Santiago TM, Drage LA. Nontuberculous mycobacteria: skin and soft tissue infections. Dermatol Clin. 2015;33:563-577.

2013.e1

7. Misch EA, Saddler C, Davis JM. Skin and soft tissue infections due to nontuberculous mycobacteria. Curr Infect Dis Rep. 2018;20:1-17. 8. Abubakar I, Gupta RK, Rangaka MX, et al. Update in tuberculosis and nontuberculous mycobacteria 2017. Am J Respir Crit Care Med. 2018;197:1248-1253. 9. Ustinova VV, Smirnova TG, Sochivko DG, et al. New assay to diagnose and differentiate between Mycobacterium tuberculosis complex and nontuberculous mycobacteria. Tuberculosis (Edinb). 2019;114:17-23. 10. Philley JV, Griffith DE. Medical management of pulmonary nontuberculous mycobacterial disease. Thorac Surg Clin. 2019;29:65-76. 11. Mitchell JD. Surgical treatment of pulmonary nontuberculous mycobacterial infections. Thorac Surg Clin. 2019;29:77-83. 12. Vogel M, Bayi PF, Ruf MT, et al. Local heat application for the treatment of Buruli ulcer: results of a phase II open label single center non comparative clinical trial. Clin Infect Dis. 2016;62:342-350. 13. Olivier KN, Griffith DE, Eagle G, et al. Randomized trial of liposomal amikacin for inhalation in nontuberculous mycobacterial lung disease. Am J Respir Crit Care Med. 2017;195:814-823.

2013.e2

CHAPTER 309  The Nontuberculous Mycobacteria  

REVIEW QUESTIONS 1. Initial Mycobacterium avium complex (MAC) susceptibility testing is recommended and validated for which of the following? A . Rifamycins B. Macrolides C. Aminoglycosides D. Ethambutol E. All of the above Answer: B  The only antibiotic class for which there is evidence for in vitro testing to predict in vivo value is macrolides. The other agents have limited predictive value in in vitro testing for MAC. Therefore, only macrolide testing is recommended and interpretable with break points. 2. Pulmonary MAC infection is associated with which of the following underlying conditions? A . Cystic fibrosis B. Primary ciliary dyskinesia C. Bronchiectasis D. Pneumoconiosis E. All of the above Answer: E  Isolated pulmonary MAC infection has been clearly associated with defects in respiratory ciliary function and channel defects affecting the respiratory epithelium. 3. Disseminated MAC infection is associated with which of the following conditions? A . Mutations in the interleukin (IL)–12 receptor B. Mutations in the interferon (IFN)-γ receptor C. Mutations in STAT1 D. Tumor necrosis factor (TNF)–inhibiting antibodies E. All of the above Answer: E  The IL-12/IFN-γ pathway is critical for the control of intracellular organisms including nontuberculous mycobacteria. IFN-γ also induces TNFα, linking these two pathways together.

4. The interferon gamma release assays (IGRA) use antigens found in tuberculosis (e.g., ESAT-6, CFP-10) to elicit interferon gamma in vitro. Which nontuberculous organisms cross-react in this assay? A . M. gordonae, M. smegmatis, M. abscessus complex B. MAC, M. neoaurum, M. chelonae C. M. kansasii, M. szulgai, M. marinum D. BCG, M. genavense, M. hemophilum E. M. ulcerans, M. xenopi, M. simiae Answer: C  M. kansasii, M. szulgai, and M. marinum contain some of the same antigens as MTB, yielding a positive IGRA, and should be considered in the case of a positive IGRA test. 5. Nontuberculous mycobacterial osteomyelitis is rare outside of genetic immunodeficiency. Which immunodeficiency is it most associated with? A . Chronic granulomatous disease B. Anti–TNF antibody therapy C. Recessive deficiency of IFN-γR D. Dominant negative deficiency of IFN-γR E. Advanced HIV Answer: D  Osteomyelitis due to NTM and bacille Calmette Guérin (BCG) is strongly associated with the autosomal dominant form of IFN-γ receptor 1 deficiency. Isolated osteomyelitis due to NTM is much less common in the recessive forms of IFN-γ receptor 1 deficiency and very unusual in other genetic defects.

CHAPTER 310  Leprosy (Hansen Disease)  

310  LEPROSY (HANSEN DISEASE) JOEL D. ERNST



DEFINITION

Leprosy (Hansen disease) is a chronic infection caused by Mycobacterium leprae, an acid-fast slowly growing bacterium that cannot yet be cultured in vitro. Leprosy is found worldwide, although three countries of high prevalence (India, Brazil, and Indonesia) currently account for more than 80% of reported cases.1 The primary manifestations of infection with M. leprae occur in the skin and peripheral nerves. The skin lesions of leprosy are classically hypopigmented, hypoesthetic or anesthetic, and nonpruritic. Peripheral nerves can be damaged by direct infection with M. leprae or by the immune response to the infection; the result is loss of sensation and motor function. Additional morbidity is due to the peripheral nerve dysfunction, including painless traumatic and burn injuries, secondary bacterial infections, and muscle atrophy and contractures. Leprosy per se is not a cause of death, but the debility associated with leprosy contributes to the severity of poverty and the likelihood of death from malnutrition or other infections. Despite the low transmissibility of M. leprae and the ability of multiple-drug therapy to cure leprosy, it remains a stigmatized disease that can pose a challenge to diagnosis and therapy.

The Pathogen

M. leprae is an acid-fast bacillus that contains a mycolic acid–rich cell wall and a single membrane. Despite nearly 150 years of effort, M. leprae remains uncultivatible in vitro. For biochemical and structural characterization, M. leprae can be grown in large quantities in nine-banded armadillos (Dasypus novemcinctus), and inoculation of the footpads of athymic mice allows semiquantitation of viable bacilli.  

EPIDEMIOLOGY

Leprosy is found worldwide, although endemic leprosy is absent from northern Europe, where it was present in epidemic form as recently as the 19th century. The global prevalence of leprosy is about 175,000 known cases, and the current incidence is about 211,000. By definition of the World Health Organization (WHO) Strategic Plan for the Elimination of Leprosy, a newly diagnosed patient who has been treated with multidrug therapy is removed from the prevalence registry, which explains the lower prevalence than incidence of this chronic infection. Since initiation of the WHO Strategic Plan (whose

Bacteriologic:

Clinical:

goal is to eliminate leprosy as a public health problem, i.e., a prevalence of less than 1 in 10,000 in all regions), an estimated 16 million people have been cured of leprosy. The success of multidrug therapy notwithstanding, leprosy remains a public health problem in 14 countries. India, Brazil, and Indonesia currently have the largest number of cases. Although domestic transmission of leprosy is extremely rare in the United States, 178 cases of leprosy were diagnosed in 2015, including cases in immigrants from India, Brazil, the Philippines, the Dominican Republic, and Mexico. Because leprosy is not highly transmissible, it is not considered a disease of travelers other than immigrants. Inability to culture M. leprae in vitro has been a major hindrance to understanding the modes of transmission and reservoirs of the organism. Observational studies reveal a low frequency of leprosy in casual travelers or temporary residents of high-incidence regions, thus indicating that M. leprae is not highly transmissible. Even in areas of high incidence, clusters of leprosy are rare outside families or others with prolonged close contact. It is believed that transmission of M. leprae commonly occurs through the respiratory route, because nasal secretions of people with lepromatous leprosy may contain 107 viable bacilli per milliliter. In addition, transmission of M. leprae is thought to occur through contact with contaminated soil, although soil has not been found to be a reservoir for the bacilli.  

PATHOBIOLOGY

Immunology

There is an inverse correlation between the number of lymphocytes and the number of acid-fast bacteria present in skin lesions. Tuberculoid lesions have abundant lymphocytes, well-formed granulomas, and few bacteria (hence this form of leprosy is also termed paucibacillary). In contrast, lepromatous lesions have very few lymphocytes, poorly organized or no granulomas, and large numbers of bacteria (also termed multibacillary leprosy). Between these polar extremes are intermediate forms that represent a continuum of the histopathologic and bacteriologic findings, termed borderline tuberculoid, borderline, and borderline lepromatous (Fig. 310-1). In addition to correlating with the number of bacteria in individual lesions, the polar forms of leprosy correlate with the total number of skin lesions in an individual patient: tuberculoid leprosy exhibits few (102° F) and associated with nonspecific symptoms, including malaise, myalgias, nausea, vomiting, anorexia, and diarrhea. At this stage, RMSF is not frequently diagnosed, but during the “tick season,” patients with high fever who live in or have a history of travel to an endemic location and, possibly, a history of tick bite should be considered as possibly having RMSF. The most characteristic feature is a rash. However, the classic triad of fever, headache, and rash is present in only 44% of confirmed cases. A rash is found in 14% of cases on the first day of disease and in less than 50% in the first 3 days. The rash is macular; it appears first on the ankles and wrists and then generalizes. Spots are 1 to 5 mm in diameter and can evolve from pink to purpuric. A rash can appear later or even not at all; Rocky Mountain “spotless” fever represented 34% of cases in a series from the U.S. Centers for Disease Control and Prevention (CDC). Involvement of the palms and soles theoretically differentiates the typhus diseases (in which it is absent) from spotted fever rickettsioses. Untreated patients worsen progressively. The disease is associated in various degrees with general manifestations related to vascular inflammation and increased vascular permeability and with multiple organ involvement that can lead to multiple organ dysfunction syndrome (MODS). In severe forms, patients suffer from edema, hypovolemia, hypoalbuminemia, and hypotension leading to shock. In very severe cases, necrosis and gangrene of the extremities occur. In some instances, noncardiogenic pulmonary edema develops; pulmonary involvement leading to respiratory distress can cause death. Renal failure can result either from hypovolemia and shock and be reversible or from acute tubular necrosis and require hemodialysis. The usual neurologic symptoms are confusion, lethargy, and stupor. In severe cases, delirium, coma, and seizures are observed. Cerebrospinal fluid (CSF) sampling exhibits meningitis in one third of cases; in general, a few mononuclear cells (10 to 100) are observed, along with increased protein but normal glucose levels. Heart involvement can cause arrhythmia. Liver involvement is manifested as an

CHAPTER 311  Rickettsial Infections  

ABSTRACT

Rickettsioses are emerging, mostly arthropod-borne and zoonotic infections caused by Rickettsia sp, Ehrlichia sp, or Coxiella burnetii. These infections are encountered worldwide, but their distribution mimics that of their reservoirs and/or vectors. They represent a heavy disease burden in both farm animals and humans and are among the main causes of travel-associated infections. They may be difficult to diagnose because most patients will present with mild symptoms, but some of the rickettsioses such as epidemic typhus and Rocky Mountain spotted fever are associated with significant lethality.

KEYWORDS

rickettsioses ehrlichiosis anaplasmosis Q fever zoonoses arthropods

2017.e3

2018

CHAPTER 311  Rickettsial Infections  

TABLE 311-1 GENETIC CLASSIFICATION OF RICKETTSIALES GENUS Rickettsiae

GROUP

Rickettsia

SPECIES

Typhus

R. prowazekii R. typhi R. aeschlimannii R. akari R. australis R. conorii

R. slovaca O. tsutsugamushi E. chaffeensis

1991

E. ewingii

1999

E. canis

1996

E. muris–like

2009

Candidatus Neoehrlichia mikurensis

2007

Anaplasma

A. phagocytophilum

1992

Neorickettsia

N. sennetsu

1957

Wolbachia

W. pipientis

2001

Coxiella

C. burnetii

1931

conorii israeli caspia indica

R. rickettsii R. felis R. helvetica R. heilongjiangensis R. honei R. japonica R. massiliae R. monacencis R. parkeri “R. philipii” R. raoultii R. sibirica

Orientia

Coxiellae

FIRST YEAR OF ISOLATION OR DISCOVERY 1916 1920 2001 1946 1950 1932 1974 1991 2001 1919 2001 2000 1998 1991 1992 2006 2007 2003 1966 2008 1946 1996 1997 1920

Spotted fever

Ehrlichiae

SUBSPECIES

Scrub typhus

Ehrlichia

sibirica mongolotimonae

TABLE 311-2 RICKETTSIAL DISEASES IN HUMAN BEINGS DISEASE

ORGANISM

ARTHROPOD HOST

GEOGRAPHIC AREA

RASH

ESCHAR REGIONAL HIGH FATALITY TACHE NOIRE LYMPH NODE FEVER RATE

TICK-TRANSMITTED SPOTTED FEVERS Rocky Mountain spotted fever

R. rickettsii

Mediterranean spotted fever, R. conorii Astrakhan fever, Indian tick typhus, Israeli spotted fever

Dermacentor andersoni America (North, Yes, may be Central, and South) purpuric Dermacentor variabilis Rhipicephalus sanguineus Amblyomma cajennense

Very rare

No

Yes

High

Rhipicephalus sanguineus Mediterranean, India, Yes, papular; may Yes Caspian Sea, Africa be purpuric

No

Yes

Moderate

African tick-bite fever

R. africae

Amblyomma hebraeum Amblyomma variegatum

Sub-Saharan Africa, West Indies

Yes, half of cases may be vesicular

Yes (frequently multiple)

Yes

No

Low

Queensland tick typhus

R. australis

Ixodes holocyclus

Eastern Australia

Yes, may be vesicular

Yes

?

Yes

Moderate

Siberian tick typhus

R. sibirica

Dermacentor nuttallii

Siberia, China, Mongolia

Yes

Yes

No

Yes

Low

Scalp eschar, neck lymphadenopathy after tick bite (SENLAT)

R. slovaca or R. raoultii

Dermacentor marginatus Dermacentor reticulatus

Europe, Pakistan

Very rare

Yes, may be Yes (painful) erythematous

No

Low

Lymphangitis-associated rickettsiosis (LAR)

R. sibirica Hyalomma asiaticum mongolotimonae

Mongolia, Africa, Europe

Yes

Yes

Yes

Yes

Low

Ixodes granulosus

Flinders Island, eastern Australia

Yes

Yes

Yes

Yes

Low

Ixodes ricinus

Japan, Korea (China?) Yes

Yes

No

Yes

Low

Flinders Island spotted fever R. honei Japanese spotted fever

R. japonica

2019

CHAPTER 311  Rickettsial Infections  

TABLE 311-2 RICKETTSIAL DISEASES IN HUMAN BEINGS—cont’d DISEASE

ORGANISM

ARTHROPOD HOST

GEOGRAPHIC AREA

RASH

ESCHAR REGIONAL HIGH FATALITY TACHE NOIRE LYMPH NODE FEVER RATE

Pacific coast tick fever

“R. philipii”

Dermacentor occidentalis

Northern California

Yes

Yes

No

Yes

Unknown

Unnamed

R. aeschlimannii R. helvetica R. massiliae R. monacencis R. parkeri

Hyalomma sp Ixodes ricinus Rhipicephalus sanguineus Ixodes ricinus Amblyomma maculatum

Mediterranean, Africa Europe, Asia Europe, United States Europe America

Yes No Yes Yes Yes

Yes Yes Yes Yes Yes

Yes No No No No

Yes No Yes Yes Yes

Unknown Unknown Unknown Unknown Unknown

FLEA-TRANSMITTED DISEASES Murine typhus

R. typhi

Xenopsylla cheopis Ctenocephalides felis (mosquitoes)

Worldwide

Yes

No

No

Yes

Low

Flea-borne spotted fever

R. felis

Ctenocephalides felis, Aedes mosquitoes

Worldwide

Sometimes

Sometimes

Unknown

Yes

Unknown

LOUSE-TRANSMITTED DISEASES Epidemic typhus

R. prowazekii

Pediculus humanus corporis Amblyomma ticks (?)

Worldwide

Yes

No

No

Yes

High

American sylvatic typhus

R. prowazekii

Flying squirrel ectoparasites

United States

Yes

No

No

Yes

Low

Worldwide

Yes, could lack

No

No

No

Low

Brill-Zinsser disease (relapse R. prowazekii of epidemic typhus) MITE-TRANSMITTED DISEASES Rickettsialpox

R. akari

Liponyssoides sanguineus

Worldwide

Yes, vesicular

Yes

Yes

Yes

Low

Scrub typhus

Orientia tsutsugamushi

Leptotrombidium sp (chiggers)

Central and eastern Asia, Australia

Yes

Yes

Yes

Yes

High, may relapse

3000

2500

Number of cases

2000

1500

1000

500

0 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Year FIGURE 311-1.  Number of reported cases of Rocky Mountain spotted fever in the United States, 1993 to 2010.

2020

CHAPTER 311  Rickettsial Infections  

increase in transaminases in one third of patients and jaundice in 8%. Jaundice can also reflect hemolysis. Intestinal tract involvement is manifested as abdominal pain, diarrhea, vomiting, and severe bleeding (and upper gastrointestinal hemorrhage can cause death). Ocular involvement consists of conjunctivitis and retinal abnormalities, including hemorrhages, papilledema, and arterial occlusion.4 The blood cell count shows a normal number of white blood cells but often immature myeloid cells. Thrombocytopenia is observed in 30 to 50% of cases and may be marked in severe cases. Anemia develops in 30% of patients. Coagulopathy with decreases in clotting factors (including fibrinogen) and prolonged coagulation times may contribute to bleeding. There may be hypoalbuminemia, and proteins of the acute phase response are increased (C-reactive protein, ferritin, fibrinogen). Hyponatremia and hypocalcemia may be noted and correlate with severity, as does an increase in creatinine. Increased concentrations of serum liver and muscle enzymes such as aminotransferases (aspartate [AST] and alanine [ALT] aminotransferase), lactate dehydrogenase (LDH), and creatine kinase usually reflect the severity of organ involvement, including the lung, heart, liver, and rhabdomyolysis.  

FIGURE 311-2.  Tick removal technique.

Severely ill patients should be treated in intensive care units and fluid administration carefully monitored. Mechanical ventilation is used in case of respiratory distress, hemodialysis in patients with renal insufficiency, and antiseizure drugs in patients with seizures. Anemia and coagulation abnormalities may also be corrected. For patients with gangrene of the extremities, amputation may be necessary. Glucocorticoids have not proved useful.

DIAGNOSIS

The diagnosis of RMSF should be based on clinical and epidemiologic findings and lead to early use of doxycycline. The most important clue is unexplained fever in a patient with a history of tick exposure in an endemic area. When a rash is present, RMSF should be suspected and the patient treated accordingly unless another cause is demonstrated. The differential diagnosis includes other rickettsioses (such as those caused by R. parkeri in southern states), meningococcemia, enterovirus infections, typhoid, leptospirosis, ehrlichiosis, gonococcemia, toxic shock syndrome, syphilis, rubella, measles, and Kawasaki syndrome. Drug hypersensitivity, especially after antimicrobial use for febrile illness, is sometimes confused with RMSF. The main diagnostic test relies on serology, and treatment should never be delayed to obtain diagnostic confirmation. Criteria for laboratory confirmation include a fourfold or greater change in antibody titer determined by serology (measured by immunofluorescence antibody assay [IFA], complement fixation, or latex agglutination) and direct detection of the bacterium by demonstration of specific antigens by immunodetection, genomic amplification by PCR, or culture. A biopsy specimen of a skin lesion is the best sample for this purpose. Culture of rickettsiae takes 3 to 7 days and is restricted to specialized laboratories. It is performed on cell lines such as Vero, L929, or HEL cells in biosafety level 3 conditions. Immunodetection by IFA or immunohistochemistry is sensitive and specific. It can be performed with frozen or fixed and paraffin-embedded material and allows retrospective diagnosis. PCR amplification and identification is also a sensitive and specific detection method for rickettsioses in general but cannot replace IFA for the diagnosis of RMSF. Skin biopsy and direct detection in removed ticks yield the best results because blood contains inhibitors and only few copies of rickettsial DNA. Two serum samples should be tested (early and convalescent). The early serum is usually negative because patients seroconvert between the 7th and 15th days. IFA can be used to detect either immunoglobulin G (IgG) or IgM antibodies. A cutoff value of 1 : 64 for total immunoglobulin and 1 : 32 for IgM antibodies is required for diagnosis. The latex agglutination cutoff is 1 : 64 or 1 : 128. Cross-reactive antibodies have been reported, mainly with infections caused by other rickettsioses, but also infections with Ehrlichia, Bartonella, Legionella, and Proteus spp. False-positive results, including IgM, may be observed when rheumatoid factor is present in serum and in patients with viral infection generating nonspecific B-lymphocyte proliferation (cytomegalovirus, Epstein-Barr virus). Complement fixation (which lacks sensitivity) and the Weil-Felix test (using antibodies that cross-react with Proteus strains) should not be used.

TREATMENT  The prognosis for patients with RMSF depends on the timing of antimicrobial treatment. Doxycycline saves patients with RMSF.5 The recommended dose is 100 mg two times a day, and treatment should be continued for at least 3 days after the fever resolves. Oral treatment is effective, but in patients with gastric intolerance or coma, the intravenous route is advised. Several antimicrobials are also effective in vitro against R. rickettsii, including fluoroquinolones, rifampin, and macrolide antimicrobials (azithromycin and clarithromycin but not erythromycin), but lack of clinical experience precludes their use for RMSF. β-Lactam antimicrobials, aminoglycosides, and cotrimoxazole are not effective.



PREVENTION

Prevention is based on avoidance of tick bites by use of repellents, protective garments, or both. To discourage tick attachment, repellents containing permethrin can be sprayed on boots and clothing and will last for several days. Repellents containing DEET (N,N-diethyl-m-toluamide) can be applied to the skin but will last only a few hours before reapplication is necessary. It is also useful to check for ticks after exposure. Careful examination of the scalp, groin, and axillae is recommended. The tick can be removed by forceps, and the skin should be disinfected (Fig. 311-2).  

PROGNOSIS

The evolution of RMSF depends strongly on the timing of diagnosis and antimicrobial treatment. The current fatality rate is 2.4% on the basis of a 4-year national survey in the United States (27 deaths were attributable to RMSF during this period). This rate is currently declining, but this may result from reporting of confounding rickettsial diseases. No significant difference in outcome has been observed between blacks and whites, but the case-fatality rate was highest in people older than 70 years (9%). Patients with glucose6-phosphate dehydrogenase (G6PD) deficiency are more susceptible to severe infection. Chloramphenicol has been associated with a poorer outcome than treatment with doxycycline. Recovery from RMSF is usually complete, but neurologic sequelae can remain, and amputation of extremities may be necessary after gangrene.  

OTHER TICK-BORNE RICKETTSIOSES EPIDEMIOLOGY  

Like other tick-transmitted diseases, rickettsioses have a limited geographic distribution that is determined mainly by the tick vector ecology (Fig. 311-3). R. parkeri has recently been identified in the United States and South America. R. conorii is found in Europe around the Mediterranean and Caspian seas (caspia subspecies); Rickettsia slovaca, Rickettsia raoultii, and possibly Rickettsia helvetica in western and central Europe; and Rickettsia sibirica mongolotimonae in France and Greece. Elsewhere, a number of specific agents of rickettsial disease have been identified (see Table 311-2).  

CLINICAL MANIFESTATIONS

R. conorii comprises different but closely related subspecies. Many names are given to the infection caused by R. conorii: Mediterranean spotted fever (MSF), boutonneuse fever, Marseilles fever, Kenya tick typhus (caused by the subspecies R. conorii conorii), Astrakhan fever (caused by R. conorii caspia), Israeli spotted fever (caused by R. conorii israeli), and Indian tick typhus (caused by R. conorii indica). R. conorii is closely related to R. rickettsii, with which it shares many common antigens that generate cross-reactive antibodies. MSF resembles RMSF but has some distinguishing features. The spontaneous evolution is milder, but a fatality rate of 1.5 to 2.5% in hospitalized patients is still observed. A malignant form of the disease that includes purpuric rash, shock, and MODS has been described in alcoholic, diabetic, human immunodeficiency virus (HIV)-infected, and old or debilitated patients. The typical clinical manifestation is that of a patient with fever, a rash, and a tache noire

2021

CHAPTER 311  Rickettsial Infections  

R. conorii caspia

R. massiliae

R. slovaca R. parkerii

“R. philipii” R. massiliae

R. helvetica

R. sibirica sibirica

R. sibirica mongolitimonae

R. heilongjiangensis R. conorii indica

R. sibirica mongolitimonae R. raoultii R. conorii conorii

R. japonica R. conorii israelensis

R. rickettsii

R. conorii conorii R. australis R. aeschlimannii R. africae

R. honei

FIGURE 311-3.  Geographic distribution of tick-borne rickettsioses.

(i.e., a black eschar at the site of the tick bite). A tache noire is found in 50 to 80% of cases. Multiple eschars are rare because the dog tick vector, R. sanguineus, seldom bites humans. The rash is frequently clearly papular, which led to one of the names of the disease, boutonneuse fever. Israeli tick-bite fever and Astrakhan fever appear to be milder than typical MSF, and a tache noire is usually lacking. R. africae, which causes African tick-bite fever, may be responsible for most of the rickettsioses worldwide. It is extremely common in travelers visiting safari parks in southern Africa.6 It is transmitted by African ticks, Amblyomma hebraeum and Amblyomma africanum. These ticks are often infected; as many as 60% can harbor R. africae. They usually feed on ungulates but attack human beings in groups and cause a high prevalence of infection in rural Africa (60% of tested patients exhibit antibodies) and in travelers. The tick attacks typically generate clusters of cases in safari tourists. The disease differs from MSF in that it is much milder, fever is frequently absent, a rash is observed in only half of patients, and the rash may be vesicular (which has never been reported in confirmed MSF). Moreover, several taches noires are frequently observed. They are prevalently found on the lower limbs and often associated with draining lymphadenopathy in the groin. Japanese spotted fever (caused by Rickettsia japonica) and Siberian tick typhus (caused by R. sibirica) resemble MSF. Infections caused by R. sibirica mongolitimonae resemble MSF but in some cases exhibit distinguishing clinical features, including a tache noire, groin lymphadenopathy, and lymphangitis joining these two lesions. The disease has recently been named lymphangitisassociated rickettsiosis. Rickettsia australis (Queensland tick typhus) and Rickettsia honei (Flinders Island spotted fever) cause diseases resembling MSF, but their rash can be vesicular. R. slovaca and R. raoultii cause a disease common in Europe named scalp eschar and neck lymphadenopathy transmitted by ticks (SENLAT, reported in the Czech Republic, France, Hungary, Germany, Italy, Lithuania, Romania, Spain). Its tick vectors, Dermacentor marginatus and Dermacentor reticulatus, preferentially bite in cold months and bite the scalp because they prefer hairy prey. In contrast to other tick-borne rickettsioses, the disease is more prevalent in children and women. It is rarely exanthematic; the typical clinical picture consists of an erythematous skin lesion at the site of the tick bite on the scalp that ranges from 2 to 8 cm in diameter and a draining neck lymphadenopathy (which may be painful). Rarely, patients may exhibit fever and a rash. Deep postinfectious asthenia and residual alopecia at the site of the tick bite can be observed. The occurrence of this rickettsiosis without rash may stimulate research on other new rickettsial diseases with only localized manifestations.

Other Rickettsia species have rarely been identified as causing SENLAT, including Rickettsia massiliae, Rickettsia sibirica mongolitimonae, and Candidatus Rickettsia rioja.  

DIAGNOSIS

The diagnosis of other tick-borne rickettsioses is similar to that of RMSF, mainly by serology (IFA; see earlier). An exception is R. slovaca infection, in which the serologic response is weak, possibly because of its lack of general infection; in this case, PCR of a skin eschar sample by a swab or a lymph node aspirate provides the highest diagnostic yield. In R. africae infection, the serologic response occurs later than in RMSF and MSF, and late serum samples (i.e., 4 weeks after onset of symptoms) are therefore recommended.

TREATMENT  Doxycycline (100 mg twice daily for adults or 4.4 mg/kg body weight per day in two divided doses for children under 45.4 kg [100 lb]) is the drug of choice for treatment. A single day of therapy usually suffices, but in adults with more severe disease, it should be administered until the patient is afebrile for 24 hours. In pregnant women, josamycin, a macrolide antimicrobial, has proved efficient at a dose of 3 g daily for 7 days for MSF; quinolones and newer macrolide antimicrobials give results comparable to those of doxycycline but require longer treatment courses.



Flea-Transmitted Diseases

Fleas (Chapter 104) can harbor two rickettsial species: Rickettsia typhi, the agent of murine typhus; and Rickettsia felis, the agent of flea-borne spotted fever. Both rickettsiae can be transmitted transovarially in the flea. Vectors are Xenopsylla cheopis and Pulex irritans but also Ctenocephalides felis, a cat flea. Rats, cats, opossums, and dogs can propagate infected fleas. These reservoirs and vectors are distributed worldwide, and thus these diseases have a global distribution. Fleas can be infected by both species at the same time.  

MURINE TYPHUS DEFINITION  

Fleas are usually infected by R. typhi when feeding on apparently healthy rats that have blood-borne infection. Humans and other mammals are infected through autoinoculation by scratching a fleabite that is contaminated with

2022

CHAPTER 311  Rickettsial Infections  

feces from an infected flea. Murine typhus, because of its cycle, is more prevalent in hot and humid areas, when rats proliferate.  

EPIDEMIOLOGY

In the United States, 50 to 100 cases are reported yearly, mainly in southern California and southern Texas. In California, a transmission cycle involving opossums and cat fleas has been demonstrated. Murine typhus is extremely common in Southeast Asia and North Africa and is a common cause of fever in travelers to those areas.  

CLINICAL MANIFESTATIONS

On the basis of studies of infected volunteers, the incubation period is generally 8 to 16 days. The disease begins with abrupt fever, nausea, vomiting, myalgias, arthralgias, and headache. A rash is observed in 40 to 50% of patients about 6 days after the onset. It is detected even less frequently in patients with dark skin. The rash begins as pink maculae that can evolve to be maculopapular. It is often discrete, starting in the axilla; it generalizes to the trunk but does not usually involve the face, palms, and soles. In severe cases, it can become purpuric. The most frequently involved organ is the lung. One third of patients have a cough, and in one fourth, a nonspecific interstitial pneumonia develops that is sometimes associated with a pleural effusion. In severe forms, respiratory failure occurs. In patients with severe disease, neurologic symptoms range from confusion and stupor to coma and seizures. Cerebral hemorrhages may occur. Digestive involvement can be manifested as vomiting, abdominal pain, jaundice, and, in severe cases, hematemesis. The white blood cell count shows leukopenia and then leukocytosis. Thrombocytopenia can be noted as well as anemia, specifically when hemolysis is observed (frequently in patients with G6PD deficiency). A moderate increase in serum liver enzymes is common. In patients with severe disease, hyponatremia and hypoalbuminemia are observed.  

DIAGNOSIS

The diagnosis of murine typhus is based mainly on serology (IFA), with titers similar to those of RMSF. On serologic evaluation, R. typhi cross-reacts with R. prowazekii; it can be differentiated either by comparing titers (two dilutions or more if IgG and IgM titers are discriminative) or by cross-adsorption. In this technique, the serum is absorbed with either antigen and then retested, and the causative agent is that removing antibodies to both bacteria. Skin biopsies and blood samples for culture and PCR may be valuable.

TREATMENT  Treatment is the same as that for RMSF.



PROGNOSIS

The prognosis is usually favorable, but 10% of patients require intensive care and 1% die. Older patients and those with G6PD deficiency (Chapter 152) or chronic debilitating conditions are at higher risk.  

FLEA-BORNE SPOTTED FEVER CAUSED BY RICKETTSIA FELIS

R. felis is mainly transmitted transovarially. Its genome comprises one or two plasmids, one being apparently conjugative. This is a new, incompletely defined disease. The bacterium is found in fleas in the Americas, Asia, Europe, Africa, and New Zealand. Isolated cases have been reported from Texas, Mexico, Brazil, France, and Germany. Reported cases all exhibited fever, a rash in six of seven cases, and inoculation eschar in some cases. The diagnosis can be based on serologic testing using specific R. felis antigen or PCR of blood or skin biopsy samples. Treatment has not been established, but the bacterium is highly susceptible to doxycycline and resistant to erythromycin. R. felis has been found at very high prevalence in the blood of febrile sub-Saharan Africans and is suspected to be transmitted by mosquitoes, as suggested by an animal model.  



Louse and Mite Infections

EPIDEMIC LOUSE-BORNE TYPHUS EPIDEMIOLOGY  

The human body louse (Chapter 104) lives in clothes and multiplies rapidly when cold weather and lack of hygiene allow it to. The body louse transmits three bacterial diseases: trench fever (caused by Bartonella quintana) (Chapter 299), relapsing fever (caused by Borrelia recurrentis) (Chapter 306), and

exanthematic typhus (caused by R. prowazekii). The name typhus is derived from the Greek tuphos, which describes the neurologic condition associated with this disease and with typhoid. The body louse is prevalent during war, in poor countries, and in the homeless population of rich countries, including the United States and Europe. A 100,000-person outbreak of typhus was reported during the civil war in Burundi in 1997, and cases were reported in Russia, Peru, the United States, Algeria, and France in the 1990s. Lousetransmitted diseases killed more people than weapons did during central and eastern European wars in the 19th and 20th centuries.7 The epidemiology of R. prowazekii is mainly related to humans as reservoirs and lice as vectors. In the United States, the eastern flying squirrel (Glaucomys volans volans) is also a reservoir, and its fleas, lice, and mites can be infected. R. prowazekii has also been found in Amblyomma ticks, but their role in the epidemiology of the bacterium is not known. The louse is infected when feeding on blood, which it does five times a day. R. prowazekii multiplies in the gut of the louse and is released in feces. After a few days, it destroys the intestinal epithelium, causing bright red blood to spread from the gut (typhus was also named the red louse disease). The patient is usually contaminated by infected feces (in which R. prowazekii survives for weeks), through aerosols, or by skin autoinoculation after scratching. Patients who recover from typhus may harbor the bacterium in a dormant form and suffer relapses under stressful conditions years later; this relapsing form is called Brill-Zinsser disease. During the relapse, a bacteremia occurs that may allow the start of a new outbreak if lice bite the patient.  

CLINICAL MANIFESTATIONS

Typhus begins abruptly with fever, headaches, and myalgias, which may lead to the crouched posture termed sutama, as observed in the largest recent outbreak in Burundi. Cough and neurologic signs (stupor, confusion, or coma) are common.8 A rash is observed in 20 to 80% of patients, depending on the population studied; it is probably commonly underobserved on dark skin. It generally starts in the axilla and then spreads. The rash is usually macular but can be papular or purpuric in severe cases. In some cases, diarrhea and jaundice are reported. Splenomegaly is infrequently found. In severe cases, shock occurs, and the fatality rate of epidemic typhus is 20 to 30% when untreated. Leukopenia, thrombocytopenia, and anemia, as well as an increase in serum hepatic enzymes, may be noted. Sylvatic typhus in the United States is caused by an R. prowazekii variant and is a milder disease. The most prominent clinical features are neurologic. Few cases have been described, and nearly all occurred in areas where the eastern flying squirrel is found, east of the Mississippi. Brill-Zinsser disease is difficult to diagnose because rash is rare and recent exposure to lice can be lacking. Interviewing the patient may reveal prior exposure to lice, associated or not with a diagnosis of typhus in previous years. The disease is mild, and the prognosis is good.  

DIAGNOSIS

The diagnosis of typhus should be considered when grouped cases of high fever with confusion are observed in patients exposed to lice. The most common diagnostic error is to attribute the findings to typhoid (Chapter 292), which can have fatal consequences because the antimicrobials typically prescribed for that condition (β-lactams, cotrimoxazole, and quinolones) are ineffective treatment of typhus. In tropical countries, typhus is frequently confused with malaria, hemorrhagic fever, and dengue. In people with lice, it can be confused with trench fever and relapsing fever, but treatment for both can be prescribed. The diagnosis of typhus should be clinical because the fatality rate is high and the treatment safe and efficient. Any outbreak of unexplained fever in unhygienic environments may suggest typhus, including outbreaks during civil wars, during social collapses, in jails, and in chronically poor and cold countries. The diagnosis is mainly based on serology, in which there is crossreaction with R. typhi (see earlier). When the investigation is performed under difficult field conditions, a drop of blood applied on filter paper and sent to a reference laboratory is valuable for serologic testing. Culture and PCR are helpful and can be performed with a skin biopsy sample or blood. Lice are good diagnostic tools because they can be tested even when dry and can be sent in closed containers without specific temperature conditions.

TREATMENT  Treatment of typhus is extremely simple, cheap, and effective; 200 mg of doxycycline orally in two divided doses is life-saving. Comatose patients should be treated with doxycycline parenterally. In allergic patients, chloramphenicol

2023

CHAPTER 311  Rickettsial Infections  

is the only known alternative, prescribed at a dose of 2 g/day for 10 days. There is no current vaccination, and the fight against lice is the major prevention strategy. Because lice are fragile, changing and boiling clothes are efficient. When this is not possible, insecticides (primarily permethrin) or ivermectin orally should be used.

cases of black skin eschars were investigated for possible anthrax in New York but were in fact rickettsialpox. High seroprevalence was reported among intravenous drug users in Baltimore. Cases have also been reported from Russia, Ukraine, Slovenia, and Korea.  



SCRUB TYPHUS (ORIENTIA TSUTSUGAMUSHI) EPIDEMIOLOGY  

Scrub typhus is a neglected tropical disease transmitted by the bite of trombiculid mite (Chapter 104) larvae infected by O. tsutsugamushi. These mites, also named chiggers, are vertically infected through their mother. Although rare cases have been reported in Chile, sub-Saharan Africa, and United Arab Emirates, scrub typhus distribution is mostly limited to a triangle extending between northern Japan, eastern Australia, and eastern Russia and includes the Far East, China, and the Indian subcontinent.9 Altogether, 1 billion people may be exposed. Seasonality is determined by the emergence of larvae. It is one of the three most common causes of prolonged fever in rural Asia; in temperate zones, it occurs mainly in autumn and to a lesser extent in spring. O. tsutsugamushi species have a wide heterogenicity that may allow the definition of several species, but a single species is currently recognized with many serotypes. The more frequent are Kato, Karp, Gilliam, and Kawasaki.  

DIAGNOSIS

Diagnosis may be difficult. Because the clinical features are frequently not specific, epidemiologic factors are critical. A diagnosis of infectious mononucleosis has erroneously been made in patients with scrub typhus. The bacterium can be detected by culture (in cells or mice) or by PCR in blood and biopsy specimens.13 The serologic technique first used was agglutination of Proteus mirabilis serotype OXK in the Weil-Felix reaction. This test lacks sensitivity and specificity and should be replaced by IFA or enzyme-linked immunosorbent assay tests using the three or four major serotypes. Point-of-care testing for scrub typhus still lacks standardized methodology and reporting of diagnostic accuracy.14

TREATMENT  Chloramphenicol was the mainstay of treatment for many years, but now doxycycline is recommended. Single-day treatment with doxycycline is followed by relapses, and even repeated treatment for 2 days at a 7-day interval does not prevent all relapses. Hence, the currently recommended regimen is doxycycline, 100 mg orally twice a day, for 7 days. Cases resistant to doxycycline have been reported, and rifampin (600 mg orally daily) is a reasonable alternative. A1  Quinolones should be avoided. Prophylaxis is based on the use of repellents.





DIAGNOSIS

The diagnosis can be made by serologic testing with IFA. Specific antigens react with high titer, but antibodies to other Rickettsia may be detected. The diagnosis may also be made on skin specimens by culture, immunodetection, or PCR.

TREATMENT  Doxycycline is highly effective in these patients. Prevention is based on the control of mice.

CLINICAL MANIFESTATIONS

The disease occurs in patients exposed to rural or urban foci of scrub typhus after a delay of 10 or more days. The onset is usually sudden and includes fever, headache, and myalgias. Attentive examination may reveal an inoculation eschar at the site of the mite bite and tender draining lymph nodes. Most patients with scrub typhus present with relative bradycardia.10 Generalized lymphadenopathy and rash may be observed. The symptoms vary according to organ involvement. Neuromeningeal symptoms are relatively common.11 Severe forms can be manifested as septic shock. The lethality rate of untreated scrub typhus is 6%. Abortion commonly occurs in pregnant women. Leukopenia, thrombocytopenia, and increased levels of hepatic enzymes can occur. Evolution depends on the hosts and strains, and the fatality rate ranges from 0 to 30%.12 Scrub typhus is not more severe in HIV-infected patients, and surprisingly, HIV suppressive factors appear to be produced during infection. Relapses may occur in this disease.  

CLINICAL MANIFESTATIONS

Ten days after the mite bite, the beginning of the illness is marked by fever, headache, and myalgia. Careful examination reveals an inoculation eschar and draining lymphadenopathy that could be mistaken for cutaneous anthrax. Two to 6 days later, a rash appears and comprises 5 to 40 macular, then papular and vesicular, spots. This aspect led to the name of the disease. It is frequently mistaken for chickenpox. The disease is usually mild.

RICKETTSIALPOX (RICKETTSIA AKARI) EPIDEMIOLOGY  

Rickettsialpox was first described in New York City, where it is still prevalent. Rickettsia akari, the causal agent, is transmitted by the bite of the mouse mite (Liponyssoides sanguineus). Its prevalence is probably underestimated; an active search revealed 13 cases in a New York hospital in the 1980s. Cases have been reported in Arizona, Utah, and Ohio. After the terrorist attacks of 9/11/01,

  EHRLICHIOSES AND ANAPLASMOSES  

DEFINITION

Ehrlichiae multiply exclusively in vacuoles of their eukaryotic cell host, where they form clusters known as morulae. The vacuoles are derived from phagosomes and help the organism escape bactericidal lysosomal fusion. In humans, ehrlichiae are associated with monocytes (E. chaffeensis, E. canis, N. sennetsu) or polymorphonuclear neutrophils (PMN) (A. phagocytophilum, E. ewingii). Ehrlichioses can be acquired through tick bites, by ingestion of nematodes through contaminated water or animals (fish, snails), or as a consequence of filariasis.15

American Human Monocytic Ehrlichiosis (Ehrlichia chaffeensis)  



EPIDEMIOLOGY

Human monocytic ehrlichiosis (HME) is caused by E. chaffeensis. This organism has been isolated or identified by PCR mainly in the United States, in the southeastern, south central, and mid-Atlantic states and California (Table 311-3). In the United States, A. americanum (Lone Star tick) is the vector

TABLE 311-3 EHRLICHIOSES AND ANAPLASMOSES DISEASE American monocytic ehrlichiosis

AGENT Ehrlichia chaffeensis

Human Anaplasma granulocytic phagocytophilum ehrlichiosis Japanese monocytic ehrlichiosis

Neorickettsia sennetsu

Unnamed

Ehrlichia canis

GEOGRAPHIC REPARTITION

VECTOR

Amblyomma South central, southeastern, americanum mid-Atlantic coastal states Ixodes ricinus

Europe, China

Ixodes scapularis

Northeast, upper Midwest, northern California

Helminth of the gray mullet?

Japan

Rhipicephalus Venezuela sanguineus Ehrlichia ewingii Amblyomma South central, southeastern, americanum mid-Atlantic coastal states Ehrlichia muris–like Ixodes Indiana, Michigan, scapularis Minnesota, North Dakota, Wisconsin Candidatus Ixodes ricinus Central and eastern Europe, Neoehrlichia northeastern China mikurensis

CHAPTER 311  Rickettsial Infections  

The index case of modern ehrlichiosis was reported in the United States in 1987. The patient died of fever, presumably acquired after a tick bite in Arkansas, despite receiving chloramphenicol. The patient had several initially confusing diagnostic features; on blood smears, morulae in polymorphonuclear (PMN) cells were seen, and antibodies to Ehrlichia canis, a pathogen of dogs but not humans, were detected. He was then thought to have Ehrlichia chaffeensis, but this bacterium infects monocytes, not PMN cells. A diagnosis of Anaplasma phagocytophilum infection (or human granulocytic ehrlichiosis [HGE]) was considered, but the tick vector of this disease is absent in Arkansas. The most likely diagnosis is currently believed to be infection with Ehrlichia ewingii, an agent transmitted by Amblyomma americanum that is prevalent in Arkansas, infects PMN cells, and cross-reacts with E. canis, although it typically affects

2023.e1

immunocompromised hosts. This case illustrates the progress in knowledge on ehrlichioses and how difficult it is to conclude the etiology of an atypical infection definitively on the basis of serology alone. All Ehrlichia spp pathogenic for humans except E. ewingii can be cultured. The ehrlichiae have been reclassified into four genera, mainly on the basis of 16S ribosomal RNA–derived phylogenetic analysis. Two are the tick-associated genera Ehrlichia and Anaplasma (A. phagocytophilum, or the HGE agent that was formerly named Ehrlichia phagocytophila). One is a helminth-associated genus, Neorickettsia, including Neorickettsia sennetsu (formerly Rickettsia sennetsu, then Ehrlichia sennetsu). The fourth is Wolbachia pipientis, a bacterium associated with arthropods (insects, crustaceans, and acarids) and helminth worms (mainly filaria). These organisms elicit cross-reactive antibodies.

2024

CHAPTER 311  Rickettsial Infections  

(Chapter 104), and the white-tailed deer is the main mammalian reservoir. Immature ticks are infected by blood while feeding on persistently bacteremic reservoirs. E. chaffeensis is transmitted transstadially in the tick and infects its next host (deer or human) during its next blood meal. The disease epidemiology reflects the tick habitat and activity, with most cases being contracted in the southern United States, in rural areas, and from April to September. In highly endemic areas, the incidence can reach 100 cases per 100,000 inhabitants. The severity is age dependent, which may explain the lower incidence reported in children. Males are more often affected than females, with a sex ratio of 4 : 1.  

TREATMENT  Doxycycline (100 mg twice daily for adults) is the drug of choice for patients with ehrlichiosis. The optimal duration of therapy has not been established, but current regimens recommend continuation of treatment for at least 3 days after the fever subsides and until evidence of clinical improvement, for a minimum total course of 5 to 7 days. Severe or complicated disease may require longer treatment courses. Because tetracyclines are contraindicated in pregnancy, rifampin has been used successfully in a limited number of pregnant women with documented HME.

CLINICAL MANIFESTATIONS

The incubation lasts for 7 to 10 days after an identified tick exposure in 80% of cases. Patients have fever, headache, malaise, nausea, and anorexia. Untreated patients worsen and may require intensive care. Digestive tract involvement consisting of nausea, vomiting, diarrhea, and abdominal pain is common. Central nervous system infection is manifested in many forms, ranging from confusion to coma. A rash is observed in one third of cases and lymphadenopathy in one fourth. In severe forms, sepsis syndrome and MODS may occur. The white blood cell count typically shows leukopenia, caused by both lymphopenia and neutropenia. Thrombocytopenia is also frequently found; anemia may appear later. Coagulopathy may be observed in severe forms. Increases in serum enzymes, including AST, ALT, and LDH, may reflect organ involvement, as does creatininemia. CSF examination in patients with neurologic symptoms reveals pleocytosis and increased protein levels. Cells may be monocytic or PMN. The prognosis depends on early antimicrobial treatment, but the fatality rate is still high at 2.5%. In persons coinfected with HIV, it may be most severe; in one series, 6 of 13 patients died.  

diagnosed serologically by a four-fold or greater increase in antibody titer or by seroconversion. The reference technique is IFA. A single titer of 25 is indicative of the diagnosis. There are cross-reactive antibodies among Ehrlichia species and with A. phagocytophilum. Western blotting may be valuable to distinguish among these bacteria.

DIAGNOSIS

The diagnosis of HME should be considered in patients with a history of tick exposure and unexplained fever. HME resembles RMSF, but rash is less frequent. Later in the disease, it can be misdiagnosed as anything that causes severe sepsis. Leukopenia associated with thrombocytopenia and an increase in liver enzyme levels may establish the etiology. Careful examination of blood and CSF smears may help identify typical morulae. Treatment should be started in any suspected case. The diagnosis can be confirmed by culture in specialized laboratories using a canine cell line, DH82. However, PCR is more practical; confirmatory PCR using a second target gene is useful. Most cases are currently Anaplasma phagocytophilum

Human Granulocytic Ehrlichiosis (Anaplasma phagocytophilum)  



EPIDEMIOLOGY

The first human case of A. phagocytophilum infection was recognized in 1990. The disease is found in America, Asia, and Europe (Fig. 311-4). It is transmitted by Ixodes scapularis (eastern North America), Ixodes pacificus (western North America), Ixodes ricinus (Europe), and Ixodes persulcatus (Asia), the vectors of Lyme disease (Chapter 305), and its epidemiology is similar.16 Coinfection with the two diseases may occur. The temporal distribution of the disease parallels that of nymph tick activity, with two peaks in spring and autumn. Ticks are born free of Ehrlichia and are infected while feeding on bacteremic small mammals. Deer play a major role as hosts of adult ticks and reservoirs. In highly endemic areas, the incidence can reach 50 per 100,000 inhabitants per year. The mean age of diagnosed patients is high, and males are more frequently infected than females, with a sex ratio of 3 : 1.  

CLINICAL MANIFESTATIONS

The incubation time is usually between 7 and 10 days, and 80% of patients report a history of tick exposure. Many infections may be asymptomatic or too mild to require a diagnostic procedure. The disease frequently begins abruptly, with fever, headache, malaise, and myalgias that may be particularly Candidatus Neoehrlichia mikurensis

Ehlichia chaffeensis

Ehrlichia muris-like

Neorickettsia sennetsu

Ehrlichia ewingii

Ehlichia canis

FIGURE 311-4.  Geographic distribution of ehrlichioses.

CHAPTER 311  Rickettsial Infections  

severe. Rash is found in less than 10% of cases. Visceral involvement may be observed and includes digestive symptoms such as nausea, vomiting, and diarrhea. Neurologic symptoms may include confusion, meningitis, and meningoencephalitis. The evolution of the disease is favorable in most cases, even without specific therapy, but the disease may progress to septic shock in some patients. Patients with underlying diseases are more at risk for dying. Most deaths are the consequence of Anaplasma-induced immunosuppression, and patients may experience invasive aspergillosis, candidiasis, cryptococcosis, and herpes esophagitis.  

DIAGNOSIS

Laboratory findings consist of the association of thrombocytopenia and leukopenia (lymphopenia or neutropenia). An increase in serum transaminases is also frequent. The diagnosis can be made by careful examination of blood smears for morulae within PMN cells. Culture from blood is possible in appropriate cells (HL-60), and PCR is useful as for HME. Most cases are diagnosed by serologic testing with IFA, which is comparable to that in HME (see earlier).

TREATMENT  Treatment is similar to that of HME except that A. phagocytophilum is susceptible to fluoroquinolones in vitro, but these drugs have not been tested in patients.



Ehrlichia ewingii

Canine granulocytic ehrlichiosis, reported in the United States in 1972, is caused by E. ewingii. This bacterium was characterized by amplification and sequencing of the 16S ribosomal RNA gene. The vector of E. ewingii is A. americanum, which also transmits E. chaffeensis. Among 60 cases of ehrlichiosis in Missouri in 1999, 4 were caused by E. ewingii; 4 other cases have been reported since by the CDC. The disease has been prevalent in immunocompromised hosts coinfected with HIV or receiving immunosuppressive drugs. Patients who report tick exposure are noted to have fever, thrombocytopenia, leukopenia, and various symptoms, including meningitis. Morulae may be seen on blood smears in PMN cells. The evolution in reported cases was good; patients responded dramatically to doxycycline. Patients have antibodies to E. chaffeensis, and PCR has been shown to be useful when it is applied to blood samples. This diagnosis should be considered when ehrlichiosis is suspected in immunocompromised patients exposed to A. americanum ticks.  

Ehrlichia canis

Canine monocytic ehrlichiosis was reported first in Algeria in the 1930s. It is caused by E. canis and transmitted by the dog tick R. sanguineus. This tick is found worldwide and is prevalent in temperate and hot areas. In 1996, a single case of infection was reported in an asymptomatic man from Venezuela who owned an infected dog. Recently, cases have been reported in patients in South America.  

Ehrlichia muris–like Agent

Since 2009, more than 50 cases of an ehrlichiosis have been diagnosed in Wisconsin and Minnesota, caused by an Ehrlichia that is closely related to E. muris. The bacterium has also been cultured from Ixodes scapularis ticks in this area. Patients have fever, headache, myalgias, lymphopenia, thrombocytopenia, and transaminitis, and they recover with doxycycline treatment.  

Candidatus Neoehrlichia mikurensis

Candidatus Neoehrlichia mikurensis, transmitted by Ixodes ticks, has been reported based on PCR testing of the blood of febrile immunocompromised patients in Europe and China.  

Wolbachia Species

Wolbachia bacteria are endosymbionts of arthropods and nematodes. They were known to be present in filarial worms, but it was later shown that they may play a role in human disease. These bacteria manipulate the fertility of their host. Eradication of Wolbachia in filariae may lead to infertility and stop

2025

the microfilariae from spreading. This effect was demonstrated by field treatment with doxycycline in patients with onchocerciasis. The patients improved when treated with this drug, which is effective on Wolbachia and subsequently on the worm’s fertility but not on the worm itself. In 2001, it was shown that the adverse reactions observed after treatment of lymphatic filariasis may be caused by the release of Wolbachia from destroyed worms. Some authors suggested that eradicating Wolbachia before the anthelmintic prescription would avoid these reactions. For some reason, Loa loa (Chapter 335) do not harbor Wolbachia, and the genome of Wolbachia integrated in the Brugia malayi genome makes it inaccessible to therapy.

  Q FEVER  

DEFINITION

Q fever is a worldwide zoonosis caused by Coxiella burnetii. The name Q fever is derived from “query” to emphasize the surprising aspect of the disease first described in Queensland, Australia, in 1935 by Derrick. The infection in humans is variable in its severity, clinical expression, and natural course (i.e., acute or chronic). It is considered by the CDC to be a potential agent of bioterrorism. Ungulates (hoofed mammals) and pets are the major sources of human infection.

The Pathogen

C. burnetii is a gram-negative bacterium that naturally infects its host’s monocytes. It multiplies in an acidic vacuole. Strains are heterogeneous genetically and antigenically and are associated with acute infections of variable severity. C. burnetii in vitro generates a deleted, avirulent mutant also named phase II. This mutant exhibits diagnostic antigens that are useful because they are more reactive during acute infection. C. burnetii is incompletely eliminated after acute infection. In immunocompromised hosts and patients with cardiac valve lesions, C. burnetii continues to multiply despite high levels of antibodies and causes chronic infection. Control of the disease in acute Q fever is associated with the formation of a granuloma.  

EPIDEMIOLOGY

C. burnetii infects a wide range of animals, including mammals, birds, and ticks. Ungulates and pets (cats and dogs) are the most common source of the disease. Mammals are infected through aerosols and may shed Coxiella in feces, urine, milk, and birth products. Humans are usually infected by aerosols or less frequently by exposure to milk products. Interhuman infections through sexual intercourse, during delivery, or by blood transfusion have been reported. Coxiella survives in the environment and can be spread far by the wind. In the past few years, major outbreaks were related to sheep and goats. The disease is partly seasonal and related to lambing time. Males have more severe disease but are not more often exposed to Q fever, and middle-aged people are more frequently affected and hospitalized. The number of reported cases recently increased dramatically in Europe and Asia. Many American soldiers were infected in Iraq. A giant outbreak was observed in the Netherlands from 2007 to 2010.17  

CLINICAL MANIFESTATIONS

After contamination by C. burnetii, 60% of patients seroconvert without apparent disease, 38% experience a self-limited disease, and only 2% require diagnostic evaluation. Months to years after the primary infection, a chronic infection develops in 0.2 to 0.5% of patients, associated with an immunocompromised state, a cardiac valve lesion, or a vascular prosthesis or aneurysm. Patients with diagnosed acute infection may have a variety of symptoms (Table 311-4). Isolated prolonged fever was observed in 14% of a series of more than 1000 patients. Pneumonia was found in 37% and was the only symptom in 17%. This percentage may vary according to the place of study and reach 90% of diagnosed cases. Some cases may be associated with respiratory distress.18 Hepatitis is found in 60% of patients and is the sole manifestation in 40%. The association of fever and a moderate increase in transaminases is an important clue. Some hepatitides, specifically in middle-aged men, are associated with an inflammatory syndrome and autoantibodies and may be resistant to antimicrobial treatment. Liver biopsy, when it is performed, exhibits granulomas that may be typified by a lipid vacuole and surrounded by a fibrinoid ring in the form of a doughnut. Less frequently, in 1.5% of cases, patients exhibit a rash. Patients can have specific neurologic manifestations such as meningitis, encephalitis, meningoencephalitis, or peripheral neuropathy. In 1 to 2% of cases, patients have cardiovascular manifestations, such as

TABLE 311-4 SITUATIONS THAT SHOULD PROMPT SEROLOGIC TESTING FOR Q FEVER ACUTE Q FEVER (PHASE II ANTIGEN AND IgG ≥200 AND IgM ≥50) Fever in a patient in contact with ungulates (hoofed mammals) Unexplained prolonged fever (>7 days) Granulomatous hepatitis Fever and thrombocytopenia Meningoencephalitis Myocarditis Erythema nodosum Fever during pregnancy Fever in a patient in contact with a parturient pet Unexplained atypical pneumonia Fever and an increase in transaminases (2-5 times the normal level) Aseptic meningitis Guillain-Barré syndrome Pericarditis Spontaneous abortion CHRONIC Q FEVER (PHASE I ANTIGEN AND IgG ≥800 AND IgA ≥100) Blood culture–negative endocarditis Patient with a valvulopathy and unexplained Fever Weight loss Fatigue Increased erythrocyte sedimentation rate Increased transaminases Thrombocytopenia Patient with unusually rapid degradation of a prosthetic valve Fever in a patient with a vascular aneurysm or prosthesis Aseptic osteomyelitis Chronic pericarditis Multiple spontaneous abortions

pericarditis or more rarely myocarditis or acute endocarditis.19 Rare cases of prosthetic joint infection have also been reported. Evolution is usually favorable even without treatment, except in certain hosts. In pregnant women, symptomatic or not, Q fever compromises the pregnancy. When infected during the first trimester, the patient usually aborts spontaneously. When the patient is infected later, the disease can result in fetal death or prematurity, although in many the outcome may be normal. Chronic uterine infection may develop in half the patients infected during pregnancy, and these women may later experience multiple spontaneous abortions. Thirty percent to 50% of patients with heart valve or vascular lesions may experience chronic endocarditis within 2 years. This development is not prevented by regular treatment. Patients with chronic Q fever endocarditis have a chronic infection with low-grade fever, progressive degradation of valve function, and progressive heart failure. Fever is intermittent, and vegetations are frequently absent on cardiac echocardiography. Endocarditis is therefore not frequently considered in the initial differential diagnosis. If it is not diagnosed, the disease progressively worsens, and emboli (mainly cerebral), as well as renal insufficiency, splenomegaly, and hepatomegaly, may be observed. Digital clubbing may also be seen. The main clue to the diagnosis in a patient with a valvulopathy is unexplained sickness (unexplained fatigue, weight loss, fever), a biologic abnormality (leukopenia, increased erythrocyte sedimentation rate, thrombocytopenia, increase in hepatic enzymes), or rapid degradation of a prosthetic valve. Chronic osteomyelitis, hepatitis, and infection of an aneurysm and vascular prosthesis have been reported.20 Leukopenia may be observed; thrombocytopenia is frequent, as are increases in hepatic enzymes. Lupus anticoagulant associated with antiphospholipid antibodies may be observed (Chapter 242), as may anti–smooth muscle antibodies. During endocarditis, antinuclear antibodies, microhematuria, and rheumatoid factor are frequently found.  

DIAGNOSIS

The diagnosis is based mainly on serology (see Table 311-4). Direct detection by culture and PCR or immunochemistry in valve, liver, or blood samples is also useful, but serologic evaluation by IFA is the reference method. Two antigens (phase I and phase II) can be tested. Acute Q fever is diagnosed when seroconversion or a four-fold increase is obtained with phase II antigen.

A single serum test exhibiting IgG antibodies of 200 or greater and IgM of 50 or greater against phase II is also diagnostic. During chronic Q fever, antibodies are at higher titer and directed against both phase I and phase II. IgG against phase I at a titer of 800 or 1600 is diagnostic of chronic infection, as is IgA at 100 or greater. Serology is useful for follow-up of patients with acute Q fever and underlying disease and those with treated chronic Q fever. Culture may be performed in a BSL-3 laboratory. Although axenic culture has been reported, C. burnetii is most commonly grown in cell monolayers (notably, human lung endothelial cells). In patients with vascular infection or prosthetic valve endocarditis, fluorodeoxyglucose (FDG)–positron emission tomography (PET)/computed tomography (CT) has proved useful.

TREATMENT  Treatment is straightforward during acute Q fever. Doxycycline is the most efficient antimicrobial, and it should be prescribed for 2 weeks. Some patients with hepatitis do not respond well because of an excessive immune response. They rapidly improve with a short course of glucocorticoids. In pregnant women, cotrimoxazole during the entire pregnancy may decrease the chance of an unfavorable outcome. As for endocarditis, bactericidal treatment is necessary. In vitro, antimicrobial efficacy is impaired by the low pH of the intracellular host vacuoles within which C. burnetii resides. Hydroxychloroquine increases the pH of these vacuoles and restores the bactericidal effect of doxycycline. In patients with endocarditis, the recommended treatment is a combination of doxycycline (200 mg daily) and hydroxychloroquine (600 mg/day, then adjusted to reach a 1-mg/mL plasma concentration). This regimen is prescribed for 18 to 36 months based on serologic results. A more rapid favorable outcome can be obtained with doxycycline serum levels higher than 5 µg/mL. Some strains may be resistant to doxycycline, and new macrolides may be an alternative. The major adverse effect of this treatment is photosensitivity; sun exposure should be avoided. An alternative treatment is a combination of doxycycline and ofloxacin for 3 years or more.



PREVENTION

Prevention is based on veterinary control in animals. A vaccine is currently available in Australia.

  Grade A Reference A1. Kim YS, Kim DM, Yoon NR, et al. Effects of rifampin and doxycycline treatments in patients with uncomplicated scrub typhus: an open-label, randomized, controlled trial. Clin Infect Dis. 2018;67:600-605.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 311  Rickettsial Infections  

GENERAL REFERENCES 1. Blanton LS. The rickettsioses: a practical update. Infect Dis Clin North Am. 2019;33:213-229. 2. Tomassone L, Portillo A, Nováková M, et al. Neglected aspects of tick-borne rickettsioses. Parasit Vectors. 2018;11:1-11. 3. McFee RB. Tick borne illness: Rocky mountain spotted fever. Dis Mon. 2018;64:185-194. 4. Sathiamoorthi S, Smith WM. The eyes and tick-borne disease in the United States. Curr Opin Ophthalmol. 2016;27:530-537. 5. Gottlieb M, Long B, Koyfman A. The evaluation and management of Rocky Mountain spotted fever in the emergency department: a review of the literature. J Emerg Med. 2018;55:42-50. 6. Eldin C, Parola P. Update on tick-borne bacterial diseases in travelers. Curr Infect Dis Rep. 2018;20:1-9. 7. Angelakis E, Bechan Y, Raoult D. The history of epidemic typhus. Microbiol Spectr. 2016;4:1-9. 8. Rauch J, Eisermann P, Noack B, et al. Typhus group rickettsiosis, Germany, 2010-2017. Emerg Infect Dis. 2018;24:1213-1220. 9. Xu G, Walker DH, Jupiter D, et al. A review of the global epidemiology of scrub typhus. PLoS Negl Trop Dis. 2017;11:1-27. 10. Jung LY, Jeon M, Choi SH, et al. Relative bradycardia in scrub typhus. Am J Trop Med Hyg. 2017;97:1316-1318.

2026.e1

11. Mahajan SK, Mahajan SK. Neuropsychiatric manifestations of scrub typhus. J Neurosci Rural Pract. 2017;8:421-426. 12. Jain D, Nand N, Giri K, et al. Scrub typhus infection, not a benign disease: an experience from a tertiary care center in Northern India. Med Pharm Rep. 2019;92:36-42. 13. Koralur M, Singh R, Varma M, et al. Scrub typhus diagnosis on acute specimens using serological and molecular assays: a 3-year prospective study. Diagn Microbiol Infect Dis. 2018;91:112-117. 14. Saraswati K, Day NPJ, Mukaka M, et al. Scrub typhus point-of-care testing: a systematic review and meta-analysis. PLoS Negl Trop Dis. 2018;12:1-18. 15. Ismail N, McBride JW. Tick-borne emerging infections: ehrlichiosis and anaplasmosis. Clin Lab Med. 2017;37:317-340. 16. McFee RB. Tick borne illness: anaplasmosis. Dis Mon. 2018;64:181-184. 17. Fanoy E, Dijkstra F, van der Hoek W, et al. Familiarity of general practitioners with Q fever decreases hospitalization risk. Neth J Med. 2018;76:184-189. 18. Kelm DJ, White DB, Fadel HJ, et al. Pulmonary manifestations of Q fever: analysis of 38 patients. J Thorac Dis. 2017;9:3973-3978. 19. de Lange MMA, Gijsen LEV, Wielders CCH, et al. Should acute Q-fever patients be screened for valvulopathy to prevent endocarditis? Clin Infect Dis. 2018;67:360-366. 20. Mongardon N, Dubory A, Dorget A, et al. Vascular infection and vertebral destruction in a patient with Q fever. Lancet Infect Dis. 2018;18:226.

2026.e2

CHAPTER 311  Rickettsial Infections  

REVIEW QUESTIONS 1. A 50-year-old male patient with acute Q fever presents after 7 days of fever, when he became spontaneously afebrile. You should: A . Prescribe 15 days of doxycycline. B. Evaluate clinically and/or echocardiographically if he has a predisposing underlying cardiovascular factor to chronic endocarditis. C. Perform hepatic echography to detect hepatomegaly. D. Explain to the patient that because he is spontaneously cured, he does not need follow-up. E. Investigate the possible environmental source of the disease. Answer: B  It is critical in such patients to determine the risk for endocarditis. 2. A 6-year-old boy is diagnosed with Rocky Mountain spotted fever and presents with high fever. You prescribe: A . A treatment of doxycycline adjusted to weight B. Chloramphenicol intravenously C. Ciprofloxacin orally D. Erythromycin intravenously E. Ceftriaxone intravenously Answer: A  Doxycycline is recommended even in children for the treatment of potentially severe diseases. 3. A patient with body lice exposure and fever returns from Eastern Congo and you suspect louse-borne typhus. You: A . Prescribe immediately 2 pills of 100-mg doxycycline. B. Prescribe intravenous chloramphenicol. C. Sample blood for serology and wait for the result to treat. D. Sample blood for polymerase chain reaction and wait for the result to treat. E. Sample a stain biopsy and wait for the result to treat. Answer: A  This is an emergency; the patient should be treated as soon as the diagnosis is suspected.

4. A pregnant women who traveled in Thailand received a diagnosis of scrub typhus. You should treat with which of the following drugs? A . Doxycycline B. Ciprofloxacin C. Chloramphenicol D. Rifampicin E. Streptomycin Answer: D  Rifampicin is the reference treatment for scrub typhus when doxycycline is contraindicated. 5. Which of these tick-bite diseases may be commonly superinfected by yeast or opportunistic bacteria? A . Human monocytic ehrlichiosis B. Infection by Ehrlichia canis C. Rocky Mountain spotted fever D. Human granulocytic anaplasmosis E. African tick-bite fever Answer: D  Granulocytopenia favors superinfection by opportunistic pathogens.

2026

CHAPTER 312 Zoonoses  

312  ZOONOSES J. STEPHEN DUMLER AND MEGAN E. RELLER



DEFINITION

Zoonoses, derived from the Greek words zōio (animal) and nósos (disease), are infectious diseases transmitted from animals, both wild and domestic, to humans or from humans to animals. Per the 2016 Global Burden of Disease Study, of 44 billion incident cases attributed to 328 diseases and injuries, 55% were caused by infectious or communicable diseases.1 Critically important among communicable diseases are emerging or reemerging infectious diseases, 60 to 75% of which are zoonoses. Zoonotic pathogens or their toxins can be transmitted by several different routes: direct contact with animals or their

CHAPTER 312 Zoonoses  

ABSTRACT

Zoonoses are infections and/or diseases acquired by humans from animals. They are distinct from anthroponoses that only involve humans, even if through an arthropod vector. Zoonoses make up a significant proportion of all emerging infections and are caused by a complete range of pathogens, including prions, viruses, bacteria, fungi, protozoa, and helminths. The emergence of zoonoses is in part determined by the concept of zoonotic “spillover,” whereby barriers that prevent access of zoonotic agents to humans are breached. Specific risks for emergence of zoonoses include origins in tropical forest regions; increases in mammal biodiversity; and changes in anthropogenic land use related to agricultural practices or urban residence or built-up land relating to human activity. Zoonoses can be acquired by direct animal contact, animal bites or scratches, bites or mechanical transmission from arthropod vectors, and consumption of contaminated food or water. The clinical manifestations, diagnostic methods, and treatments are specific for individual infectious agents and disease, but prevention can often be applied if proactively sought by use of vaccines, protective clothing, and other prophylactic measures.

KEYWORDS

zoonosis vectors climate change human activity zoonotic spillover

2026.e3

CHAPTER 312 Zoonoses  

products (e.g., tissue, urine, or excreta); animal bites and scratches; arthropod vectors; and consumption of contaminated food or water. Zika virus, Borrelia burgdorferi, Babesia microti, and Plasmodium knowlesi are well-recognized vector-transmitted zoonotic pathogens.2 However, many arthropod vector– borne pathogens, such as Plasmodium falciparum or Bartonella quintana, are not zoonotic because they are transmitted from human to human through the vector and lack an animal reservoir. Similarly, many infections acquired from the environment are not zoonoses either; examples include melioidosis (Burkholderia pseudomallei) and histoplasmosis (Histoplasma capsulatum). Understanding the epidemiology and transmission of zoonotic versus nonzoonotic infections is critical for implementing control measures at local, regional, and international levels. Implicit in the distinction of “zoonosis” from “anthroponosis” is the concept of spillover from animal host reservoirs.3 Zoonotic spillover occurs when barriers that prevent a zoonotic agent from gaining access to and infecting a human are breached or fail. Unfortunately, the specific mechanisms are not always well-defined and can be difficult to predict, which impairs our ability to prepare for the emergence of new zoonoses. Such barriers include (1) factors related to the interaction between reservoir host and pathogens, including both the reservoir host distribution and density, as well as the prevalence of pathogens within hosts and the likelihood of release from the infected reservoir host; (2) the ability of the pathogen to survive and disseminate after release; and (3) human host-specific exposures and risk factors that affect the ability of the pathogen to enter and propagate within the human host, such as structural barriers, immunity to infection, and genetic predisposition to infection. Zoonotic infections, not surprisingly, are difficult to predict because distinct barriers are not equally shared for each pathogen/reservoir combination owing to the great genomic diversity of animals, humans, and microbes. 4  

EPIDEMIOLOGY

For many millennia, humans have associated themselves with wildlife and domestic animals, whether farm animals or companion pets. Dogs, sheep, goats, and other animals were domesticated as long as 10,000 to 30,000 years ago; in the United States today, approximately 60% of people own a pet. Dogs and cats are most frequent. However, other pets include fish, birds, reptiles, hamsters, guinea pigs, rabbits, and horses; exotic animals, including large cats, primates, and bears, are legal with or without permit in many (16 and 20, respectively) U.S. states. Wildlife-related recreation places more than 90 million U.S. residents at increased risk for zoonoses, and 72 million observe, feed, or photograph birds and other wildlife. Climate change is expected to increase the risk for zoonoses over the next 50 years because of extreme temperatures that precipitate increased rainfall and violent storms; rising sea level and floods that could impact zoonotic diarrheal disease; loss of biodiversity; collapse of ecosystems; an increase in pests; and the spread and increase in both neopteran (biting insect) and acarid (tick and mite) vectors of zoonotic pathogens.5 Most emerging zoonoses are acquired from contact with wildlife. In recent analytical models weighted for reporting bias, key factors predicting emergence

2027

of zoonoses are origin in tropical forest regions, increases in mammal biodiversity, and changes in anthropogenic land use related to agricultural practices6,7 (Fig. 312-1). In unweighted models that do not account for reporting bias, urban residence or built-up land relating directly to human activity was by far the strongest predictor of risk. Detailed analyses using refined data pools such as these will increasingly help define general mechanisms such that attention can be focused in specific high-risk environments for early detection and implementation of interventions to minimize or abrogate risk and emergence of zoonoses. Zoonoses are transmitted through one or more of several mechanisms: (1) direct contact with animals or infected materials and/or related inhalation; (2) animal bites and scratches; (3) bites or mechanical transmission by arthropod vectors; and (4) consumption of contaminated foods (Table 312-1). Direct contact with animals is a well-documented mechanism for acquiring viral, bacterial, protozoan, and fungal zoonotic pathogens. The risk is particularly high for those whose occupations or recreations involve contact with animals. For example, in a study of veterinarians in Oregon, contact-related ringworm was the most common zoonosis and rabies the most dangerous; cats were the most likely source of contact-related zoonoses. Significant risks also exist for farmers, abattoir workers, and others. Leptospirosis, likely one of the most neglected of all zoonoses worldwide with an estimated 1 million cases and nearly 60,000 deaths annually, is commonly acquired through direct contact with contaminated rodent urine; however, many wild and domestic animals can become colonized and excrete the spirochete in their urine.8 Coxiella burnetii is distributed worldwide and acquired by inhalation of dust contaminated with excretions, such as milk, urine, and feces, and/or birth products, from infected animals.9 In addition to direct tissue injury, animal scratches and bites are a common cause of zoonotic infections. There are estimated 4.5 million dog bites annually in the United States alone, and 900,000 result in infection; therefore, approximately 1 in every 72 people is affected. Examples of pathogens include Capnocytophaga canimorsus from dog bites and Pasteurella multocida and Bartonella henselae, the etiologic agent of cat-scratch disease, from cat bites. In some regions of the world, endemic rabies is rampant. Perhaps best known are zoonoses that are transmitted through arthropod bites, such as mosquitoes, flies, or ticks and mites. Vector-borne zoonotic viral infections include West Nile virus, Eastern equine encephalitis virus, Hendra virus, and others. P. knowlesi, B. microti, Leishmania spp, and Trypanosoma spp are examples of vector-borne zoonotic protozoal infections. Leishmania is associated with epidemics (799,000 cases in 2016) across southern Asia, northern Africa, and South America. Trypanosoma brucei rhodesiense and T. brucei gambiense, agents of African trypanosomiasis and human sleeping sickness, are transmitted by biting tsetse flies in Africa; nearly 11.5 million new cases occurred in 2016. Ticks and mites are the most important vectors of zoonotic disease in the developed world. In the United States from 2012 to 2016, 217,000 of 242,000 (90%) reported cases of vector-borne zoonosis were attributed to tick-borne infections, especially B. burgdorferi (Fig. 312-2); recent studies provide evidence that the actual number of B. burgdorferi

Estimated risk High

Low

FIGURE 312-1.  Heat map of predicted zoonotic emerging infectious disease estimated risk locations after factoring out reporting bias. Heat map estimated risk scale on right: highest risk is blue; lowest risk is yellow. (Adapted from Allen T, Murray KA, Zambrana-Torrelio C. Global hotspots and correlates of emerging zoonotic diseases. Nat Commun. 2017;8:1124.)

2028

CHAPTER 312 Zoonoses  

TABLE 312-1 EXAMPLES OF RECENTLY EMERGED AND REEMERGING ZOONOSES DISEASE/ETIOLOGIC AGENT

TRANSMISSION

RESERVOIR HOST

CLINICAL MANIFESTATIONS

VIRUSES Bourbon virus

Ticks?

?

Fever, fatigue, rash, headache, myalgia, nausea, vomiting; leukopenia, thrombocytopenia

Heartland viruses

Amblyomma americanum tick bites

Deer?

Fever, headache, fatigue, anorexia, nausea, myalgia, arthralgia; thrombocytopenia, leukopenia

Zika virus

Aedes aegypti and Aedes albopictus mosquito bites

Nonhuman primates, sheep, bats, rodents

Rash, fever, arthralgia, conjunctivitis, myalgia, headache, retroorbital pain; congenital microcephaly, Guillain-Barré syndrome

Avian influenza (H5N1)

Inhalation

Poultry, wild and domestic waterfowl, other birds

Influenza-like illness, shortness of breath, difficulty breathing, viral pneumonia with or without acute respiratory distress syndrome, respiratory failure, multiorgan diseases

Indirect (aerosol) exposure

Goats, sheep, cattle, wildlife

Acute: fever, fatigue, photophobia, headache; cough with pneumonia; hepatitis; chronic: endocarditis, immunosuppression, chronic renal insufficiency

Amblyomma maculatum ticks Dermacentor occidentalis ticks

Small mammals, cattle? Small mammals

Fever, rash, eschar, myalgia, chills, fatigue, arthralgia, headache, lymphadenopathy

Fleas

Small mammals

Fever, chills, headaches, carbuncles, lymphadenopathy with buboes, pneumonia, sepsis

Zoonotic malaria Plasmodium knowlesi

Anopheles leucosphyrous group mosquitoes

Macaques and monkeys

Fever, chills, headache, rigors, malaise, myalgia, cough, nausea, abdominal pain, vomiting, diarrhea

Babesiosis Babesia microti

Ixodes scapularis ticks

Small mammals

Fever, anemia, thrombocytopenia, elevated lactate dehydrogenase, hyperbilirubinemia, increased alanine aminotransferase and aspartate aminotransferase

Chagas disease Trypanosoma cruzi

Triatomine insects in Triatoma, Rhodnius, and Panstrongylus genera

Rodents, raccoons, skunks, coyotes

Chagoma (Romaña sign), fever, lymphadenopathy, edema, hepatosplenomegaly, myocarditis, meningoencephalitis; cardiomyopathy (arrhythmia, heart failure, atrioventricular/branch blocks, thromboembolism); gastrointestinal (dysperistalsis, megaesophagus, megacolon)

Baylisascariasis Baylisascaris spp.

Ingestion of contaminated materials

Raccoons, skunks, badgers

Larva migrans (visceral, neural); diffuse unilateral subacute neuroretinitis

Direct contact, indirect contact?

Bamboo rats

Fever, cutaneous or subcutaneous lesion(s), malaise, anemia, dyspnea, weight loss, lymphadenopathy, nonproductive cough, anemia

BACTERIA AND RICKETTSIAE Q fever Coxiella burnetii Spotted fever rickettsiosis   Maculatum rickettsiosis   Rickettsia parkeri   Pacific Coast tick fever   Rickettsia “philipii” 364D Plague Yersinia pestis

Fever, headache, myalgia, arthralgia, malaise, lymphadenopathy (children), rash, eschar

PROTOZOA AND HELMINTHS

FUNGI Talaromycosis Talaromyces (Penicillium) marneffei

infections is more than 10 times that reported. Worldwide, ticks, lice, and mites are also significant, but underrecognized, causes of vector-borne bacterial zoonoses. Examples of neglected vector-borne bacterial zoonotic disease include tick- and louse-borne relapsing fever borreliosis; spotted fever group rickettsioses, transmitted principally by hard body (ixodid) ticks; and typhus group rickettsioses, transmitted by fleas and lice. Scrub typhus, caused by the rickettsia Orientia tsutsugamushi and transmitted by the larval stage (chigger) of a trombiculid mite, occurs across much of Asia and Oceania and is now emerging in Africa and South America; it is estimated to cause as many as 1,000,000 infections each year globally.10 Many important zoonotic infections are acquired through consumption of contaminated foods or water. The list of potential pathogenic agents is vast and includes viruses (such as hepatitis E virus); prions (such as bovine spongiform encephalopathy); bacteria (such as Salmonella, Campylobacter, Listeria; and Brucella); protozoa (such as Cyclospora, Cryptosporidium,11 Toxoplasma, and Giardia); and helminths (such as Taenia, Trichinella, Opisthorchis, and Clonorchis).  

CLINICAL MANIFESTATIONS

The clinical manifestations of zoonoses are as variable as those for nonzoonotic infections. Symptoms and signs may be referent to the skin, the gastrointestinal tract, the respiratory tract, the central nervous system, the musculoskeletal system, or major organs such as liver, kidneys, and heart; zoonotic infections can also present as undifferentiated fever, sepsis, or toxic shock–like syndromes. For the clinical manifestations with specific zoonotic agents or syndromes, the individual chapters that address each should be consulted. Although most

zoonoses are acquired locally, early identification requires taking a comprehensive history, including possible animal exposures and domestic and international travel.  

DIAGNOSIS

The differentiation of zoonotic from nonzoonotic diseases is complex and challenging but often includes the same general diagnostic approaches. The key is to obtain sufficient historical details that might suggest a zoonotic exposure and which kind. Acquisition of information regarding potential exposure to animals must be thorough and should include occupation, travel, recreation, as well as pets (e.g., reptiles, exotic birds). One should enquire about travel to suburban and rural areas, where ticks are generally more prevalent, as well as international travel. One should ask about possible direct or indirect contact with animals, including bites and scratches, and exposure to vectors associated with zoonotic infection. Finally, one should ask about ingestion of specific food types associated with food-borne zoonoses, such as unpasteurized dairy products associated with Brucella and Listeria. A careful physical examination, including a thorough skin evaluation of rashes, eschars, ulcers, or other lesions, is important. General laboratory and imaging studies as required can provide additional data. Depending on the specific suspected agent, confirmation of the etiology could require blood culture, blood smear, paired acute and convalescent serology to document a four-fold rise in antibody titer, and/or molecular testing. Mandatory reporting is required for some agents to protect the public’s health.12 Some tests are only available from large reference laboratories or at the U.S. Centers for Disease Control and Prevention.

Vector-borne zoonotic disease, United States 2011-2016 Eastern equine encephalitis virus California serogroup disease virus disease 0.02% 0.18%

St. Louis encephalitis virus disease 0.02%

West Nile virus disease 5.19%

Western equine encephalitis virus disease Chikungunya virus 0.00% disease 1.10% Plague 0.01%

Powassan virus disease 0.02%

Tularemia 0.42% Spotted fever rickettsiosis 7.62%

Dengue virus infection 1.22%

Ehrlichiosis/anaplasmosis 8.41%

Malaria 3.21% Babesiosis 3.40%

Lyme disease 69.17%

FIGURE 312-2.  Proportion of vector-borne zoonoses by pathogenic agent as reported to the U.S. Centers for Disease Control and Prevention in the United States between 2011 and 2016; N = 286,017 cases. (Data from references 9-14).

TREATMENT  As diverse as they are, zoonoses will have specific treatments based on identification of the causative agent. This information is supplied in detail in the relevant chapters of this textbook.



PROGNOSIS

The prognosis of zoonoses is highly variable and dependent on the specific agent and disease. However, some zoonoses have high case fatality rates. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.



PREVENTION

Prevention of zoonotic diseases requires use of personal protection (frequent handwashing, gloves, and disinfectants) to avoid direct contact with infected animals; application of insect repellent to minimize exposure to vectors and their bites; and avoidance of consumption of undercooked meats, unpasteurized dairy products, uncooked seafood, and tap water in areas of the world where sanitation is inadequate. For some diseases, vaccines are available, and individuals at risk because of occupation (such as farmers, veterinarians, forestry workers, microbiology laboratorians), recreation, or travel to endemic areas should be vaccinated. When possible, pets and domestic farm animals should also be vaccinated. Immunocompromised individuals who are potential hosts must be particularly careful. Finally, some pathogens warrant patient isolation to prevent spread. Historically, this list has included hemorrhagic fever viruses, Hantaviruses, rabies, severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS) coronaviruses, and anthrax; however, the list has grown, and biocontainment units could be increasingly required in tertiary care referral hospitals. It has been argued that more sophisticated analyses will be required to estimate the economic impact of emerging infectious diseases, most of which are zoonotic in etiology, in order to provide a rationale and justification for the global health policies and mitigation strategies that will have to be implemented in the future at considerable expense. Examples of such global strategies that have been proposed to have high impact on prevention of zoonotic infections include implementation of controls on wildlife trade and consumption; identification of land use conversions that most likely lead to emergence of zoonoses; and targeted global surveillance programs to identify novel pathogens of zoonotic potential before they emerge.13

CHAPTER 312 Zoonoses  

GENERAL REFERENCES 1. GBD 2016 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet. 2017;390: 1211-1259. 2. Pisarski K. The global burden of disease of zoonotic parasitic diseases: top 5 contenders for priority consideration. Trop Med Infect Dis. 2019;4:1-9. 3. Plowright RK, Parrish CR, McCallum H, et al. Pathways to zoonotic spillover. Nat Rev Microbiol. 2017;15:502-510. 4. McMahon BJ, Morand S, Gray JS. Ecosystem change and zoonoses in the Anthropocene. Zoonoses Public Health. 2018;65:755-765. 5. Watts N, Adger WN, Ayeb-Karlsson S, et al. The Lancet Countdown: tracking progress on health and climate change. Lancet (London, England). 2017;389:1151-1164. 6. Han BA, Kramer AM, Drake JM. Global patterns of zoonotic disease in mammals. Trends Parasitol. 2016;32:565-577.

2029.e1

7. Allen T, Murray KA, Zambrana-Torrelio C. Global hotspots and correlates of emerging zoonotic diseases. Nat Commun. 2017;8:1-10. 8. Costa F, Hagan JE, Calcagno J, et al. Global morbidity and mortality of leptospirosis: a systematic review. PLoS Negl Trop Dis. 2015;9:1-19. 9. Trinh P, Zaneveld JR, Safranek S, et al. One health relationships between human, animal, and environmental microbiomes: a mini-review. Front Public Health. 2018;6:1-9. 10. Bonell A, Lubell Y, Newton PN, et al. Estimating the burden of scrub typhus: a systematic review. Plos Negl Trop Dis. 2017;11:1-17. 11. Benedict KM, Collier SA, Marder EP, et al. Case-case analyses of cryptosporidiosis and giardiasis using routine national surveillance data in the United States—2005-2015. Epidemiol Infect. 2019;147:1-5. 12. Adams DA, Thomas KR, Jajosky RA, et al. Summary of notifiable infectious diseases and conditions—United States, 2015. MMWR Morb Mortal Wkly Rep. 2017;64:1-143. 13. Cunningham AA, Daszak P, Wood JLN. One Health, emerging infectious diseases and wildlife: two decades of progress? Philos Trans R Soc Lond B Biol Sci. 2017;1-8.

2029.e2

CHAPTER 312 Zoonoses  

REVIEW QUESTIONS 1. Which of the following is not a zoonosis? A . Baylisascariasis B. Plasmodium falciparum malaria C. Scrub typhus D. Talaromyces penicilliosis E. Plasmodium knowlesi malaria Answer: B  Zoonoses are infections or disease-causing agents transmitted from animals or their products to humans. P. falciparum is transmitted only from human to human through the bites of mosquitos. 2. An 18-year-old man was playing Frisbee with his dog. During an attempt to catch the Frisbee, the dog inadvertently bit the hand of the man and punctured the skin. Domestic animal bites are known to transmit infectious agents that can cause disease after bites and scratches. Which of the following is often a consequence of dog bites? A . Parvovirus B19 infection B. Dirofilaria immitis filariasis C. Toxoplasmosis D. Capnocytophaga canimorsus skin infection and sepsis E. Leishmaniasis Answer: D  Dog and cat bites are frequent and often lead to transmission of microbes that can lead to serious local and systemic disease. Common pathogens transmitted by dog bite include C. canimorsus and Pasteurella multocida, and cats can often transmit Bartonella henselae, the etiologic agent of cat-scratch disease. 3. A 59-year-old man was walking in scrub brush along the side of a river in southern Connecticut during June. Within 1 week, he was febrile, had abdominal pain and myalgias, and was hypotensive. Laboratory studies showed thrombocytopenia and elevated direct and indirect bilirubin. Potential zoonotic agents for this process include which one of the following? A . Plasmodium vivax B. Plasmodium knowlesi C. Hepatitis E virus D. Brucella melitensis E. Babesia microti Answer: E  While each of the responses could produce the clinical findings, the well-recognized changes in the increasing prevalence of reservoir hosts (white footed mice) and tick vectors which use these reservoir hosts (Ixodes scapularis) in this region of the United States, led to the recent zoonotic spillover and a rapidly increasing prevalence of the vector-borne zoonosis, human babesiosis.

4. Zoonotic spillover describes the occurrence when zoonotic agents circumvent barriers that ordinarily prevent their access to infection of human hosts. Which of the following are considered important barriers that must be breached for spillover and subsequent emerging zoonotic infections to occur? A . Human urban population density B. Reservoir animal host density C. Eradication of arthropod vectors for endemic regions D. Decreased diversity of wildlife hosts in endemic regions E. Altered microbiome of arthropod vectors in endemic regions Answer: B  The most important factors in zoonotic spillover include factors related to the interaction between reservoir host and pathogens, including both the reservoir host distribution and density as well as the prevalence of pathogens within hosts and the likelihood of release from the infected reservoir host; the ability of the pathogen to survive and disseminate after release; and human host-specific exposures and risk factors that affect the ability of the pathogen to enter and propagate within the human host, such as structural barriers, immunity to infection, and genetic predisposition to infection. 5. Some zoonotic infections are acquired by consumption of contaminated food or water. Which of the following infections is associated with this mechanism of transmission? A . Relapsing fever borreliosis B. Listeriosis C. Powassan virus infection D. Leptospirosis E. Melioidosis Answer: B  There are many microbial infections that can be transmitted by foods or contaminated water consumption. These range from prions (“mad cow disease,” or bovine spongiform encephalopathy), to bacteria (including Salmonella, Campylobacter, Listeria, and Brucella), to protozoa (such as Cyclospora, Cryptosporidium, Toxoplasma, and Giardia). to helminths (including Taenia, Trichinella, Opisthorchis, and Clonorchis, among many others). Relapsing fever borreliosis and Powassan virus are tick vector–transmitted, whereas both leptospirosis and melioidosis are acquired by contact with a contaminated environment.

CHAPTER 313 Actinomycosis  

2029

313  ACTINOMYCOSIS ITZHAK BROOK



DEFINITION

Actinomycosis is an uncommon, subacute to chronic bacterial infection that induces both suppurative and granulomatous inflammation. Localized swelling with suppuration, abscess formation, tissue fibrosis, and draining sinuses characterize this disease. The infection spreads contiguously and often forms draining sinuses that extrude characteristic but not pathognomonic “sulfur granules.” Infections of the oral and cervicofacial regions are the most common, but any site in the body can be infected, including the thoracic region, abdominopelvic region, and central nervous system (CNS). Musculoskeletal, cutaneous, or disseminated disease is rare but does occur. Actinomycosis sometimes mimics other diseases, particularly malignancy.1

The Pathogen

Actinomycetes of the genera Actinomyces, Propionibacterium, and Bifidobacterium act as the principal pathogens. However, 98 to 99% of actinomycoses are caused by non–spore-forming anaerobic or microaerophilic bacterial species

CHAPTER 313 Actinomycosis  

ABSTRACT

Actinomycosis is an uncommon, subacute to chronic bacterial infection that induces both suppurative and granulomatous inflammation. Localized swelling with suppuration, abscess formation, tissue fibrosis, and draining sinuses characterize this disease. The infection spreads contiguously and often forms draining sinuses that extrude characteristic but not pathognomonic “sulfur granules.” Infections of the oral and cervicofacial regions are the most common, but any site in the body can be infected, including the thoracic region, abdominopelvic region, and central nervous system (CNS). Musculoskeletal, cutanous or disseminated disease is rare but does occur.

KEYWORDS

actinomycosis “sulfur granules” cervicofacial thoracic abdominopelvic central nervous system musculoskeletal cutaneous disseminated disease

2029.e3

2030

CHAPTER 313 Actinomycosis  

of the genus Actinomyces, family Actinomycetaceae, order Actinomycetales. Of the 30 Actinomyces spp, 8 may cause disease in humans: the strictly anaerobic A. israelii, A. gerencseriae (formerly known as A. israelii serotype II), A. odontolyticus, A. naeslundii, A. meyeri, A. viscosus, A. pyogenes, and A. georgiae. A. israelii is the most common species causing human disease. P. propionicum (formerly known as Arachnia propionica) and B. dentium (formerly known as A. eriksonii) also are associated with clinically indistinguishable infection.2 The organisms are filamentous, branching, gram-positive, pleomorphic, non–spore-forming, non–acid-fast anaerobic, or microaerophilic bacilli. Actinomyces organisms are fastidious bacteria that require enriched culture media; 6 to 10% ambient CO2 may aid in their growth, which takes up to 2 to 3 weeks in culture. Most actinomycotic infections are polymicrobial and involve other aerobic and anaerobic bacteria. The most common co-isolates depend on the infection site and are Actinobacillus actinomycetemcomitans, Aggregatibacter aphrophilus, Eikenella corrodens, Bacteroides, Fusobacterium, Capnocytophaga, aerobic and anaerobic streptococci, Staphylococcus, and Enterobacteriaceae.  

EPIDEMIOLOGY

Actinomyces spp are members of the endogenous mucous membrane flora in the oral cavity, lower gastrointestinal tract, bronchi, and female genital tract. No external environmental reservoir, such as soil or straw, has been documented, nor has person-to-person transmission of pathogenic Actinomyces spp been demonstrated. Although infection can occur in all age groups, it is rarely seen in children or patients older than 60 years. Most cases are encountered in individuals in the middle decades of life. A male-to-female infection ratio of 3 : 1 is reported in most series. The explanation for this ratio is the higher prevalence of poor oral hygiene and oral trauma in men. The annual reported incidence in the United States is fewer than 100 cases. However, because of the fastidious nature of the organism, many cases are undiagnosed and the true incidence is probably much higher.  

PATHOBIOLOGY

Actinomyces spp are agents of low pathogenicity and require mucosal barrier disruption to cause disease. Actinomycosis usually occurs in immunocompetent persons but may afflict those with diminished host defenses. Risk factors include steroids, bisphosphonates, leukemia with chemotherapy, human immunodeficiency virus (HIV), alcoholism, lung and renal transplant receipt, and local tissue damage caused by trauma, recent surgery, or irradiation. Oral and cervicofacial diseases are commonly associated with dental caries and extractions, gingivitis and gingival trauma, infection in erupting secondary teeth, chronic tonsillitis, otitis or mastoiditis, diabetes mellitus, immunosuppression, immunodeficiency, malnutrition, and neoplastic disease. Pulmonary infections generally arise after aspiration of oropharyngeal or gastrointestinal secretions and have been reported in patients with underlying lung disorders, such as emphysema, chronic bronchitis, and bronchiectasis. Gastrointestinal infection frequently follows loss of mucosal integrity, such as with surgery, trauma, foreign bodies, perforated appendix or diverticulitis, neoplasia, foreign bodies, and emergency colonic surgery. Extended use (>2 years) of intrauterine contraceptive devices (IUDs) increases risk for the development of actinomycosis of the female genital tract. Other bacterial species that are frequently copathogens with Actinomyces spp may assist in the spread of infection by inhibiting host defenses and reducing local oxygen tension. Once the organism is established locally, it may spread progressively to surrounding tissues. The infection tends to spread without regard for anatomic barriers, including fascial planes and lymphatic channels. The end result is a chronic, indurated, suppurative infection (usually with draining sinuses and fibrosis, especially in pelvic and abdominal infection). The fibrotic walls of the mass before suppuration are “wooden” in nature and may be confused with a neoplasm. Hematogenous spread can be fulminant but is rare. Actinomyces spp grow in microscopic or macroscopic clusters of tangled filaments surrounded by neutrophils. Plasma cells and multinucleated giant cells are often observed with lesions, as are large macrophages with foamy cytoplasm around purulent centers. When visible, these clusters are pale yellow and exude through sinus tracks; they are called sulfur granules (originally called drusen). These granules (1 to 2 mm in diameter) are made of aggregates of organisms and contain calcium phosphate. A central purulent loculation surrounds the granules. Their centers have a basophilic staining property, with eosinophilic rays terminating in pear-shaped “clubs.” One to six granules can be present per loculation, and up to 50 loculations can be present in a lesion. Multicenter giant cells can be seen as well.

FIGURE 313-1.  Actinomycosis of the jaw, observed at Letterman General Hospital, San Francisco, Calif., in a sergeant who had punctured the floor of his mouth while picking his teeth. (Courtesy Office of Medical History, Office of the Surgeon General, U.S. Army.)



CLINICAL MANIFESTATIONS

Physicians must be aware of typical clinical manifestations, such as cervicofacial actinomycosis following a dental focus of infection, pelvic actinomycosis in women with an intrauterine device, and pulmonary actinomycosis in smokers with poor dental hygiene.3,4

Cervicofacial

Cervicofacial infection is the most common manifestation of actinomycosis (Fig. 313-1).5,6 It is generally odontogenic and evolves as a chronic or subacute, painless or painful soft tissue that is characterized as a slowly progressive, nontender swelling or mass involving the submandibular or paramandibular region. However, the submental and retromandibular spaces, temporomandibular joint, cheek, chin, and upper jaw can also be involved. The swelling may have a ligneous consistency caused by tissue fibrosis. Depending on the composition of the concomitant synergistic flora, the onset of actinomycosis may be acute, subacute, or chronic. When Staphylococcus aureus or β-hemolytic streptococci are involved, an acute painful abscess or a phlegmatous cellulitis may be the initial manifestation. Pain and trismus can be disproportionate to the degree of inflammation apparent. The chronic form of the disease is the most common presentation and is characterized by painless infiltration and bluish or reddish induration that generally progresses to form multiple abscesses and draining sinus tracts discharging pus that may contain sulfur granules in up to 25% of instances. Periapical infection, trismus, dyspnea, dysphagia, fever, pain, and leukocytosis may be present. The infection can extend to the carotid artery, tongue,7 sinuses, ears, mastoid, orbit, salivary glands, pharynx, masseter muscle, thyroid, larynx, trachea, or thorax. Bone (most commonly the mandible) may be invaded from the adjacent soft tissue and results in periostitis or osteomyelitis. Cervical spine or cranial bone infection may lead to subdural empyema and invasion of the CNS. The differential diagnosis includes tuberculosis (scrofula), fungal infections, nocardiosis, suppurative infections by other organisms, and neoplasms.

Thoracic

Thoracic actinomycosis is an indolent, slowly progressive process involving the pulmonary parenchyma and pleural space. This form accounts for 15 to 30% of actinomycosis cases and is caused by aspiration of infective material from the oropharynx, as well as rarely after esophageal perforation, by extension into the mediastinum from the neck, by spread from an abdominal site, or by hematogenous spread to the lung. Infection can spread from a pneumonic focus across lung fissures to involve the pleura and the chest wall, with eventual fistula formation and drainage containing sulfur granules (Fig. 313-2). The mediastinum, endocardium, and pericardium also rarely can be affected. Granules are seldom present in sputum. The incidence of this complication, as well as the destruction of thoracic vertebrae and adjacent ribs, has declined in the antimicrobial era.

CHAPTER 313 Actinomycosis  

2031

Pelvic

Pelvic infection is observed in patients with prolonged use of IUDs and also may occur from extension of intestinal infection, commonly from indolent ileocecal disease.10 Manifestations may range from a chronic vaginal discharge to pelvic inflammatory disease with tubo-ovarian abscesses or pseudomalignant masses. Patients generally have abnormal vaginal bleeding or discharge, abdominal or pelvic pain, menorrhagia, fever, and weight loss. Endometritis is the earlier form of the infection, followed by tubo-ovarian abscesses. Extension to the uterus, bladder, rectal area, abdominal wall, peritoneum, pelvic bones, thorax, and systemic circulation also can occur.

Central Nervous System

A

Infections of the CNS are very rare and generally manifest as single or multiple encapsulated brain abscesses11 that appear as ring-enhancing lesions with a thick wall that may be irregular or nodular on contrast-enhanced CT scans. There are no features that readily distinguish actinomycosis from other causes of chronic brain abscesses. Rarely, solid nodular or mass lesions termed actinomycetomas or actinomycotic granulomas are found.12 Headache and focal neurologic signs are the most common finding. Most actinomycotic infections of the CNS are seeded hematogenously from a distant primary site, but direct extension of cervicofacial disease also occurs. Sinus formation is not a characteristic of CNS disease. Meningitis caused by Actinomyces is chronic and basilar in location, and the cerebrospinal fluid pleocytosis is usually lymphocytic. Thus, it may be misdiagnosed as tuberculous meningitis. The meningitis of actinomycosis is characteristically a pachymeningitis that involves thickening of the dura mater.13 Extension to the cranial epidural or subdural space and spinal epidural space also can occur from adjacent foci.  

B FIGURE 313-2.  Thoracic computed tomography scan of a 43-year-old woman with pulmonary actinomycosis. A, There is consolidation of the lung with pleural thickening adjacent to the parenchymal disease. B, Abscess extended into the left breast and inferiorly to the costophrenic sulcus, to the retroperitoneum, and into the lateral abdominal wall (arrow).

The complaints of patients with thoracic actinomycosis are nonspecific. The most common are chest pain, productive cough, dyspnea, weight loss, and fever. Anemia, mild leukocytosis, and elevated sedimentation rate are relatively common. There is often a history of underlying lung disease, and patients are rarely initially seen in an early stage of infection. The pulmonary lesion is either a mass lesion or pneumonitis and may resemble tuberculosis, especially when cavity formation occurs, or blastomycosis, which may destroy ribs posteriorly but rarely forms sinuses. Nocardiosis, aspergilloma, bronchogenic carcinoma, cryptococcosis, aspiration pneumonia, and lymphoma also can mimic thoracic actinomycosis.8 Pleural thickening, effusion, or emphysema is common.

Abdominal

Abdominal actinomycosis makes up approximately 20% of all cases of actinomycosis. It is a chronic, localized inflammatory process that can occur weeks, months, or years after the integrity of the gastrointestinal mucosa is breached by surgery, cancer, or trauma. Extension from the thorax or pelvis or through hematogenous dissemination also can occur.9 The ileocecal region is involved most frequently (usually after a perforated appendix) with the formation of a mass lesion. The infection extends slowly to contiguous organs, especially the liver, and may involve retroperitoneal tissues, the spine, or the abdominal wall. Hepatic, renal, or splenic dissemination is an uncommon complication. Persistent draining sinuses may form, and those involving the perianal region can simulate Crohn disease (Chapter 132) or tuberculosis (Chapter 308). The extensive fibrosis of actinomycotic lesions, recognized by the examiner as a mass, often suggests tumor. A frequent finding on computed tomography (CT) is an infiltrative mass with dense inhomogeneous contrast medium enhancement. Constitutional symptoms and signs are nonspecific, the most common being fever, diarrhea or constipation, weight loss, nausea, vomiting, pain, and the sensation of a mass.

DIAGNOSIS

Appropriate microbiologic and pathologic studies are essential for diagnosis. A high index of suspicion should be communicated to the microbiology laboratory, along with material from draining sinuses, from deep-needle aspiration, or from biopsy specimens. CT or ultrasound needle aspiration can be used to obtain a biopsy specimen. It is important to avoid contamination of the specimen by normal flora and administration of antimicrobial therapy before a specimen is obtained. Anaerobic culture is required, and no selective media are available to restrict overgrowth of the slow-growing Actinomyces by associated microflora. The presence, in pus or tissue specimens, of non–acid-fast, grampositive organisms with filamentous branching is suggestive of the diagnosis. The characteristic morphologic features of sulfur granules within are helpful. In tissue sections stained with hematoxylin-eosin, sulfur granules are round or oval basophilic masses with a radiating arrangement of eosinophilic terminal clubs. However, Actinomyces spp are infrequently visible in sections stained with hematoxylin-eosin; visualization is facilitated by special stains such as Gomori methenamine silver, p-aminosalicylic acid, McCallen-Goodpasture, and Brown-Brenn. Multiple biopsy sections from different tissue levels are recommended to improve the histopathologic diagnosis. The granules must be distinguished from similar structures that are sometimes produced in infections caused by Nocardia, Monosporium, Cephalosporium, Staphylococcus (botryomycosis), and others. Actinomyces and Arachnia can generally be differentiated from other gram-positive anaerobes by means of their growth rate (slow), by catalase production (negative, except for A. viscosus), and by gas-liquid chromatographic detection of the acetic, lactic, and succinic acids produced in peptone-yeast-glucose broth. Specific staining with fluorescent-conjugated monoclonal antibody testing, matrix-assisted laser desorption ionization timeof-flight mass spectrometry and molecular methods can be used, but these methods are not readily available to clinical microbiology laboratories. Imaging methods such as conventional radiography, CT, and magnetic resonance imaging do not provide a specific diagnosis but allow more accurate definition of the dimensions and extension of the infection.

TREATMENT  Prolonged antimicrobial therapy (i.e., 6 to 12 months) has typically been recommended for patients with all clinical forms of actinomycosis to prevent disease recrudescence. However, individualization of courses of therapy is recommended because the duration of treatment depends on the initial burden of disease, the site of infection, and the clinical and radiologic response. Adequate drainage is indicated if abscesses are present.

Penicillin G is the drug of choice for treatment of an infection caused by any of the Actinomyces. It is given in high dosage during a prolonged period because the infection has a tendency to recur. Most deep-seated infections can be expected to respond to intravenous penicillin G, 18 to 24 million units/day given for 2 to 6 weeks, followed by an oral phenoxypenicillin in a dosage of 2 to 4 g/day. A few additional weeks of oral penicillin therapy may suffice for uncomplicated cervicofacial disease; complicated cases and extensive pulmonary or abdominal disease may require treatment for 12 to 18 months. Little evidence exists of acquired resistance to penicillin G by Actinomyces during prolonged therapy. The combination of a penicillin (i.e., amoxicillin, piperacillin) and a β-lactamase inhibitor (i.e., clavulanate, tazobactam) offers the advantage of coverage against penicillin-resistant aerobic and anaerobic copathogens.14 Alternative first-line antibiotics include amoxicillin, tetracycline, erythromycin, and clindamycin. Ceftriaxone, imipenem, and fluoroquinolones have also been used successfully. Metronidazole, aminoglycosides, oxacillin, and cephalexin are not effective. In vitro antimicrobial susceptibility testing of Actinomyces is difficult, and the results may not be predictive of antimicrobial effects in vivo.

The need to use combination antimicrobial therapy to eradicate microorganisms that are isolated in association with Actinomyces has not been established. Clinical and microbiologic cure was achieved in more than 85% cases of cutaneous actinomycosis with amoxicillin/clavulanic acid 875/125 mg BID orally for up to 12 weeks in one series.15 However, because many of these organisms are known pathogens, treatment is usually appropriate, especially with lower abdominal infections. Surgical removal of infected tissue also may be necessary in some cases, especially if extensive necrotic tissue or fistulas are present, if malignant disease cannot be excluded, and if large abscesses cannot be drained by percutaneous aspiration. When well-defined IUD-related symptoms and Papanicolaou smears demonstrate Actinomyces by specific fluorescence-labeled antibody, the IUD should be removed. Antimicrobial administration for a 2-week period may be indicated. More serious infections require prolonged therapy.  

PROGNOSIS

The availability of antimicrobial treatment has greatly improved the prognosis for all forms of actinomycosis. At present, cure rates are high and neither deformity nor death is common. GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 313 Actinomycosis  

GENERAL REFERENCES 1. Boyanova L, Koralov R, Mateva L, et al. Actinomycosis: a frequently forgotten disease. Future Microbiol. 2015;10:613-628. 2. Könönen E, Wade WG. Actinomyces and related organisms in human infections. Clin Microbiol Rev. 2015;28:419-442. 3. Paulo CO, Jordao S, Correia-Pinto J, et al. Actinomycosis, a lurking threat: a report of 11 cases and literature review. Rev Soc Bras Med Trop. 2018;51:7-13. 4. Bonnefond S, Catroux M, Melenotte C, et al. Clinical features of actinomycosis: a retrospective, multicenter study of 28 cases of miscellaneous presentations. Medicine (Baltimore). 2016;95:1-7. 5. Shikino K, Ikusaka M, Takada T. Cervicofacial actinomycosis. J Gen Intern Med. 2015;30:263. 6. Moturi K, Kaila V. Cervicofacial actinomycosis and its management. Ann Maxillofac Surg. 2018; 8:361-364. 7. Sadeghi S, Azais M, Ghannoum J. Actinomycosis presenting as macroglossia: case report and review of literature. Head Neck Pathol. 2018. [Epub ahead of print.] 8. Higashi Y1, Nakamura S, Ashizawa N, et al. Pulmonary actinomycosis mimicking pulmonary aspergilloma and a brief review of the literature. Intern Med. 2017;56:449-453.

2032.e1

9. Lisa-Gracia M, Martín-Rivas B, Pajarón-Guerrero M, Arnáiz-García A. Abdominal actinomycosis in the last 10 years and risk factors for appendiceal actinomycosis: review of the literature. Turk J Med Sci. 2017;47:98-102. 10. García-García A, Ramírez-Duran N, Sandoval-Trujillo H, et al. Pelivic actinomycosis. Can J Infect Dis Med Mircrobiol. 2017;2017:1-17. 11. Hwang CS, Lee H, Hong MP, et al. Brain abscess caused by chronic invasive actinomycosis in the nasopharynx: a case report and literature review. Medicine (Baltimore). 2018;97:1-4. 12. Mishra A, Prabhuraj AR, Bhat D, et al. Intracranial actinomycosis manifesting as a parenchymal mass lesion: a case report and review of literature. World Neurosurg. 2019;122:190-194. 13. Ravindra N, Sadashiva N, Mahadevan A, et al. Central nervous system actinomycosis—a clinicoradiologic and histopathologic analysis. World Neurosurg. 2018;116:e362-e370. 14. Steininger C, Willinger B. Resistance patterns in clinical isolates of pathogenic Actinomyces species. J Antimicrob Chemother. 2016;71:422-427. 15. Bonifaz A, Tirado-Sanchez A, Calderon L, et al. Treatment of cutaneous actinomycosis with amoxicillin/clavulanic acid. J Dermatolog Treat. 2017;28:59-64.

2032.e2

CHAPTER 313 Actinomycosis  

REVIEW QUESTIONS 1. Which of the following statements is correct regarding treatment of actinomycosis? A . Treatment should be guided by in vitro antimicrobial susceptibility testing. B. Prolonged therapy with penicillin is limited by increasingly frequent acquisition of resistance to it. C. Cervicofacial and thoracic actinomycosis generally requires combination antimicrobial therapy. D. Penicillin G is the drug of choice and should be given in high doses for a prolonged period because the infection has a tendency to recur. E. Surgical intervention is almost never required if antibiotic therapy selection and duration are correctly prescribed. Answer: D  Penicillin G is the drug of choice for treatment of an infection caused by any of the actinomycoses. The infection has a tendency to recur, so high doses for prolonged periods are generally required. Little if any evidence exists for significant acquired resistance to penicillin G during prolonged therapy. The need for combination antimicrobial therapy to eradicate actinomycosis has not been established; however, it is often considered clinically appropriate, especially with lower abdominal infections. Effective antibiotic therapy does not replace the need for surgical intervention when needed, such as removal of necrotic tissue, fistula repair, and drainage. In vitro antimicrobial susceptibility testing of Actinomycosis is difficult and the results may not be predictive of in vivo effects.

2. Cervicofacial actinomycosis is: A . A rare manifestation of the infection B. Usually caused by extension of primary intrathoracic involvement C. Acutely accompanied by disproportionately severe symptoms of pain and trismus D. Not prone to spread beyond the soft tissues of the neck, although its extension to those soft tissues can be extensive E. A chronic disease Answer: C  Cervicofacial infection is the most common manifestation of actinomycosis. It is generally odontogenic (not intrathoracic) in origin. Actinomycosis can be acute, subacute, or chronic. In its acute form, the symptoms of pain and trismus are disproportionate for the signs of inflammation that is apparent. Chronic actinomycosis can extend widely, including to the carotid arteries, tongue, sinuses, mastoid, orbit, thyroid, larynx, trachea, and contiguous muscle and bone. 3. Which of the following statements is correct regarding the epidemiology of actinomycosis? A . There is no person-to-person transmission. B. External reservoirs include soil and straw. C. Most diagnosed individuals are elderly. D. Males and females are comparably affected. E. It predominantly occurs in tropical climates. Answer: A  There is no evidence for person-to-person transmission or for any external environmental reservoirs (e.g., soil or straw) for actinomycosis. Ages of infected patients can range from childhood to old age, but peak incidence is in the middle decades of life. A male-to-female predominance (3 : 1 ratio) has been attributed to the increased prevalence of poor oral hygiene and oral trauma in males (because Actinomyces spp are endogenous flora of the oral cavity).

2032

CHAPTER 314 Nocardiosis  

N. transvalensis complex, N. farcinica, N. cyriacigeorgica, N. otitidiscaviarum, N. veterana, N. brasiliensis, and N. pseudobrasiliensis.  

EPIDEMIOLOGY

Nocardia spp are ubiquitous and primarily originate in soil. Despite being found throughout the environment, they rarely cause symptomatic infection in humans. Because nocardiosis is not a reportable disease, the frequency of this disease is mostly unknown. The annual incidence has been estimated to be 0.4 in 100,000. The risk for symptomatic Nocardia infection is greatly increased (estimated to be 140 to 340 times greater) in individuals who are immunocompromised, including patients who are receiving immunosuppressive agents following bone marrow or solid organ transplant,3 and patients with acquired immunodeficiency syndrome (AIDS). Corticosteroids are the most frequent immunosuppressant associated with nocardiosis4; however, cases also have been reported in patients receiving anti–tumor necrosis factor-α antibody (infliximab) as well as other immunosuppressants.5 It is important to keep in mind that trimethoprim-sulfamethoxazole prophylaxis does not always protect against Nocardia. Other risk groups include patients with cancer, Cushing disease, chronic granulomatous disease, and dysgammaglobulinemia. Patients with chronic pulmonary disorders, particularly alveolar proteinosis, are also more susceptible to this infection. In approximately one third of patients with nocardiosis, no predisposing condition can be identified.  

PATHOBIOLOGY

Most Nocardia spp gain entry to the host via the respiratory tract or less commonly by skin inoculation. Invading bacteria elicit a neutrophil response that inhibits but does not kill the organism. The bacteria are phagocytosed by neutrophils and macrophages and become enclosed in a membrane-bound phagolysosome. In this closed environment neutrophils and macrophages can kill many species of bacteria by synthesizing superoxide and hydrogen peroxide. However, Nocardia can survive in this hostile environment by producing superoxide dismutase, an enzyme that inactivates these toxic oxygen byproducts. In addition Nocardia spp produce a mycolic acid called cord factor that inhibits the fusion of lysosomes with the phagolysomal compartment, preventing toxic proteases and other antibacterial products from reaching the intracellular bacteria. Cell wall extractable lipids impair phagocytosis and inhibit bacterial killing. In addition to neutrophils and macrophages, cell-mediated and humoral immunity also play roles in protecting the host against Nocardia invasion, explaining the wide range of immunocompromised patients that are at increased risk for contracting nocardiosis.  

314  NOCARDIOSIS FREDERICK S. SOUTHWICK



DEFINITION

Nocardiosis refers to infections caused by Nocardia spp. Nocardia most commonly causes pneumonia but also can infect the central nervous system (CNS) and the skin. Less commonly, this organism can disseminate throughout the body. These infections usually occur in patients with defective immunity.

Etiology

Nocardia spp are thin, aerobic, gram-positive bacilli that form branching filaments.1 The bacteria stain irregularly and appear beaded on Gram stain. The speciation of Nocardia has been problematic. The original classification was based on the ability to use specific nutrients and decompose substrates such as adenine, casein, urea, gelatin, and xanthine. However, gene sequencing and DNA-DNA hybridization have now defined the true taxonomy.2 The species called N. asteroides was previously reported to be the most common cause of human disease. However, the majority of these bacteria were misidentified by today’s standards. The number of species causing human disease is large and includes N. abscessus, N. brevicatena/paucivorans complex, N. nova complex,

CLINICAL MANIFESTATIONS

Nocardiosis has no pathognomonic characteristics, and delays in diagnosis are common.6 Failure of a pulmonary or skin infection to respond to conventional antibiotic therapy should raise the possibility of a Nocardia spp infection. Nocardiosis always should be considered in the immunocompromised patient.7

Pulmonary Nocardiosis

Approximately two thirds of patients with nocardiosis present with pulmonary infection.8 Pulmonary disease is usually subacute in onset, mimicking a fungal or mycobacterial infection, and is most commonly misdiagnosed as tuberculosis. The most common complaints are a persistent cough producing purulent sputum, fever, anorexia, and weight loss. Less commonly, patients may report pleuritic chest pain and dyspnea. Hemoptysis is rare but can develop in patients with large cavitary lesions. Acute onset of pneumonia has occasionally been reported in the immunocompromised host.

Central Nervous System Infection

Approximately 5% of patients with a Nocardia infection have CNS involvement. Multilocular brain abscess is the most common CNS manifestation and is usually the consequence of transient bacterial dissemination from the lung. Lesions can occur in any region of the brain, and symptoms depend on location. Headache is the usual initial complaint and frequently localized to the site of the abscess. Patients also may present with neurologic deficits and seizures. The combined findings of a lung nodule on chest x-ray and a ringenhancing CNS lesion are often mistaken for metastatic lung carcinoma. Other diagnoses that should be considered when the immunocompromised host presents with both a lung and CNS focus are disseminated aspergillosis and toxoplasmosis. Meningitis is a less common CNS manifestation and is often associated with brain abscess (40% of meningitis cases). The CSF cell count

CHAPTER 314 Nocardiosis  

ABSTRACT

Nocardia species is a gram-positive and acid fast filamentous aerobic bacteria found in the soil that infects humans primarily through inhalation of dust, and less commonly secondary to cutaneous inoculation following trauma. This bacterium can cause nodular lung lesions and lymphocutaneous infections. When nocardia disseminates into the bloodstream it can result in multiloculated brain abscesses that can be identified by head CT scan with contrast. These infections are more common in immunocompromised patients. The organism is slow growing and takes several days to be identified. Speciation is now performed by PCR and certain strains, particularly N. farcinica, are trimethoprim-sulfa resistant; therefore, for serious infections combination therapy is recommended with ceftriaxone, imipenem, or amikacin. Linezolid and moxifloxacin have also been shown to be effective in some cases. In patients with brain abscess, surgical drainage in addition to long term-antibiotics are required. The recommended duration of therapy is 4 to 6 months for a normal host and 6 to 12 months for immunocompromised patients.

KEYWORDS

nocardiosis immunocompromised host lung nodule lymphocutaneous infection brain abscess mycetoma

2032.e3

CHAPTER 314 Nocardiosis  

2033

FIGURE 314-1.  Chest computed tomography showing a peripheral nodular lung lesion (arrow) caused by Nocardia infection. FIGURE 314-2.  Multiple, gadolinium-enhanced brain lesions (arrows) caused by Nocardia infection.

usually reveals predominantly neutrophils and the CSF culture may be positive, particularly if the culture is held for a prolonged period.

Cutaneous Infection

Skin infection is usually caused by N. brasiliensis and typically follows a break in the skin that is contaminated by soil. Cutaneous disease has been reported in association with trauma, a postoperative wound, insect bites, thorn bush scratches, or even a cat scratch. Initially a pustule or a moderately erythematous, nonfluctuant nodule develops at the site of inoculation. Erythema can extend along the lymphatic system and is associated with tender lymphadenopathy. This form of cutaneous infection has been termed lymphocutaneous or sporotrichoid disease. Similar skin manifestations are seen with other etiologies, including cat scratch disease, tularemia, Mycobacterium marinum, and sporotrichosis. In the immunocompromised host, disseminated infection may be manifest by multiple erythematous raised nodules and is an ominous finding. In tropical regions of South and Central America Nocardia spp can cause ulcerations and large tumor-like lesions called mycetomas that are usually found on the lower legs.  

DIAGNOSIS

Radiology

In pulmonary disease, chest x-ray findings are variable, with pulmonary nodules or mass lesions being seen most often (Fig. 314-1). Less frequently, consolidation, cavitary lesions with air-fluid levels, interstitial infiltrates, and pleural effusions are found. Chest CT often demonstrates areas of low attenuation within consolidations, multiple nodules, and chest wall extension of the infection. Patients with AIDS are more likely to have multiple pulmonary nodules, cavitary lesions, and upper lobe infiltrates. In some patients the infiltrate may resolve, particularly in patients with normal immune function. However, the patient may present with brain abscess several months later because of transient dissemination. In CNS infection CT or magnetic resonance imaging with contrast usually demonstrates one or more ringenhancing lesions (Fig. 314-2). Nocardia brain abscess is more commonly multiloculated; otherwise the radiologic findings are like those in other bacterial causes of brain abscess. Positron emission tomography is not helpful in differentiating Nocardia brain abscess from tumor; both demonstrate increased uptake.

Histopathology

Invasive procedures are generally required for specific diagnosis. For pulmonary infection, bronchoscopy with transbronchial biopsy or skinny needle biopsy is recommended. CT-guided needle aspirate is the diagnostic procedure of choice for brain abscess. Histopathologic examination usually reveals an acute inflammatory response with a predominance of neutrophils. Micronodular abscesses with minimal capsular formation are usually found. Gram stain or Brown-Brenn stains reveal gram-positive, beaded, branching forms. The morphology is identical to that of Actinomyces; however, the high lipid content of its cell wall often renders Nocardia modified acid-fast positive, whereas Actinomyces spp are modified acid-fast negative. However, acid fastness may be

variable when staining Nocardia colonies from cultures and is unreliable for direct clinical samples.

Culture

Isolation of the organism on culture provides a definitive diagnosis from needle aspirate samples of brain abscess, and a presumptive diagnosis when grown from respiratory and cutaneous samples. Nocardia grows best under aerobic conditions with 5 to 10% carbon dioxide. Because the organism grows slowly on blood agar, taking 3 to 5 days to form colonies, other organisms can overgrow. When Nocardia is suspected the clinical laboratory should be notified to allow the use of selective media and prolonged incubation.9 16S ribosomal RNA gene sequencing allows rapid speciation of isolates.10 Antibiotic susceptibility testing always should be performed to guide the choice of therapy.11

TREATMENT  Because of the rarity of nocardiosis, no prospective treatment trials have been performed, and all recommendations are based on retrospective studies and in vitro sensitivity testing. Sulfonamides remain the treatment of choice for pulmonary and cutaneous disease. Trimethoprim-sulfamethoxazole given orally at a dose of 160/800 mg (one double strength) three times daily is the most commonly used adult regimen. When patients with disseminated and/ or CNS Nocardia are treated with sulfonamides alone, survival has been less than 50%. One of the most common species to cause disseminated diseases, N. farcinica, is also one of the most common Nocardia spp to be resistant to sulfonamides. In these more serious conditions combination therapy is generally recommended,12 the exact regimen being guided by antibiotic susceptibility testing. One recommended empirical regimen is trimethoprim-sulfamethoxazole 15 mg/kg/day IV of the trimethoprim component divided into 2 to 4 doses, amikacin (7.5 mg/kg every 12 hours), and either ceftriaxone (2 g twice daily) or imipenem (500 mg four times daily). In a retrospective study, patients failing trimethoprim-sulfamethoxazole responded to imipenem with or without amikacin. Linezolid (600 mg twice daily) has been used successfully in a small number of CNS infections. However, prolonged therapy with this agent can lead to bone marrow toxicity and warrants weekly monitoring of the peripheral blood cell counts. The newer fluoroquinolone moxifloxacin (400 mg/day PO) has been shown to have activity against several strains of Nocardia, including N. farcinica and N. brasiliensis, and this agent may prove useful in patients who cannot tolerate sulfonamides; however, relapse has been reported. Minocycline (100 mg PO twice daily) and amoxicillin-clavulanate (875/125 mg PO twice daily) are other potentially effective alternative treatments for Nocardia. Because of the intracellular nature of Nocardia and the slow rates of bacterial growth, antibiotic treatment for 6 to 12 months in the immunocompromised host and 4 to 6 months in the normal host is usually required to prevent relapse.13 In addition, for patients with brain abscess or subcutaneous abscesses, surgical drainage as well as antibiotics are required for cure.14



PROGNOSIS

The overall mortality for nocardiosis is approximately 25%. In otherwise healthy individuals pulmonary nocardiosis has a better prognosis (15% mortality). The survival rate is worse in patients with bacteremia, patients with acute infection (symptoms for 500/µL), a predominance of lymphocytes (although neutrophils are sometimes prominent early in the course), elevated protein, and decreased glucose. AIDS patients most often have normal or only mildly abnormal findings as a result of their markedly defective immune response. Despite normal CSF findings with regard to cells, protein, and glucose, every AIDS patient with a headache must have a CRAG or LFA test and culture performed on CSF. It is extremely important that an opening pressure be obtained when lumbar puncture is performed. Especially in AIDS patients, extremely high intracranial pressure (>350 mm H2O) has been associated with poor outcomes and must be aggressively lowered. All patients with cryptococcal meningitis should undergo computed tomography or magnetic resonance imaging of the brain to look for mass lesions and to assess ventricular size. Whether normal hosts with isolated pulmonary cryptococcosis require cerebral imaging should be considered in the context of the individual patient. Obstructive hydrocephalus is uncommon but requires a shunting procedure to decrease the pressure. More commonly, the increased intracranial pressure with cryptococcal infection is associated with normal-sized ventricles and is due to blockage at the arachnoid villi or increased brain edema (or both), perhaps related to the osmotic effect of the polysaccharide capsule. Different methods for reducing pressure are used in this situation.

TREATMENT  Guidelines for the treatment of cryptococcal infection have been published by the Infectious Diseases Society of America (IDSA) and the World Health Organization.6

Central Nervous System Infection

Early multicenter randomized trials in non-AIDS patients showed superiority of the combination of amphotericin B and flucytosine for 6 weeks over amphotericin B alone for 10 weeks. Subsequent randomized trials in the azole era have been performed only in the AIDS population. They have confirmed the benefit of flucytosine added to amphotericin B for induction therapy and have shown that initial therapy with fluconazole alone or with amphotericin B alone is not as effective as therapy with amphotericin B and flucytosine. The combination of amphotericin B and flucytosine has been shown to be the most rapidly fungicidal regimen, and increasing numbers of reports have documented that rapid fungicidal activity that clears the organism from the CSF is associated with improved outcomes. Regimens using amphotericin B with fluconazole or flucytosine with fluconazole A1  are less effective but reasonable alternatives when the preferred regimen of amphotericin B plus flucytosine is not available. Unlike bacterial meningitis, adjunctive dexamethasone is not beneficial and appears to be detrimental in HIV-infected patients. A2  Current recommendations for AIDS patients are to give induction therapy with intravenous amphotericin B deoxycholate, 0.7 to 1 mg/kg daily, combined with oral flucytosine, 100 mg/kg daily given in four divided doses for at least 2 weeks, followed by consolidation therapy with oral fluconazole, 400 mg daily for a minimum of 8 weeks, and then suppressive therapy with fluconazole, 200 mg daily. Lipid formulations of amphotericin B at daily dosages of 3 to 5 mg/kg daily are increasingly used because they are less nephrotoxic; however,

2047

they are often not available in developing countries. Induction therapy with amphotericin B (1 mg/kg per day for 4 weeks) plus flucytosine (100 mg/kg per day for 2 weeks) is associated with improved survival among HIV-positive patients with cryptococcal meningitis compared with either amphotericin B alone or amphotericin B plus fluconazole (400 mg twice daily for 2 weeks). A3  Treatment strategies that could be more sustainable in underdeveloped areas of the world have been tested in a randomized trial of antifungal combinations to treat cryptococcal meningitis in Africa. One week of amphotericin B (1 mg per kilogram per day) plus flucytosine (100 mg per kilogram per day) and 2 weeks of fluconazole (1200 mg per day) plus flucytosine were found to be effective as induction therapy, A4  and can be used to treat cryptococcal meningitis in resource-limited settings. For patients who have undergone 12 months of antifungal therapy, who have CD4+ counts higher than 100/µL, and whose HIV viral load is undetectable on antiretroviral therapy, the suppressive therapy can be stopped. Suppressive therapy with fluconazole for transplant recipients is recommended for 6 to 12 months. The IDSA guideline recommendations for non–HIV-infected, non−transplant recipients are to treat with amphotericin B deoxycholate, 0.7 to 1.0 mg/kg daily, plus flucytosine, 100 mg/kg daily, in four divided doses for at least 4 weeks for induction therapy, followed by consolidation therapy with fluconazole, 400 mg daily for 8 weeks, and suppressive therapy with fluconazole, 200 mg daily for 6 to 12 months. Lipid formulations of amphotericin B may be used, and most physicians use a dose of 3 to 5 mg/kg daily, in the non-AIDS population. Only one treatment trial used voriconazole in combination with amphotericin B, and there are case reports on the use of voriconazole and posaconazole for salvage treatment of cryptococcal meningitis. These are reasonable alternatives if no other azoles can be used. The echinocandins are not active against C. neoformans and should not be used. A significant observation from the AIDS treatment trials was the role of increased intracranial pressure as a cause of early death from cryptococcal meningitis. An aggressive approach to the diagnosis and treatment of increased intracranial pressure in both AIDS and non-AIDS patients is mandatory and should include daily lumbar puncture or placement of a temporary lumbar drain or ventriculostomy until the opening pressure remains lower than 190 mm H2O. Repeated lumbar punctures have been associated with improved survival rates. Treatment with corticosteroids, acetazolamide, or mannitol has not proved efficacious. The development of immune reconstitution inflammatory syndrome (IRIS) can occur in patients with AIDS who are receiving effective ART that increases the CD4 count (Chapter 367). Symptoms of meningitis reappear and are due to the inflammatory response and not to a relapse of disease; IRIS is associated with an influx of CD4 lymphocytes and monocytes into the CSF.7 In one randomized controlled trial, HIV-positive patients with cryptococcal meningitis who had not previously received ART were randomized to initiate either earlier ART (1 to 2 weeks after meningitis diagnosis) or deferred ART (5 weeks after diagnosis). Although the incidence of recognized cryptococcal IRIS did not differ significantly between the earlier and deferred ART groups, deferring ART for 5 weeks after diagnosis was associated with significantly improved survival, especially among patients with a paucity of CSF white cells. A5  IRIS can also occur in transplant recipients in whom immunosuppressive therapy is decreased rapidly. Generally, no specific therapy is needed for mild IRIS, but sometimes corticosteroids are needed if increased intracranial pressure occurs and/or if there is evidence of cerebral edema.

Pulmonary and Other Nonmeningeal Infections

Treatment of nonmeningeal cryptococcosis depends on the severity of the infection. Many patients with isolated pulmonary or other focal infections are not severely ill, and oral fluconazole, 400 mg daily for 6 to 12 months, is recommended. For patients who are severely ill, therapy is the same as noted earlier for CNS infection.



PROGNOSIS

The outcome for both AIDS and non-AIDS patients with cryptococcal meningitis has improved markedly in the developed world.8 In Africa, however, the mortality from cryptococcal meningitis in AIDS patients approaches 100% in some areas because of lack of access to specific therapy. Among patients with HIV, those with a predominant TH1 cytokine profile in the CSF and blood at diagnosis appear to have a better survival.9 Dementia, which usually occurs in older patients, hearing loss, and visual loss may not be reversed even though mycologic cure is achieved. Fluconazole, 200 mg three times per week, is safe and effective as primary prophylaxis against cryptococcal disease in cryptococcal antigen-negative, HIV-infected adults with CD4 counts lower than 200 cells/µL, both before and during early antiretroviral treatment. A6 

  Grade A References A1. Nussbaum JC, Jackson A, Namarika D, et al. Combination flucytosine and high-dose fluconazole compared with fluconazole monotherapy for the treatment of cryptococcal meningitis: a randomized trial in Malawi. Clin Infect Dis. 2010;50:338-344. A2. Beardsley J, Wolbers M, Kibengo FM, et al. Adjunctive dexamethasone in HIV-associated cryptococcal meningitis. N Engl J Med. 2016;374:542-554. A3. Day JN, Chau TT, Wolbers M, et al. Combination antifungal theory for cryptococcal meningitis. N Engl J Med. 2013;368:1291-1302. A4. Molloy SF, Kanyama C, Heyderman RS, et al. Antifungal combinations for treatment of cryptococcal meningitis in Africa. N Engl J Med. 2018;378:1004-1017. A5. Boulware DR, Meya DB, Muzoora C, et al. Timing of antiretroviral therapy after diagnosis of cryptococcal meningitis. N Engl J Med. 2014;370:2487-2498. A6. Parkes-Ratanshi R, Wakeham K, Levin J, et al. Primary prophylaxis of cryptococcal disease with fluconazole in HIV-positive Ugandan adults: a double-blind randomised, placebo-controlled trial. Lancet Infect Dis. 2011;11:933-941.

  GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 317 Cryptococcosis  

GENERAL REFERENCES 1. Maziarz EK, Perfect JR. Cryptococcosis. Infect Dis Clin North Am. 2016;30:179-206. 2. Rajasingham R, Smith RM, Park BJ, et al. Global burden of disease of HIV-associated cryptococcal meningitis: an updated analysis. Lancet Infect Dis. 2017;17:873-881. 3. Lawrence DS, Boyer-Chammard T, Jarvis JN. Emerging concepts in HIV-associated cryptococcal meningitis. Curr Opin Infect Dis. 2019;32:16-23. 4. Setianingrum F, Rautemaa-Richardson R, Denning DW. Pulmonary cryptococcosis: a review of pathobiology and clinical aspects. Med Mycol. 2019;57:133-150. 5. Rajasingham R, Wake RM, Beyene T, et al. Cryptococcal meningitis diagnostics and screening in the era of point-of-care laboratory testing. J Clin Microbiol. 2019;57:1-8.

2048.e1

6. Gassiep I, Douglas J, Emeto TI, et al. Cryptococcal infections over a 15 year period at a tertiary facility & impact of guideline management. Mycoses. 2018;61:633-638. 7. Meya DB, Okurut S, Zziwa G, et al. Cellular immune activation in cerebrospinal fluid from Ugandans with cryptococcal meningitis and immune reconstitution inflammatory syndrome. J Infect Dis. 2015;211:1597-1606. 8. Pasquier E, Kunda J, De Beaudrap P, et al. Long-term mortality and disability in cryptococcal meningitis: a systematic literature review. Clin Infect Dis. 2018;66:1122-1132. 9. Mora DJ, Fortunato LR, Andrade-Silva LE, et al. Cytokine profiles at admission can be related to outcome in AIDS patients with cryptococcal meningitis. PLoS ONE. 2015;10:1-17.

2048.e2

CHAPTER 317 Cryptococcosis  

REVIEW QUESTIONS 1. A patient with cryptococcal meningitis had improvement of headache and nausea following the initial diagnostic lumbar puncture. Amphotericin B and flucytosine were begun immediately after the tap, but 2 days later, he has recurrence of these symptoms and also has some blurring of vision. What is the likely reason for the return of his symptoms? A . Reaction to amphotericin B B. Secondary bacterial infection related to the lumbar puncture C. Increased intracranial pressure D. Resistance of the organism to amphotericin B E. Resistance of the organism to flucytosine Answer: C  This is a typical picture in that the lumbar puncture takes off cerebrospinal fluid and relieves some of the pressure briefly. Increased intracranial pressure is most likely due to blockage of the arachnoid villi and/or increased brain edema and is thought to be secondary to the large capsule surrounding the yeast. Resistance to amphotericin B and flucytosine is very uncommon. The symptoms, especially blurring of vision, are unlikely due to amphotericin B toxicity. Secondary bacterial infection after a spinal tap is also rare. 2. What is the treatment for the condition noted in question 1? A . Perform repeated lumbar punctures to bring down the increased pressure B. Change therapy to fluconazole C. Continue amphotericin B but stop flucytosine D. Add ceftriaxone 2 g every 12 hours E. Symptomatic treatment only Answer: A  Treatment of increased intracranial pressure, which is seen commonly with cryptococcal meningitis, is to perform repeated spinal taps until the pressure remains low. There is no reason to change antifungal therapy or add an antibiotic. 3. Which of the following is the preferred treatment for an AIDS patient with cryptococcal meningitis? A . Fluconazole and flucytosine B. Voriconazole and amphotericin B C. Posaconazole and caspofungin D. Amphotericin B and flucytosine E. Fluconazole

Answer: D  Several randomized blind and open-label treatment trials have shown the benefit of combined therapy with amphotericin B and flucytosine for cryptococcal meningitis. Echinocandins have no role in treating this infection. The other suggested regimens are inferior to amphotericin B and flucytosine. 4. What is the major host defense against Cryptococcus species? A . Immunoglobulin A antibody B. Natural killer cells C. T lymphocytes D. Complement E. Neutrophils Answer: C  Patients who have deficient T-cell immunity, such as AIDS patients with low CD4 cell counts, transplant recipients, and patients on corticosteroids, are at high risk for developing infection with Cryptococcus. 5. The diagnosis of cryptococcal meningitis can be made in a few hours by performing which of the following? A . Culture of cerebrospinal fluid B. Magnetic resonance imaging C. Lateral flow dipstick test for cryptococcal antigen D. Cryptococcal immunoglobulin M antibody assay E. Gram stain of cerebrospinal fluid Answer: C  Culture is definitive but takes several days. Antibody assays for Cryptococcus neoformans are not helpful for diagnosis. Magnetic resonance imaging may show brain abscesses and meningeal enhancement, but these are not specific for cryptococcosis. The dipstick method for detecting antigen in cerebrospinal fluid is a rapid, sensitive, and specific test to make the diagnosis of cryptococcal meningitis. An India ink preparation (but not Gram stain) allows visualization of the budding yeast cells surrounded by the large capsule, but this test is not currently done by many laboratories.

2048

CHAPTER 318 Candidiasis  

Candidiasis is the most common opportunistic fungal infection as a result of both the organisms’ ubiquity and the increasing number of patients with risk factors for infection with these organisms.1 The classic immunosuppressed host at risk for serious Candida infections is a neutropenic patient with a hematologic malignancy who has received cytotoxic agents and corticosteroids. Increasingly, however, candidiasis is seen in patients in intensive care units (ICUs). Risk factors for the development of serious Candida infections in ICU patients include diabetes,2 the use of broad-spectrum antimicrobials, indwelling central venous catheters, previous surgical procedures, renal failure, parenteral nutrition, pancreatitis, any form of dialysis, and high Acute Physiology and Chronic Health Evaluation (APACHE) score. Certain ICU populations, especially very-low-birthweight neonates and burn victims, are at even higher risk for Candida infection than is the typical ICU patient. The primary manifestation of Candida infection in patients with HIV/AIDS is mucocutaneous infection, primarily oropharyngeal candidiasis. The development of mucosal Candida infection is related to deficient T-cell immunity as reflected by a low CD4 lymphocyte count. With appropriate antiretroviral therapy, oropharyngeal candidiasis has become an uncommon opportunistic infection.  

318  CANDIDIASIS CAROL A. KAUFFMAN AND PETER G. PAPPAS



DEFINITION

Candidiasis encompasses a wide variety of clinical syndromes caused by yeasts of the genus Candida. Of the species that cause infection in humans, Candida albicans is the most common; Candida glabrata, Candida parapsilosis, and Candida tropicalis are responsible for most of the remaining infections. Organisms such as Candida krusei, Candida lusitaniae, and Candida guilliermondii are less common causes of infection.

The Pathogen

Candida species are 2- to 6-µm yeastlike organisms that reproduce by budding. Most species, with the exception of C. glabrata, form pseudohyphae (elongated buds that remain attached to the mother cell) and hyphae in tissues. Candida species cause a wide spectrum of diseases ranging in severity from localized mucous membrane infection to life-threatening disseminated disease. The major determinant of the severity of infection is the host’s immune response. Local infections are often related to overgrowth of Candida as a result of changes in the normal microbiota. Invasive infections that remain within an organ system, such as urinary tract infections, usually occur because of local anatomic abnormalities. In an immunosuppressed host, especially a patient with neutropenia, widespread visceral dissemination is common.  

EPIDEMIOLOGY

Candida species reside normally in the gastrointestinal and genitourinary tracts and on the skin. As colonizers, Candida species do not cause infection unless there is a defect in host defense mechanisms or unless exogenous factors, such as antibiotic use or cutaneous disruption and/or maceration, have upset the ecology of the normal microbiota. C. albicans is the species most commonly found colonizing humans; C. glabrata is the second most common species, and C. tropicalis, C. parapsilosis, and others are found less often. The species of Candida colonizing and infecting patients has changed in recent decades in that C. glabrata, a species that is increasingly resistant to fluconazole, has become a prominent pathogen in many hospitals. In many parts of the world, the emergence of C. auris, a previously uncommon multidrug-resistant Candida species, has created new challenges in understanding the epidemiology, treatment, and prevention of this organism. Though uncommon, acquisition of Candida from environmental sources has been noted. The Candida species most often associated with transmission from contaminated fluids or devices, especially central intravenous catheters, has been C. parapsilosis.

PATHOBIOLOGY

The usual mode of infection with Candida is egress from its normal niche into the bloodstream or other tissues; the source is usually the gastrointestinal tract, but the skin and genitourinary tract are other sources. The primary host defense in response to this event is phagocytosis and killing by neutrophils, monocytes, and macrophages. C-C chemokine receptor 2 (CCR2)-expressing inflammatory monocytes and their tissue-resident derivatives play an essential antifungal role, particularly in the first 48 hours after Candida infection. Phagocytosis is enhanced in the presence of specific anti-Candida antibody and complement. Several different mechanisms are operative within neutrophils and macrophages that allow the killing of yeasts. Thus, patients who are leukopenic, especially those with chemotherapy-induced disruption of the gut mucosa, are at great risk for invasion with Candida species. Once Candida gains access to the bloodstream, widespread hematogenous dissemination is the rule. Biopsy of involved organs shows multiple microabscesses composed of neutrophils (in a host who has these cells), budding yeasts, and often pseudohyphae or hyphae. Over time, the lesions show a mixed neutrophilic and granulomatous response. T-cell immunity is an important host defense against infection with Candida at mucosal surfaces. In contrast to those with neutropenia, patients with deficient T-cell immunity are at risk for persistent and recurrent mucocutaneous candidiasis, but invasive infection rarely develops. The ubiquitin ligase, Casitas B lymphoma-b (CBLB), a T-cell activator and inducer of T-cell differentiation, is crucial for restraining the magnitude of the innate immune response against C. albicans infection but at the same time confers suboptimal protection to the host. CBLB therefore may be a potential drug target for systemic candidiasis.3  

CLINICAL MANIFESTATIONS

Mucocutaneous Candidiasis Oropharyngeal Candidiasis

Local mucous membrane and cutaneous lesions are the most common forms of Candida infection.4 Oropharyngeal candidiasis, or thrush (Chapter 397), can be due to either local factors or T-cell dysfunction. Local factors include the use of broad-spectrum antimicrobials or inhaled corticosteroids, xerostomia, and radiation treatment of the head and neck. Denture stomatitis occurs frequently in persons who wear full upper dentures, especially those who do not remove their dentures at night. Thrush secondary to T-cell dysfunction is most commonly seen in patients with HIV infection (Chapter 366) and is the most frequent opportunistic infection noted in patients with AIDS. The appearance of thrush in a previously healthy individual with no known risk factors should immediately raise suspicion of HIV infection. Thrush manifests with white plaques on the buccal mucosa, palate, oropharynx, or tongue (Fig. 318-1). Scraping the lesions with a tongue depressor reveals an erythematous, nonulcerated mucosa under the plaques. Denture stomatitis almost always manifests as a painful erythematous palate without plaques. Angular cheilitis, or perlàche, which is the presence of painful cracks at the corners of the mouth, can occur with or without thrush.

Esophagitis

Esophagitis5 may accompany oropharyngeal candidiasis or may occur independently of lesions in the oropharynx (Chapter 129). The development of

CHAPTER 318 Candidiasis  

ABSTRACT

Candida species reside in the gastrointestinal and genitourinary tracts and on the skin of humans. The species that is most commonly encountered as a cause of infection is Candida albicans, but in some medical centers, C. glabrata and C. parapsilosis also are major pathogens. Infections caused by Candida species vary from benign localized mucous membrane infections to focal invasive infections to disseminated infections with a high mortality rate. Examples of mucous membrane infections are oropharyngeal candidiasis, esophagitis, and vulvovaginitis. These are generally treated with oral fluconazole. Focal invasive infections include urinary tract infections, osteoarticular infections, meningitis, and endophthalmitis. Urinary tract infections can be treated with fluconazole in most patients, but endophthalmitis, meningitis, and osteoarticular infections are more complicated, must be treated with longer courses of therapy, and for some patients will require initial treatment with another agent, such as amphotericin B. Candidemia and invasive candidiasis should be treated initially with an echinocandin; transition to oral fluconazole after the patient has responded to initial intravenous therapy is reasonable if the Candida species causing the infection is susceptible.

KEYWORDS

Candida candidemia invasive candidiasis mucocutaneous candidiasis fluconazole echinocandin

2048.e3

CHAPTER 318 Candidiasis  

2049

FIGURE 318-1.  Thrush.

Candida esophagitis is almost always related to immune dysfunction and not simply to local factors. Candida esophagitis occurs in AIDS patients with low CD4 counts, patients with leukemia, and others taking immunosuppressive agents. Candida esophagitis occurs rarely in otherwise normal hosts. The classic symptom of Candida esophagitis is odynophagia localized to a discrete substernal area. The differential diagnosis includes ulcerations due to herpes simplex or cytomegalovirus and, in AIDS patients, idiopathic ulcers.

FIGURE 318-2.  Skin lesions in invasive candidiasis.

Vulvovaginitis

Candida vulvovaginitis is a common infection in women of childbearing age and is the most frequent mucocutaneous manifestation of Candida infection. Risk factors include conditions associated with increased estrogen levels, such as the use of oral contraceptives and pregnancy, diabetes mellitus, therapy with corticosteroids or broad-spectrum antimicrobials, and HIV infection. Symptoms include vaginal discomfort, discharge, and vulvar pruritus. The discharge is usually curdlike, but it can also be thin and watery. The labia are erythematous and swollen, and the vaginal walls show erythema and white plaques. Although most women have only a few episodes throughout their lives, a minority have frequent recurrences; in most of these patients, no discrete risk factor can be identified, and the cause is presumed to be local immune dysregulation.

Cutaneous Candidiasis

Candida infection of the skin (Chapter 412) occurs mostly in the intertriginous areas or under a large pannus or pendulous breasts. The lesions are erythematous, pruritic, and frequently pustular; have a distinct border; and are almost always associated with smaller satellite lesions, which helps distinguish candidiasis from tinea cruris or corporis. Candida onychomycosis results in thickened, opaque, and onycholytic nails. Candida can also cause paronychia, especially in those whose occupation involves frequent immersion of the hands in water.

Chronic Mucocutaneous Candidiasis

This uncommon syndrome usually begins in childhood and is characterized by recalcitrant and relapsing thrush, esophagitis, vaginitis, onychomycosis, and hyperkeratotic skin lesions on the face, scalp, and hands. Autosomal dominant chronic mucocutaneous candidiasis is associated with mutations in the CC domain of STAT1 leading to defective TH1 and TH17 responses. Some patients have associated autoimmune endocrinopathies, including hypoparathyroidism, hypothyroidism, and hypoadrenalism (autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy [APECED], which is caused by a loss-of-function mutation of the autoimmune regulator gene, AIRE, and in these patients autoantibodies against interleukin-17 (IL-17) and IL-22 are found. (See Autoimmune Polyglandular Syndrome Type 1 in Chapter 218.)

Disseminated Infections Candidemia

The most common manifestation of disseminated Candida infection is candidemia.6 However, candidemia merely implies the presence of Candida in blood; it does not define the extent of visceral involvement. Candida obtained from a blood culture should never be considered a contaminant and should always prompt a search for the probable source and the extent of infection. Risk factors for candidemia include broad-spectrum antimicrobial therapy,

FIGURE 318-3.  Retinal involvement.

central intravenous catheters, parenteral nutrition, renal failure, pancreatitis, any type of dialysis, surgical procedures involving the gastrointestinal tract, neutropenia, and corticosteroid therapy.7 The attributable mortality from candidemia approaches 40%. Mortality is higher in elderly patients and neonates and among those with severe sepsis and poor source control. Although candidemia is the most obvious manifestation of serious infection with Candida species, septic shock can occur, along with invasion of multiple viscera, in the absence of positive blood cultures. The clinical picture of invasive candidiasis is indistinguishable from that of bacterial infection. The characteristic histologic picture consists of multiple microabscesses in many organs. The eyes, kidneys, liver, spleen, and brain are the most commonly involved sites, but virtually all organs can be involved. Clinical clues to the diagnosis of invasive candidiasis include the appearance of skin and retinal lesions. The nonpainful, nonpruritic skin lesions are papular to pustular and surrounded by a zone of erythema (Fig. 318-2). Classical eye lesions appear as distinctive white exudates in the retina (Fig. 318-3); with extension into the vitreous body, the retina becomes obscured.

Endocarditis

Candida endocarditis is an uncommon and often fatal complication of candidemia. It occurs most often in intravenous drug users, patients who have prosthetic cardiac valves and intracardiac pacemaker devices, and those with central venous catheters in place. Blood cultures are usually persistently positive, and echocardiography reveals large vegetations that can readily embolize to major vessels.

2050

CHAPTER 318 Candidiasis  

develops as a late infection after previous episodes of bacterial peritonitis. A cloudy dialysate, abdominal pain, and fever are typically noted.

Meningitis

Acute Candida meningitis occurs as part of disseminated infection, especially in low-birthweight neonates. Chronic meningitis, an uncommon manifestation of candidiasis, resembles cryptococcal or tuberculous meningitis with regard to symptoms and cerebrospinal fluid findings.  

FIGURE 318-4.  Computed tomography scan of a patient with chronic disseminated candidiasis (hepatosplenic candidiasis). Note the distinctive punched-out lesions in the liver.

Chronic Disseminated (Hepatosplenic) Candidiasis

This syndrome almost always occurs in leukemic patients who have had an episode of neutropenia. After the neutrophil count returns to normal, fevers that are often quite high, right upper quadrant tenderness, and nausea develop. The alkaline phosphatase level is generally elevated, and distinctive punchedout lesions are seen in the liver, spleen, and sometimes the kidneys on computed tomography (Fig. 318-4). Biopsy of these lesions shows microabscesses that may contain budding yeasts.

Focal Invasive Infections

These forms of candidiasis result from local inoculation, contiguous spread, or hematogenous spread. Hematogenous spread,8 which often goes undetected, is probably the most common pathogenetic mechanism.

Urinary Tract Infections

Candiduria is a frequent finding in hospitalized patients and is related to factors such as diabetes mellitus, broad-spectrum antimicrobial treatment, indwelling urinary devices, and genitourinary tract structural abnormalities. Most patients with candiduria have only bladder colonization and not true infection. Urinary tract infection with Candida species can arise by two mechanisms. Patients with candidemia can develop multiple microabscesses secondary to hematogenous spread to the kidneys. Other patients, who have the risk factors noted earlier, can develop cystitis or ascending infection with pyelonephritis. Patients with cystitis or pyelonephritis have symptoms indistinguishable from those of bacterial infections. A fungus ball composed of fungal hyphae can develop at any level of the collecting system and lead to obstruction, with subsequent infection.

Osteoarticular Infections

Osteoarticular infections arise secondary to hematogenous seeding or exogenous inoculation during intra-articular injection, a surgical procedure, or trauma. Vertebral osteomyelitis is the most common manifestation of osteoarticular candidiasis. The symptoms of back pain and fever may occur many weeks after an episode of fungemia.

Endophthalmitis

Exogenous endophthalmitis occurs secondary to trauma or ophthalmic surgery. Most often, the procedure involved is cataract extraction, with or without lens implantation, and the most common infecting species is C. parapsilosis. Primary infection occurs in the anterior chamber, but ultimately the posterior chamber is also involved. Endogenous Candida endophthalmitis results from hematogenous seeding of the choroid and retina and is one of the most serious complications of candidemia. Characteristic white lesions are visible in the retina, and with progression of the infection, vitritis occurs; the risk for loss of vision is quite high.

Peritonitis

Candida peritonitis can follow bowel surgery or perforation. Symptoms are the same as those noted in bacterial peritonitis. Usually, this type of infection is polymicrobial, and abscess formation is common. In patients maintained on continuous ambulatory peritoneal dialysis, Candida peritonitis generally

DIAGNOSIS

The diagnosis of mucocutaneous candidiasis is often made clinically. Culture is rarely indicated. Confirmation can be sought by scraping the lesions and performing either a potassium hydroxide preparation or a Gram stain to look for budding yeasts (Chapter 407). In cases in which the disease is recurrent or unresponsive to standard therapy, lesions should be cultured to establish whether a more resistant species, such as C. glabrata or C. krusei, is the causative agent. In the event of suspected esophagitis, endoscopy shows plaquelike lesions or ulcerations, and biopsy shows mucosal invasion with budding yeasts and pseudohyphae. The diagnosis of invasive candidiasis is more difficult. Evidence of dissemination is usually sought by culturing blood or other sterile body sites. However, blood cultures are not sensitive enough for clinicians to rely on them to establish the diagnosis of invasive candidiasis in all cases or to rule out candidiasis as a diagnostic possibility. In addition, 1 to 4 days is required for growth to occur; in a desperately ill patient, this delay is problematic. The tips of intravenous catheters that have been removed should be sent for culture. However, no studies have evaluated the number of yeasts that is indicative of infection, and many physicians accept the growth of any yeast as affirming infection that requires treatment. Many focal forms of candidiasis are indistinguishable from bacterial infection, and biopsy should be performed for histopathologic and culture studies. In a seriously ill patient suspected of having candidiasis, the development of pustular skin lesions or typical retinal lesions can be helpful. Budding yeasts typical of Candida species should be sought by smearing material from a pustule on a slide and staining it with Gram stain or by performing a biopsy of a lesion and staining the tissue section with a silver stain. All patients who are candidemic or suspected of having disseminated Candida infection should undergo a dilated ophthalmologic examination, preferably by an ophthalmologist, to look for typical retinal lesions. Imaging studies are invaluable for certain forms of candidiasis, especially chronic disseminated candidiasis, and they can be of major help in defining the extent of infection in other types of Candida infection, such as osteoarticular and urinary tract infections and endocarditis. Non–culture-based techniques are increasingly used to aid in the diagnosis of invasive candidiasis,9 including β-d-glucan, a cell wall component of many fungi, including Candida species. The β-d-glucan assay is not specific for Candida infections, but has moderately good sensitivity in patients at high risk for invasive candidiasis. PCR is not standardized, but some studies show that it is more sensitive than β-d-glucan and blood cultures. A more rapid technique can identify Candida species from whole blood by use of magnetic biosensor technology.10

TREATMENT  Guidelines for treatment of the various forms of candidiasis have been published by the Infectious Diseases Society of America (IDSA).11 Mucocutaneous disease is obviously treated in a much different fashion than disseminated life-threatening infection. Because diagnostic tests are not sensitive, empirical therapy is indicated in some circumstances, and for patients at the highest risk for Candida infection, antifungal prophylaxis can decrease that risk (see below under “Prevention”).

Mucocutaneous Infections

Most mucocutaneous infections should initially be treated with local creams, solutions, troches, or suspensions.12 For thrush, clotrimazole troches (10 mg four or five times daily) are preferred to nystatin suspension (commonly given as “swish and swallow” four times daily). Patients with AIDS may not respond to local therapy, especially when their CD4 counts are low; in this situation, oral fluconazole 100 to 200 mg daily is given. For vaginitis, a variety of creams and vaginal tablets (miconazole, clotrimazole, and others) are effective, but many women prefer to take a single 150-mg fluconazole tablet orally. Recurrent vaginitis is a more complicated therapeutic issue and often requires chronic suppressive therapy with fluconazole.13 Esophagitis should always be treated

CHAPTER 318 Candidiasis  

with a systemically absorbed agent; the usual treatment is fluconazole 200 mg/ day for 14 days. In patients with advanced AIDS and low CD4 counts, who are often taking fluconazole to prevent recurrent candidiasis, fluconazole-refractory disease may develop. For these patients, increasing the dosage of fluconazole or switching to itraconazole suspension 200 mg daily, voriconazole 200 mg twice daily, or posaconazole suspension 400 mg daily should be effective. If oral tablets and solutions are no longer effective, intravenous amphotericin B, caspofungin, anidulafungin, and micafungin are alternative agents that can be used. Patients with the syndrome of chronic mucocutaneous candidiasis require lifelong suppressive therapy with oral azole agents.

Candidemia and Invasive Candidiasis

All patients with candidemia should be treated with an antifungal agent, including patients who have only one blood culture that yields Candida and those with a vascular catheter tip that yields Candida. The rationale for this recommendation is related to the high rate of metastatic foci in major organs associated with hematogenously disseminated candidiasis. Randomized controlled trials have shown the effectiveness of the following antifungal agents for the treatment of candidemia: fluconazole 400 or 800 mg/day; the three echinocandins—caspofungin 50 mg/day, anidulafungin 100 mg/day, and micafungin 100 mg/day A1 ; voriconazole 3 mg/kg twice daily; amphotericin B 0.7 mg/ kg/day; and a lipid formulation of amphotericin B 3 to 5 mg/kg/day. The IDSA guidelines recommend an echinocandin as initial therapy in both neutropenic and non-neutropenic patients. Fluconazole is an acceptable alternative, but only for patients who are not severely ill and who are considered unlikely to have an azole-resistant Candida species. Patients who have stabilized clinically and are found to have an isolate, such as C. albicans, that is likely to be susceptible to fluconazole can be transitioned to fluconazole from an echinocandin. Voriconazole can be used for step-down therapy if the organism is susceptible, but offers little advantage over fluconazole. Amphotericin B formulations are used infrequently, except for patients who are neutropenic and for neonates. Vascular catheters should be removed if at all feasible because removal has been shown to help clear Candida from blood more quickly. Repeated blood cultures should be obtained to ascertain that the fungemia has resolved, and treatment should continue for 2 weeks after the date of the first negative blood culture. An individual patient-level quantitative review of seven randomized trials for the treatment of invasive candidiasis found an overall mortality in the entire data set of 31.4%. Significant predictors of mortality included increasing age; use of immunosuppressive therapy; and infection with C. tropicalis. Improved survival and clinical success was found with the use of an echinocandin and the removal of central venous catheters. Because diagnostic tests are not sensitive, seriously ill patients who could have invasive candidiasis may need to be treated before culture confirmation. This approach is used frequently in neutropenic patients and is increasingly used in the ICU setting.14 Liposomal amphotericin B, caspofungin, and voriconazole have been shown in randomized clinical treatment trials to be effective for empirical use in neutropenic patients. A placebo-controlled, randomized trial of fluconazole empirical therapy in ICU patients failed to show a benefit; however, there were acknowledged problems with the chosen end point, and the rate of candidemia was too low to allow a proper evaluation of empirical therapy. The IDSA guidelines recommend that empirical therapy be reserved for febrile, critically ill patients who have risk factors for invasive candidiasis. The preferred treatment is an echinocandin. In a randomized clinical trial of 260 non-neutropenic, nontransplanted, critically ill patients with ICU-acquired sepsis, Candida colonization, and multiorgan failure, on broad-spectrum antibiotics, empirical micafungin (100 mg daily for 14 days) decreased the rate of new invasive fungal infection but did not increase fungal-free survival at day 28. A2  Compelling data for early treatment come from a study of 224 candidemic patients who had septic shock. Mortality rates as high as 98% were found in patients in whom there was a delay beyond 24 hours of the onset of shock in initiating antifungal therapy and in effecting source control, defined as draining abscesses and removing central venous catheters. Endocarditis should be treated with a lipid formulation of amphotericin B, with or without flucytosine. Echinocandins are an acceptable alternative. Infected valves should be replaced. In a few patients for whom valve replacement was not an option, lifelong suppression with fluconazole appeared to be effective. Chronic disseminated candidiasis generally requires months of therapy for cure. Most patients begin therapy with a lipid formulation of amphotericin B or an echinocandin and are then switched to fluconazole and treated until the lesions disappear on computed tomography scanning. This process of radiographic resolution may require several months.

Focal Invasive Infections

Treatment of focal infections depends on the organ system involved. Perhaps the simplest to treat are urinary tract infections. Most patients with candiduria are not infected but merely colonized; removing the selective pressure of antimicrobials and indwelling catheters eliminates candiduria in many of these patients. For those who have infection, oral fluconazole at a dosage of 200 mg/ day for 2 weeks is recommended. Bladder irrigation with amphotericin B should

2051

not be used because it eradicates only bladder colonization, requires that a catheter be placed in the bladder, and is associated with a high recurrence rate. None of the newer antifungal agents has a role in the treatment of urinary tract infections. Osteoarticular infections require months of therapy; a lipid formulation of amphotericin B or an echinocandin can be given initially, followed by long-term therapy with an azole. Peritonitis associated with chronic ambulatory peritoneal dialysis can be treated with amphotericin B, fluconazole, or an echinocandin, depending on the species of Candida causing infection. Intraperitoneal administration of amphotericin B can be extremely irritating and should not be attempted. The dialysis catheter should be removed. Meningitis should be treated initially with a lipid formulation of amphotericin B and flucytosine; patients with more chronic disease can be switched to fluconazole for a longer duration of therapy. Treatment of Candida eye infections varies with the extent of ocular involvement. Lesions discovered early at the stage of choroidal or retinal involvement perhaps can be treated effectively with systemic antifungal agents (amphotericin B, an echinocandin, fluconazole, or voriconazole) alone. Many experts prefer to use an agent, such as voriconazole or fluconazole, that achieves higher concentrations in the eye. Lesions extending into the vitreous require more aggressive therapy. The best results have been obtained with pars plana vitrectomy, injection of amphotericin B or voriconazole into the vitreous; and a systemic antifungal agent such as fluconazole or voriconazole for several weeks. Management must be individualized and performed in concert with an ophthalmologist experienced in the treatment of this infection. Treatment of endophthalmitis associated with an intraocular lens implant requires removal of the implant, vitrectomy, and local amphotericin B injections, as well as therapy with fluconazole or voriconazole.



PREVENTION

For certain populations at the highest risk for invasive fungal infection, prophylactic antifungal agents can prevent infection. The populations for whom prophylaxis is recommended include stem cell transplant recipients, patients with acute leukemia who are undergoing induction chemotherapy, high-risk liver transplant recipients, and pancreas and small bowel transplant recipients; in these groups, a variety of different agents are effective. In the ICU population, prophylaxis with fluconazole can be effective, but it is recommended only in units that have a high rate of invasive candidiasis, and only in those patients at the highest risk for infection. In a placebo-controlled trial of caspofungin as antifungal prophylaxis in adults who were in the ICU for at least 3 days, were ventilated, received antibiotics, had a central line, and had at least one additional risk factor, caspofungin was safe and tended to reduce the incidence of invasive candidiasis when used for prophylaxis, but the difference was not statistically significant. A3  Restricting the use of prophylaxis to individuals at higher risk for candidiasis is essential to prevent the widespread use of azoles, with subsequent selection of resistant species. An experimental vaccine has shown promise in reducing recurrent symptomatic vulvovaginal candidiasis. A4   

PROGNOSIS

The prognosis for patients with mucocutaneous infections is excellent. The prognosis for focal invasive infections depends on the organ involved and the patient’s immune status. For example, whereas pyelonephritis may respond well to antifungal therapy, endocarditis and meningitis are more difficult to treat and have poor outcomes. Invasive candidiasis has a high mortality rate. Early treatment with an effective antifungal agent is extremely important for a favorable outcome.

  Grade A References A1. Kullberg BJ, Vasquez J, Mootsikapun P, et al. Efficacy of anidulafungin in 539 patients with invasive candidiasis: a patient-level pooled analysis of six clinical trials. J Antimicrob Chemother. 2017;72: 2368-2377. A2. Timsit JF, Azoulay E, Schwebel C, et al. Empirical micafungin treatment and survival without invasive fungal infection in adults with ICU-acquired sepsis, Candida colonization, and multiple organ failure. The EMPIRICUS randomized clinical trial. JAMA. 2016;316:1555-1564. A3. Ostrosky-Zeichner L, Shoham S, Vasquez J, et al. MSG-01: a randomized, double-blind, placebocontrolled trial of caspofungin prophylaxis followed by preemptive therapy for invasive candidiasis in high-risk adults in the critical care setting. Clin Infect Dis. 2014;58:1219-1226. A4. Edwards JE Jr, Schwartz MM, Schmidt CS, et al. A fungal immunotherapeutic vaccine (NDV-3A) for treatment of recurrent vulvovaginal candidiasis-a phase 2 randomized, double-blind, placebocontrolled trial. Clin Infect Dis. 2018;66:1928-1936.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 318 Candidiasis  

GENERAL REFERENCES 1. Strollo S, Lionkia MS, Adjemian J, et al. Epidemiology of hospitalizations associated with invasive candidiasis, United States, 2002-2012. Emerg Infect Dis. 2017;23:7-13. 2. Rodrigues CF, Rodrigues ME, Henriques M. Candida sp. infections in patients with diabetes mellitus. J Clin Med. 2019;8:1-41. 3. Xiao Y, Tang J, Guo H, et al. Targeting CBLB as a potential therapeutic approach for disseminated candidiasis. Nat Med. 2016;22:906-914. 4. Hellstein JW, Marek CL. Candidiasis: red and white manifestations in the oral cavity. Head Neck Pathol. 2019;13:25-32. 5. Hoversten P, Otaki F, Katzka DA. Course of esophageal candidiasis and outcomes of patients at a single center. Clin Gastroenterol Hepatol. 2019;17:200-202. 6. Colombo AL, de Almeida Junior JN, Slavin MA, et al. Candida and invasive mould diseases in nonneutropenic critically ill patients and patients with haematological cancer. Lancet Infect Dis. 2017;17:e344-e356. 7. Pfaller MA, Castanheira M. Nosocomial candidiasis: antifungal stewardship and the importance of rapid diagnosis. Med Mycol. 2016;54:1-22.

2052.e1

8. Kullberg BJ, Arendrup MC. Invasive candidiasis. N Engl J Med. 2015;373:1445-1456. 9. Clancy CJ, Pappas PG, Vazquez J, et al. Detecting infections rapidly and easily for candidemia trial, part 2 (DIRECT2): a prospective, multicenter study of the T2Candida panel. Clin Infect Dis. 2018;66:1678-1686. 10. Mylonakis E, Clancy CJ, Ostrosky-Zeichner L, et al. T2 magnetic resonance assay for the diagnosis of candidemia in whole blood: a clinical trial. Clin Infect Dis. 2015;60:892-899. 11. Pappas PG, Kauffman CA, Andes DR, et al. Executive summary: clinical practice guideline for the management of candidiasis: 2016 update by the Infectious Diseases Society of America. Clin Infect Dis. 2016;62:409-417. 12. Eschenauer GA, Nguyen MH, Clancy CJ. Is fluconazole or an echinocandin the agent of choice for candidemia. Ann Pharmacother. 2015;49:1068-1074. 13. Tang Y, Yu F, Huang L, et al. The changes of antifungal susceptibilities caused by the phenotypic switching of Candida species in 229 patients with vulvovaginal candidiasis. J Clin Lab Anal. 2019;33:1-6. 14. Vergidis P, Clancy CJ, Shields RK, et al. Intra-abdominal candidiasis: the importance of early source control and antifungal treatment. PLoS ONE. 2016;11:1-14.

2052.e2

CHAPTER 318 Candidiasis  

REVIEW QUESTIONS 1. Candida glabrata infections are more difficult to treat for which of the following reasons? A . C. glabrata is more virulent. B. C. glabrata cannot be phagocytized by neutrophils. C. C. glabrata is resistant to azole antifungals. D. C. glabrata readily forms hyphae and invades tissues. E. C. glabrata has a large polysaccharide capsule. Answer: C  C. glabrata infections are difficult to treat because of both inherent and acquired resistance to fluconazole and other azoles. It is not more virulent and may actually be less so. It does not form hyphae in contrast to the other species of Candida. Phagocytosis of C. glabrata occurs just as it does with other Candida species. There is no capsule. 2. An AIDS patient comes in because he has retrosternal pain when he swallows. He does not have thrush, and there is no cervical lymphadenopathy. What is the most likely diagnosis? A . Gastroesophageal reflux disease B. Pericarditis C. Pulmonary embolus D. Esophageal candidiasis E. Angina Answer: D  Odynophagia is the classic symptom shown by AIDS patients who have Candida esophagitis. The other diseases generally do not cause pain when swallowing food. Esophagitis can occur without thrush, although frequently both are present concomitantly. Response to treatment with fluconazole can be used as a diagnostic test for Candida esophagitis; endoscopy can be done, also, but is usually reserved for patients who do not respond to fluconazole therapy. 3. Which is not a risk factor for candidemia? A . High APACHE II score B. Central venous catheter C. Deficient T-cell immunity D. Prior surgical procedure E. Neutropenia Answer: C  The risk factors for candidemia include everything noted above except deficient T-cell immunity. T cells are important in protection against mucosal infection with Candida species but do not play a role in prevention of invasive candidiasis and candidemia.

4. A 70-year-old man who is in the intensive care unit after having abdominal surgery for a perforated bowel becomes hypotensive and febrile. He has a central catheter in place, is receiving parenteral nutrition and broad-spectrum antibiotics, and has kidney failure requiring renal replacement therapy. What is the most appropriate approach to treatment after you obtain blood cultures? A . Treat with fluconazole when the laboratory tells you that yeasts are growing in the cultures. B. Treat with an echinocandin after the laboratory identifies the yeast in the blood cultures as C. glabrata. C. Treat with amphotericin B immediately after obtaining the blood cultures. D. Treat with fluconazole after the laboratory identifies the yeast in the blood culture as C. glabrata. E. Treat with an echinocandin immediately after obtaining the blood cultures. Answer: E  In a patient at great risk for candidemia, as is this patient (who is on renal replacement therapy, has a central venous catheter plus another central line for dialysis, is receiving broad-spectrum antibiotics and parenteral nutrition, has a perforated bowel, and has undergone recent abdominal surgery), empirical treatment with an antifungal agent should be started as soon as the blood cultures and any other pertinent cultures have been obtained. Amphotericin B is rarely used in this setting, and echinocandins or fluconazole are generally used. In a patient who is not stable and is hypotensive, an echinocandin is preferred. Waiting for the cultures to grow, which will take 1 to 3 days, and for identification of species, which may add another 2 to 3 days, increases the mortality rate enormously in candidemic patients. 5. Which finding is helpful in directing the clinician to a diagnosis of invasive candidiasis? A . Nonpainful papular skin lesions on an erythematous base on the chest and left arm B. Painful erythematous nodules developing on the shins C. Bilateral spreading erythema over the lower legs and feet D. Molluscum-type lesions on the forehead accompanied by fever E. Pruritic blisters appearing in crops over the trunk Answer: A  Candidemia and invasive candidiasis can present with the sudden appearance of nonpainful, nonpruritic papules that evolve to become pustules on an erythematous base. These can occur anywhere on the body. All of the other lesions do not occur with candidemia.

2052

CHAPTER 319 Aspergillosis  

319  ASPERGILLOSIS THOMAS J. WALSH



DEFINITION

Aspergillosis is a disease caused by one or more of the species of the genus Aspergillus. Sporelike structures called conidia are aerosolized from the mold form of the organism growing in the environment. When conidia reach tissue, they germinate to form invasive filaments called hyphae.

The Pathogens

The most common species infecting humans are Aspergillus fumigatus, Aspergillus flatus, Aspergillus terreus, and A. niger. The species are usually identified in culture by characteristic microscopic features of hyphae and the structures producing conidia. When some species are not readily identifiable, they may be reported by the clinical laboratory as “Aspergillus species” or “Aspergillus sp.” A. fumigatus may be reported as “A. fumigatus species complex.” Some species within A. fumigatus complex may be particularly drug resistant. A. terreus is resistant to amphotericin B. Aspergilli within tissue appear as dichotomously branched (Y-shaped) septate hyphae. Scedosporium and Fusarium species also may produce septate hyphae in tissue. The presence of septa and dichotomous branching differentiates Aspergillus species from the Mucorales, which are the causative organisms of mucormycosis (Chapter 320).  

EPIDEMIOLOGY

Aspergillus species are ubiquitous organisms in the external environment, including soil, decaying matter, and air in temperatures as high as 40 to 50° C. Aspergilli are easily isolated from houses, particularly from basements, crawl spaces, bedding, humidifiers, ventilation ducts, potted plants, dust, condiments (e.g., pepper), and marijuana samples. Aspergilli cause abortion in cattle and are important pathogens of marine organisms, insects, and domesticated and wild birds. Aflatoxin, which is one of the most potent carcinogens known, is produced by strains of Aspergillus flavus at ambient temperature on stored grain, spices, and nuts. Foodborne ingestion of preformed aflatoxin may cause hepatic necrosis or hepatocellular carcinoma (Chapter 186) in animals and humans. Aspergillus species may be acquired from airborne conidia in inpatient and outpatient health care settings. Nosocomial aspergillosis is associated with building renovation, new construction, unfiltered air, contaminated ventilation systems, and fireproofing materials. Hospital water, which may become aerosolized during activities such as showering, is a more recently described potential source of aspergilli. As human pathogens, Aspergillus species may cause acute invasive disease, chronic infection, or allergic symptoms.1,2 A classification of aspergillosis is presented in Table 319-1. Acute invasive aspergillosis develops in immunocompromised patient populations, particularly those with hematologic malignancies,3 hematopoietic stem cell transplantation (HSCT), severe aplastic anemia, primary immunodeficiencies, and solid organ transplantation, especially of heart, lung, and liver. Genetic deficiency of the soluble pattern-recognition receptor called PTX3 (long pentraxin 3) caused by homozygous haplotype (h2/h2) in the PTX3 gene of donor cells has been found to lead to impaired neutrophil antifungal capacity and increased risk for invasive aspergillosis in recipients of HSCT. Persistent neutropenia, corticosteroids, other immunosuppressive agents, graft-versus-host disease (GVHD), and cytomegalovirus (CMV) disease are the most frequently observed clinical risk factors. The mortality of acute invasive aspergillosis varies from as much as 100% with central nervous system (CNS) infection to approximately 65% with pulmonary infection in HSCT recipients. Early recognition of clinical manifestations followed by initiation of antifungal therapy may improve the ominous prognosis of acute invasive aspergillosis.

Invasive aspergillosis complicates the care of up to 13% of patients with immunocompromise. For the period of 2009 to 2013 in the United States, using propensity score matching, invasive aspergillosis was associated with increased hospital mortality and 30-day readmission rates, as well as excess duration of hospitalization and costs (attributable excess costs of up to $600 million annually).4  

CLINICAL MANIFESTATIONS

Invasive Aspergillosis

The classic clinical manifestations of invasive pulmonary aspergillosis in immunocompromised hosts are fever and focal pulmonary infiltrates, nodules, or wedge-shaped densities resembling infarcts.5 Cough, pleuritic pain, and hemoptysis also may be present. Focal pulmonary infiltrates may progress to a cavity on recovery from neutropenia. Pulmonary infiltrates may also present as bronchopneumonia in an immunosuppressed patient. The pulmonary pathology in all these entities is that of hemorrhagic infarction caused by the organism’s capacity to invade blood vessel walls (angioinvasion). These processes lead to formation of a necrotic center surrounded by a ring of hemorrhage and edema, which correlates with a “halo sign” surrounding the nodular density. Concomitant pleural effusion may develop and represent Aspergillus empyema. Tracheobronchial aspergillosis in immunocompromised patients presents as ulcerative, pseudomembranous, or plaquelike large airway disease that may presage pulmonary parenchymal invasion. Acute Aspergillus sinusitis may occur concomitantly or independently of invasive pulmonary aspergillosis. Although symptoms may include fever, localized pressure, and pain, they may be absent in severely immunocompromised patients. Eschar on the nasal septum and turbinates may be observed by speculum examination or endoscopy. Acute Aspergillus sinusitis of the ethmoid and sphenoid sinuses may progress to cavernous sinus thrombosis with symptoms referable to cranial nerves III, IV, V1,2, and VI. A. flavus has a high propensity for causing acute sinus infection. The tissue targets of extrapulmonary and disseminated aspergillosis most commonly include the CNS, where abscesses and infarcts are characteristic. Patients with CNS aspergillosis present with focal paresis, cranial nerve deficits, and seizures. The glucose level in cerebrospinal fluid (CSF) is usually normal, and cultures of CSF are negative. Other extrapulmonary manifestations include endophthalmitis, myocardial infarction, gastrointestinal disease, renal infarction, cutaneous lesions, and Budd-Chiari syndrome. Esophageal ulcers and mesenteric thrombosis may produce gastrointestinal bleeding. Renal infection may present as flank pain and hematuria. Aspergillus endocarditis6 usually begins as an isolated infection in intravenous drug users or after cardiac valvular surgery. Aspergillus endocarditis most commonly presents as major arterial emboli. Blood cultures, which are seldom

TABLE 319-1 CLASSIFICATION OF ASPERGILLOSIS CATEGORY

SPECIFIC FORMS OF ASPERGILLOSIS

Acute invasive aspergillosis

Invasive pulmonary aspergillosis  Empyema   Tracheobronchial infection Extrapulmonary aspergillosis   Acute sinusitis   Focal rhinitis   Cerebral, cerebellar, or brain stem infarction  Endophthalmitis  Osteomyelitis   Epidural abscess   Cardiac aspergillosis   Myocarditis   Endocarditis   Pericarditis   Gastrointestinal aspergillosis   Renal infection   Cutaneous lesions (nodules, ulcers) Disseminated aspergillosis

Chronic aspergillosis

Aspergilloma Chronic necrotizing pulmonary aspergillosis Chronic cavitary pulmonary aspergillosis Aspergillus otomycosis

Allergic forms of aspergillosis

Allergic bronchopulmonary aspergillosis Extrinsic allergic alveolitis Allergic Aspergillus sinusitis

CHAPTER 319 Aspergillosis  

ABSTRACT

Aspergillosis propagates by airborne conidia and invades tissue by hyphae. Aspergillus fumigatus is the most common species. Aspergillus species are found ubiquitously in nature and in buildings. Aspergillus species cause acute invasive diseases of the lungs and sinuses from which dissemination to other tissues may occur in immunocompromised patients. They also cause chronic pulmonary infections and allergic sinopulmonary diseases. Acute invasive pulmonary aspergillosis is diagnosed in immunocompromised patients by a combination of symptoms, diagnostic imaging features (nodules, halo signs, cavities), and microbiologic features, including bronchoalveolar lavage. Disseminated infection is best diagnosed by direct biopsy or needle aspirate. Chronic forms of aspergillosis include aspergilloma, chronic necrotizing aspergillosis, and chronic cavitary aspergillosis. Allergic bronchopulmonary aspergillosis may develop in patients with asthma and in those with cystic fibrosis. Treatment of invasive aspergillosis relies on a triad of antifungal therapy, reversal of immunosuppression, and, where appropriate, surgical resection of infected tissue. Voriconazole is recommended as the preferred agent by the Infectious Diseases Society of America Guidelines Committee, whereas isavuconazole and liposomal amphotericin B are considered acceptable alternatives of comparable efficacy.

KEYWORDS

Aspergillus galactomannan β-d-glucan voriconazole isavuconazole posaconazole amphotericin invasive aspergillosis allergic bronchopulmonary aspergillosis

2052.e3

CHAPTER 319 Aspergillosis  

FIGURE 319-1.  Tomogram of pulmonary aspergillomas (arrows).

positive, may be delayed in growth by as much as 14 to 21 days. Diagnosis is difficult, and despite valve replacement with antifungal therapy, mortality approaches 100%. Aspergillus pericarditis may arise from contiguous pulmonary lesions or through transmural infection from endocardial infection. Locally invasive aspergillosis usually develops in immunocompromised patients as cutaneous ulcers, focal rhinitis, osteomyelitis, and septic arthritis.7 Cutaneous ulcers have been associated with use of contaminated adhesive tape and arm boards. Blood-borne infection in illicit intravenous drug users may present as foci of dissemination in brain, lung, kidney, and bone. Keratitis, endophthalmitis, and infection of burn wounds may develop from traumatic inoculation in otherwise immunocompetent patients.

eosinophilia, positive sputum cultures, mucous plugs containing hyphae, the presence of grossly visible brown flecks in sputum (hyphae), transient infiltrates and parallel “tramline” or ring markings on chest radiographs, proximal bronchiectasis, upper lobe contraction, and elevated levels of total immunoglobulins G and E (IgG and IgE). Eosinophilia may be present in blood, sputum, and lung tissue. Mucous plugs contribute to development of pulmonary infiltrates, atelectasis, and peribronchial inflammation. The parallel or ring markings are caused by thickened ectatic bronchi, whereas the upper lobe changes are due to progressive apical fibrosis. Pulmonary infiltrates in allergic bronchopulmonary aspergillosis may be nonsegmental and transient in association with eosinophilia and asthma; alternatively, they may be segmental and associated with bronchial obstruction by mucous plugs, wherein asthma and eosinophilia may be absent. Extrinsic allergic alveolitis is an unusual allergic form of Aspergillus lung disease that has been most frequently associated with Aspergillus clavatus in malt workers. A hypersensitivity pneumonitis with dyspnea and fever develops approximately 4 hours after exposure. Diffuse reticulonodular interstitial infiltrates may be present at the time of symptoms. Patients have IgG precipitins and cell-mediated immune reactions against Aspergillus antigens. Granulomas are present in lung tissue. In contrast to allergic bronchopulmonary aspergillosis, eosinophilia is not a feature of Aspergillus extrinsic allergic alveolitis. Allergic Aspergillus sinusitis is a noninvasive form of sinus disease that typically presents in patients with asthma, nasal polyps, sinus opacification, and eosinophilia. Sinus aspirate yields mucinous material containing eosinophils, Charcot-Leyden crystals (eosinophil-derived rhomboid crystals of lysophospholipase), and hyphal elements. Allergic Aspergillus sinusitis and allergic bronchopulmonary aspergillosis may coexist in some patients. Advanced forms of allergic Aspergillus sinusitis may present with proptosis and optic neuropathy, necessitating prompt surgical intervention.  

Chronic Pulmonary Aspergillosis

Aspergilloma appears on chest radiograph as a ball in a cavity. The fungus ball consists of matted hyphae and debris in a preformed cavity from previous pulmonary tuberculosis, histoplasmosis, or fibrocystic sarcoidosis (Fig. 319-1). Symptomatic patients present with cough, hemoptysis, dyspnea, weight loss, fatigue, chest pain, or fever. Sputum culture is typically positive for Aspergillus species, particularly A. niger. Pleural aspergillosis may complicate surgical resection of aspergilloma or develop spontaneously as a bronchopleural fistula or concomitantly with tuberculosis. As a stage in the repair process of infarcted lung tissue in neutropenic patients, one or more apparent “aspergillomas” may develop in consolidated lesions during recovery from neutropenia. These apparent aspergillomas do not develop in preexisting cavities and create an “air-crescent sign,” or Monod sign, during their formation. Chronic necrotizing pulmonary aspergillosis and chronic cavitary pulmonary aspergillosis occur in patients with underlying chronic lung disease, chronic immunosuppression, such as that due to prolonged use of systemic corticosteroids, or both. Chronic necrotizing pulmonary aspergillosis characteristically causes a slowly progressive inflammatory destruction of lung tissue superimposed on chronic lung disease. Clinical manifestations of worsening pulmonary function, cough, and dyspnea in chronic necrotizing pulmonary aspergillosis may be indistinguishable from concomitant primary chronic respiratory disease. Chronic cavitary pulmonary aspergillosis is defined as the presence of multiple Aspergillus-related cavities, which may or may not contain an aspergilloma. Patients with chronic cavitary pulmonary aspergillosis may have genetically mediated deficits in innate host defenses. Occurring in association with symptoms of cough, hemoptysis, and dyspnea, the progressive cavities of chronic cavitary pulmonary aspergillosis tend to coalesce with the loss of functional lung tissue. Aspergillus otomycosis is a chronic infection that usually involves the external auditory canal with symptoms of pain, pruritus, hypoacusis, and otic discharge in patients with impaired mucocutaneous immunity, such as those with chronic eczema, hypogammaglobulinemia, diabetes mellitus, or HIV infection and those receiving corticosteroids. Aspergillus may involve the middle ear and extend into the mastoid sinus if the tympanic membrane has been perforated.

Allergic Forms of Aspergillosis

Allergic bronchopulmonary aspergillosis8,9 develops most frequently in patients with a history of chronic asthma or cystic fibrosis. Occurring in genetically susceptible patients exposed to specific Aspergillus antigens, allergic bronchopulmonary aspergillosis is characterized by episodic airway obstruction, fever,

2053

DIAGNOSIS

Invasive Aspergillosis

Diagnosis of invasive pulmonary aspergillosis and disseminated aspergillosis is difficult. None of the aforementioned clinical manifestations are diagnostic for invasive aspergillosis. Advances in computed tomography (CT) have revealed characteristic features of nodules, halo signs, wedge-shaped infiltrates, and air-crescent signs during invasive pulmonary aspergillosis in immunocompromised patients (Fig. 319-2).10 For example, infarct-shaped consolidations and smooth bronchial wall thickening are more frequent in invasive pulmonary aspergillosis, whereas mass-shaped consolidations and centrilobular nodules (1 µg/mL) and for isavuconazole in setting of drug interactions or apparent failure

Limited or no improvement (limited evidence) • • • •

Consider G-CSF or GM-CSF Consider IFN-γ (50 µg/m2/3 × weekly) Consider WBC transfusions if neutropenic Consider hyperbaric oxygen for localised disease (e.g., sinusitis)

GM-CSF = granulocyte-macrophage-colony stimulating factor; IFN = interferon; G-CSF = granulocyte-colony stimulating factor

FIGURE 320-2.  Diagram of the management approach to patients with suspected mucormycosis. *200 mg of isavuconazole = 372 mg of isavuconazonium sulfate.

CHAPTER 320 Mucormycosis  

Morphologic identification of Mucorales requires their cultivation to examine reproductive fruiting structures. Most of these fungi grow rapidly on most fungal media (e.g., Sabouraud dextrose agar) when incubated at 25° to 30° C. These fungi are sensitive to the protein inhibitor cycloheximide, and addition of this agent to fungal media may not ensure optimal recovery. Morphologic features alone, especially when assessed by individuals with expertise in fungal identification, can provide a high level of accuracy comparable to that of molecular methods. Data on the antifungal susceptibility of Mucorales species are limited, and minimal inhibitory concentration (MIC) testing is rarely available outside

2059.e1

research or research laboratories. The MIC end points for these rapidly growing fungi are inconsistent, not standardized, and at times difficult to interpret. Because interpretive MIC break points for Mucorales have yet to be defined, the correlation between clinical responses and MIC values is uncertain. Mucorales are resistant to many antifungals, including flucytosine, ketoconazole, fluconazole, voriconazole, and the echinocandins. Also, they have variable susceptibility to itraconazole. Amphotericin B and posaconazole, a new triazole, are the most active agents in vitro, although their activities differ among different Mucorales families. The activity of antifungal combinations against these fungi has yet to be proved in vivo.

Among open-label studies and retrospective surveys evaluating posaconazole suspension as salvage therapy (800 mg/day) in patients with refractory mucormycosis, the agent showed a response rate approaching 70%. Furthermore, posaconazole has been well tolerated. Determining whether posaconazole alone or combined with a lipid formulation of amphotericin B or other agent (e.g., deferasirox) is of value requires further study. Posaconazole has limitations because absorption of the oral suspension is suboptimal in patients with mucositis, severe diarrhea, acid suppression therapy, or poor oral intake. Absorption of oral posaconazole is maximized when administered with high-fat foods in separate doses (four times daily). Finally, steady-state plasma concentrations of posaconazole are not reached until around 1 week of therapy. The new formula of posaconazole (posaconazole tablets 300 mg/daily) has not been studied adequately in mucormycosis. In a recent single-arm open-label trial of 37 patients with mucormycosis, the new triazole isavuconazole (loading dose 372 mg q 8 hrs for 6 doses IV/oral; followed by 372 mg oral or IV 12 to 24 hrs after last dose) was shown to have similar efficacy compared with amphotericin B in a matched case-control analysis.19 The duration of antifungal therapy should be determined on an individual basis. Near-normalization of radiographic imaging, negative follow-up biopsy specimens, and cultures from the affected site, as well as recovery from immunosuppression, are important indicators for stopping antifungal therapy.

Surgery

Surgical débridement of cutaneous lesions is crucial and must be done without delay because of the aggressively invasive nature of mucormycosis. A coordinated effort among all subspecialties involved (surgery, infectious diseases, head and neck, ophthalmology, pathology, clinical microbiology, and plastic surgery) is crucial, and the internist can play a vital role coordinating it. Repeated removal of necrotic tissue or aggressive surgical measures such as enucleation of the eye may be required for control of the infection. Decisions regarding the extent of débridement are often made at the bedside. A CT scan or MRI before surgery and intraoperative frozen section analysis help determine the extent of tissue and tissue margin involvement. Low platelet counts, as may be seen in patients with underlying hematologic malignancies, must be corrected with transfusions before surgical intervention. Unfortunately, bleeding problems can limit surgical options. Surgery in conjunction with systemic antifungal therapy has been shown to significantly improve survival rates.

Management of Comorbidity and Adjunct Treatments

Adjunct measures have been proposed to improve host immunity and tissue viability, as well as to impede fungal proliferation. Rapid correction of underlying conditions, such as control of hyperglycemia, reversal of ketoacidosis,20 rapid tapering of glucocorticoid therapy, and discontinuation of deferoxamine treatment, can influence outcomes. Hyperbaric oxygen is a beneficial adjunct therapy for mucormycosis, particularly in diabetic patients with rhinocerebral disease. Specifically, the increased oxygen pressure achieved seems to improve neutrophil activity and oxidative killing by polyene antifungals. Also, high concentrations of oxygen can inhibit growth of the organism in vitro and improve the rate of wound healing by increasing the release of tissue growth factors. However, this treatment has not been studied vigorously to determine efficacy and cannot be routinely recommended. Investigators have proposed several immune augmentation strategies as adjunct therapy, including administration of cytokines (e.g., granulocyte colony-stimulating factor [G-CSF], interferon). In refractory neutropenic patients, granulocyte transfusion may be beneficial until granulocyte recovery. These adjunct measures, although promising, are yet to be studied sufficiently. Finally, the new iron chelator deferasirox has been considered as an adjunct antifungal agent based on preclinical studies and very limited human experience with patients with refractory mucormycosis. Results of the small randomized, double-blinded DEFEAT Mucor trial were published in 2012. A1  Twenty patients with proven or probable mucormycosis were randomized to treatment with liposomal amphotericin B plus deferasirox (20 mg/kg per day for 14 days) or liposomal amphotericin B plus placebo. Although reported adverse events were similar between the two study groups, significantly higher mortality rates were found in patients randomized to receive deferasirox at 30 days (45% vs. 11%) and 90 days (82% vs. 22%, P = .01). However, patients in the deferasirox arm were more likely than patients in the placebo arm to have active malignancy, neutropenia, and/or corticosteroid therapy, and less likely to have received additional antifungals, making the results of this pilot trial less conclusive. Nevertheless, currently available data do not support a role for initial deferasirox therapy for mucormycosis. Further knowledge of the unique virulence attributes of Mucorales based on genomic analysis might aid the development of novel therapeutic targets.

  Grade A Reference A1. Spellberg B, Ibrahim AS, Chin-Hong PV, et al. The Deferasirox-AmBisome Therapy for Mucormycosis (DEFEAT Mucor) study: a randomized, double-blinded, placebo-controlled trial. J Antimicrob Chemother. 2012;67:715-722.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 320 Mucormycosis  

GENERAL REFERENCES 1. Farmakiotis D, Kontoyiannis DP. Mucormycoses. Infect Dis Clin North Am. 2016;30:143-163. 2. Song Y, Qiao J, Giovanni G, et al. Mucormycosis in renal transplant recipients: review of 174 reported cases. BMC Infect Dis. 2017;17:1-6. 3. Jeong W, Keighley C, Wolfe R, et al. The epidemiology and clinical manifestations of mucormycosis: a systematic review and meta-analysis of case reports. Clin Microbiol Infect. 2019;25:26-34. 4. Baldin C, Ibrahim AS. Molecular mechanisms of mucormycosis—the bitter and the sweet. PLoS Pathog. 2017;13:1-9. 5. Ibrahim AS, Voelz K. The mucormycete-host interface. Curr Opin Microbiol. 2017;40:40-45. 6. Petrikkos G, Tsioutis C. Recent advances in the pathogenesis of mucormycoses. Clin Ther. 2018;40:894-902. 7. Manesh A, Rupali P, Sullivan MO, et al. Mucormycosis—a clinicoepidemiological review of cases over 10 years. Mycoses. 2019;62:391-398. 8. Nezafati S, Kazemi A, Asgari K, et al. Rhinocerebral mucormycosis, risk factors and the type of oral manifestations in patients referred to a university hospital in Tabriz, Iran. Mycoses. 2018;61: 764-769. 9. Kontoyiannis DP, Yang H, Song J, et al. Prevalence, clinical and economic burden of mucormycosisrelated hospitalizations in the United States: a retrospective study. BMC Infect Dis. 2016;16:1-6. 10. He R, Hu C, Tang Y, et al. Report of 12 cases with tracheobronchial mucormycosis and a review. Clin Respir J. 2018;12:1651-1660. 11. Castrejón-Pérez AD, Welsh EC, Miranda I, et al. Cutaneous mucormycosis. An Bras Dermatol. 2017;92:304-311.

2060.e1

12. Kaur H, Ghosh A, Rudramurthy SM, et al. Gastrointestinal mucormycosis in apparently immunocompetent hosts—a review. Mycoses. 2018;61:898-908. 13. Kerezoudis P, Watts CR, Bydon M, et al. Diagnosis and treatment of isolated cerebral mucormycosis: patient-level data meta-analysis and Mayo Clinic experience. World Neurosurg. 2019;123:425-434. 14. Skiada A, Lass-Floerl C, Klimko N, et al. Challenges in the diagnosis and treatment of mucormycosis. Med Mycol. 2018;56(suppl 1):93-101. 15. Dadwal SS, Kontoyiannis DP. Recent advances in the molecular diagnosis of mucormycosis. Expert Rev Mol Diagn. 2018;18:845-854. 16. Tissot F, Agrawal S, Pagano L, et al. ECIL-6 guidelines for the treatment of invasive candidiasis, aspergillosis and mucormycosis in leukemia and hematopoietic stem cell transplant patients. Haematologica. 2017;102:433-444. 17. Lanternier F, Poiree S, Elie C, et al. Prospective pilot study of high-dose (10 mg/kg/day) liposomal amphotericin B (L-AMB) for the initial treatment of mucormycosis. J Antimicrob Chemother. 2015;70:3116-3123. 18. Kyvernitakis A, Torres HA, Jiang Y, et al. Initial use of combination treatment does not impact survival of 106 patients with haematologic malignancies and mucormycosis: a propensity score analysis. Clin Microbiol Infect. 2016;22:811.e1-811.e8. 19. Marty FM, Ostrosky-Zeichner L, Cornely OA, et al. Isavuconazole treatment for mucormycosis: a single-arm open-label trial and case-control analysis. Lancet Infect Dis. 2016;16:828-837. 20. Gebremariam T, Lin L, Liu M, et al. Bicarbonate correction of ketoacidosis alters host-pathogen interactions and alleviates mucormycosis. J Clin Invest. 2016;126:2280-2294.

2060.e2

CHAPTER 320 Mucormycosis  

REVIEW QUESTIONS 1. What is the best way to diagnose mucormycosis? A . Serum galactomannan B. Serum β-d-glucan C. Anti-Mucorales antibodies D. Polymerase chain reaction (PCR) E. Tissue for biopsy and culture Answer: E  Unfortunately, there are no biomarkers to diagnose mucormycosis, and this is a major unmet need for the management of the disease. Specifically, galactomannan and β-d-glucan do not detect Mucorales. PCR, although promising, remains investigational. There are no data about the value of antiMucorales antibodies for diagnosis. Tissue for histopathologic analysis and culture remains the only method for documenting invasive disease. 2. All of these conditions are common risk factors for mucormycosis except: A . Iron overload B. Poorly controlled diabetes C. Crowding conditions D. Severe trauma of soft tissues E. Chronic severe immune suppression Answer: C  Classic risk factors for the disease include iron overload that results in increased availability of host iron to support invasive fungal growth, significant and chronic immunosuppression such as severe cytopenia, systemic corticosteroids, crushing soft tissue injury (e.g., following severe trauma), and poorly controlled diabetes, especially in the setting of acidosis. Mucorales are extremely rare causes of catheter related infections not known to be associated with crowding conditions in the absence of trauma. (See Epidemiology.)

3. What is the most common Mucorales species causing mucormycosis? A . Mucor spp B. Cunninghamella bertholletiae C. Rhizomucor spp D. Rhizopus spp E. Apophysomyces spp Answer: D  Although there is notable geographic and host-related variability of infecting Mucorales, on a global scale Rhizopus species are the most common cause of mucormycosis, accounting for nearly 50% of such cases. (See Epidemiology.) 4. Which of the following azoles has the most potent activity against Mucorales? A . Fluconazole B. Posaconazole C. Voriconazole D. Itraconazole E. Ketoconazole Answer: B  Mucorales are innately resistant to most antifungal drugs, including azoles. Specifically, the only agents that have activity are amphotericin B and the triazoles posaconazole and isavuconazole.

2060

CHAPTER 321  PNEUMOCYSTIS PNEUMONIA  

321  PNEUMOCYSTIS PNEUMONIA JOSEPH A. KOVACS



DEFINITION

Pneumocystis jirovecii is a fungus that causes pneumonia almost exclusively in immunodeficient patients. Pneumocystis pneumonia (PCP) has been the most common life-threatening opportunistic infection in persons with HIV/AIDS. Although the frequency of PCP has decreased in HIV-infected patients first in association with the widespread use of anti-Pneumocystis prophylaxis and later with the introduction of effective combination antiretroviral therapy for HIV/AIDS, it continues to be seen with regularity in HIV-infected and other immunodeficient patients.1 Over the past two decades, there has been a marked increase in outbreaks of PCP among renal transplant recipients, especially in Europe and Australia.

The Pathogen

Pneumocystis is an ascomycete fungus of the subphylum Taphrinomycotina, most closely related to Schizosaccharomyces, Taphrina, and Saitoella species of fungi. Molecular studies including whole genome sequencing have demonstrated that the genus Pneumocystis includes a group of closely related organisms that are unique species, each of which can infect only a single host species.2 This has led to the application of species names to the group of organisms previously called Pneumocystis carinii, which now is reserved for a species that infects rats. The organism infecting humans has been renamed P. jirovecii. Despite the name change, the acronym PCP (for Pneumocystis pneumonia) continues to be used to designate the disease in humans. Pneumocystis species show a strict host specificity: attempts to transmit, for example, rat or human Pneumocystis to mice have been unsuccessful. Molecular evolutionary studies suggest that Pneumocystis species coevolved with their hosts, with rat and mouse Pneumocystis diverging an estimated 33 million years ago. Studies of Pneumocystis have been substantially hampered by an inability to grow any species in culture for a sustained period. Thus, the life cycle of Pneumocystis is unknown, although putative life cycles based on morphologic studies have been proposed; there is substantial evidence supporting a sexual phase in the life cycle. There are two easily recognized forms of the organism: trophic forms (≈2 to 6 µm in diameter) and cysts (also called asci; ≈6 to 8 µm in diameter), which can contain up to eight intracystic bodies (ascospores); additional intermediate forms are also seen. Trophic forms, which have an amorphous shape, are estimated to outnumber cysts, which are spherical, by approximately 10 : 1 in an infected lung. The genome of P. jirovecii, approximately 8.2 million base pairs, is markedly contracted compared with other fungi and has lost multiple metabolic pathways, including enzymes needed to synthesize all amino acids de novo, enzymes needed for chitin synthesis and degradation, and carbonic anhydrase, an enzyme important for regulation of intracellular pH.3 These findings suggest that that Pneumocystis lives exclusively in the mammalian host and cannot survive in the external environment. Pneumocystis has evolved mechanisms to avoid host innate and adaptive immune responses. β-Glucans are masked by surface proteins, and enzymes needed to synthesize outer chain mannans have been lost; both can be recognized by host pattern recognition receptors. The most abundant surface protein of Pneumocystis, the major surface glycoprotein, is found on both cysts and trophic forms and is encoded by a multicopy gene family, only one of which is apparently expressed in a given organism; this provides Pneumocystis with the potential for antigenic variability. Although to date only a single Pneumocystis species has been found that infects humans, molecular typing techniques have demonstrated a high level of diversity among human Pneumocystis isolates.

CHAPTER 321  PNEUMOCYSTIS PNEUMONIA  

ABSTRACT

Pneumocystis jirovecii is a fungus that causes pneumonia (PCP) almost exclusively in immunodeficient patients, especially those with HIV infection, renal transplant recipients, and patients receiving intensive immunosuppressive regimens for cancer or inflammatory diseases. Patients with PCP typically present with fever, a nonproductive cough, and progressive shortness of breath. Although chest radiographs may be normal early in the disease process, they will often show bilateral interstitial infiltrates that progress to a diffuse alveolar pattern if untreated. Because Pneumocystis cannot be cultured, diagnosis is based on detection of the organism by colorimetric or immunofluorescent stains, or by polymerase chain reaction (PCR) testing, in respiratory samples, especially bronchoalveolar lavage fluid and induced sputum. Trimethoprimsulfamethoxazole is the treatment of choice for PCP. Alternatives include dapsone plus trimethoprim, clindamycin plus primaquine, atovaquone, and pentamidine. Adjunctive corticosteroids improve survival in HIV-infected patients with hypoxia (Pao2 60 min May switch to oral therapy following clinical improvement No parenteral formulation is available Begin as early as possible and within 72 hr; efficacy if started later has not been demonstrated Use if parenteral therapy is necessary

Note: HIV-infected patients should receive 21 days of therapy; non-HIV patients should receive at least 14 days of therapy. G6PD = glucose-6-phosphate dehydrogenase.

The treatment of choice for PCP or extrapulmonary Pneumocystis disease, regardless of severity, is trimethoprim-sulfamethoxazole, which combines inhibitors of two enzymes in the folate synthetic pathway of Pneumocystis: sulfamethoxazole, an inhibitor of dihydropteroate synthase (DHPS), and trimethoprim, an inhibitor of dihydrofolate reductase (DHFR).20 Trimethoprim-sulfamethoxazole is available in both oral and intravenous formulations. Oral therapy should be reserved for patients with mild to moderate disease in whom poor absorption is not a concern. Outpatient therapy should be reserved for patients with mild to moderate disease who will reliably return for follow-up. The high incidence of PCP during the early years of the AIDS epidemic led to the identification of a number of new agents with anti-Pneumocystis activity, and these were extensively evaluated in randomized controlled trials, primarily in HIV-infected patients. In patients with mild to moderate disease (A-a O2 gradient 1 hour) infusion is usually well tolerated. Toxicities associated with pentamidine, which occur in about 50 to 60% of patients and frequently result in discontinuation of the drug, include nephrotoxicity, hypoglycemia, hyperglycemia, fever, neutropenia, thrombocytopenia, hypotension, hyperkalemia, transaminase elevations, and pancreatitis. Hypoglycemia may be life-threatening and may precede the development of hyperglycemia; hyperglycemia may be irreversible. Torsades de pointes (Chapter 59) has also rarely been reported.

Adjunctive Corticosteroid Therapy

Initiation of specific anti-Pneumocystis therapy is associated with deterioration in oxygenation after approximately 3 to 4 days; this likely results from a host inflammatory response to organisms damaged by therapy. Randomized controlled trials have demonstrated that the early addition of corticosteroids to specific anti-Pneumocystis therapy in HIV-infected patients can prevent this deterioration and improve survival, without a significant increase in opportunistic complications other than localized herpes simplex infection. A1  In the largest such study, corticosteroid therapy was associated with a 50% decrease in respiratory failure and mortality; this benefit was limited to patients with moderate to severe disease (E-Fig. 321-1). For HIV-infected patients, corticosteroids and specific anti-Pneumocystis therapy should be started at the same time. The addition of corticosteroids after 72 hours has shown no benefit, although it is reasonable to add them if patients exhibit deterioration after this time. Although the optimal regimen has not been defined by controlled trials, the tapering regimen from the largest study is most commonly used (see Table 321-1). The optimal utilization of corticosteroids in non-HIV patients who are often receiving them as part of the treatment regimens for their underlying disease is less clear because data from randomized controlled trials are not available;

CHAPTER 321  PNEUMOCYSTIS PNEUMONIA  

Cumulative risk for death

Cumulative risk for respiratory failure

0.5 0.4 0.3 0.2 0.1 0.0 0

8

16 Days

24

32

0.5 0.4 0.3 0.2 0.1 0.0 0

16 Days

24

32

E-FIGURE 321-1.  Cumulative risks for an unfavorable outcome during a randomized trial of corticosteroid therapy in patients with HIV infection and Pneumocystis pneumonia measured over a period of 31 days. The risk for respiratory failure (left) was 0.14 in the corticosteroid group (blue square) and 0.30 in the standard-treatment group (green X) (P = .004). The risks for death (right) were 0.11 and 0.23, respectively (P = .009). Corticosteroid therapy was associated with a significant decrease in both outcomes. (Modified from Bozzette SA, Sattler FR, Chiu J, et al. A controlled trial of early adjunctive treatment with corticosteroids for Pneumocystis carinii pneumonia in the acquired immunodeficiency syndrome. N Engl J Med. 1990;323:1451-1457. Copyright 1990, Massachusetts Medical Society. All rights reserved.)

2065.e1

2066

CHAPTER 321  PNEUMOCYSTIS PNEUMONIA  

dosing may need to be individualized to balance the immunosuppressive effects that potentially contributed to the development of PCP against the anti-inflammatory effects that may ameliorate life-threatening pulmonary dysfunction.21 One retrospective analysis of 31 patients suggested that increasing corticosteroids to a prednisone equivalent of 60 mg/day or more was associated with clinical benefit. However, a subsequent retrospective cohort study of 323 hospitalized patients with documented PCP without HIV concluded that the addition of early corticosteroids was not associated with improved respiratory outcomes.22 At this point it would not be unreasonable to administer corticosteroids to non-HIV patients with moderate or severe disease if they were not receiving corticosteroids, using the same regimen as for HIV-infected patients; for patients already taking corticosteroids at lower doses, the dosage could be increased to those levels.

Initiation of Antiretroviral Therapy

Given that many patients with HIV infection who are diagnosed with PCP are not receiving antiretroviral therapy, an important issue is how soon to start cART after a diagnosis of PCP. Retrospective and prospective studies have suggested that, in general, it is safe to initiate cART while patients are being treated for PCP, and early cART may be associated with an improved outcome. In a randomized 282-patient trial that examined early versus late initiation of cART in patients with acute opportunistic infections, 63% of whom had PCP, the early initiation arm (cART started a median of 12 days after starting therapy for the opportunistic infection) had a decreased rate of AIDS progression or death. Similar findings for tuberculosis in developing countries have emphasized the benefit of early initiation of cART. Major concerns about initiating cART include the risk for adverse drug reactions, which may be confused with adverse reactions to anti-PCP therapy; the risk for overlapping toxicities, which may complicate management; and the risk for immune reconstitution (Chapter 367), which has been rarely reported but can be life-threatening. Thus, many clinicians initiate cART during or immediately after completion of anti-Pneumocystis therapy, assuming the patient has shown clinical improvement, is able to tolerate oral medications, and accepts the commitment to lifelong therapy. However, the parameters for such an approach are difficult to define precisely. Patients who start cART early should be closely monitored for a recurrence of symptoms that may represent immune reconstitution.

Treatment Failure

The optimal approach to the management of patients who are failing therapy has not been well defined. In patients with progressive respiratory deterioration, it is critical that the diagnosis of PCP be confirmed rather than presumptive and that other concurrent processes (e.g., other infections, congestive heart failure, pulmonary emboli) have been ruled out; bronchoscopy should be considered to facilitate these determinations. Parenteral therapy should be used to eliminate absorption concerns, and corticosteroid medications should be added if this has not already been done. Because patients who will ultimately respond can show clinical deterioration at 3 to 4 days, as noted earlier, it is reasonable to wait 5 to 8 days before considering a change in drug therapy. Only trimethoprim-sulfamethoxazole and pentamidine are available in parenteral formulations. Parenteral clindamycin is available, but primaquine is available only as a tablet. No randomized trials have examined the relative efficacy of these agents in patients who are failing therapy. For patients who have not received trimethoprim-sulfamethoxazole, this should be the first choice as an alternative agent, assuming the patient did not have a life-threatening adverse reaction previously. Rapid desensitization (similar to penicillin desensitization), ideally in consultation with an allergy specialist, can be considered in patients with prior adverse reactions; however, patients with a history of Stevens-Johnson syndrome or toxic epidermal necrolysis should not be rechallenged. Retrospective cohort studies and meta-analyses have found that clindamycin-primaquine is superior to pentamidine in patients failing a first-line regimen, but there are potential biases in such analyses (e.g., severity of illness or ability to take oral medications may have affected the choice of salvage regimen), although some studies attempted to correct for this. There are no data to recommend switching to an alternative agent rather than adding an alternative agent (if toxicity is not an issue); both approaches have been used.

Resistance

Although Pneumocystis cannot be cultured, molecular studies have identified mutations in genes that are the targets of anti-Pneumocystis therapy, and these mutations appear to represent the development of resistance by Pneumocystis to these agents. The best-characterized mutations have been identified in the DHPS gene of Pneumocystis, which is the target of sulfamethoxazole and dapsone. Two mutations at the active site of this enzyme, which can occur either individually or together, have been identified with greater frequency in patients receiving trimethoprim-sulfamethoxazole or dapsone for prophylaxis; in vitro studies suggest that these mutations confer resistance. The clinical relevance of these mutations remains uncertain; some studies have found worse outcomes in patients with these mutations, but others have found no such association. Most patients in whom these mutations were identified retrospectively were

successfully treated with sulfa-containing drugs. In contrast to DHPS, there are very limited reports suggesting that the DHFR gene of Pneumocystis, which is the target of trimethoprim and pyrimethamine, has developed potential drugresistant mutations. Atovaquone presumably binds to the mitochondrial bc1 complex of Pneumocystis and thus inhibits electron transport. Multiple mutations have been identified in the cytochrome B gene of Pneumocystis, which presumably represent resistance in patients receiving atovaquone for prophylaxis; these mutations have not, however, been associated with clinical outcome. Because the presence of these mutations has not been definitively associated with worsening prognosis, clinical decisions should not be based on their identification. Methods for identifying DHPS mutations are not readily available in the United States but are available in Europe through a commercial PCR kit23; their detection should remain a research tool until their clinical relevance can be better defined.



PREVENTION

Although Pneumocystis is transmitted by the airborne route, exposure to the organism appears to be ubiquitous in humans, suggesting that avoidance of exposure may be difficult. Currently, respiratory isolation of patients with active PCP is not required, although it is reasonable to avoid having a susceptible patient share a room with a PCP patient. Recent outbreaks in renal and liver transplant patients strongly suggest a common source of infection; a better understanding of the patterns of transmission in these settings may lead to improved guidelines for preventing the spread of infection. It is noteworthy that in these outbreaks, broad institution of anti-Pneumocystis prophylaxis was the intervention that terminated the outbreaks. A major advance in the management of patients at risk for the development of PCP was the demonstration that trimethoprim-sulfamethoxazole was highly effective in preventing the disease in a susceptible pediatric population. Subsequent studies, primarily in HIV-infected patients, demonstrated that additional drug regimens were also effective. This has led to the broad use of antiPneumocystis prophylaxis in a wide range of susceptible populations. Two important issues in administering prophylaxis are identifying populations at risk and defining the period of risk during which prophylaxis should be provided. AIDS patients are at especially high risk; before the use of prophylaxis or cART, the lifetime incidence of PCP in this population was estimated at 60 to 80%. The most recent CD4 count is a validated surrogate marker for HIV-infected patients: patients with CD4 counts below 200 cells/μL in the absence of cART are at substantially increased risk for developing PCP, and primary prophylaxis is recommended for this group. Although 10 to 15% of patients who develop PCP have higher CD4 counts, the incidence is very low in this population, given the large number of patients who fall in this category. Patients with CD4 counts greater than 200 cells/μL but a CD4 percentage of less than 14% or a history of an AIDS-defining illness are also candidates for prophylaxis. In addition, patients who develop PCP should be placed on prophylaxis after the treatment regimen is successfully completed (secondary prophylaxis). For pediatric patients with HIV infection, in whom the normal CD4 count changes with age, guidelines are based on current age. Prophylaxis is recommended for children older than 6 years with CD4 counts below 200 cells/μL or 15%; for children between 1 and less than 6 years old with CD4 counts below 500 cells/μL or 15%; and for all children younger than 12 months.24 Before the availability of cART, when patients with HIV infection initiated prophylaxis, they were committed to continuing it for life because immunologic decline was irreversible. With cART, however, control of HIV replication leads to an increase in the CD4 count, which is associated with a concomitant decrease in the risk for developing PCP. Multiple studies have shown that when the CD4 count has been above 200 cells/μL for at least 3 months (ideally in the setting of controlled HIV replication), prophylaxis can be safely discontinued because the risk for developing PCP is no greater than in patients whose CD4 counts never fell below 200 cells/μL. In most of these studies, the median CD4 count was greater than 300 cells/μL, and HIV viral loads were below detection limits in the majority of patients. Recent observational studies have suggested that prophylaxis can also be safely discontinued in patients with CD4 counts between 100 and 200 cells/μL who have virologically suppressed HIV, but specific criteria for discontinuation (e.g., duration of viral suppression) were not defined in these studies. A reasonable approach would be to discontinue primary or secondary prophylaxis in such patients if the viral load remains below detection limits for 3 to 6 months. In patients with CD4 counts less than 100 cells/μL, prophylaxis should be maintained regardless of the viral load.

CHAPTER 321  PNEUMOCYSTIS PNEUMONIA  

2067

underlying disease or therapy (e.g., patients with leukemia or lymphoma receiving intensive conditioning or immunosuppressive therapy). Cancer patients meeting the following conditions should receive prophylaxis as indicated: acute lymphocytic leukemia, throughout therapy; therapy with alemtuzumab, for 2 months after completion and until CD4 is greater than 200 cells/μL; therapy with idelalisib with or without rituximab, or with prolonged corticosteroids, or temozolomide combined with radiation therapy, at least through active therapy; therapy with purine analogues or other T-cell-depleting regimens, consider until CD4 is greater than 200 cells/μL. For solid organ transplant patients, prophylaxis has not been universally adopted; it has been used primarily in centers with a known incidence greater than 3%. Guidelines recommend the administration of prophylaxis for 6 to 12 months in most solid organ transplant patients, with longer periods, up to lifelong, in heart, lung, liver, and intestine transplant recipients. As noted earlier, a number of outbreaks of PCP have recently been reported in renal transplant patients, and many developed disease more than 1 year after transplantation. Given this, some authors are suggesting that lifelong prophylaxis may be needed in this population.28 Risk factors identified in casecontrol studies have included older age, recent or concurrent cytomegalovirus infection, and treatment for rejection. Specific immunosuppressive drugs, such as mycophenolate mofetil and cyclosporine, have not been consistently implicated. Rare outbreaks have also been reported in liver transplant patients. For patients with inflammatory bowel disease (Chapter 132), who appear to be at increased risk as newer immunosuppressive agents are used, data are limited; however, consensus-based guidelines recommend prophylaxis for patients receiving triple immunomodulators that include either a calcineurin inhibitor or an anti-TNF agent. No consensus has been reached for less intensive regimens. For patients with connective tissue disorders or vasculitis, there are currently no consensus guidelines. Corticosteroid therapy (Chapter 32) is a well-described risk factor in non– HIV-infected patients, with approximately 90% of patients receiving such therapy before developing PCP in some studies. Higher dose and longer duration increase the risk. Not all patients who receive corticosteroids are at risk, however; for instance, asthmatic patients receiving corticosteroid therapy are at low risk. Although there are no consensus guidelines on the use of prophylaxis for patients receiving corticosteroids, one reasonable approach is to provide prophylaxis to patients with an underlying immunosuppressive or inflammatory disease who receive at least 20 mg of prednisone or equivalent for longer than 1 month. Other immunosuppressive agents (Chapter 32), such as calcineurin inhibitors, sirolimus, TNF antagonists, and rituximab, also appear to increase the risk for developing PCP, primarily in the patient populations noted earlier. Trimethoprim-sulfamethoxazole is the first-line agent for prophylaxis in all populations (Table 321-2). Alternatives include dapsone alone or combined with pyrimethamine plus leucovorin, atovaquone, and aerosol pentamidine administered by the Respirgard II nebulizer. In a randomized trial of 843 HIV-infected patients comparing trimethoprim-sulfamethoxazole with dapsone and aerosol pentamidine, no significant differences were seen on an intent-to-treat basis, but

Among non–HIV-infected patients, the CD4 count is not routinely measured, and it has not been shown to have the same predictive value for the development of PCP as in HIV-infected patients; however, CD4 counts below 200 cells/μL do appear to increase their susceptibility. In a meta-analysis of PCP prophylaxis in a variety of types of non-HIV immunocompromised patients, trimethoprim-sulfamethoxazole prophylaxis was found to be overall highly effective, A2  with one half-strength dose per day appearing to be as efficacious and less toxic than higher doses. A3  Recommendations for PCP prophylaxis in these populations are based on clinical parameters, including empirical identification of periods of risk and estimation of levels of immunosuppression.25 Very broad prophylaxis has not been implemented because of the side effects associated with these regimens; for example, there is concern that trimethoprim-sulfamethoxazole can cause bone marrow suppression that would interfere with engraftment or cause nephrotoxicity that would damage a transplanted kidney. Risk factors for non–HIV-infected patients include underlying disease, older age, use of immunosuppressive drugs, radiation therapy, graft-versus-host disease, and concomitant cytomegalovirus infection (Chapter 352). Patients with malignancies, especially hematologic malignancies, but increasingly solid tumors as well, are at risk for PCP primarily owing to the therapies they receive; the incidence can range from 1 to 43% in the absence of prophylaxis and is highly dependent on the intensity and duration of immunosuppression. In the absence of prophylaxis, the risk for developing PCP in transplant patients, whether hematopoietic stem cell transplantation (HSCT) (Chapter 168) or solid organ transplantation (Chapter 43), is reportedly about 5 to 15%, although lung and heart-lung transplant patients appear to have a higher incidence (up to 43%). Among patients with collagen vascular disease, the reported risk is less than 2% without prophylaxis, although patients with granulomatosis with polyangiitis (formerly Wegener granulomatosis) (Chapter 254) reportedly have a risk up to 12%, presumably because of the use of more immunosuppressive treatment regimens. In patients with inflammatory bowel disease (Chapter 132), the incidence in one large retrospective cohort study was about 1% per year, with a greater risk for Crohn disease compared with ulcerative colitis. The current risk for developing PCP in non-HIV populations is difficult to quantify because of the widespread use of prophylaxis and because immunosuppressive regimens are evolving. In a recent cohort study of HSCT patients, the risk for PCP in the setting of widely used prophylaxis was 0.28 to 0.63%, with approximately 1 4 of cases occurring greater than 270 days after transplantation.26 To facilitate the management of prophylaxis in at-risk, non-HIV populations, a number of guidelines have been developed by expert panels that have made recommendations based on the strength of the available data.27 For allogeneic stem cell transplant recipients, prophylaxis is recommended from the time of engraftment to at least 6 months after transplantation, and longer for patients who continue receiving immunosuppressive therapy or who have chronic graft-versus-host disease. For autologous stem cell transplant recipients, who have a lower risk for PCP, prophylaxis for 3 to 6 months should be considered if the degree of immunosuppression is substantial owing to

TABLE 321-2 DRUG REGIMENS FOR PREVENTION OF PNEUMOCYSTIS PNEUMONIA (PCP) INDICATION

DRUG

ROUTE

DOSE

COMMENTS

Preferred

Trimethoprim-sulfamethoxazole (TMP-SMX)

PO

1 double-strength tablet (160 mg TMP + 800 mg SMX) or one single-strength tablet (80 mg TMP + 400 mg SMX) qd

Also active in preventing toxoplasmosis

Alternative

TMP-SMX

PO

1 double-strength tablet (160 mg TMP + 800 mg SMX) three times weekly

Also active in preventing toxoplasmosis

Alternative

Dapsone

PO

100 mg qd or 50 mg bid

Test for G6PD deficiency before use

Alternative

Dapsone plus

PO

50 mg qd

Also active in preventing toxoplasmosis

Pyrimethamine plus Leucovorin

PO PO

50 mg once weekly 25 mg once weekly

Alternative

Atovaquone

PO

1500 mg qd with food

Likely active in preventing toxoplasmosis; efavirenz can decrease concentrations

Alternative

Pentamidine Pentamidine

Aerosol IV

300 mg via the Respirgard II nebulizer once monthly 4 mg/kg administered over >1 hr

Not active in preventing toxoplasmosis Limited observational data primarily in pediatric populations; not active in preventing toxoplasmosis

Should be administered with pyrimethamine to minimize toxicity

Note: Patients receiving pyrimethamine-sulfadiazine and atovaquone therapy for toxoplasmosis do not appear to need additional prophylaxis for PCP; patients receiving clindamycin-pyrimethamine therapy for toxoplasmosis will need additional prophylaxis for PCP. G6PD = glucose-6-phosphate dehydrogenase.

the lowest failure rates were seen while patients were receiving trimethoprimsulfamethoxazole. Conversely, trimethoprim-sulfamethoxazole was superior to aerosol pentamidine in another randomized study. In other large randomized trials in HIV-infected patients, the following regimens showed similar efficacy: atovaquone suspension and dapsone, atovaquone suspension and aerosol pentamidine, and dapsone-pyrimethamine and aerosol pentamidine. No randomized trials of these regimens have been conducted in non–HIVinfected populations, but clinical experience suggests that they are effective in these populations as well. Observational studies primarily in non-HIV pediatric populations suggest that monthly intravenous pentamidine is also effective. Patients receiving pyrimethamine plus sulfadiazine plus leucovorin for treatment of toxoplasmosis do not require additional anti-Pneumocystis prophylaxis because this regimen also prevents PCP. Although the combination of sulfadoxine and pyrimethamine is also efficacious, it is contraindicated in patients with sulfonamide allergies. Moreover, because Stevens-Johnson syndrome and other potentially life-threatening cutaneous reactions are more common with this combination than with trimethoprim-sulfamethoxazole, and because its long half-life results in slow clearance after the drug is discontinued, sulfadoxine plus pyrimethamine should probably not be used in sulfa-tolerant patients if trimethoprim-sulfamethoxazole is available. HIV-infected patients with a prior history of a mild sulfa allergy (e.g., mild rash, excluding those with prior Stevens-Johnson syndrome or toxic epidermal necrolysis) can often be safely rechallenged with trimethoprim-sulfamethoxazole. Randomized trials have demonstrated that dose escalation over a 6- to 13-day period is associated with better tolerance than direct rechallenge with full-dose trimethoprim-sulfamethoxazole, and that up to 75% of patients can continue to receive trimethoprim-sulfamethoxazole for at least 6 months.  

PROGNOSIS

Mortality for untreated PCP approaches 100%. With therapy, the survival rate for HIV-infected patients with confirmed PCP is now as high as 95%, but a poorer survival rate of 75% has been reported in patients without HIV infection. Risk factors for death in HIV-infected patients include more severe hypoxia, older age, recurrent episodes of PCP, low hemoglobin, and the presence of comorbid conditions. Although mortality for patients admitted to an intensive care unit is high, survival for HIV-infected patients has improved in recent years, now approaching 75%.

  Grade A References A1. Ewald H, Raatz H, Boscacci R, et al. Adjunctive corticosteroids for Pneumocystis jiroveci pneumonia in patients with HIV infection. Cochrane Database Syst Rev. 2015;4:CD006150. A2. Stern A, Green H, Paul M, et al. Prophylaxis for Pneumocystis pneumonia (PCP) in non-HIV immunocompromised patients. Cochrane Database Syst Rev. 2014;10:CD005590. A3. Utsunomiya M, Dobashi H, Odani T, et al. Optimal regimens of sulfamethoxazole-trimethoprim for chemoprophylaxis of Pneumocystis pneumonia in patients with systemic rheumatic diseases: results from a non-blinded, randomized controlled trial. Arthritis Res Ther. 2017;19:1-10.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 321  PNEUMOCYSTIS PNEUMONIA  

GENERAL REFERENCES 1. Salzer HJF, Schafer G, Hoenigl M, et al. Clinical, diagnostic, and treatment disparities between HIV-infected and non-HIV-infected immunocompromised patients with Pneumocystis jirovecii pneumonia. Respiration. 2018;96:52-65. 2. Ma L, Chen Z, Huang da W, et al. Genome analysis of three Pneumocystis species reveals adaptation mechanisms to life exclusively in mammalian hosts. Nat Commun. 2016;7:1-14. 3. Ma L, Cissé OH, Kovacs JA. A molecular window into the biology and epidemiology of Pneumocystis spp. Clin Microbiol Rev. 2018;31:1-49. 4. Yiannakis EP, Boswell TC. Systematic review of outbreaks of Pneumocystis jirovecii pneumonia: evidence that P. jirovecii is a transmissible organism and the implications for healthcare infection control. J Hosp Infect. 2016;93:1-8. 5. Jiang X, Mei X, Feng D, et al. Prophylaxis and treatment of Pneumocystis jiroveci pneumonia in lymphoma patients subjected to rituximab-contained therapy: a systemic review and meta-analysis. PLoS ONE. 2015;10:1-15. 6. National Comprehensive Cancer Network (NCCN). Clinical Practice Guidelines in Oncology. Prevention and treatment of cancer-related infections. Version 2.2017. http://www.nccn.org. Accessed May 7, 2019. 7. Messiaen PE, Cuyx S, Dejagere T, et al. The role of CD4 cell count as discriminatory measure to guide chemoprophylaxis against Pneumocystis jirovecii pneumonia in human immunodeficiency virusnegative immunocompromised patients: a systematic review. Transpl Infect Dis. 2017;19:1-6. 8. Hoving JC, Kolls JK. New advances in understanding the host immune response to Pneumocystis. Curr Opin Microbiol. 2017;40:65-71. 9. Opata MM, Hollifield ML, Lund FE, et al. B lymphocytes are required during the early priming of CD4+ T cells for clearance of Pneumocystis infection in mice. J Immunol. 2015;195:611-620. 10. Kutty G, Davis AS, Ferreyra GA, et al. Beta-glucans are masked but contribute to pulmonary inflammation during Pneumocystis pneumonia. J Infect Dis. 2016;214:782-791. 11. Shibata S, Kikuchi T. Pneumocystis pneumonia in HIV-1-infected patients. Respir Investig. 2019. [Epub ahead of print.] 12. Calero-Bernal ML, Martin-Garrido I, Donazar-Ezcurra M, et al. Intermittent courses of corticosteroids also present a risk for Pneumocystis pneumonia in non-HIV patients. Can Respir J. 2016;2016:1-7. 13. Ebner L, Walti LN, Rauch A, et al. Clinical course, radiological manifestations, and outcome of Pneumocystis jirovecii pneumonia in HIV patients and renal transplant recipients. PLoS ONE. 2016;11:e0164320. 14. Alanio A, Hauser PM, Lagrou K, et al. ECIL guidelines for the diagnosis of Pneumocystis jirovecii pneumonia in patients with haematological malignancies and stem cell transplant recipients. J Antimicrob Chemother. 2016;71:2386-2396. 15. Sasso M, Chastang-Dumas E, Bastide S, et al. Performances of four real-time PCR assays for diagnosis of Pneumocystis jirovecii pneumonia. J Clin Microbiol. 2016;54:625-630.

2068.e1

16. Fauchier T, Hasseine L, Gari-Toussaint M, et al. Detection of Pneumocystis jirovecii by quantitative PCR to differentiate colonization and pneumonia in immunocompromised HIV-positive and HIVnegative patients. J Clin Microbiol. 2016;54:1487-1495. 17. Le Gal S, Robert-Gangneux F, Pepino Y, et al. A misleading false-negative result of Pneumocystis real-time PCR assay due to a rare punctual mutation: a French multicenter study. Med Mycol. 2017;55:180-184. 18. Maschmeyer G, Helweg-Larsen J, Pagano L, et al. ECIL guidelines for treatment of Pneumocystis jirovecii pneumonia in non-HIV-infected haematology patients. J Antimicrob Chemother. 2016;71: 2405-2413. 19. Panel on Opportunistic Infections in HIV-Infected Adults and Adolescents. Guidelines for the prevention and treatment of opportunistic infections in HIV-infected adults and adolescents: recommendations from the Centers for Disease Control and Prevention, the National Institutes of Health, and the HIV Medicine Association of the Infectious Diseases Society of America. http:// aidsinfo.nih.gov/contentfiles/lvguidelines/adult_oi.pdf. Accessed March 27, 2019. 20. White PL, Backx M, Barnes RA. Diagnosis and management of Pneumocystis jirovecii infection. Expert Rev Anti Infect Ther. 2017;15:435-447. 21. Fujikura Y, Manabe T, Kawana A, Kohno S. Adjunctive corticosteroids for Pneumocystis jirovecii pneumonia in non-HIV-infected patients: a systematic review and meta-analysis of observational studies. Arch Bronconeumol. 2017;53:55-61. 22. Wieruszewski PM, Barreto JN, Frazee E, et al. Early corticosteroids for Pneumocystis pneumonia in adults without HIV are not associated with better outcomes. Chest. 2018;154:636-644. 23. Montesinos I, Delforge ML, Ajjaham F, et al. Evaluation of a new commercial real-time PCR assay for diagnosis of Pneumocystis jirovecii pneumonia and identification of dihydropteroate synthase (DHPS) mutations. Diagn Microbiol Infect Dis. 2017;87:32-36. 24. Panel on Opportunistic Infections in HIV-Exposed and HIV-Infected Children. Guidelines for the Prevention and Treatment of Opportunistic Infections in HIV-Exposed and HIV-Infected Children. Department of Health and Human Services. Available at http://aidsinfo.nih.gov/contentfiles/ lvguidelines/oi_guidelines_pediatrics.pdf. Accessed March 27, 2019. 25. Brakemeier S, Pfau A, Zukunft B, et al. Prophylaxis and treatment of Pneumocystis jirovecii pneumonia after solid organ transplantation. Pharmacol Res. 2018;134:61-67. 26. Williams KM, Ahn KW, Chen M, et al. The incidence, mortality and timing of Pneumocystis jiroveci pneumonia after hematopoietic cell transplantation: a CIBMTR analysis. Bone Marrow Transplant. 2016;51:573-580. 27. Maertens J, Cesaro S, Maschmeyer G, et al. ECIL guidelines for preventing Pneumocystis jirovecii pneumonia in patients with haematological malignancies and stem cell transplant recipients. J Antimicrob Chemother. 2016;71:2397-2404. 28. Goto N, Futamura K, Okada M, et al. Management of Pneumocystis jirovecii Pneumonia in Kidney Transplantation to Prevent Further Outbreak. Clin Med Insights Circ Respir Pulm Med. 2015;9: 81-90.

2068.e2

CHAPTER 321  PNEUMOCYSTIS PNEUMONIA  

REVIEW QUESTIONS 1. A 47-year-old previously healthy male comes into the emergency room complaining of a 2-week history of shortness of breath, nonproductive cough, and temperature to 103° F. On physical examination he is tachypneic, with no rales or rhonchi, and has thrush on his buccal mucosa. His chest radiograph and pulse oximetry are normal. Which of the following is true? A . The normal chest radiograph and pulse oximetry rule out Pneumocystis pneumonia. B. A chest computed tomography (CT) can help in the diagnosis of Pneumocystis pneumonia because it is almost invariably abnormal, even in the setting of a normal chest X-ray. C. A positive polymerase chain reaction (PCR) test for Pneumocystis definitively confirms the diagnosis of Pneumocystis pneumonia because the specificity of the PCR test for PCP approaches 100%. D. Culture of induced, but not expectorated, sputum on Sabouraud agar will confirm the diagnosis of Pneumocystis pneumonia in 2 to 3 days. E. Bronchoscopy with bronchoalveolar lavage is helpful if positive for Pneumocystis but has a low sensitivity (~40-50%); open lung biopsy is required to confirm the diagnosis of PCP in about 50% or patients. Answer: B  This patient likely has undiagnosed HIV infection and Pneumocystis pneumonia based on the symptoms and presence of thrush. Patients presenting with Pneumocystis pneumonia can have a normal chest x-ray (CXR) and normal pulse oximetry or arterial oxygenation at rest. Even in patients with a normal CXR, a CT scan of the chest will invariably show infiltrates, as illustrated in Figure 341-2. PCR-based assays have a high sensitivity, but the specificity does not approach 100% because it can also be positive in individuals who are colonized or subclinically infected with Pneumocystis but do not have Pneumocystis pneumonia and do not require specific anti-Pneumocystis therapy. Pneumocystis cannot be cultured, so definitive diagnosis depends on direct detection of the organism in a clinical sample, usually an induced sputum or bronchoalveolar lavage sample. Bronchoscopy with bronchoalveolar lavage has greater than 90% sensitivity for Pneumocystis pneumonia and is usually the definitive diagnostic procedure for Pneumocystis pneumonia. Before the AIDS epidemic and the demonstration of the utility of bronchoscopy, open lung biopsy was often required to make the diagnosis, but currently it is rarely necessary; negative results by bronchoscopy with BAL are sufficient to rule out Pneumocystis pneumonia in most circumstances. 2. A 25-year-old female with a history of prior intravenous drug use comes in with a 1-week history of fever and a nonproductive cough. CXR shows diffuse bilateral infiltrates. Resting Pao2 at presentation is 96 mm Hg with an A-a gradient of 15 mm Hg. A rapid HIV test is positive. An induced sputum stain for Pneumocystis is positive, and the patient is started on trimethoprim-sulfamethoxazole, 2 double strength tables three times a day. She is sent home to be followed in clinic. Three days later she returns complaining of increasing shortness of breath with walking. Resting Pao2 has decreased to 86 mm Hg, with an A-a gradient of 30 mm Hg. What is the most appropriate management course for this patient? A . Discontinue the trimethoprim-sulfamethoxazole and begin pentamidine. B. Discontinue the trimethoprim-sulfamethoxazole and begin clindamycinprimaquine. C. Perform bronchoscopy to rule out other infections, and evaluate the patient for pulmonary emboli. D. Begin prednisone 40 mg/day in addition to the trimethoprimsulfamethoxazole. E. Continue the trimethoprim-sulfamethoxazole; no change in therapy is needed.

Answer: E  Deterioration in oxygenation at day 3 to 4 after start of therapy is common and expected in patients with Pneumocystis pneumonia and does not signify clinical failure. Thus there is no reason to change therapy to an alternative regimen or to search for alternative diagnoses at this point. Although prednisone has been shown to decrease the frequency of deterioration in oxygenation in patients with Pneumocystis pneumonia, a large randomized trial found that clinical benefit in terms of respiratory failure and survival outcome was limited to patients with moderate to severe disease, defined as a Pao2 less than 70 mm Hg or an A-a gradient greater than 35 mm Hg. Moreover, addition of prednisone after 3 days of specific anti-Pneumocystis therapy has not been shown to be beneficial. 3. A 38-year-old male with a recent history of thrush is admitted with a fever and nonproductive cough. He is diagnosed with HIV infection and Pneumocystis pneumonia and is started on trimethoprim-sulfamethoxazole. He would like to start anti-HIV medications soon. How should his combination antiretroviral therapy (cART) be managed? A . Initiate cART 2 months after discontinuation of anti-Pneumocystis therapy. B. Initiate cART immediately if HIV plasma viral load is greater than 100,000 copies/mL, and 2 months after discontinuation of antiPneumocystis therapy if HIV plasma viral load is less than or equal to 100,000 copies/mL. C. Timing of cART should be based on the CD4 count. If the CD4 count is less than or equal to 350 cells/μL, cART should be initiated within 1 month of completion of therapy. If the CD4 count is greater than 350 cells/μL, therapy can be deferred until the cell count drops below that level. D. Initiate cART within 2 weeks of starting anti-Pneumocystis therapy if possible. E. Don’t initiate cART until the CXR has normalized to minimize the risk for developing immune reconstitution inflammatory syndrome (IRIS). Answer: D  A randomized trial in patients with an opportunistic infection, 63% of whom had Pneumocystis pneumonia, found that initiation of cART early (~12 days) versus late (~45 days) after the start of therapy for the opportunistic infection was associated with a lower risk for AIDS progression or death. Patients with Pneumocystis pneumonia should be started on cART regardless of the CD4 count or viral load because they have already shown that their immune systems are severely compromised; without cART or antiPneumocystis prophylaxis, they are at high risk for recurrence of Pneumocystis pneumonia. Of note, most patients diagnosed with Pneumocystis pneumonia will have a CD4 count under 200 cells/μL. While cases of IRIS in the setting of Pneumocystis pneumonia have been reported, the incidence appears to be relatively low and rarely appears to be life-threatening; in the randomized trial noted previously, the incidence of IRIS in patients with Pneumocystis pneumonia was 7.3%.

CHAPTER 321  PNEUMOCYSTIS PNEUMONIA  

4. A patient who received a renal transplant 6 months ago and is receiving mycophenolate mofetil, cyclosporine, and prednisone for chronic immunosuppression is diagnosed with Pneumocystis pneumonia. This is the fourth renal transplant recipient diagnosed with Pneumocystis pneumonia during the past year. Renal transplant patients at this center do not routinely receive anti-Pneumocystis prophylaxis because no cases of Pneumocystis pneumonia were seen in the prior 10 years. What is the best management strategy for stopping this outbreak of Pneumocystis pneumonia? A . Place the patient on respiratory isolation to prevent spread of the organism to other transplant patients. B. Clean the walls, floors, and ceilings of the patient rooms in the clinic and inpatient service with bleach to eliminate the source of Pneumocystis. C. Have the patients at risk for infection wash daily with a povidone-iodine antiseptic solution to minimize colonization with Pneumocystis. D. Begin prophylaxis with trimethoprim-sulfamethoxazole in all transplant patients for at least 6 months, and longer for high-risk patients (e.g., those requiring additional immunosuppression for graft rejection). E. Begin prophylaxis with clindamycin-pyrimethamine in all transplant patients for at least 4 months, and longer for high-risk patients (e.g., those requiring additional immunosuppression for graft rejection). Answer: D  Outbreaks of Pneumocystis pneumonia have been reported from multiple renal transplant centers over the past 10 years. As demonstrated in molecular typing studies, these outbreaks are caused by a single or a limited number of strains of Pneumocystis at each center. This strongly implies recent transmission of the infection and potential acquisition as a nosocomial infection. Current data suggest that Pneumocystis is acquired from other humans through transmission by the respiratory route. There is no evidence for an environmental reservoir, thus cleaning a room will not eliminate the sources of infection. Although colonization or subclinical infection of the respiratory tract with Pneumocystis has been documented by sensitive PCR-based methods, there is no evidence that the skin is colonized, and thus washing with povidone-iodine antiseptic solution will not eliminate potential reservoirs for this specific organism. Even though it is logical to use respiratory isolation to prevent the spread of Pneumocystis, there are no data to demonstrate that this is an effective strategy to halt an outbreak. Infection may be spread during the incubation period, which in animal models can be 8 to 12 weeks before the development of severe pneumonia. Clindamycin-pyrimethamine has not been demonstrated to have activity in either treatment or prophylaxis of Pneumocystis pneumonia. Trimethoprim-sulfamethoxazole is a highly effective prophylactic regimen, and institution of prophylaxis with trimethoprimsulfamethoxazole has been successful in halting the outbreaks of Pneumocystis pneumonia reported to date.

2068.e3

5. A 35-year-old female comes in with a headache and weakness in her right arm. An MRI scan shows multiple ring-enhancing lesions. Her anti-toxoplasma IgG antibody test is positive, but her IgM is negative. Cerebrospinal fluid is positive for Toxoplasma by PCR. Her HIV test is positive, and her baseline labs include a CD4 count of 27 cells/μL and an HIV viral load of 42,500 copies/mL. The patient noted that she had received trimethoprimsulfamethoxazole many years ago for a urinary tract infection and said she thought she developed a rash around her elbows but did complete her therapy. She is started on pyrimethamine-sulfadiazine-leucovorin for treatment of CNS toxoplasmosis. How should anti-Pneumocystis prophylaxis be managed in this patient? A . No specific anti-Pneumocystis prophylaxis is required. B. Begin dapsone, 100 mg/day, because she may have an allergy to trimethoprim-sulfamethoxazole. C. Begin atovaquone, 1500 mg per day, because she may have an allergy to trimethoprim-sulfamethoxazole and patients with sulfa allergies are frequently allergic to dapsone. D. Begin azithromycin, 1200 mg per week, to provide prophylaxis for both Pneumocystis and Mycobacterium avium complex. E. Begin trimethoprim-sulfamethoxazole, one double-strength tablet daily; change to dapsone if the patient develops a rash. Answer: A  This severely immunocompromised patient with HIV infection and toxoplasmosis is at high risk for Pneumocystis pneumonia based on her CD4 count. Although she describes a possible rash during prior treatment with trimethoprim-sulfamethoxazole, a rash limited to the elbows would be unusual for a drug reaction, and reactions to sulfamethoxazole, the component that appears to be most commonly responsible for reactions to trimethoprimsulfamethoxazole, are rarely immediate or life-threatening. Thus therapy with a sulfa drug is not contraindicated, and the patient was started on pyrimethamine-sulfadiazine-leucovorin for treatment of toxoplasmosis. Pyrimethamine and sulfadiazine target the same enzymes as trimethoprim and sulfamethoxazole (dihydrofolate reductase and dihydropteroate synthase, respectively), and studies have shown that patients receiving pyrimethamine-sulfadiazine-leucovorin are at low risk for developing Pneumocystis pneumonia and thus do not require additional prophylaxis. While azithromycin prophylaxis should be started in this patient because of the low CD4 count, it has not been studied as an anti-Pneumocystis agent, and thus additional prophylaxis for Pneumocystis would be required if it were indicated.

2068

CHAPTER 322  Mycetoma and Dematiaceous Fungal Infections  

interchangeably. Dematiaceous fungal infections generally fall into three broad categories: mycetoma (e.g., Madura foot), chromomycosis (also known as chromoblastomycosis), and phaeohyphomycosis.1,2 Mycetoma (a tumor produced by fungi) is a disease of historical interest and was first described in 1842 in the Madura district of India, hence the terms “Madura foot,” “maduromycosis,” and “maduromycetoma.”

The Pathogens

More than 100 dematiaceous fungi have been identified as causes of disease in humans. The most common organisms and their related conditions are listed in Table 322-1. The taxonomy of the dematiaceous fungi is somewhat confusing because these agents belong to different classes, including Hyphomycetes, Ascomycetes, Basidiomycetes, Coelomycetes, and Zygomycetes. The most common agents of phaeohyphomycosis include species in the following genera: Alternaria, Bipolaris, Curvularia, Exophiala, Cladosporium, Cladophialophora, Fonsecaea, Exserohilum, Ochroconis, Phialophora, Phaeoacremonium, and Chaetomium. These agents are ubiquitous saprophytes of soil and decaying matter, and some are important plant pathogens. In tissue, these organisms exist as yeastlike cells, septated hyphae, or a combination of yeast and hyphae. Most of these organisms demonstrate melanin pigmentation (brownish coloration) in the cell walls on microscopic examination. Most cases of chromomycosis are caused by three species: Fonsecaea pedrosoi, Cladosporium carrionii, and Phialophora verrucosa. The distinctive histologic appearance is characterized by the presence of thick-walled, dark brown bodies known as sclerotic cells or copper pennies, which represent individual organisms and may be seen in clusters or as single cells. The etiologic fungi causing chromomycosis are indistinguishable on histologic examination of tissue. Mycetoma is caused by two groups of organisms: (1) the filamentous aerobic actinomycetes (actinomycetoma) and (2) a wide range of saprophytic soil and woody plant fungi (eumycetoma). Eumycetoma accounts for about 50% of cases of mycetoma.3 A variety of Nocardia species (e.g., Nocardia brasiliensis, Nocardia asteroides), Actinomadura species (e.g., Actinomadura pelletierii, Actinomadura madurae), and Streptomyces species (e.g., Streptomyces somaliensis) cause actinomycetoma. Most cases of eumycetoma are due to Madurella species

TABLE 322-1 DEMATIACEOUS FUNGI AND ASSOCIATED DISEASES CLINICAL CONDITION

322  MYCETOMA AND DEMATIACEOUS FUNGAL INFECTIONS

Fonsecaea pedrosoi Cladophialophora carrionii Phialophora verrucosa

Cutaneous or subcutaneous disease

Exophiala jeanselmei Exophiala dermatitidis Phialophora spp Bipolaris spp Alternaria spp

Sinusitis

Bipolaris spp Curvularia spp Exserohilum spp Alternaria spp

Central nervous system

Cladophialophora bantiana Verruconis (Ochroconis) gallopava Rhinocladiella mackenziei Chaetomium atrobrunneum Exophiala dermatitidis

Health care associated

Exserohilum rostratum Exophiala spp

Disseminated

Exophiala dermatitidis Exophiala jeanselmei Bipolaris spp Verruconis (Ochroconis) gallopava Phialophora spp Lomentospora (Scedosporium) prolificans

Eumycetoma

Madurella mycetomatis Exophiala spp Curvularia spp Leptosphaeria senegalensis Pseudallescheria boydii Acremonium spp Fusarium spp

PETER G. PAPPAS AND DIMITRIOS P. KONTOYIANNIS



DEFINITION

Dematiaceous fungi represent a large group of fungal organisms characterized by the presence of abundant melanin in the cell wall, which gives rise to a brown-black coloration on artificial culture media and which can be seen on histopathologic specimens. A related term, phaeohyphomycosis, refers broadly to infection by these pigmented fungi. The two terms are often used

COMMON ETIOLOGIC AGENTS

Chromomycosis

Non-dematiaceous fungi

2068.e4

ABSTRACT

CHAPTER 322  Mycetoma and Dematiaceous Fungal Infections  

Mycetoma and dematiaceous fungal infections are caused by over 100 fungal species that have in common the characteristic of being “melanized” or pigmented, leading to a brown or darkish pigmentation in histopathologic samples and in culture. These organisms cause a range of disorders that vary from benign conditions such as onychomycosis, to extremely destructive and disfiguring mycetoma, to life-threatening, invasive phaeohyphomycosis in an immunocompromised individual. Diagnosis is based on the identity of characteristic fungi on pathologic specimens and is confirmed by culture. The triazole antifungals (e.g., itraconazole, voriconazole, and posaconazole) seem to have the best in vitro and clinical activity versus these pathogens. A combined medical/ surgical approach is probably best for those with mycetoma and extensive tissue involvement and for individuals with disseminated phaeohyphomycosis, especially those with central nervous system involvement.

KEYWORDS

phaeohyphomycoses dematiaceous fungi mycetoma Madura foot chromomycosis

CHAPTER 322  Mycetoma and Dematiaceous Fungal Infections  

United States Pseudallescheria boydii (Scedosporium apiospermum)

2069

Middle East/East Africa Madurella mycetomatis Streptomyces somaliensis India Nocardia spp Madurella grisea

West Africa Leptoshaeria senegalensis

Central and South America Madurella grisea Nocardia spp

FIGURE 322-1.  Predominant agents of mycetoma according to region.

(e.g., Madurella mycetomatis causes 70% of all cases of eumycetoma worldwide). Other causes of eumycetoma include some nondematiaceous fungi such as Fusarium species, Acremonium species, Pseudallescheria boydii, and several phaeohyphomycetes including Exophiala and Curvularia species. Eumycetoma is further characterized on the basis of the color of the granular drainage: white- to yellow-grain mycetomas (white piedra) are typically caused by hyalohyphomycetes (e.g., P. boydii, Fusarium species, Acremonium species), and black-grain eumycetomas are caused by Madurella species and other less common fungi.4  

EPIDEMIOLOGY

Dematiaceous fungi are found in the environment worldwide. Although there is no unique endemic area for most of these infections, some observations are relevant. Allergic fungal sinusitis associated with dematiaceous fungi appears to be more common in the southern United States. Chronic infections of the lower extremities are more commonly seen in men and in tropical areas. Chromomycosis are more prevalent in rural populations in the tropics and are hyperendemic in certain geographic areas such as Madagascar, India, Brazil, and other poorer countries in Africa and South America. Cutaneous infections usually occur as a result of minor skin trauma and direct inoculation of the organism. Phaeohyphomycosis is an important emerging fungal infection in medically advanced regions, particularly among immunocompromised patients such as solid organ and hematopoietic stem cell transplant recipients, patients with prolonged neutropenia, and other immunocompromised individuals.5,6 Risk factors for extracutaneous infection include intravenous drug abuse, chronic sinusitis, freshwater immersion, and chronic immunosuppression. Phaeohyphomycosis is reported in human immunodeficiency virus (HIV)-infected patients but is far less common than other opportunistic fungi. Extracutaneous invasive disease can also occur in otherwise normal patients but is much less common. Recent data suggest subtle host immune abnormalities, including CARD-9 mutations and other disorders associated with TH-17 deficiency, in a subset of these previously “normal” individuals.7,8 In the United States, an epidemic of fungal meningitis, epidural abscess, sacroiliitis, vertebral osteomyelitis, discitis, and peripheral arthritis involved over 750 persons and was caused by Exserohilum rostratum following injection of contaminated methylprednisolone acetate from a single compounding pharmacy; this is a dramatic example of the risk for dematiaceous fungal infections following invasive procedures in the health care setting.9 Previous reports of infection due to Exophiala species following contaminated steroid injections, infected breast implants, other prosthetic materials, and, rarely, contaminated intravascular catheters and intravenous fluids further underscore the importance of these organisms as potential health care–associated pathogens.

Mycetoma has a global distribution, but it occurs primarily in the tropical zones. The disorder is quite prevalent in India, Latin America, the Middle East, and sub-Saharan Africa (the “mycetoma belt”). Sudan has a particularly high burden of mycetoma.10 Indigenously acquired mycetoma is sporadic in North America and Europe. Moreover, the relative frequency of actinomycetoma and eumycetoma differs among geographic areas. Eumycetoma is more common in India and Africa, and actinomycetoma is more common in Central and South America. Furthermore, the causative agents of mycetoma differ in their geographic distribution. For example, P. boydii is the most common agent of mycetoma in North America, and Actinomadura and Nocardia species are predominant in Central and South America. Leptosphaeria senegalensis and M. mycetomatis are predominant in sub-Saharan Africa and India (Fig. 322-1). The ratio of male to female patients with mycetoma is 5 : 1. The disease is typically seen in rural areas and in persons susceptible to local trauma and contamination from soil. Hence, farmers, gardeners, woodcutters, herders, and people who work outside while barefoot are more susceptible to this infection. Local trauma (e.g., wood splinters) introduces a mycetoma-causative organism into the skin and subcutaneous tissues and initiates a chain of events that leads to chronic, suppurative granulomatous inflammation, tumefaction, formation of multiple fistulous tracts and sinuses, deep abscesses, fibrosis and scar formation, and extension to adjacent connective tissue across the lines of least resistance (fascia) and ultimately to bones, muscles, nerves, and tendon sheaths, leading to gross anatomic distortion of the affected site. The genetics and immunopathogenesis of mycetoma are not well defined, but it appears that there are differences in host susceptibility as some infected persons have impaired or delayed hypersensitivity reactions or polymorphisms in genes encoding for chemokines (e.g., CCL50) and cytokines (e.g., interleukin-10). Mycetoma does not appear to be more common in immunocompromised hosts.  

CLINICAL MANIFESTATIONS

Chromomycosis is manifested as a cutaneous or subcutaneous lesion that may range in size from a small papule to a large confluent plaque involving a major portion of an extremity.11 Single or multiple lesions are seen; ulceration may occur. Lesions may remain unchanged in size and consistency for months or years, although most tend to progress in the absence of specific therapy. Chronic lesions may become dry and crusted with a raised border, which may be smooth or irregular, and which may take on a verrucous, warty appearance. Multiple lesions can coalesce to form larger plaques in which central scarring may develop. Cutaneous lesions usually remain confined to one anatomic site, although nodular lymphangitis and multifocal cutaneous disease from autoinoculation may occur. Common complications include local disfigurement

2070

CHAPTER 322  Mycetoma and Dematiaceous Fungal Infections  

due to scarring and extensive tissue involvement. Disseminated disease involving visceral organs is rare. Superficial phaeohyphomycosis is characterized by tinea nigra and black piedra.12 Tinea nigra is a darkening of the skin caused by growth of Phaeoannellomyces werneckii in the stratum corneum. Black piedra is associated with the development of focal thickening on the hair shaft and results from colonization of the shaft by Piedraia hortae. Cutaneous phaeohyphomycosis involves deeper skin structures and results in dermatomycosis and onychomycosis; this is frequently due to agents such as Scytalidium and Phyllosticta species. Subcutaneous phaeohyphomycosis may be confused clinically with chromomycosis. Patients have discrete subcutaneous nodules or cysts that result from direct inoculation or penetrating trauma. The most common organisms are Exophiala jeanselmei, Exophiala dermatitidis, Alternaria alternans, and Phialophora species. Mycotic keratitis as a result of infection with Curvularia, Exophiala, and Exserohilum species may occur after corneal trauma or surgery. Foreign body–related infections are seen in patients undergoing chronic ambulatory peritoneal dialysis in whom fungal peritonitis develops, in patients with indwelling intravenous catheters, and in other devices such as breast implants. Fungal sinusitis13 is commonly associated with dematiaceous fungi and can be manifested as allergic fungal sinusitis, a fungus ball in a sinus cavity, and invasive fungal sinusitis associated with extension into bone, soft tissue, and the central nervous system. This last destructive process is indistinguishable from rhinocerebral zygomycosis or invasive Aspergillus sinusitis. Bipolaris, Curvularia, Exserohilum, and Alternaria species are the most common organisms causing invasive fungal sinusitis. Systemic phaeohyphomycosis may result from direct extension from a colonized area or dissemination from a distant source. Most patients with systemic disease have significant underlying immunosuppression, and the organisms have a proclivity for involvement of the brain, lungs, endocardium, and other visceral organs. Among patients with primary central nervous system disease, Cladophialophora bantiana, Rhinocladiella mackenziei, and Chaetomium atrobrunneum are the most common etiologic agents, and the majority of these are otherwise healthy patients with little or no underlying immunodeficiency. Among immunocompromised patients, Verruconis (Ochroconis) gallopava, Bipolaris species, and Exophiala dermatitidis are seen more commonly. In severely immunocompromised patients, especially those with hematologic malignancies or stem cell transplant recipients with or without prolonged and profound neutropenia, infection due to Lomentospora (Scedosporium) prolificans can present as multiorgan dissemination. Infection due to this multidrugresistant pathogen is associated with an exceptionally high mortality rate. Mycetoma is a chronic, slowly progressive infection that starts in the subcutaneous tissue and spreads across tissue planes to contiguous structures. The lesion is usually confined to one anatomic site. The clinical manifestations and natural history of mycetoma are variable and, to some degree, related to the pathogenic agent involved. For example, the progression of eumycetoma tends to be slower than that of actinomycetoma. In addition, eumycetoma lesions tend to be more confined and have less inflammation and fewer granulomas and fistulas but more fibrosis compared with actinomycetoma lesions. The foot is the most common site involved in mycetoma (Fig. 322-2), followed by the hands, but any other part of the body may become involved. Painless nodular and/or papular swelling is a common early manifestation of mycetoma, followed by a slow evolution to painless, fixed woody induration. This infection typically runs a chronic, relentless course, sometimes spanning several decades. It is characterized by recurring cycles of suppuration, draining sinuses, bacterial superinfection, and scar formation. Old sinuses close, and new ones occur. Satellite lesions are common. Constitutional symptoms are rare. The presence of fever often indicates bacterial superinfection. Substantial bone involvement mimicking chronic osteomyelitis with cavitary bone lesions, periosteal reaction or sclerosis (seen with radiography, computed tomography, or magnetic resonance imaging [MRI] studies), osteoporosis, and reactive periosteal bone formation may occur. However, pathologic fractures are rare. Because nerves are relatively spared from involvement, neuropathic manifestations are uncommon. Inexorable limb deformity and misuse because of destruction of deeper tissues may be seen in chronic, refractory, and advanced cases.14 Mycetoma does not spread hematogenously; thus, visceral dissemination is not seen, although regional lymphadenitis may occur.  

DIAGNOSIS

The diagnosis of a dematiaceous fungal infection is suggested by direct examination of a clinical specimen with a 10% potassium hydroxide preparation or special stains to demonstrate pigmentation in the cell walls of these organisms. For patients with chromomycosis, the finding of sclerotic cells or copper

FIGURE 322-2.  Clinical presentation of mycetoma. A 40-year-old farmer from rural Venezuela with a 10-year history of foot edema and slowly progressive deformity following an injury caused by being struck by a hammer presented with chronic crusted plaques, multiple confluent tender abscesses with fistulization, and release of black grains. The range of motion of the patient’s ankle and foot joints was limited, but the joints were not painful. A deep skin biopsy with hematoxylin and eosin staining, periodic acid–Schiff staining, foot radiography, sampling of black grain smears, and a mycology culture were performed. The foot radiograph showed osteofibrosis, destruction of articular surfaces, osteoporosis, and ankylosis, and Madurella spp grew in the culture. (Courtesy of Dr. M. Mendoza, Instituto De Biomedicina, Laboratorio de Micología, San Jose Caracas, Venezuela.)

pennies on skin biopsy is characteristic, and special stains are usually unnecessary. For patients with other forms of phaeohyphomycosis, the Fontana-Masson stain is useful in distinguishing organisms with significant melanin content. Culture remains the means by which a specific etiologic diagnosis is established, and the identity of the organism is largely based on colony and microscopic morphology.15 A polymerase chain reaction–based diagnostic assay for Exserohilum rostratum was developed in the context of the fungal meningitis outbreak and serves as a reliable marker of infection among exposed patients. Otherwise, serologic studies and molecular diagnostics for other organisms are not generally available. The diagnosis of mycetoma is complicated by the fact that nonfungal organisms are frequent etiologic agents. The diagnosis is suspected in the presence of granular drainage from a characteristic skin lesion. However, grains in tissue alone may be difficult to culture because they may be composed of dead organisms. Furthermore, grains may be contaminated by surface bacteria or fungi. Thus, a deep-tissue biopsy specimen is ideal for staining (hematoxylin and eosin, Gram stain, modified Ziehl-Neelsen stain, Gomori methenamine silver, periodic acid–Schiff) and appropriate selective bacterial and fungal cultures. Alternatively, aspiration of grains from unopened sinus tracts can provide suitable material for culture. The culture should be maintained for several weeks because some of the causative agents of mycetoma (e.g., Nocardia and Streptomyces species) may require 4 to 6 weeks for detection by culture methodology.

Differential Diagnosis

The specific manifestations of chromomycosis, cutaneous and subcutaneous phaeohyphomycosis, and mycetoma are often confused with other entities. It is important to consider benign or malignant skin tumors, other chronic granulomatous lesions (e.g., thorn granuloma), and verrucous leishmaniasis. Deep skin and subcutaneous infections due to other fungi (e.g., blastomycosis, sporotrichosis, cryptococcosis, and coccidioidomycosis), and higher-order bacteria (e.g., nocardiosis, mycobacteriosis), must be considered. Skin lesions without fistulas can occasionally resemble unusual forms of sporotrichosis (mycetomatous lymphatic sporotrichosis) and granulomatous dermatophyte infections. The latter infection, which is typically seen in Africans and sometimes called pseudomycetoma (Majocchi granuloma), is a painless granulomatous induration of the skin and subcutaneous tissues caused by dermatophytes that may be associated with grains consisting of fungi. Unlike mycetoma, pseudomycetoma is confined to the skin and subcutaneous tissue and does not spread to deeper tissues. Chronic severe botryomycosis (typically caused by Staphylococcus aureus) with purulent exudates, grains, and draining sinus tracts may be confused with mycetoma. Actinomycosis (Chapter 313), which is caused by endogenous microaerophilic actinomycetes, also has a propensity for grains and formation of draining sinus tracts, but its location is characteristic in the neck, chest, abdomen, and pelvis. Differentiation between mycetomas

with bone involvement and chronic osteomyelitis or osseous tumors may be challenging. Ultrasonography has been used to reliably differentiate mycetoma from either tumor or osteomyelitis. The dot-in-circles sign seen on MRI (tiny hypodense foci, representing grains, within high-hyperintensity spherical lesions, representing granulomas, scattered by areas of fibrosis) may provide an early and specific diagnostic clue for mycetoma.

TREATMENT  For chromomycosis, surgical excision of a cutaneous or subcutaneous lesion is often curative,16 although antifungal therapy is usually given in conjunction with surgery. There is only anecdotal evidence and a few small trials that have examined the efficacy of antifungal therapy for this condition. Further details about systemic antifungal agents are found in Chapter 315. Historically, oral 5-flucytosine (5-FC, 150 mg/kg/day in four divided doses) has been advocated for the oral treatment of chromomycosis based on moderate in vitro activity and clinical experience. Because of limited availability, cost, and potential toxicity, 5-FC is rarely used for this purpose. The triazoles, including itraconazole (200 mg orally twice daily), voriconazole (200 mg twice daily), and posaconazole (300 mg daily), demonstrate the best in vitro activity, although clinical studies with these agents are very limited. Terbinafine (500 mg orally twice daily) has been used successfully for the treatment of chromomycosis and is an effective alternative. Duration of therapy, with or without surgical excision, is generally 3 to 12 months, depending on clinical response. For phaeohyphomycosis, the expanded-spectrum triazoles, including itraconazole, voriconazole, posaconazole, and isavuconazole, demonstrate excellent in vitro activity against the most common pathogenic dematiaceous fungi. Amphotericin B has modest in vitro activity against most of these fungi and is usually reserved for patients with life-threatening or disseminated disease. Most patients with epidemic Exserohilum rostratum infections have been treated successfully with voriconazole (intravenous, then oral), with or without a lipid formulation of amphotericin B, for severe central nervous system involvement. Posaconazole has been used as salvage therapy for those intolerant to voriconazole or in whom voriconazole has failed. Among the triazoles, posaconazole probably offers the most potential in this clinical setting based on anecdotal experience from patients with central nervous system infection. The length of antifungal therapy for any of the systemic phaeohyphomycoses is unclear, but therapy should probably be continued for at least 4 to 6 months or until 1 month after resolution of all signs and symptoms of disease. Therapy for mycetoma should be individualized. Optimal management is not well defined, and the literature provides only heterogeneous and uncontrolled studies. As such, there is no “gold standard” therapy for mycetoma. Successful treatment is dependent upon distinguishing actinomycetoma from eumycetoma, identification of the causative pathogen, and the extent of tissue invasion. For actinomycetoma, treatment consists of chronic antimicrobial therapy (at least 9 to 12 months) in conjunction with limited debulking surgery in selected cases.17 Several regimens (trimethoprim-sulfamethoxazole, tetracyclines, dapsone, streptomycin) have been used in different sequences and combinations depending on the organism (e.g., trimethoprim-sulfamethoxazole with or without dapsone for Nocardia species, streptomycin with dapsone for A. madurae). Parenteral streptomycin is usually reserved for refractory cases not responding to oral therapy. Response to therapy is slow (usually within 4 weeks), and relapse is common. Several cycles of therapy may be necessary for chronic recurrent disease. The best treatment results for eumycetoma have been achieved with prolonged use (at least 9 to 12 months) with an imidazole such as ketoconazole 200 to 400 mg daily or itraconazole 200 to 400 mg daily, combined with appropriate surgery. In vitro activity and clinical experience with the newer azoles (voriconazole, posaconazole, and isavuconazole) and terbinafine are promising, but data are very limited.18 Clinical results with amphotericin B for refractory cases have been generally disappointing, and many of the causative agents demonstrate in vitro resistance to this compound. Surgery often plays a pivotal role in eumycetoma because most disease is refractory to antifungal therapy alone. The need and extent of surgery depend on the specific agent and the degree of tissue invasion. Wide-margin surgery for localized lesions is usually curative. Extensive débridement and disfiguring surgery are reserved for severe and refractory cases. The prognosis for mycetoma depends on the anatomic site and degree of tissue involvement. Extension into bony structures is associated with the worst long-term consequences. Thus, early diagnosis and intervention are critical in optimizing outcomes. Finally, mycetoma is a frequently ignored infection, causing significant socioeconomic burden in many impoverished regions of the world. As such, it is notable that mycetoma is now recognized as a neglected tropical disease by the World Health Organization.19 It is hoped that in gaining this status, there will be greater efforts toward education, prevention, specific therapies, and research.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 322  Mycetoma and Dematiaceous Fungal Infections  

GENERAL REFERENCES 1. Emmanuel P, Dumre SP, John S, et al. Mycetoma: a clinical dilemma in resource limited settings. Ann Clin Microbiol Antimicrob. 2018;17:35. 2. Wong EH, Revankar SG. Dematiaceous molds. Infect Dis Clin North Am. 2016;30:165-178. 3. Verma P, Jha A. Mycetoma: reviewing a neglected disease. Clin Exp Dermatol. 2019;44:123-129. 4. Zijlstra EE, van de Sande WW, Welsh O, et al. Mycetoma: a unique neglected tropical disease. Lancet Infect Dis. 2016;16:100-112. 5. McCarty TP, Baddley JW, Walsh TJ, et al. Phaeohyphomycosis in transplant recipients: results from the Transplant Associated Infection Surveillance Network (TRANSNET). Med Mycol. 2015;53: 440-446. 6. Revankar S, Baddley J, Chen S, et al. A Mycoses Study Group international prospective study of phaeohyphomycosis: an analysis of 99 proven/probable cases. Open Forum Infect Dis. 2017;1-7. 7. Alves de Medeiros AK, Lodewick E, Bogaert DJ, et al. Chronic and invasive fungal infections in a family with CARD9 deficiency. J Clin Immunol. 2016;36:204-209. 8. Lanternier F, Barbati E, Meinzer U, et al. Inherited CARD9 deficiency in 2 unrelated patients with invasive Exophiala infection. J Infect Dis. 2015;211:1241-1250. 9. Renfrow JJ, Frenkel MB, Hsu W. Fungal contamination of methylprednisolone causing recurrent lumbosacral intradural abscess. Emerg Infect Dis. 2017;23:552-553. 10. Fahal A, Mahgoub el S, El Hassan AM, et al. Mycetoma in the Sudan: an update from the Mycetoma Research Centre, University of Khartoum, Sudan. PLoS Negl Trop Dis. 2015;1-19.

2071.e1

11. Fahal AH, Suliman SH, Hay R. Mycetoma: the spectrum of clinical presentation. Trop Med Infect Dis. 2018;3:1-11. 12. Caviedes MP, Torre AC, Eliceche ML, et al. Cutaneous phaeohyphomycosis. Int J Dermatol. 2017;56:415-420. 13. Kalin-Hajdu E, Hirabayashi KE, Vagefi MR, Kersten RC. Invasive fungal sinusitis: treatment of the orbit. Curr Opin Ophthalmol. 2017;28:522-533. 14. Bitan O, Wiener-Well Y, Segal R, Schwartz E. Mycetoma (Madura foot) in Israel: recent cases and a systematic review of the literature. Am J Trop Med Hyg. 2017;96:1355-1361. 15. Ahmed AA, van de Sande W, Fahal AH. Mycetoma laboratory diagnosis: review article. PLoS Negl Trop Dis. 2017;11:1-17. 16. Gismalla MDA, Ahmed GMA, MohamedAli MM, et al. Surgical management of eumycetoma: experience from Gezira Mycetoma Center, Sudan. Trop Med Health. 2019;47:1-6. 17. Arenas R, Fernandez Martinez RF, Torres-Guerrero E, et al. Actinomycetoma: an update on diagnosis and treatment. Cutis. 2017;99:E11-E15. 18. Wilby KJ. A review of the clinical pharmacokinetics and pharmacodynamics of isavuconazole. Eur J Drug Metab Pharmacokinet. 2018;43:281-290. 19. Burki TK. Starting from scratch—the unique neglect of mycetoma. Lancet Infect Dis. 2016;16: 1011-1012.

2071.e2

CHAPTER 322  Mycetoma and Dematiaceous Fungal Infections  

REVIEW QUESTIONS 1. Which of the following organisms is most commonly associated with central nervous system infections? A . Exserohilum rostratum B. Alternaria alternans C. Cladophialophora bantiana D. Scedosporium prolificans E. Exophiala spp Answer: C  Cladophialophora bantiana is commonly associated with central nervous system infections, especially brain abscess, in both normal hosts and in immunocompromised hosts. The other organisms frequently associated with central nervous system infections include Verruconis (Ochroconis) gallo­ pava and Rhinocladiella mackenziei. 2. Which of the following organisms was associated with the U.S. epidemic involving injectable methylprednisolone acetate, leading to hundreds of cases of meningitis, parameningeal infections, and peripheral joint infections? A . Alternaria alternans C. Curvularia spicifera C. Exophiala dermatitidis D. Exserohilum rostratum E. Fonsecaea pedrosoi Answer: D  The major cause of the recently reported U.S. epidemic of contaminated methylprednisolone injections was Exserohilum rostratum. This organism was isolated by culture and/or polymerase chain reaction in the majority of patients for whom positive tests results were available. The epidemic led to 751 proven and probable cases in patients, of whom approximately 8% died as a direct result of the infection. 3. Which of the following statement is true concerning mycetoma? A . Mostly occurs on the hands and face B. May be caused by fungal or bacterial pathogens, depending on geographic location C. Spreads hematogenously to distant organs D. Is most commonly seen in immunocompromised hosts E. Can be readily treated with antifungal therapy alone

Answer: B  Mycetoma is a disease of the tropics, it is seen mostly in immunologically normal individuals, and it can be generally divided into two categories: (1) actinomycetoma, caused by higher-order bacteria, including Nocardia spp, Actinomadura spp, and Streptomyces spp; and (2) eumycetoma, caused by fungal organisms, either dematiaceous (pigmented) molds or hyaline molds. Madurella mycetomatis is the most common cause, but some hyaline molds such as Pseudallescheria boydii, Fusarium spp, and Acremonium spp are associated with eumycetoma. Treatment of fungal mycetoma (eumycetoma) usually involves combined surgery and antifungal therapy administered over at least 9 months. 4. Which of the following features of chromomycosis is true? A . This disorder is characterized by discrete nodular skin and/or subcutaneous lesions. B. The infection often follows cutaneous inoculation through minor trauma. C. Most disease is caused by three organisms: Fonsecaea, Cladosporium, and Phialophora. D. The characteristic histologic findings are sclerotic cells or “copper pennies.” E. All of the above are true. Answer: E  Chromomycosis is a disease largely seen in the tropics and usually follows minor skin trauma. The lesion(s) is usually unilateral and is typically confined to an extremity. The differential diagnosis includes sporotrichosis, blastomycosis, nocardiosis, cutaneous neoplasia, and foreign body granulomas. The lesions have a characteristic nodular, cutaneous, or subcutaneous appearance, and a histologic finding of sclerotic cells or copper pennies, indicating a melanized fungal pathogen, is characteristic of this condition. Chromomycosis is usually caused by one of the three organisms listed. 5. What is the most common cause of eumycetoma in Latin America? A . Fusarium spp B. Nocardia brasiliensis C. Pseudallescheria boydii D. Streptomyces spp E. Madurella mycetomatis Answer: C  The specific etiology of mycetoma is related, in part, to geography. In Latin America, P. boydii is the leading cause of eumycetoma. In contrast, M. mycetomatis is the leading cause of eumycetoma in other parts of the world including Sudan, sub-Saharan Africa, and India.

CHAPTER 323  Antiparasitic Therapy  

2071

323  ANTIPARASITIC THERAPY RICHARD D. PEARSON

Although substantial progress has been made in reducing the impact of protozoal and helminthic infections, including a major reduction in deaths due to malaria and the near eradication of the guinea worm, Dracunculus medinensis, parasitic diseases remain a major cause of morbidity and mortality worldwide, particularly in impoverished tropical regions. In the United States, parasitic diseases are most often encountered among immigrants and international travelers to endemic areas and among a growing number of persons with compromised immunity due to human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS), organ transplants, immunosuppressive medications, neoplasms, or other causes. A number of drugs are available to treat protozoal and helminthic diseases, but physicians practicing in industrialized countries may not be familiar with their use. This chapter focuses on their therapeutic indications, pharmacology, and major side effects. Generalizations emerge that help in organizing an otherwise vast array of information. In considering the chemotherapy of parasitic diseases, it is helpful to group infections into those caused by protozoa, single-celled organisms that can multiply by cell division, and helminths, multicellular worms with complex internal structures. Protozoa multiply in their human hosts. They can be grouped into those that live aerobically in the blood stream or tissues, and those that reside under anaerobic conditions in the gastrointestinal tract or vagina. The most important systemic protozoal infections are attributable to members of the phylum Apicomplexa, which are responsible for malaria, babesiosis, toxoplasmosis, and cryptosporidiosis, and to members of the order Kinetoplastida, which are responsible for Chagas disease, human African trypanosomiasis (sleeping sickness), and leishmaniasis. Amoeba and other types of protozoa can also cause disease. Helminths are subdivided into nematodes, or roundworms, which can be grouped into those that live in the gastrointestinal tract and those found elsewhere in the body, and platyhelminths, or flat worms, which are subdivided into cestodes, or tapeworms, and trematodes, or flukes. Detailed information about the diagnosis and treatment of specific parasitic diseases can be found in the chapters to follow, the U.S. Centers for Disease Control and Prevention (CDC) (see A-Z Index; https://www.cdc.gov/ az/a.html),1 and textbooks of tropical diseases. Many antiparasitic drugs are commercially available in the United States, whereas others can only be obtained directly from the manufacturer, special pharmacies, or the CDC Drug Service. Some are available through investigational new drug protocols. Major challenges in recent years have arisen from a shortage of some antiparasitic drugs and, in the United States, massive increases in the cost of others. A number of drugs used to treat bacterial or fungal pathogens are also effective in the treatment of parasitic diseases. They are discussed elsewhere (Chapters 271 and 315).



  PROTOZOAL DISEASES Malaria (Plasmodium species): Treatment and Prevention

A number of drugs are available for the treatment and prophylaxis of malaria. Most act on Plasmodium species within erythrocytes. The antimalarial drug of choice depends on the infecting species and the likelihood of resistance (Chapter 324). Resistance to chloroquine is now widespread among Plasmodium falciparum in most regions of the world and well documented in Plasmodium vivax in some areas. Artemisinin-based combination therapy (ACT), including the fixed drug combination artemether-lumefantrine, is the treatment of choice for acute malaria acquired in areas with chloroquine resistance. The fixed drug combination atovaquone-proguanil or the antibiotic doxycycline is commonly used for prophylaxis in travelers to areas with chloroquine-resistant Plasmodium species. Chloroquine is used for treatment and prophylaxis in

CHAPTER 323  Antiparasitic Therapy  

ABSTRACT

Parasitic infections remain a major cause of morbidity and mortality worldwide, particularly in impoverished tropical regions. It is helpful to group parasitic diseases and treatment into those caused by protozoa, single-celled organisms that multiply by cell division, and helminths, multicellular worms and flukes with complex internal structures. The most important protozoal diseases are attributable to members of the Apicomplexa that cause malaria, babesiosis, toxoplasmosis, and cryptosporidiosis and to the Kinetoplastida that are responsible for Chagas disease, human African trypanosomiasis (sleeping sickness), and leishmaniasis. Ameba and other types of protozoa also are important human pathogens. Artemisinin-based combination therapy is currently the treatment of choice worldwide for chloroquine-resistant malaria. Miltefosine has become available for the treatment of leishmaniasis and granulomatous amebic encephalitis and primary amebic meningoencephalitis. Helminths are subdivided into nematodes, or roundworms, which can be grouped into those that live in the gastrointestinal tract or elsewhere in the body, and platyhelminths, or flat worms, which are subdivided into cestodes, or tapeworms, and trematodes, or flukes. Depending on the helminth, albendazole, diethylcarbamazine, ivermectin, praziquantel, or another drug is used. Shortages of some antiparasitic drugs and substantial increases in the cost of others complicate therapy. Many antiparasitic drugs are commercially available in the United States; others can only be obtained from the Centers for Disease Control and Prevention Drug Service.

KEYWORDS

artemisinin-based combination therapy granulomatous amebic encephalitis leishmaniasis malaria miltefosine

2071.e3

2072

CHAPTER 323  Antiparasitic Therapy  

areas where Plasmodium species are sensitive. Only primaquine and tafenoquine kill hypnozoites of P. vivax and Plasmodium ovale in the liver. Country-specific recommendations for prophylaxis and treatment are provided by the CDC (www.cdc.gov/travel/) and in “CDC Health Information for International Travel 2018.”2 Artemisinins are the most rapidly acting drugs available for the treatment of malaria.3 They are sesquiterpene lactone derivatives of the wormwood plant Artemisia annua, from which qinghaosu, the Chinese herbal medication for fever, is derived. They are endoperoxide-containing compounds. In the presence of intraparasitic iron, they are converted into free radicals and other intermediates that alkylate specific malarial proteins and act rapidly to kill intraerythrocytic parasites. The artemisinins are administered with a second antimalarial drug having a different mechanism of action and longer half-life to prevent the development of resistance.4 The route of administration of artemisinins varies; some are well absorbed orally, whereas others must be administered intravenously, intramuscularly, or by suppository. Their short half-lives preclude their use for prophylaxis. Side effects in humans are common, but seldom result in discontinuation of treatment. Delayed hemolysis up to 4 weeks after treatment has been observed in persons treated with artesunate5 or some other artemisinins. Neurologic toxicity and cerebellar dysfunction have been observed in dogs receiving chronic, high-dose artemisinin therapy, but have not emerged as a problem in humans treated for acute malaria. Artemether-lumefantrine, a fixed drug combination, is used throughout the world (including the United States) to treat chloroquine-resistant malaria. It is taken with food, but grapefruit juice should be avoided. Common adverse reactions in adults are headache, anorexia, dizziness, asthenia, arthralgia, and myalgia. The most common in children are fever, cough, vomiting, anorexia, and headache. These side effects do not usually result in discontinuation of therapy. Of greater concern, lumefantrine can result in prolongation of the QT interval and is contraindicated in persons with abnormal QTc. It also inhibits CYP206 and can thereby reduce the metabolism of other medications including those that prolong the QTc. It can also decrease the effectiveness of birth control pills. Care must be taken in reviewing the recipient’s medication list for potential interactions. Artesunate is available from the CDC under an expanded access investigational new drug protocol for the intravenous treatment of severe malaria (contact the CDC Malaria Hot Line for information). Quinidine gluconate is no longer available in the United States. Artesunate is rapidly hydrolyzed to dihydroartemisinin, which is responsible for the antimalarial effect. Studies in malaria-endemic regions suggest that parenteral artesunate has a higher success rate and lower adverse event rate than quinidine. In addition to the side effects with artemether described previously, artesunate has been associated with delayed hemolysis (unrelated to glucose-6-phosphate dehydrogenase [G6PD] deficiency) up to 4 weeks after therapy. Atovaquone-proguanil, a fixed drug preparation, is used for the treatment of uncomplicated, chloroquine-resistant and -sensitive malaria, as well as prophylaxis. Atovaquone is a highly lipophilic compound with low aqueous solubility. Administration with food enhances its absorption two-fold. Plasma concentrations do not increase proportionately with dose. Atovaquone is highly protein bound with a half-life exceeding 60 hours. It undergoes extensive enterohepatic cycling and is eventually excreted unchanged in feces. Atovaquone selectively inhibits electron transport in the mitochondria of susceptible Plasmodium at the level of the cytochrome bc1 complex, which results in collapse of the mitochondrial membrane potential. It also affects pyrimidine biosynthesis, which is obligatorily coupled to electron transport in Plasmodium. Resistance develops rapidly when atovaquone is used alone to treat malaria. Atovaquone is generally well tolerated but can cause nausea, vomiting, diarrhea, rash, and pruritus. Proguanil is absorbed slowly after oral administration. The serum level falls to zero within 24 hours, so it must be administered daily. Its triazine metabolite, cycloguanil, inhibits dihydrofolate reductase in susceptible Plasmodium species. Resistance is well documented when proguanil is used alone. In addition, proguanil acts synergistically with atovaquone to collapse the mitochondrial membrane potential in susceptible Plasmodium species. The combination of atovaquone-proguanil is the best tolerated of the options for prevention of chloroquine-resistant malaria. It is begun 1 to 2 days before departure and continued during the time of exposure and for 7 days thereafter. Higher doses of atovaquone-proguanil, administered over a 3-day period, are used to treat acute, uncomplicated malaria. Potential side effects include abdominal pain, nausea, vomiting, diarrhea, headache, pruritus, and rash. Asymptomatic, transient elevations in liver enzymes have been observed with treatment doses.

Chloroquine was once the mainstay of malaria treatment and prevention, but widespread resistance among P. falciparum, and in some locations, especially Papua New Guinea and Indonesia, P. vivax, now limits its use. Chloroquine, a 4-aminoquinoline, has a bitter taste but is well absorbed from the gastrointestinal tract. Its half-life, which varies among persons, averages 4 days, thus permitting once-weekly administration for prophylaxis. Chloroquine is concentrated in the hemoglobin-containing digestive vesicles of asexual intraerythrocytic parasites. It inhibits the parasite’s heme polymerase that incorporates ferriprotoporphyrin type IX complexes, which are potentially toxic to the parasite, into insoluble, nontoxic, crystalline hemozoin. Chloroquineresistant strains of P. falciparum actively transport chloroquine out of the intraparasitic compartment. Hydroxychloroquine (Plaquenil), which is used for rheumatologic diseases, is also effective against chloroquine-sensitive Plasmodium species. Chloroquine is generally well tolerated when used at the doses recommended for the prophylaxis and treatment of malaria. Side effects include headache, nausea, vomiting, blurred vision, dizziness, and fatigue. Some persons with African ancestry experience pruritus, which responds to antihistamines. Rare side effects include depigmentation of hair, exacerbation of psoriasis, blood dyscrasias, seizures, neuropsychiatric effects, and reactions in persons with porphyria. Retinal damage has occurred in persons receiving chloroquine at high doses for the treatment of rheumatologic disorders, but it has not been documented as a problem in those taking it weekly over a period of many years for malaria prophylaxis. Cardiopulmonary collapse and death can occur after accidental overdose and in adults attempting suicide. As little as 5 g of chloroquine can be fatal unless treatment is promptly initiated with mechanical respiration, anticonvulsants, and blood pressure support. Doxycycline 100 mg taken daily by adults provides effective prophylaxis against all Plasmodium species. It is begun 1 to 2 days before exposure and continued during the time of exposure and for 4 weeks after departure from a malaria-endemic area. Doxycycline or tetracycline is also often administered with quinine for the treatment of acute chloroquine-resistant malaria, but neither doxycycline nor tetracycline acts rapidly enough to be used alone for treatment. Doxycycline is generally well tolerated, although it can cause gastrointestinal symptoms and “pill” esophagitis. To avoid the latter, it should be taken with a full glass of water, and the recipient should remain upright for an hour or more after ingestion. Other important potential side effects include photosensitivity dermatitis, Candida albicans vaginitis, and Clostridium difficile colitis. Finally, doxycycline and tetracycline should not be used in children younger than 8 years or in women who are pregnant or breast-feeding unless the potential benefits outweigh the risks. Mefloquine, a quinoline methanol compound derived from quinine, was once commonly used for prophylaxis, and occasionally for the treatment of chloroquine-resistant P. falciparum malaria. Concern about neuropsychiatric and other toxicities and the availability of better alternatives now limit its use. Mefloquine is available for oral administration only. Slowly and incompletely absorbed, it is 99% protein bound. It has a variable half-life ranging from 6 to 23 days with a mean of approximately 14 days. It is metabolized and excreted slowly through bile and feces. It is associated with nausea, dizziness, vivid dreams, fatigue, and lassitude. Less common side effects, but of greater concern, are anxiety, depression, acute psychosis, and seizures. Mefloquine is contraindicated in persons with a history of epilepsy or psychiatric disorders. It also depresses atrioventricular conduction and should not be used in persons taking β-blockers for cardiac indications. Mefloquine now carries a U.S. Food and Drug Administration (FDA) black-box warning. Although not approved for use during pregnancy or in children weighing less than 15 kg, mefloquine has been used in these situations when its potential benefits are judged to outweigh its risks. It has been reported to have no adverse effects on pregnancy outcomes (such as stillbirths and abortions) and no effects on low birth rate and prematurity.6 Primaquine, an 8-aminoquinoline, and tafenoquine, an analogue, kill the hypnozoite stage of P. vivax and P. ovale in the liver. Primaquine is used as a 14-day course at the end of treatment or prophylaxis to prevent late relapses in persons who are or may have been infected with these Plasmodium species. It is also an alternative for daily primary prophylaxis for P. vivax and other species. In that case, it is begun 1 or 2 days before exposure and continued during and for 7 days after a traveler leaves a malaria-endemic area. Tafenoquine is administered as a single 300 mg dose following treatment of acute malaria due to P. vivax and P. ovale to prevent relapse. It can also be used for primary prophylaxis in travelers; 200 mg daily for 3 days before travel, then once weekly while in the malarious region, then in the week following exit, a final dose 7 days after the previous dose in the malarious region.

CHAPTER 323  Antiparasitic Therapy  

Primaquine is absorbed orally and rapidly converted to carboxyprimaquine, which has a half-life of approximately 7 days. It is generally well tolerated, although some recipients experience abdominal cramps, epigastric distress, and nausea. Tafenoquine is better absorbed, has a half-life of approximately 15 days, and is less likely to cause gastrointestinal problems. The major concern with primaquine and tafenoquine is hemolysis in persons with glucose-6phosphate dehydrogenase (G6PD) deficiency (Chapter 152). The G6PD status of the recipient must be determined before it is administered. Rarely, primaquine causes neutropenia, methemoglobinemia, hypertension, or arrhythmias. Both drugs are contraindicated during pregnancy and in breast-feeding mothers because life-threatening hemolysis may occur if the fetus or baby is deficient in G6PD. Tafenoquine can cause severe psychiatric reactions and is contraindicated in persons with a history of psychosis or schizophrenia. Quinine sulfate, a cinchona alkaloid, is the oldest of the antimalarials. It has a very bitter taste. It is rapidly absorbed after oral administration and has a half-life of 16 to 18 hours in persons with malaria. Quinine has the poorest therapeutic to toxicity ratio of any antimalarial drug. The side effects, known collectively as cinchonism, include tinnitus, decreased hearing, headache, nausea, vomiting, dysphoria, and visual disturbances. They are dose related and reversible. Quinine has also been associated with severe hypoglycemia in persons with heavy P. falciparum infections as a result of the utilization of glucose by intraerythrocytic Plasmodium and release of insulin from the pancreas. Hypoglycemia can be treated and prevented by the intravenous administration of glucose. Rare complications with quinine include massive hemolysis in patients with heavy P. falciparum infection resulting in hemoglobinuria and renal failure (blackwater fever), cutaneous hypersensitivity reactions, agranulocytosis, and hepatitis. Quinine can cause respiratory paralysis in persons with myasthenia gravis. It stimulates uterine contractions during pregnancy and can result in abortion, but it has saved the lives of many pregnant women with P. falciparum malaria. Quinine dihydrochloride given intravenously can cause myocardial depression, peripheral vascular collapse, respiratory depression, and potentially death. Quinidine gluconate, the stereoisomer of quinine, is no longer available in the United States. It is used for the intravenous treatment of patients with complicated malaria and in those who cannot take antimalarials orally. Quinidine gluconate was once widely used for the treatment of ventricular ectopy, but it has been replaced by newer antiarrhythmic agents, which has decreased its availability in many hospitals. Side effects include prolongation of the QT interval, arrhythmias, and hypotension, particularly if it is infused too rapidly. Persons receiving intravenous quinidine should be monitored in an intensive care setting. Therapy should be switched to an oral antimalarial medication as soon as possible.  

Toxoplasmosis, Babesiosis, and Amoebic Encephalitis

Toxoplasma gondii (Chapter 328) and Babesia species (Chapter 332) are other important pathogens of the phylum Apicomplexa. Pyrimethamine and sulfadiazine are used for the treatment of toxoplasmosis. They inhibit sequential steps in the folic acid metabolic pathway. Pyrimethamine preferentially inhibits dihydrofolate reductase. It is well absorbed orally. The major side effect is macrocytic anemia, which can be prevented by the concurrent administration of leucovorin. Sulfonamides reduce the activity of dihydropteroate synthetase and the binding of p-aminobenzoic acid to it. In ocular toxoplasmosis with macular involvement, corticosteroids are used along with anti-Toxoplasma therapy to minimize the local inflammatory response. Clindamycin and pyrimethamine or atovaquone and pyrimethamine are therapeutic options in sulfonamideintolerant patients. Persons with AIDS, CD4+ counts less than 100/μL, and serologic evidence of T. gondii infection should receive prophylaxis with one of the following regimens: daily trimethoprim-sulfamethoxazole, pyrimethamine and dapsone, pyrimethamine and atovaquone, or atovaquone alone. Spiramycin, a macrolide, is used for the treatment of toxoplasmosis during pregnancy. Pyrimethamine is currently in short supply. Trimethoprim-sulfamethoxazole is a potential alternative when it is not available. Two therapeutic regimens are available for babesiosis. The combination of clindamycin and quinine is usually used in persons with severe babesiosis, but side effects are common. Long-term treatment (≥6 weeks) is often necessary for those who are immunocompromised. Atovaquone and azithromycin is better tolerated and used for less severe babesiosis (Chapter 332). Finally, the addition of miltefosine, an FDA-approved antileishmanial drug,7 to multidrug regimens has improved the likelihood of survival in persons with granulomatous amebic encephalitis due to Acanthamoeba or Balamuthia. It is now recommended as part of multidrug regimens to treat granulomatous amebic encephalitis as well as primary amebic meningoencephalitis due to Naegleria.8



2073

Intestinal and Vaginal Protozoa

Several major luminal pathogens, including Entamoeba histolytica (Chapter 331), Giardia lamblia (Chapter 330), and Trichomonas vaginalis (Chapter 332), live in anaerobic conditions in the intestine or vagina. The 5-nitroimidazoles, metronidazole and tinidazole, are first-line therapy for them, but refractory giardiasis and trichomoniasis are documented. Tinidazole has more favorable pharmacodynamics and is generally better tolerated. Because neither metronidazole nor tinidazole reliably eradicates cysts of E. histolytica in the colon, either paromomycin, a poorly absorbed aminoglycoside, or iodoquinol, both of which are active in the lumen of the bowel, is administered as well. Either one of these drugs or diloxanide furoate is used alone to treat persons with asymptomatic cyst excretion. Giardiasis can also be treated with nitazoxanide. It is well tolerated, and a liquid formulation is available for children. Nitazoxanide is the only drug effective in the treatment of cryptosporidiosis (Chapter 329). It can cure immunocompetent persons, but not those with AIDS. Trimethoprimsulfamethoxazole, which inhibits successive steps in the folic acid pathway, is the drug of choice for Cystoisospora (Isospora) belli and Cyclospora cayetanensis. Ciprofloxacin is a second-line alternative for the former. Either tetracycline, metronidazole, or nitazoxanide can be used to treat symptomatic Balantidium coli infection. Metronidazole, a 5-nitroimidazole, is rapidly absorbed after oral administration and has a half-life of 8 hours. More than half of each dose is metabolized in the liver. The metabolites and remaining parent drug are excreted in urine. Metronidazole is activated by reduction of its 5-nitro group through a sequence of intermediate steps involving microbial electron transport proteins of low redox potential. It is concentrated in susceptible anaerobic organisms and serves as an electron sink. Nausea, vomiting, diarrhea, and a metallic taste are often associated with the use of metronidazole. They are less common with the lower doses recommended for the treatment of giardiasis than with the higher doses used for amebiasis. Other untoward effects include headache, dizziness, vertigo, and numbness. Potentially severe disulfiram-like reactions occur in patients who ingest alcohol while taking metronidazole. Tinidazole, another 5-nitroimidazole, has a similar mechanism of action and spectrum of activity as metronidazole, but more favorable pharmacodynamics, and it is generally better tolerated. It has been used widely around the world for the treatment of giardiasis, intestinal amebiasis, and trichomoniasis. In comparison with metronidazole, it has a longer half-life, a shorter and less complicated dosing regimen, and fewer gastrointestinal side effects. It, too, can cause severe disulfiram-like reactions after alcohol ingestion. Nitazoxanide, a nitrothiazolyl-salicylamide derivative thiazolide, has a broad spectrum of activity against protozoa and helminths. It is formulated as a liquid for children. Nitazoxanide is well absorbed orally and hydrolyzed to its active metabolite tizoxanide, which undergoes conjugation to tizoxanide glucuronide. The parent compound is not detectable in serum. Maximum concentrations of the metabolites are observed in 1 to 4 hours. They are excreted in urine and bile. Tizoxanide is highly protein bound. Although its antiparasitic mechanism of action is uncertain, nitazoxanide inhibits pyruvate : ferredoxin oxidoreductase–dependent electron transport reactions essential for the metabolism of susceptible anaerobic organisms. It is very well tolerated in children and adults.

  AFRICAN TRYPANOSOMIAIS (SLEEPING

SICKNESS), CHAGAS DISEASE, AND LEISHMANIASIS

The Kinetoplastida, Trypanosoma brucei rhodesiense and Trypanosoma brucei gambiense, which cause human African trypanosomiasis (sleeping sickness); Trypanosoma cruzi, the etiology of Chagas disease; and Leishmania species, which cause cutaneous, mucosal, and visceral leishmaniasis often pose substantial therapeutic challenges.9  

African Trypanosomiasis (Sleeping Sickness)

Suramin, pentamidine, eflornithine, and melarsoprol are used for the treatment of human African trypanosomiasis (Chapter 325). Suramin or pentamidine is recommended for the hemolymphatic stage of T. brucei rhodesiense and T. brucei gambiense, respectively. Both drugs are associated with potentially serious side effects, and suramin can cause severe reactions in persons coinfected with Onchocerca volvulus. Eflornithine, which is much better tolerated, is effective against both the hemolymphatic and central nervous system (CNS) stages of T. brucei gambiense infection. It does not have activity against

2074

CHAPTER 323  Antiparasitic Therapy  

T. brucei rhodesiense. Unfortunately, eflornithine is costly and supplies are limited. It is available in the United States from the CDC. It can be used alone, but eflornithine is often co-administered with nifurtimox for CNS T. brucei gambiense infection. Melarsoprol is used for persons with CNS involvement due to T. brucei rhodesiense and T. brucei gambiense when eflornithine is not available. Melarsoprol can result in severe toxicity, including life-threatening encephalopathic reactions.  

Chagas Disease

Benznidazole or nifurtimox orally is used to treat T. cruzi infections (Chapter 326). Benznidazole has been the drug of choice in endemic areas of Latin America and nifurtimox in the United States.10 Both drugs are available from the CDC. They lower mortality and shorten the duration of acute Chagas disease. Treatment is recommended for persons with recent infection as well as asymptomatic children and adults through middle age with indeterminatestage T. cruzi infection. The percentage of those who are parasitologically cured by treatment has been debated. Neither benznidazole nor nifurtimox can reverse the cardiac or gastrointestinal manifestations of chronic Chagas disease. Side effects are common with both drugs and increase in frequency and severity with age. Benznidazole is administered daily for 1 month and is associated with allergic dermatitis, peripheral neuropathy, insomnia, and gastrointestinal symptoms, including anorexia and weight loss. Nifurtimox is administered daily for 3 months and is associated with anorexia, nausea, vomiting, weight loss, headache, dizziness or vertigo, paresthesias, weakness, and polyneuropathy.  

Leishmaniasis (Cutaneous, Mucosal, and Visceral)

Treatment of cutaneous leishmaniasis depends on the size, number, complexity, and location of skin lesions, their cosmetic impact, the infecting Leishmania species, and its propensity to cause mucosal disease.11 Uncomplicated lesions acquired in Europe, Africa, and Asia, where mucosal involvement is rare, can be followed without therapy if they appear to be healing spontaneously. Lesiondirected treatment options include cryotherapy; heat therapy, which requires a specialized delivery system; or intralesional injection of stibogluconate sodium, which is not available in the United States. Topical therapy with direct application of paromomycin is another option. An ointment containing 15% paromomycin and 12% methylbenzethonium chloride in white paraffin developed in Israel has been used. A U.S. Army formulation of topical paromomycin also appears promising. Parenteral or oral antileishmanial therapy is used for persons with complicated cutaneous leishmaniasis and for those who are or may be infected with Leishmania braziliensis or related New World species potentially associated with mucosal leishmaniasis.12 Miltefosine is FDA approved for the oral treatment of cutaneous leishmaniasis due to Leishmania braziliensis. The efficacy with other Leishmania species varies. Liposomal amphotericin B and amphotericin B deoxycholate are effective, but more toxic, parenteral alternatives. Stibogluconate sodium and meglumine antimoniate were once widely used, but their toxicity, the requirement for parenteral administration, and increasing resistance have limited their use. Fluconazole and other imidazole antifungals have activity against some Leishmania species. Mucosal leishmaniasis is less responsive to treatment than cutaneous leishmaniasis, and relapses are common.13 Therapeutic options include liposomal amphotericin B, miltefosine, amphotericin B deoxycholate, stibogluconate sodium, or meglumine antimoniate. Visceral leishmaniasis is a potentially fatal disease. The drug of choice is liposomal amphotericin B. Miltefosine and amphotericin B deoxycholate are alternatives. Sodium stibogluconate or meglumine antimoniate are used in some geographic areas where resistance has not yet emerged. Liposomal amphotericin B was the first drug approved by the FDA for treatment of visceral leishmaniasis (Chapter 327) in the United States. Liposomes deliver amphotericin to macrophages where leishmania reside. Liposomal amphotericin B is better tolerated than amphotericin B deoxycholate, which is an alternative. Other lipid-associated amphotericin B preparations appear to be effective, but they have been less extensively studied and are not FDA approved for this indication. Miltefosine, an alkylphospholipid and phosphocholine analog, initially developed as an antineoplastic drug, is FDA approved for the treatment of visceral leishmaniasis and cutaneous leishmaniasis due to L. braziliensis. A major advantage is oral administration.14 The pharmacokinetics are characterized by a long elimination half-life and extensive drug accumulation. The mechanism of action is uncertain, but miltefosine induces apoptosis-like changes in the parasite. Side effects are frequent, but typically mild to

moderate. Dose-dependent gastrointestinal toxicity can result in nausea, vomiting, and diarrhea, which tend to decrease with continued administration of the drug. Elevations in liver enzymes and creatinine are common but usually transient. Miltefosine is embryotoxic and is thus contraindicated during pregnancy. Contraceptive cover is mandatory in women of childbearing years during and for 4 months after therapy. Relapses have been reported after the treatment of visceral leishmaniasis. They are frequent in persons with concurrent AIDS. Stibogluconate sodium and meglumine antimoniate, pentavalent antimony drugs, were used to treat visceral leishmaniasis in the past, but resistance is now common among Leishmania donovani isolates in India, adjacent countries, and some other areas of the world. Pentavalent antimony drugs require parenteral administration. Side effects increase with age and include gastrointestinal symptoms, pancreatitis, myalgias, headache, malaise, elevated liver enzyme levels, and occasionally bone marrow suppression. Nonspecific ST-T wave changes are common. Sudden death has been reported in older persons and in those receiving more than the recommended dose.

  HELMINTHIC DISEASES

The treatment of helminthic diseases is discussed based on their taxonomy and the site of infection. A number of anthelminthic drugs are available. Several have multiple indications. Information about their pharmacology is provided at the end of this section.  

Intestinal Roundworms (Nematodes)

Intestinal roundworms (Nematodes) (Chapter 335) are among the world’s most prevalent parasites. Ascaris lumbricoides, the hookworms Ancylostoma duodenale and Necator americanus, and Trichuris trichiura each infect millions of people worldwide. Strongyloides stercoralis is less common, but can progress to life-threatening hyperinfection in immunocompromised persons. Many residents of impoverished areas harbor more than one of these species. A number of drugs are available to treat them for Enterobius vermicularis (pinworm) and other intestinal roundworms. Albendazole, mebendazole, pyrantel pamoate, and ivermectin have replaced older anthelmintics that were more toxic, such as piperazine and thiabendazole, or less effective. Albendazole has a broad spectrum of activity. It is active against A. lumbricoides, the hookworms, and T. trichiura. Administered as a single 400-mg dose, it has been widely used in mass treatment programs in areas where these soil-transmitted roundworms are prevalent.15 However, reinfection is common, and treatment is often repeated at 3- to 4-month intervals. The CDC recommends presumptive treatment of immigrants from endemic regions with a single dose of albendazole before arrival in the United States. Daily doses of albendazole for 3 days are recommended for persons with heavy T. trichiura infection. Albendazole, 400 mg twice daily for 7 days, is used as an alternative to ivermectin for the treatment of Strongyloides stercoralis infection. Failures can occur with either drug, and they are often used together for longer periods to treat those with disseminated hyperinfection. Albendazole is effective against Enterobius vermicularis, the pinworm, in a single dose that is repeated in 2 weeks. It can be used for cutaneous larva migrans, which is caused by migrating stages of Ancylostoma braziliense and other intestinal roundworms of animals. It is the drug of choice for trichinosis, and it has been used for the treatment of Trichostrongylus species and Capillaria philippinensis. There has been a massive increase in the cost of albendazole in the United States. Mebendazole, 100 mg orally twice daily for 3 days, has a similar spectrum of activity as albendazole against A. lumbricoides, hookworms, and T. trichiura. In this regimen, it is more effective than a single dose of albendazole and is considered the treatment of choice for heavy T. trichiura infections. A single 500-mg dose of mebendazole has been used in mass treatment programs. Mebendazole is effective in treating pinworms when given at 100 mg orally in one dose followed by a second dose after 2 weeks. It is an alternative to albendazole for the treatment of trichinosis. Mebendazole is poorly absorbed and ineffective against S. stercoralis. Pyrantel pamoate is a relatively safe, poorly absorbed, over-the-counter drug with activity against A. lumbricoides, hookworms, and pinworm, but it is not effective against T. trichiura or S. stercoralis. When used for pinworm infection, it is administered as an oral suspension at a dose of 11 mg/kg (to a maximum of 1 g), which is repeated after 2 weeks. Ivermectin at an oral dose of 200 µg/kg daily for 2 days is considered the treatment of choice for uncomplicated S. stercoralis. In immunocompromised persons with hyperinfection, daily doses are given until stool and sputum exams are negative for larvae for 2 weeks. It is also effective against cutaneous

CHAPTER 323  Antiparasitic Therapy  

larva migrans and A. lumbricoides but not hookworms. It is an alternative to mebendazole for the treatment of T. trichiura.  

Systemic Roundworms (Nematodes)

Diethylcarbamazine is the drug of choice for the treatment of lymphatic filarial infections (Chapter 335) caused by Wuchereria bancrofti, Brugia malayi, and Brugia timori, as well as for tropical pulmonary eosinophilia. It promotes the host’s killing of microfilariae of these species and also damages or kills adult worms. Inflammatory side effects are common. They are due in part to the release of lipopolysaccharide from endosymbiotic Wolbachia bacteria within dying filaria. Wolbachia are necessary for filarial development and are a potential drug target. Long-term therapy with doxycycline results in their elimination and has been used for therapy. Diethylcarbamazine is also used for Loa loa infections in persons with acceptably low levels of microfilaremia (5% or >200,000 parasites/µL). Cerebral malaria,7 the most common severe complication in children, is

generally defined as altered consciousness in the setting of P. falciparum malaria. Seizures are common, and deep coma, abnormal posturing, focal neurologic findings, and abnormal respiratory patterns can be seen. The mortality rate is 15% to 25%, with about 10% with persistence of neurologic sequelae, but many patients do remarkably well with appropriate therapy. Severe anemia is a common presentation, particularly in young children. Transfusions are avoided when possible but play a key role in the management of those with severe anemia. Respiratory failure is caused by noncardiogenic pulmonary edema and is more common in adults than in children. Mechanical ventilation can be life saving, if available. Acute renal failure is also more common in adults and generally due to hypoperfusion and acute tubular necrosis; hemofiltration and hemodialysis are valuable, if available. Blackwater fever, including intravascular hemolysis and hemoglobinuria, has an uncertain etiology but can be caused by quinine. Hepatic dysfunction, including jaundice, can be seen; jaundice may also be caused by hemolysis. Splenic enlargement is common, and splenic rupture can occur. Hypoglycemia is common because of glucose consumption by parasites, increased demand, impaired gluconeogenesis, and quinine-induced insulin secretion; blood glucose levels should be observed closely, with glucose replacement as needed. Metabolic acidosis, particularly lactic acidosis, is common; the value of specific therapies for acidosis or aggressive fluid resuscitation is uncertain. Electrolyte derangements can be seen. Coagulopathy, caused by consumption of clotting factors, and marked thrombocytopenia, caused by increased platelet turnover, can lead to significant bleeding. Bacterial infection and sepsis can coexist with malaria; presumptive use of antibiotics is warranted when signs of sepsis are noted.

Complications of Non–P. falciparum Malaria

The large majority of infections with non–P. falciparum parasites are uncomplicated both in endemic areas and in nonimmune travelers. Nonetheless, P. vivax infection is common in many areas, and studies from a number of sites in Asia and Oceania have shown that it makes up about a quarter of children hospitalized with severe malaria, with a mortality rate of about 1%. Important features of severe P. vivax malaria include severe anemia and respiratory distress. All malarias, but in particular P. vivax infection, can be complicated by splenic rupture. Chronic malaria infections can be complicated by hyperreactive malarial splenomegaly, with massive splenomegaly and findings of hypersplenism. Chronic infections can also lead to the nephrotic syndrome, particularly with P. malariae infection.  

DIAGNOSIS

Clinical Features

In individuals with febrile illness and malaria risk, it is essential to make a diagnosis promptly and important to distinguish the different species that infect humans because management differs according to the infecting species. Malaria is the most common cause of febrile illness in many areas, and with limited diagnostic capabilities, it is frequently diagnosed based only on presentation with a febrile illness. However, formal diagnosis is preferred. In travelers returning from endemic areas with fever, historical details can aid in the diagnosis. Malaria is most likely in individuals who failed to use measures to prevent infection and in travelers to the most highly endemic areas, such as rural sub-Saharan Africa. P. falciparum malaria generally has a fairly short incubation period in nonimmune individuals, so it presents within 1 to 2 months of return in more than 90% of travelers. Infection with the other malarial species can present in a number of months and uncommonly more than 1 year after exposure.

Blood Smears

The standard means of diagnosis in malaria-endemic areas is by thick blood smear. In this procedure, 1 drop of blood is allowed to dry on a slide, erythrocytes are lysed, and parasites are then stained with Giemsa. Parasites are easily identified by trained microscopists, and parasite density can be estimated on the basis of counts relative to those of leukocytes. However, thick blood smears do not allow identification of erythrocyte morphology, which is helpful in species diagnosis, and are difficult for those with limited training. Giemsastained thin blood smears offer an improved means of characterizing parasite morphology, but the process is much less efficient than for thick smears. Thus, thick smears are the standard means of diagnosis in highly endemic areas, and thin smears are preferred where malaria is uncommon and where laboratory personnel have ample time to examine multiple microscopic fields. It is important to distinguish infecting species of malaria parasites. In P. falciparum infection, generally only ring-form asexual parasites are seen. Trophozoites of

CHAPTER 324 Malaria  

P. vivax and P. ovale are present in enlarged (and ovoid in the case of P. ovale) erythrocytes that contain inclusions known as Schüffner dots. Intraerythrocytic P. malariae and P. knowlesi trophozoites are often elongate in shape. Sexual stage gametocytes (which have a characteristic oblong shape in P. falciparum) are also seen on blood smears; most treatments do not eradicate gametocytes, so persistence of these forms for a few weeks is not a sign of treatment failure.

CHLOROQUINE-RESISTANT PLASMODIUM FALCIPARUM, RESISTANT PLASMODIUM VIVAX, OR SPECIES UNIDENTIFIED

Antigen Detection

or

An important new means of malaria diagnosis is antigen detection. Multiple simple tests are now available, incorporating colorimetric detection of one or two antigens in an assay that requires limited training and only a few minutes.8 The most used assays in Africa use histidine-rich protein-2 (HRP2), a protein that is abundant and long-lived but expressed only by P. falciparum. Other assays identify plasmodial lactate dehydrogenase and aldolase, which are produced by all human malarial species. Some tests use two antigens to separately identify P. falciparum and all-species plasmodial infection. Rapid diagnostic tests have become a standard component of many malaria control programs. One test is approved in the United States. However, with many different tests available around the world, standardization is not optimal, and the specific role of rapid antigen tests for the diagnosis of malaria in different epidemiologic settings is not yet established. In addition, there is increasing evidence of P. falciparum lacking HRP2 in some areas, including parts of South America and other regions, leading to concern that HRP2-based diagnostic tests may miss some cases of falciparum malaria.

Other Diagnostic Tests

Serologic tests are available to identify prior malaria infection, but responses develop slowly and persist for an extended period, so these tests have limited value for acute diagnosis. Malaria parasites can be identified with molecular amplification assays using primers encoding genus- and species-specific sequences. These tests are highly sensitive and convenient for research purposes because they can be conducted on DNA extracted from blood spotted onto filter paper in field settings. Quantitative polymerase chain reaction assays are most sensitive, but loop-mediated isothermal amplification and related assays are highly sensitive and more practical for many field settings. Molecular assays are not practical for routine diagnosis, however, because of the time required, the uncertain significance of a positive result in endemic areas where low-level parasitemia may be clinically insignificant, and cost and logistical requirements.

TREATMENT  The treatment of falciparum malaria has been challenged by drug resistance for many years.9,10 Artemisinin-based combination therapy (ACT) is recommended in nearly all countries endemic for P. falciparum malaria (Table 324-1).11 Non–P. falciparum malaria is still generally treated with chloroquine, although chloroquine resistance is increasing in P. vivax; resistant vivax malaria can be treated with other drugs used to treat falciparum malaria.

Chloroquine and Other Aminoquinolines

Chloroquine has been widely used to treat malaria for more than 70 years. It remains the treatment of choice for non–P. falciparum malaria and P. falciparum malaria in the few areas where resistance has not been seen (Central America and the Caribbean) and is generally rapidly effective and well tolerated. For P. vivax and P. ovale infections, primaquine or tafenoquine must also be given to eradicate dormant liver forms and thereby prevent subsequent relapse. Chloroquine remains effective as weekly chemoprophylaxis to prevent malaria in areas without resistance. Amodiaquine and piperaquine probably share mechanisms of action with chloroquine, but they are routinely active against chloroquine-resistant parasites because of increased potency compared with chloroquine and some differences in mechanisms of resistance; each is now a component of a leading ACT (Table 324-2). Amodiaquine is somewhat less well tolerated than other aminoquinolines. It is generally safe with short-term use but can cause rare serious hepatic and bone marrow toxicity, especially with chronic use, and so it is not recommended for chemoprophylaxis. The ACT dihydroartemisinin-piperaquine has shown excellent efficacy in most regions, but frequent failures, mediated by resistance to both components of the regimen, have recently been seen in Southeast Asia.

Mefloquine and Lumefantrine

Mefloquine offers effective therapy and chemoprophylaxis for most chloroquine-resistant strains of P. falciparum and for other species. Resistance to mefloquine is uncommon but has been seen in parts of Southeast Asia,

2079

TABLE 324-1 TREATMENT OF MALARIA*

UNCOMPLICATED DISEASE Coartem (artemether 20 mg, lumefantrine 120 mg) Malarone (atovaquone 250 mg, proguanil 100 mg)

4 tablets orally twice daily for 3 days

4 tablets daily for 3 days

or Quinine

650 mg quinine sulfate 3 times daily for 3-7 days

plus Doxycycline

100 mg twice daily for 7 days

or plus Clindamycin

600 mg twice daily for 7 days

or Mefloquine

750 mg followed by 500 mg in 6-8 hours; can also be given as a single 1250-mg dose, although this is less well tolerated than the divided dose

COMPLICATED P. FALCIPARUM MALARIA OR UNABLE TO TOLERATE ORAL MEDICATIONS† IV artesunate‡

2.4 mg/kg every 12 hr on day 1; then daily for 2 additional days

or IV quinidine gluconate§,‖

10 mg/kg over 1 to 2 hr; then 0.02 mg/kg/min or 15 mg/kg over 4 hr; then 7.5 mg/kg over 4 hr every 8 hr

or IV quinine dihydrochloride§,‖

20 mg/kg over 4 hr; then 10 mg/kg every 8 hr

or IM artemether‖

3.2 mg/kg IM; then 1.6 mg/kg/day

CHLOROQUINE-SUSCEPTIBLE P. FALCIPARUM AND OTHER SPECIES Chloroquine phosphate

1 g, followed by 500 mg at 6, 24, and 48 hr or 1 g at 0 and 24 hr; then 0.5 g at 48 hr

plus (for P. vivax and P. ovale only) Primaquine¶

30-mg base (52.6 mg primaquine phosphate) daily for 14 days

or Tafenoquine¶

300 mg single dose

*Dosages refer to salts unless indicated and are for adults. For pediatric dosing and full Centers for Disease Control and Prevention (CDC) recommendations, see http://www.cdc.gov/malaria/pdf/ treatmenttable.pdf. † IV regimens should be given until the patient can tolerate oral agents and then followed by a course of oral therapy (doxycycline, clindamycin, or full treatment courses of other drugs, as listed) when patients can tolerate this. ‡ Available in the United States on an investigational basis through the CDC. § Cardiac monitoring should be in place during IV administration of quinidine or quinine. ‖ Not available in the United States. ¶ Use primaquine only after demonstrating normal levels of glucose-6-phosphate dehydrogenase. IM = intramuscular; IV = intravenous.

with failures of the ACT artesunate-mefloquine. Mefloquine is one of three drugs recommended for chemoprophylaxis against P. falciparum by the Centers for Disease Control and Prevention (CDC). Tolerability of chemoprophylactic and especially treatment doses of mefloquine is often limited by neurologic and gastrointestinal (GI) toxicity. Lumefantrine is only used in combination and offers effective therapy with artemether as Coartem, the most widely used ACT to treat falciparum malaria.

Quinine and Quinidine

Quinine has been used to treat malaria for hundreds of years. It offers rapid action against all species, with limited known resistance except in Southeast Asia, where failures against P. falciparum malaria are fairly common. Quinine can be used to treat uncomplicated malaria, but GI and other nonspecific

2080

CHAPTER 324 Malaria  

TABLE 324-2 RECOMMENDATIONS FOR THE TREATMENT OF PLASMODIUM FALCIPARUM MALARIA IN DEVELOPING COUNTRIES* REGIMEN

NOTES

Artemether–lumefantrine (Coartem, Riamet)

First-line therapy in many countries; FDA approved

Artesunate–amodiaquine (ASAQ)

First-line therapy in many African countries

Artesunate–mefloquine

Standard therapy in parts of Southeast Asia

Dihydroartemisinin–piperaquine

Recently limited by resistance in SE Asia, but highly effective elsewhere

*Recommendations modified from World Health Organization. Guidelines for the Treatment of Malaria. Geneva: World Health Organization; 2015. FDA = U.S. Food and Drug Administration.

toxicities lead to difficulty in tolerating a full 7-day treatment course. This problem is circumvented by combining a 3-day course of quinine with other agents. For severe disease, intravenous (IV) quinine has been a standard therapy for many years but offers inferior efficacy compared to IV artesunate. In the United States, IV quinine and quinidine are now unavailable, and IV artesunate is the standard treatment for severe malaria.

Primaquine and Tafenoquine

Primaquine (typically 30 mg daily for 14 days) A1  and tafenoquine (a single 300  mg dose) A2  A3  are related aminoquinolines that are the only available drugs to eradicate dormant liver forms of P. vivax and P. ovale. These forms can lead to relapses after therapy with chloroquine and other agents. Primaquine and tafenoquine are also alternative agents for chemoprophylaxis against P. falciparum and other species. Both drugs can cause hemolysis or methemoglobinemia (Chapter 149) in individuals with deficiency in glucose-6-phosphate dehydrogenase (G6PD) (Chapter 152). Testing for deficiency should be performed before use of the drugs. As a new strategy to lower transmission, a single low dose of primaquine in conjunction with artemisin-based combination therapy is a welltolerated treatment for uncomplicated falciparum malaria and can lower transmission to mosquitoes in subjects with normal G6PD levels.12 ,

Inhibitors of Folate Metabolism

Inhibitors of dihydrofolate reductase and dihydropteroate synthase are used in fixed-dose combination regimens for the treatment and prevention of malaria. For treatment, sulfadoxine–pyrimethamine (Fansidar) was heavily used to treat uncomplicated P. falciparum malaria, but resistance has increased markedly in most endemic areas. The dihydrofolate reductase inhibitor proguanil is combined with atovaquone in Malarone (see later). For chemoprophylaxis, sulfadoxine–pyrimethamine is no longer recommended because of drug resistance and rare life-threatening dermatologic toxicities. However, less frequent dosing in intermittent preventive therapy regimens has been well tolerated and has decreased malaria in high-risk African groups, particularly pregnant women and young children, although efficacy is limited by drug resistance. Seasonal malaria chemoprevention with monthly combined sulfadoxine–pyrimethamine and amodiaquine during the transmission season is now recommended for malaria control in areas of Africa with highly seasonal transmission and limited drug resistance. Daily trimethoprim–sulfamethoxazole, a common prophylactic to prevent opportunistic infections in individuals with HIV infection, has offered some protection against malaria in Africa.

Artemisinins

Artemisinin, the active component of an herbal medicine from China, and a number of its analogues offer rapid elimination of circulating malaria parasites and activity against gametocytes to limit disease transmission. The drugs are all short-acting, leading to frequent recrudescences of infection after short-course monotherapy. For this reason and to limit selection of resistance, artemisinins are now used in combination with longer acting drugs to treat malaria in 3-day regimens. A number of these combinations have become the standard therapies for P. falciparum malaria in most endemic countries (see Table 324-2). Leading regimens are fixed-dose combinations of artemether–lumefantrine (Coartem; the only ACT approved in the United States), artesunate–amodiaquine, artesunate–mefloquine, and dihydroartemisinin–piperaquine. All these regimens have demonstrated excellent treatment efficacy in Africa in children A4-A6  and pregnant women A7 ; different regimens have had optimal efficacy in different regions. ACTs have also shown excellent efficacy for the treatment of vivax malaria. A8  Other ACTs that are not currently recommended by the WHO but are available for the treatment of malaria in some countries and have also shown good efficacy include artesunate– pyronaridine, A9  arterolane–piperaquine, A10  artemisinin–piperaquine, and artemisinin–naphthoquine. A11  Resistance to artemisinins is a recent concern, with

evidence for prolonged times to parasite clearance in Southeast Asia suggestive of diminished drug responsiveness of P. falciparum.13 Of great concern, treatment efficacies of ACTs have recently waned in parts of Southeast Asia, most notably for dihydroartemisinin–piperaquine, with resistance to both components of the regimen leading to high clinical failure rates. A12  Artemisinins also have a key role for the treatment of severe malaria. IV artesunate has been shown to be superior to quinine for the treatment of severe malaria in a mostly adult population in Asia and in African children, A13  notably with survival advantages over quinine in both populations. For settings with limited infrastructure, intramuscular artemether A14  and intramuscular A15  or intrarectal artesunate have also shown excellent efficacy. IV artesunate should now be considered the first-line therapy for severe malaria in the United States, where it is not routinely available, but can be readily obtained from the CDC for this indication. Artemisinins are generally very well tolerated, with minimal toxicity, although delayed hemolysis can occur after therapy, in particular with IV artesunate.14

Atovaquone–Proguanil (Malarone)

This fixed-dose combination of a dihydrofolate reductase inhibitor and atovaquone, which has a unique antimalarial mechanism, has excellent efficacy against most P. falciparum infections. It is approved for both treatment and chemoprophylaxis of P. falciparum malaria and other species in the United States, where it is now widely used for both indications. Malarone offers excellent efficacy with minimal toxicity. Adverse effects may include GI symptoms, elevations in liver enzymes, headache, and rash. Widespread use in developing countries is limited by high cost and concerns about resistance because resistance to each component drug is readily selected, but it has been suggested that atovaquone-resistant parasites cannot complete development in mosquitoes, preventing transmission of these parasites.15

Antibiotics

A number of antibacterials are slow-acting antimalarials. Tetracyclines and clindamycin should not be used alone to treat malaria but are combined with quinine to allow a shorter duration of therapy with that drug. In addition, doxycycline is effective in chemoprophylaxis of most P. falciparum malaria and is recommended for this purpose by the CDC, in particular for travelers to regions of Southeast Asia with high-level resistance to other drugs.

Treatment of Severe Malaria

Severe malaria is a medical emergency and requires parenteral therapy. With appropriate prompt therapy and supportive care, rapid recoveries may be seen even in very ill individuals. Standard therapy for severe malaria is IV artesunate, which must be obtained from the CDC. If acquisition is delayed, short-term therapy with an oral agent is appropriate until IV artesunate is available. Appropriate care of severe malaria includes close nursing care; maintenance of fluids, electrolytes, and glucose; respiratory and hemodynamic support; and consideration of blood transfusions, anticonvulsants, antibiotics for bacterial infections, and hemodialysis or hemofiltration. Aggressive fluid resuscitation, blood transfusion for moderate anemia, exchange transfusion, and specific treatment of acidosis are of uncertain value. After the acute illness, IV artesunate should be followed by oral longer acting drugs, typically a full course of an oral ACT, Malarone, mefloquine, or quinine plus doxycycline or clindamycin.



PREVENTION

Key malaria control interventions in malaria-endemic regions are control of mosquito vectors by indoor residual spraying of insecticides; personal protection against mosquito bites with insecticide-impregnated bed nets; routine use of artemisinin-based combination therapies, which offer prompt control of malaria infections and activity against gametocytes to limit transmission to mosquitoes; and selected use of intermittent preventive therapies to decrease malarial incidence in high-risk groups. A long-lasting piperonyl butoxide-treated insecticidal net and pyrethroid indoor residual spraying interventions have shown improved control of malaria transmission compared with standard long-lasting insecticidal nets. A16  A17  Ivermectin, which is used to treat some helminth infections, is also mosquitocidal and can reduce malaria transmission when provided in addition to standard malaria therapy. A18  No vaccine to prevent malaria is yet available, but extensive research on potential vaccines is under way. RTS,S, which is based on an immunogenic sporozoite antigen, is the most advanced vaccine candidate. Multiple clinical trials showed protection in children immunized with RTS,S, with about 25% to 50% protection against malaria in the year after immunization, but lower levels of protection in very young children, in areas of highest malaria exposure, and over longer periods of time. A19-A21  Based on these results, a new concept is seasonal malaria immunization, with provision of short-acting vaccines ,

TABLE 324-3 CHEMOPROPHYLAXIS OF MALARIA* AREAS WITH CHLOROQUINE-RESISTANT PLASMODIUM FALCIPARUM Malarone 1 tablet (250 mg artesunate/100 mg proguanil) daily Mefloquine 250 mg weekly Doxycycline 100 mg daily Primaquine† 30 mg daily during exposure (chemoprophylaxis) or 30 mg daily for 14 days (terminal prophylaxis against P. vivax and P. ovale) Tafenoquine† 200 mg daily for 3 days prior to travel, then 200 mg weekly during travel AREAS WITHOUT CHLOROQUINE-RESISTANT P. FALCIPARUM Chloroquine phosphate 500 mg weekly *Recommendations may change on the basis of drug resistance patterns. For additional details and pediatric dosing, see Centers for Disease Control and Prevention guidelines (http://www.cdc.gov). Begin 1 to 2 weeks before travel for mefloquine and 2 days before for doxycycline, Malarone, and primaquine; continue for 4 weeks after leaving the endemic area (1 week for Malarone and tafenoquine; 2 weeks for primaquine). All doses refer to salts unless indicated. † Use primaquine only after demonstrating normal levels of glucose-6-phosphate dehydrogenase.

during the high transmission season.16 Controlled human malaria infection was prevented in 9 out of 9 (100%) malaria-naïve, healthy adult volunteers who ten weeks earlier were immunized with radiation-attenuated Plasmodium falciparum sporozoites inoculated by mosquitoes.17 Other approaches under study include vaccines containing erythrocytic, liver-stage, and sexual-stage antigens.

Preventive Measures for Travelers to Malaria-Endemic Regions

It is important for nonimmune travelers (Chapter 270) to endemic areas to be protected against potentially lethal malaria. Travelers should decrease exposure to night-biting anopheline mosquitoes by use of insecticide repellents and sleeping in rooms that are screened or equipped with insecticideimpregnated bed nets. Standard advice for travelers to endemic areas is also to use low doses of preventive drugs chosen on the basis of the resistance profile of the particular region. Chloroquine is still recommended for malariaendemic regions of Central America and the Caribbean. For nearly all other areas, the CDC recommends use of daily Malarone, weekly mefloquine, or daily doxycycline; details of dosing vary, but it is important to continue therapy after return from travel to eliminate parasites as they emerge from the liver (Table 324-3). Primaquine and tafenoquine (FDA approved in 2018) offer alternative chemoprevention regimens. For areas with high risk of P. vivax malaria, some authorities recommend a full treatment course of primaquine after travel to eliminate dormant liver stages. For all chemoprophylaxis, it is important to appreciate that no mosquito avoidance methods or drug regimens are fully protective, so consideration of malaria as a cause of fever in returned travelers is essential (Chapter 270).  

PROGNOSIS

Patients with malaria caused by P. vivax, P. ovale, or P. malariae generally respond well to chloroquine and make an uneventful recovery. Chloroquine-resistance is increasing with P. vivax from many areas; failures of initial treatment are usually not dangerous but should be followed by treatment with another regimen, such as an ACT, Malarone, or mefloquine. Patients with P. falciparum malaria also generally respond well to prompt therapy as long as the disease is not overly advanced at presentation. The mortality rate in those with uncomplicated P. falciparum malaria is about 0.1%. Key contributing factors to most deaths from P. falciparum malaria are probably a delay between the appearance of symptoms and presentation for definitive therapy and the use of suboptimal therapies. Presentation with high-level parasitemia (>200,000 parasites/µL or >5% parasitemia) or signs of severe malaria are predictive of a poor outcome. However, with aggressive support, even individuals with severe disease can often experience complete recoveries.

  Grade A References A1. Lacerda MVG, Llanos-Cuentas A, Krudsood S, et al. Single-dose tafenoquine to prevent relapse of Plasmodium vivax malaria. N Engl J Med. 2019;380:215-228. A2. Llanos-Cuentas A, Lacerda MVG, Hien TT, et al. Tafenoquine versus primaquine to prevent relapse of Plasmodium vivax malaria. N Engl J Med. 2019;380:229-241.

A3. Graves PM, Choi L, Gelband H, et al. Primaquine or other 8-aminoquinolines for reducing Plasmodium falciparum transmission. Cochrane Database Syst Rev. 2018;2:CD008152. A4. Four Artemisinin-Based Combinations (4ABC) Study Group. A head-to-head comparison of four artemisinin-based combinations for treating uncomplicated malaria in African children: a randomized trial. PLoS Med. 2011;8:1-16. A5. Yeka A, Kigozi R, Conrad MD, et al. Artesunate/amodiaquine versus artemether/lumefantrine for the treatment of uncomplicated malaria in Uganda: a randomized trial. J Infect Dis. 2016;213: 1134-1142. A6. Sirima SB, Ogutu B, Lusingu JPA, et al. Comparison of artesunate-mefloquine and artemetherlumefantrine fixed-dose combinations for treatment of uncomplicated Plasmodium falciparum malaria in children younger than 5 years in sub-Saharan Africa: a randomised, multicentre, phase 4 trial. Lancet Infect Dis. 2016;16:1123-1133. A7. PREGACT study group. Four artemisinin-based treatments in African pregnant women with malaria. N Engl J Med. 2016;374:913-927. A8. Gogtay N, Kannan S, Thatte UM, et al. Artemisinin-based combination therapy for treating uncomplicated Plasmodium vivax malaria. Cochrane Database Syst Rev. 2013;10:CD008492. A9. Pryce J, Hine P. Pyronaridine-artesunate for treating uncomplicated Plasmodium falciparum malaria. Cochrane Database Syst Rev. 2019;1:CD006404. A10. Toure OA, Valecha N, Tshefu AK, et al. A phase 3, double blind, randomized study of arterolane maleate-piperaquine phosphate vs artemether-lumefantrine for falciparum malaria in adolescent and adult patients in Asia and Africa. Clin Infect Dis. 2016;62:964-971. A11. Laman M, Moore BR, Benjamin JM, et al. Artemisinin-naphthoquine versus artemether-lumefantrine for uncomplicated malaria in Papua New Guinean children: an open-label randomized trial. PLoS Med. 2014;11:1-18. A12. Amaratunga C, Lim P, Suon S, et al. Dihydroartemisinin-piperaquine resistance in Plasmodium falciparum malaria in Cambodia: a multisite prospective cohort study. Lancet Infect Dis. 2016;16: 357-365. A13. Dondorp AM, Fanello CI, Hendriksen IC, et al. Artesunate versus quinine in the treatment of severe falciparum malaria in African children (AQUAMAT): an open-label, randomised trial. Lancet. 2010;376:1647-1657. A14. Esu E, Effa EE, Opie ON, et al. Artemether for severe malaria. Cochrane Database Syst Rev. 2014;9:CD010678. A15. Kremsner PG, Adegnika AA, Hounkpatin AB, et al. Intramuscular artesunate for severe malaria in African children: a multicenter randomized controlled trial. PLoS Med. 2016;13:1-22. A16. Tiono AB, Ouedraogo A, Ouattara D, et al. Efficacy of Olyset Duo, a bednet containing pyriproxyfen and permethrin, versus a permethrin-only net against clinical malaria in an area with highly pyrethroidresistant vectors in rural Burkina Faso: a cluster-randomised controlled trial. Lancet. 2018;392: 569-580. A17. Protopopoff N, Mosha JF, Lukole E, et al. Effectiveness of long-lasting piperonyl butoxide-treated insecticidal net and indoor residual spray interventions, separately and together, against malaria transmitted by pyrethroid-resistant mosquitoes: a cluster, randomised controlled, two-by-two factorial design trial. Lancet. 2018;391:1577-1588. A18. Ouédraogo AL, Bastiaens GJ, Tiono AB, et al. Efficacy and safety of the mosquitocidal drug ivermectin to prevent malaria transmission after treatment: a double-blind, randomized, clinical trial. Clin Infect Dis. 2015;60:357-365. A19. Olotu A, Fegan G, Wambua J, et al. Four-year efficacy of RTS,S/AS01E and its interaction with malaria exposure. N Engl J Med. 2013;368:1111-1120. A20. RTS,S Clinical Trials Partnership. Efficacy and safety of RTS,S/AS01 malaria vaccine with or without a booster dose in infants and children in Africa: final results of a phase 3, individually randomised, controlled trial. Lancet. 2015;386:31-45. A21. Olotu A, Fegan G, Wambua J, et al. Seven-year efficacy of RTS,S/AS01 malaria vaccine among young African children. N Engl J Med. 2016;374:2519-2529.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 324 Malaria  

GENERAL REFERENCES 1. Ashley EA, Pyae Phyo A, Woodrow CJ. Malaria. Lancet. 2018;391:1608-1621. 2. Cowman AF, Healer J, Marapana D, et al. Malaria: biology and disease. Cell. 2016;167:610-624. 3. Karnad DR, Nor MBM, Richards GA, et al. Intensive care of severe malaria: report from the task force on tropical diseases by the World Federation of Societies of Intensive and Critical Care Medicine. J Crit Care. 2018;43:356-360. 4. World Health Organization. World Malaria Report 2018. Geneva, Switzerland: WHO; 2018. 5. Gething PW, Casey DC, Weiss DJ, et al. Mapping Plasmodium falciparum mortality in Africa between 1990 and 2015. N Engl J Med. 2016;375:2435-2445. 6. Newby G, Bennett A, Larson E, et al. The path to eradication: a progress report on the malariaeliminating countries. Lancet. 2016;387:1775-1784. 7. Luzolo AL, Ngoyi DM. Cerebral malaria. Brain Res Bull. 2019;145:53-58. 8. Mukkala AN, Kwan J, Lau R, et al. An update on malaria rapid diagnostic tests. Curr Infect Dis Rep. 2018;20:1-8. 9. Plewes K, Leopold SJ, Kingston HWF, et al. Malaria: what’s new in the management of malaria? Infect Dis Clin North Am. 2019;33:39-60.

2081.e1

10. Rout S, Mahapatra RK. Plasmodium falciparum: multidrug resistance. Chem Biol Drug Des. 2019; 93:737-759. 11. World Health Organization. Guidelines for the treatment of malaria. Geneva, Switzerland: WHO; 2015. 12. Dicko A, Brown JM, Diawara H, et al. Primaquine to reduce transmission of Plasmodium falciparum malaria in Mali: a single-blind, dose-ranging, adaptive randomized phase 2 trial. Lancet Infect Dis. 2016;16:674-684. 13. Haldar K, Bhattacharjee S, Safeukui I. Drug resistance in plasmodium. Nat Rev Microbiol. 2018;16:156-170. 14. Rolling T, Agbenyega T, Krishna S, et al. Delayed haemolysis after artesunate treatment of severe malaria—review of the literature and perspective. Travel Med Infect Dis. 2015;13:143-149. 15. Goodman CD, Siregar JE, Mollard V, et al. Parasites resistant to the antimalarial atovaquone fail to transmit by mosquitoes. Science. 2016;352:349-353. 16. Greenwood B, Dicko A, Sagara I, et al. Seasonal vaccination against malaria: a potential use for an imperfect malaria vaccine. Malar J. 2017;16:1-5. 17. Mordmüller B, Surat G, Lagler H, et al. Sterile protection against human malaria by chemoattenuated PfSPZ vaccine. Nature. 2017;542:445-449.

2081.e2

CHAPTER 324 Malaria  

REVIEW QUESTIONS 1. A young American adult consults with you before travel to Kenya. Appropriate regimens for the prevention of malaria in travelers from a developed country to areas with chloroquine-resistant P. falciparum malaria include: A . Malarone, mefloquine, or artemisinin B. Malarone, mefloquine, or doxycycline C. Coartem, Malarone, or doxycycline D. quinine, artemisinin, or Fansidar Answer: B  These three drugs are all listed as first-line regimens for the prevention of malaria in travelers. Artemisinin and artemisinin-based combination therapy (Coartem) are not appropriate because of the short half-lives of artemisinins. Quinine is not appropriate because of a short half-life and poor tolerability. 2. Reasons for the predilection for P. falciparum to cause severe malaria include all of the following except: A . Cytoadherence of P. falciparum–infected erythrocytes to vascular endothelium B. Infection of erythrocytes of all ages C. A high prevalence of resistance to available antimalarial drugs D. Increasing pathogenicity with increasing age of the patient Answer: D  Major contributors to the pathogenesis of P. falciparum malaria include cytoadherence of P. falciparum–infected erythrocytes to vascular endothelium, infection of erythrocytes of all ages, and high prevalence of resistance to available antimalarial drugs. P. falciparum malaria does not generally show increasing pathogenicity with increasing age. Rather, the disease is typically most severe in young children because of their lack of prior exposure to the pathogen and thus lack of antimalarial immunity and possibly other factors. 3. The region with the greatest morbidity and mortality from malaria in the world is: A . Africa B. Southeast Asia C. South America D. Oceania E. The Middle East Answer: A  The greatest morbidity and mortality from malaria is in Africa, likely because of many factors, including a preponderance of P. falciparum malaria, particularly competent anopheline mosquito vectors, and limited medical infrastructure.

4. The region with the most highly resistant malaria parasites in the world is: A . Africa B. Southeast Asia C. South America D. Oceania E. Middle East Answer: B  The greatest prevalence of parasites resistant to available antimalarial drugs is in Southeast Asia. Resistance to chloroquine, amodiaquine, piperaquine, antifolates, mefloquine, quinine, and artemisinins is seen in this area, threatening the efficacies of key artemisinin-based combination therapies. 5. A child from Ghana is admitted with fever, altered consciousness, acute renal failure, and 7% parasitemia with P. falciparum. Severe malaria should be treated with: A . Oral artemisinin-based combination therapy B. Malarone or mefloquine C. Quinine or quinidine D. Intravenous artesunate, if available, and as a back-up intravenous quinine or quinidine Answer: D  The international standard-of-care for the treatment of severe malaria is intravenous artesunate. Oral therapy should be avoided in very ill individuals. Intravenous quinine or quinidine is highly effective, but intravenous artesunate was superior to intravenous quinine in randomized trials of Asians and Africans with severe malaria.

CHAPTER 325  African Sleeping Sickness  

2081

325  AFRICAN SLEEPING SICKNESS WILLIAM A. PETRI, JR.



DEFINITION

Human African trypanosomiasis, commonly known as sleeping sickness, is a vector-borne parasitic disease transmitted to humans and animals by the bite of the tsetse fly (genus Glossina). Infection is caused by protozoa of the genus Trypanosoma and species brucei.1 In humans, there are two forms of illness caused by two distinct subspecies that are morphologically identical but differ in their geographic range and clinical presentations. Trypanosoma brucei gambiense is typically found in west and central Africa and Trypanosoma brucei rhodesiense in east Africa. T. b. gambiense has a more chronic course, and T. b. rhodesiense causes a rapid disease course; both have late stages marked by meningoencephalitis, resulting in coma and death if untreated. There is a third subspecies, Trypanosoma brucei, known to cause chronic infection called nagana in cattle; however, humans are not susceptible to this organism. Chagas disease, or American trypanosomiasis, is covered in Chapter 326.

CHAPTER 325  African Sleeping Sickness  

ABSTRACT

Human African trypanosomiasis, commonly known as sleeping sickness, is a vector-borne parasitic disease transmitted to humans and animals by the bite of the tsetse fly. Infection is caused by protozoa of the genus Trypanosoma and species brucei. In humans, there are two forms of illness caused by two distinct subspecies that are morphologically identical but differ in their geographic range and clinical presentations. Trypanosoma brucei gambiense is typically found in west and central Africa and Trypanosoma brucei rhodesiense in east Africa. T. b. gambiense has a more chronic course, and T. b. rhodesiense causes a rapid disease course; both have late stages marked by meningoencephalitis, resulting in coma and death if untreated. Treatment depends on the infecting organism and the stage of disease.

KEYWORDS

African sleeping sickness Trypanosoma brucei tsetse fly pentamidine nifurtimox eflornithine meningoencephalitis

2081.e3

2082

CHAPTER 325  African Sleeping Sickness  

The Pathogen

As an extracellular microbe, the parasite must evade immune clearance to establish a persistent infection. The surface of the trypanosome is covered by a dense, homogeneous coat of variant surface glycoproteins (VSGs), which are immunodominant. Each individual trypanosome expresses only one VSG at a time but possesses more than 1000 different silent copies of the VSG gene, and switching to a new VSG occurs at a frequency of about 1 × 10−6 parasites. Hence, the trypanosomes expressing a given VSG will eventually be cleared by the host antibody response, but any individual trypanosome that has switched to a new VSG will evade immune clearance, resulting in a new peak of parasitemia. Recombination between VSG alleles ensures a virtually limitless repertoire of new VSGs; therefore antibody-mediated parasite eradication is impossible.  

T. b. gambiense Countries reporting no cases Countries reporting fewer than 100 cases T. b. rhodesiense Countries reporting between 100 and 1000 cases Countries reporting more than 1000 cases

EPIDEMIOLOGY

Human African trypanosomiasis is relatively less of a problem now than it was in the prior century, partly because of cyclic epidemics in the past, as well as recent increases in public health efforts for control.2 However, it remains a looming threat to an estimated 60 million people living in tsetse fly–inhabited areas among 36 countries of sub-Saharan Africa and causes significant morbidity, accounting for 1.5 million disability-adjusted life-years in Africa as a whole.3 The first epidemic occurred in the Congo basin and Uganda between 1896 and 1906, driven in part by natural disasters that decimated populations of livestock and regional droughts, as well as changes in population distribution influenced by colonialism. The second epidemic occurred in numerous endemic countries in the 1920s; control was achieved after major efforts to systematically screen and treat individuals followed by extensive vector control programs, including bush clearing and insecticide spraying. These acts were nearly successful in halting transmission by the 1960s; however, the independence of many African nations around this time hindered the sustainability of prevention and control programs. The incidence began to rise by the 1970s and surged a decade later, leading to the third epidemic of the 20th century. However, in the past 15 years, increased access to at-risk populations for diagnosis and treatment has led to a 68% reduction in the incidence of reported cases. Despite increased surveillance efforts during the past 15 years, there are still at-risk areas that lack active monitoring programs, and many people are thought to die of human African trypanosomiasis without an accurate diagnosis. The World Health Organization (WHO) reported an annual incidence of 12,000 cases in 2007, a dramatic decrease from the estimated incidence of 300,000 cases a decade earlier. It is estimated that 300,000 to 500,000 people are infected, contributing to approximately 100,000 deaths each year. More than 90% of reported cases are due to T. b. gambiense, the majority being from the Democratic Republic of the Congo. The geographic distribution includes areas where the vector, parasite, reservoir hosts, and human hosts cohabit (Fig. 325-1). In general, these include focal areas on the African continent within 15 degrees North and 15 degrees South latitude, with a predilection for rural areas. Humans at greatest risk for infection are those who rely on animal husbandry, agriculture, fishing, and hunting for their livelihoods. Often, disease is concentrated among foci of rural areas, having significant socioeconomic impact on affected villages. With a few exceptions, this infection is never found in urban areas. There have been fewer than 50 cases reported annually outside of Africa, usually the result of travel by North Americans or Europeans to African game reserves. T. b. gambiense is transmitted by tsetse flies from the Glossina palpalis group, the annotated genome sequence of which has been reported. Tsetse flies are typically found along riverbanks among wooded areas in the more tropical regions of central and West Africa. The reservoir hosts for T. b. gambiense and the focus of public health campaigns are mainly humans. Because of this, T. b. gambiense is not generally considered a zoonotic disease; the role of the animal reservoir for this organism remains undetermined, although natural infections have been reported in domestic animals such as dogs, sheep, cattle, and pigs. T. b. rhodesiense is transmitted by tsetse flies belonging to the Glossina morsitans group, which are commonly found among woodland and savannah areas of east and central Africa. T. b. rhodesiense is a zoonotic disease with numerous wild and domestic species of animals acting as reservoirs. With this infection, the animal reservoir serves an important role in the cycle, sustaining parasite transmission and human infections. Domestic species, especially cattle, have the potential to drive outbreaks and, not surprisingly, have served as the focus of successful prevention campaigns.

FIGURE 325-1.  Map of human African trypanosomiasis. These 36 sub-Saharan African countries are considered endemic for human African trypanosomiasis. Shaded areas represent the reported incidence from 1997 to 2006. The black line roughly represents the dividing line for the two parasites, although some overlap may already occur.



PATHOBIOLOGY

After the bite of a tsetse fly carrying metacyclic trypomastigotes, a local reaction (chancre) may form at the inoculation site. This symptom is seen with T. b. rhodesiense infection and is more frequently observed in travelers but is rarely seen with T. b. gambiense infection. Parasites subsequently disseminate into the blood and lymphatic systems in what is considered stage I of the disease. Spread of the parasites into the central nervous system (CNS) defines stage II of the disease, which is invariably fatal if untreated. The parasite appears to remain extracellular throughout the course of infection. Peaks and waves of parasitemia occur during stage I disease and result in the classic symptom of intermittent fever. These bouts of fever correspond to a type 1 inflammatory response (Chapter 42), in which classically activated macrophages produce high levels of tumor necrosis factor (TNF) and nitric oxide. This helps in the control of parasitemia but also contributes to tissue damage. Type 2 inflammatory responses with high interleukin-10 production may subsequently occur, which limit TNF and nitric oxide production after initial parasitemia has been controlled. Antibody responses are directed to VSGs and other trypanosome antigens (e.g., antigens from lysed parasites), but autoantibodies are also produced. Generalized febrile episodes are observed together with lymphadenopathy and myocardial and pericardial inflammation. Cardiac involvement is typically more severe with T. b. rhodesiense infection. Anemia, thrombocytopenia, disseminated intravascular coagulation, and renal disease may also be observed. In stage II of the disease, parasites cross the blood-brain barrier and invade the CNS. Acute meningoencephalitis develops, with a variety of inflammatory cells infiltrating the brain, including macrophages, lymphocytes, plasma cells, Mott cells (plasma cells with spherical cytoplasmic inclusions composed of immunoglobulin), and morular cells (plasmacytoid cells with transparent hyaline vesicles in the cytoplasm). These inflammatory cells are found in the meninges, which become thickened, as well as in the perivascular spaces and neuropil. Edema, hemorrhage, and granulomatous lesions are often present; thrombosis and neuronal degeneration may also be observed. Genetic resistance to African sleeping sickness is due to a common (in individuals of African descent) polymorphism in the APOL1 gene, a component of high-density lipoprotein. This gene product mediates serum resistance to the parasite. An untoward consequence of this polymorphism is that individuals with two copies of the variant allele (15% of African Americans) are at increased risk of chronic kidney disease.  

CLINICAL MANIFESTATIONS

The clinical manifestations of human African trypanosomiasis differ on the basis of the infecting organism. T. b. gambiense is more of a chronic disease,

CHAPTER 325  African Sleeping Sickness  

2083

with an estimated average duration of 3 years; infection with T. b. rhodesiense progresses much more rapidly, leading to coma and death within weeks to months. However, infection with T. b. gambiense has been known to also cause a rapid decline. Other similarities exist, including an early hemolymphatic stage (stage I) and a late stage characterized by prominent CNS disease including meningoencephalitis (stage II).4

West African Sleeping Sickness

Infection starts after the bite of a fly infected with T. b. gambiense. At the site of inoculation, a painful, indurated, and erythematous trypanosomal chancre may develop 1 to 2 weeks after the bite and resolves spontaneously after several weeks. On occasion, the chancre may ulcerate. However, these features are rarely seen at the time of clinical presentation and sometimes are not known to have occurred on history; thus many develop disseminated disease without awareness of a localized infection. The hemolymphatic stage, when parasites disseminate throughout the body, may not be manifested clinically until weeks or months after the initial bite. Typical symptoms include intermittent spiking fevers, occasionally accompanied by headaches and malaise. These features may persist for weeks or months because of the cyclic nature of parasitemia and antibody production against the various antigens sequentially expressed by the parasite. Lymphadenopathy (Chapter 159) is a common finding in West African sleeping sickness. Whereas regional lymphadenopathy may develop after the initial bite, generalized lymphadenopathy around the head and neck is often found as chronic illness develops. Enlargement of nodes in the posterior cervical triangle, commonly known as Winterbottom sign, is the classic finding, but other cervical and supraclavicular nodes may also be involved. Affected nodes are discrete, movable, rubbery nodes that are nontender to palpation; over time, they may become more indurated because of fibrosis. Other symptoms reported include pruritus, occasionally accompanied by rash, arthralgias, and periarticular swelling, as well as by transient edema of the extremities and face. Less common symptoms include features consistent with neuroendocrine dysfunction, including loss of libido and impotence, amenorrhea and infertility, alopecia, and gynecomastia. Signs of disease include hepatomegaly and splenomegaly; cardiac dysfunction, including tachycardia; electrocardiographic abnormalities such as prolonged QTc intervals and repolarization changes; and, less commonly, pericarditis or myocarditis. Hemolytic anemia and derangements in liver function test results may occur. Months or even years after the initial infection, stage II disease develops and is characterized by headaches, daytime somnolence, and neuropsychiatric features. Behavioral changes, including irritability, confusion, inability to concentrate, and lassitude, are among the first signs of CNS disease; psychosis has also been described. Neurologic findings are numerous and include a variety of motor and sensory disturbances, including extrapyramidal features, dysesthesias, and visual impairment. These symptoms have given the infections its common name of sleeping sickness and are manifested by daytime somnolence and nocturnal irritability. Late-stage disease is marked by cerebral edema and meningoencephalitis. Progressively, a loss of neurologic function can lead to paralysis, and many may succumb to aspiration pneumonias or malnutrition; otherwise, coma leads to death in the absence of treatment.

East African Sleeping Sickness

Compared with Gambian sleeping sickness, East African sleeping sickness is more rapidly progressive. The infective bite is more frequently associated with the development of a chancre, although some studies report this in only 20% of patients. An incubation period lasting days to weeks is needed before symptoms are demonstrated. Initial symptoms include severe intermittent fevers that may resemble those found with malaria. Lymphadenopathy is not as common with this infecting organism; the Winterbottom sign is typically absent. Skin changes are more prominent, and rashes in the early stage of infection are particularly common in expatriates with the infection. In addition, cardiac manifestations are more commonly and clinically relevant; tachycardia, arrhythmias, myocarditis, and congestive heart failure have been documented and may be severe enough to cause death before the development of severe CNS disease. CNS disease mirrors that of West African sleeping sickness, but the onset occurs earlier and the rate of deterioration is more rapid. Hematologic abnormalities include anemia,5 thrombocytopenia, and disseminated intravascular coagulation. Death may ensue after weeks or months without treatment.  

DIAGNOSIS

Epidemiologic clues and clinical findings may together suggest the diagnosis of human African trypanosomiasis, but definitive diagnosis relies on demonstration

FIGURE 325-2.  Trypanosoma rhodesiense in peripheral blood. It has a nucleus, poste­ rior kinetoplast, undulating membrane, and flagellum (×1500).

of the parasite. In the early stage of disease, light microscopy and Giemsa stain may be used to visualize the highly motile parasites directly from fresh specimens of fluid expressed from chancres or lymph node aspirates. Peripheral blood smears, including Giemsa-stained thick and thin smears and bone marrow aspirates, have been successful. Blood smears have increased sensitivity when they are performed during stage I disease, when parasitemia is high (Fig. 325-2); the threshold for visualization of parasites by thick smear is approximately 5000 parasites/mL. Performance is superior with T. b. rhodesiense infection, given the higher parasite load. If the initial smear analysis is negative, subsequent examinations should be pursued. Concentration techniques, including buffy coat examination, should be used when technically feasible. Culture of any of these fluids may yield higher sensitivity than smear preparations. Cerebrospinal fluid (CSF) must also be analyzed to determine the most appropriate treatment. Abnormalities in CSF analysis often start with increased cell counts and progress to include an elevated opening pressure and total protein levels with an increased polyclonal immunoglobulin M (IgM). Stage II disease is defined by the presence of more than 5 white blood cells/µL, the presence of trypanosomes, or an elevated total protein (>370 mg/L) in the CSF. Newer diagnostic methods, including polymerase chain reaction analysis of CSF and a latex-agglutination test for CSF IgM, hold promise but require validation to determine outcomes after treatment of patients with positive test results. Allocation of resources to this neglected tropical disease in recent years has allowed slow but exciting progress in the field of human African trypanosomiasis diagnostics. The genomes of T. brucei species have been mapped, and molecular assays have been developed that can distinguish among species of human African trypanosomiasis with a single polymerase chain reaction test. Novel uses of mass spectrometry have been developed that use proteomic signature analysis to identify specific fingerprints of human African trypanosomiasis with the host. In contrast to these highly technical and expensive methods, other tests suitable for the field are being validated, including the dot-ELISA test, which would be able to provide information on stage of disease. Serology tests are also available for diagnosis of T. b. gambiense infection. The card agglutination test for trypanosomiasis with T. b. gambiense (CATT) is commonly used by screening programs; the sensitivity ranges from 87% to 98%, depending on the population under study, and the specificity may be as high as 95%. No serology tests are available for T. b. rhodesiense infection. Rapid serologic tests for infection with T. b. gambiense include human African trypanosomiasis Sero-Strop (which uses a dipstick method) and the human African trypanosomiasis Sero + K SeT test (which uses a lateral flow device) for testing blood or plasma, respectively. The WHO Collaborating Centres for human African trypanosomiasis are important resources for clinical diagnostic tests.

TREATMENT  Very few drugs have been available for treatment of human African trypanosomiasis, and those that have been commonly used are quite toxic (Table 325-1). Treatment depends on the infecting organism and the stage of disease. Distribution of drugs is through the WHO in Geneva. For stage I disease with T. b. gambiense infection, pentamidine is the drug of choice. A1  The standard regimen consists of daily parenteral administration for 1 week; however, studies are ongoing to determine the efficacy of shortening therapy to three doses. Suramin is also used for stage I disease with T. b. rhodesiense. This medication is complex to mix and to administer. The drug must be administered in a slow intravenous infusion periodically during 3 weeks. Although anaphylaxis is rare

TABLE 325-1 DRUGS USED TO TREAT HUMAN AFRICAN TRYPANOSOMIASIS DRUG

CLASS

STAGE

ROUTE

ADULT DOSE

ADVERSE EFFECTS

TRYPANOSOMA BRUCEI GAMBIENSE Pentamidine

Aromatic diamidine

I

IM or IV

4 mg/kg daily for 7 days

Pain, GI symptoms, hypoglycemia or hyperglycemia, electrolyte abnormalities, leukopenia, thrombocytopenia

Eflornithine

Ornithine carboxylase inhibitor

II

IV

100 mg/kg q6hr for 14 days

GI symptoms, bone marrow toxicity, seizures

Nifurtimox 15 mg/kg per day orally in three doses × 10 days; eflornithine 400 mg/kg per day intravenously in two 2-h infusions (each dose diluted in 250 mL of water for injection) × 7 days

Eflornithine plus nifurtimox TRYPANOSOMA BRUCEI RHODESIENSE Suramin

Polysulfonated naphthylamine derivative of urea

I

IV

1 g IV on days 1, 3, 7, 14, and 21 (after test dose of 100-200 mg)

Anaphylaxis, nephrotoxicity, fever, rash, pruritus, arthralgias, reversible peripheral neuropathy, and bone marrow toxicity

Melarsoprol

Trivalent arsenical

II

IV

1.2 mg/kg q8hr for 3 consecutive days each week for 3 weeks (maximum daily dose, 180 mg)

Encephalopathic syndromes, peripheral neuropathy, paralysis, cardiac dysrhythmias, GI symptoms, rash, pruritus, thrombophlebitis

GI = gastrointestinal; IM = intramuscular; IV = intravenous.

(approximately 1 in 20,000 patients), a test dose is recommended before full treatment is initiated. A number of side effects require close monitoring, the most important being nephrotoxicity. Urinalysis is recommended before each dose, and the drug should be discontinued if proteinuria persists and casts are seen in the urine sediment. The first choice for treatment of stage II T. b. gambiense is the combination of nifurtimox and eflornithine. A2  Nifurtimox is administered orally, whereas eflornithine requires frequent intravenous administration. Melarsoprol remains the only agent available for the treatment of stage II T. b. rhodesiense infection. It is a highly effective but impressively toxic drug used for stage II disease with either organism. The side effects are numerous, but the most important is the life-threatening encephalopathy that may develop from the arsenic (highly fatal) or as an inflammatory reaction. Concomitant steroid use is helpful in reducing the risk of encephalopathy and death without compromising the efficacy of melarsoprol. Gradual increase in the first round of treatment between 2 and 3.6 mg/kg in divided doses three times daily during 3 days has also been shown to reduce the risks of encephalopathy. Melarsoprol resistance is widespread. The mechanism of resistance to this drug has been pinpointed to mutations affecting an aquaglyceroporin (AQP2), a parasite solute and drug transporter.6 Tafenoquine, which is an oral antimalarial drug belonging to the 8-aminoquinoline family, has in vitro activity against T. brucei. Whether it will become a clinically useful alternative is unknown at this time.7 Another potential new alternative, oral pafuramidine (100 mg twice a day for 10 days), can provide an overall cure rate of about 90% at 12 months but has been discontinued because of late renal toxicity. Few therapeutic options are available to treat the late stage of African trypanosomiasis. However, oral fexinidazole,8 given once daily orally, has been found to be effective and safe for the treatment of T. b. gambiense infection compared with nifurtimox eflornithine combination therapy in late-stage infection. A3 



PREVENTION

To date, there is no vaccine against human African trypanosomiasis. The mainstays of prevention include active case finding with early treatment and vector control.9 Given the nature of each infection, active case surveillance is more suitable for infection with T. b. gambiense, but vector control is more effective for prevention of infection with T. b. rhodesiense. The goal of active case finding is to identify infected individuals who may still be in the asymptomatic stage or early stage. This approach, which is more suited for disease from T. b. gambiense, typically consists of a screening examination for lymphadenopathy followed by a CATT test. If results are positive for both, the patient should undergo further evaluation with lymph node aspirate and blood testing. If trypanosomes are found, the patient should be treated. Vector control measures include tsetse fly traps and insecticide-impregnated screens; the traps are easily maintained by locals, but the screens require regular retreatment and thus are more labor intensive and costly to maintain. Mass spraying, once a successful method until the 1960s, is no longer practiced. However, if an epidemic does occur, ground or aerial spraying combined with disruption of

the animal reservoir’s habitat may be the most effective method of achieving rapid vector control. Travelers to endemic areas should be aware of this disease and should use basic protective measures. Protective clothing of at least medium weight should be worn; neutral colors are most effective because flies are attracted to bright and dark colors. Tsetse flies are attracted to moving vehicles but rest in the shade or bushes. Although use of insect repellant is prudent for other vectorborne illnesses that may be endemic in these areas, it has not been proven to substantially reduce the risk of tsetse fly bites. There is no recommended chemoprophylaxis for travelers.

  Grade A References A1. Pohlig G, Bernhard SC, Blum J, et al. Efficacy and safety of pafuramidine versus pentamidine maleate for treatment of first stage sleeping sickness in a randomized, comparator-controlled, international Phase 3 clinical trial. PLoS Negl Trop Dis. 2016;10:1-17. A2. Kansiime F, Adibaku S, Wamboga C, et al. A multicentre, randomised, non-inferiority clinical trial comparing a nifurtimox-eflornithine combination to standard eflornithine monotherapy for late stage Trypanosoma brucei gambiense human African trypanosomiasis in Uganda. Parasit Vectors. 2018;11:1-11. A3. Mesu VKBK, Kalonji WM, Bardonneau C, et al. Oral fexinidazole for late-stage African Trypanasome brucci gambiense trypanasomiasis: a pivotal multicentre, randomised, non-inferiority trial. Lancet. 2018;391:144-154.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 325  African Sleeping Sickness  

GENERAL REFERENCES 1. Büscher P, Cecchi G, Jamonneau V, Priotto G. Human African trypanosomiasis. Lancet. 2017;390: 2397-2409. 2. Aksoy S, Buscher P, Lehane M, et al. Human African trypanosomiasis control: achievements and challenges. PLoS Negl Trop Dis. 2017;11:1-6. 3. Bottieau E, Clerinx J. Human African trypanosomiasis: progress and stagnation. Infect Dis Clin North Am. 2019;33:61-77. 4. Mogk S, Bosselmann CM, Mudogo CN, et al. African trypanosomes and brain infection—an unsolved question. Biol Rev Camb Philos Soc. 2017;92:1675-1687. 5. Stijlemans B, De Baetselier P, Magez A, et al. African trypanosomias—associated anemia: the contribution of the interplay between parasites and the mononuclear phagocyte system. Front Immunol. 2018;9:1-15.

2084.e1

6. Fairlamb AH, Horn D. Melarsoprol resistance in African trypanosomiasis. Trends Parasitol. 2018;34:481-492. 7. Carvalho L, Martinez-Garcia M, Perez-Victoria I, et al. The oral antimalarial drug tafenoquine shows activity against Trypanosoma brucei. Antimicrob Agents Chemother. 2015;59:6151-6160. 8. Scarim CB, Jornada DH, Machado MGM, et al. Thiazole, thio and semicarbazone derivatives against tropical infective diseases: Chagas disease, human African trypanosomiasis (HAT), leishmaniasis, and malaria. Eur J Med Chem. 2019;162:378-395. 9. Keating J, Yukich JO, Sutherland CS, et al. Human African trypanosomiasis prevention, treatment and control costs: a systematic review. Acta Trop. 2015;150:4-13.

2084.e2

CHAPTER 325  African Sleeping Sickness  

REVIEW QUESTIONS 1. Risk factors for sleeping sickness include which of the following? A . Tsetse fly bite B. Residence in urban areas of North Africa C. Rural areas of sub-Saharan Africa D. A and C E. A and B Answer: D  African sleeping sickness is rare in urban areas or in North Africa. This vector-borne parasitic disease is transmitted to humans and animals by the bite of the tsetse fly. 2. Clinical presentation of sleeping sickness includes which of the following? A . Chancre at site of bite B. Fever and headache C. Lymphadenopathy D. A and C E. A, B, and C Answer: E  Sleeping sickness can present acutely with a chancre at the site of the bite, fever, headache, and lymphadenopathy. 3. Diagnostic test(s) of sleeping sickness is/are: A . Visualizing the parasite in chancre, node, or blood B. Polymerase chain reaction (PCR) on cerebrospinal fluid C. Serology D. A and C E. A, B, and C Answer: E  Diagnosis of African sleeping sickness can be accomplished by visualizing the parasite, PCR, and serology.

4. What drug is used to treat patients with stage I sleeping sickness caused by T. b. gambiense? A . Pentamidine B. Melarsoprol C. Eflornithine D. A and B E. A, B, and C Answer: A  Therapies with demonstrated effectiveness include melarsoprol, pentamidine, and eflornithine. For stage I disease, pentamidine is the treatment of choice. 5. East African sleeping sickness is: A . More rapidly progressive than West African B. Characterized by severe fevers C. Commonly associated with lymphadenopathy D. A and B E. A, B, and C Answer: D  East African sleeping sickness tends to be more severe and rapidly progressive, with intermittent fevers that resemble those seen in malaria. Unlike West African sleeping sickness, in which lymphadenopathy (often generalized) is common, lymphadenopathy is not typically associated with lymphadenopathy in East African sleeping sickness.

2084

CHAPTER 326  Chagas Disease  

326  CHAGAS DISEASE LOUIS V. KIRCHHOFF



DEFINITION

Chagas disease, or American trypanosomiasis, is caused by the protozoan parasite Trypanosoma cruzi. The terms Chagas disease, American trypanosomiasis, and T. cruzi infection are synonyms.

The Pathogen

Several dozen species are included in the genus Trypanosoma, but only the African trypanosome Trypanosoma brucei (subspecies T. b. gambiense [West

CHAPTER 326  Chagas Disease  

ABSTRACT

Chagas disease, or American trypanosomiasis, is a zoonotic infection caused by the single-cell protozoan parasite, Trypanosoma cruzi. T. cruzi is spread among its various mammalian hosts, which include domestic and wild animals as well as humans, by blood-sucking triatomine insects. T. cruzi can also be transmitted from mother to fetus, through blood products and organs obtained from infected donors, by ingestion of contaminated food and drink, and in laboratory accidents. Infection with T. cruzi in humans is lifelong. Chagas disease is endemic Mexico, as well as in all the countries of Central and South America but not in any of the Caribbean islands. All told, about 6 million persons are chronically infected with T. cruzi, roughly 30,000 new infections occur each year, and about 14,000 persons die of the illness annually. An estimated 240,000 to 300,000 immigrants with Chagas disease currently live in the United States. 10 to 30% of persons who are chronically infected with T. cruzi ultimately develop cardiac and/or gastrointestinal symptoms related to the persistent presence of the parasite. The diagnosis of acute or congenital Chagas disease is made by parasitologic methods, typically direct microscopic examination of blood or PCR assays. Chronic T. cruzi infection is diagnosed serologically and many accurate ELISAs, IFAs, and chemiluminescence tests are available commercially for this purpose. Nifurtimox and benznidazole are the only two drugs available for treating T. cruzi infections and the latter is preferred by most experts. Parasitological cure rates for these drugs are high for acute and congenital infections but unfortunately very low in persons with long-standing infections. There are no convincing data from randomized controlled trials indicating that treatment of chronically infected persons with either drug significantly delays pathogenesis or affects long-term outcomes.

KEYWORDS

cardiomyopathy Chagas disease congenital transmission megacolon megaesophagus triatomine vector Trypanosoma cruzi chagoma conjunctiva

2084.e3

CHAPTER 326  Chagas Disease  

African] and T. b. rhodesiense [East African]) (Chapter 325) and the American trypanosome T. cruzi cause disease in humans. Many species of triatomine insects, also called kissing bugs, act as vectors for T. cruzi, and many species of wild and domestic mammals, as well as humans, are involved in the complex life cycle of this fascinating organism. The vectors become infected by ingesting blood from mammals that have parasites in their blood stream. The parasites then multiply in the gut of the insects and are ultimately discharged in the feces of the vector. Transmission to a new mammalian host occurs when parasite-laden vector feces contact vulnerable surfaces such as the mucosae of the mouth or nose, the conjunctivae, or breaks in the skin. When in contact with tissues of the new host, the contaminating parasites enter local cells and multiply intracellularly, and as parasitized cells rupture, they are released into the lymphatics and blood stream. The circulating organisms enter new cells at distant sites and in this manner maintain an endless process of asynchronous multiplication. The life cycle is completed as parasites are swept up in blood meals taken by vectors. In addition to vector-borne transmission, T. cruzi can be transmitted by blood or organs donated by infected persons, from mother to unborn child, by the ingestion of contaminated food or drink,1 by infected persons to sexual contacts,2 and in laboratory accidents.  

EPIDEMIOLOGY

Epizootiology of T. cruzi

The triatomine vectors that transmit T. cruzi are found in the Americas from southern Argentina through the southern half of the United States. The parasite has been isolated from more than 100 species of domestic and wild mammals, which for the most part likely become infected when they eat infected vectors or through congenital transmission. Armadillos, wood rats, raccoons, and opossums are typical wild mammalian reservoirs, and these and other species that harbor T. cruzi can be found in large numbers in the southern and southwestern parts of the United States. Typically, humans acquire T. cruzi infection, which is lifelong, when they live in houses in enzootic areas where the sylvatic cycle of transmission is active. The process begins when vector species adaptable to living in human dwellings take up residence in niches in the primitive wood, mud, and stone houses that are typical in many endemic regions. These vectors become domiciliary, and they then take blood meals, mostly at night, from the humans who occupy the dwellings that they have invaded, as well as from domestic animals that sleep there, particularly dogs. Thus, Chagas disease is primarily a public health problem among poor people who live in rural areas.

Epidemiology of Chagas Disease in the Endemic Countries

Chagas disease is a zoonosis that is endemic in Mexico and all countries of Central and South America. None of the Caribbean islands are endemic. In 2014, the Pan American Health Organization estimated that in the 21 endemic countries, 6 million people are chronically infected with T. cruzi, 30,000 new cases occur each year, and up to 14,000 deaths result from Chagas disease annually.3 Since 1991 a major international vector control program in the southern cone countries of South America (Chile, Argentina, Paraguay, Brazil, Bolivia, and Uruguay) has achieved a marked reduction in vectorial transmission of T. cruzi through housing improvement, education of people at risk for acquiring the infection, and spraying of residual insecticides. Substantial reductions in prevalence rates in school-aged children and in blood donors constitute clear evidence of the success of the program. Uruguay, Chile, and Brazil were certified as being free of vector transmission in 1997, 1999, and 2006, respectively. Marked reduction in transmission has been achieved in Argentina as well. Similar programs have been initiated in Central America and the Andean countries. In parallel with the vector control programs, donor screening has been implemented throughout almost the entire endemic range, and with the notable exception of Mexico, transmission of T. cruzi by transfusion has largely been eliminated.

2085

and Prevention (CDC). Only one of the latter infections occurred in a tourist returning to the United States, but three such instances have been reported in Europe as well as one in Canada. Thus, acute Chagas disease is extremely rare in the United States, and there is no indication that its incidence is increasing. A recent estimate puts the number of T. cruzi–infected persons currently living in the United States at about 240,000 to 300,000.4 Several studies done before blood donor screening began in 2007 identified T. cruzi–infected persons in the donor pool, and nine instances of transmission by transfusion in the United States and Canada were described. Since screening began in 2007, more than 50 million units have been screened, and more than 3,000 T. cruzi– infected donors have been identified and permanently deferred from donation. The confirmed rate of T. cruzi infection in donors is about 1 in 13,300. With the goal of reducing the enormous cost of universal screening ($100 to $200 million per year), a U.S. Food and Drug Administration (FDA)-approved selective screening protocol based on previous negative test results has been implemented. In Europe, infection is seen primarily in Latin American immigrants to Spain, Italy, France, the United Kingdom, and Switzerland. The pooled seroprevalence among such immigrants is about 4%, with an 18% seroprevalence among immigrants from Bolivia.5 The transplantation in the United States of organs from three persons with undiagnosed chronic T. cruzi infection resulted in acute Chagas disease in five recipients, one of whom died of the infection. To date, two instances of congenital transmission of T. cruzi here have been reported. A reasonable estimate of the number of babies born in the United States each year with congenital Chagas disease puts it in the range of 63 to 315. The fact that most babies with congenital Chagas disease are asymptomatic and the low level of knowledge about Chagas disease among caregivers likely underlie the dearth of reported cases.  

PATHOBIOLOGY

In acute Chagas disease, an inflammatory lesion, called a chagoma, may appear at the site of entry of the parasites. Local histologic changes include intracellular parasitism of muscle and other subcutaneous tissues, lymphocytic infiltration, interstitial edema, and hyperplasia of lymph nodes that drain the area. As the parasites spread systemically through the lymphatics and blood stream, muscles, including the myocardium, are the most heavily parasitized tissues, but the organisms can invade essentially any tissue. Myocarditis may develop in association with focal areas of infected cardiomyocytes, inflammation, and necrosis. The characteristic pseudocysts seen in sections of T. cruzi–infected tissues are actually host cells filled with multiplying forms of the parasite (Fig. 326-1). In some patients, parasites can be seen in cerebrospinal fluid (CSF). In persons with chronic Chagas disease, the heart is the organ most commonly affected. Hearts obtained at autopsy from patients who died of Chagas cardiomyopathy usually have a global appearance reflecting biventricular enlargement and thinning of ventricular walls (Fig. 326-2). Mural thrombi are frequently present, and an apical aneurysm of the left ventricle is typical in patients with advanced disease. At the cellular level, the process that underlies these gross pathologic abnormalities is a chronic inflammation with mononuclear

Epidemiology of Chagas Disease in the United States

As noted, the sylvatic cycle of T. cruzi exists in much of the southern and southwestern regions of the United States, but only six cases of autochthonous (disease spread from one individual and acquired in another individual in the same place) acute Chagas disease have been reported: three in Texas and one each in Tennessee, Louisiana, and California. Moreover, in the first 4 years of blood donor screening that started in January 2007, in which approximately 29 million units were tested, only 16 T. cruzi–infected donors, who appeared to have acquired the infection autochthonously, were identified. In the past 30 or more years, fewer than 30 laboratory-acquired and imported cases of acute Chagas disease have been reported to the U.S. Centers for Disease Control

FIGURE 326-1.  Trypanosoma cruzi in cardiac muscle of a child who died of acute Chagas myocarditis. An infected myocyte containing several dozen T. cruzi amastigotes is in the center of the field (hematoxylin-eosin staining, ×900).

2086

CHAPTER 326  Chagas Disease  

cell infiltration, diffuse interstitial fibrosis, and atrophy of myocardial cells. The chronic inflammation affects the conduction system as well and causes a variety of rhythm disturbances, including atrial bradyarrhythmias and fibrillation; premature ventricular contractions; bundle branch blocks, often of the right bundle; ventricular tachycardia; and third-degree atrioventricular block. Parasites are rarely seen in diseased tissues by conventional histologic methods, but several studies using polymerase chain reaction (PCR) assays found a correlation between the intensity of inflammation and the presence of parasites. Evidence accumulated to date implicates the persistence of parasites and the resulting chronic inflammation in affected tissues—rather than autoimmune mechanisms—as the basis for the pathogenesis in patients with chronic T. cruzi infection. The dilation and hypertrophy observed on gross examination of the esophagus or colon of a patient with chronic Chagas disease of the digestive tract (megadisease) can be striking. Focal inflammatory lesions with lymphocytic infiltration are seen on microscopic examination of affected tissues. In addition, the number of neurons in the myenteric plexus is reduced, and periganglion and intraganglion fibrosis with Schwann cell proliferation and lymphocytosis is present. In most patients, the clinical consequences of this parasympathetic denervation are limited to the esophagus or colon (or to both), but the ureters, biliary tree, and other hollow viscera can be affected as well.  

CLINICAL MANIFESTATIONS

Acute Chagas Disease

Acute Chagas disease is usually an illness of children but can occur at any age. Symptoms are typically mild and nonspecific (Table 326-1).6,7 When the parasite has entered through a break in the skin or the site of a vector’s

puncture, as noted, a chagoma may appear with local lymphadenopathy. The Romaña sign, the classic finding in acute Chagas disease, consists of painless edema of the palpebrae and periocular tissues and may appear when the conjunctiva is the portal of entry. These initial local signs can be followed by fever, malaise, anorexia, and edema of the face and lower extremities. Generalized lymphadenopathy and hepatosplenomegaly may also be present. Severe myocarditis may develop as well, and most deaths are caused by the resulting congestive heart failure. Meningoencephalitis is a rare complication. In untreated patients the acute illness resolves spontaneously over a period of 6 to 8 weeks as the patient enters the indeterminate phase of Chagas disease, which is characterized by subpatent parasitemia, absence of associated signs and symptoms, and easily detectable antibodies to T. cruzi.

Chronic Chagas Cardiopathy

In only 10 to 30% of persons chronically infected with T. cruzi does clinically manifested disease develop. It most often involves rhythm disturbances or cardiomyopathy.8 Symptoms of cardiac Chagas disease can develop insidiously over years and often decades after the initial infection. Clinical findings reflect the rhythm disturbances, congestive heart failure, and thromboembolism that characterize the illness. Dizziness, syncope, and even seizures can result from a wide variety of arrhythmias. The cardiomyopathy often leads to biventricular failure, and right-sided heart failure can predominate in patients with advanced disease. Chronic Chagas disease is an independent risk factor for stroke.

Chronic Gastrointestinal Chagas Disease (Megadisease)

The esophagus and colon are the segments of the gastrointestinal (GI) tract most commonly affected in persons with chronic T. cruzi infection. In patients with megaesophagus, the symptoms are similar to those of idiopathic achalasia (Chapter 129) and may include cough, dysphagia, odynophagia, and regurgitation. Hypersalivation and consequent salivary gland hypertrophy develop in some patients with advanced esophageal dysfunction. Aspiration can occur, especially during sleep, and in untreated patients, repeated episodes of aspiration pneumonitis are common. Weight loss and even cachexia in patients with severe megaesophagus can combine with pneumonitis to cause death. Patients with chagasic megacolon have intermittent abdominal pain and chronic constipation and in advanced cases can go for several weeks between bowel movements. Rarely, acute obstruction, occasionally with volvulus, can lead to perforation, sepsis, and death.  

FIGURE 326-2.  Chest radiograph of a patient from Bolivia with chronic Trypanosoma cruzi infection, rhythm disturbances, and cardiomyopathy. Pacemaker wires can be seen in the area of the left ventricle.

IMMUNOSUPPRESSION AND TRANSPLANTATION IN PATIENTS INFECTED WITH T. CRUZI

When persons with chronic carriage of T. cruzi become immunosuppressed, reactivation of the infection can occur, sometimes with an intensity that is atypical of acute Chagas disease in immunocompetent persons.9 The overall incidence of reactivation in immunosuppressed persons who harbor the parasite chronically is not known. Reactivation after renal transplantation has been reported, and in rare instances central nervous system abscesses and skin lesions were involved. The consensus view is that Chagas disease should not be a contraindication to kidney transplantation. In T. cruzi–infected patients who do undergo the procedure, periodic monitoring for signs and symptoms of acute Chagas disease should nonetheless be carried out, including careful neurologic evaluation, and parasitologic testing should be performed when acute illnesses occur postoperatively. Reactivation of T. cruzi infection can also occur in persons coinfected with the parasite and human immunodeficiency virus (HIV). Dozens of such patients have been described. It is striking that in many of these patients T. cruzi brain

TABLE 326-1 CLINICAL MANIFESTATIONS AND DIAGNOSIS OF CHAGAS DISEASE PHASE OR FORM

CLINICAL MANIFESTATIONS

DIAGNOSIS

TREATMENT SUCCESS

Acute infection

Usually nonspecific mild symptoms

Wet prep of blood or PCR

80-100%

Congenital infection

Asymptomatic or nonspecific mild symptoms

Wet prep of blood or PCR

80-100%

Chronic infection

Asymptomatic, without ECG change

Positive serology

May decrease progression to cardiomyopathy

Cardiomyopathy

Arrhythmias, syncope, left ventricular dysfunction, heart failure, ECG conduction abnormalities

Positive serology

Benefit not established

Gastrointestinal

Dilation of esophagus, colon, or both

Positive serology

Benefit not established

Reactivation in immunosuppressed hosts

Acute myocarditis, CNS abscesses, skin chagomas

Wet prep of blood or PCR

Suppression of acute symptoms and parasitemia, but long-term benefit not clear

CNS = central nervous system; ECG = electrocardiogram; PCR = polymerase chain reaction. Adapted from Bern C. Chagas disease. N Engl J Med. 2015;373:456-466.

2087

CHAPTER 326  Chagas Disease  

abscesses developed, which do not occur in immunocompetent patients with chronic Chagas disease. It has been shown that HIV viral loads increase in the context of reactivated acute Chagas disease. Calculations based on the overlapping epidemiologies of HIV and T. cruzi infections in the endemic countries suggest that the incidence of reactivation of the latter in coinfected persons is low.  

DIAGNOSIS

Acute Chagas Disease

The first step in considering the diagnosis of acute Chagas disease is establishing that a person is at risk for T. cruzi infection. Risk factors include recent residence or blood transfusion in an endemic area, birth to a mother with geographic- or transfusion-associated risk in the case of a newborn, or a laboratory accident involving the parasite. Definitive diagnosis of acute Chagas disease can be made only by detecting parasites. Serologic assays for T. cruzi– specific IgM are not accurate enough to justify their use. In immunocompetent persons suspected of having acute Chagas disease, the most productive approach is examination of wet preparations of anticoagulated blood or buffy coat for the highly motile blood stream parasites. They can be seen in Giemsa-stained smears as well. In infected immunocompromised patients. Moreover, parasites can sometimes be found in other specimens such as lymph node aspirates, biopsy specimens of skin lesions, bone marrow, endomyocardial tissue, CSF, and pericardial fluid. When these direct methods fail to detect organisms in an at-risk person, samples should be tested with a PCR assay (see later).10 PCR assays have been shown to be more sensitive than the direct methods described earlier for detecting T. cruzi. Another method is to culture blood or other samples in specialized liquid medium, but the usefulness of this approach is limited by low sensitivity (50 to 70% for hemoculture) and by the fact that cultures take a minimum of 2 weeks before turning positive. In newborns whose blood is negative both by direct examination and in a PCR assay right after birth, serologic evaluation for T. cruzi–specific IgG should be performed 6 to 9 months later, by which time maternal antibodies will have disappeared.

Chronic Chagas Disease

Chronic T. cruzi infection is usually diagnosed by detecting IgG antibodies that specifically bind to parasite antigens, and in almost all instances isolation of the organism is not necessary. There is no credible evidence that “seronegative chronic Chagas disease” exists. More than 30 serologic assays for diagnosing Chagas disease are currently available commercially in endemic countries, where they are used widely for testing clinical specimens and screening blood donors. Even though these tests generally have high sensitivity and specificity, false-positive reactions do occur, typically with specimens from people who have other infectious diseases or autoimmune conditions. The World Health Organization has recommended that testing be done with two assays based on different formats. In the United States, the Ortho T. cruzi ELISA Test System (Ortho-Clinical Diagnostics, Raritan, NJ) and the Abbott Prism Chagas Assay (Abbott Laboratories, Abbott Park, IL) are approved by the FDA for screening donated blood. The Abbott ESA Chagas and the Chagas RIPA have both been cleared by the FDA for confirmatory testing of donor samples that are positive in the screening tests. In human studies, the sensitivity of the PCR assays has ranged from 44.7 to 100%, with most being higher than 90%. It is generally accepted that the level of sensitivity of these assays is not high enough to justify their use for confirmatory testing of serologically positive donor samples. In contrast, PCR assays may be useful in persons who have borderline serologic results, in patients suspected of having congenital or acute Chagas disease in whom parasites are not detected microscopically, and in infected patients who have received specific treatment. In all such persons, because of the sensitivity issue, only positive PCR results can be taken as being truly indicative of infection status.

TREATMENT  Antiparasitic Drugs

The two drugs currently available for treating Chagas disease (benznidazole and nifurtimox) are unsatisfactory, and the need for a parasitologically curative drug regimen is the most important current challenge in Chagas disease research (also see Chapter 323).11 Benznidazole is considered the drug of choice by most Latin American experts.12 Benznidazole is a nitroimidazole derivative. Cure rates are similar or perhaps a bit higher than those achieved with nifurtimox. A cure rate higher than 90%

in babies with congenital infection has been observed with benznidazole. Side effects can include rash, peripheral neuropathy, and granulocytopenia. The recommended oral dosage of benznidazole is 5 to 10 mg/kg body weight per day for children and 5 mg/kg body weight per day for adults, in both cases for 60 days.13 Nifurtimox is a nitrofuran derivative that has been used for more than three decades. Nifurtimox reduces symptoms and decreases mortality rates in patients with acute Chagas disease, approximately 70% of whom are cured parasitologically. Nifurtimox also can cure a substantial portion of children in the indeterminate phase, but unfortunately, cure rates may be less than 10% in adults with long-standing chronic T. cruzi infection. Disadvantages of nifurtimox include its long course of treatment and occasionally bothersome side effects, including GI complaints such as anorexia, nausea, vomiting, weight loss, and abdominal pain. Patients taking the drug may also have neurologic symptoms such as insomnia, restlessness, paresthesias, twitching, polyneuritis, and even seizures. For adults, the recommended oral dosage is 8 to 10 mg/kg body weight per day. For adolescents, the dose is 12.5 to 15 mg/kg/day, and for children 1 to 10 years of age, it is 15 to 20 mg/kg/day. The drug should be given each day in four divided doses, and treatment should be continued for 90 to 120 days. In the United States, nifurtimox can only be obtained from the CDC Drug Service (404-639-3670). There is broad agreement among experts that treatment is indicated in all patients with acute or congenital infections, as well as in chronically infected children up to 18 years old. This recommendation is supported by several studies suggesting that a majority of such patients appear to be cured parasitologically. By extension, it would be reasonable to treat anyone 18 years or older known to have acquired T. cruzi infection within the past 17 years. There is also broad agreement that persons with advanced symptomatic T. cruzi infection should not be given specific treatment. The remaining question, then, is whether adults with long-standing indeterminate-phase infections, who by far constitute the largest group of T. cruzi–infected persons, should be treated. This is a thorny question because the burden of taking a full course of either drug can be substantial and because parasitologic cure rates are so low. In asymptomatic patients, benznidazole can convert up to 90% of PCR-positive patients who are able to tolerate the drug to negative for a year, A1  A2  but there is no clear evidence that such therapy delays the onset of symptoms, slows disease progression, or reduces mortality rates. In a randomized trial of benznidazole in 3000 patients ages 18 to 75 years with Chagas disease and incipient cardiac disease, PCR positivity was significantly suppressed in the subjects treated with benznidazole through 5 years of follow-up, but there was no significant reduction in clinical cardiac deterioration or in deaths. A3  One area of hope in this dreary landscape of drugs for Chagas disease relates to whether treatment before pregnancy reduces the likelihood of subsequent congenital transmission of T. cruzi. In a handful of studies done in Argentina and Spain now involving a total of more than 250 babies born to women who had been or during the study were treated with benznidazole or nifurtimox, either when they were younger than 18 years or as adults, not a single baby was found to have congenital Chagas disease.14 This outcome compares with historical rates of congenital transmission of 2 to 10%, as well as a rate of 16% in one of the trials in which there was an untreated arm. The fact that under current perspectives regarding the efficacy of treatment, a substantial proportion of the girls and women treated would not have been cured parasitologically, makes this outcome unexpected. In any event, the results suggest that all girls of any age, as well as all women of childbearing age, who have geographic or maternal risk for Chagas disease should be screened serologically. Those determined to be positive, after ruling out pregnancy, should be treated with a full course of benznidazole with the goal of reducing the rate of congenital transmission to babies they may have, even years later. ,

Management of Symptomatic Chagas Disease

An algorithm for the evaluation of persons with newly diagnosed Chagas disease has been developed (Fig. 326-3). T. cruzi–infected patients in whom symptomatic cardiac or GI disease develops should be referred to appropriate subspecialists. Beyond the possible use of nifurtimox or benznidazole, treatment of acute and chronic Chagas disease is symptomatic (Chapter 54). In patients with symptomatic chronic Chagas cardiac disease, treatment should be directed at managing symptoms with the anticoagulants and cardiotropic drugs used in patients with cardiomyopathy of other causes.15 Some data suggest that carvedilol (up-titrated to 25 mg twice daily) may reduce mortality in Chagas cardiomyopathy. A4  Pacemakers are useful in patients with ominous arrhythmias. Implantable cardioverter-defibrillators may be useful in selected patients with dysrhythmias due to Chagas heart disease, but this issue needs further investigation in prospective randomized trials.16 Heart transplantation (Chapter 53) is an option in patients with end-stage Chagas cardiac disease, and more than 150 such patients have undergone the procedure in Brazil and the United States.17 As is the case with other T. cruzi– infected patients who are immunosuppressed, reactivation is a risk but is manageable. The usefulness and side effects of long-term prophylaxis for reactivation with either benznidazole or nifurtimox in T. cruzi–infected patients after heart

2088

CHAPTER 326  Chagas Disease  

Patient with new diagnosis of chronic Chagas disease

Confirm T. cruzi infection with ≥2 serologic tests

Advise patient not to donate blood Test all children and subsequent newborns of infected women

Normal medical history, physical examination, and ECG

Repeat history, physical examination, and ECG yearly

Perform complete medical history, physical examination, and 12-lead ECG with 30-sec rhythm strip

Immediate evaluation if any of the following are present Cardiac alterations Presyncope or syncope Transient ischemic attack Atypical chest pain Ventricular tachycardia High-grade AV block Marked bradyarrhythmia Gastrointestinal symptoms Acute abdominal pain Rebound tenderness Other symptoms suggestive of bowel ischemia or volvulus

Cardiac symptoms or signs or ECG findings suggestive of chronic Chagas disease Symptoms Palpitations, syncope, presyncope Symptoms of CHF Thromboembolic phenomena Atypical chest pain ECG findings Common RBBB Incomplete RBBB* Left anterior fascicular block First-degree AV block Second-degree AV block, Mobitz type I or II Complete AV block Bradycardia, sinus node dysfunction Ventricular extrasystoles, often frequent, multifocal, or paired Ventricular tachycardia, nonsustained or sustained Less common but clinically significant when present Atrial fibrillation or flutter LBBB Low QRS voltage Q waves

Gastrointestinal symptoms Dysphagia for both liquids and solids Regurgitation, aspiration Odynophagia Weight loss Prolonged constipation

Barium studies, other gastrointestinal evaluation as indicated

Perform complete cardiac evaluation Echocardiogram Ambulatory ECG monitoring Exercise test Other evaluation as indicated *QRS interval of 0.10 to 0.11 seconds in adults. Criteria based on the Minnesota Code Manual of Electrocardiographic Findings with modifications from Maguire et al. Different criteria may be required for ECGs in children. FIGURE 326-3.  Algorithm for baseline evaluation of a patient with newly diagnosed chronic Trypanosoma cruzi infection. AV = atrioventricular; CHF = congestive heart failure; ECG = electrocardiogram; LBBB = left bundle branch block; RBBB = right bundle branch block. (From Bern C, Montgomery SP, Herwaldt BL, et al. Evaluation and treatment of Chagas disease in the United States: a systematic review. JAMA. 2007;298:2171-2181.)

transplantation have not been evaluated. The long-term survival of Chagas patients with heart transplants appears to be longer than that of patients undergoing cardiac transplantation for other reasons, probably because the lesions of T. cruzi–associated pathology affect mostly the heart. Chagas megaesophagus should be treated as normally done for idiopathic achalasia (Chapter 129), which usually responds to balloon dilation of the lower esophageal sphincter when symptoms are mild. Surgical treatment may be required in patients who do not respond to repeated attempts at balloon dilation. Laparoscopic myotomy is being used with increasing frequency to treat Chagas megaesophagus, as is the case with achalasia. Chagas megacolon in its early stage can be treated with a high-fiber diet and occasional laxatives or enemas. Fecal impaction requiring manual disimpaction can occur, and toxic megacolon requires surgery. In patients with advanced megacolon, volvulus (Chapter 133) can develop when an enlarged and lengthened sigmoid colon twists and folds on itself; volvulus causes a constellation of symptoms and in many cases requires immediate surgery. Even if the symptoms associated with volvulus are resolved without operative

intervention, however, surgical treatment is usually ultimately necessary because the volvulus tends to recur. Several surgical procedures are used to treat advanced Chagas megacolon, all of which include resection of the sigmoid and removal of part of the rectum.



PREVENTION

Reducing human contact with triatomine vectors through education of at-risk persons, housing improvement, and spraying of residual insecticides in endemic countries has resulted in reduction or elimination of vector transmission of T. cruzi in a major part of the endemic range, and progress in this regard is expected to continue.18 Serologic screening of donated blood has essentially eliminated transfusion-related transmission of the parasite in most endemic areas. Outbreaks of acute Chagas disease through oral transmission can be avoided by the implementation of better food safety standards. Drug treatment

TABLE 326-2 PREDICTING DEATH IN CHAGAS HEART DISEASE CHARACTERISTIC

POINTS

Male sex

2

Functional class III or IV

5

Low QRS voltage on electrocardiography

3

Cardiomyopathy on chest radiography

5

Left ventricular systolic dysfunction on echocardiograph

3

Nonsustained ventricular tachycardia on Holter monitoring

3

10 year mortality:

0-6 points 7-11 points 12-20 points

9% 37% 85%

Adapted from Rassi A, Jr., Rassi A, Little WC, et al. Development and validation of a risk score for predicting death in Chagas’ heart disease. N Engl J Med. 2006;355:799-808.

of T. cruzi–infected women before pregnancy reduces the likelihood of congenital transmission. No protocols have been defined and validated for preventing reactivation of T. cruzi infection in chronically infected persons who are immunosuppressed iatrogenically or by HIV. A treatment regimen that reliably results in parasitologic cure is needed to prevent the onset or progression of chronic symptomatic Chagas disease.  

PROGNOSIS

The prognosis for patients with acute Chagas disease is generally excellent because most acutely infected persons have only mild symptoms that resolve spontaneously, even without specific treatment. The occasional patient who has symptomatic acute Chagas myocarditis should generally do well if treated early. In persons with chronic T. cruzi infection, the lifetime risk for the development of related cardiac or GI dysfunction is only 10 to 30%. A validated risk score assessment tool can estimate prognosis in the absence of heart transplantation (Table 326-2).

  Grade A References A1. Molina I, Gomez I, Prat J, et al. Randomized trial of posaconazole and benznidazole for chronic Chagas’ disease. N Engl J Med. 2014;370:1899-1908. A2. Morillo CA, Marin-Neto JA, Avezum A, et al. Randomized trial of benznidazole for chronic Chagas’ cardiomyopathy. N Engl J Med. 2015;373:1295-1306. A3. Morillo CA, Waskin H, Sosa-Estani S, et al. Benznidazole and posaconazole in eliminating parasites in asymptomatic T. cruzi carriers: the STOP-CHAGAS Trial. J Am Coll Cardiol. 2017;69:939-947. A4. Martí-Carvajal AJ, Kwong JS. Pharmacological interventions for treating heart failure in patients with Chagas cardiomyopathy. Cochrane Database Syst Rev. 2016;7:CD009077.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 326  Chagas Disease  

GENERAL REFERENCES 1. Robertson LJ, Devleesschauwer B, Alarcon de Noya B, et al. Trypanosoma cruzi: time for international recognition as a foodborne parasite. PLoS Negl Trop Dis. 2016;10:1-6. 2. Gomes C, Almeida AB, Rosa AC, et al. American trypanosomiasis and Chagas disease: sexual transmission. Int J Infect Dis. 2019;81:81-84. 3. Pan American Health Organization. Chagas disease. Available from: http://www.paho.org/hq/index. php?option=com_topics&view=article&id=10&Itemid=40743. Accessed May 15, 2019. 4. Manne-Goehler J, Umeh CA, Montgomery SP, et al. Estimating the burden of Chagas disease in the United States. PLoS Negl Trop Dis. 2016;10:1-7. 5. Antinori S, Galimberti L, Bianco R, et al. Chagas disease in Europe: a review for the internist in the globalized world. Eur J Intern Med. 2017;43:6-15. 6. Bern C. Chagas’ disease. N Engl J Med. 2015;373:456-466. 7. Echeverria LE, Morillo CA. American trypanosomiasis (Chagas disease). Infect Dis Clin North Am. 2019;33:119-134. 8. Perez-Molina JA, Molina I. Chagas disease. Lancet. 2018;391:82-94. 9. Salvador F, Sanchez-Montalva A, Valerio L, et al. Immunosuppression and Chagas disease: experience from a non-endemic country. Clin Microbiol Infect. 2015;21:854-860. 10. Schijman AG. Molecular diagnosis of Trypanosoma cruzi. Acta Trop. 2018;184:59-66. 11. Sales Junior PA, Molina I, Fonseca Murta SM, et al. Experimental and clinical treatment of Chagas disease: a review. Am J Trop Med Hyg. 2017;97:1289-1303.

2089.e1

12. Sperandio da Silva GM, Mediano MFF, Hasslocher-Moreno AM, et al. Benznidazole treatment safety: the Médecins Sans Frontières experience in a large cohort of Bolivian patients with Chagas’ disease. J Antimicrob Chemother. 2017;72:2596-2601. 13. Thakare R, Dasgupta A, Chopra S. An update on benznidazole for the treatment of patients with Chagas disease. Drugs Today (Barc). 2018;54:15-23. 14. Murcia L, Simon M, Carrilero B, et al. Treatment of infected women of childbearing age prevents congenital Trypanosoma cruzi infection by eliminating the parasitemia detected by PCR. J Infect Dis. 2017;215:1452-1458. 15. Bocchi EA, Bestetti RB, Scanavacca MI, et al. Chronic Chagas heart disease management: from etiology to cardiomyopathy treatment. J Am Coll Cardiol. 2017;70:1510-1524. 16. Carmo AAL, de Sousa MR, Agudelo JF, et al. Implantable cardioverter-defibrillator in Chagas heart disease: a systematic review and meta-analysis of observational studies. Int J Cardiol. 2018;267: 88-93. 17. Benatti RD, Oliveira GH, Bacal F. Heart transplantation for Chagas cardiomyopathy. J Heart Lung Transplant. 2017;36:597-603. 18. Cucunubá ZM, Nouvellet P, Peterson JK, et al. Complementary paths to Chagas disease elimination: the impact of combining vector control with etiological treatment. Clin Infect Dis. 2018;66: S293-S300.

2089.e2

CHAPTER 326  Chagas Disease  

REVIEW QUESTIONS 1. Which of the following measures have been important over the past few decades in reducing the transmission of Trypanosoma cruzi (Chagas disease) in endemic countries? A . Improvement of housing conditions B. Identification and treatment of babies with congenital Chagas disease C. Spraying of insecticides in at-risk houses to eliminate insect vectors D. All of the above E. A and C Answer: E  Improvement of housing conditions and insecticide spraying, along with education of at-risk populations, all of which reduce contact with infected vectors, have been at the core of the successful programs that reduce transmission of T. cruzi. Programs focused on identifying and treating babies with congenital T. cruzi infection have not been implemented widely and in any event would have relatively little effect on reducing transmission of the parasite. 2. In which of the following patients is chronic Chagas disease (Trypanosoma cruzi infection) likely to be the cause of the patient’s signs and symptoms? A . A 56-year-old man from rural Brazil with early-onset dementia B. A 58-year-old Bolivian woman with abdominal pain caused by a small bowel obstruction C. A 62-year-old man with weight loss, hemoptysis, and anemia D. A 20-year-old Bolivian man with recurrent syncope and third-degree heart block and an ectopic ventricular pacer on electrocardiogram E. A 7-year-old child with intermittent diarrhea and eosinophilia Answer: D  In the endemic countries, particularly in Bolivia, where the overall prevalence is about 7%, chronic Chagas disease is a major cause of serious rhythm disturbances, heart block, and sudden death among young people. The other options are unlikely because a link between Chagas disease and dementia has not been established; chronic Chagas disease is sometimes manifest as obstructive megacolon but not small bowel obstruction; weight loss, hemoptysis, and anemia are more suggestive of tuberculosis than Chagas disease; and intermittent diarrhea and eosinophilia are not associated with chronic Chagas disease in patients of any age. 3. Which of the following Latin American countries has the lowest prevalence of Chagas disease? A . Bolivia B. Dominican Republic C. Colombia D. Argentina E. Ecuador Answer: B  Chagas disease is endemic in all the countries of Central and South America, as well as in Mexico. None of the Caribbean islands are endemic for Chagas disease. The only persons with Chagas disease in the Dominican Republic would be immigrants from the endemic countries, and thus the prevalence there would be quite low and far less than in the other countries listed where vector-borne transmission continues.

4. Which of the following statements is false? A . Of all the countries in which Chagas disease is endemic, Bolivia has the highest prevalence. B. Several hundred cases of transfusion transmission of Chagas disease were reported in the United States and Canada before the implementation of donor screening in 2007. C. Transfusion transmission of Trypanosoma cruzi (Chagas disease) has been largely eliminated in the endemic countries by serologic screening of blood donors. D. Highly accurate assays for diagnosing Chagas disease are available in the United States, Canada, and the endemic countries. E. Trypanosoma cruzi is widely distributed among wild mammals and insect vectors in the southern and southwestern United States. Answer: B  Before the implementation of blood donor screening for Chagas disease in the United States and Canada in 2007, only nine instances of transfusion transmission of the infection had been reported. The other statements are all true: Bolivia’s Chagas disease prevalence of about 7% is the highest of all the endemic countries. To the credit of politicians and public health authorities in the endemic countries, and because of the availability of accurate serologic assays, transfusion transmission of Chagas disease is largely a thing of the past. Mexico is a notable exception to this achievement, however, because universal donor testing for Chagas disease has not been implemented there yet. To the surprise of many, in the southern third of the United States, T. cruzi is widely enzootic in many species of wild mammals and insect vectors, as well as in domestic dogs. 5. The current consensus of experts in Chagas disease holds that the following groups of persons with T. cruzi infection should be given specific treatment (benznidazole or nifurtimox) except: A . Babies with congenital Chagas disease B. Patients older than 50 years with advanced symptomatic Chagas heart disease C. Children and adolescents up to age 18 years D. Patients with AIDS and reactivated acute T. cruzi infection E. Adults with acute Chagas disease and mild symptoms Answer: B  A panel of experts in Chagas disease from the endemic countries as well as the United States and Canada, convened by the Centers for Disease Control and Prevention in 2006 with the goal of developing evidence-based guidance on diagnosis and treatment, concluded that all groups of patients listed previously should be treated, with the exception of patients older than 50 years of age or those with advanced symptomatic Chagas cardiac disease. (Bern C, Montgomery SP, Herwaldt BL, et al. Evaluation and treatment of Chagas disease in the United States: a systematic review. JAMA. 2007;298:2171-2181.)

CHAPTER 327 Leishmaniasis  

2089

widespread multilesional cutaneous involvement, and from a single nodule to potentially fatal visceral disease affecting spleen, liver, and bone marrow. This disease complex is caused by 20 species of Leishmania, which are widely distributed in Europe, Asia, Africa, and South and Central America, with limited foci in Southeast Asia.1 The characteristics of the main Leishmania species are summarized in Table 327-1. There are an estimated 1.5 to 2.0 million new cases each year, with up to 70,000 deaths, although this is probably an underestimate because leishmaniasis is not a reportable disease in many of the 101 countries and territories in which it is known to occur. Many Leishmania infections are either asymptomatic or diagnosed several months after the onset of symptoms. Clinical aspects are considered under separate sections for visceral leishmaniasis (VL) and cutaneous/mucosal leishmaniasis (CL, ML).  

EPIDEMIOLOGY

Infection is established in the mammalian host following a bite of the female sandfly belonging to either Phlebotomus spp in Europe, Asia, and Africa or Lutzomyia spp in the Americas. Different species of sandfly are associated with transmission of different Leishmania species. Most species that cause CL have a zoonotic (acquired from another mammal) transmission cycle, with the exception of Leishmania tropica, which is frequently anthroponotic (transmitted between human beings). VL is either normally anthroponotic (in the case of Leishmania donovani) or zoonotic (in the case of Leishmania infantum). The predominant mammalian hosts (the reservoirs) are associated with different Leishmania species in diverse ecosystems (Fig. 327-1). VL is caused by either L. donovani or L. infantum (which is identical to Leishmania chagasi in South America). These species have different geographic distributions, with the highest incidence found in the poorest communities in seven countries (Bangladesh, Nepal, India, Sudan, South Sudan, Ethiopia, and Brazil), and it is potentially fatal if untreated. An estimated 1 in 5 to 1 in 50 infections are symptomatic, depending on the parasite species and host immunity. Since 2005, there has been a regional program to eliminate VL, during which time there has been a reduction of annual incidence by more than 90% on the Indian subcontinent. Nevertheless, the economic burden of infection remains enormous.2 CL, which undergoes self-cure in 20 to 90% of patients within 3 to 18 months (depending on infecting species and geographic location), is widely distributed, but its prevalence is difficult to estimate because of underreporting. Prevalence is associated with age, possibly related to the acquisition of immunity and risk factors, including the presence of domestic animals, rodents, or other mammalian hosts. Ecologic conditions for sandflies, including shaded and humid habitats in crevices and mammal burrows, have been identified. Urbanization, deforestation, and migration have resulted in changing patterns of disease, with transmission occurring in peridomestic cycles. Other forms of transmission, such as through organ transplantation, intravenous needles shared by drug users, or laboratory exposure, have been reported.  

327  LEISHMANIASIS PIERRE A. BUFFET AND SIMON L. CROFT



DEFINITION

Leishmaniasis is caused by protozoan parasites of the genus Leishmania that are generally transmitted between mammalian hosts by female phlebotomine sandflies. The parasite exists as an extracellular flagellated form, the promastigote, in the sandfly gut, and as an intracellular form, the amastigote, that survives and multiplies in a phagolysosomal compartment of macrophages in the mammalian host. The disease in humans affects either the skin/mucosae or internal organs and ranges in severity from a single, self-curing, limited skin lesion to

PATHOBIOLOGY

The infection is initially established in the skin after the inoculation of infective metacyclic promastigotes by the sandfly. These infective forms have a glycoprotein coat (a lipophosphoglycan) that enables them to resist complement and attach to and invade host cells. Peptides in sandfly saliva (e.g., maxadilan) cause vasodilation and erythema and help establish infection in the dermal layer of the skin. Early responses to infection involve neutrophil infiltration and invasion of resident macrophages. Progress of the disease depends on the parasite species and host responses. For both VL and CL, disease progression depends on the maintenance of a parasite-specific immunosuppressive state. During established disease, host cell macrophages are in a deactivated state but become activated either spontaneously or after treatment. They are then able to kill the parasites, which are sensitive to nitric oxide and oxygen radicals, in the phagolysosomal compartment. Resolution of disease, following the activation of macrophages, is enhanced by a helper T-cell type 1 (TH1) response after interaction between antigen-presenting cells (e.g., dendritic cells) with CD4+ and CD8+ T cells and subsequent secretion of pro-inflammatory cytokines (e.g., interleukin-1 [IL-1], interferon-γ, tumor necrosis factor-α). However, in clinical forms such as active VL or diffuse CL, a TH2 cell response predominates, whereby downregulation of macrophage activity follows the production of cytokines such as IL-4, IL-10, IL-13, and transforming growth factor-β. This profile has been defined in experimental models, mainly inbred mice, and clinical studies support the notion of a generally similar profile in typical, polar infections in humans. Immunologic patterns may be more complex in some clinical forms, with

CHAPTER 327 Leishmaniasis  

ABSTRACT

Leishmaniasis is caused by protozoan parasites of the genus Leishmania that are transmitted by sandflies. There are 20 species of Leishmania widely distributed in Europe, Asia, Africa, and South and Central America, with limited foci in Southeast Asia. The predominant mammalian hosts (the reservoirs) are associated with different Leishmania species in diverse ecosystems. The disease in humans affects either the skin/mucosae or internal organs and ranges in severity from one self-curing, limited skin lesion to widespread multilesional cutaneous involvement, and from a single nodule to potentially fatal visceral disease affecting spleen, liver, and bone marrow. There are an estimated 1.5 to 2.0 million new cases each year, with up to 70,000 deaths in 101 countries. Subjects carrying Leishmania often do not have symptoms. When they are ill, the diagnosis is generally made several months after the onset of symptoms. Visceral leishmaniasis is caused by either Leishmania donovani or Leishmania infantum. The highest incidence is found in the poorest communities of Bangladesh, Nepal, India, Sudan, South Sudan, Ethiopia, and Brazil. Since 2005, there has been a reduction of annual incidence by more than 90% on the Indian subcontinent. Diagnosis confirmation is by direct visualization of parasites, polymerase chain reaction testing, and serology. Treatment is based on liposomal amphotericin B at most foci but is more complex in East Africa and in immunodeficient patients. Cutaneous leishmaniasis undergoes self-cure in 20 to 90% of patients within 3 to 18 months and is widely distributed. Treatment is by local measures in most patients.

KEYWORDS

leishmaniasis visceral cutaneous mucosal amphotericin B pentamidine paromomycin

2089.e3

2090

CHAPTER 327 Leishmaniasis  

TABLE 327-1 CHARACTERISTICS OF THE MAIN LEISHMANIA SPECIES DISTRIBUTION: OLD WORLD

DISTRIBUTION: NEW WORLD

PRIMARY FORM

SECONDARY FORMS

ANTHROPONOTIC: AREAS OF TRANSMISSION

ZOONOTIC: RESERVOIR

ALTERNATIVE NAME

LEISHMANIA SPP

LEISHMANIA SUBGENUS

L. donovani

Leishmania

Indian subcontinent E. Africa

L. infantum (L. chagasi)

Leishmania

Europe Asia

L. major

Leishmania

Asia N. & E. Africa Europe

CL

L. tropica

Leishmania

Asia Europe

CL

Recidivans

L. aethiopica

Leishmania

Ethiopia

CL

DCL

L. mexicana

Leishmania

C. America

CL

L. amazonensis

Leishmania

C. & S. America

CL

DCL

Rodent

L. braziliensis

Viannia

S. America

CL ML

DissCL Lymph

Rodent, marsupial

ML-espundia

L. panamensis

Viannia

C. & S. America

CL

ML Lymph

Edentate rodent

Ulcera de bejuco

L. guyanensis

Viannia

S. America

CL

ML Lymph

Rodent, edentates

Pian bois

L. peruviana

Viannia

S. America

CL

Canid

Uta

L. martiniquensis/ siamensis

Leishmania?

S. & C. America

South East Asia/ West Indies

VL

PKDL CL, ML OIVL

VL

CL, ML OIVL

Indian subcontinent E. Africa

Kala-azar Canid Rodent

Syria Afghanistan

Rodent

Aleppo boil

Hyrax Rodent

CL, VL

Chiclero ulcer

DCL

CL = cutaneous leishmaniasis; DCL = diffuse cutaneous leishmaniasis; DissCL = disseminated cutaneous leishmaniasis; Lymph = nodular lymphangitis; ML = mucosal leishmaniasis; OIVL = opportunistic infection with VL in HIV-infected patients; PKDL = post–kala-azar dermal leishmaniasis; VL = visceral leishmaniasis.

Cutaneous Leishmaniasis

Visceral Leishmaniasis

Visceral Leishmaniasis

Diffuse Cutaneous Leishmaniasis

Cutaneous Leishmaniasis

Cutaneous Leishmaniasis

L. infantum –HIV Co-infection

Cutaneous Leishmaniasis

Human

Human

Human

Human IVDU

Human

Human Host

L. donovani L. infantum ?

Peridomiciliar Animal

L. tropica

Human IVDU Dog

Rodent L. major

L. infantum

Human

Hyrax, Dog ?

L. peruviana Dog

Rodent

Hyrax, Dog

Human

Wild Animal

Fox, wolf, cat Sloth and other wild animals ?

L. panamensis L. guyanensis L. amazonensis

L. aethiopica causes cutaneous leishmaniasis, diffuse cutaneous leishmaniasis, and mucosal leishmaniasis

Fox, wolf, cat

Sloth Primary Cycles

Secondary Cycles

Tertiary Cycles

FIGURE 327-1.  Old World and New World zoonotic and anthroponotic life cycles of the main species of Leishmania. Leishmaniasis is often referred to as a “disease complex” because different forms of disease can be caused by the same parasite species and similar forms of disease can be caused by different parasite species. IVDU = intravenous drug use.

CHAPTER 327 Leishmaniasis  

antiparasitic immunity also playing a pathogenic role and contributing to tissue damage In patients with VL, the absence of a T-cell-specific immune response to leishmanial antigens is associated with uncontrolled progression of infection. This is linked to elevated levels of IL-10 and decreased production of interferon-γ. Genetic susceptibility to L. donovani in Sudan has been associated with a solute carrier family (formerly NRAMP1) that regulates macrophage activation and with a polymorphism in the IL-4 gene. In localized simple CL, patients show a TH1-type response and a delayed-type hypersensitivity (DTH) response. DTH is frequently measured by a Montenegro skin test, which can also be used in epidemiologic prevalence studies. Chronic infections show a TH2-type response, most predominantly in patients with diffuse CL, in whom there is complete anergy to leishmania antigen and no DTH response. Patients with ML have a TH1 or TH2 cell response or both and have a strong or weak DTH response. Post–kala-azar dermal leishmaniasis (PKDL), a rare sequela to cure from VL, is poorly understood. The roles of CD4+ and CD25+ T cells appear to be different in Indian and Sudanese forms of PKDL.

  VISCERAL LEISHMANIASIS  

CLINICAL MANIFESTATIONS

The onset of VL, often referred to as kala-azar when caused by L. donovani, occurs weeks to months after the initial infection. Clinical signs and symptoms, such as fever, splenomegaly, and hepatomegaly, do not distinguish VL from hyperreactive malarial splenomegaly or other infectious or hematologic conditions. Moderate to severe anemia, mild to moderate leukopenia, thrombocytopenia, systemic inflammation, and polyclonal hypergammaglobulinemia, either isolated or combined, suggest but do not confirm the diagnosis. Parasitologic tests are therefore indispensable before making a therapeutic decision.3  

DIAGNOSIS

Parasitology

Microscopic visualization of amastigotes in samples from the lymph nodes, bone marrow, liver, spleen, or other organs was usually the first step in diagnosis. Because spleen aspiration causes life-threatening complications in approximately 0.1% of patients, it should be performed only in trained facilities and only if other, lower risk methods cannot be used. It is still used in the field because of the higher cost, logistic constraints, and lower sensitivity of bone marrow aspiration. Polymerase chain reaction (PCR) is more sensitive than microscopic examination and has become the first-line test in several referral hospitals and research centers.4 Quantitative PCR with validated thresholds allows accurate diagnosis with venous blood samples, thereby avoiding bone marrow aspiration.

Serology

Serologic tests based on indirect fluorescent antibody (IFA), enzyme-linked immunosorbent assay (ELISA), or Western blot display high performance but require equipment that is poorly adapted to field settings. The direct agglutination test (DAT) and immunochromatography (dipstick) using the rK39 antigen have high diagnostic accuracy and can be used in peripheral health centers. Whatever the serologic test used, specific antibodies remain detectable for several years after cure or asymptomatic infection.

Antigen Detection Tests

A latex agglutination test that detects a heat-stable, low-molecular-weight carbohydrate antigen in the urine of patients with VL showed good specificity but low to moderate sensitivity in East Africa and the Indian subcontinent. Further development of antigen-based tests is leading to new products.5  



Complex Manifestations of Visceral Leishmaniasis

VISCERAL LEISHMANIASIS–HUMAN IMMUNODEFICIENCY VIRUS COINFECTION

Although the clinical manifestations in VL patients infected with human immunodeficiency virus (HIV) without severe immunosuppression are generally similar to those in immunocompetent patients, atypical clinical features can be found in patients with low CD4+ T-cell counts ( 4 cm (Plaque)

Lesion(s) < 4 cm (papule, nodule or ulcerated nodule) Situation 1

Situation 2

• L major, L. mexicana or self-curing lesion1 • No potentially disfiguring or disabling lesion (face, fingers, toes, etc.) • No uncontrolled diabetes or immunosuppression • Patient adheres to this option

All criteria above are present?

NO

Wash lesion + wound dressing Assure follow-up2

YES

• < 4 lesions for which the patient asks for treatment3 • Not L. braziliensis from Peru or Bolivia • No immunosuppression • Lesion site compatible with method4

NO

All criteria above are present? YES

YES

Cure ?

Situation 37

NO

Cryotherapy + intralesional Sb5 Topical paromomycin6

Cure ?

YES

NO

• No cardiac or other major disease (liver, kidney) • Age < 55

All criteria above are present? YES

NO

Systemic Sb

Miltefosine8

Cure ?

NO

YES

E-FIGURE 327-1.  Algorithm of treatment options for cutaneous leishmaniasis (CL)—an example of decision making. (1) Self-curing lesions show flattening or reduction in the

surface of the ulceration and/or induration. (2) Washing lesion, wound dressing, and follow-up performed in all situations. (3) Lymphatic dissemination in itself does not influence treatment decision, except when it increases the number of lesions requiring therapy. (4) Most lesions of limbs, trunk, cheek, upper cheek, chin, and front can be injected, including those close to large joints. Injection in ears, fingers, and toes is very painful. In children, premedication with inhaled equimolar nitrogen protoxide plus oxygen facilitates the injection. (5) When/where cryotherapy is not available, simpler physical methods can be adopted. Sb = pentavalent antimony. (6) In 2017, the validated formulation of topical paromomycin is available only through clinical trial or compassionate use. (7) In case of pregnancy, wait until after delivery or use liposomal amphotericin B. (8) Miltefosine should be used with caution or dose adaptation (or both) in case of liver or kidney involvement, and in case of gastrointestinal symptoms.

2094.e2

CHAPTER 327 Leishmaniasis  

GENERAL REFERENCES 1. Burza S, Croft SL, Boelaert M. Leishmaniasis. Lancet. 2018;392:951-970. 2. Sunyoto T, Boelaert M, Meheus F. Understanding the economic impact of leishmaniasis on households in endemic countries: a systematic review. Expert Rev Anti Infect Ther. 2019;17:57-69. 3. van Griensven J, Diro E. Visceral leishmaniasis: recent advances in diagnostics and treatment regimens. Infect Dis Clin North Am. 2019;33:79-99. 4. Hossain F, Ghosh P, Khan MAA, et al. Real-time PCR in detection and quantitation of Leishmania donovani for the diagnosis of visceral leishmaniasis patients and the monitoring of their response to treatment. PLoS ONE. 2017;12:1-16. 5. Ben-Abid M, Galai Y, Habboul Z, et al. Diagnosis of Mediterranean visceral leishmaniasis by detection of Leishmania-related antigen in urine and oral fluid samples. Acta Trop. 2017;167:71-72. 6. Zijlstra EE. The immunology of post-kala-azar dermal leishmaniasis (PKDL). Parasit Vectors. 2016;9:1-9. 7. Sundar S, Chakravarty J. An update on pharmacotherapy for leishmaniasis. Expert Opin Pharmacother. 2015;16:237-252. 8. Alves F, Bilbe G, Blesson S, et al. Recent development of visceral leishmaniasis treatments: successes, pitfalls, and perspectives. Clin Microbiol Rev. 2018;31:1-30. 9. den Boer M, Das AK, Akhter F, et al. Safety and effectiveness of short-course AmBisome in the treatment of post-kala-azar dermal leishmaniasis: a prospective cohort study in Bangladesh. Clin Infect Dis. 2018;67:667-675. 10. Meireles CB, Maia LC, Soares GC, et al. Atypical presentations of cutaneous leishmaniasis: a systematic review. Acta Trop. 2017;172:240-254. 11. Handler MZ, Patel PA, Kapila R, et al. Cutaneous and mucocutaneous leishmaniasis: differential diagnosis, diagnosis, histopathology, and management. J Am Acad Dermatol. 2015;73:911-926, 927-918.

12. Taxy JB, Goldin HM, Dickie S, et al. Cutaneous Leishmaniasis: contribution of routine histopathology in unexpected encounters. Am J Surg Pathol. 2019;43:195-200. 13. Solomon M, Sahar N, Pavlotzky F, et al. Mucosal leishmaniasis in travelers with Leishmania braziliensis complex returning to Israel. Emerg Infect Dis. 2019;25:642-648. 14. Uribe-Restrepo AF, Prieto MD, Cossio A, et al. Eligibility for local therapies in adolescents and adults with cutaneous leishmaniasis from southwestern Colombia: a cross-sectional study. Am J Trop Med Hyg. 2019;100:306-310. 15. Aronson NE, Joya CA. Cutaneous leishmaniasis: updates in diagnosis and management. Infect Dis Clin North Am. 2019;33:101-117. 16. Berbert TRN, de Mello TFP, Wolf Nassif P, et al. Pentavalent antimonials combined with other therapeutic alternatives for the treatment of cutaneous and mucocutaneous leishmaniasis: a systematic review. Dermatol Res Pract. 2018;2018:1-21. 17. Wolf Nassif P, DE Mello TFP, Navasconi TR, et al. Safety and efficacy of current alternatives in the topical treatment of cutaneous leishmaniasis: a systematic review. Parasitology. 2017;144: 995-1004. 18. Pedras MJ, Carvalho JP, Silva RED, et al. Mucosal leishmaniasis: the experience of a Brazilian referral center. Rev Soc Bras Med Trop. 2018;51:318-323. 19. Guéry R, Henry B, Martin-Blondel G, et al. Liposomal amphotericin B in travelers with cutaneous and muco-cutaneous leishmaniasis: not a panacea. PLoS Negl Trop Dis. 2017;11:1-12. 20. Le Rutte EA, Chapman LAC, Coffeng LE, et al. Policy recommendations from transmission modeling for the elimination of visceral leishmaniasis in the Indian Subcontinent. Clin Infect Dis. 2018; 66:S301-S308.

CHAPTER 327 Leishmaniasis  

REVIEW QUESTIONS 1. A 60-year-old woman has been experiencing fever without chills for 10 days, starting 1 month after a 6-month travel from China to Europe. Clinical examination shows splenomegaly and liver enlargement. Laboratory parameters show anemia, leucopenia, and thrombocytopenia. A . The two most likely diagnosis hypotheses are a hematologic condition and visceral leishmaniasis. B. The two most likely diagnosis hypotheses are bacterial infection and Chagas disease. C. The two most likely diagnosis hypotheses are viral infection and histoplasmosis. D. None of the above proposals is accurate. Answer: A  The association of fever splenomegaly with pancytopenia is highly suggestive but not specific of visceral leishmaniasis. Several hematologic conditions (myeloproliferative disorders, splenic lymphoma, Castelman syndrome) present the same association of signs and laboratory alterations. Bacterial conditions and acute Chagas disease are infrequently associated with pancytopenia, so this option is possible but less frequent. Viral infections (HIV, HHV8, CMV) and histoplasmosis may present with splenomegaly and pancytopenia, but these presentations are less prevalent in series of fever of unknown origin (at least in the European context), so this option is possible but less frequent. 2. A 3-year-old child has fever, splenomegaly, anemia, leucopenia, and thrombocytopenia. Direct examination of a bone marrow aspirate (Giemsa-stained smear) for Leishmania amastigotes is negative. The most sensitive and specific safe laboratory test to confirm the diagnosis of visceral leishmaniasis is: A . Quantitative polymerase chain reaction (PCR) for Leishmania on a blood sample B. Search for antileishmanial antibody C. Wait for Leishmania culture on the bone marrow aspirate to become positive D. Direct examination of a splenic aspirate (Giemsa-stained smear) E. The direct examination of the bone marrow aspirate is sensitive enough to rule out the diagnosis of visceral leishmaniasis Answer: A  Quantitative PCR is a highly sensitive and specific test as soon as thresholds for symptomatic disease have been defined (low-grade positivity is frequent in endemic areas). Antileishmanial antibody can be positive because of past infection. Culture is also a sensitive test, although generally less so than an optimized PCR based on the amplification of kinetoplastic DNA. Splenic aspirate is not a safe procedure.

2094.e3

3. A 22-year-old HIV-seropositive male patient with fever, splenomegaly, and pancytopenia has visceral leishmaniasis confirmed by quantitative PCR of a peripheral blood sample. The following items suggest a complicated form: A . Mucosal hemorrhages B. Presence of a splenomegaly 5 cm below the costal margin C. Cough D. Moderate thrombocytopenia Answer: A  Hemorrhage and bacterial superinfection are the two main complications of advanced visceral leishmaniasis. Splenomegaly and mild to moderate thrombocytopenia are present in most patients with uncomplicated visceral leishmaniasis. Dry cough is present in 10 to 15% of patients with visceral leishmaniasis (but will justify exploration to rule out a bacterial superinfection in the lung). 4. A 2-year-old child has parasitologically confirmed, uncomplicated visceral leishmaniasis. The treatment option with the most favorable therapeutic window is: A . Liposomal amphotericin B 10 mg/kg/day on Day 1 and Day 2 by slow IV infusion B. Pentamidine mesylate 4 mg/kg/day for 10 days by slow IV infusion C. Amphotericin B deoxycholate 0.5 mg/kg/day for 14 days by slow IV infusion D. Meglumine antimoniate (pentavalent antimony) 20 mg SbV/kg/day for 28 days by slow IV infusion E. Miltefosine 2.5 mg/kg for 28 days orally Answer: A  All these options have been used in the therapy of visceral leishmaniasis. The 2-day schedule of liposomal amphotericin B has been validated in children. (Syriopoulou V, Daikos GL, Theodoridou M, et al. Two doses of a lipid formulation of amphotericin B for the treatment of Mediterranean visceral leishmaniasis. Clin Infect Dis. 2003;36:560-566.) 5. A 35-year-old female patient has a subacute, painless, crusted ulceration of the forearm that appeared 1 month after travelling to Israel. The following item makes the diagnosis of cutaneous leishmaniasis unlikely: A . The lesion reached its maximum size in less than 1 week. B. The lesion is infiltrated. C. The lesion has had a red painful border for the last 4 days. D. The lesion has an oval shape. E. The lesion has small peripheral papules. Answer:  Cutaneous leishmanisis has typically a slow evolution. Lesions reach their maximal size most often in more than 3 weeks. The most frequent differential in travelers is pyodermitis related to cocci gram-positive bacteria (streptococcus and staphylococcus) which generally have a fast initial evolution. Almost all lesions of cutaneous leishmaniasis (except the macular form of post-kala-azar dermal leishmaniasis) are infiltrated. Patent bacterial superinfection of CL lesions affect 5 to 10% of patients. Oval shape is typical (though not specific) of CL, small peripheral papules are not infrequent. (https:// apps.who.int/iris/handle/10665/120002 Manual for case management of cutaneous leishmaniasis in the WHO Eastern Mediterranean Region.)

2094

CHAPTER 328 Toxoplasmosis  

328  TOXOPLASMOSIS JOSÉ G. MONTOYA



DEFINITION

Toxoplasma gondii is a protozoan parasite that infects most species of warmblooded animals, including humans, and can cause toxoplasmosis. The term toxoplasmosis is reserved for the disease, when clinical manifestations are present, whereas Toxoplasma infection is preferred for the asymptomatic presence of the parasite. Toxoplasmosis may result in significant morbidity and mortality of the fetus, newborn, and immunocompromised patient. However, toxoplasmosis can also manifest in immunocompetent patients as chorioretinitis, lymphadenopathy, pneumonia, brain abscesses, myositis, myocarditis, and hepatitis. In animal models and epidemiological studies in humans, Toxoplasma infection has been associated with behavioral changes, mental illness, and neurocognitive impairment, although compelling, conclusive evidence for these associations has not yet been produced. A more aggressive form of congenital and adult toxoplasmosis appears to occur in certain geographic locales in Latin America, where pneumonia, fever of

2094.e4

ABSTRACT

CHAPTER 328 Toxoplasmosis  

Toxoplasmosis is caused by an intracellular parasite, Toxoplasma gondii, that can infect any warm-blooded animal. Most humans get infected via the oral route. However, transmission through the placenta, a transplanted organ, or laboratory accident is also possible. Up to 50% of acutely infected individuals do not endorse known risk factors or symptoms. Toxoplasmosis can result in serious morbidity and mortality in those with immature or impaired immune systems such as the fetus (from mothers infected during gestation) or immunocompromised patients (e.g., AIDS, cancer, transplant recipients, patients on corticosteroids, biologics). Toxoplasmosis can also severely affect those with ocular disease and immunocompetent patients infected with virulent strains. Universal screening for Toxoplasma IgG and IgM should be implemented in all pregnant women and immunocompromised patients. Those with negative IgG/IgM should be counseled on how to avoid exposure to the parasite. Those with positive IgG/negative IgM are considered as having protective immunity. Those with positive IgM should undergo confirmatory testing to see whether patient is acutely infected, and if so, potentially treated. Presence of toxoplasma tachyzoites or DNA in any body fluid is diagnostic of toxoplasmosis. Common regimens for the treatment of toxoplasmosis in humans include pyrimethamine (+folinic acid) plus sulfadiazine, or clindamycin, or atovaquone. Trimethoprim/ sulfamethoxazole can also be used. Although toxoplasmosis can cause lifethreatening syndromes, it is preventable and treatable with very good outcomes, if treated early.

KEYWORDS

Toxoplasmosis Toxoplasma gondii pregnancy immunocompromised ocular fetus

CHAPTER 328 Toxoplasmosis  

A

2095

B

C FIGURE 328-1.  Toxoplasma gondii exists in nature primarily in three forms. A, Tachyzoites are bow shaped, measure 2 to 3 µm wide and 5 to 7 µm long, and can be stained with Wright-Giemsa stain. B, Oocysts isolated from cat’s feces are subspherical to spherical and measure 10 × 12 µm in diameter. C, Tissue cyst observed in human myocardial tissue stained with hematoxylin and eosin. Tissue cysts vary in shape and size and may reach more than 100 µm.

unknown origin, brain abscesses, and death have been reported in HIV-negative and otherwise immunocompetent individuals. Recent epidemiologic studies have established the role of novel risk factors for the acquisition of the acute infection, including the ingestion of untreated water, oysters, mussels, or clams.

The Pathogen

T. gondii is an intracellular parasite with a high capacity for host cell invasion due to a motile invasive form with capacity for rapid replication (tachyzoite or trophozoite) that is equipped with an evolutionarily unique apical complex and a mechanism of actin-based gliding motility.1 The sexual cycle of the parasite only takes place in the small intestine of animals belonging to the feline family, the definitive host. The asexual cycle takes place in tissues and cells of its intermediate hosts (e.g., humans). In nature, the parasite exists in several forms, including the tachyzoite (Fig. 328-1A), the oocyst that contains sporozoites (Fig. 328-1B), the tissue cyst that contains bradyzoites (Fig. 328-1C). The tachyzoite is the rapidly proliferating form of the parasite responsible for the clinical manifestations of toxoplasmosis observed in the setting of the acute infection or reactivation of a latent infection. The tissue cyst is the slower metabolic form of the parasite responsible for chronic infection and for its transmission through meat consumption in humans and animals. It has been thought that tissue cysts persist in tissues for the life of the host and cannot be eradicated by currently available drugs, but the notion of persistence of the parasite for the life of the host has been recently challenged.3,4 Tissue cysts vary in shape and size from younger ones that contain only a few bradyzoites to older tissue cysts that may contain several thousand bradyzoites and may reach more than 100 µm in size. The central nervous system (CNS), eye, and skeletal, smooth, and heart muscles appear to be the most common sites of tissue cyst formation (i.e., latent infection). Oocysts are primarily responsible for the worldwide and large-scale spread of the parasite among different populations of other animals and humans. Domestic and feral animals belonging to the Felidae family shed oocysts after they ingest any of the infectious forms of the parasite: tachyzoites, tissue cysts, and oocysts. As many as 10 million oocysts may be shed in the feces of an infected animal in a single day for periods varying from 7 to 20 days. Oocysts may remain viable for as long as 18 months in moist soil; this results in an environmental reservoir from which incidental hosts may be infected.



EPIDEMIOLOGY

The prevalence of T. gondii infection varies significantly according to geographic locale and the socioeconomic status of the population. It can be as low as 7% in England and as high as 78% in Brazil. Seroprevalence increases with age because of increasing length of exposure with age, and it is inversely associated with socioeconomic status because of the strong influence of hygienic and alimentary habits in the transmission of the parasite. The overall age-adjusted seroprevalence of T. gondii infection in the United States has been recently reported at 11%, but it may be higher in certain geographic areas, ethnic, and socioeconomic groups. The seroprevalence of the parasite has declined during the past 30 years in the United States and several other countries5 but appears to be stable or increasing in certain geographic locales, such as in the tropics (e.g., Latin America). Humans and nonfelid animals are incidental hosts and become infected primarily by the ingestion of infected meat containing tissue cysts or of contaminated food, water, or soil material containing oocysts. They can also become infected during gestation by vertical transmission of the parasite from the mother to her offspring. In addition, humans can become infected through organ transplantation and, more rarely, in the setting of laboratory accidents. Ingestion of raw or undercooked meat contaminated with tissue cysts, untreated water, food, or soil contaminated with oocysts are the major routes of infection of humans with the parasite. Untreated water has been found to be the source of large epidemics of toxoplasmosis in Canada and Brazil. The main risk factors for T. gondii infection in the United States include eating raw ground beef, rare lamb; eating locally produced cured, dried, or smoked meat; eating raw oysters, clams, or mussels; working with meat; drinking unpasteurized goat’s milk; and having three or more domestic kittens. Untreated water as a potential vehicle for the transmission of T. gondii has been established in several large epidemiologic studies and was found to have a trend toward increased risk for acute infection in the United States. In up to 50% of individuals acutely infected with T. gondii, it is not possible to identify the presence of a known risk factor for their acute infection. Thus, attempting to establish whether a patient is at risk for toxoplasmosis solely on the basis of the epidemiologic history is a futile task. The possibility of toxoplasmosis should not be excluded from the differential diagnosis of patients presenting with syndromes suggestive of the disease solely on the basis of a negative epidemiological history.

CHAPTER 328 Toxoplasmosis  

T. gondii maintains a clonal population structure in certain areas of the world that consists of three main lineages: types I, II, and III, and atypical strains. However, recent studies have reported on the wider genetic diversity that is observed in North and Latin America. It has also been suggested that the parasite originated in South America.2 In Europe, type II strains predominate and are most commonly associated with human toxoplasmosis, both in congenital infections and in immunocompromised patients. In North America, types II, I/III, and atypicals appear to be equally common. In Latin America,

2095.e1

type I and atypical strains are significantly more common, with most of these reported from human infections in Brazil and French Guiana. These atypical Latin American strains, initially called exotic strains, belong to several haplogroups that are endemic to Latin America and have been found to be associated with more severe clinical manifestations; physicians need to entertain the diagnosis of toxoplasmosis in their patients presenting with pneumonia, fever of unknown origin, brain abscesses, lymphadenopathy, or chorioretinitis who are from or were traveling in Latin America and Caribbean islands.

2096

CHAPTER 328 Toxoplasmosis  

The seroprevalence of T. gondii infection in immunocompromised patients reflects the seroprevalence of the particular population from which they come. Latent T. gondii infection can reactivate in these patients, particularly in those with the acquired immunodeficiency syndrome (AIDS) and in hematopoietic stem cell, kidney, heart, and liver transplant recipients. In these patients, it is important to establish whether they have been infected with the parasite before severe immunosuppression ensues or their transplant procedure because serologic testing in severe immunosuppression or after transplantation may be unreliable. Approximately 30% of AIDS patients who are infected with T. gondii will develop toxoplasmosis by reactivation of their chronic infection if their CD4 count falls below 200 cells/µL and they are not taking antiToxoplasma primary prophylaxis. The advent of highly active antiretroviral therapy, in addition to the use of primary anti-Toxoplasma prophylaxis, has clearly contributed to the decline in the incidence of toxoplasmosis in AIDS patients. Among hematopoietic stem cell transplant (HSCT) patients, those recipients who are Toxoplasma seropositive before allogeneic transplantation and then develop graft-versus-host disease have the highest risk for reactivation. For solid organ transplants, the highest risk for toxoplasmosis is observed when an allograft from a Toxoplasma-seropositive donor (D+) is transplanted into a seronegative recipient (R−). In D+/R− patients, there is a 25% risk for development of potentially life-threatening toxoplasmosis if effective anti-Toxoplasma prophylaxis is not instituted. It is highly advised that the Toxoplasma serologic status of the donor and the recipient be established before transplantation. Serologic test results are less reliable in the posttransplantation period, and they may significantly vary without any clinical relevance. Transmission of T. gondii to the fetus can occur during pregnancy when a woman acquires her primary infection during gestation. The incidence of seroconversion for pregnant women in the United States has been estimated at 0.27%. The overall rate of transmission of the parasite (prevalence of congenital toxoplasmosis) in seroconverting women has been estimated between 50 and 60% before spiramycin was instituted as an attempt to decrease vertical transmission and 25 to 30% thereafter. The transmission rate increases with the gestational age at which maternal infection is acquired. In women who have been treated for toxoplasmosis during gestation, it can be as low as 4.5% during the first trimester, 31.7% during the second trimester, and as high as 63% during the third trimester. The likelihood of severe disease is inversely proportional to the gestational age at which maternal infection was acquired. Although objective data are lacking on the prevalence of congenital toxoplasmosis in the United States, it has been estimated that among the approximately 4.0 million live births per year, congenital T. gondii infection occurs in 400 to 5000 newborns. The global annual incidence of congenital toxoplasmosis6 has been estimated to be 190,100 cases. This is equivalent to a burden of 1.2 million disability-adjusted life years. Particularly high burdens are seen in South America and in some Middle Eastern and low-income countries. Toxoplasmosis also causes significant morbidity and mortality in the United States.7  

PATHOBIOLOGY

Pathogenesis

After oral infection with tissue cysts (e.g., contaminated meat) or oocysts (e.g., contaminated soil, water, or food), the wall of both infectious forms is disrupted by the digestive juices of the gastrointestinal tract. Bradyzoites (from cysts) and sporozoites (from oocysts) are released and converted to the rapidly replicating tachyzoite form. Tachyzoites have the capacity to infect contiguous cells or distant tissues by hematogenous and lymphatic spread. Tachyzoites appear to actively and rapidly migrate across epithelial cells and may traffic to distant sites while they are extracellular (acute infection) or within infected immune cells subverted by the parasite to facilitate its own spread (i.e. Trojan horse hypothesis).8 The formation of the parasite vacuole within infected cells appears to be an important mechanism for Toxoplasma survival and subversion.9 The histologic hallmark of tachyzoites is necrosis surrounded by inflammation. In immunocompetent individuals, the immune system controls the proliferation of the tachyzoite and induces its conversion to bradyzoites, facilitating the final formation of tissue cysts (chronic infection). Tissue cysts persist for the life of most infected individuals, and T. gondii can be isolated from tissues of individuals who have died from causes other than toxoplasmosis.

Pathology

Most of the data on the pathology of toxoplasmosis come from studies of congenitally infected babies and immunosuppressed patients. CNS lesions of

patients with toxoplasmosis are characterized by significant necrosis and surrounding inflammation. In congenitally infected cases, necrotic areas may calcify and lead to typical radiographic findings suggestive but not diagnostic of toxoplasmosis. Hydrocephalus may result from obstruction of the aqueduct of Sylvius or foramen of Monro by very high protein concentrations (e.g. >1000 mg/dL). Tachyzoites and tissue cysts may be visualized near necrotic foci, near or in glial nodules, in perivascular regions, and in cerebral tissue uninvolved by inflammatory changes. Formation of multiple brain abscesses is relatively common in patients with AIDS. In the areas around the abscesses, edema, vasculitis, hemorrhage, and cerebral infarction secondary to vascular involvement may also be present. Important associated features in toxoplasmic encephalitis are arteritis, perivascular cuffing, and astrocytosis. A “diffuse form” of toxoplasmic encephalitis has been described with histopathologic findings of widespread microglial nodules without abscess formation in the gray matter of the cerebrum, cerebellum, and brain stem. Pulmonary involvement by T. gondii in the immunodeficient patient can lead to interstitial pneumonitis, necrotizing pneumonitis, consolidation, pleural effusion or empyema, or all of these. Chorioretinitis in AIDS patients is characterized by segmental panophthalmitis and areas of coagulative necrosis associated with tissue cysts and tachyzoites. Toxoplasmic lymphadenitis in immunocompetent individuals may result in patterns of findings that are often diagnostic of the disease: a reactive follicular hyperplasia; irregular clusters of epithelioid histiocytes encroaching on and blurring the margins of the germinal centers; and focal distention of sinuses with monocytoid cells.  

CLINICAL MANIFESTATIONS

Toxoplasmosis should be entertained in the differential diagnosis of several clinical syndromes in immunocompetent, unborn, newborn, infant, pediatric, adult, and immunocompromised patients (Table 328-1). Symptoms result from the primary infection or reactivation of the parasite due to T-cell-mediated or severe B-cell-mediated immunodeficiency. Primary infection can be asymptomatic in a significant number of individuals, and conventional risk factors for the acute infection may not be present in a particular patient. Thus, the possibility of acute toxoplasmosis or T. gondii infection should not be ruled out because of the absence of epidemiologic risk factors (e.g., exposure to cats or undercooked meat) or symptoms in a given patient. For this reason, if the goal is to detect each case of primary T. gondii infection in a population of patients (e.g., pregnant women), only systematic and universal screening methods can achieve such an objective; testing of only symptomatic patients or those with conventional epidemiologic risk factors will miss a significant number of acute cases. Severity of toxoplasmosis due to primary infection or reactivation in a given patient or population may be influenced by the infecting strain, size of the inoculum, infectious form (e.g., oocyst vs. cyst), genetics of the host (e.g., presence of HLA-DQ3), or net state of immunosuppression. Patients infected in certain geographic locales (e.g., South America) have more aggressive clinical presentations, including a more severe primary infection and disease due to reactivation. These observations need to be kept in mind on seeing ill travelers returning from those endemic areas or patients who were born in those areas and in whom toxoplasmosis by reactivation has been included in their differential diagnosis. Lymphadenopathy due to toxoplasmosis may be completely asymptomatic or be accompanied by other symptoms, such as fever (temperature as high as 104° F), headache, general malaise, and fatigue. It can be localized or generalized. A solitary, occipital, and painlessly enlarged lymph node can be the sole manifestation of toxoplasmosis in a child, pregnant woman, or adult. However, more generalized cervical, axillary, and abdominal lymphadenopathy has also been reported. Lymph nodes are usually 1 to 3 cm in size, nonsuppurative, and nontender on palpation. They usually regress within 12 weeks, but a mild relapse of the lymphadenopathy has been observed between months 3 and 6. Recurrence of toxoplasmic lymphadenopathy beyond the sixth month is extremely rare. Ocular disease due to T. gondii can be asymptomatic or symptomatic and can be the result of congenital or postnatally acquired infection.10 In both settings (congenitally and postnatally acquired), toxoplasmic chorioretinitis can be discovered at the time of the diagnosis of the infection or as a reactivation of the subsequent latent infection months to years later. Up to 17% of patients acutely infected with the parasite in Brazil and in a Canadian outbreak of toxoplasmosis presented with concurrent symptomatic toxoplasmic chorioretinitis at the time their acute infection was diagnosed. Similar cases have

CHAPTER 328 Toxoplasmosis  

Congenital transmission in women who were infected before conception has only rarely been reported in immunosuppressed patients, in those who were acutely infected shortly before conception (i.e., within 3 months of conception), and in those who have been reinfected with a more virulent strain.

2096.e1

It appears that the activation of well-orchestrated immune responses is required for the successful resistance against T. gondii. Innate, adaptive humoral, and cellular immune responses likely to be involved in preventing the uncontrolled proliferation of tachyzoites include activation of the monocyte-macrophage system, dendritic cells, natural killer cells, and αβ and γδ T cells; T. gondii–specific and cytotoxic CD4+ and CD8+ T cells; and interferon-γ, interleukin-12, tumor necrosis factor-α, interleukin-10 and other cytokines, transforming growth factor-β, costimulatory molecules (e.g., CD28, CD40 ligand), and, to some degree, immunoglobulins. Studies have shown that innate type 1 immune responses that involve interferon-γ-producing natural killer cells and neutrophils, rather than interferon-γ-producing T cells, predetermine host resistance to T. gondii. More recently, it has been suggested that T. gondii infection can induce long-term immune alterations associated with impaired thymic function resulting in a state of relative immune deficiency that facilitates parasite dissemination and persistence. In immunocompromised patients previously infected with T. gondii, overt decrease in T-cell-mediated or severe impairment in B-cell-mediated immune responses can facilitate the reactivation of their infection (i.e., conversion of bradyzoites in their tissue cysts into rapidly proliferating tachyzoites). Toxoplasmosis in this setting is 100% lethal if untreated.

CHAPTER 328 Toxoplasmosis  

2097

TABLE 328-1 CLINICAL MANIFESTATIONS OF TOXOPLASMOSIS IN HUMANS CLINICAL CATEGORIES Primary infection Immunocompetent individuals and pregnant women

Congenital toxoplasmosis Fetus

Newborn

Children and adults

CLINICAL MANIFESTATIONS AND SYNDROMES Most patients are asymptomatic. However, in ≈10% of patients, the following symptoms or syndromes, alone or in various combinations, have been reported: fever, lymphadenopathy, headache, myalgias, arthralgias, sore throat, stiff neck, nausea, abdominal pain, anorexia, rash, confusion, earache, eye pain, general malaise, fatigue. Chorioretinitis resulting in blurred vision, eye pain, decreased visual acuity, floaters, scotoma, photophobia, or epiphora Hepatitis; myositis; myocarditis Disseminated disease, pneumonia, brain abscesses, and even death have been observed in immunocompetent individuals infected with atypical strains of Toxoplasma gondii (e.g., in Latin America). Ultrasound study can be normal or reveal hydrocephalus, brain or hepatic calcifications, splenomegaly, ascites, pericarditis. Fetal death can also result from overwhelming infection. Newborn can be entirely normal, have a nonspecific illness, or have abnormal findings on physical examination including chorioretinitis, strabismus, blindness, seizures, encephalitis, abnormal cephalic perimeter (microcephaly or hydrocephalus), psychomotor or mental retardation, hepatosplenomegaly, pneumonitis, diarrhea, hypothermia, jaundice, petechiae, rash. Intracranial calcifications can be present in brain imaging studies. Newborns can also die as a result of overwhelming infection. Children can continue to suffer the chronic sequelae of the congenital disease. However, children may be born apparently normal and become symptomatic for the first time during childhood, adolescence, or adulthood, primarily in the form of reactivation of congenitally acquired chorioretinitis.

Chronic infection

Asymptomatic. However, some investigators have proposed a role of chronic infection in individuals with schizophrenia, bipolar disease, and behavioral issues including a higher incidence of motor vehicle accidents. Chorioretinitis can occur as a reactivation of congenital or postnatally acquired disease in otherwise immunocompetent individuals.

Reactivation of chronic infection in immunocompromised patients

Multiple brain abscesses, diffuse encephalitis, seizures, chorioretinitis, fever of unknown origin, pneumonia, myocarditis, hepatosplenomegaly, lymphadenopathy, rash

been described in Europe and the United States. T. gondii strain type appears to be a contributing factor determining severity and recurrence of ocular toxoplasmosis. Symptomatic ocular disease primarily consists of a retinochoroiditis that can result in blurred vision, eye pain, decreased visual acuity, floaters, scotoma, photophobia, or epiphora. The morphology of the retinal lesions on funduscopic examination is thought to be characteristic of toxoplasmosis. An active whitish infiltrate is usually attached to the darkly pigmented border of an older scar (Fig. 328-2). However, retinal lesions tend to be less typical in older or immunocompromised patients. Other less common but well-documented syndromes have been associated with the acute infection, including hepatitis, myositis, myocarditis, and skin lesions. More aggressive disease, including pneumonia, brain abscesses, and death, has been observed in immunocompetent patients in Latin America. Primary infection can be observed in solid organ transplant patients when an allograft from a seropositive donor is transplanted into a seronegative recipient (D+/R−). Disseminated and localized toxoplasmosis has been reported in this setting, including myocarditis, pneumonia, fever of unknown origin, and encephalitis.

FIGURE 328-2.  The morphology of the retinal lesions on funduscopic examination believed to be characteristic of toxoplasmic retinochoroiditis. An active whitish infiltrate is usually attached to the darkly pigmented border of an older scar.

Congenital disease can be asymptomatic in the fetus, newborn, child, or adult. However, most of the infected offspring will eventually develop signs and symptoms of toxoplasmosis (see Table 328-1). The classic triad of chorioretinitis, hydrocephalus (Fig. 328-3A and B), and brain calcifications is highly suggestive of toxoplasmosis and is primarily seen in babies whose mothers have not been treated against the parasite during gestation. Eye examination by an experienced pediatric ophthalmologist may reveal active or inactive toxoplasmic chorioretinitis. New lesions have been reported in up to 30% of congenitally infected children observed up until 11 years of age when their mothers have been treated but in up to 70% when their mothers have not. Chronic infection is believed to be asymptomatic. However, several studies have recently suggested the possibility that chronic infection may play a role in the predisposition of infected individuals to have a higher frequency of traffic accidents, mental illness (e.g. schizophrenia, bipolar disease), and behavioral abnormalities. Neurocognitive impairment has been associated with latent T. gondii infection in HIV-positive patients whose HIV infection has been successfully controlled with antiretroviral therapy.11 Overt reactivation of the chronic infection is usually observed in patients with significant impairment of T-cell-mediated immunity or severe impairment of B-cell-mediated immunity. Toxoplasmosis by reactivation can cause brain abscesses, diffuse encephalitis, seizures, chorioretinitis, fever of unknown origin, pneumonia, myocarditis, hepatosplenomegaly, lymphadenopathy, and rash. Although multiple brain abscesses (Fig. 328-3C) are commonly described in patients with toxoplasmic encephalitis, diffuse encephalitis without spaceoccupying lesions by magnetic resonance imaging has been reported with a very high mortality. Fever with pneumonia can be the sole manifestation of toxoplasmosis in immunocompromised patients, including hematopoietic stem cell transplant (HSCT) and solid organ transplant recipients. Toxoplasmic pneumonitis can be manifested by cough, dyspnea, hypoxia, and diffuse bilateral or localized infiltrates. Most patients with toxoplasmic pneumonia have been reported to have bilateral ground-glass opacities that can be confused with Pneumocystis pneumonia, viral etiologies, atypical pneumonia, or strongyloidiasis. Fever alone has frequently been described in patients with allogeneic HSCT and liver transplant patients. Reactivation in heart tissue causing congestive heart failure, arrhythmias, and pericarditis has been described.  

DIAGNOSIS

Laboratory methods for the diagnosis of T. gondii infection and toxoplasmosis include serologic tests, polymerase chain reaction (PCR), microscopic examination of tissue and body fluids, and attempts to isolate the parasite (Table 328-2).12 The first step is to establish whether the patient has never been infected with Toxoplasma or has an acute or latent T. gondii infection; this can be accomplished by serologic testing. Serologic tests can determine this infection status regardless of the presence or absence of symptoms. Available serologic tools include methods to detect T. gondii–specific immunoglobulin G (IgG)-, IgM-, IgA-, IgE-, and IgG-based avidity and differential agglutination (AC/HS).

2098

CHAPTER 328 Toxoplasmosis  

A

B

C

FIGURE 328-3.  Radiologic manifestations of central nervous system toxoplasmosis. A, Fetal ultrasound of a fetus congenitally infected with Toxoplasma gondii in the United States reveals hydrocephalus. B, Computed tomography scan of the brain of a newborn congenitally infected with T. gondii in the United States reveals hydrocephalus and calcifications. C, Magnetic resonance image of the brain of an AIDS patient revealing multiple ring-enhancing brain lesions.

TABLE 328-2 LABORATORY METHODS FOR THE DIAGNOSIS OF TOXOPLASMA GONDII INFECTION AND TOXOPLASMOSIS* IN HUMANS METHOD Serologic tests IgG IgM

Confirmatory testing for positive IgM test results

CLINICAL INTERPRETATION A positive test result establishes that the patient has been infected with T. gondii. However, a negative test result can be seen in patients infected within 4 weeks before serum sampling or in patients unable to produce IgG (e.g., immunocompromised hosts). A positive test result suggests but is not necessarily diagnostic of an acute infection. Sera with positive IgM test results should be sent to a reference laboratory for confirmatory testing that includes a more specific IgM assay and additional tests, including avidity, acetone (AC)-fixed versus formalin (HS)-fixed (AC/HS) differential agglutination test of tachyzoites, and IgA and IgE.† Positive IgM test results can be seen in chronically infected patients because of persistence of the IgM response or false-positive results observed in certain commercial kits. IgG avidity test; differential agglutination (AC/HS); IgA, IgE performed at a reference laboratory.† At PAMF-TSL, a high IgG avidity test result‡ or a nonacute AC/HS test result indicates that the patient has been infected for more than 4 mo (avidity) or 12 mo (AC/HS).

Polymerase chain reaction (PCR)

The B1 gene and the 529-bp repetitive element are the most commonly used gene targets for amplification. PCR test can be performed in any body fluid, including amniotic fluid, peripheral blood, cerebrospinal fluid, bronchoalveolar lavage fluid, vitreous fluid, aqueous humor, peritoneal fluid, pleural fluid, ascitic fluid, and urine. PCR can also be performed in any tissue. A positive test result in any body fluid establishes that the patient has either acute or reactivated toxoplasmosis. However, a positive PCR test result in tissue is more difficult to interpret because it does not differentiate symptomatic toxoplasmosis from a latent infection. Although DNA extraction is more cumbersome, it can be attempted in paraffin-embedded tissue.

Direct visualization of the parasite

Identification of tachyzoites in any body fluid or tissue is diagnostic of toxoplasmosis due to acute infection or reactivation of a chronic infection. Tachyzoites can be identified by hematoxylin and eosin or cytologic studies but are better visualized with Wright-Giemsa and immunoperoxidase stains. Identification of cysts by hematoxylin and eosin or immunoperoxidase stains confirms the presence of T. gondii in the host but does not necessarily establish that the patient has toxoplasmosis. However, a strong inflammatory response surrounding the cysts is highly suggestive of toxoplasmosis, possibly explaining the patient’s symptoms.

Attempts to isolate the parasite

A positive isolation study in any body fluid establishes the diagnosis of toxoplasmosis. Isolation of T. gondii in cell cultures or peritoneal cavity of mice can be attempted at reference laboratories. These studies can be important in trying to establish a correlation between the genetics of the parasite and its clinical manifestations.

Histology of the lymph node

Classic histologic triad is considered diagnostic: follicular hyperplasia, epithelioid histiocytes impinging on the margins of the germinal centers, and monocytoid cells focally distending sinus walls.

*T. gondii infection = asymptomatic presence of the parasite. Toxoplasmosis = active symptoms and signs are present. † For example, Palo Alto Medical Foundation Toxoplasma Serology Laboratory (PAMF-TSL), Palo Alto, Calif; www.pamf.org/serology; 650-853-4828; [email protected]. ‡ The window of exclusion for acute infection varies for different avidity kits (usually between 3 and 5 months).

With the use of commercial kits for the detection of IgG and IgM, most hospital-based or commercial laboratories can reliably diagnose the absence of T. gondii infection (negative IgG/negative IgM) and the presence of chronic infection (positive IgG/negative IgM). However, the diagnosis of acute infection is more challenging. A positive IgM test result is observed during the acute infection, but it may remain positive for months to years in certain individuals without any apparent clinical relevance. In addition, commercial IgM kits have been designed to be extremely sensitive so that an acute infection will be rarely missed; as a consequence, their specificity is somewhat sacrificed. Of patients who are found to be IgM positive at hospital-based or commercial laboratories, 78% are found to be chronically infected when their serum is tested at the national reference laboratory for the study and diagnosis of toxoplasmosis in the United States (Palo Alto Medical Foundation Toxoplasma Serology Laboratory [PAMF-TSL], Palo Alto, Calif; www.pamf.org/

serology; 650-853-4828; [email protected]).13 At PAMF-TSL, a battery of confirmatory tests (avidity, differential agglutination, IgA, IgE) are performed in addition to the “gold standard” dye test for IgG and the “double” sandwich capture enzyme-linked immunosorbent assay for IgM for confirmatory testing of positive IgM test results obtained at hospital-based or commercial laboratories. These tests are used in various combinations, depending on the clinical scenario of each patient and the questions of the treating physician. For an appropriate interpretation of the serologic test results obtained at PAMF-TSL, it is also crucial to have relevant clinical information available for the medical consultant. (For instance, low positive IgG and positive IgM test results with a high IgG avidity test result will mean no risk for congenital toxoplasmosis for a 16-week pregnant woman, but the same results can be highly supportive of the diagnosis of toxoplasmic encephalitis for an AIDS patient with multiple brain lesions.) At PAMF-TSL, three interpretations can be given to final

CHAPTER 328 Toxoplasmosis  

serologic test results: (1) acute, results are consistent with a recently acquired infection; (2) chronic, consistent with an infection acquired in the distant past; and (3) equivocal, cannot exclude a recently acquired infection—an earlier or subsequent sample is required to attempt to establish whether the infection is acute or chronic. For serologic test results consistent with an acute infection, an attempt is made by the medical consultants at PAMF-TSL to establish the approximate date that the infection was acquired. The definitive diagnosis of toxoplasmosis (due to primary infection or reactivation of a chronic infection) requires the identification of tachyzoites in tissues or body fluids or the amplification of parasite DNA in any body fluid (see Table 328-2). Tachyzoites can be visualized in histologic sections stained with hematoxylin and eosin or in cytologic preparations without any specific staining, but they are better visualized with Wright-Giemsa (see Fig. 328-1A) and T. gondii–specific immunoperoxidase stains. Real-time PCR (in any body fluid including cerebrospinal) has become a useful method for the diagnosis of toxoplasmosis in immunocompromised patients and for the prenatal diagnosis of congenital toxoplasmosis (in amniotic fluid). Isolation of the parasite in any body fluid is also diagnostic of toxoplasmosis and can be attempted at reference laboratories. The diagnosis of toxoplasmosis can be indirectly supported by the use of serologic tools, demonstration of cysts in tissues (see Fig. 328-1C) surrounded by an inflammatory response, and attempts to isolate the parasite (see Table 328-2).

Immunocompetent Patients, Pregnant Women, and Patients with Lymphadenopathy

The first diagnostic goal in these patients is to establish whether they have ever been infected with T. gondii. If T. gondii–specific IgG and IgM test results are negative, the possibility that the patient’s symptoms are due to the parasite can be ruled out. During pregnancy, these results confirm that the mother has not been exposed to T. gondii but that she is at risk, if exposed, to acquire the primary infection during pregnancy and therefore can potentially transmit T. gondii to her offspring. In attempting to determine whether the patient is infected with T. gondii, it is important to perform both IgG and IgM tests because during the first 4 weeks of the acute infection, the IgG can still be negative while the IgM will be positive. In these cases, seroconversion can be diagnosed by having a new positive IgG test result in a subsequent serum sample. In rare instances, infected patients may be IgG negative because of their incapacity to produce IgG. If the patient is found to be IgG positive, the next goal is to determine whether the patient is having an acute infection or has been chronically infected (e.g., >6 months). If the IgG titer is low (e.g., a dye test at PAMF-TSL ≤512) and the IgM test result is negative, the patient has essentially been infected for more than 6 months. With these results, a patient whose symptoms or lymphadenopathy had a date of onset within 6 months of serum sampling will be considered unlikely to have toxoplasmosis. For a pregnant woman whose serum was obtained within 6 months of gestation, these results will mean that her infection was acquired before conception and that the risk for congenital toxoplasmosis is essentially zero. If the patient is found to have a positive IgM test result confirmed to be indicative of a recently acquired infection at a reference laboratory and the onset of symptoms or lymphadenopathy diminishes within the time predicted by the serologic test results for the acquisition of T. gondii, the patient will be diagnosed as having acute toxoplasmosis. For a pregnant woman, if the predicted time for when the infection was acquired falls within her gestational age, she will be diagnosed with toxoplasmosis during pregnancy and at risk for transmitting the parasite to her baby. In patients with lymphadenopathy, the histologic examination of the lymph node tissue obtained by excisional biopsy can be diagnostic or pathognomonic of toxoplasmic lymphadenitis (see earlier under Pathology).

Prenatal and Postnatal Diagnosis of Congenital Toxoplasmosis

After the diagnosis of acute toxoplasmosis or T. gondii infection has been confirmed or is highly suspected in the mother, the next step is to attempt to establish whether her offspring has been infected. Consultation with reference centers for the study and diagnosis of congenital toxoplasmosis is highly recommended. Ultrasound abnormalities can be consistent with or suggestive of congenital toxoplasmosis (see Fig. 328-3A), but they are not diagnostic. The method of choice for the prenatal diagnosis of congenital toxoplasmosis is a PCR in amniotic fluid obtained at 18 weeks of gestation. Attempts to diagnose congenital toxoplasmosis from amniotic fluid obtained before 18 weeks of gestation

2099

should be avoided because the studies reported to date have included only pregnant women whose gestational age was 18 weeks or more. In addition, false-negative results have been reported in women whose amniocentesis was performed before 18 weeks of gestation. The overall sensitivity of the amniotic fluid PCR has been reported between 64 and 92% and is highly dependent of the gestational age at which the mother acquired the infection. In the newborn, congenital toxoplasmosis can be confirmed by positive T. gondii–specific serologic test results or PCR. Samples for serology should be obtained in the peripheral blood of the baby. Cord blood should be avoided because of the high rate of maternal blood contamination. However, there is still a small degree of maternal blood contamination in newborn blood obtained by peripheral venipuncture, during the first 5 days of life for IgM antibodies and the first 10 days of life for IgA antibodies. A positive IgM immunosorbent agglutination assay (after 5 days of life) or IgA enzyme-linked immunosorbent assay (after 10 days) is diagnostic of congenital disease. Congenitally infected babies can be positive for both; positive for either one, but negative for the other test; or negative for both.14 A positive T. gondii–specific IgM in cerebrospinal fluid (CSF) is diagnostic of congenital disease, but testing of the CSF by PCR rather than for IgM is strongly recommended because of the higher sensitivity of the PCR test. The diagnosis can also be made by a positive PCR in peripheral blood, CSF, or urine. The CSF of infected infants may exhibit very high levels of protein (e.g., ≥1000 mg/dL). Cellular response in CSF is characterized by lymphocytosis, and eosinophilia may be present. Brain imaging studies may reveal calcifications or hydrocephalus; computed tomography scan is superior to ultrasound examination in the detection of these CNS abnormalities (see Fig. 328-3B).

Ocular Disease

Serologic and PCR testing can be helpful in the diagnosis of toxoplasmic chorioretinitis. An IgG-negative/IgM-negative patient is unlikely to have ocular disease due to toxoplasmosis. However, patients should be tested at reference laboratories (e.g., PAMF-TSL) because their T. gondii–specific IgG can be present but at very low levels such that only a gold standard method like the dye test can detect it. In patients with eye lesions typical of toxoplasmic chorioretinitis (see Fig. 328-2), a positive IgG test result at a relatively low titer (e.g., a dye test at PAMF-TSL ≤512) and a negative IgM test result are diagnostic of ocular disease due to the parasite reactivation. If the serologic test reveals a positive IgM result and confirmatory testing at PAMF-TSL establishes the diagnosis of an acute infection in patients 1 year of age or older, the eye disease is most likely the result of eye involvement in the setting of a recent and postnatally acquired infection. In patients with atypically appearing lesions or in whom the response to anti-Toxoplasma drugs is atypical or absent, a T. gondii–specific immune load (aqueous humor) or PCR in ocular fluids (vitreous fluid is preferable to aqueous humor because of probable higher sensitivity, but it is riskier to obtain) should be considered.

Immunocompromised Patients

It appears that acute infection rarely occurs in immunocompromised patients. However, life-threatening disease can occur when the patient’s latent infection is reactivated by AIDS, HSCT, solid organ transplantation, or other diseases characterized by severe T-cell or B-cell deficiency. Toxoplasmosis can also develop when T. gondii is transmitted from a seropositive donor to a seronegative recipient through an infected allograft (e.g., heart, liver, kidney). Therefore, to establish the risk for toxoplasmosis and to have a high index of suspicion when patients develop illnesses suggestive of toxoplasmosis, all immunocompromised patients should be tested for T. gondii–specific IgG as soon as they have been diagnosed with the underlying disease or it has been established that they will be subsequently immunosuppressed. In addition, serologic testing may not be reliable when immunosuppression is advanced or severe. Post-transplantation serologic test results for IgG antibodies may remain positive or may rise, decrease, or even become negative. Thus, pretransplantation Toxoplasma serologic studies are critical for interpretation of subsequent test results and clinical evaluation. Solid organ donors should also be tested for T. gondii–specific IgG as their allograft has the potential to transmit the parasite to the transplanted patient (e.g., heart, heart-lung, kidney, kidney-pancreas, liver, liver-pancreas). Toxoplasmosis in solid organ transplant recipients causes substantial morbidity, including disseminated disease, and mortality. In AIDS patients suspected of having toxoplasmic encephalitis with the presence of multiple brain-occupying and ring-enhancing lesions (see Fig. 328-3C), a CD4 count below 200 cells/µL, and a positive T. gondii–specific

2100

CHAPTER 328 Toxoplasmosis  

IgG, the response to anti-Toxoplasma-specific treatment is considered an additional “diagnostic” criterion of toxoplasmic encephalitis. In these patients, invasive diagnostic tests (e.g., lumbar puncture, brain biopsy) are considered unnecessary unless they do not respond to treatment within a 7- to 10-day period. This diagnostic paradigm should not be applied to other populations of immunosuppressed patients (e.g., transplant patients) because their differential diagnosis often includes other pathogens, such as invasive mold infections. In those patients, examination of the CSF by PCR or brain biopsy should be attempted at the outset of the central nervous system illness. The definitive diagnosis of toxoplasmosis in immunosuppressed patients relies on PCR, direct visualization of the parasite, and attempts for isolation of the organism (see Table 328-2). PCR testing in body fluids is the diagnostic method of choice for immunosuppressed patients at risk for toxoplasmosis who develop unexplained fever (e.g., in whole blood), pneumonia (e.g., in bronchoalveolar lavage fluid), brain lesions (e.g., in CSF), or other compatible syndromes. Theoretically, PCR can be performed in any body fluid or tissue, and laboratories have validated its use in most fluids and some tissues. Attempts to identify the tachyzoite or tissue cyst in tissues by microscopy can be enhanced with the use of the T. gondii–specific immunoperoxidase stain. CSF examination by PCR or brain biopsy should be initially considered in AIDS patients who have a low likelihood of having toxoplasmic encephalitis, such as those who have a single lesion by magnetic resonance imaging examination, have tested seronegative for T. gondii infection, have a CD4 count of more than 200 cells/µL, or who are not responding to an appropriate anti-Toxoplasma regimen.

TREATMENT  Principles of antiparasitic therapy are discussed in Chapter 323. Treatment of toxoplasmosis is indicated for immunocompetent patients with acute infection in the setting of ongoing fever, myocarditis, myositis, hepatitis, pneumonia, brain lesions or skin lesions, and lymphadenopathy accompanied by severe or persisting symptoms. Treatment is indicated as well for patients with active chorioretinitis due to primary infection or reactivation of a latent infection (Table 328-3). In immunocompetent patients, treatment is prescribed for 3 to 4 weeks or until symptoms have subsided, whichever is longer. For toxoplasmic lymphadenitis, trimethoprim-sulfamethoxazole (TMP-SMZ) (8 mg TMP/40 mg SMZ per kilogram per day divided into two doses for 1 month) increased the cure rate to 65% compared with a 13% resolution rate with placebo. A1  Treatment is also often recommended for all pregnant women suspected of having or diagnosed with primary infection during gestation (Table 328-4) in an attempt to prevent transmission of the parasite to the fetus (spiramycin) or, if congenital infection has occurred, to start treatment of the fetus in utero (pyrimethamine, sulfadiazine, and folinic acid). During pregnancy, treatment regimens are prescribed for the duration of the gestation. There was a worldwide controversy about the efficacy of spiramycin to decrease the incidence of congenital toxoplasmosis and of pyrimethamine, sulfadiazine, and folinic acid to decrease the frequency of clinical signs in infected offspring. Although no definitive studies ever disproved their efficacy, several epidemiologic studies erroneously concluded that there was no evidence of benefit. However, since 2006, several studies have reported a strong association between prenatal treatment of women infected during gestation (with spiramycin or pyrimethamine, sulfadiazine, and folinic acid) and decreases in the incidence of congenital toxoplasmosis and frequency of clinical signs in infected offspring. Spiramycin is recommended for pregnant women who have been definitively diagnosed to have or are highly suspected of having an acute infection during pregnancy that was acquired before 18 weeks of gestation. Spiramycin should be given throughout pregnancy unless fetal infection is suspected or documented. Fetal infection should be investigated with amniotic fluid PCR at 18 weeks of gestation to see whether Toxoplasma DNA is amplified and with monthly follow-up ultrasound examinations. Therapy with pyrimethamine, sulfadiazine, and folinic acid is recommended for pregnant women who have been definitively diagnosed to have or are highly suspected of having an acute infection during pregnancy that was acquired after 18 weeks of gestation, whose amniotic fluid PCR is positive for the presence of Toxoplasma DNA, or whose follow-up ultrasound examinations are suggestive of fetal congenital toxoplasmosis in the setting of acute Toxoplasma infection during gestation. In addition, newborns and infants diagnosed with or suspected of having congenital toxoplasmosis should also be treated during their first year of life (see Table 328-4). Treatment at higher doses is urgently indicated for all immunocompromised patients with toxoplasmosis due to reactivation of their latent infection or primary infection acquired by natural exposure to the parasite or by solid organ transplantation (see Table 328-3). If untreated, toxoplasmosis in these patients has a very high rate of morbidity and mortality.

TABLE 328-3 TREATMENT REGIMENS FOR PATIENTS WITH ACUTE OR PRIMARY TOXOPLASMOSIS AND IMMUNOCOMPROMISED PATIENTS WITH TOXOPLASMOSIS DUE TO REACTIVATION* IMMUNOCOMPETENT PATIENTS WITH ACUTE INFECTION† Pyrimethamine (PO)

plus Folinic acid‡

Sulfadiazine (PO)

or Clindamycin (PO or IV) or Atovaquone (PO)

IMMUNOCOMPROMISED PATIENTS

50 mg every 12 hr for 2 days, followed by 25 to 50 mg daily

200 mg loading dose, followed by 50 mg/day (60 kg)

10-20 mg daily (during and 1 week after therapy with pyrimethamine) plus either sulfadiazine or clindamycin or atovaquone as follows: 75 mg/kg (first dose), followed by 50 mg/kg every 12 hr (maximum 4 g/day)

10-20 mg daily (up to 50 mg/ day; during and 1 week after therapy with pyrimethamine) plus either sulfadiazine or clindamycin or atovaquone as follows: 1000 mg (60 kg) every 6 hr

300 mg every 6 hr

600 mg every 6 hr (up to 1200 mg every 6 hr)

1500 mg orally twice daily

1500 mg orally twice daily

Trimethoprimsulfamethoxazole (PO or IV)

10 mg/kg/day (trimethoprim component) in two or three doses

10 mg/kg/day (trimethoprim component) in two or three doses (doses as high as 15-20 mg/kg/day have been used)

Pyrimethamine and folinic acid plus Clarithromycin (PO) or Dapsone (PO) or Azithromycin (PO)

Same doses as above

Same doses as above

500 mg every 12 hr

500 mg every 12 hr

100 mg/day

100 mg/day

900 to 1200 mg/day

900 to 1200 mg/day

After the successful use of a combination regimen during the acute or primary therapy phase, the same agents at half-dose are usually used for maintenance or secondary prophylaxis. *Preferred regimens: pyrimethamine, sulfadiazine, and folinic acid or trimethoprimsulfamethoxazole. Assistance is available for the diagnosis and management of patients with toxoplasmosis at the Palo Alto Medical Foundation Toxoplasma Serology Laboratory (PAMF-TSL), Palo Alto, Calif; www.pamf.org/serology; 650-853-4828; [email protected]. † Particularly in the setting of myocarditis, myositis, hepatitis, pneumonia, brain or skin lesions, and lymphadenopathy accompanied by severe or persisting symptoms. Also indicated for those with active ocular disease due to primary infection or reactivation. ‡ Folinic acid = leucovorin; folic acid must not be used as a substitute for folinic acid.



PREVENTION

Primary Infection

Because approximately 50% of patients may inadvertently become infected with the parasite without having a recognized risk factor for acute infection, only systematic serologic testing can establish whether a patient has been exposed to T. gondii. Thus, each pregnant woman and immunocompromised patient should be screened for T. gondii–specific IgG and IgM regardless of their epidemiologic history. Seronegative pregnant women and immunocompromised individuals should be counseled on how to maximize their prevention efforts to avoid infection with T. gondii. In addition, seronegative pregnant women should be tested serially during gestation in an attempt to diagnose seroconversion at the earliest time possible. In some countries, such as France, seronegative pregnant women are mandated by law to be tested every month for T. gondii–specific IgG and IgM.15 Women who seroconvert are offered spiramycin (if infected before 18 weeks of gestation) or pyrimethamine, sulfadiazine, and folinic acid (if infected after 18 weeks). Mothers whose amniotic fluid is found to be positive by PCR or those in whom fetal ultrasound study is highly suggestive of congenital toxoplasmosis are offered pyrimethamine, sulfadiazine, and folinic acid. Although infection often occurs in the absence of known risk factors for the acute infection, educational interventions to avoid exposure to the parasite have been shown to be effective in decreasing the incidence of seroconversion during gestation.

CHAPTER 328 Toxoplasmosis  

2101

TABLE 328-4 TREATMENT REGIMENS FOR PREGNANT WOMEN WHO HAVE LIKELY ACQUIRED TOXOPLASMA GONDII INFECTION DURING GESTATION AND INFANTS SUSPECTED OF HAVING OR CONFIRMED TO HAVE CONGENITAL TOXOPLASMOSIS DURING PREGNANCY

IN CONGENITAL DISEASE

Spiramycin (oral)

Recommended for pregnant women suspected of having or confirmed to have acquired the infection during gestation and before 18 weeks of gestation. Spiramycin should be administered until delivery in those with negative amniotic fluid PCR test results and normal follow-up ultrasound studies or low suspicion of fetal infection. Spiramycin is not teratogenic, and it is available in the United States only through the Investigational New Drug (IND) process at the Food and Drug Administration (301-796-1400). Prior medical consultation is required.* Dosage: 1 g (3 million units) every 8 hr (for a total of 3 g or 9 million units per day)

Not recommended during pregnancy if the fetus has been documented to be or suspected to have been infected. In the setting of fetal infection, pyrimethamine, sulfadiazine, and folinic acid should be instituted (see below)

Pyrimethamine (oral) plus sulfadiazine (oral) plus folinic acid† (oral)

Recommended for women ≥18 weeks of gestation in whom it is suspected or confirmed that the acute infection has been acquired at or after 18 weeks of gestation or who have a positive amniotic fluid PCR test result or an abnormal ultrasound study suggestive of congenital toxoplasmosis Pyrimethamine is teratogenic and should not be used during pregnancy before week 18 (in some centers in Europe, it is used as early as week 14). Sulfadiazine should not be used alone. Dosages: Pyrimethamine: 50 mg every 12 hr for 2 days followed by 50 mg daily Sulfadiazine: 75 mg/kg (first dose) followed by 50 mg/kg every 12 hr (maximum 4 g/day) Folinic acid† (leucovorin): 10-20 mg daily (during and for 1 week after pyrimethamine therapy)

Infant (treatment regimen is usually recommended for 1 year): Pyrimethamine: 1 mg/kg every 12 hr for 2 days; followed by 1 mg/kg/ day for 2 or 6 months; followed by 1 mg/kg/day every Monday, Wednesday, Friday Sulfadiazine: 50 mg/kg every 12 hr Folinic acid† (leucovorin): 10 mg three times weekly Prednisone (if CSF protein ≥1 g/dL or severe chorioretinitis): 0.5 mg/kg every 12 hr (until CSF protein 90%

70%

70-80% >90%

Stool antigen detection

N/A = not available; PCR = polymerase chain reaction. Modified from Haque R, Huston CD, Hughes M, et al. Current concepts: amebiasis. N Engl J Med. 2003;348:1565-1573.



DIAGNOSIS

Diagnosis of amebiasis is best accomplished by the combination of serology and identification of the parasite in feces or at extraintestinal sites of invasion by PCR or antigen detection (such as pus obtained by fine needle aspiration of a liver abscess).8 Examination of stool for ova and parasites should not be used to diagnose amebiasis (Table 331-1). The most sensitive diagnostic approach is the combined use of E. histolytica–specific antigen detection or polymerase chain reaction plus serology.9 Amebic colitis can also be diagnosed by colonoscopy, which can show characteristic discrete ulcers or erosions with white or yellow exudates, especially in the cecum, and can identify trophozoites in about 90% of patients.10

TREATMENT (also see Chapter 323)  Therapy for invasive infection differs from that for noninvasive infection, which may be treated with paromomycin (Table 331-2). Invasive infections require treatment with nitroimidazoles, particularly metronidazole, tinidazole, secnidazole, or ornidazole. For amebic colitis, tinidazole reduces treatment failure

ADULT DOSAGE

AMEBIC LIVER ABSCESS Metronidazole 750 mg PO tid × 10 days

or Tinidazole

Followed by a luminal agent Paromomycin

or Diloxanide furoate

SIDE EFFECTS Primarily GI side effects: anorexia, nausea, vomiting, diarrhea, abdominal discomfort, or unpleasant metallic taste Disulfiram-like intolerance reaction to alcoholic beverages Neurotoxicity, including seizures, peripheral neuropathy, dizziness, confusion, irritability

2 g PO once daily × 5 days

Primarily GI side effects and disulfiram-like intolerance reaction to alcoholic beverages as for metronidazole

30 mg/kg/day PO in 3 divided doses per day × 5-10 days

Primarily GI side effects: diarrhea, GI upset

500 mg PO tid × 10 days

Primarily GI side effects: flatulence, nausea, vomiting Pruritus, urticaria

AMEBIC COLITIS Metronidazole

Same as for amebic liver abscess 750 mg PO tid × 5-10 days Plus a luminal agent (same as for amebic liver abscess) ASYMPTOMATIC INTESTINAL COLONIZATION Treatment with a luminal agent as for amebic liver abscess GI = gastrointestinal; PO = per os (by mouth); tid = three times a day. Modified from Haque R, Huston CD, Hughes M, et al. Current concepts: amebiasis. N Engl J Med. 2003;348:1565-1573.

TABLE 331-3 FREE-LIVING AMEBAE ORGANISM

DISEASE

EPIDEMIOLOGY

DIAGNOSIS

CLINICAL COURSE

THERAPY

Naegleria fowleri

Primary amebic encephalitis

Warm freshwater exposure

CSF wet mount for amebae, PCR

Death within 1-2 weeks of onset

Combination therapy with amphotericin B, miltefosine, rifampin, fluconazole, and azithromycin

Acanthamoeba spp

Keratitis

Corneal trauma, usually from contact lens

Corneal scraping for amebae and cysts

Subacute

Polyhexamethylene biguanide, chlorhexidine, propamidine, hexamidine

Acanthamoeba spp

Granulomatous amebic encephalitis

Immunodeficient (organ transplants, HIV/ AIDS)

Biopsy of brain or skin abscess—IFA or PCR

Subacute

Combination therapy with miltefosine, pentamidine, an azole (fluconazole or itraconazole), and/or sulfadiazine

Balamuthia mandrillaris

Granulomatous amebic encephalitis

Immunodeficient but also immunocompetent

Biopsy of brain

Subacute

Combination therapy with miltefosine, pentamidine, fluconazole, and sulfadiazine plus either azithromycin or clarithromycin

Sappinia

Amebic encephalitis

Single patient was not immunodeficient

Azithromycin, pentamidine, itraconazole, flucytosine

AIDS = acquired immunodeficiency syndrome; CSF = cerebrospinal fluid; HIV = human immunodeficiency virus; IFA = indirect fluorescent antibody; PCR = polymerase chain reaction. Modified from Visvesvara GS, Moura H, Schuster FL. Pathogenic and opportunistic free-living amoebae: Acanthamoeba spp, Balamuthia mandrillaris, Naegleria fowleri, and Sappinia diploidea. FEMS Immunol Med Microbiol. 2007;50:1-26.

rates and adverse effects compared with metronidazole. A1  In the rare case of fulminant amebic colitis, it is prudent to add broad-spectrum antibiotics to treat intestinal bacteria that may spill into the peritoneum. Parasites persist in up to half of the patients who are treated with a nitroimidazole, so treatment should be followed with paromomycin or the second-line agent diloxanide furoate to cure luminal infection. High-throughput drug screening has identified auranofin, a U.S. Food and Drug Administration–approved drug used for treatment of rheumatoid arthritis, as a potentially active agent against E. histolytica. Drainage of a liver abscess should be considered in patients who do not show a clinical response to drug therapy within 5 to 7 days or in those with a high risk for rupture of the abscess, as defined by a cavity with a diameter of greater than 5 cm or by the presence of lesions in the left lobe. Percutaneous needle aspiration or catheter drainage is the procedure of choice for drainage of a liver abscess. Surgical intervention is occasionally required for drainage of a liver abscess, acute abdomen, gastrointestinal bleeding, or toxic megacolon.



PREVENTION

The feasibility of prevention by vaccination with the parasite’s Gal/GalNAc lectin is supported by substantial data from human, animal model, and in vitro studies.11 This vaccine is in the late stages of preclinical development for the prevention of amebiasis in infants and children in the developing world. Provision of sanitation and clean water and safe sexual practices to prevent fecal-oral transmission are of great importance but not universally effective because of the low infectious dose and chlorine resistance of the cyst.12  

PROGNOSIS

Therapy for amebiasis is highly effective. Drug resistance is not reported.

  FREE-LIVING AMEBAE

Rare infections of the central nervous system can be seen with infection by free-living amebae of the genera Naegleria, Balamuthia, Acanthamoeba, and Sappinia. Naegleria fowleri is the agent of primary amebic meningoencephalitis, which occurs in previously healthy children and young adults who have swum in fresh water 2 to 5 days before the onset of meningoencephalitis. Cerebrospinal fluid has a polymorphonuclear predominance, and motile amebae can be seen in a wet mount of cerebrospinal fluid. The disease is relentlessly progressive to death in most patients. In one case of successful treatment, a combination of intrathecal and intravenous amphotericin B and miconazole and oral rifampin was used. Acanthamoeba can cause keratitis13 in individuals with corneal injuries (usually from contact lens use), as well as granulomatous amebic encephalitis in the immunocompromised. Granulomatous amebic encephalitis is caused by Acanthamoeba, Balamuthia, and Sappinia14; it is usually associated with focal neurologic findings and has a subacute course. Miltefosine should be part of combination therapy (Table 331-3).

  Grade A Reference A1. Pandey S, Gupta GK, Wanjari SJ, et al. Comparative study of tinidazole versus metronidazole in treatment of amebic liver abscess: a randomized control trial. Indian J Gastroenterol. 2018;37: 196-201.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 331 Amebiasis  

GENERAL REFERENCES 1. Gwairgi MA, Ghildyal R. Nuclear transport in Entamoeba histolytica: knowledge gap and therapeutic potential. Parasitology. 2018;145:1378-1387. 2. Noor Z, Watanabe K, Abhyankar MM, et al. Role of eosinophils and tumor necrosis factor alpha in Interleukin-25-Mediated protection from amebic colitis. MBio. 2017;8:1-10. 3. Gilchrist CA, Petri SE, Schneider BN, et al. Role of the gut microbiota of children in diarrhea due to the protozoan parasite Entamoeba histolytica. J Infect Dis. 2016;213:1579-1585. 4. Ali IK. Intestinal amebae. Clin Lab Med. 2015;35:393-422. 5. Kantor M, Abrantes A, Estevez A, et al. Entamoeba histolytica: updates in clinical manifestation, pathogenesis, and vaccine development. Can J Gastroenterol Hepatol. 2018;2018:1-6. 6. Mohidin B, Green SF, Duggineni S. Amoebic liver abscess. QJM. 2018;111:821-822. 7. Lachish T, Wieder-Finesod A, Schwartz E. Amebic liver abscess in Israeli travelers: a retrospective study. Am J Trop Med Hyg. 2016;94:1015-1019. 8. Guevara Á, Vicuna Y, Costales D, et al. Use of real-time polymerase chain reaction to differentiate between pathogenic Entamoeba histolytica and the nonpathogenic Entamoeba dispar in Ecuador. Am J Trop Med Hyg. 2019;100:81-82.

2111.e1

9. Saidin S, Othman N, Noordin R. Update on laboratory diagnosis of amoebiasis. Eur J Clin Microbiol Infect Dis. 2019;38:15-38. 10. Horiki N, Furukawa K, Kitade T, et al. Endoscopic findings and lesion distribution in amebic colitis. J Infect Chemother. 2015;21:444-448. 11. Min X, Feng M, Guan Y, et al. Evaluation of the C-terminal fragment of Entamoeba histolytica Gal/ GalNAc lectin intermediate subunit as a vaccine candidate against amebic liver abscess. PLoS Negl Trop Dis. 2016;10:1-17. 12. Shirley DT, Farr L, Watanabe K, et al. A review of the global burden, new diagnostics, and current therapeutics for amebiasis. Open Forum Infect Dis. 2018;5:1-9. 13. Lorenzo-Morales J, Khan NA, Walochnik J. An update on Acanthamoeba keratitis: diagnosis, pathogenesis and treatment. Parasite. 2015;22:1-19. 14. Ong TYY, Khan NA, Siddiqui R. Brain-eating amoebae: predilection sites in the brain and disease outcome. J Clin Microbiol. 2018;55:1989-1997.

2111.e2

CHAPTER 331 Amebiasis  

REVIEW QUESTIONS 1. Individuals at increased risk for amebiasis include all except: A . Diabetics B. Pacific Islanders C. Men who have sex with men D. Residents of institutions for the mentally retarded E. Children in the developing world Answer: A  All of the other answers are risk factors for amebiasis, with children in low-income settings in the developing world at highest risk. 2. Which statement is false for the parasite Entamoeba dispar? A . It is as prevalent as Entamoeba histolytica worldwide. B. It is a cause of amebic colitis and liver abscess. C. It is more common than E. histolytica in North America. D. It is identical in appearance to E. histolytica on stool ova and parasite examination. E. It is fecal-orally transmitted. Answer: B  E. dispar is not known to cause disease and is more common in temperate climates than the identical-appearing E. histolytica.

3. Which statement is true for the diagnosis of intestinal amebiasis? A . Serology is positive acutely in more than 90% of cases. B. Test results for occult blood in stool are positive in most. C. Stool ova and parasite examination is sensitive and specific. D. E. histolytica–specific stool antigen detection test is more than 80% sensitive for the diagnosis. Answer: D  Stool ova and parasite examination is neither sensitive nor specific. Serologic testing by indirect hemagglutination is positive during acute illness in no more than 70% of cases (but >90% in convalescence). Blood in the stool occurs in virtually all patients with amebic colitis, but stools do not have to be positive for occult blood in amebic diarrhea. 4. Which statement is false about extraintestinal amebiasis? A . Amebic liver abscess is much more common in men than in women and rare in children. B. Amebic brain abscess is seen only in patients with amebic liver abscess. C. Amebic liver abscess can be treated without drainage in many cases. D. Bacterial superinfection of amebic liver abscess is common. Answer: D  For unclear reasons, bacterial superinfection of an amebic liver abscess is rare.

CHAPTER 332  Babesiosis and Other Protozoan Diseases  

2111

332  BABESIOSIS AND OTHER PROTOZOAN DISEASES SAM R. TELFORD, III, AND PETER J. KRAUSE

  BABESIOSIS

Babesiosis is a tick-borne, malaria-like disease caused by sporozoan parasites of the genus Babesia.  

EPIDEMIOLOGY

With few exceptions, Babesia spp are transmitted by ixodid ticks. Thus, wherever humans are intensely exposed to hard-bodied ticks, babesiosis should be part of the differential diagnosis for a patient presenting with fever and hematologic abnormalities. Three worldwide epidemiologic patterns are apparent. The first involves the rodent-maintained Babesia microti, which is a species complex distributed across the Holarctic. A median of 1400 cases of B. microti babesiosis were reported each year from the northeastern United States and upper Midwestern states from 2011 through 2014, the first 4 years that human babesiosis has been designated as a notifiable infectious disease by the Centers for Disease Control and Prevention (CDC). By comparison, about 30,000 Lyme disease cases are reported each year in the United States, although both babesiosis and Lyme disease are thought to be underreported. The vector for B. microti is the same as that for Lyme disease (Chapter 305), the deer tick, Ixodes dammini, also known as northern populations of I. scapularis.1 Indeed, concurrent babesiosis and Lyme disease is common. Immunocompetent as well as immunocompromised individuals are at risk. Within the last decade, B. microti has been increasingly reported in an expanded distribution in the United States from the original foci in coastal New England and the Upper Midwest,

CHAPTER 332  Babesiosis and Other Protozoan Diseases  

ABSTRACT

The burden due to human babesiosis is increasing in sites where Lyme disease is endemic; due to the fact that younger people tend to be asymptomatic, babesia is a major transfusion-transmitted disease. The global public health importance of babesiosis is being better defined because of the availability of molecular phylogenetic tools. Similarly, enteric disease due to miscellaneous protozoa, particularly the coccidia, may more easily be diagnosed and characterized due to these methods. New methods for detection are also providing more information about the prevalence and incidence of trichomoniasis.

KEYWORDS

Babesiosis tick-borne enteric coccidian food-borne ixodid ticks

2111.e3

2112

CHAPTER 332  Babesiosis and Other Protozoan Diseases  

and it is possible that babesiosis cases eventually may be found wherever Lyme disease is intensely zoonotic. In addition, cases of B. microti or B. microti– like babesiosis have been reported from Australia, Canada, China, Germany, Japan, Poland, Spain, South Africa, and Taiwan; the vectors for these have not been definitively identified. The second pattern is represented by cases of babesiosis due to Babesia divergens, B. divergens–like, or closely related species (e.g., Babesia venatorum) that have been reported from Asia and Europe. Almost all cases in Europe have been in splenectomized patients residing in sites where castor bean ticks (Ixodes ricinus) and deer are common. Babesia venatorum recently has been found to be endemic in northeastern China. A few cases of Babesia divergens–like infections (referred to as MO-1) have been described in the United States. The third pattern of babesiosis involves sporadic cases due to diverse Babesia spp. These include Babesia duncani (WA-1) and CA-type parasites of the western United States; a Babesia motasi–like infection (KO-1) in Korea; and unidentified or poorly characterized Babesia spp from China, Colombia, Egypt, India, Mexico, Mozambique, and South Africa. Although the known zoonotic tick vectors (I. dammini, I. ricinus) have marked seasonal periods of activity (May to August) and the majority of reported cases are acquired during these times, babesiosis may be diagnosed at any time of the year. More than 230 cases of transfusion-acquired babesiosis due to B. microti and three due to B. duncani have been reported. The actual number of cases is thought to be much greater. Babesiosis is currently the most commonly reported transfusion-transmitted disease in the United States, and the number of such cases is increasing, including those ending in death (Chapter 167). Cases occur throughout the year, and about 10% of cases occur in nonendemic areas because Babesia-infected blood is exported to nonendemic areas or persons become infected in endemic areas and subsequently donate blood in nonendemic areas. A few cases of transplacentally transmitted babesiosis have been reported.  

PATHOBIOLOGY

The pathophysiology of Babesia infection is directly related to the development of parasitemia. Peripheral blood parasitemias of 70% or greater have been reported, although most cases sustain parasitemias on the order of 0.5 to 5%. Excessive production of pro-inflammatory cytokines seems to best explain the most common clinical manifestations, which include fever, sweats, chills, headache, myalgia, nausea, vomiting, diarrhea, and pallor. Such findings are not seen when erythrocyte lysis is due to noninfectious causes, which suggests that the release of merozoites serves as a trigger for the pro-inflammatory cascade. Elevated serum concentrations of TNF as well as of interferon-γ, interleukins 2 and 6, E-selectin, vascular cell adhesion molecule 1, and intracellular cell adhesion molecule 1 are detected during the acute phase of human B. microti infection and return to baseline within 3 months after resolution of infection. Severe illness caused by infection with Babesia includes a complex array of metabolic abnormalities and organ dysfunction. Pulmonary disease is the most common complication in people experiencing severe Babesia infection, with up to 20% of patients suffering from noncardiogenic pulmonary edema. Pro-inflammatory cytokines appear to mediate the pulmonary complications of Babesia infection, at least in part. It is also likely that lung and other endorgan disease is mediated, at least in part, by vascular stasis.  

CLINICAL MANIFESTATIONS

About a quarter of B. microti infections in adults and half of those in children are subclinical.2 This estimate is derived from an epidemiologic study that determined the frequency of people who seroconverted during the course of the summer transmission season but reported no illness, coupled with a careful accounting of symptomatic cases. Most people experience a mild to moderate illness lasting about a week. There is a gradual onset of malaise, anorexia, fatigue, fever (temperature as high as 40° C), sweats, and myalgia. Nausea, vomiting, headache, shaking chills, emotional lability, depression, hemoglobinuria, and hyperesthesia also have been reported. Findings on physical examination consist of fever, pallor, splenomegaly, and hepatomegaly. Laboratory abnormalities include anemia, thrombocytopenia, and leukopenia. Parasitemia generally ranges from barely detectable on blood smear to 5% in previously healthy people but may reach 85% in asplenic and other immunocompromised patients. Serum lactate dehydrogenase, bilirubin, and transaminase levels may be elevated in more severe cases. Persistent relapsing illness may occur in highly immunocompromised people who fail to clear the infection for months or more than a year despite multiple courses of antibiotics. The case-fatality rate for B. microti babesiosis has been estimated to be 6 to 9% in hospitalized patients but may be as high as 20% in immunocompromised hosts, including those who acquire the infection through blood transfusion. Review of the Medicare database (2006-2013) demonstrated that about 1%

of 10,000 cases in patients older than 65 years of age died within 30 days of diagnosis. Severe babesiosis usually occurs only in people with asplenia, malignant disease, coinfection with human immunodeficiency virus (HIV), organ transplants, immunosuppressive treatment, or age younger than 2 months or older than 50 years. About a third of treated asplenic babesiosis patients without history of autoimmune disease may suffer warm autoimmune hemolytic anemia that requires immunosuppressive treatment.3 Cases of babesiosis caused by B. divergens tend to be severe, at least in part because they are primarily reported in immunocompromised patients. Virtually all European patients experiencing B. divergens infection have been splenectomized, and about a third of the patients died. In these patients, there is an acute onset of illness with hemoglobinuria, a persistent nonperiodic high fever (temperature of 40° to 41° C), shaking chills, intense sweats, headaches, and myalgia, as well as lumbar and abdominal pain. Vomiting and diarrhea may occur. Pulmonary, renal, or liver failure may develop rapidly. In fatal cases, patients become comatose with multiorgan failure. B. duncani, B. venatorum, and B. divergens–like infections also have often been reported in immunocompromised hosts with a similarly severe course of illness. In contrast, B. venatorum infection in 48 immunocompetent patients in northeastern China was similar to that of B. microti infection, with full recovery of all patients, including seven patients who were admitted to hospital.  

DIAGNOSIS

The diagnosis of babesiosis is based on epidemiologic and clinical findings and confirmed by laboratory testing. It should be considered in patients who live or travel in Babesia endemic regions or who have received a blood transfusion within the previous six months and whose clinical findings are consistent with babesiosis. The diagnosis may be confirmed by examination of a Giemsastained thin blood smear for the presence of parasites within erythrocytes. In immunocompromised patients, parasitemias are likely to exceed one infected cell per oil immersion field and thus are quickly detected. For B. microti babesiosis (Fig. 332-1), examination of a slide for 10 minutes or as many fields as needed to tally 200 leukocytes (that are not infected but serve as a marker for effort) and repeated smears performed twice a day may be required. Standard Romanowsky stains (Giemsa, Wright) using malaria protocols are optimal. Artifactual inclusions are limited mainly to stain precipitates (which can be determined by their presence in the plasma spaces between cells), Howell-Jolly or Heinz bodies (Chapter 148), or platelets superimposed on erythrocytes, which always have a light-colored halo when visualized this way. Babesia spp have clearly defined chromatin with a lighter-colored cytoplasm (Fig. 332-2A). They may be mistaken for early malarial trophozoites. Neither malarial nor babesial rings have hemozoin (malarial pigment), so this is not a good feature to distinguish between the two. Paired piriform parasites, arranged in a v, are suggestive of B. divergens or B. divergens–like infection (Fig. 332-2B). Rings of all sizes may be seen in all species. Multiple parasites may frequently be seen in single erythrocytes, as well as clumps of extracellular parasites. Tetrad forms (Fig. 332-2C) and Maltese cross forms (Fig. 332-2D) are diagnostic but are rarely seen in B. microti babesiosis. They seem to be more common with B. duncani or CA-type infections. Polymerase chain reaction (PCR) assays are more sensitive than blood smears in cases in which parasitemias are sparse and may be as good as or better than a blood smear for making the diagnosis in clinical practice. Real-time PCR

FIGURE 332-1.  Babesia microti. Human infection, Nantucket Island. Predominance of ring forms with a cluster of extraerythrocytic parasites (arrow) free in the plasma.

CHAPTER 332  Babesiosis and Other Protozoan Diseases  

A

C

2113

B

D

FIGURE 332-2.  Diagnosis of Babesia infection. A, Typical thin-film field demonstrating ring forms of B. microti with vacuole or “whitish” cytoplasm demarcated by a dark-staining, defined chromatin. B, B. Divergens–like (MO-1) with robust rings; accolé form and paired piriform parasites are marked by arrows. C, Tetrad forms of B. microti. D, MO-1, classic Maltese cross. (Microscope slide from human case of MO-1 kindly provided by Dr. J. F. Beattie, Department of Pathology, The Medical Center, Bowling Green, Ky.)

assays performed in-house would provide confirmation nearly as quickly as microscopy with the added advantage of increased sensitivity.4 Serologic testing is useful for confirming B. microti infection. The indirect immunofluorescence test (IFAT), using antigen from infected hamster red cells, is sensitive and specific and is currently the serologic method of choice. Analysis of paired acute and convalescent serum samples is most useful for a confirmation of B. microti infection. The presence of parasite-specific IgM may indicate that the patient has an acute infection even in the absence of readily demonstrable parasitemia. Serology is not generally useful for B. divergens babesiosis (although the IFAT is generally sensitive and specific), given its fulminant natural history. Because parasitemia occurs before an antibody response and the doubling time of B. divergens can be as short as 8 hours, treatment needs to be initiated immediately on the basis of clinical suspicion and initial laboratory results. Serology for B. duncani is complicated by a high rate of false positives, and specificity depends on carefully establishing a diagnostic threshold dilution. Antigens for other of the Babesia spp infecting humans are not available. The known vectors for human babesiosis are ticks that also transmit the agents of Lyme disease, human granulocytic anaplasmosis, Borrelia miyamotoi infection, Ehrlichia muris–like infection, and tick-borne encephalitis virus. Thus, coinfections should be considered in all patients with babesiosis. Acute illness in patients coinfected with Lyme disease and babesiosis is more severe and more persistent than in patients experiencing Lyme disease alone.

TREATMENT  Mild to moderate B. microti infection typically occurs in immunocompetent people. Therapy for these individuals should consist of the combination of atovaquone (750 mg orally twice daily for 7 to 10 days) and azithromycin (500 to 1000 mg initial dose followed by 250 mg orally daily for 7 to 10 days). For immunocompromised hosts, the azithromycin dose should be increased to 600 to 1000 mg orally daily for 7 to 10 days).5,6 A prospective randomized trial demonstrated that patients treated with atovaquone and azithromycin cleared parasitemia as effectively as did those receiving clindamycin and quinine and with fewer side effects. Severe babesiosis usually occurs in immunocompromised hosts, including premature infants, those over 50 years of age, people who have asplenia,

malignancy, HIV infection or who are on immunosuppressive medications. For severe babesiosis, a 7- to 10-day course of the combination of atovaquone (750 mg orally twice daily) and azithromycin (500 mg intravenously once a day) is recommended. An alternative combination is clindamycin (300 to 600 mg every 6 hours intravenously) and quinine (650 mg orally every 8 hours). An incidental finding of parasites (e.g., during a manual CBC, or at a blood donation center that proactively performs screening) in an otherwise healthy individual would not require treatment. Treatment may occasionally fail in high-risk patients or in those who must discontinue quinine because of side effects, such as severe tinnitus and gastrointestinal distress, or because of the rare occurrence of antibiotic resistance with atovaquone and azithromycin. A prolonged course of treatment may be required to clear parasitemia in certain immunocompromised patients, including those with B-cell lymphoma or other conditions treated with rituximab, patients with malignancy who also are asplenic, patients with organ or stem cell transplantation, and patients with HIV/AIDS. In such cases, combination therapy should be used that may include two or more of the following antimicrobials: artemisinin, atovaquone, azithromycin, clindamycin, doxycycline, atovaquone-proguanil (Malarone), pentamidine, quinine, and trimethoprimsulfamethoxazole. Once an effective combination is identified, it should be continued for at least 6 weeks and 2 weeks beyond the time when Babesia can no longer be visualized on blood smear or blood samples become PCR negative. Other drugs have been demonstrated to be effective against B. microti in laboratory models (e.g., robenidine, primaquine, artesunate, and endochin-like quinolones) but await clinical trials. Exchange transfusion should be considered in severely ill patients with parasitemias in excess of 10%, evidence of severe hemolysis, or organ compromise. In particularly severe babesiosis cases, partial or complete blood exchange transfusion (1 to 3 blood volumes) should be undertaken, in addition to treatment with atovaquone and azithromycin. Apheresis may help reduce circulating factors (e.g., inflammatory products liberated by the parasite) that contribute to pathology.



PREVENTION

Prevention depends on reducing the risk for tick bites. Immunocompromised individuals should be especially careful to use personal protection and may even consider avoiding highly endemic sites such as coastal New England and Long Island in the United States during May through July, when risk is the greatest. Use of repellants such as DEET or application of permethrin to

2114

CHAPTER 332  Babesiosis and Other Protozoan Diseases  

clothing will greatly reduce tick attachment. Such products should be applied to shoes, socks, and trouser cuffs. Wearing light-colored long pants and tucking the cuffs into socks will also help prevent ticks from gaining access to attachment sites. Daily examination for attached ticks should be performed; the best way to do this is to feel for new bumps on a soapy body in the shower. Any attached ticks should be promptly removed by simple traction, which is best accomplished with the use of tweezers, as illustrated in Figure 311-3 in Chapter 311. As with the agent of Lyme disease, ticks must be attached at least 36 to 48 hours before a sufficient inoculum of Babesia sporozoites is delivered. Community-level prevention should focus on public education about the risks of tick-borne infection, reducing habitat for ticks (brush removal and landscaping around yards), or reducing the reproductive hosts for the tick. Deer reduction will reduce the abundance of the deer tick vector for B. microti babesiosis. Screening blood donations for B. microti antibodies and B. microti DNA (using PCR) can decrease the risk of transfusion-associated babesiosis.7.8  

PROGNOSIS

Death may occur in patients with severe babesiosis, but other long-term sequelae have not been reported for patients who have been adequately treated. In most patients who complete a full treatment regimen, B. microti DNA becomes undetectable by PCR within 3 months. Infection does not imply protective immunity based on laboratory rodent models, although subsequent infections are limited in duration and intensity. Recrudescent infections have been reported, mainly in immunocompromised individuals.

  MISCELLANEOUS ENTERIC PROTOZOA

The gastrointestinal and urogenital tracts may contain representatives of the four major groupings of protozoa (amebae, sporozoa, flagellates, and ciliates). Diarrhea and other lower gastrointestinal signs and symptoms may be caused by diverse protozoa. Specific clinical diagnosis is difficult because expert clinical parasitology support is required to determine whether an agent that has been detected in a stool sample is a pathogenic species. However, new pointof-care antigen detection tests, or nucleic acid amplification tests (NAAT) are sensitive and specific when they can be used. Identification is necessary because treatment options differ by the agent. With the exception of Trichomonas vaginalis infection (sexually transmitted), all of the enteric protozoa are acquired by the ingestion of food or materials contaminated by human feces; a small subset may have extraintestinal manifestations. Given a shared mode of transmission (fecal-oral), demonstrating the presence of any one of these protozoa within a stool sample from a patient is justification for an intensified search for those that are recognized as clinically significant pathogens (Entamoeba histolytica, Giardia lamblia/intestinalis, Cyclospora cayetanensis, Cystoisospora belli, and Cryptosporidium parvum/hominis). Other protozoa, many of which morphologically resemble true pathogens, are commonly detected within stools of patients with lower gastrointestinal disturbances, but support for their role as etiologic agents is weak. Cryptosporidiosis (Chapter 329), giardiasis (Chapter 330), and amebiasis (Chapter 331) are discussed in separate chapters. Trichomoniasis and coccidian enteritis are discussed here because they are relatively common infections.  

Dientamoeba fragilis (previously thought to be an ameba). The 10- to 15-µmlong trophozoites multiply by longitudinal binary fission on the epithelial surface of the vagina or urethra as well as in vaginal or urethral secretions and are thereby transmitted by sexual intercourse. No cyst form is known, and the trophozoites are easily killed by drying.  

CLINICAL MANIFESTATIONS

Trichomoniasis is one of three common causes of vaginitis or vaginosis (along with bacterial vaginosis and vulvovaginal candidiasis).10 It is characterized by a thin gray to yellowish green frothy discharge; vulvovaginal erythema; ectocervical erythema or “strawberry cervix,” observable mainly by colposcopy; pH higher than 4.5; increased presence of polymorphonuclear leukocytes; and a positive result of the whiff test, in which a foul fishy odor is intensified on addition of potassium hydroxide. The incubation period for trichomoniasis is 5 to 28 days. In addition to a frothy discharge, vaginitis can be accompanied by vulvovaginal irritation, dyspareunia, abdominal pain, and dysuria. Symptoms may worsen during menstruation. Population-based studies indicate that as many as half of T. vaginalis infections in women and the majority in men are asymptomatic. T. vaginalis can frequently be isolated from the male partners of infected women and can produce symptomatic urethritis. Urethral discharge is generally scant in these cases. Pharyngitis with positive T. vaginalis NAAT has been reported. Rarely, T. vaginalis is associated with epididymitis, superficial penile ulcerations that are usually located under the prepuce, or prostatitis. T. vaginalis infection is not associated with a greater risk of advancement of prostate cancer.  

DIAGNOSIS

The CDC guidelines (http://www.cdc.gov/std/treatment/2010/vaginaldischarge.htm#a2) state that all women with a sexually transmitted infection should be specifically tested for evidence of T. vaginalis infection, and HIVpositive women should be tested annually. The WHO vaginal discharge syndromic management/vaginal discharge flowchart for diagnosis of trichomoniasis has 99.93% positive predictive value.11 In women, vaginal and urethral secretions should be examined. T. vaginalis is seen in wet mounts of vaginal secretions in approximately 60% of infected women, thus confirming the diagnosis. Live T. vaginalis have a twitching or tumbling motion in wet mounts, and polymorphonuclear leukocytes are usually present. Direct immunofluorescent antibody staining is more sensitive than wet mounts but technically more difficult. Culture is an even more sensitive method of diagnosis; commercial kits for culture are available, but the results are not available for 3 to 7 days. T. vaginalis is occasionally identified in Papanicolaou-stained smears; Giemsa stain may also be used (Fig. 332-3). For men, a wet mount of material from a platinum loop scraping of the anterior urethra reveals the organism in approximately half the cases. Prostatic massage before collection of urine for Trichomonas culture is a more sensitive diagnostic approach. T. vaginalis is not found in the gastrointestinal tract, and the presence of trichomonads in wet fecal mounts or stained fixed fecal smears (iron hematoxylin or trichrome) most likely represents the commensal Pentatrichomonas (formerly Trichomonas) hominis. Serology has limited clinical use because of issues of sensitivity and specificity and because evidence of exposure does not imply current disease.

Trichomoniasis  

EPIDEMIOLOGY

Trichomonas vaginalis is among the most prevalent of all pathogenic protozoa and is one of the most common sexually transmitted infections in the United States and likely worldwide.9 As many as 30% of female college students and 40% of pregnant Nigerian women were found to be infected. The highest incidence of infection occurs in women with multiple sexual partners and those with other sexually transmitted diseases (Chapter 269): 24% of HIVpositive South African women presenting for their first antenatal screening were infected by T. vaginalis and 4.5% of 10,000 women presenting to U.K. STD clinics for chlamydia and gonorrhea testing by NAAT were found to be infected by T. vaginalis. T. vaginalis also can be passed from infected mothers to their newborn daughters, but it is seldom symptomatic in girls before menarche. The parasite is able to survive for some time in moist environments, and nonvenereal transmission, although uncommon, can occur. Trichomoniasis, like other sexually transmitted diseases, may increase the likelihood of transmission of HIV.  

PATHOBIOLOGY

T. vaginalis, known colloquially as a flagellate, is classified in the phylum Metamonada and class Parabasalia, along with another human pathogen,

FIGURE 332-3.  Trichomonas vaginalis, Giemsa-stained smear of cultivated trophozoites. (Bright-field microscopy, ×630. Scale bar is 15 µm.)

CHAPTER 332  Babesiosis and Other Protozoan Diseases  

The parabasalids are a phylogenetic sister group to the class Eopharingia, which includes Giardia spp. The parabasalids lack mitochondria and are anaerobic; they all have a unique cellular organelle, the hydrogenosome, which is a relic of the mitochondrion and serves as the site of anaerobic pyruvate metabolism. A hemolysin is produced and may cause epithelial damage.

2114.e1

CHAPTER 332  Babesiosis and Other Protozoan Diseases  

Diverse CLIA-waived, CLIA moderate level, or FDA-approved rapid pointof-care tests are now available, as are NAAT through commercial clinical laboratories, all of which have excellent sensitivities and specificities12 and almost always outperform the classical microscopy based tests.

TREATMENT (also see Chapter 323) 

wall is deposited around the zygote. The sexual cycle begins about a week after infection, with oocysts sloughing into the bowel lumen and subsequently out of the body through feces. Biopsy specimens from infected patients demonstrate mononuclear and eosinophilic infiltrates in the lamina propria as well as alterations to the morphology of villi. Humans appear to be intermediate hosts for certain Sarcocystis spp, with sporozoites liberated in the small bowel, entering the vasculature, and forming cysts in muscle.  

Tinidazole, a single 2-g oral dose in adults, or metronidazole, either as a single 2-g oral dose or 500 mg twice daily for 7 days, is the treatment of choice with 7-day treatment providing a higher cure rate. A1  Tinidazole is the better tolerated of the two. Single-dose therapy (metronidazole or tinidazole) ensures compliance of the patient but can produce nausea and a metallic taste, particularly with metronidazole. Both tinidazole and metronidazole have a disulfiram-like effect, and patients who consume alcohol within 24 hours of metronidazole or 72 hours of tinidazole may experience severe nausea, vomiting, and flushing. The use of tinidazole and metronidazole is relatively contraindicated during pregnancy, given the lack of well-controlled studies. Treatment failures with metronidazole are uncommon but well documented. HIV-positive women should be treated for 7 days with 500 mg per day of metronidazole because of frequent recrudescence with the single-dose therapy. Some instances of treatment failure in immune-intact women result from reinfection, others from poor compliance, but some are caused by metronidazole-resistant parasites. A repeated course of metronidazole (2 g orally daily for 5 days) may be tried. If patients remain refractory to appropriate treatment, metronidazole sensitivity can be tested by the CDC (available at www.dpd.cdc.gov/dpdx/HTML/DiagnosticProcedures.htm). Metronidazole gel delivered intravaginally does not achieve therapeutic levels but clinical improvement has been documented with intravaginal boric acid.



PREVENTION

Condoms (male or female) reduce risk of acquiring trichomoniasis. Sexual partners should be treated concurrently to prevent reinfection because nearly 20% of male partners are coinfected.  

PROGNOSIS

Rare complications include pelvic inflammatory disease. Infection during gestation may lead to fetal growth retardation. There is no natural or acquired immunity, so reinfection may be common.  

Coccidian Enteritis  



PATHOBIOLOGY

Oocysts are passed in feces and must sporulate for at least a day (C. belli) or 5 to 11 days (C. cayetanensis) before attaining infectivity. Sporozoites are liberated in the small bowel, penetrating enterocytes (for C. cayetanensis, mainly in the jejunum). C. cayetanensis appears to have a complicated developmental cycle with at least two merogonic cycles in the bowel, leading to the formation of gametes. The gametes fuse within the enterocyte cytoplasm, and an oocyst

CLINICAL MANIFESTATIONS

After an incubation period of approximately 1 week, either organism produces watery diarrhea, nausea, vomiting, abdominal pain, myalgias, anorexia, and fatigue. Illness is usually self-limited, but symptoms can be prolonged (10 to 12 weeks) and associated with steatorrhea, flatulence, and substantial weight loss in persons who are immunocompromised, especially those with AIDS. Infections may also be asymptomatic. With S. nesbitti, fever and prominent myalgias are the dominant presentation. Cystoisospora has been detected in the gallbladder and apparently causes a subclinical infection in immunocompetent individuals, but is associated with acalculous cholecystitis in immunosuppressed patients.14 Sarcocystosis may comprise infection with zoonotic species or with those that have humans as definitive hosts. With the former, diverse species cause aberrant infection with a tropism to muscle; myalgias and edema may result. The latter usually colonize the intestinal epithelium but are not associated with disease. Rarely, a diarrheal disease may result. Both kinds of sarcocystosis are acquired by ingesting poorly cooked meat.  

DIAGNOSIS

The diagnosis is confirmed by identifying coccidia in stool samples stained with modified acid-fast or modified safranin preparations or by phase contrast microscopy or bright-field microscopy (using iodine as a contrast medium) of wet mounts (Fig. 332-4). C. cayetanensis and C. belli may be sensitively detected by fluorescent microscopy of wet mounts. The CDC can help confirm the identification of these enteric protozoa by telediagnosis (sending a digital image of suspicious microscopy findings). PCR is specific and can be sensitive, depending on the mode of DNA extraction, but there are no FDA-approved assays available; PCR support may be requested from the CDC through state public health departments. S. nesbitti and likely other zoonotic Sarcocystis spp appear to undergo an aberrant asexual cycle only within humans and may not produce oocysts to be liberated into the bowel; therefore, diagnosis has required muscle biopsy and demonstration of sarcocysts by histology or detection of the agent’s DNA by PCR.

EPIDEMIOLOGY

The coccidia, with 43 genera and more than 1700 recognized species, are well-known veterinary pathogens. At least two of the coccidia, Cyclospora cayetanensis (an eimeriid) and Cystoisospora belli (a sarcocystid), are causative agents of enteritis in humans. Despite their ubiquity (nearly 100 species have been described), Sarcocystis spp have been rare causes of human enteritis and an even less common cause of myositis,13 although Sarcocystis nesbitti caused a disease comprising fever, myalgia, headache, and myositis in 89 college students who had traveled to Malaysia. C. cayetanensis may have both an animal and human reservoir, but C. belli is thought to be an anthroponosis. Cyclosporiasis is a cause of gastroenteritis in tropical and subtropical areas, with Peru, Mexico, Haiti, Caribbean countries, and Nepal commonly reporting such cases. During 1996 to 2016, 595 cases of cyclosporiasis were reported through FoodNet [https://wwwn.cdc.gov/foodnetfast/] but this surveillance program comprises only 10 states (15% of the U.S. population). Outbreaks involving hundreds of cases are common and invariably traced to contaminated fresh produce such as raspberries, basil, snow peas, or mesclun. Such outbreaks have a distinct seasonality, with most cases presenting during June and July. Cyclosporiasis is commonly diagnosed in international travelers, and large outbreaks have been reported from cruise ships. Isosporiasis/cystoisosporiasis remains understudied despite it being described in military patients during World War I. It is mainly reported from AIDS patients.

2115

TREATMENT  Rehydration is important, as it is for any severe diarrheal disease. Both infections may respond to treatment with 160 mg trimethoprim and 800 mg sulfamethoxazole taken twice daily for 7 to 10 days. HIV-infected patients may require a longer course of therapy. The widespread use of trimethoprim-sulfamethoxazole (Bactrim) prophylaxis for Pneumocystis has reduced the incidence of coccidial diarrhea in patients with HIV infection. No specific treatment has been recommended for sarcocystosis.



PREVENTION

At the community level, preventing the contamination of water and food (mainly vegetables and fruits) by animal or human feces reduces the risk of transmission. Washing vegetables and fruits in water will reduce the potential inoculum but does not eliminate all risk. Sarcocystosis may be prevented by ensuring that meat is well cooked.  

PROGNOSIS

Reactive arthritis, Guillain-Barré syndrome, Reiter syndrome, cholecystitis, and cholangitis have been reported as complications of either coccidian enteritis, mainly in patients with AIDS. Otherwise, treatment appears to eradicate the organism. Whether reinfection may occur is not known. Reports of extraintestinal development of C. belli and C. cayetanensis suggest the possibility of reinvasion of the bowel with ensuing recrudescence of signs and symptoms.  

Other Enteric Protozoans

A number of other protozoa transmitted by fecal-oral contamination have been associated with enteric disease (Table 332-1). Microsporidia have classically been considered to be protozoa but are now known to be obligately

2116

CHAPTER 332  Babesiosis and Other Protozoan Diseases  

A

B

FIGURE 332-4.  Diagnosis of coccidian enteritis. A, Cyclospora cayetanensis, seen on formalin-fixed human feces, unsporulated. (Bright-field microscopy, ×1000 with green contrast filter. Scale bar is 10 µm.) Note the doublet of yeast cells in lower right of the photomicrograph. B, Cystoisospora belli, seen on formalin-fixed human feces, partially sporulated. (Brightfield microscopy, ×400. Scale bar is 15 µm.)

TABLE 332-1 OTHER ENTERIC PROTOZOA ORGANISM

EPIDEMIOLOGY

MANIFESTATIONS

THERAPY*

Balantidium coli

Primarily an infection of animals, especially pigs, but also affects humans

Asymptomatic or mild and self-resolving; occasionally more severe with abdominal pain, blood, and mucus in stool

Tetracycline (500 mg qid for 10 days) Alternative: metronidazole (750 mg tid for 5 days) or iodoquinol (650 mg tid for 20 days)

Blastocystis hominis

Probably worldwide, including North America; often found concomitantly with Giardia lamblia

Pathogenicity is debated

The need for treatment is debated, but symptomatic improvement has been reported with metronidazole (750 mg tid for10 days), or trimethoprimsulfamethoxazole (160 mg TMP/800 mg SMX bid for 7 days)

Dientamoeba fragilis

Worldwide distribution; frequently found concomitantly with the pinworm Enterobius

Often asymptomatic; diarrhea reported

Paromomycin (25-35 mg/kg body weight per day in 3 doses for 7 days), metronidazole (500-750 mg tid for 10 days), or iodoquinol (650 mg tid for 20 days)

Microsporidia† (Enterocytozoon bieneusi, Encephalitozoon intestinalis, Anncaliia sp, Tubulinosema sp, Nosema sp, Pleistophora sp, Vittaforma sp)

Apparent worldwide distribution

AIDS patients with persistent diarrhea and wasting; self-limited cases in immunocompetent persons; ocular disease is increasingly recognized

Oral fumagillin (20 mg tid) has been effective for E. bieneusi, but it has been associated with thrombocytopenia. Albendazole (400 mg bid) has been effective for E. intestinalis. Treatment with HAART may lead to clinical response in HIV-infected patients with microsporidial diarrhea.

Sarcocystis species

Worldwide distribution, more commonly in sites of cattle or swine production

Often asymptomatic; nausea, vomiting, abdominal pain, and diarrhea may occur; eosinophilic necrotizing enteritis has been reported. Infection is usually asymptomatic for S. hominis and S. suihominis for which humans are definitive hosts (parasites undergo sexual reproduction); zoonotic Sarcocystis spp cause aberrant infection with muscle tropism, causing myalgias and edema.

No specific therapy

*Based on CDC recommendations, www.cdc.gov/parasites/az/index.html. Accessed May 9, 2019. The dosages and durations are for adults. † Associated with persistent, severe diarrhea in persons with AIDS. AIDS = acquired immunodeficiency syndrome; HAART = highly active antiretroviral therapy; HIV = human immunodeficiency virus.

intracellular fungi15; they are a diverse group and at least 7 genera have been implicated as human pathogens. They are discussed here because diagnostic parasitology laboratories continue to confirm diagnoses of these agents. Dientamoeba, Blastocystis, Balantidium, and some of the microsporidia reside in the lumen of the bowel, and others invade and multiply within enterocytes; heavy infection may result in diarrhea. Some of the microsporidia cause conjunctivitis or other ocular disease. Enteric protozoa should be considered in the differential diagnosis of patients with persistent diarrhea and abdominal symptoms, particularly those with a history of recent international travel. A clinical diagnosis is rarely possible; laboratory tests, mainly for ova and parasites in stools, establish the diagnosis. Expert microscopists are required because these parasites may be confused with fecal debris. Pathogenic protozoa must also be differentiated from commensals such as Entamoeba coli, Endolimax nana, Iodamoeba bütschlii, Pentatrichomonas hominis, and Chilomastix mesnili. Therapy includes

administration of the appropriate antiprotozoal drug and rehydration, as listed in Table 332-1 (also see Chapter 323).

  Grade A Reference A1. Kissinger P, Muzny CA, Mena LA, et al. Single-dose versus 7-day-dose metronidazole for the treatment of trichomoniasis in women: an open-label, randomised controlled trial. Lancet Infect Dis. 2018;18:1251-1259.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 332  Babesiosis and Other Protozoan Diseases  

GENERAL REFERENCES 1. Gray JS, Estrada-Peña A, Zintl A. Vectors of babesiosis. Annu Rev Entomol. 2019;64:149-165. 2. Vannier EG, Diuk-Wasser MA, Ben Mamoun C, et al. Babesiosis. Infect Dis Clin North Am. 2015;29:357-370. 3. Woolley AE, Montgomery MW, Savage WJ, et al. Post-babesiosis warm autoimmune hemolytic anemia. N Engl J Med. 2017;376:939-946. 4. Wang G, Wormser GP, Zhuge J, et al. Utilization of a real-time PCR assay for diagnosis of Babesia microti infection in clinical practice. Ticks Tick Borne Dis. 2015;6:376-382. 5. Kletsova EA, Spitzer ED, Fries BC, et al. Babesiosis in Long Island: review of 62 cases focusing on treatment with azithromycin and atovaquone. Ann Clin Microbiol Antimicrob. 2017;16:1-17. 6. Krause PJ. Human babesiosis. Int J Parasitol. 2019;49:165-174. 7. Moritz ED, Winton CS, Tonnetti L, et al. Screening for Babesia microti in the U.S. blood supply. N Engl J Med. 2016;375:2236-2245. 8. Villatoro T, Karp JK. Transfusion-transmitted babesiosis. Arch Pathol Lab Med. 2019;143: 130-134.

2116.e1

9. Menezes CB, Frasson AP, Tasca T. Trichomoniasis—are we giving the deserved attention to the most common non-viral sexually transmitted disease worldwide? Microb Cell. 2016;3:404-419. 10. Meites E, Gaydos CA, Hobbs MM, et al. A review of evidence-based care of symptomatic trichomoniasis and asymptomatic Trichomonas vaginalis infections. Clin Infect Dis. 2015;61(suppl 8):S837-S848. 11. Bouchemal K, Bories C, Loiseau PM. Strategies for prevention and treatment of Trichomonas vaginalis infections. Clin Microbiol Rev. 2017;30:811-825. 12. Gaydos CA, Klausner JD, Pai NP, et al. Rapid and point-of-care tests for the diagnosis of Trichomonas vaginalis in women and men. Sex Transm Infect. 2017;93:531-535. 13. Fayer R, Esposito DH, Dubey JP. Human infections with Sarcocystis species. Clin Microbiol Rev. 2015;28:295-311. 14. Lai KK, Goyne HE, Hernandez-Gonzalo D, et al. Cystoisospora belli infection of the gallbladder in immunocompetent patients: a clinicopathologic review of 18 cases. Am J Surg Pathol. 2016;40: 1070-1074. 15. Han B, Weiss LM. Microsporidia: obligate intracellular pathogens within the fungal kingdom. Microbiol Spectr. 2017;5:1-17.

2116.e2

CHAPTER 332  Babesiosis and Other Protozoan Diseases  

REVIEW QUESTIONS 1. A 35-year-old woman presented to the emergency department in Toronto with a 2-day history of fever, headache, and malaise. She had just returned from spending 2 weeks during June on Nantucket Island. A 3-inch expanding circular rash was noted on one of her extremities, and she was diagnosed with acute Lyme disease and provided with a prescription for 100 mg of doxycycline, to be taken twice a day for 21 days. A month later, she presents again with fever, headache, chills, myalgia, and fatigue. She had not traveled again and works as a law clerk. Complete blood count demonstrates a hematocrit of 35, hemoglobin of 12, white blood cell count of 3000 (60% neutrophils, 35% lymphocytes, and 5% monocytes), and platelet count of 120,000; a blood smear is normal. What is the most likely cause of the recurrent symptoms? A . Epstein-Barr virus infection B. Human anaplasmosis C. Myelodysplastic syndrome D. Human ehrlichiosis E. Babesiosis Answer: E  Symptomatic Epstein-Barr virus infection is unlikely in a 35-yearold patient and usually is accompanied by lymphocytosis with atypical lymphocytes. Although thrombocytopenia or leukopenia may be seen with anaplasmosis and ehrlichiosis, these infections are susceptible to doxycycline and would have been adequately treated with the course prescribed for the Lyme disease. Myelodysplastic syndrome typically is manifested with anemia and without fever. Treatment failure may rarely occur with acute Lyme disease, mostly as a result of poor compliance of the patient with therapy, but later manifestations of Lyme disease are more likely to be a monarticular arthritis or radiculopathy. Babesiosis may be coacquired with Lyme disease, but babesia are not susceptible to the tetracyclines. The negative blood smear is not unusual, and the severity of illness is not necessarily related to parasitemia. Thrombocytopenia and leukopenia may be evident in the absence of hematologic signs of anemia, particularly in early infection. Polymerase chain reaction will often be positive when blood smears are negative, given its greater sensitivity. Serology with a high anti-Babesia titer would provide evidence of infection if polymerase chain reaction is not performed, given that the patient was likely infected 6 weeks previously. 2. A 30-year-old biologist presented with a 3-week history of watery, nonbloody diarrhea with nausea and low-grade fever. He had previously been in good health and had returned 6 weeks ago from a field trip to Panama, where he had stayed for 1 month. He had several episodes of diarrhea in Panama that had responded quickly to Pepto-Bismol. He had remained hydrated and thus his activity was not curtailed. He now seeks medical attention because the diarrhea fails to respond to Pepto-Bismol or Imodium. Physical examination findings are normal, as is a complete blood count with the exception of slight eosinophilia. The patient admits to eating local foods and failing to always drink bottled water. Leukocytes are not noted in a stool examination, but Charcot-Leyden crystals are present. Stool analysis for ova and parasites demonstrates 30 × 15-µm oocysts with a sharply defined cell wall and a single sporoblast. Rare Entamoeba spp cysts with eight nuclei are also found. Which of the following would provide the most critical information for case management?

A . Testing for Clostridium difficile toxin B. Human immunodeficiency virus (HIV) infection status C. Testing for norovirus D. Specific identification of the Entamoeba sp E. Whether the Pepto-Bismol–treatable diarrhea was similar in quality to that experienced now Answer: B  The oocyst that was detected is likely Cystoisospora belli, which typically causes a self-limited diarrhea in immunocompetent younger patients. Norovirus and C. difficile disease should also terminate within a few days in a younger, healthy individual. The presence of eight nuclei identifies Entamoeba coli. Amebic infection reflects fecal-oral contamination, although the nonpathogenic Escherichia coli may also be found as a gut commensal. It is possible that the diarrhea episodes in Panama were an uncomplicated traveler’s diarrhea and that the current disease is not related. Persistent enteritis due to C. belli is commonly seen in HIV-infected patients, who may develop life-threatening persistent diarrhea, malabsorption, and dehydration and require lifelong suppression of C. belli with trimethoprim-sulfamethoxazole (Bactrim). 3. A 40-year-old married woman presents with a foul-smelling vaginal discharge of 3 days’ duration. She is sexually active with her husband and complains of a recurring history of discharge during the last 6 months, typically a week after intercourse, but not with every instance of intercourse. The discharge would resolve within 2 weeks if she was abstinent. A wet mount of the discharge demonstrates flagellated cells moving with a tumbling motion, and a diagnosis of trichomoniasis is made. Metronidazole 250 mg orally three times a day is prescribed for 7 days. She returns 2 months later with the same discharge, stating that she had improved after the treatment course but that the discharge returned within the last week. Which of the following is the most likely explanation for the recurrence of symptoms? A . The patient’s husband has Trichomonas infection and is reinfecting the patient. B. The patient has antibiotic-resistant trichomoniasis. C. The original diagnosis based on the wet mount examination alone was incorrect. D. The patient has been noncompliant with treatment. Answer: A  Patients may be reinfected by their untreated sexual partners, and thus a diagnosis of trichomoniasis should imply presumptive treatment of the partner. The majority of infected men are asymptomatic. The current Centers for Disease Control and Prevention recommendations for treatment failure are to re-treat with tinidazole. Metronidazole susceptibility may be established, but there is no guidance on minimum inhibitory concentration thresholds to define resistant strains. Although Trichomonas vaginalis is detected on wet mount examination of vaginal secretions in only about 60% of infected women, it confirms the diagnosis when it is seen in the appropriate clinical setting.

2117

CHAPTER 333 Cestodes  

333  CESTODES



CLINICAL MANIFESTATIONS AND DIAGNOSIS

In most of those infected, Diphyllobothrium species produce few or no symptoms. Some may complain of gastrointestinal symptoms (abdominal discomfort, nausea, weight loss). The main clinical manifestation is the observation of proglottids being passed in stool. Pernicious anemia with symptoms of anemia or peripheral neuropathy may develop with D. latum infection. The diagnosis depends on observation of the characteristic operculated eggs in stool.

A. CLINTON WHITE, JR., AND ENRICO BRUNETTI



tapeworm, which can out-compete the host, leading to vitamin B12 deficiency (Chapter 155). This manifestation has been described only in Scandinavia.

DEFINITION

The Pathogens

Cestode parasites are members of the animal kingdom, subphylum Cestoda. The organisms are characterized by several life cycle stages, which typically develop in distinct hosts. The adult stage is the tapeworm, which is acquired by ingestion of uncooked tissues harboring larval forms. After ingestion, the larvae excyst and the scolex attaches to the intestines. Segments, termed proglottids, develop at the base of the scolex and are displaced from the scolex by new proglottids to form a chain or tapeworm. The host in which the tapeworm develops is termed the definitive host. The proglottids contain male and female sexual organs and produce large numbers of ova. The proglottids or their ova are shed in stools. Humans are the definitive hosts for a number of different tapeworms, including the Taenia species, Diphyllobothrium species, and Hymenolepis nana. Humans can also be an accidental host for the dog and cat tapeworms of the genus Dipylidium (Table 333-1). The intermediate hosts harbor the larval form of the parasite. Infection follows ingestion of the ova. Under the influence of gastric and intestinal fluids, the ova hatch, releasing the invasive larvae (oncospheres), which migrate to tissues, forming tissue forms. The forms in tissue vary between organisms and may include the cysticercus (a bladder containing a single invaginated scolex), the coenurus (a bladder with multiple scolices), the hydatid (a cystic structure with a germinal layer, which forms numerous protoscolices), or the plerocercoid (a solid form seen in Spirometra species). Humans can harbor the intermediate forms of Taenia solium (cysticercosis), Echinococcus granulosus group (cystic hydatid disease), Echinococcus multilocularis (alveolar hydatid disease), and rarely other organisms (Table 333-2). Humans can serve as both the definitive host and an intermediate host for two species, T. solium and H. nana. In the case of T. solium, humans are the obligate host for the tapeworm stage (pork tapeworm) but can also harbor the cystic form (cysticercosis). In the case of H. nana, both stages typically develop in a single person, with the cysticercoid form in the intestinal wall and the tapeworm in the lumen.

TREATMENT AND PREVENTION  A single oral dose of praziquantel (5 to 10 mg/kg) is usually adequate for therapy (Table 333-3). Niclosamide can be used as an alternative (2 g [adults] or 50 mg/kg [children] in a single dose chewed and swallowed), but it is not available in the United States. Parasites in fish can be killed by cooking (>56° C, >5 minutes) or freezing (−20° C, 24 hours). Infected fish may also be identified by inspection.



Hymenolepis nana

Hymenolepis nana is the human dwarf tapeworm. Hymenolepis diminuta, a rat tapeworm, can also cause human infection.  

EPIDEMIOLOGY AND PATHOBIOLOGY

H. nana is prevalent worldwide, with estimates of at least 50 to 75 million people infected. Infection follows ingestion of ova. The larvae are released, invade, and develop into cysticercoid forms in the intestinal villi. After a few days, the cysticercoids mature, invade the lumen, and are transformed into a scolex, forming small tapeworms (up to 5 cm long), which begin producing eggs within 2 to 3 weeks. Autoinfection either in the intestines or by the fecal-oral route can lead to heavy infection.  

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Most infections are asymptomatic. However, some children may be infected by hundreds or thousands of worms, which can cause abdominal pain, loose stools, diarrhea, and malabsorption. Diagnosis depends on observation of the characteristic eggs in stool. More than one specimen may be required.

  INTESTINAL TAPEWORM INFECTIONS



Diphyllobothrium Species (Fish Tapeworm)

Diphyllobothrium tapeworms are large segmented parasites that are acquired by ingestion of undercooked or pickled freshwater fish dishes (sushi, sashimi, ceviche, carpaccio, gefilte fish). The tapeworms develop within a few weeks and can live for more than 10 years.  

EPIDEMIOLOGY AND PATHOBIOLOGY

Diphyllobothrium species are found worldwide, including foci in Europe, North and South America, and Asia. Perhaps 20 million people are thought to be infected worldwide. Major foci include Russia, Japan, and South America. Disease was formerly highly endemic in Scandinavia, where it is now rarely diagnosed. In most cases, infection has little impact on the host. However, one species, Diphyllobothrium latum, contains vitamin B12 receptors on the surface of the

TABLE 333-2 HUMAN LARVAL CESTODE INFECTIONS ORGANISM

COMMON NAME

ORGANS INVOLVED

Taenia solium

Cysticercosis

Brain, spinal fluid, eye, muscle

Echinococcus granulosus group

Cystic hydatid disease

Liver, lung, other

Echinococcus multilocularis

Alveolar hydatid disease

Liver

Taenia multiceps, Taenia spp

Coenurosis

Brain, eyes

Spirometra species

Sparganosis

Subcutaneous tissue, viscera

TABLE 333-1 COMMON HUMAN TAPEWORM INFECTIONS ORGANISM

INTERMEDIATE HOST

COMMON NAME

CLINICAL PRESENTATION

TREATMENT

Diphyllobothrium spp

Fish

Fish tapeworm

Passing segments, pernicious anemia

Praziquantel, niclosamide

Hymenolepis nana

Humans

Dwarf tapeworm

Asymptomatic, diarrhea

Praziquantel, niclosamide

Taenia saginata

Cattle

Beef tapeworm

Asymptomatic, passing segments

Praziquantel, niclosamide

Taenia asiatica

Pigs

Asian tapeworm

Asymptomatic, passing worms

Praziquantel, niclosamide

Taenia solium

Pigs

Pork tapeworm

Asymptomatic, passing segments

Praziquantel, niclosamide

Dipylidium caninum

Fleas

Dog tapeworm

Passing segments

Praziquantel, niclosamide

CHAPTER 333 Cestodes  

ABSTRACT

Cestode infections are neglected zoonotic infections that have widespread distributions. Neurocysticercosis is a common cause of neurologic disease in endemic countries. Diagnosis and management vary with the number of parasites and site of infection. Recent advances include the use of combination antiparasitic therapy for multiple parenchymal cysts. Cystic echinococcal disease requires careful staging and optimal management requires attention to the stage of disease.

KEYWORDS

taenia solium neurocysticercosis Echinococcus hydatid disease tapeworm taeniasis

2117.e1

2118

CHAPTER 333 Cestodes  

TABLE 333-3 THERAPY FOR INTESTINAL TAPEWORM INFECTIONS PRAZIQUANTEL Dosage Adults

NITAZOXANIDE

2 g (4 tablets)

500 mg

1.5 g (3 tablets)

200 mg

1 g (2 tablets)

100 mg

Administration Taken as a single dose

Taken as a single dose; tablets must be chewed and swallowed

Taken twice a day for 3 days

Side effects

Nausea, vomiting, abdominal pain, diarrhea, drowsiness, dizziness, headache, pruritus

Children >34 kg Children 11-34 kg

5-10 mg/kg for all age groups (25 mg/kg for Hymenolepis nana)

NICLOSAMIDE



Mild but frequent, including dizziness, myalgias, nausea, vomiting, diarrhea, abdominal pain

Pregnancy

No known mutagenic effects; considered safe if indicated



TREATMENT, PREVENTION, AND PROGNOSIS 

Praziquantel (15 to 25 mg/kg as a single oral dose) is usually effective in treating H. nana infection,1 but it may need to be repeated in heavy infection (see Table 333-3). Nitazoxanide (100 mg by mouth twice daily for 3 days for children 1 to 3 years of age, 200 mg by mouth twice daily for 3 days for children 4 to 11 years of age, and 500 mg by mouth twice daily for 3 days for older children) is a reasonable alternative therapy; efficacy is about 75 to 82%. Niclosamide also can be used as an alternative, but must be continued for 7 days. Transmission is by the fecal-oral route and could be prevented by improved hygiene. Mass chemotherapy has been used to control infection in some populations.

Taeniasis can be treated with praziquantel in a single dose (see Table 333-3). Single doses of niclosamide are also effective. Nitazoxanide has also been used for T. saginata. Taeniasis can be prevented by inspection of beef. Also, cooking to 56° C for 5 minutes or freezing at −20° C for 7 to 10 days destroys the infective larvae. Only minor symptoms are noted and are eventually self-limited with or without treatment.



Dipylidium caninum

Dipylidium caninum is a common tapeworm of dogs and cats. Dogs are infected by ingestion of fleas, which carry the cysticercoid form in their body cavities. The tapeworms can also develop in children who have ingested the fleas. It is widespread worldwide, but human infections are unusual.  

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Infection may be asymptomatic. In some cases, the motile proglottids may be noted in stool. The proglottids are similar in size and shape to rice grains. Diagnosis depends on identification of the ova in stool or identification of the proglottids.

TREATMENT AND PREVENTION  There are no controlled trials of treatment for Dipylidium infection, but infection is likely to respond to regimens used for other tapeworms (see Table 333-3). The main measure for prevention is treatment of pets for fleas and tapeworms.

Taenia asiatica

T. asiatica is a cause of taeniasis in Asia, termed Asian taeniasis. Infection is acquired by ingestion of undercooked pork. Pigs are infected by ingestion of the ova from tapeworm carriers. T. asiatica has been widely described in China, Taiwan, Korea, Indonesia, and Southeast Asia. The clinical manifestations, diagnosis, treatment, and prevention of T. asiatica infection are similar to those noted for T. saginata infection.  

Taenia solium

T. solium, also known as the pork tapeworm, can cause both tapeworm infection and larval infection termed cysticercosis. T. solium tapeworm infections are caused by ingestion of infected undercooked pork. The scolex evaginates and attaches to the intestines, forming proglottids. The proglottids gradually mature as they are separated from the scolex by new proglottids. The adult worms are often 10 to 20 feet long. Within the mature proglottids, thousands of microscopic ova develop. The ova are either excreted into the stool or shed with the proglottids. By contrast, ingestion of the ova results in development of larval infection, termed cysticercosis (see later). Thus, the tapeworm carrier poses risk of self-infection as well as infection to other people.  



Taenia saginata

Taeniasis refers to infection with the tapeworm form of one of three Taenia species. Taenia solium and Taenia asiatica are acquired from ingestion of undercooked pork. Taenia saginata, called the beef tapeworm, is a common intestinal infection worldwide. Cattle are the intermediate hosts, harboring the tissue cysticerci in their muscle. Humans are the obligate definitive host, harboring the tapeworm form.

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Mild symptoms (e.g., nausea, abdominal discomfort, anorexia, and pruritus) may be noted. The motile proglottids may cause discomfort as they exit the anus or may be noted in stool. Ova may be noted in stool. The ova are 40 µm in diameter, surrounded by brown radial striations, and the embryos have six hooks. However, the ova of the three Taenia species are morphologically indistinguishable. The proglottids can be distinguished from those of T. solium by counting the number of uterine branches (≥14 branches suggests T. saginata). However, the proglottids of T. saginata cannot be readily distinguished from T. asiatica.

TREATMENT AND PREVENTION 



EPIDEMIOLOGY AND PATHOBIOLOGY

T. saginata is common worldwide in areas where cattle are raised and human fecal material contaminates the pastures. Approximately 45 to 60 million people are thought to be infected. It is found on most continents. Very high rates (>20% of the population) have been noted in east Africa, Bali, and Tibet. It is likewise endemic in the Middle East, the Americas, and Europe. T. saginata is also common in other parts of Asia, but many of the epidemiologic studies did not differentiate T. saginata from T. asiatica. T. saginata tapeworms are acquired by ingestion of undercooked beef. The scolex attaches to the intestinal wall, and proglottids form at the base of the scolex. The proglottids gradually enlarge as they are displaced from the scolex by newer proglottids. The chain of proglottids can reach a length of up to 30 feet. The terminal proglottids are shed periodically in the stool. Terminal proglottids are typically off-white, 2 to 3 cm long, 0.5 to 1 cm wide, and 1 to 2 mm thick.

EPIDEMIOLOGY AND PATHOBIOLOGY

T. solium is common worldwide in areas where pigs are raised and where pigs have access to human fecal material. Only a few million people are thought to harbor the tapeworm form. Pork tapeworm infection is highly endemic in Latin America, sub-Saharan Africa, south, east, and southeast. T. solium tapeworms are acquired by ingestion of undercooked pork. The scolex attaches to the intestinal wall and proglottids form at the base of the scolex. The proglottids gradually enlarge as they are displaced from the scolex by newer proglottids. The terminal proglottids are shed periodically in the stool. Terminal proglottids are typically off-white, 2 cm long, 0.5 to 1 cm wide, and 1 to 2 mm thick.

2119

CHAPTER 333 Cestodes  



CLINICAL MANIFESTATIONS AND DIAGNOSIS

Mild symptoms (e.g., nausea, abdominal discomfort, anorexia, and pruritus) may be noted. The proglottids may be noted in stool. Diagnosis is made by the finding of ova in stool. The ova are 40 µm in diameter, surrounded by brown radial striations, and embryos have six hooks and are morphologically indistinguishable from the other Taenia species.

A

B

C

D

E

F

G

H

TREATMENT AND PREVENTION  Taeniasis can be treated with praziquantel in a single dose. However, praziquantel should be used with caution in areas where cysticercosis is prevalent, because it can precipitate symptoms such as seizures. Single doses of niclosamide are also effective. Taeniasis can be prevented by inspection of pork. Also, cooking to 56° C for 5 minutes or freezing at −20° C for 7 to 10 days destroys the infective larvae. Current control measures include mass chemotherapy for entire populations with praziquantel (see below in the section on cysticercosis)

FIGURE 333-1.  Human neurocysticercosis can be classified on the basis of neuroimaging studies. A, Multiple cystic lesions. B, Single enhancing lesion. C, Multiple calcifications. D, Intraventricular cysticerci. E, Subarachnoid cysticerci. F, Diffuse infection with cerebral edema, termed cysticercal encephalitis. G, Ocular cysticerci. H, Diffuse muscle calcifications. (Reprinted from Garcia HH, Del Brutto OH. Neurocysticercosis: updated concepts about an old disease. Lancet Neurol. 2005;4:653-661.)

  TISSUE CESTODE (CYST) INFECTION



Taenia solium (Cysticercosis)  

DEFINITION

T. solium is the cause of human larval infection termed cysticercosis. The normal hosts for the larval (cysticercus) forms are pigs. When ingested by pigs, the ova hatch, releasing the invasive larvae (termed oncospheres), which invade the intestines, migrate to tissues (especially muscle), and mature into cysticercus forms within the tissues. The cysticercus consists of a thin translucent bladder containing an invaginated scolex, which is poised to form a tapeworm after being ingested by a human host. The ova are also infectious to people, including the tapeworm carrier. The sticky ova attach to the hands of the tapeworm carrier and are transmitted by the oral route to the carrier or close contacts. After ingestion, the ova can migrate to tissues and form cysts (cysticercosis). The presence of cysticerci in the central nervous system is termed neurocysticercosis. Neurocysticercosis includes cysticerci in the brain parenchyma (parenchymal neurocysticercosis) and cysticerci in the ventricles, subarachnoid space, spine, and eye (extraparenchymal neurocysticercosis).  

EPIDEMIOLOGY

Cysticercosis is found in all regions of the world where pigs are raised with access to human fecal material.2 However, exact data on incidence and prevalence are available from only a limited number of studies because of the requirement for neuroimaging studies to make a diagnosis. In the 19th century, infection was highly endemic in Europe. However, with improving standards of living, local transmission is now limited to a few rural areas in southern and eastern Europe. Cysticercosis is widespread in rural areas of Latin America. In endemic villages, more than 10% of the population may have abnormalities on neuroimaging studies consistent with neurocysticercosis.3 Studies have highlighted the importance of cysticercosis in sub-Saharan Africa. Cysticercosis is widespread in India, Nepal, Southeast Asia, and parts of China. In India, the most common manifestation of neurocysticercosis is seizures and a single enhancing lesion. In the United States, over 2000 cases are diagnosed each year. Most cases are in immigrants from pig-raising villages in Mexico and Latin America. However, there are also imported cases from Asia and a few locally acquired infections.  

PATHOBIOLOGY

For cysticercosis, the pathogenesis and pathophysiology vary with the location of the cysticerci and the host inflammatory response. Cysticerci in the brain parenchyma initially suppress the host inflammatory response. After a silent period, estimated to be several years, the cysticerci lose the ability to suppress the host inflammatory response, leading to parenchymal inflammation, which typically is manifested by seizures. The cysticerci induce a granulomatous response, which gradually degrades the parasites. In some cases, the lesions resolve. However, in others, degradation leads to formation of calcified granulomas. These calcified lesions may intermittently become inflamed (as evidenced by edema or contrast enhancement on magnetic resonance imaging [MRI] scan) and may cause recurrent seizures during a period of years. In some cases, cysticerci develop within the ventricles of the brain and can mechanically cause obstructive hydrocephalus. Cysticerci in the subarachnoid space may cause a chronic arachnoiditis, which can be manifested by vasculitis and stroke,

communicating hydrocephalus, basilar meningitis, and, in some cases, mass effect. Cysticerci can also develop in the spine (manifested as radiculitis), eye, subcutaneous tissue, and muscle.  

CLINICAL MANIFESTATIONS

The clinical manifestations vary with the location of the cysticerci and the associated host response (Fig. 333-1).4 All forms of disease may be associated with headaches. In general, parenchymal cysticerci are associated with seizures, whereas ventricular and subarachnoid cysticercosis are associated with hydrocephalus.

Single Enhancing Lesion

A single enhancing lesion is the most common manifestation of cysticercosis in India and the United States. Patients typically present with seizures, which can be focal or focal with secondary generalization. Many will have a single seizure or a few seizures during the period when the cysticercus is degenerating, but the duration of seizures is eventually self-limited in most cases. However, a few go on to develop calcified lesions, which are a risk factor for recurrent seizures.

Multiple Parenchymal Cysticerci

In patients with multiple lesions, the main presentation is with seizures, associated with parenchymal inflammation. In contrast to those with single lesions, seizures are more likely to recur.

Calcified Lesions

Many patients do not present until after they have calcified lesions. Patients with calcified lesions may develop recurrent seizures during a period of years. Some cases are associated with the development of mesiotemporal sclerosis and refractory epilepsy (Chapter 375).

Ventricular Cysticerci

The cysticerci typically are manifested by obstructive hydrocephalus. The patient may present with headache, nausea and vomiting, dizziness, altered mental status, or papilledema with altered vision. This is a medical emergency and can be fatal if it is not treated.

Subarachnoid Cysticerci

Cysticerci in the basilar cisterns are often accompanied by cysticerci in other locations, including parenchymal cysticerci or calcifications, ventricular cysticerci, and spinal or ocular cysticerci; patients may present with disease attributable to cysticerci at these sites. Cysticerci in the basilar cisterns are particularly prone to cause arachnoiditis. Manifestations of arachnoiditis may include vascular involvement (large- or small-vessel strokes), meningeal signs, or communicating hydrocephalus (headaches, nausea, vomiting, dizziness, and altered mental status).  

DIAGNOSIS

The major clinical manifestations of neurocysticercosis (e.g., seizures and hydrocephalus) are not specific, and it is difficult to identify the parasites. The

2120

CHAPTER 333 Cestodes  

main tools used to diagnose neurocysticercosis are neuroimaging studies. Computed tomography (CT) scans are sensitive for identification of parenchymal calcifications, which appear as 2- to 5-mm nodules. CT may also reveal parenchymal cysticerci or obstructive hydrocephalus. MRI scans are more sensitive for identification of the cysticerci, especially in the subarachnoid space and ventricles. Three-dimensional sequences such as fast imaging employing steady-state acquisition (FIESTA) are particularly effective for ventricular and subarachnoid cysticerci. The cysticerci are typically round, 1 to 2 cm in diameter. The cyst fluid is usually isodense with spinal fluid. In uninflamed cysticerci, the walls may not be visible. However, most cases demonstrate enhancement of the cyst walls or surrounding tissues and associated edema. In some cases, the scolex may be visible as a 1- to 2-mm solid nodule, cylinder, or spiral on the side of the cystic lesions. Serologic tests are useful to confirm the diagnosis. Assays using crude antigen, including enzyme-linked immunosorbent assay, are associated with poor sensitivity and specificity and are not reliable. An immunoblot assay using semipurified membrane glycoproteins is highly specific for the diagnosis. The sensitivity is excellent in cases with extraparenchymal or multiple parenchymal cysticerci. However, the sensitivity is poor in those with single enhancing lesions or just calcifications. Antigen detection assays are sensitive and more specific for viable cases and are increasingly available in the United States. Revised diagnostic criteria published recently have defined neurocysticercosis as a “definitive diagnosis” or only a “probable diagnosis.”5

TREATMENT  Treatment varies with the clinical manifestations and form of infection.6,7 Seizures should be treated with antiepileptic drugs (Chapter 375). Phenytoin and carbamazepine are typically used and can control the seizures. Newer antiepileptic drugs may be more effective. There are no viable parasites in those with just calcified lesions, so the main measure is to control symptoms (e.g., antiepileptic drugs for those with seizures). In patients presenting with hydrocephalus, surgery to reestablish cerebrospinal fluid flow is the critical initial step in management. The role of antiparasitic drugs (Chapter 323) varies with the form of infection. Randomized controlled trials have demonstrated more rapid resolution of parenchymal cystic lesions with fewer generalized seizures in those treated with corticosteroids and albendazole (15 mg/kg/day in two daily doses) compared with those treated with placebo. A1  Praziquantel (50 to 100 mg/kg/day in three daily doses) can be used as an alternative. There are emerging data on the use of the two drugs in combination, which may have greater cysticidal activity in those with more than 2 parenchymal cysticerci. A2  A number of clinical trials in subjects with single enhancing lesions have demonstrated a slightly more rapid radiologic resolution and fewer seizures in those treated with steroids and antiparasitic drugs. A3  A4  Higher-dose steroids may decrease seizures near the time of antiparasitic treatment, but the differences are not dramatic. For patients with cysticerci in the ventricles, management usually involves removal of the cysticerci. This can be best achieved by neuroendoscopy for lateral and third ventricular cysticerci. However, the main alternative approach is placement of a ventriculoperitoneal shunt. Chronic steroids and antiparasitic drugs may decrease the rate of shunt failure, which usually results from clogging of the shunts by the cysticerci or proteinaceous debris. There are no controlled trials on the management of subarachnoid cysticercosis. However, expert opinion supports treatment of subarachnoid cysticercosis with prolonged courses of antiparasitic drugs (e.g., albendazole for months), higher doses of albendazole, or combinations of albendazole and praziquantel. Chronic anti-inflammatory medications (e.g., prednisone 1 mg/ kg/day, or 24 mg/day of dexamethasone) are also critically important. Patients who will be treated with chronic steroids should be screened for Mycobacterium tuberculosis and Strongyloides infections before initiation of steroid therapy. Methotrexate is increasingly recommended as a steroid-sparing agent. Patients with hydrocephalus should be treated by cerebrospinal fluid diversion (e.g., ventriculoperitoneal shunting). Before treatment with antiparasitic drugs, patients should undergo a funduscopic examination. Intraocular parasites may develop brisk inflammatory responses after treatment with antiparasitic drugs. Because this inflammation could lead to blindness, most authorities recommend extraction of the parasites before antiparasitic therapy. However, there are also reports of treatment of intraocular parasites with antiparasitic drugs. ,



PREVENTION

Although transmission of T. solium is potentially preventable, population-wide eradication efforts have proved very difficult. Recently, however, transmission of T. solium infection was interrupted in a highly endemic region of Peru by a program emphasizing mass antiparasitic treatment of the human population

with niclosamide, treatment of pigs with oxfendazole, plus vaccination of piglets. A5   

PROGNOSIS

The prognosis varies significantly between the different forms of neurocysticercosis. Parenchymal enhancing lesions and parenchymal cystic lesions will eventually resolve, but this may take months to years. Patients who have or develop calcifications are at lifelong risk for recurrent seizures. Patients with ventricular or subarachnoid disease are at high risk for morbidity and mortality. However, a recent case series noted no deaths with optimal management.  

Cystic Hydatid Disease (Echinococcus granulosus group)  

DEFINITION

Cystic echinococcosis, also called cystic hydatidosis, is caused by the larval stage of cestodes of the Echinococcus granulosus complex. All of these parasites were initially thought to be a single species, E. granulosus. However, molecular studies demonstrate that E. granulosus comprises a number of different species and genotypes. In humans, the clinical manifestations range from asymptomatic infection to severe, potentially fatal disease. Echinococcal cysts consist of a periparasitic host tissue (pericyst or adventitia), which encompasses the larval endocyst, and the endocyst itself. The endocyst has an outer, acellular laminated layer and an inner, or germinative, layer that gives rise to brood capsules and protoscolices. The cyst is filled with clear fluid, numerous brood capsules, and protoscolices. Some cysts may also harbor daughter cysts of variable size. The protoscolices convert to tapeworms in the canine definitive hosts but can also form new cysts when released in mammalian tissues.  

EPIDEMIOLOGY

E. granulosus species occur on all continents and in circumpolar, temperate, subtropical, and tropical zones. The highest prevalence of the parasite is found in parts of Eurasia, Africa, Australia, and South America. Within the endemic zones, the prevalence of the parasites varies from sporadic to high, but only a few countries can be regarded as being free of E. granulosus. It is difficult to determine the true incidence of cystic echinococcosis because of the slow rate of growth and variable clinical presentation. Most epidemiologic reports are based on hospital- and surgery-based surveys that greatly underestimate the actual rates of infection, especially in low socioeconomic groups with limited access to diagnosis and treatment. Since the mid-1980s, however, mass community-based surveys using portable ultrasound scanners have been conducted in many remote, rural areas of the world. The sensitivity and specificity of ultrasound have been shown to be superior to those of serology in prevalence surveys.8 These studies showed the real burden of disease, uncovering population infection rates of up to 6.6%. E. granulosus exists as a complex of species and strains that differ in a variety of criteria that may have an impact on the epidemiology, pathology, and control of cystic echinococcosis. To date, 10 distinct genotypes (G1 to G10) have been identified. Some distinct species have been identified (Echinococcus equinus, Echinococcus ortleppi). The great majority of E. granulosus isolates from human patients thus far have been of the sheep genotype (G1).  

CLINICAL MANIFESTATIONS

The presentation of human cystic echinococcosis is protean. Patients seek medical attention when a large cyst has some mechanical effect on organ function or rupture of a cyst causes acute hypersensitivity reactions. The cyst is often diagnosed incidentally during ultrasound examination, chest radiography, or body scanning performed for other clinical reasons. The liver is the most frequent location of echinococcal cysts, representing approximately 70% of cases.9 The lungs are the second most common location. However, cystic echinococcosis can occasionally occur in virtually any other organ. Common symptoms are upper abdominal discomfort and pain, poor appetite, and a mass in the abdomen. Physical findings are hepatomegaly, a palpable mass on the surface of the liver or other organs, and abdominal distention. Other manifestations include jaundice, colic-like pains, portal hypertension, ascites, and compression of the inferior vena cava. If cysts in the lung rupture into the bronchi, symptoms may include intense cough, a salty taste in the mouth, or vomiting of hydatid material and cystic membranes. Patients may present with a chest mass, chest pain, chronic cough, pneumothorax, eosinophilic pneumonitis, pleural effusion, parasitic lung embolism, hemoptysis, or biliptysis. Cysts in the heart can cause a cardiac mass, pericardial effusion, and embolism. Cysts in the breast must be differentiated from neoplasms.

CHAPTER 333 Cestodes  

Cysts located in the spine and in the brain can cause serious neurologic symptoms, including paralysis and seizures.  

DIAGNOSIS

The diagnosis of cystic echinococcosis is based on imaging methods and on serology, but serology has only a confirmatory role. Routine laboratory tests are nonspecific. Cyst rupture into the biliary tree may cause elevation of alkaline phosphatase, sometimes in association with hyperamylasemia and eosinophilia (up to 60%). Unless the cyst has ruptured, eosinophilia is low grade or absent.

Imaging

Modern imaging tools (ultrasound, CT, and, to a lesser extent, MRI) are central to the diagnosis and clinical management of cystic echinococcosis. Ultrasound is the procedure of choice for diagnosis of asymptomatic cystic echinococcosis.10 Ultrasound is also useful for longitudinal studies, such as monitoring the response of cysts to treatment and recording cyst growth rate. In 2003, the World Health Organization (WHO) Informal Working Group on Echinococcosis proposed a standardized ultrasound classification (Fig. 333-2). This classification defines six cyst stages that are assigned to three clinical groups. The active group comprises developing cysts, which may be unilocular (CE1) or multivesicular with daughter cysts (class CE2) and are usually found to be viable. The transitional group (class CE3) contains cysts that are usually starting to degenerate. There are two types of CE3: the “water lily sign” for floating membranes, which is now known as subclass CE3a; and predominantly solid cysts with daughter cysts, or subclass CE3b. This subdivision is based on their different response to percutaneous treatment (see later) and albendazole, which is generally good for CE3a and poor for CE3b. A study using nuclear magnetic resonance spectroscopy has found that CE3a and CE3b may have different metabolic characteristics. The inactive group (classes CE4 and CE5) exhibits involution and signs of solidification of cyst content with increasing degrees of calcification and is nearly always found to be nonviable. CT scanning has the advantage of inspecting any organ, detecting smaller cysts located outside the liver, locating cysts precisely, and sometimes differentiating parasitic from nonparasitic cysts. MRI may have some advantages over CT scanning in the evaluation of postsurgical residual lesions, recurrences, and selected extrahepatic infections, such as cardiac infections. Furthermore, a study has shown that MRI reproduces the ultrasound-defined features of cystic echinococcosis better than CT does. If ultrasound cannot be performed because of cyst location or patient-specific reasons, MRI with heavily T2-weighted series is preferable to CT. Plain radiographs are used for cysts in the lungs, bone, and muscle and for detection of calcified cysts.

Serology

Serologic tests are useful for confirmation of presumptive imaging diagnoses. However, many tests are available, and they are not standardized. Their sensitivity varies with the location of the cysts. Hepatic cysts are more likely to elicit an immune response than are pulmonary, brain, or splenic cysts. Serologic test results are usually positive when the endocyst is detached (CE3a) and in CL

Cystic lesion

CE1

CE2

Active

2121

active (CE2) and transitional (CE3b) stages. Serologic test results are generally negative in patients with inactive cysts (CE4 and CE5). Titers tend to slowly decrease when a cyst becomes inactive (CE4, CE5) and after radical surgery. Titers may remain positive after conservative surgery in which the antigen source (the germinal layer) is not completely removed. Antibody titers usually increase immediately after medical or percutaneous treatments because of the mobilization of the antigen following disruption of cyst integrity.

Other Diagnostic Procedures

Fine-needle aspiration of the cyst performed under ultrasonographic guidance, with a transhepatic approach, under anthelmintic coverage, is useful for differentiation of cystic echinococcosis, malignant neoplasms, abscesses, and nonparasitic cysts. The procedure must be carried out in the presence of an anesthesiologist ready to manage the rare but possible anaphylactic reaction.

TREATMENT  The appropriate treatment depends on factors of the individual patient, the characteristics of the cyst, the therapeutic resources available, and the physician’s preference.11 There are few randomized clinical trials evaluating treatment options, so a low level of evidence supports one therapeutic modality over another.12

Surgery

Surgery has long been the only option in the treatment of cystic echinococcosis. However, in the past two decades, medical treatment, percutaneous procedures, and a “watch and wait” approach have been successfully introduced and replaced surgery as the treatment of choice in selected cases. Surgery remains the main therapy in complicated cysts (i.e., those with rupture, biliary fistula, compression of vital structures, superinfection, or hemorrhage), cysts at high risk of rupture, or large cysts with many daughter vesicles that are not suitable for percutaneous treatments. Surgery can be performed as an open procedure, with either radical or conservative techniques, or laparoscopically, but there are controversies as to the safest and most effective technique and in which cases it should be applied. In all hepatic cases, perioperative albendazole prophylaxis, from 1 week before surgery until 4 weeks postoperatively, is recommended as a cautionary measure to minimize the risk of fluid spillage and consequent secondary echinococcosis from seeding of protoscolices in the abdominal cavity. Some authorities treat with praziquantel as well. Most experts recommend avoiding preoperative antiparasitic drugs for lung and intracranial disease.

Percutaneous Treatments

Percutaneous techniques provide an alternative to surgery and benzimidazole derivatives. These treatment modalities aim either to destroy the germinal layer with scolicidal agents or to evacuate the entire endocyst. The most popular method aimed at destroying the germinal layer is PAIR (puncture, aspiration, injection of a scolicidal agent, and reaspiration). Many modified catheterization techniques are used to evacuate the endocyst and are generally reserved for cysts that are difficult to drain or tend to relapse after PAIR, such as multivesiculated cysts or cysts with predominantly solid content and daughter cysts.

CE3

Transitional

CE4

CE5

Inactive

FIGURE 333-2.  WHO Informal Working Group on Echinococcosis standardized ultrasound classification of cystic echinococcosis. CL lesions are cystic lesions lacking a distinct wall and may have other diagnoses. CE1 lesions are cystic lesions with a visible wall that may demonstrate protoscolices (“hydatid sand”). CE2 lesions include internal septation. CE3 lesions may be detached from the wall (CE3a, top row) or have daughter cysts with internal thickening (CE3b, bottom row). CE4 lesions are heterogeneous lesions with degeneration. CE5 lesions show thick calcification.

A growing number of articles have reported its safety in treating abdominal, especially liver, echinococcal cysts. In a study on 5943 percutaneous punctures of echinococcal cysts, lethal anaphylaxis occurred in 0.03% of procedures, whereas reversible allergic reactions complicated 1.7% of procedures. Prophylactic administration of albendazole for at least 30 days after puncture is a cautionary measure that should always accompany PAIR. PAIR is generally successful at inducing permanent solidification in CE1 and CE3a cysts. A few reports with long-term follow-up indicate that multivesicular cysts (i.e., CE2 and CE3b) tend to relapse repeatedly after PAIR.

Chemotherapy

Albendazole (Chapter 323) is the antiparasitic drug of choice for cystic echinococcosis. It is administered orally at a dosage of 10 to 15 mg/kg/day; administration should be continuous without treatment interruptions. However, the optimal dosage and optimal duration of treatment with albendazole have not been formally assessed, and data from the small clinical trials generally fail to take into account the cyst’s characteristics. A recent systematic review on the effect of albendazole showed that the efficacy of the drug may have been overstated in previous retrospective, nonrandomized studies. Albendazole induces solidification in small and medium-sized CE1 and CE3a cysts, whereas it has generally little effect on giant (diameter >10 cm) CE1 and CE3a cysts. It has no effect on most cases of CE2 and CE3b cysts. Adverse effects of benzimidazoles (Chapter 323) include hepatotoxicity, leukopenia, thrombocytopenia, and alopecia. Increases in aminotransferases may be due to drug-related efficacy or real drug-related toxicity. Whereas teratogenic risks are theoretical, it is nonetheless good practice to avoid use during pregnancy when possible and to delay treatment until after delivery unless it is absolutely necessary.

Watch and Wait

Inactive liver cysts that are free of complications, such as compression on neighboring organs, are increasingly monitored without being treated.13 Prospective studies need to be carried out to confirm the safety of this option.

Follow-up

Follow-up is crucial to evaluate the efficacy of treatment. Long-term follow-up, generally more than 5 years, is required to evaluate local recurrences, which have been reported up to 10 years after apparently successful treatment. When the combination of imaging and serology is inconclusive, fineneedle aspiration should be performed to ascertain the viability of the cyst contents.



Alveolar Hydatid Disease (Echinococcus multilocularis)

Alveolar hydatid disease is caused by the tissue forms of Echinococcus multilocularis. In tissues, typically the liver, E. multilocularis grows as a budding mass rather than as a large cystic lesion. The tissues resemble lung tissues, hence the name “alveolar.” The normal definitive hosts are canines, including wolves and foxes. The normal intermediate hosts are rodents. Humans are accidentally infected by contact with soil containing the ova.  

EPIDEMIOLOGY

E. multilocularis is endemic in arctic and alpine areas of the Northern Hemisphere. It is highly endemic in western China, Tibet, and central Asia. In recent years, E. multilocularis has emerged as an important problem in Alpine areas of central Europe and adjacent forested areas.14  

CLINICAL MANIFESTATIONS AND DIAGNOSIS

Human E. multilocularis infection almost invariably involves the liver, in which it is manifested as a tumor-like mass that gradually expands during decades. The main symptoms are liver discomfort and swelling. Diagnosis is by demonstration of a characteristic mass on imaging studies, with the etiology confirmed by serologic tests.15

TREATMENT AND PROGNOSIS  Surgery remains the mainstay of treatment of E. multilocularis. When feasible, all infected tissues should be removed. Apparently curative therapy should be followed by a 2-year course of albendazole to decrease the risk of relapse. In some cases, resection is feasible only when it is accompanied by liver transplantation. In cases that are not amenable to surgical resection, prolonged courses of albendazole can suppress growth of the lesion. After treatment with benzimidazoles, mortality is similar to that of the age- and-sex-matched general population.



Other Larval Cestode Infections

Sparganosis is caused by infection with the larval (plerocercoid) stage of Spirometra mansonoides. Infection is acquired by ingestion or application of infected meat (frogs, birds, fish) or exposure of skin to infected flesh (e.g., poultices of infected tissues). After infection, plerocercoids develop in the tissues, typically presenting as subcutaneous or central nervous system nodules or occasionally larva migrans symptoms. Treatment usually involves removal of the nodule. Coenurosis is a rare larval cestode infection caused by human infection with the larval stage of the dog tapeworms Taenia multiceps and Taenia serialis. In the tissue, the larva forms a cystic lesion containing multiple scolices (the coenurus). The cystic lesion is usually single and most frequently identified in brain, eye, or soft tissues. Treatment usually involves removal. Echinococcus oligarthrus and Echinococcus vogeli have been associated with polycystic hydatid disease in northern South America. Taenia crassiceps has been identified in the eye and in tissues of compromised hosts. Hymenolepislike organisms have also been identified in tissues of AIDS patients.

  Grade A References A1. Baird A, Wiebe S, Zunt JR, et al. Evidence-based guideline: treatment of parenchymal neurocysticercosis: report of the Guideline Development Subcommittee of the American Academy of Neurology. Neurology. 2013;80:1424-1429. A2. Garcia HH, Gonzales I, Lescano AG, et al. Efficacy of combined antiparasitic therapy with praziquantel and albendazole for neurocysticercosis: a double-blind, randomised controlled trial. Lancet Infect Dis. 2014;14:687-695. A3. Zhao BC, Jiang HY, Ma WY, et al. Albendazole and corticosteroids for the treatment of solitary cysticercal granuloma: a network meta-analysis. PLoS Negl Trop Dis. 2016;10:1-15. A4. Garcia HH, Gonzales I, Lescano AG, et al. Enhanced steroid dosing reduces seizures during antiparasitic treatment for cysticercosis and early after. Epilepsia. 2014;55:1452-1459. A5. Garcia HH, Gonzalez AE, Tsang VC, et al. Elimination of Taenia solium transmission in northern Peru. N Engl J Med. 2016;374:2335-2344.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 333 Cestodes  

GENERAL REFERENCES 1. Webb C, Cabada MM. Intestinal cestodes. Curr Opin Infect Dis. 2017;30:504-510. 2. Coral-Almeida M, Gabriël S, Abatih EN, et al. Taenia solium human cysticercosis: a systematic review of sero-epidemiological data from endemic zones around the world. PLoS Negl Trop Dis. 2015;9: 1-20. 3. Moyano LM, O’Neal S, Ayvar V, et al. High prevalence of asymptomatic neurocysticercosis in an endemic rural community in Peru. PLoS Negl Trop Dis. 2016;10:1-11. 4. Gripper LB, Welburn SC. Neurocysticercosis infection and disease –a review. Acta Trop. 2017;166: 218-224. 5. Del Brutto OH, Nash TE, White AC Jr, et al. Revised diagnostic criteria for neurocysticercosis. J Neurol Sci. 2017;372:202-210. 6. Coyle CM. Neurocysticerosis: an individualized approach. Infect Dis Clin North Am. 2019;33: 153-168. 7. White AC Jr, Coyle CE, Rajshekhar V, et al. Diagnosis and treatment of neurocysticercosis: practice guidelines by the Infectious Diseases Society (IDSA) of America and the American Society of Tropical Medicine and Hygiene (ASTMH). Clin Infect Dis. 2018;66:1159-1163. 8. Tamarozzi F, Akhan O, Cretu CM, et al. Prevalence of abdominal cystic echinococcosis in rural Bulgaria, Romania, and Turkey: a cross-sectional, ultrasound-based, population study from the HERACLES project. Lancet Infect Dis. 2018;18:769-778.

2122.e1

9. Mihmanli M, Idiz UO, Kaya C, et al. Current status of diagnosis and treatment of hepatic echinococcosis. World J Hepatal. 2016;8:1169-1181. 10. Barosa R, Pinto J, Caldeira A, Pereira E. Modern role of clinical ultrasound in liver abscess and echinococcosis. J Med Ultrason. 2017;44:239-245. 11. Kern P, Menezes da Silva A, Akhan O, et al. The echinococcoses: diagnosis, clinical management and burden of disease. Adv Parasitol. 2017;96:259-369. 12. Nabarro LE, Amin Z, Chiodini PL. Current management of cystic echinococcosis: a survey of specialist practice. Clin Infect Dis. 2015;60:721-728. 13. Stojkovic M, Rosenberger KD, Steudle F, et al. Watch and wait management of inactive cystic echinococcosis—does the path to inactivity matter—analysis of a prospective patient cohort. PLoS Negl Trop Dis. 2016;10:1-10. 14. Gottstein B, Stojkovic M, Vuitton DA, et al. Threat of alveolar echinococcosis to public health—a challenge for Europe. Trends Parasitol. 2015;31:407-412. 15. Bulakci M, Kartal MG, Yilmaz S, et al. Multimodality imaging in diagnosis and management of alveolar echinococcosis: an update. Diagn Interv Radiol. 2016;22:247-256.

2122.e2

CHAPTER 333 Cestodes  

REVIEW QUESTIONS 1. Active cysts in the WHO Informal Working Group on Echinococcosis classification are: A . CE1 B. CE1, CE2 C. CE3a, CE3b D. CE4 E. CE4, CE5 Answer: B  CE1 (unilocular) and CE2 (cysts with daughter cysts) are categorized as viable or active in the WHO Informal Working Group on Echinococcosis. When aspirated, viable protoscolices are found in the fluid. 2. The risk of lethal anaphylactic shock after an echinococcal cyst puncture is: A . 2% B. 3% C. 4% D. 5% E. Less than 1% Answer: E  In an analysis of the literature that included 5943 percutaneous puncture procedures of 5517 hepatic and nonhepatic echinococcal cysts, two cases of lethal anaphylaxis and 99 reversible anaphylactic reactions were reported. Lethal anaphylaxis occurred in 0.03% of percutaneous puncture procedures, corresponding to 0.04% of treated cysts, whereas reversible allergic reactions complicated 1.7% of percutaneous punctures, corresponding to 1.8% of treated echinococcal cysts. Lethal anaphylaxis related to percutaneous treatment of cystic echinococcosis is an extremely rare event and is observed no more frequently than drug-related anaphylactic side effects are. 3. The best imaging technique for staging of cystic echinococcosis (CE) cysts of the liver is: A . Computed tomography (CT) B. Magnetic resonance imaging (MRI) C. Ultrasound D. Positron emission tomography (PET) scan E. Scintigraphy Answer: C  Imaging plays the key role in diagnosis and staging of CE. The description of CE-specific imaging features and the WHO CE cyst classification are based on ultrasound. The reproducibility of the ultrasound-defined features of CE cysts is variable in MRI and CT. This is of particular importance for cysts that are not accessible by ultrasound and because of the increasing availability and overuse of CT and MRI. A retrospective study was conducted in patients with abdominal CE cysts seen on ultrasound who had additional CT or MRI scans performed; it showed that MRI reproduces the ultrasound-defined features of CE better than CT does. If ultrasound cannot be performed because of cyst location or patientspecific reasons, MRI with heavily T2-weighted series is preferable to CT.

4. In a patient presenting with seizures and calcified cysticercosis with surrounding edema on MRI, the most appropriate therapy is: A . Antiepileptic drugs alone B. Antiepileptic drugs and albendazole C. Antiepileptic drugs and prednisone D. Antiepileptic drugs, albendazole, and prednisone E. Placement of a ventriculoperitoneal shunt Answer: A  Patients with calcified lesions no longer have viable cysticerci and thus do not benefit from antiparasitic drugs. Prednisone may transiently decrease edema, but there is no evidence of clinical benefit, and steroid withdrawal has been associated with flares of disease. 5. In a patient presenting with multiple subarachnoid cysticerci complicated by basilar meningitis, which of the following approaches have been associated with the best outcomes? A . Treatment with a 2-week course of albendazole and dexamethasone B. Removal of the cysticerci by open craniotomy C. Treatment with prolonged courses of albendazole, dexamethasone, and methotrexate D. Neuroendoscopic removal of intraventricular cysticerci E. Cerebrospinal fluid diversion alone for hydrocephalus Answer: C  Subarachnoid neurocysticercosis responds poorly to the doses of albendazole used for parenchymal infection. Use of cerebrospinal fluid diversion alone carries a high case-fatality rate. The optimal approach includes prolonged courses of antiparasitic drugs along with anti-inflammatory treatment. Higher doses of albendazole, combinations of praziquantel and albendazole, and endoscopic debulking (removal of cysticerci in the basilar cistern, not just the ventricles) have also been reported as helpful.

2122

CHAPTER 334  Trematode Infections  

334  TREMATODE INFECTIONS EDGAR M. CARVALHO, ALDO A.M. LIMA, LUIS A. MARCOS, AND EDUARDO GOTUZZO

  SCHISTOSOMIASIS  

DEFINITION

Schistosomiasis is one of the most important parasitic diseases of humans and is a global public health problem in the developing world. The disease is caused by trematodes of the genus Schistosoma, and it is estimated that 250 million people are infected and 779 million are at high risk for infection. Mortality from schistosomiasis is estimated at 13,000 deaths per year and the burden of disease at 3.3 million disability-adjusted life-years lost per year.1

The Pathogen

Six major species of Schistosoma affect humans: S. mansoni, S. haematobium, S. japonicum, S. intercalatum, S. mekongi, and S. guineensis, and the sequencing of the S. mansoni, S. haematobium, and S. japonicum genome has been determined. These species differ biologically from one another and in their geographic distribution and the type of disease that they produce. The schistosomes are digenetic parasitic trematodes. Although they are morphologically distinct, the species of Schistosoma that infect humans share some common features. The large male (0.6 to 2.2 cm × 2 to 4 mm) has a ventral gynecophoric canal in which the female (1.2 to 2.6 cm × 1 to 2 mm) is held during copulation.

CHAPTER 334  Trematode Infections  

ABSTRACT

The main Schistosoma species affecting humans are S. mansoni, S. haematobium, S. japonicum, S. intercalatum, S. mekongi, and S. guineensis. Schistosomiasis is endemic in the African continent, Brazil, Suriname, Venezuela, the Middle East, China, Indonesia, Philippines, and Thailand. The schistosomes are transmitted to humans after contact with contaminated fresh water. After the cercariae penetration in the skin parasites migrate to the lungs and mature to adult worms in the urinary tract (S. haematobium) or in mesenteric vessels (other species). The release of eggs leads to granuloma formation and fibrosis. The main symptoms are hematuria and dysuria (S. haematobium). In other schistosoma infections liver fibrosis may result in hepatosplenic schistosomiasis due to portal hypertension and hematemesis from bleeding esophageal or gastric varices. Acute infection is characterized by fever, malaise, diarrhea, weight loss, cough, and dyspnea with marked eosinophilia. Diagnosis of acute infection is performed by serologic tests. In chronic infection the diagnosis is performed by detection of eggs in the stool or in the urine (S. haematobium). As eggs may not be found in light infections, the diagnosis can be made by quantification of circulating cathodic antigen in the urine, antibody detection, or rectal biopsy. Chemotherapy is the major method for cure and control of schistosomiasis. Praziquantel (40 mg/kg) in a single dose is the recommended therapy for S. mansoni, S. haematobium, S. intercalatum, and S. guineensis and 60 mg/kg for S. japonicum and S. mekongi. The cure rate with praziquantel ranges from 60 to 85%.

KEYWORDS

schistosomiasis Schistosoma human schistosomiasis Schistosoma mansoni Schistosoma haematobium Schistosoma japonicum

2122.e3

CHAPTER 334  Trematode Infections  

Adult worms in human vesical or mesenteric veins

Schistosomula first parasite stage in humans

Egg in urine or stools

Cercaria free-living in fresh water (infective stage)

Miracidium freeliving in fresh water

Freshwater snail asexual multiplication FIGURE 334-1.  Schistosome life cycle.



EPIDEMIOLOGY

Schistosomiasis occurs mainly in rural agricultural and periurban areas. Programs for schistosomiasis control are advancing and helping in disease reduction.2 However, schistosomiasis is still endemic in 78 countries. S. mansoni is found in the Arabian Peninsula, Egypt, Libya, Sudan, most countries in subSaharan Africa, Brazil, Suriname, and Venezuela. S. haematobium is endemic in the Middle East and most of the African continent, including the islands of Madagascar and Mauritius. S. japonicum is endemic in China, Indonesia, and the Philippines; it has also been reported from Thailand. S. intercalatum has been reported from 10 countries in Africa. S. mekongi is found in Cambodia and Laos. The endemicity of schistosomiasis depends on the urban disposal of urine (S. haematobium) and feces (S. mansoni, S. japonicum, S. intercalatum, S. mekongi, S. guineensis), the presence of suitable snail hosts, and human exposure to cercariae. The freshwater snail intermediate hosts are Biomphalaria species in Africa and Biomphalaria glabrata (Australorbis) and Tropicarbis in South America and the West Indies. In some cases, the endemicity of schistosomiasis may be maintained by animal reservoirs. Such is the case with S. japonicum, which infects dogs and cows.

Etiology and Life Cycle

Adult worms live in the mesenteric veins (S. mansoni, S. japonicum, S. mekongi, S. intercalatum, and S. guineensis) or in the venous plexus around the lower ends of the ureters and the urinary bladder (S. haematobium) (Fig. 334-1). In these sites, they start their sexual reproduction by releasing eggs. Once deposited in the host, eggs may stay in the mesenteric vein, be trapped in the intestines, escape to the intestinal lumen, and migrate by portal blood to the liver (S. mansoni, S. japonicum). Eggs of S. haematobium may be trapped in the intestines and bladder and may escape to the intestinal or bladder lumen. After being excreted with feces or urine into fresh water, the eggs hatch and release ciliated motile miracidia that penetrate into the snail intermediate host. Following asexual multiplication in the snail, the development of cercariae, the infective forms for humans, takes 4 to 7 weeks. After leaving the snails, the cercariae can survive in fresh water for almost 72 hours. When penetration of the skin in the human host occurs, the cercariae lose their tails and change into schistosomula. Schistosomula migrate to the lungs and, in about 6 weeks, mature to adult worms and descend to their final habitat. Viable eggs can be seen in excretions (i.e., stool or urine) 5 to 9 weeks after cercarial penetration. The lifespan of the worms ranges from 5 to 10 years.  

PATHOBIOLOGY

The pathogenesis of acute human schistosomiasis is mainly related to egg deposition, liberation of antigens of adult worms and eggs, and an exaggerated inflammatory response characterized by high levels of pro-inflammatory cytokines such as interleukin-1 (IL-1), IL-6, and tumor necrosis factor (TNF). In chronic schistosomiasis, tissue injury is mediated by egg-induced granulomas and the subsequent appearance of fibrosis. In S. haematobium, infected eggs induce granuloma formation in the urinary tract. Because the habitat of S.

2123

mansoni, S. japonicum, S. mekongi, S. guineensis, and S. intercalatum worms is the mesenteric blood vessels, the intestines are involved primarily, and egg embolism results in secondary involvement of the liver. Host genetics, immunologic response, and parasite load measured by egg count in the stool are associated with a greater chance for hepatosplenic involvement. IL-4, IL-13, and IL-17 have been associated with granuloma formation and liver fibrosis. Enzymes and antigens released from eggs sensitize host lymphocytes, which migrate to areas of egg deposition and recruit other cell types, such as macrophages, eosinophils, and fibroblasts. The size of these granulomas and the resulting fibrosis lead to most of the characteristic chronic fibro-obstructive lesions in schistosomiasis. In the liver, the granulomas result in perisinusoidal obstruction of portal blood flow, portal hypertension, splenomegaly, esophageal varices, and portosystemic collateral circulation. Liver cell perfusion is not reduced; consequently, liver function test results remain normal for a long time. In schistosome-infected populations, the intensity of infection increases during the first 2 decades of life as children accumulate worms, and then declines. Although there is a decrease in exposure with age, the lower intensities of infection in older individuals are due in part to acquired resistance. There is an association between parasite-specific IgE, eosinophila, and resistance to reinfection. In the S. haematobium–infected population, IgE increases progressively with age, and IgE antibodies directed against adult worm antigens are associated with subsequent low intensities of reinfection. Similar associations between high IgE levels or a high IgE/IgG4 ratio and resistance to reinfection have been found among Brazilian and Kenyan subjects exposed to S. mansoni. In contrast, susceptibility to reinfection has been associated with IgG4, which may serve as blocking antibody, inhibiting the action of IgE. A major codominant gene, called SM1, in the human host’s genome controls the intensity of infection by S. mansoni. Modulation of the immune response is a characteristic of chronic schistosomiasis, and S. mansoni antigens downmodulate the immune response by increasing their production of regulatory cytokines such as IL-10 and the frequency of regulatory T cells.3 S. mansoni infection attenuates the clinical manifestations of type 1 diabetes and experimental autoimmune encephalitis in mice. In humans, S. mansoni attenuates the clinical manifestations of asthma and is inversely associated with the development of myelopathy related to human T-cell leukemia virus type 1 (HTLV-1) infection. S. mansoni infection may impair immunologic response to vaccines and change manifestations of other infectious diseases.  

CLINICAL MANIFESTATIONS

Clinical manifestations of schistosomiasis are divided into schistosome dermatitis, acute schistosomiasis, and chronic schistosomiasis.4 Schistosome dermatitis, or swimmer’s itch, is seen mainly when avian cercariae penetrate the skin and are destroyed. Schistosome dermatitis is a sensitization phenomenon occurring in previously exposed persons. The cercariae evoke an acute inflammatory response with edema, early infiltration of neutrophils and lymphocytes, and later invasion of eosinophils. A pruritic papular rash occurs within 24 hours after the penetration of cercariae and reaches maximal intensity in 2 to 3 days. Acute schistosomiasis occurs usually 20 to 50 days after primary exposure. Though asymptomatic in endemic areas, acute schistosomiasis is becoming a frequent and major clinical problem in nonimmune individuals from urban regions who are exposed for the first time to a heavy infection in an endemic area. The clinical syndrome (i.e., fever, chills, liver and spleen enlargement, and marked eosinophilia) originally described for S. japonicum infection, and still common for this species, is increasingly being diagnosed in Brazil and Africa in individuals with S. mansoni infection. Malaise, diarrhea, weight loss, cough, dyspnea, chest pain, restrictive respiratory insufficiency, and pericarditis are also documented in this phase. High levels of TNF and circulating immune complexes are found in the patient’s blood in this phase of illness, and they coincide with the development of abdominal pain, diarrhea, and weight loss and with respiratory and pericardial manifestations, respectively. Acute disease is not observed in individuals living in endemic areas for schistosomiasis. In chronic schistosomiasis, abdominal pain, irregular bowel movements, and blood in the stool are the main symptoms of intestinal involvement. Colonic polyposis secondary to schistosomiasis may occur, especially in Egypt. Hepatosplenic involvement is the most important cause of morbidity with S. mansoni and S. japonicum infection. Patients may remain asymptomatic until the manifestation of hepatic fibrosis and portal hypertension develops. Hepatic fibrosis is caused by a granulomatous reaction to Schistosoma eggs that have been carried to the liver. Although the intensity of infection is clearly

2124

CHAPTER 334  Trematode Infections  

linked directly to liver damage, the immune response, extent of collagen deposition, and genetic factors can potentiate disease in some individuals with only moderate infections or inhibit disease in others with heavy infections. Hematemesis from bleeding esophageal or gastric varices may occur. Anemia and decreasing levels of serum albumin are observed. A few patients have severe hepatosplenic disease with decompensated liver disease. Jaundice, ascites, and liver failure are then observed. Concomitant infection by Salmonella species and, to a lesser extent, other gram-negative bacteria with S. mansoni or S. haematobium leads to a picture of prolonged fever, hepatosplenomegaly, and mild leukocytosis with eosinophilia. Coinfection with S. mansoni and hepatitis B or C virus exacerbates liver pathology. Glomerulonephritis and hypersplenism are other complications associated with hepatosplenic schistosomiasis. Pulmonary hypertension is increasingly recognized with the use of more advanced diagnostic technology. Pulmonary hypertension has been detected in 10.7% of patients with liver fibrosis. In countries endemic for S. mansoni up to 30% of all cases of pulmonary hypertension are due to schistosomiasis. Cerebral schistosomiasis is observed in patients with S. japonicum infection. It may occur as early as 6 weeks after infection, and its most common manifestation is focal seizures. Signs and symptoms of generalized encephalitis may occasionally be found. In S. mansoni infection, neurologic involvement is mainly characterized by transverse myelitis. An association between S. mansoni or S. haematobium and human immunodeficiency virus (HIV) infection has been documented in areas where both infectious agents are found. Although the immune response of patients with schistosomiasis is altered by HIV, it has not been reported to be associated with any clear changes in the degree of infection or severity of manifestations of schistosomiasis. In S. haematobium infection, the main organ system involved is the urinary tract. The acute granulomatous response to parasite eggs in the early stages causes urinary tract disease, such as urethral ulceration and bladder polyposis. In chronic disease, usually in older patients, granulomas at the lower ends of the ureters obstruct urinary flow and may cause hydroureter and hydronephrosis. Bladder fibrosis and calcification are also seen in this phase. Up to 70% of infected individuals have hematuria, dysuria, or urinary frequency. Radiologic findings include hydronephrosis, hydroureter, ureteral strictures, dilation or distortion, ureteral calcifications, ureterolithiasis, calcified bladder, polyps, reduction in bladder capacity, irregular contraction of the bladder wall, or a dilated bladder because of bladder neck fibrosis. An increased incidence of squamous cell carcinoma of the bladder (Chapter 187) has been reported in endemic areas of S. haematobium infection, but the mechanism of carcinogenesis is unknown. S. haematobium eggs have occasionally been found in the lungs, with subsequent focal pulmonary arteritis and pulmonary hypertension.  

DIAGNOSIS

A definitive diagnosis of schistosomiasis can be made only by finding schistosome eggs in feces, urine, or a biopsy specimen, usually from the rectum (Table 334-1).5,6 A history of contact with contaminated water and appropriate clinical manifestations are important steps in establishing the diagnosis. Because schistosome eggs may be few, concentration by sedimentation should be

performed. All eggs from feces, urine, or tissues should be examined under high power to determine their viability by visualizing the activity of cilia of the excretory flame cells of the enclosed miracidium. Dead eggs may persist for a long time after successful therapy or natural death of the worms. The presence of only dead eggs does not necessarily require treatment. Because the intensity of infection is associated with morbidity, quantitative techniques are recommended. For S. mansoni and S. japonicum, the Kato-Katz thick smear method is used. In patients with chronic S. mansoni and S. japonicum infection and liver disease, the diagnosis is sometimes made by documentation of eggs in liver specimens. Ultrasonography allows determination of the degree of liver fibrosis. S. mekongi and S. intercalatum infection is diagnosed by examining the stool for eggs. Urine examination for S. haematobium eggs can be performed by direct or concentration methods. Samples should be obtained at midday, when excretion of eggs is maximal. After S. haematobium infection is diagnosed, assessment of urinary tract pathology by ultrasonography is recommended. Because of an increased incidence of carcinoma of the bladder, cancer surveillance should be performed in patients with S. haematobium infection (Chapter 187). Molecular and immunodiagnostic assays are helpful in the diagnosis of acute infection and in diseases associated with light infection as in schistosomal myeloradiculopathy associated with S. mansoni and S. haematobium infection. Serologic assays have proven useful clinically for diagnosis by detection of antibodies to Schistosoma mansoni antigens. However, serologic tests cannot distinguish between active and past exposure. Quantification of circulating cathodic antigen (CCA) in urine has a higher sensitivity than the Kato-Katz method and has also been used to monitor the efficacy of antischistosome chemotherapy.7 A significant decrease in antigen levels or negativity of the test is observed as early as 10 days after therapy. Molecular techniques for schistosoma DNA detection in fecal, urine, and blood specimens also increase sensitivity.8

TREATMENT AND PREVENTION  Chemotherapy is the principal modality for prophylaxis, control, and cure of schistosomiasis. Several compounds are in use, including metrifonate, oxamniquine, praziquantel, and artemisinin derivatives (artesunate and artemether) for prophylaxis, control, and treatment of schistosomiasis. Praziquantel, a pyrazinoisoquinoline derivative, is the drug of choice for the treatment of schistosomiasis for four reasons: high efficacy against all schistosome species and against cestodes, lack of serious short-term and long-term side effects, administration as a single oral dose, and competitive cost. A recent review confirmed that a single dose of praziquantel (40 mg/kg) is an effective treatment for S. mansoni infection. A1  Lower doses may be less effective and there is no additional benefit for higher doses. Oxamniquine (40 mg/kg) is also effective, and based on limited evidence to date it is uncertain which intervention is more effective. Further research will help optimize the doses for young children. A meta-analysis also confirmed that artemisinin derivatives, different from oxamniquine, used in combination with praziquantel increase the cure rates in schistosomiasis treatment, but not artemisinin derivatives or oxamniquine alone. A2 

TABLE 334-1 DIAGNOSIS OF SCHISTOSOMIASIS SCHISTOSOME

EGGS

DIAGNOSIS

S. haematobium

Mainly found in urine but may be found in stools or rectal biopsy specimens Eggs: 143 × 50 µm; spindle shaped: rounded anterior, conical posterior, tapering to a terminal delicate spine

Obtain urine sample at midday (when eggs are excreted); more than one sample may be needed Examine urine directly or by filtering 10 mL of urine through a Nuclepore membrane Urine-circulating antigen test for diagnosis of light infection and serologic test to diagnose of early or light infection Rectal biopsy in suspected cases with negative urine for egg and for circulating cathodic antigen

S. mansoni

Found in stool eggs: 155 × 66 µm; oval with lateral, long spine

Examine stool for eggs Use the Kato-Katz thick smear method for quantification purposes Urine-circulating cathodic antigen test, serologic testing, or rectal biopsy to diagnose stoolnegative cases, particularly in lightly infected patients

S. japonicum

Found in stool Eggs: 89 × 67 µm; oval or rounded with a lateral, short, sometimes curved spine

Examine stool for eggs Kato-Katz thick smear (for quantitative assessment) Urine-circulating cathodic antigen test, serologic test or rectal biopsy for those with light infections, especially with less common manifestations (e.g., cerebral schistosomiasis)

S. mekongi

Found in stool Eggs: 60 × 32 µm; smaller than eggs of S. japonicum

Examine stool for eggs

S. intercalatum

Found in stool Eggs: 180 × 65 µm; terminal spine

Examine stool for eggs

2125

CHAPTER 334  Trematode Infections  

TABLE 334-2 CLINICAL MANIFESTATIONS, DIAGNOSIS, AND TREATMENT OF LIVER, INTESTINAL, AND LUNG FLUKE INFECTIONS IN HUMANS FLUKE, NUMBER OF PEOPLE INFECTED, AND GEOGRAPHIC DISTRIBUTION

CLINICAL MANIFESTATIONS OF ACUTE AND CHRONIC PHASES

DIAGNOSIS*

TREATMENT†

LIVER FLUKES Fasciola spp 17 million Cosmopolitan

Acute Hepatomegaly, eosinophilia, fever, abdominal pain, metastaticlike lesions on liver CT Chronic § Biliary obstruction, gallstones, fibrosis, cholangitis

Stools are negative Serology (Fas2 ELISA)

Opisthorchis spp 11.2 million Asia



Serology Eggs in stool

PZQ

Clonorchis spp 35 million Asia, Eastern Europe



Fever, rash, malaise, and RUQ abdominal discomfort Choledocholithiasis, cholangitis, cholecystitis, liver abscess, and possible cholangiocarcinoma

Serology Eggs in stool

PZQ

Small bowel inflammation, ulceration, mucus secretion, protein-losing enteropathy, malabsorption

Eggs in stool

PZQ



Abscess serology Eggs in sputum or stool

PZQ or TCZ



Fever, malaise, arthralgia, lymphadenopathy, and rash Jaundice, cholangitis, cholangiocarcinoma

§

§

TCZ

Eggs in stool Fas2 ELISA

INTESTINAL FLUKES Fasciolopsis buski and others¶ 50 million Asia and North Africa LUNG FLUKE Paragonimus spp 23 million Asia, Americas, Africa

Abdominal pain, pleuritic pain, cough, eosinophilia Hemoptysis, cough, chronic pleural effusions, pulmonary cysts

§

*For egg morphology and size, refer to text in the Diagnosis section. † For drug regimens, refer to text in the Treatment section. ‡ Acute. § Chronic. For chronic infection, a sedimentation technique is preferred (suggested technique: rapid sedimentation technique or Kato-Katz technique). ¶ Diagnosis and treatment also apply for other intestinal flukes. CT = computed tomography; ELISA = enzyme-linked immunosorbent assay; PZQ = praziquantel; RUQ = right upper quadrant; TCZ = triclabendazole.

The standard recommended treatment consists of a single dose of praziquantel, 40 mg/kg, for S. mansoni, S. haematobium, and S. intercalatum infection. In S. japonicum infection, a total dose of 60 mg/kg is recommended, split into two or three doses in a single day. Although no significant difference was found in the overall cure rates between single-dose (40 mg/kg) and double-treatment (40 mg/kg with 2-week interval) regimens of praziquantel for S. haematobium, the effect of double treatment resulted in significant reduction in infection intensity and microhematuria, which may have an impact in reducing morbidity. S. mekongi may require two treatments at 60 mg/kg body weight. With these dosages of praziquantel, recorded cure rates are 75 to 85% for S. haematobium, 63 to 85% for S. mansoni, 80 to 90% for S. japonicum, 89% for S. intercalatum, and 60 to 80% for double infections with S. mansoni and S. haematobium. A decrease in the efficacy of praziquantel has been observed in patients coinfected with HTLV-1. Praziquantel is well tolerated and effective in patients of all ages and for different clinical forms of schistosomiasis, including advanced hepatosplenic cases (S. mansoni), cerebral schistosomiasis (S. japonicum), and neurologic syndromes (S. mansoni and S. haematobium), possibly in association with corticosteroids. However, praziquantel has a low prophylactic effect, which reduces its efficacy in areas of high transmission. There have been several reports of persistent schistosome egg shedding after treatment, raising concern about the emergence of drug resistance. A systematic review and meta-analysis showed that either artesunate or artemether has confirmed prophylactic effect across different trials performed in China. Further studies will be necessary to examine the combination of antischistosome chemotherapy for S. haematobium infection in which repeated standard treatment fails to clear the infection. The most common adverse events observed with praziquantel or oxamniquine are related to the gastrointestinal tract: abdominal pain or discomfort, nausea, vomiting, anorexia, and diarrhea. These symptoms can be observed in up to 50% of patients but are usually well tolerated. Other side effects are related to the central nervous system (e.g., headache, dizziness, drowsiness) and the skin (e.g., pruritus, eruptions) or may be nonspecific (e.g., fever, fatigue). In general, the cumulative experience from many studies allows the conclusion that praziquantel is an extremely well-tolerated drug that requires minimal medical supervision and is therefore particularly suitable for mass chemotherapy programs. Mass drug administration with praziquantel has been used as the mainstay of programs for the control and prevention of schistosomiasis morbidity. However, this approach alone is not sufficient to eliminate or interrupt transmission of this disease. The development of anti-Schistosoma vaccines is necessary.9 Additional measures such as provision of clean water, use of molluscicides, adequate sanitation, and improvement of socioeconomic conditions should also be implemented to control the disease.10,11

  LIVER FLUKES

The geographic distribution, clinical manifestations, diagnosis, and treatment of liver, intestinal, and lung fluke infections are summarized in Table 334-2.  

Fascioliasis

Fascioliasis is a zoonosis caused by Fasciola hepatica (adult: 30 × 13 mm) or F. gigantica (adult: 75 × 20 mm). The most common natural hosts are cattle, sheep, and goats.  

EPIDEMIOLOGY

The infection is distributed globally. An estimated 2.6 million people are infected by this parasite, mainly from South America, Africa, and Asia. The highest prevalence rates (>60%) have been reported in Peru and Bolivia.  

PATHOBIOLOGY

The life cycle begins when the parasite’s eggs in stool are deposited in water; miracidia appear, develop, and hatch in 9 to 14 days and invade many species of freshwater snails (Lymnaea spp), in which they multiply as sporozoites, rediae, and cercariae during a period of 4 to 7 weeks. They then leave as freeswimming cercaria that subsequently attach to watercress, water lettuce, alfalfa, mint, parsley, or khat. The main source of infection is consumption of raw vegetables or water contaminated with metacercariae. Women have a higher incidence of the disease, with more severe infections and complications than seen in men.  

CLINICAL MANIFESTATIONS

After consumption of contaminated vegetables, the larvae excyst in the duodenum and then migrate through the bowel wall to the liver through the peritoneal cavity. In 4 weeks they reach the liver, penetrate the Glisson capsule, and cause inflammation and pain. An acute diarrhea of 2 to 5 days’ duration may occur before liver invasion. During their migration through the liver, the ongoing inflammatory process is accompanied by fever, pain, and eosinophilia. In a few cases, intense hemorrhage manifested as subcapsular liver hematoma may develop. Computed tomography (CT), magnetic resonance imaging, or ultrasonography can detect these initial lesions. Furthermore, migration of the parasite leaves a trail or track that can be observed in histologic sections or by imaging (CT). The flukes sometimes die and leave cavities filled with necrotic debris that are eventually replaced by scar tissue and then become

CHAPTER 334  Trematode Infections  

The effects of praziquantel on schistosomes can be summarized under three headings: muscular contraction, tegumental damage (i.e., vacuolization and blebbing), and metabolic alterations (i.e., decreased glucose uptake, lactate excretion, and glycogen content). Schistosome calcium ion (Ca2+) channels are so far the only molecular targets of praziquantel action identified. The drug may also interfere with adenosine uptake, which is relevant because schistosomes are unable to synthesize purines de novo. Several drugs (e.g., phenytoin, rifampin, azole antifungals) decrease plasma concentrations of praziquantel because they inhibit the cytochrome P450 pathway. Chloroquine and cimetidine also decrease praziquantel bioavailability. Praziquantel’s activity also depends on the patient’s immune system. Several studies have reported that praziquantel induces the exposure of worm surface antigens that may function as targets for immune responses.

2125.e1

2126

CHAPTER 334  Trematode Infections  

calcified. After 3 to 5 months of migration in the liver, the juvenile larvae finally reach the bile ducts. During this invasive, migratory, or acute phase the clinical manifestations are prolonged fever, hepatomegaly, abdominal pain, and eosinophilia. Multiple hypodense lesions are seen on CT, similar to metastases, but they change in position, attenuation, and shape in time because the parasites are still migrating. Acute fascioliasis is clinically similar to acute cholecystitis but with the addition of significant eosinophilia. It can occur in travelers with acute subcapsular hematoma or “metastatic-like lesions” seen on CT of the liver. Hyperbilirubinemia is notably absent in this phase. Other manifestations are anorexia, weight loss, nausea, vomiting, cough, diarrhea, urticaria, lymphadenopathy, and arthralgias. Occasionally, the juvenile larvae reach other ectopic or extrahepatic locations, such as subcutaneous tissue, the pancreas, the eye, the brain, and the stomach wall, among others. In endemic areas, the acute phase manifestations can be superimposed on chronic infection. Arrival of the parasite in the bile ducts marks the beginning of the chronic phase. Mature flukes consume hepatocytes and duct epithelium and reside for years in the hepatic and common bile ducts and sometimes in the gallbladder. In this chronic phase, the liver contains large dilated, thick-walled, and calcareous bile ducts with yellowish-brown bile. The bile ducts have a thickened hyperplastic wall with marked fibrosis. Symptoms usually reflect biliary obstruction with colicky pain in the right upper quadrant and epigastric area. Eosinophilia is absent in half of the chronic cases. Bacterial superinfection of these cysts and consequent cholangitis can develop. Other complications are hemobilia and liver fibrosis. Alkaline phosphatase is commonly elevated because of biliary obstruction, which sometimes requires surgical intervention. On imaging, the initial lesions may be often confused with hepatic metastases. Other findings on CT are hepatomegaly, tracklike hypodense lesions in subcapsular locations, multiple hypodense nodular areas (abscess-like lesions), or low-density, serpiginous, tortuous, tunnel-like branching lesions ranging from 2 to 10 mm. CT also can show subcapsular hematoma, enhancement of the Glisson capsule, necrotic granuloma, and cystic calcifications. After maturation, the adult flukes start laying eggs, which are passed from the sphincter of Oddi to the intestines and evacuated to the environment along with stool. Adult parasites can live in the bile ducts for up to 13 years. In endemic populations, chronic infection has been reported as mildly symptomatic, whereas in travelers or temporary residents, it has been reported to cause biliary obstruction, with adult parasites being seen on endoscopic retrograde cholangiopancreatography (ERCP).12,13 In summary, the typical clinical presentation of acute fascioliasis must be differentiated from cholecystitis; “liver metastasis” with fever and eosinophilia should raise the possibility of this infection; and in children and adolescents, systemic toxocariasis will be in the differential diagnosis.  

Clonorchiasis and Opisthorchiasis

Clonorchiasis is the disease caused by Clonorchis sinensis, also called the Chinese or oriental liver fluke (adult: 10 to 25 mm × 3 to 5 mm). Opisthorchiasis is caused by Opisthorchis viverrini (adult: 5 to 10 mm × 1 to 2 mm) and Opisthorchis felineus (adult: 7 to 12 mm × 2 to 3 mm). The most common natural hosts are dogs, cats, pigs, and some small wild mammals.  

EPIDEMIOLOGY

The global estimate for the number of people infected with both infections is 46.2 million, 35 million for C. sinensis (15 million in China), 10 million for O. viverrini (8 million in Thailand and 2 million in the Lao People’s Democratic Republic), and 1.2 for O. felineus. There are 601 million and 79.8 million people at risk of infection with Clonorchis and Opisthorchis, respectively. Both infections are endemic in the Far East, Southeast Asia, and Eastern Europe. C. sinensis is endemic in northeast China, southern Korea, Japan, Taiwan, northern Vietnam, and the far eastern part of Russia, and O. viverrini is endemic in Laos, Thailand, Vietnam, and Cambodia. O. felineus infection is prevalent in Russia, Ukraine, and Kazakhstan.14  

PATHOBIOLOGY

The life cycle starts when the adult worm deposits fully developed eggs, which are then passed to the environment through feces. They hatch in water and the miracidia infect their first intermediate host, a freshwater snail (Bithynia spp or Parafossarulus spp), where they transform into sporocysts, rediae, and cercariae. Cercariae are released from the snail and then penetrate freshwater fish, which are the second intermediate host (Cyclocheilichthys spp, Puntius spp, Hampala dispar); the cercariae encyst as metacercariae in the muscles or under the scales. In general, the infection is acquired by eating raw or

uncooked cyprinoid fish products in rural areas or dishes such as koi-pla, a salad made with raw fish. The metacercariae pass through the stomach and reach the small intestine unharmed. Then, through the ampulla of Vater, they reach the bile ducts, where they mature into adult worms within 4 weeks and deposit yellow, operculated eggs. The parasites may live for up to 45 years in a human host.  

CLINICAL MANIFESTATIONS

Clonorchiasis as an acute infection caused by C. sinensis is usually asymptomatic, but some patients may have fever, rash, malaise, and right upper quadrant abdominal discomfort.15 Chronic infections may be manifested as recurrent pyogenic cholangitis, cholecystitis, obstructive jaundice, hepatomegaly, cholecystitis, multiple hepatic tumors, cholelithiasis, or pancreatitis. In chronic carriers with a high load of parasites, cholangiocarcinoma may develop, especially in Thailand.16 Opisthorchiasis as an acute infection caused by O. viverrini can cause right upper quadrant abdominal pain, flatulence, fatigue, and a hot sensation over the abdomen. In the chronic phase, mild hepatomegaly occurs, mainly in more heavily infected patients (egg counts >10,000/g). Jaundice and splenomegaly are not observed. Intrahepatic duct stones and recurrent suppurative cholangitis are common manifestations of opisthorchiasis. Whenever jaundice and ascending cholangitis are detected, fluke-related cholangiocarcinoma should be suspected. In opisthorchiasis caused by O. felineus, infestation usually follows the consumption of raw, slightly salted, and frozen fish (“stroganina”), and acute symptoms occur 2 to 4 weeks later, including high-grade fever, nausea, vomiting, abdominal pain, malaise, arthralgias, lymphadenopathy, and rash. Peripheral eosinophilia is a common finding, especially during the initial 2 to 6 weeks of the infection, together with raised liver enzyme levels. In chronic infection, the eosinophilia is usually milder. Patients may have suppurative cholangitis and liver abscesses because of biliary obstruction. Ultrasonography or CT demonstrates the pathologic changes in the liver, including intrahepatic duct dilation and periductal changes. The pathologic and clinical consequences of these flukes are related to the intensity and duration of cumulative infestations. In general, they cause inflammation around the biliary tree, severe hyperplasia of epithelial cells, metaplasia of mucin-producing cells in the mucosa, and progressive periductal fibrosis. There are clear associations between O. viverrini infection and cholangiocarcinoma.17 Several N-nitroso compounds and their precursors occur at low levels in fermented food, such as preserved mud fish paste (pla ra), a condiment that is a ubiquitous component of the cuisine of northeastern Thailand and Laos.

  INTESTINAL FLUKES

The most common human intestinal trematode is Fasciolopsis buski (adult: 20 to 75 mm × 8 to 20 mm). It is found mainly in the central and southeast parts of Asia. F. buski is a common parasite in pigs. Others are Heterophyes (adult: 1 to 2 mm in length), Metagonimus yokogawai (adult: 1 to 2.5 mm × 0.4 to 0.75 mm), and Echinostoma spp (adult: 6.5 × 1 to 2 mm).  

EPIDEMIOLOGY

More than 50 species of intestinal trematodes from the Far East, Middle East, and North Africa have been reported to cause human infection. H. heterophyes also can be found in the Nile delta region of Egypt. An estimated 40 to 50 million people are infected with one or several species of intestinal flukes. Their life cycles are similar. The adult worm, attached to the intestinal wall of humans, produces eggs that are passed in feces. The eggs reach water, and miracidia develop and penetrate the first intermediate host—snails. During the course of 6 to 7 weeks inside the host snails, they develop into sporocysts, rediae, and cercariae. The cercariae leave the snails to encyst in the second intermediate host, which can be freshwater snails, fish, tadpoles, or vegetables. Humans are infected by the ingestion of raw stems, leaves (especially bamboo shoots), watercress, or water chestnuts with encysted metacercariae. In the human duodenum, the metacercariae attach to the walls and become adult worms in approximately 3 months.  

PATHOBIOLOGY AND CLINICAL MANIFESTATIONS

Despite the fact that the majority of intestinal fluke infections are asymptomatic, they may cause inflammation, ulceration, and mucus secretion at the site of attachment, particularly in the duodenum and jejunum. In fact, gastrointestinal hemorrhage, perforation, and abscesses have been observed. The differential diagnoses include typhoid fever, intestinal tuberculosis, and amebiasis. In these

CHAPTER 334  Trematode Infections  

endemic cases, ulcerative colitis and other inflammatory diseases of the bowel are uncommon. In heavy infection, intestinal obstruction, protein-losing enteropathy, malabsorption, impaired vitamin B12 absorption, hypoalbuminemia, and anasarca have been reported. The adult worm causes traumatic, toxic, and obstructive damage to the intestinal mucosa. Some cases have been diagnosed by direct visualization of the adult parasite via esophagogastroduodenoscopy.

  PULMONARY FLUKES

Paragonimiasis is a zoonosis caused by Paragonimus species (adult size: 10 × 5 mm). Reservoir hosts include felids, canids, viverrids, mustelids, some rodents, and pigs. At least 10 species of Paragonimus are known to cause human disease; of these, Paragonimus westermani is the most common.  

EPIDEMIOLOGY

An estimated 23 million people are infected worldwide with Paragonimus species, and 293 million are at risk. Human paragonimiasis is distributed mainly in Southeast Asia, Japan, Korea, China, and the Philippines, where P. westermani is the main species. In other areas of low endemicity, other species have been reported, such as Paragonimus mexicanus in Latin America, Paragonimus kellicotti in North America, Paragonimus heterotremus in India, and Paragonimus africanus and Paragonimus uterobilateralis in West Africa.18  

PATHOBIOLOGY

The life cycle starts when the eggs are excreted unembryonated in sputum, or alternatively they can be swallowed and passed in stool. In the external environment the eggs become embryonated, and miracidia hatch; seek the first intermediate host, a snail (families Pleuroceridae and Thiaridae); and penetrate its soft tissues. Within the snail, asexual reproduction occurs for several weeks, with transformation into sporocysts, rediae, and cercariae; the last emerge from the snail and invade the second intermediate host, a crustacean such as a crab or crayfish, where they encyst and become metacercariae. This is the infective stage for the mammalian host. Human infection occurs by eating inadequately cooked or pickled crab or crayfish that harbor metacercariae of the parasite. The metacercariae excyst in the duodenum, penetrate the intestinal wall, and migrate through the peritoneal cavity toward the lungs. During migration through the peritoneum and diaphragm, the inflammatory process causes abdominal pain and dry cough. When invading the lungs, they become encapsulated and develop into adults. Infections may persist for 20 years in humans.  

CLINICAL MANIFESTATIONS

Paragonimiasis typically results from the consumption of raw or improperly cooked crustaceans, especially crabs. Recently in the United States, some autochthonous cases have been reported in the Midwest in people who consumed raw crayfish while camping during the summer. Most of the infected people are asymptomatic and have subclinical disease. During the first month of infection, abdominal pain may represent the juvenile larvae migrating through the abdominal cavity before reaching the lungs. Irritation of the diaphragm or pleura may cause dry cough. Fever, chest pain, fatigue, and urticaria may follow, as well as eosinophilia. Pleural effusions may be seen at this stage, with significant eosinophilia noted on analysis of pleural fluid, which can be the first clue to the diagnosis. In fact, pleural manifestations predominate early in the disease process, whereas lesions of the pulmonary parenchyma predominate later in the course of disease. Moreover, pneumothorax and mild eosinophilia may occur only 1 month after the initiation of infection. The migrating worms may cause bronchiectasis, interstitial pneumonitis, transient hemorrhage, or bronchopneumonia. Again, pulmonary lesions and eosinophilia raise the possibility of paragonimiasis. Cough and recurrent hemoptysis are common clinical findings in this phase. The chronic stage occurs when the worms are paired in a cyst in the pulmonary parenchyma. Eggs are produced 6 weeks after infection, and if there is communication with the bronchial tree, eggs may be seen in a sputum sample under microscopy, or they can be swallowed and passed with stool. The rusty discoloration of sputum is caused by the presence of the tan- to brown-pigmented Paragonimus eggs; the sputa of these patients have been classically described as resembling “iron filings.” Charcot-Leyden crystals can be seen. Peripheral blood eosinophilia and elevated total serum IgE levels are observed in approximately 80% of patients. Common findings on chest CT are pleural effusion, hydropneumothorax, pulmonary nodules or consolidation of air spaces, and cysts. The most common ectopic form is cerebral paragonimiasis, which is manifested as eosinophilic meningitis or meningoencephalitis,19 brain tumor, or just residual calcifications from a past infection.

2127

  MANAGEMENT OF LIVER, INTESTINAL, AND

LUNG FLUKE INFECTIONS



DIAGNOSIS

In general, transmission of food-borne trematodiases is restricted geographically, and distribution is one of the most important factors in suspecting this diagnosis. In the appropriate clinical setting, laboratory and imaging data can add information to narrow the differential diagnosis. Acute fluke infections require a high level of suspicion. Serologic tests, direct visualization of the migrating larvae, or empirical therapy with a significant clinical response (including reduction of eosinophilia) are major criteria to confirm the diagnosis. In the chronic phase, the diagnosis is usually made by visualization of the eggs in stool or, in the case of Paragonimus, in sputum. A sedimentation technique must be performed on a series of at least three stool specimens from alternate days or even weeks. Immunodiagnostics is an excellent tool, particularly for patients who do not have demonstrable eggs in clinical specimens. If acute fascioliasis is strongly suspected, serology would be the next step. Cathepsin L1–based antibody enzyme-linked immunosorbent assay (ELISA) has a sensitivity of 92% and a specificity of 84%. If negative, the diagnosis is unlikely. If serology is not available, CT of the liver can visualize the characteristic tracklike lesions. However, because the parasitic lesions are very similar to metastases, liver biopsy may be necessary. If serology or CT is not available, a trial of triclabendazole with clinical (including eosinophilia) resolution is the major criterion for diagnosis. In chronic fascioliasis, the Lumbreras rapid sedimentation technique is the method of choice to detect the eggs in stool. At least three stool examinations are preferred. If negative, serology can be helpful. Ultrasonography and CT have low sensitivity in this phase. ERCP usually finds the adult parasites in the bile duct when performed for other reasons. Nonetheless, ERCP can be useful to eliminate the adult parasites causing biliary obstruction. In opisthorchiasis, serology or stool examinations can be performed to approach the diagnosis. The Ov-CP-1–based ELISA has a sensitivity of 95% and a specificity of 96%. For C. sinensis, ELISA has a sensitivity between 81.3 and 96% and a specificity between 92.6 and 96.2%. For detection of eggs and to measure the intensity of infection, the Kato-Katz technique is preferred. Intestinal fluke eggs can be detected by performing a sedimentation stool technique, preferably in consecutive stool samples. For paragonimiasis, an immunoblot assay performed with a crude antigen extract of P. westermani has been in use at the U.S. Centers for Disease Control and Prevention; the sensitivity of the test is 96%, and its specificity is 99%. This would be the ideal first step to confirm the diagnosis. In the acute phase, the precise location of the migrating larva is unknown and a biopsy may not necessarily target the parasite. When serologic findings are negative (or not available), a trial of praziquantel or triclabendazole with a positive clinical response in 48 to 72 hours is a major criterion for diagnosis. In the chronic phase, several sputum samples have to be examined by a sedimentation technique to increase sensitivity. Stool examinations are complementary because the eggs can be swallowed by the host and then passed through stool. If a pulmonary cyst contains adult parasites with eggs but they do not have communication to the main bronchi, serologic examination is indicated to confirm the diagnosis. If not available, biopsy is warranted. Under the light microscope, the morphologic characteristics and size of the eggs may be sufficient to identify the specific fluke.

TREATMENT AND PREVENTION  For fascioliasis, 10 mg/kg of triclabendazole once or twice has a cure rate higher than 90%, and it is the treatment of choice, but failures have been reported. Cure is achieved if stool examinations remain negative for at least 3 months. Serology usually can take more than a year to resolve. In treatment of the chronic phase, the dead parasites can occasionally cause biliary obstruction, which may need surgical consultation. In case of failure, some experts recommend at least double doses of triclabendazole for 2 days, and even multiple rounds are needed to resolve the infection. For O. viverrini, a single dose of praziquantel (40 to 50 mg/kg) has a cure rate of 91 to 97%. For clonorchiasis, the recommended dose of praziquantel is 25 mg/ kg three times for 1 day (total dose of 75 mg/kg), which has a cure rate of 83 to 85%. Tribendimidine has been found to be comparably effective as praziquantel in the treatment of C. sinesis infection with fewer adverse events in an open-label randomized trial. A3  Tribendimidine (100 to 600 mg orally) is an

efficacious alternative to praziquantel for treating O. viverrini  A4  but may or may not be quite as efficacious. A5  For intestinal flukes, praziquantel, 25 mg/kg by mouth three times daily for 1 day, is recommended. For paragonimiasis, praziquantel, 25 mg/kg by mouth three times daily for 3 days, or triclabendazole, 10 mg/kg by mouth once or twice, is highly effective. For ectopic cases, surgery may be necessary. In follow-up, negative stool examinations in the ensuing weeks can confirm cure. However, because the rate of reinfection is high in individuals from endemic areas, a suddenly positive stool examination is highly suggestive of a new infection rather than failure to respond to treatment. Prevention of infection with these flukes depends on several factors, including recognizing their geographic distribution and avoiding consumption of raw vegetables, fish, crayfish, or contaminated water in endemic areas. Proper medical advice must be given to individuals traveling to or planning to reside in endemic areas, not only to prevent flukes’ infection but also risk for coinfections with other parasites. Control of these flukes in animals is impractical because of wild animal reservoirs, but in human infections it is challenging. It involves changing long-established cultural, dietary, and sanitary habits. Massive chemotherapy to highly endemic populations may reduce the infection in humans and in selected animals.

  Grade A References A1. Danso-Appiah A, Olliaro PL, Donegan S, et al. Drugs for treating Schistosoma mansoni infection. Cochrane Database Syst Rev. 2013;2:CD000528. A2. Pérez del Villar L, Burguillo FJ, López-Abán J, et al. Systematic review and meta-analysis of artemisinin based therapies for the treatment and prevention of schistosomiasis. PLoS ONE. 2012;7:1-15. A3. Xu LL, Jiang B, Duan JH, et al. Efficacy and safety of praziquantel, tribendimidine and mebendazole in patients with co-infection of Clonorchis sinensis and other helminths. PLoS Negl Trop Dis. 2014;8:1-10. A4. Sayasone S, Odermatt P, Vonghachack Y, et al. Efficacy and safety of tribendimidine against Opisthorchis viverrini: two randomised, parallel-group, single-blind, dose-ranging, phase 2 trials. Lancet Infect Dis. 2016;16:1145-1153. A5. Sayasone S, Keiser J, Meister I, et al. Efficacy and safety of tribendimidine versus praziquantel against Opisthorchis viverrini in Laos: an open-label, randomised, non-inferiority, phase 2 trial. Lancet Infect Dis. 2018;18:155-161.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 334  Trematode Infections  

GENERAL REFERENCES 1. Tchuenté L-AT, Rollinson D, Stothard JR, et al. Moving from control to elimination of schistosomiasis in sub-Saharan Africa: time to change and adapt strategies. Infect Dis Poverty. 2017;6:42-55. 2. King CH, Sutherland LJ, Bertsch D. Systemic review and meta-analysis of the impact of chemicalbased mollusciciding for control of Schistosoma mansoni and S. haematobium transmission. Plos Negl Trop Dis. 2015;9:1-23. 3. De Almeida TVVS, Fernandes JS, Lopes DM, et al. Schistosoma mansoni antigens alter activation markers and cytokine profile in lymphocytes of patients with asthma. Acta Trop. 2017;166: 268-279. 4. McManus DP, Dunne DW, Sacko M, et al. Schistosomiasis. Nat Rev Dis Primers. 2018;4:1-19. 5. Ogongo P, Kariuki TM, Wilson RA. Diagnosis of schistosomiasis mansoni: an evaluation of existing methods and research towards single worm pair detection. Parasitology. 2018;6:1-12. 6. Wu Y, Liu J, Lin Y, et al. Diagnosis, monitoring, and control of schistosomiasis—an update. J Biomed Nanotechnol. 2018;14:430-455. 7. Kittur N, Castleman JD, Campbell CH Jr, et al. Comparison of Schistosoma mansoni prevalence and intensity of infection, as determined by the circulating cathodic antigen urine assay or the Kato-Katz fecal assay: a systematic review. Am J Trop Med Hyg. 2016;94:605-610. 8. Utzinger J, Becker SL, van Lieshout L, et al. New diagnostic tools in schistosomiasis. Clin Microbiol Infect. 2015;21:529-542. 9. Hotez PJ, Bottazzi ME, Bethony J, et al. Advancing the development of a human schistosomiasis vaccine. Trends Parasitol. 2019;35:104-108.

2128.e1

10. Abou-El-Naga IF. Towards elimination of schistosomiasis after 5000 years of endemicity in Egypt. Acta Trop. 2018;181:112-121. 11. Toor J, Alsallaq R, Truscott JE, et al. Are we on our way to achieving the 2020 goals for schistosomiasis morbidity control using current World Health Organization guidelines? Clin Infect Dis. 2018;66:S245-S252. 12. Webb CM, Cabada MM. Recent developments in the epidemiology, diagnosis, and treatment of Fasciola infection. Curr Opin Infect Dis. 2018;31:409-414. 13. Machicado C, Machicado JD, Maco V, et al. Association of Fasciola hepatica infection with liver fibrosis, cirrhosis, and cancer: a systematic review. PLoS Negl Trop Dis. 2016;10:1-11. 14. Zheng S, Zhu Y, Zhao Z, et al. Liver fluke infection and cholangiocarcinoma: a review. Parasitol Res. 2017;116:11-19. 15. Harrington D, Lamberton PHL, McGregor A. Human liver flukes. Lancet Gastroenterol Hepatol. 2017;2:680-689. 16. Qian MB, Utzinger J, Keiser J, et al. Clonorchiasis. Lancet. 2016;387:800-810. 17. Feng M, Cheng X. Parasite-associated cancers (blood flukes/liver flukes). Adv Exp Med Biol. 2017;1018:193-205. 18. Fischer PU, Weil GJ. North American paragonimiasis: epidemiology and diagnostic strategies. Expert Rev Anti Infect Ther. 2015;13:779-786. 19. Bahr NC, Trotman RL, Samman H, et al. Eosinophilic meningitis due to infection with Paragonimus kellicotti. Clin Infect Dis. 2017;64:1271-1274.

2128.e2

CHAPTER 334  Trematode Infections  

REVIEW QUESTIONS 1. Clinic and pathologic findings in S. haematobium infection include which of the following? A . Abdominal pain due to the intestine inflammation B. Hepatosplenomegaly C. Hematuria, hydroureter, and hydronephrosis D. Portal hypertension E. Renal failure Answer: C In S. haematobium infection eggs released in the urinary tract induce granuloma formation and fibrosis. Hematuria and dysuria are the most common manifestation. The fibrosis may lead to hydroureter and hydronephrosis. 2. Liver fibrosis to S. mansoni infection is characterized by which of the following? A . Only observed in adults B. Hepatosplenomegaly and hematemesis from bleeding esophageal or gastric varices C. Liver cirrhosis D. Jaundice and increasing in transaminases E. Hypoalbuminemia and edema Answer: B  Liver fibrosis is the main pathology related to S. mansoni infection. Usually the hepatic function is normal and the main clinical finding is hepatosplenomegaly that can be documented since childhood. Hematemesis due to bleeding esophageal or gastric varices is the most important complication. 3. Diagnosis of acute schistosomiasis due to S. mansoni: A . Should be performed by stool examination. B. May be performed by detection of eggs in the urine. C. May be performed by real-time polymerase chain reaction in stools. D. May be performed by rectal biopsy. E. May be performed by detection of antibody or cathodic circulating antigens. Answer: E  Eggs can only be detected in stool or urine 5 to 9 weeks after cercariae penetration; therefore diagnosis of acute schistosomiasis is performed based on the clinical manifestations, presence of eosinophilia, detection of cathodic circulating antigens in urine, or by serologic test. 4. Which of the following is true about human schistosomiasis? A . It may be prophylactically treated with praziquantel. B. Praziquantel is currently the treatment of choice. C. The combination of praziquantel and oxamniquine is recommended. D. There is no combination treatment available. E. Susceptibility or resistance of praziquantel has not been reported. Answer: B  Praziquantel has low prophylactic effect, which reduces its efficacy in areas of high transmission of Schistosoma. Praziquantel is considered the drug of choice to treat schistosomiasis for its efficacy, low adverse events including severe adverse events, single oral dose, and competitive cost. There is a combination of praziquantel and artemisinin derivatives that shows increased cure rates in schistosomiasis treatment. 5. Which of the following is true regarding artemisinin derivative–based therapies for human schistosomiasis? A . Artesunate has increased cure rates. B. Artemether has increased cure rates. C. Praziquantel plus artemisinin derivatives increased cure rates. D. Artesunate cannot be used for prophylactic treatment. E. Oxamniquine increased cure rates compared with praziquantel. Answer: C  Artemisinin derivatives such as artesunate or artemether show no increased cure rates compared with praziquantel. This is also true for oxamniquine use. However, recent evidence shows that the combination of praziquantel with artemisinin derivatives shows increased cure rates in schistosomiasis treatment, which also could be a benefit for low susceptibility or resistance of praziquantel in human schistosomiasis.

6. Fasciola hepatica is acquired by ingestion of: A . Uncooked shellfish B. Watercress C. Fresh fish meat products D. Fresh fish E. All the above Answer: B  Fasciola hepatica do not invade fish or meat. After invading freshwater snails, they leave as free-swimming cercaria that subsequently attach to watercress (as well as water lettuce, alfalfa, mint, parsley, or khat). 7. Paragonimus is acquired by ingestion of: A . Uncooked crab or crayfish B. Watercress C. Fresh fish meat products D. All of the above Answer: A  Paragonimus emerges from the snail and invades the second intermediate host, a crustacean such as a crab or crayfish, where they encyst and become the infective stage for a mammalian host. 8. The treatment of choice for O. viverrini or clonorchiasis is: A . Albendazole B. Praziquantel C. Ivermectin D. Triclabendazole E. Emetina Answer: B For O. viverrini or clonorchiasis, praziquantel regimens have been shown in several studies to have cure rates of 91 to 97% and 83 to 85%, respectively. Albendazole has poor activity. Ivermectin is effective for onchocerciasis and strongyloides. Triclabendazole is the drug for Fasciola, but it is a second-line option here. Emetina is a parenteral drug that is useful only for Fasciola with high toxicity. 9. Cholangiocarcinoma is associated in Southeast Asia with chronic carriage of: A . Fasciola hepatica B. Strongyloides C. Toxoplasma D. O. viverrini E. Taenia Answer: D  Several studies in Thailand have demonstrated a strong association of O. viverrini with cholangiocarcinoma. Fasciola hepatica is associated with fibrosis but not cancer. Strongyloides produces hyperinfection but not cancer. Toxoplasma does not invade the biliary tract. Taenia is not associated with cancer. 10. A 33-year-old man presents with pleural effusion and eosinophilia. In the following weeks he developed a recurrent hemoptysis. The most likely diagnosis is: A . Tuberculosis B. Strongyloidiasis C. Toxocara D. Fasciola hepatica E. Paragonimus Answer: E  The typical clinical pattern for Paragonimus is the damaged lung, where hemoptysis and eosinophilia are the more common presentation. Evaluation of the sputum will be the key diagnostic test in this case. TB is the main differential diagnosis, but eosinophilia is not common in TB. Strongyloidiasis produces lung disease (migratory eosinophilic pneumonia) but is rarely associated with pleural effusion. Toxocara usually affects children and produces systemic disease with liver involvement; lung involvement is rare and hemoptysis even more so. Fasciola hepatica is associated with eosinophilia, but its typical presentation is a liver “abscess”–like or metastatic-like syndrome, and lung involvement with hemoptysis is rare.

2128

CHAPTER 335  Nematode Infections  

With the exception of S. stercoralis and Capillaria philippinensis, adult worms cannot complete their life cycle within a human host. Instead, sexually mature adult worms mate and produce eggs or larvae that must have at least one stage of development outside the host, either in the environment or in an intermediate host. Nematode infections are rarely fatal; they more commonly result in chronic morbidity such as iron deficiency anemia caused by hookworm or blindness due to onchocerciasis. For most nematodes, the severity of the clinical manifestations of infection is proportional to the number of worms harbored by a given host; although light infections with only a few worms are usually asymptomatic, pathologic features appear with heavier worm burdens. Nematode infections are prevalent in the temperate and tropical regions of Africa, Asia, and Latin America. They are transmitted by the oral ingestion of embryonated eggs or by penetration of infective larvae through the skin, either by direct contact with contaminated soil or by the bite of an arthropod vector. Nematode infections are most common in areas with poor sanitation, where the environment is contaminated by human feces, and in climates that support survival of the insect vector if one is involved in the life cycle.

  INTESTINAL NEMATODE INFECTIONS Ascariasis The Pathogen  

A. lumbricoides, colloquially known as roundworm, is acquired by oral ingestion of embryonated eggs. In the stomach, the egg’s protective outer shell is dissolved by gastric acid, releasing larvae into the small intestine, where they penetrate the intestinal wall and enter the portal circulation. The larvae migrate to the pulmonary vasculature, where they penetrate into the alveoli, ascend the bronchial tree, and are swallowed back into the intestinal tract where they develop into adult worms 9 to 11 weeks after egg ingestion. Adult worms (Fig. 335-1) range in length between 15 and 50 cm and survive in the host for approximately 18 months. Female adult Ascaris worms release more than 200,000 eggs per day that are expelled in feces. Fertilized eggs (Fig. 335-2)

335  NEMATODE INFECTIONS DAVID J. DIEMERT



DEFINITION

Nematode infections are highly prevalent worldwide, affecting millions. Nematodes are complex, nonsegmented roundworms with specialized organs that include a protective outer coating or cuticle, a complete and functional gastrointestinal tract, and muscular, nervous, and reproductive systems. Most species are free-living in soil or water, but a few parasitize humans. Nematodes of medical importance can be categorized into those that primarily affect the gastrointestinal tract, where adult worms become established and cause disease, and those that affect other tissues and organ systems. The former group includes the roundworm Ascaris lumbricoides, the hookworms Ancylostoma duodenale and Necator americanus, the pinworm Enterobius vermicularis, the whipworm Trichuris trichiura, and the threadworm Strongyloides stercoralis. Zoonotic intestinal nematodes such as Trichostrongylus and Anisakis also occasionally infect and cause disease in humans. Nematodes that invade and cause disease primarily in tissues outside the gastrointestinal tract include those that cause lymphatic filariasis (Wuchereria bancrofti, Brugia malayi, and Brugia timori), Onchocerca volvulus, Loa loa, the guinea worm Dracunculus medinensis, and Trichinella and Angiostrongylus species. Nematodes that infect humans measure from several millimeters to more than a meter in length and often survive for months to years within their host.

FIGURE 335-1.  Mass of adult Ascaris lumbricoides worms recovered from a child after the administration of mebendazole. (Reproduced with permission from Dickson Despommier.)

FIGURE 335-2.  Fertilized, unembryonated egg of Ascaris lumbricoides. (Reproduced with permission from Dickson Despommier.)

CHAPTER 335  Nematode Infections  

ABSTRACT

Nematodes of medical importance can be categorized into those that primarily affect the gastrointestinal tract, where adult worms become established and cause disease, and those that affect other tissues and organ systems. The former group includes the roundworm Ascaris lumbricoides, the hookworms Ancylostoma duodenale and Necator americanus, the pinworm Enterobius vermicularis, the whipworm Trichuris trichiura, and the threadworm Strongyloides stercoralis. Nematodes that invade and cause disease primarily in tissues outside the gastrointestinal tract include those that cause lymphatic filariasis (Wuchereria bancrofti, Brugia malayi, and Brugia timori), Onchocerca volvulus, Loa loa, the guinea worm Dracunculus medinensis, and Trichinella and Angiostrongylus species. Nematode infections are rarely fatal; they more commonly result in chronic morbidity such as iron deficiency anemia caused by hookworm or blindness due to onchocerciasis. Nematode infections are prevalent in the temperate and tropical regions of Africa, Asia, and Latin America. They are transmitted by the oral ingestion of embryonated eggs or by penetration of infective larvae through the skin, either by direct contact with contaminated soil or by the bite of an arthropod vector. Nematode infections are most common in areas with poor sanitation, where the environment is contaminated by human feces, and in climates that support survival of the insect vector if one is involved in the life cycle.

KEYWORDS

intestinal nematodes tissue nematodes soil-transmitted helminths filariases hookworm ascariasis cutaneous larva migrans elephantiasis trichinellosis/trichinosis filariasis

2128.e3

CHAPTER 335  Nematode Infections  

become infectious after embryonating in warm, moist, shady soil. Eggs are resistant to extreme temperatures and desiccation and can survive up to 15 years in the environment.  

TABLE 335-1 TREATMENT OF INTESTINAL NEMATODES NEMATODE

EPIDEMIOLOGY

Ascaris lumbricoides

A. lumbricoides is the most prevalent nematode infection worldwide, affecting approximately 800 million people in sub-Saharan Africa, south and Southeast Asia, and Latin America, primarily in rural areas of high population density that lack adequate sanitation or treatment of sewage or where untreated human feces are used as fertilizer.1 Climate is an important determinant of disease in that warm temperature and adequate moisture are required for embryonation of eggs in soil. In endemic areas, the prevalence and intensity of infection increases dramatically during the first 2 to 3 years of life, remains high between the ages of 4 and 15 years and then declines during adulthood.  



DIAGNOSIS

The diagnosis of ascariasis is usually made by microscopic examination of a sample of feces for characteristic thick-shelled eggs.2 However, during the pulmonary phase of infection, ova will not be detectable in feces because adult worms have not yet matured and begun producing eggs; instead, larvae, as well as eosinophils or Charcot-Leyden crystals (formed from the breakdown of eosinophils), may be visualized on microscopic examination of sputum. Pulmonary disease is also usually characterized by peripheral eosinophilia and transient infiltrates on chest radiographs. The diagnosis of intestinal or biliary obstruction caused by A. lumbricoides is usually made by ultrasound examination or endoscopic retrograde cholangiopancreatography (ERCP).

TREATMENT  Intestinal ascariasis is usually cured with a single oral dose of albendazole (Table 335-1). A1  A2  Alternatives include mebendazole, ivermectin, or pyrantel pamoate. No specific treatment is recommended for symptoms of pulmonary ascariasis because the condition is self-limited. In severe cases of biliary obstruction, including cholangitis, ERCP with or without resection of the ampulla of Vater is highly successful and may preclude the need for surgery.1

Trichuris trichiura

Albendazole, 400 mg daily for 3 days. Alternative: mebendazole, 100 mg bid for 3 days, ivermectin 200  µg/kg daily for 3 days

Enterobius vermicularis

Pyrantel pamoate, 11 mg/kg once, with a second dose 2 wk later; maximum dose of 1 g. Alternatives: mebendazole, 100 mg once, or albendazole, 400 mg once, repeated in 2 wk

Strongyloides stercoralis

Uncomplicated infection: ivermectin, 200 µg/kg daily for 2 days.* Alternative: albendazole, 400 mg bid for 7 days

Trichostrongylus spp

Pyrantel pamoate, 11 mg/kg once; maximum dose of 1 g. Alternatives: albendazole, 400 mg daily for 10 days, or mebendazole, 500 mg daily for 10 days

Capillaria philippinensis

Albendazole, 400 mg bid for 10 days. Alternative: mebendazole, 500 mg daily for 20 days

*Treatment may need to be extended in immunocompromised patients with disseminated disease.

Hookworm The Pathogen  

Hookworm infection in humans is due almost exclusively to two species: N. americanus and A. duodenale. However, incidental infection with the zoonotic hookworms Ancylostoma caninum, Ancylostoma braziliensis, Bunostomum phlebotomum, and Uncinaria stenocephala can cause self-limited dermatologic lesions known as cutaneous larva migrans (Fig. 335-3). Additionally, Ancylostoma ceylanicum, normally a hookworm infecting cats, has been reported to cause intestinal hookworm disease in humans, especially in Asia, whereas A. caninum has been implicated as a cause of eosinophilic enteritis in Australia. Infection occurs when exposed skin comes in contact with infective filariform larvae in fecally contaminated soil or grass. Larvae penetrate the skin, enter the afferent circulation, and are transported to the pulmonary vasculature, where they penetrate the alveolar wall, ascend the bronchial tree to the larynx, and are swallowed into the gastrointestinal tract. Larvae mature into sexually mature adult worms 5 to 9 weeks after skin penetration. Adult hookworms reside in the lumen of the small intestine, where they attach to the mucosa by means of cutting teeth (A. duodenale) or a rounded cutting plate (N. americanus). After mating in the host intestinal tract, female adult worms produce eggs that are expelled from the body in feces; A. duodenale female worms lay approximately 28,000 eggs daily, whereas the output from N. americanus worms averages around 10,000 a day. Hookworm eggs hatch in warm, moist, sandy soil and release larvae that can infect another host. Humans are the only major definitive host for these two parasites, and there are no intermediate or reservoir hosts. A. duodenale survives on average for 1 year in the human intestine, whereas N. americanus lives for 3 to 5 years.

,



PREVENTION

The definitive means of preventing Ascaris infection is improvement of hygiene and proper disposal of human waste.3 In endemic communities where this is not feasible, morbidity control consists of regular (usually annual) mass administration of an anthelmintic medication such as albendazole or mebendazole to preschool- and school-aged children.4

TREATMENT Albendazole, 400 mg once. Alternatives: mebendazole, 500 mg once or 100 mg bid for 3 days, ivermectin 150-200  µg/kg once, or pyrantel pamoate, 11 mg/kg once with the maximum daily dose not to exceed 1 g

Hookworm Albendazole, 400 mg daily for 3 days. Alternatives: mebendazole, 500 mg daily or 100 mg bid for 3 days, (Necator americanus and or pyrantel pamoate, 11 mg/kg for 3 days with the Ancylostoma maximum daily dose not to exceed 1 g. duodenale) Tribendimidine, 400-mg single dose alone or in combination, comparable to albendazole

CLINICAL MANIFESTATIONS

Low-intensity A. lumbricoides infections are usually asymptomatic. Clinical manifestations are associated with heavy worm burdens and result from larval migration through the lungs and from parasitism of the gastrointestinal tract by adult worms. During pulmonary migration, A. lumbricoides larvae can induce an intense reaction that is due to both physical disruption and a hypersensitivity response to secreted antigens. This phenomenon is more common in areas in which transmission is seasonal, such as on the Arabian Peninsula, where outbreaks of pneumonitis typically follow the rainy season because of resumption of transmission. Symptoms include sudden onset of wheezing, dyspnea, paroxysmal nonproductive cough, and high fever and may last for 2 to 3 weeks before resolving spontaneously. Respiratory symptoms may coincide with or be preceded by urticarial rash, angioedema, abdominal pain, and vomiting. With moderate or heavy infections, obstruction can be caused by a mass of worms in the small intestine or migration of worms to the biliary tree, pancreatic duct, or appendix. Intestinal obstruction is more common in young children because of the smaller lumen size and is characterized by colicky abdominal pain and vomiting that may progress to signs of intestinal perforation. Hepatobiliary and pancreatic ascariasis are more common in adults. Chronic intestinal infection can manifest as abdominal pain and distention, diarrhea, and nausea. More insidious effects, especially in children, include decreased protein and fat absorption, development of vitamin A and C deficiencies, and lactose intolerance, which together lead to stunted growth and impaired cognitive development.

2129



EPIDEMIOLOGY

More than 450 million people are infected with hookworms worldwide.5 N. americanus is widespread, whereas infection with A. duodenale is more geographically restricted. The highest prevalence of infection occurs in rural areas of tropical and less developed countries, where environmental and socioeconomic conditions favor transmission. Climate is an important determinant of transmission, with adequate moisture and warm temperature being essential for larval development in soil. Equally important determinants of infection are poverty and lack of access to adequate sanitation and clean water. In children living in endemic areas, prevalence increases with age until it plateaus by about 10 years of age, whereas intensity of infection rises at a slower rate during childhood, reaches a plateau by around 20 years, and then increases again in the elderly. Whether such age dependency reflects

2130

CHAPTER 335  Nematode Infections  

FIGURE 335-3.  Typical lesion of cutaneous larva migrans. An erythematous, serpiginous track caused by intradermal migration of a dog (Ancylostoma caninum) or cat (Ancylostoma braziliense) hookworm larva is apparent. (Reproduced with permission from Gregory L. Zalar.)

differences in exposure, acquired immunity, or a combination of both is controversial. Although cutaneous larva migrans is found throughout the tropics, in the United States it is diagnosed primarily in travelers who have recently returned from a vacation to a tropical beach destination, especially in the Caribbean, Brazil, Mexico, and Southeast Asia. Occasionally, autochthonous cases (originating where found) have been reported in the United States, usually from southeastern coastal states such as Florida and South Carolina. Cutaneous larva migrans occurs when exposed skin comes in contact with the larval stages of the dog or cat hookworms A. caninum or A. braziliense, respectively, present in moist soil or sand contaminated with animal feces. Other animal hookworms such as U. stenocephala and B. phlebotomum are less common causes.  

the oral ingestion of A. duodenale larvae, known as Wakana syndrome, which is characterized by nausea, vomiting, pharyngeal irritation, cough, dyspnea, and hoarseness. Abdominal symptoms and signs caused by hookworm infection are rare. Instead, the manifestations of hookworm disease occur when intestinal blood loss exceeds the nutritional reserves of the host and results in iron deficiency anemia. Usually only moderate- and high-intensity (≥2000 eggs per gram of feces) hookworm infections produce clinical disease, which resembles that of iron deficiency anemia secondary to other causes (Chapter 150). In addition, the protein losses associated with heavy hookworm infection can result in hypoproteinemia and anasarca. As iron deficiency anemia develops and worsens, weakness, palpitations, fainting, dizziness, dyspnea, lassitude, and headache may result. Uncommonly, there may be constipation or diarrhea with occult blood in the stool or frank melena, especially in children; there may also be an urge to eat soil (pica). Overwhelming hookworm infection may cause listlessness, coma, and even death, especially in infants. Because children and women of reproductive age have reduced iron reserves, they are at particular risk for symptomatic disease. Severe iron deficiency anemia caused by hookworm during pregnancy can result in adverse consequences for the mother, her unborn fetus (miscarriage, intrauterine growth restriction), and the neonate (anemia, failure to thrive). In children, the anemia and protein malnutrition associated with chronic intestinal parasitism cause long-term impairments in physical and cognitive development.  

TREATMENT  Three daily oral doses of albendazole, 400 mg, is the recommended treatment of intestinal hookworm infection (see Table 335-1). Tribendimidine, at a single oral dose of 400 mg, alone or in combination, has been found to have a similar, noninferior efficacy profile as albendazole. A3  Less effective alternatives include mebendazole, pyrantel pamoate, and single-dose albendazole. Iron supplementation should be considered in those with significant or symptomatic anemia. For cutaneous larva migrans, although the disease is self-limited and will resolve spontaneously within weeks to a few months, treatment with a single dose of ivermectin will lead to more rapid resolution of symptoms and skin manifestations. Albendazole is an alternative treatment of cutaneous larva migrans.

PATHOBIOLOGY

The major pathology of hookworm infection is due to the associated gastrointestinal blood loss and the resulting iron deficiency anemia. Hookworms attach to the intestinal mucosa and secrete enzymes that enable them to invade submucosal tissues and ingest villous tissue and blood. Hemoglobinases within the hookworm’s digestive canal degrade host hemoglobin for use as an essential nutrient source. The amount of blood loss is directly related to the total worm burden. A. duodenale causes more blood loss than N. americanus: each N. americanus worm results in a daily blood loss of 0.03 to 0.1 mL, whereas the corresponding figure for A. duodenale is between 0.15 and 0.26 mL.  

CLINICAL MANIFESTATIONS

The clinical features of hookworm infection are due to the acute manifestations associated with larval migration through the skin and other tissues or acute and chronic manifestations resulting from parasitism of the gastrointestinal tract by adult worms. Repeated skin exposure to hookworm larvae can result in a hypersensitivity reaction known as “ground itch,” a pruritic erythematous and papular rash that appears most commonly on the hands and feet. In contrast, when zoonotic hookworm larvae penetrate the skin to produce cutaneous larva migrans, most commonly on the feet, thighs, and buttocks, they are unable to complete their life cycle in the human host and eventually die after causing a typical clinical syndrome of intensely pruritic, erythematous serpiginous tracks (see Fig. 335-3). Tracks appear after an incubation period of a few days, can be single or multiple, and advance by millimeters to a few centimeters each day. Vesiculobullous or papular lesions may develop along the tracks, as can secondary bacterial infection as a result of scratching. Untreated, lesions usually heal spontaneously within weeks to months following death of the larvae in the skin. Migration of hookworm larvae through the lungs may induce mild and transient pulmonary symptoms consisting of dry cough, sore throat, wheezing, and low-grade fever. Uncommonly, acute symptomatic disease may follow

DIAGNOSIS

The diagnosis of hookworm infection is made by microscopic identification of characteristic eggs in the stool. The eggs of N. americanus and A. duodenale cannot be distinguished because both are colorless and have a single thin hyaline shell with blunted ends; they range in size from 55 to 75 µm by 36 to 40 µm. Egg concentration techniques, such as the formalin–ethyl acetate sedimentation method, can be used to detect even light infections, although a direct wet mount examination is adequate for detecting moderate to heavy infections. In addition, eosinophilia is a common finding in chronic infection and also during larval migration through the lungs.



PREVENTION

The ideal method for preventing hookworm infection is improvement in hygiene and proper disposal of human waste. Until this occurs, in endemic communities control of disease consists of regular (at least annual) mass administration of an antihelmintic medication such as albendazole or mebendazole. For cutaneous larva migrans, tourists should be advised to wear shoes or sandals when walking to and on beaches and to avoid beaches frequented by stray cats and dogs.

Trichuriasis The Pathogen  

T. trichiura, or whipworm, does not have a tissue migratory phase like A. lumbricoides and hookworm, and its entire life cycle in the host is confined to the gastrointestinal tract. After embryonated eggs are ingested orally, larvae are released into the small intestine where they undergo a series of molts before being carried passively to the transverse and descending colon. The adult worm’s narrow anterior end embeds in the columnar epithelium, while the posterior portion protrudes into the lumen, thereby allowing eggs to be released into feces, which are then passed into the environment where they must embryonate in warm, moist soil to complete the life cycle. Adult worms

CHAPTER 335  Nematode Infections  

can measure up to 50 mm in length and survive in the host for approximately 1.5 to 2 years. The period between ingestion of eggs and detection of eggs in feces is about 90 days.  

EPIDEMIOLOGY

immediately after waking and before bathing. Characteristic E. vermicularis eggs are oval and slightly flattened on one side. It is unusual to find eggs in feces or adult worms in the perianal area. Repeated examination may be necessary.

The estimated worldwide prevalence of trichuriasis is 477 million, most commonly in poor, rural areas of the tropics and subtropics where disposal of human waste is inadequate. Children are more frequently infected than adults and more likely to have higher worm burdens. Humans are the only host.  

TREATMENT  Pinworm infection is treated with a single dose of pyrantel pamoate, mebendazole, or albendazole, which must be repeated 2 weeks later because the drugs do not kill eggs or developing larvae (see Table 335-1). Given the high rate of transmission, all household members and individuals in close contact with the patient (e.g., other children attending the same daycare center) should also be treated. Bedding and underclothes should be thoroughly laundered in hot water followed by a hot dryer to kill remaining eggs.

CLINICAL MANIFESTATIONS

Most T. trichiura infections are asymptomatic. Symptomatic disease occurs primarily in children because the majority of heavy infections (>10,000 eggs per gram of feces) occur in this age group. Heavy infections can be accompanied by acute dysentery or chronic colitis resembling inflammatory bowel disease and result in abdominal pain and diarrhea. Chronic mucosal inflammation and edema of the colon and rectum can lead to protracted tenesmus that results in rectal prolapse. Chronic Trichuris colitis can also lead to malnutrition, impaired growth, and anemia.

2131

Strongyloidiasis The Pathogen  



DIAGNOSIS

Infection is diagnosed by microscopic identification of the typical barrel-shaped eggs with bipolar plugs in direct or concentrated smears of fecal specimens.

TREATMENT  Although T. trichiura responds less effectively than A. lumbricoides or hookworm to treatment with albendazole or mebendazole, a 3-day course of one of these two benzimidazole drugs is the recommended therapy, as listed in Table 335-1. The addition of ivermectin (200 µg/kg) to either drug increases the response rate significantly. A4 



PREVENTION

As for A. lumbricoides and hookworm, control of trichuriasis in endemic areas consists of regular mass anthelminthic drug administration, primarily to preschool- and school-aged children, although single doses of albendazole or mebendazole are poorly effective for this intestinal nematode.

Enterobiasis The Pathogen  

E. vermicularis, or pinworm, is transmitted by the fecal-oral route. Embryonated eggs on fingernails, bedding, or clothing are ingested and hatch in the small intestine, where larvae develop into adult worms measuring between 2 and 5 mm. Adults migrate to the large intestine, where they mate. Gravid female worms emerge nightly from the anus to deposit large numbers of eggs (11,000 per worm) on the perianal and perineal skin, where they rapidly embryonate within 6 hours. If they are still on the skin, infective larvae are released that can migrate back through the anus into the rectum (retroinfection); alternatively, autoinfection occurs when eggs are transferred to the mouth via scratching skin on which eggs have been deposited, commonly in children. In infected females, larvae may also migrate into the genital tract and establish an ectopic infection.  

EPIDEMIOLOGY

E. vermicularis is found worldwide and is the most prevalent nematode infection in temperate climes. Transmission is especially frequent in primary schools and daycare centers where children are in close contact.  

CLINICAL MANIFESTATIONS

Although pinworm infections may be asymptomatic, perianal pruritus is the most common symptom and is caused by an allergic response to worm proteins. The pruritus can be intense and result in chronic sleep deprivation. Rarely, adult E. vermicularis may precipitate appendicitis. When hatched larvae migrate into the female genital tract, vulvovaginitis, salpingitis, or peritonitis may develop.  

DIAGNOSIS

Pinworm infection is diagnosed by identifying eggs through microscopic examination of a piece of cellophane tape applied to the perianal region

S. stercoralis, or threadworm, is endemic in warm climates worldwide, including parts of the United States. Infection occurs when exposed skin comes in contact with free-living filariform larvae in soil contaminated with human feces. Similar to hookworm, larvae penetrate the skin, enter the vasculature, and migrate to the pulmonary capillaries, where they penetrate into the alveoli, ascend the bronchial tree to the pharynx, and are swallowed into the gastrointestinal tract. Further development into adult worms occurs in the upper small intestine, where parasites live embedded in the mucosa. Unlike most nematodes, S. stercoralis reproduces by parthenogenesis, with no apparent parasitic male worm present in the human host. Female worms begin laying eggs within 25 to 30 days after infection. The embryonated eggs hatch rapidly in the intestinal lumen and release non-infectious rhabditiform larvae that migrate to the colon and are excreted in feces. Alternatively, larvae may directly penetrate the colonic mucosa or perianal skin after migrating out of the anus and enter the circulation directly, a mechanism known as autoinfection. This phenomenon can lead to maintenance of parasitism in the host for decades. Infectious filariform larvae develop in the soil by direct transformation from rhabditiform larvae or indirectly from eggs produced by free-living adult worms that have developed from rhabditiform larvae in warm, moist, sandy soil. Swollen belly syndrome, a less common type of strongyloidiasis seen in infants living in central Africa and Papua New Guinea, has been attributed to infection with Strongyloides fuelleborni, normally a zoonosis of nonhuman primates.  

EPIDEMIOLOGY

S. stercoralis infection is endemic in the tropical and subtropical regions of sub-Saharan Africa, Asia, Latin America, and areas of eastern and southern Europe, with a worldwide prevalence of up to 100 million.6 In the United States, infection is diagnosed most frequently in immigrants, commonly from Southeast Asia, although strongyloidiasis is still endemic in parts of rural Appalachia. S. stercoralis can also be transmitted sexually through oralanal contact, most often among men who have sex with men. Transmission via transplantation of solid organs from infected donors has also been reported.  

PATHOBIOLOGY

In the immunologically competent, infection does not usually result in symptomatic disease. Serious complications of infection, however, may occur in individuals with cell-mediated immunodeficiency such as those chronically taking corticosteroids, solid organ transplant recipients, patients with Hodgkin disease and other lymphomas, leukemic patients, and those infected with human T-cell lymphotropic virus type 1. In these patients, the S. stercoralis autoinfection cycle can become amplified and lead to a hyperinfection syndrome with a large increase in the total worm burden in the infected person. Hyperinfection can lead to life-threatening dissemination of larvae and adult worms to aberrant sites such as the brain, pancreas, and kidneys. For unknown reasons, acquired immunodeficiency syndrome has not been associated with hyperinfection syndrome or with disseminated strongyloidiasis. Disseminated strongyloidiasis is frequently accompanied by bacterial sepsis due to translocation of enteric organisms carried by migrating larvae.

2132



CHAPTER 335  Nematode Infections  



TREATMENT 

CLINICAL MANIFESTATIONS

Most S. stercoralis infections, especially in immunocompetent hosts, are asymptomatic or are associated with only mild gastrointestinal manifestations such as abdominal pain, bloating, and watery diarrhea. Gastrointestinal bleeding, manifested by hematochezia or melena, occurs in less than 20% of those infected. Rare causes of morbidity include small bowel obstruction, paralytic ileus, and a malabsorption syndrome (especially in children). During the migratory phase of larvae through the lungs, symptoms are rare in immunocompetent patients, although there may be peripheral eosinophilia. However, pulmonary signs and symptoms in immunocompromised persons with hyperinfection syndrome can be severe and resemble those of adult respiratory distress syndrome with acute onset of dyspnea, productive cough, and hemoptysis accompanied by fever, tachypnea, and hypoxemia. Migration of filariform larvae from the anus can lead to a dermatologic manifestation known as larva currens, which is characterized by migratory serpiginous erythematous maculopapular tracks, primarily on the skin of the buttocks, groin, and lower abdomen. Autoinfection leading to exceptionally high worm loads (i.e., hyperinfection) and disseminated strongyloidiasis can occur in persons with deficient cell-mediated immunity. Because asymptomatic infection with S. stercoralis may persist for decades after initial infection, it is important to remember that a change in immune status associated with conditions such as the administration of immunosuppressive drugs following solid organ transplantation may result in hyperinfection syndrome even though the infection was previously asymptomatic. Massive increases in the number of Strongyloides larvae because of hyperinfection can present as acute enteritis with severe diarrhea and ulcerative disease of the small and large intestine. During disseminated infection, larvae and sometimes adult worms penetrate the intestinal mucosa, migrate to aberrant sites, including the central nervous system, and result in metastatic abscesses and gram-negative meningitis due to enteric bacteria being carried by the migrating parasites. Less common complications of disseminated disease include glomerulonephritis and minimal-change nephrotic syndrome, acute respiratory distress syndrome, and alveolar hemorrhage.7 Mortality from hyperinfection and disseminated disease can be high, although early diagnosis and prompt initiation of treatment are associated with improved outcomes. Infants with swollen belly syndrome caused by S. fuelleborni are often seen acutely with abdominal ascites that is not accompanied by diarrhea or fever. The ascites is due to gastrointestinal protein loss; it can be significant enough to cause respiratory impairment and is associated with a high rate of mortality.

DIAGNOSIS

Definitive diagnosis of S. stercoralis infection relies on microscopic identification of larvae in feces or other fluids (such as sputum) or tissues. Intestinal strongyloidiasis can be diagnosed by identification of larvae in direct smears of freshly passed stool, although the sensitivity of a single fecal sample examination is as low as 30%. Sensitivity can be increased by examining multiple fecal specimens, by using concentration techniques, and by plating feces on an agar plate and inspecting for tracks of colonies created by bacteria being dragged by migrating larvae. Hyperinfection syndrome and disseminated strongyloidiasis can be diagnosed by detection of filariform larvae in duodenal fluid obtained by endoscopy, in sputum, or in bronchoalveolar lavage specimens. Larvae have also been recovered from cerebrospinal fluid, urine, peritoneal washings, skin, and the brains of immunocompromised persons. Fluctuating eosinophilia is common with uncomplicated intestinal strongyloidiasis, especially during the pulmonary migration phase of initial infection. However, eosinophilia may be absent in patients with hyperinfection and dissemination. In fact, those with hyperinfection and eosinophilia have a better prognosis than do those without eosinophilia. Serologic diagnosis using an enzyme-linked immunosorbent assay (ELISA) that detects antibodies to filariform larvae is very sensitive, even in immunocompromised hosts with disseminated strongyloidiasis, although false-positive results may occur in cases of coinfection with other nematodes, particularly filaria. Specificity is improved by using the newer luciferase immunoprecipitation system (LIPS) assays that incorporate Strongyloides-specific recombinant antigens. LIPS assays have the additional advantage of rapid reversion to seronegative after treatment compared with the slow decline in ELISA titers.

Uncomplicated intestinal strongyloidiasis can be treated effectively with ivermectin (200 µg/kg body weight daily for 2 days), with cure rates, defined as the absence of larvae by parasitologic methods 1 year after treatment, exceeding 90%. A5  Albendazole is an alternative treatment (see Table 335-1).8 Decreases in antibody titer and eosinophil count indicate a treatment response in the absence of continued exposure. After 6 months, ELISA titers should decrease significantly, whereas LIPS assays should revert to negative. However, more recently reported long-term follow-up studies for up to 4 years after ivermectin treatment showed that larvae were detected again in 14 of 21 patients, observed as early as 30 days after treatment, and S. stercoralis DNA was detected in all patients, both in their 30-day posttreatment and subsequent stool samples. These data suggest that long-term parasitologic cure is unlikely, and strongyloidiasis must be considered a chronic infection for which ivermectin administration schedules should be re-evaluated.9 In immunocompromised patients with hyperinfection or disseminated disease, daily treatment with ivermectin should be extended. Some experts recommend continuing treatment until 2 weeks after fecal examinations have become negative (i.e., for one autoinfection cycle). For severely ill patients who are unable to tolerate oral therapy, parenteral veterinary and enema ivermectin preparations have been used. Combination therapy with ivermectin and albendazole may also be used to treat disseminated strongyloidiasis, but data on whether this improves prognosis over monotherapy are lacking.



PREVENTION

In endemic areas, the risk for infection can be reduced by minimizing skin contact with contaminated soil, although elimination of this infection will occur only with improvements in sanitation and treatment of human waste. To prevent hyperinfection in individuals already infected, diagnosis should be attempted before the onset of immunosuppression if possible, such as before organ transplantation or cancer chemotherapy. Anyone who has resided in or traveled to an endemic area should undergo screening for asymptomatic infection, preferably by serology or, if not possible, by microscopic examination of at least three fecal samples for the presence of larvae. Patients with positive screening test results should be treated empirically with ivermectin. Individuals with negative screening test results but unexplained eosinophilia and a history of exposure should also be considered for empirical treatment. In those undergoing hematopoietic stem cell transplantation, documentation of response with at least three consecutive negative fecal examinations or a negative LIPS assays is recommended before proceeding with transplantation.

  UNCOMMON INTESTINAL NEMATODIASES

Humans may serve as incidental hosts for some nematodes that ordinarily parasitize the intestines of other mammals.  

Trichostrongylus

Human infections with various species of the genus Trichostrongylus have been reported in Iran, the Far East, and Australia. Humans become infected when larvae are ingested with leafy vegetables that have been contaminated with soil containing the feces of herbivorous animals. Trichostrongylus worms are similar to hookworms in their morphology, appearance of their eggs on fecal examination, and the pathology that they induce. Heavy infections may be accompanied by diarrhea and anemia. Drugs recommended for treatment are pyrantel pamoate, albendazole, or mebendazole (see Table 335-1).  

Anisakiasis

Anisakiasis results from ingestion of the larvae of nematodes that normally infect sea mammals such as dolphins, whales, and seals. Larvae of the genera Anisakis, Phocanema, and Pseudoterranova infect the flesh of a number of saltwater fish species as intermediate hosts. Consumption of raw or undercooked fish, often in the form of sushi or sashimi, results in release of the infective larvae into the stomach, followed by invasion of the stomach or duodenal wall, which causes upper abdominal pain that can be intense. Anisakid worms cannot further develop in humans and die within a few days; an eosinophilic granulomatous reaction may result that mimics a gastric tumor. Diagnosis and treatment are accomplished by endoscopic removal of the parasite. Infection is prevented by cooking or freezing seafood before consumption. Of note, salting, smoking, and marinating fish do not kill anisakid larvae.

CHAPTER 335  Nematode Infections  



Capillaria philippinensis

C. philippinensis can cause a serious intestinal infection that has been reported primarily in the Philippines and Thailand, although it has also been observed in Japan, Taiwan, Korea, and Egypt. Adult worms resemble those of Trichinella spiralis, although biologically they mimic S. stercoralis in that they have an autoinfectious cycle of reproduction in which larvae can develop into adult worms without leaving the host. Even though the life cycle has not been completely elucidated, this nematode probably parasitizes waterfowl that feed on fish and crustaceans, which serve as intermediate hosts. Humans become infected by eating raw or undercooked infected shrimp or fish. Adult worms travel to the mucosal crypts of the small intestine, where they deposit larvae, sometimes resulting in overwhelming infection. Clinical disease consists of severe diarrhea associated with anorexia, vomiting, and weight loss. Mortality rates as high as 10% have been reported, with death resulting from severe malabsorption and protein-losing enteropathy. Diagnosis depends on visualizing eggs or larvae in feces. The treatment of choice is albendazole or mebendazole (see Table 335-1).

  TISSUE NEMATODES

The tissue nematodes can be divided into those for which humans are the principal host (the filariases) and those that usually infect animals but can incidentally infect humans. Several zoonotic nematodes, such as Toxocara, Trichinella, and Angiostrongylus, infect humans following accidental oral ingestion of eggs or larvae but are unable to complete their life cycle in the host. Clinical manifestations are primarily due to the aberrant migration of larvae through various tissues.  

Toxocariasis  

DEFINITION

Accidental ingestion of embryonated eggs of the dog roundworm Toxocara canis, or less frequently of the cat ascarid Toxocara cati, can lead to the clinical syndromes of visceral larva migrans and ocular larva migrans. Symptoms are caused by the migration of larvae through the organs of the body, resulting in serious disease and even death.  

EPIDEMIOLOGY

Toxocara infections in animals are ubiquitous throughout the world. In humans, children are most frequently infected, probably by exposure to soil contaminated with dog or cat feces when playing outside. Visceral larva migrans occurs most commonly in children younger than 5 years, whereas ocular larva migrans typically affects older children between the ages of 5 and 10 years.  

PATHOBIOLOGY

The life cycle of Toxocara in the animal host resembles that of A. lumbricoides in humans; larvae penetrate the intestinal wall after being released from ingested eggs, migrate through the vasculature to the lungs, enter into the alveolar space, and ascend the bronchial tree until they are swallowed back into the gastrointestinal tract, where they develop into adult worms that can produce eggs. However, when embryonated Toxocara eggs are ingested by humans, the released larvae migrate throughout the body (most commonly to the lungs, the liver, the central nervous system [CNS], and occasionally the eyes) but cannot develop into adult worms. Ultimately, the larvae die, inducing significant immediate-type and delayed-type hypersensitivity reactions that result in eosinophilic granuloma formation. Visceral larva migrans and ocular larva migrans seem to be mutually exclusive, suggesting that different Toxocara strains may have different tissue tropisms. Alternatively, visceral larva migrans may result from repeated infections, whereas ocular larva migrans may be a manifestation of infection in children who have not been previously sensitized.  

CLINICAL MANIFESTATIONS

Most Toxocara infections in humans are asymptomatic. Visceral larva migrans is characterized by low-grade fever, pulmonary symptoms including cough and wheeze, and less frequently hepatosplenomegaly accompanied by right upper quadrant pain.10 Symptoms appear gradually and resolve over 4 to 8 weeks. Myocarditis, nephritis, and CNS disease are less common. CNS involvement can result in seizures, encephalopathy, neuropsychiatric symptoms, or an eosinophilic meningoencephalitis. Ocular larva migrans typically presents as unilateral visual impairment that is sometimes accompanied by strabismus. The degree of vision loss is dependent on the particular ocular structure involved, and permanent blindness

2133

can occur. Ocular larva migrans involving the retina can be difficult to distinguish from other causes of focal intraretinal lesions, such as retinoblastoma or tuberculosis.  

DIAGNOSIS

Toxocariasis can be presumed on the basis of a compatible clinical presentation and history of exposure to dogs or cats. Eosinophilia and hypergammaglobulinemia are often present. Serologic testing using an ELISA in combination with immunoblot to detect anti-Toxocara antibodies may be informative, although this does not distinguish between active infection and past exposure. Recent development of ELISAs incorporating recombinant T. canis antigens has improved assay sensitivity and specificity. Biopsy of tissues to document the presence of larvae is not recommended because of low sensitivity. Computed tomography and fluorescein angiography may be helpful in the diagnosis of ocular larva migrans, especially to differentiate it from retinoblastoma and other causes of intraocular space-occupying lesions. Elevated levels of anti-Toxocara antibodies in aqueous and vitreous humor fluid relative to serum are also suggestive of this diagnosis.

TREATMENT  Albendazole (400 mg given twice daily for 5 days) is the treatment of choice for acute toxocariasis (Table 335-2). Mebendazole is not recommended because of its poor oral bioavailability. In patients with severe pulmonary, cardiac, or neurologic involvement, corticosteroids may reduce the severity and duration of symptoms. Ocular larva migrans is treated by vitrectomy, corticosteroids, albendazole, or a combination of these. See also Chapter 323.



PREVENTION

Visceral larva migrans and ocular larva migrans may be prevented by periodic anthelminthic treatment of dogs and cats, proper disposal of pet feces, covering sandboxes, washing hands after playing with dogs or cats, and keeping children from playing in areas where pets have defecated.

TABLE 335-2 TREATMENT OF TISSUE NEMATODE INFECTIONS NEMATODE INFECTION

TREATMENT

Toxocariasis

Albendazole, 400 mg bid for 5 days

Trichinellosis

Albendazole, 400 mg bid for 8-14 days*

Angiostrongyliasis

Treatment with albendazole or mebendazole is controversial but may relieve symptoms.

Gnathostomiasis

Albendazole, 400 mg daily for 3 wk Alternative: ivermectin, 200 µg/kg/day for 2 days +/− Surgical removal

Lymphatic filariasis

Diethylcarbamazine, 6 mg/kg/day divided in three doses for 12 days†, plus doxycycline, 100-200 mg/day for 6 wk

Onchocerciasis

Ivermectin, 150 µg/kg once (alternative: moxidectin 8 mg once), repeated every 6-12 mo until resolution of symptoms, plus doxycycline, 200 mg/day for 6 wk‡

Loiasis

Diethylcarbamazine, 8-10 mg/kg/day divided in three doses for 21 days†‡

Mansonella perstans

Doxycycline, 200 mg/day for 6 wk

Mansonella ozzardi

Ivermectin, 200 µg/kg once

Mansonella streptocerca

Diethylcarbamazine, 6 mg/kg/day divided in three doses for 12 days Alternative: ivermectin, 150 µg/kg once

Dracunculiasis

Extraction of the adult worm

*Treatment is effective only if it is initiated during the intestinal phase of infection. † Start at a dose of 50 mg on the first day, 50 mg three times/day on the second, 100 mg three times/ day on the third, and then 8-10 mg/kg/day on day 4 onward. ‡ Repeated treatment after 6 months is often necessary if symptoms and eosinophilia persist.

2134  

CHAPTER 335  Nematode Infections  

Baylisascariasis

Baylisascariasis is a rare zoonosis caused by infection with Baylisascaris procyonis, an ascarid parasite of raccoons and other small carnivores. In North America, infection is most commonly associated with contact with raccoons or environments contaminated with their feces and occurs predominantly in infants and young children who ingest the embryonated eggs while playing with soil. Clinically, disease is manifested as neural larva migrans due to larval invasion of the CNS after release from ingested eggs in the gastrointestinal tract. Characteristic findings include fever, altered mental status, focal neurologic deficits, and seizures. Examination of the cerebrospinal fluid (CSF) reveals eosinophilic meningitis. B. procyonis has also been associated with ocular larva migrans. Infection can be fatal or result in permanent neurologic or visual impairment. When neural larva migrans is present, response to treatment with anthelmintics is poor, although corticosteroids may be helpful. Successful use of laser photocoagulation in ocular larva migrans has been reported. Prophylactic albendazole (25 mg/kg/day for 20 days) started within days of an exposure may prevent clinical disease.  

Trichinellosis  

DEFINITION

Trichinella infects a range of mammalian hosts, with the domestic pig serving as the most important reservoir worldwide. Humans are infected through eating raw or undercooked pork or other meats of domestic or wild animals that are contaminated with larvae that are encysted in muscle tissue.11 Although larvae develop into adults in the human intestinal tract, mate, and produce offspring larvae, clinical disease is characterized not so much by the intestinal infection as by the newborn larvae that penetrate the intestinal wall and disseminate throughout the body.  

EPIDEMIOLOGY

Several different species of Trichinella can cause disease in humans, although Trichinella spiralis is the most common. T. spiralis is enzootic throughout the world in omnivorous and carnivorous wild animals, including bears, boars, and rats. Trichinella nativa affects predominantly carnivores (e.g., walruses, polar bears, and seals) living in the Arctic and subarctic regions of North America, Europe, and Asia. Trichinella is introduced into domestic animal populations, usually pigs or horses, by feeding them unprocessed meat scraps of infected animals, most commonly rats. Because of regulations banning this practice in the United States, Canada, and the European Union, human infection by consumption of undercooked or smoked pork products or beef contaminated with the encysted larvae has been virtually eliminated,12 although it still occurs throughout the rest of the world. Instead, ingestion of poorly cooked wild game, especially bear or boar meat, is now the most common source of infection in these places. An important source of infection with T. nativa in Alaskan and Canadian Arctic native populations is eating of uncooked walrus meat.  

PATHOBIOLOGY

Trichinellosis, also called trichinosis, results from ingestion of striated muscle containing encysted infective larvae. Larvae are released from muscle tissue by digestive enzymes in the stomach and then migrate to the upper small intestine, where they rapidly develop into sexually mature adult worms after only 2 days. Adults live embedded in the columnar epithelium, where they grow to a length of 3 mm (females) or 1.5 mm (males). Females begin producing newborn larvae within 5 days of mating. Adult worms remain viable for an additional 3 to 5 weeks, after which acquired immunity develops that leads to their expulsion from the host. Newborn larvae possess a swordlike stylet in their oral cavity that permits them to penetrate the lamina propria and enter the lymphatic and blood vessels of the host, allowing them to migrate throughout the body. Larvae enter all types of cells, where they usually die, with the exception of striated skeletal and cardiac muscle cells. Unique among nematodes, mature Trichinella larvae have an intracellular phase, developing and transforming muscle cells into “nurse cells” that support larval growth and development (Fig. 335-4). In nurse cells, Trichinella larvae can survive for decades. Although nurse cells do not result in any disease in most mammals, they can induce an eosinophilic granulomatous reaction in humans that may result in significant tissue damage and dysfunction.  

CLINICAL MANIFESTATIONS

Clinical disease in humans can be divided into an initial intestinal phase followed by a systemic or muscle phase. The initial phase of infection that occurs within days after ingestion of larvae may be associated with mild diarrhea,

FIGURE 335-4.  Nurse cell in muscle tissue containing a larva of Trichinella spiralis. (Courtesy of Dr. I. Kagan, Centers for Disease Control and Prevention.)

abdominal pain, and vomiting. This phase is self-limited and usually resolves spontaneously within 10 days. The systemic dissemination of Trichinella larvae can result in myocardial, pulmonary, and focal neurologic manifestations, although usually only in the most heavily infected persons. This systemic phase of infection usually begins 2 to 3 weeks after ingestion of infective larvae and may persist for several weeks. Clinical manifestations typically include fever, periorbital or facial edema, a diffuse inflammatory myositis (Chapter 253) that is characterized by myalgias and muscle tenderness, and petechial hemorrhages most easily observed in the subungual skin and conjunctivae. Larval invasion of the myocardium13 can lead to myocarditis (Chapter 54) that may result in heart failure or arrhythmias. As with most nematodes, the severity of symptoms is related to the total worm burden. Because adult worms are incapable of reproducing within the host, the number of encysted larvae ingested is the most important determinant of the number of larvae that invade muscle and other tissues.  

DIAGNOSIS

A diagnosis of trichinellosis should be suspected in individuals with a compatible clinical presentation, a history of eating raw or undercooked meat, eosinophilia, and increased muscle enzymes such as creatine kinase and lactate dehydrogenase. Definitive diagnosis depends on visualization of nurse cells in a muscle biopsy specimen or detection of Trichinella-specific DNA by the polymerase chain reaction (PCR) technique, although this tool is not widely available. Findings on muscle biopsy may be normal even in heavily infected patients because of sampling error. Detection of anti-Trichinella antibodies can be very useful in making a diagnosis; ELISA is the most commonly used method. Antibodies can be detected as early as 12 days after initial infection.

TREATMENT  During the intestinal phase of infection, albendazole is recommended at a dosage of 400 mg given twice daily for 8 to 14 days to kill the adult worms and prevent release of newborn larvae (see Table 335-2). Although it is not known if albendazole is effective against newborn larvae, administration of this drug during the systemic phase of infection could potentially worsen symptoms by exacerbating the host inflammatory response to dying larvae. Treatment of severe systemic disease, including myocarditis and neurologic disease, should be directed toward reducing inflammation, most commonly with corticosteroids, although albendazole should also be given in such cases as corticosteroids may delay expulsion of adult worms from the intestine thus increasing the number of newborn larvae that may be released. Symptomatic treatment with antipyretics and analgesics should also be considered. See also Chapter 323.



PREVENTION

Trichinella infection is prevented by thoroughly cooking meat to kill encysted larvae. Freezing meat solid at −20° C for at least 3 days will kill T. spiralis but not all other species of Trichinella. Of note, curing and smoking techniques do not reliably kill this nematode.

CHAPTER 335  Nematode Infections  



Angiostrongyliasis  

DEFINITION

Angiostrongylus cantonensis and Angiostrongylus costaricensis are nematodes that normally infect rodents, primarily rats. The adult worms of A. cantonensis, or rat lungworm, inhabit the pulmonary arteries of rodents; larvae are produced that migrate to the pharynx, are swallowed, and then are passed in the feces. Mollusks such as snails, slugs, and prawns serve as intermediate hosts until they are ingested by definitive hosts. Released larvae migrate to the brain, where they develop into immature adult worms before traveling to the pulmonary vasculature to become sexually mature adults. Humans are incidentally infected after eating raw or poorly cooked mollusks or fresh vegetables contaminated with parts of infected mollusks; larvae can migrate to the CNS but cannot develop further. As opposed to A. cantonensis, the larvae of A. costaricensis can develop into sexually mature adult worms in the local lymphatics and mesenteric arterioles of humans and release eggs and larvae into the intestinal tissue, causing an intense eosinophilic granulomatous reaction.  

EPIDEMIOLOGY

Human infections with A. cantonensis occur mainly in Southeast Asia and the South Pacific and less frequently in Brazil, the Caribbean, and Louisiana.14 Abdominal angiostrongyliasis due to A. costaricensis has been reported mainly in Latin America, mostly in young children.  

CLINICAL MANIFESTATIONS

After ingestion, A. cantonensis larvae penetrate the intestinal wall and migrate to the brain, the meninges, and less commonly the spinal cord and eye. Fever, severe headache, meningismus, nausea, vomiting, seizures, and focal neurologic deficits may develop. A. costaricensis infection can mimic appendicitis with right-sided abdominal pain, vomiting, and fever. Less frequently, gastrointestinal bleeding may occur.  

DIAGNOSIS

Diagnosis of A. cantonensis infection is based on a history of ingestion of potentially contaminated food, the presence of peripheral eosinophilia, and detection of eosinophils and rarely larvae in the CSF. In neither CNS nor intestinal angiostrongyliasis are larvae or eggs found in the feces, although both may be seen in tissue specimens for A. costaricensis. Serology is not commercially available.

TREATMENT AND PREVENTION  Most patients infected with either species of Angiostrongylus recover completely after approximately 2 weeks. The use of anthelmintics (Chapter 323) is controversial, with only a few reports of benefit with albendazole or mebendazole, usually administered in combination with analgesics and corticosteroids to relieve symptoms. Serial lumbar punctures to remove CSF can relieve symptoms of raised intracranial pressure caused by infection with A. cantonensis. Proper cooking of food and washing of vegetables can prevent this infection.





Filariases  

DEFINITION AND PATHOBIOLOGY

The filariases are arthropod-borne nematode infections that are endemic mostly in tropical areas of the world, in which mature adult worms live in the lymphatics or in connective tissue (Table 335-3). Eight filarial species infect humans: Wuchereria bancrofti, Brugia malayi, Onchocerca volvulus, Brugia timori, Loa loa, Mansonella streptocerca, Mansonella perstans, and Mansonella ozzardi. The first three are the most common filariases worldwide. Although not usually fatal, these infections can result in significant disability and disfigurement, such as irreversible limb lymphedema (W. bancrofti and B. malayi) or blindness (O. volvulus). Most of the filariases require prolonged exposure for disease to be manifested and are therefore uncommon in short-term travelers to endemic areas. For all of the filarial nematodes, infection begins with the bite of an infected arthropod vector that deposits infective larvae called microfilariae into the skin or blood. During several months, microfilariae mature into adult worms capable of mating to produce microfilariae that can be ingested by another arthropod vector to complete the life cycle. Adult worms can survive for 5 to 17 years in the human host; microfilariae live for between 5 months and 5 years. For most of the filarial nematodes except B. malayi and M. perstans, humans are the only definitive host.  

CLINICAL MANIFESTATIONS

Clinical manifestations of infection are varied. Severity of disease is in most cases proportional to the worm burden harbored by an individual, with relatively light infections commonly being asymptomatic. For several of the filariases, the host inflammatory response to infection becomes apparent only on the death of the adult worms or microfilariae. This may be triggered by exposure to filarial antigens that were previously hidden from the immune system or by release of bacterial endosymbionts of the genus Wolbachia that live inside several of the filariae. Wolbachia are of the order Rickettsiales and are found in the hypodermis of adult worms and in oocysts, embryos, and microfilariae; they play a critical role in worm viability and fertility.  

DIAGNOSIS

Diagnosis of filarial infections usually depends on the microscopic examination of either blood or skin specimens for characteristic microfilariae (see Table 335-3). Microfilariae of the different filarial species measure between 170 and 320 µm in length and can be distinguished on the basis of the tissue source of the specimen, the presence or absence of a sheath, and the arrangement of nuclei in the tail. For some filarial species, microfilariae are present in the blood only during certain periods of the day to coincide with biting habits of the arthropod vector, which must be taken into account in timing blood collection for microscopy. Serology is not useful in endemic areas because a positive result does not distinguish between previous and current infection, and there is considerable antigenic cross-reactivity between the filariae and other nematodes. Detection of antifilarial antibodies may, however, be useful in returned long-term travelers or expatriates who are not originally from endemic areas.

TREATMENT  Diethylcarbamazine (DEC), ivermectin, and albendazole are the principal antifilarial drugs, although they have varying efficacies against the different filarial species (see Table 335-2). DEC is macrofilaricidal (active against the adult worm) for W. bancrofti, Brugia species, and L. loa, although prolonged or repeated courses are required for this effect. More commonly, the goal is to suppress microfilaria production by adult female worms, which can be achieved by single doses of antifilarial drugs administered alone or in combination annually or biannually. Reduction of microfilariae in the blood or skin can in some cases ameliorate symptoms or prevent progression of disease as well as interrupt transmission. Furthermore, targeting the Wolbachia endosymbionts of some filarial species with extended courses of antibiotics such as doxycycline can be macrofilaricidal.

Gnathostomiasis

Gnathostoma spinigerum is an intestinal nematode of dogs and cats; intermediate hosts include tiny crustaceans (copepods), amphibians, freshwater fish, and birds. Human infection occurs throughout the Far East, Thailand, and Latin America, particularly Mexico, through eating raw or undercooked invertebrate hosts harboring larvae. Larvae are released in the intestine and subsequently migrate through the body but are unable to reach sexual maturity in humans. The most common clinical presentation is migrating painful and pruritic subcutaneous swellings. Eosinophilic meningitis and ocular larva migrans may also occur, with potentially devastating results, including paralysis, subarachnoid hemorrhage, and permanent visual loss. Peripheral eosinophilia, often significant, is usually present; with meningitis, eosinophils are also present in the CSF. Although serologic testing is not available in the United States, laboratories in Thailand and Japan can perform it. Treatment of cutaneous disease with either a 3-week course of albendazole or a 2-day course of ivermectin is recommended (see Table 335-2); for neurologic or ocular involvement, anthelminthics are not advised because they may worsen manifestations. Gnathostomiasis may be prevented by thoroughly cooking fish.

2135



LYMPHATIC FILARIASIS DEFINITION  

The three etiologic agents of lymphatic filariasis, W. bancrofti, B. malayi, and B. timori, are transmitted to humans through the bite of an infected mosquito. Microfilariae deposited at the bite wound subsequently migrate through the subcutaneous tissue to the lymphatic system, where adult worms develop

2136

CHAPTER 335  Nematode Infections  

TABLE 335-3 FILARIAL PARASITES OF HUMANS MICROFILARIAE SPECIES

DISTRIBUTION

VECTOR

PRIMARY LOCATION

PERIODICITY

PRESENCE OF SHEATH

Wuchereria bancrofti

Tropics worldwide

Mosquitoes

Blood

Nocturnal, subperiodic

+

Brugia malayi

India, Southeast Asia

Mosquitoes

Blood

Nocturnal, subperiodic

+

Brugia timori

Indonesia

Mosquitoes

Blood

Nocturnal

+

Onchocerca volvulus

Africa, South America

Simulium blackflies

Skin, eye

None or minimal



Loa loa

West and Central Africa

Chrysops flies

Blood

Diurnal

+

Mansonella perstans

Africa, South America, Caribbean

Midges

Blood

None



Mansonella ozzardi

Central and South America, Caribbean

Midges, Simulium blackflies

Blood

None



Mansonella streptocerca

West and Central Africa

Midges

Skin

None



after approximately 4 to 12 months. The worms reside coiled in lymph nodes and may extend into afferent lymph vessels and surrounding subcutaneous tissue. The lymphatics of the lower and upper extremities and male genitalia are most commonly affected. After mating, females, which measure between 4 and 10 cm in length, twice the length of males, release more than 10,000 microfilariae a day that migrate into the blood stream until ingestion by biting mosquitoes. In most endemic areas, microfilariae are present in the peripheral blood only at night, when mosquito vectors are most likely to bite. Adult filariae live between 5 and 8 years within the host, although infections lasting for decades have been reported.  

EPIDEMIOLOGY

An estimated 70 million people are affected by lymphatic filariasis worldwide; most cases are caused by W. bancrofti, and less than 5 million are due to B. malayi. B. timori is of minor importance, being restricted to southeastern Indonesia. W. bancrofti is widely distributed in the tropics, especially in Southeast Asia, the Indian subcontinent, Africa, South America, the Caribbean, and the South Pacific. The major vectors of bancroftian filariasis are Culex mosquitoes in urban areas, anopheline mosquitoes in rural areas of Africa, and Aedes species in the Pacific. Humans are the only definitive host for W. bancrofti. B. malayi, however, can be zoonotic, with both monkey and feline species serving as reservoir hosts. Brugian filariasis is found primarily in India, Malaysia, and other areas in Southeast Asia.  

PATHOBIOLOGY

The pathology of filarial infections is primarily due to obstruction of the lymphatic circulation resulting from damage induced by adult worms, specifically a local inflammatory lymphangitis with components of the innate and adaptive immune response leading to vessel wall hypertrophy. This inflammatory response can be triggered by release of antigens from dead or dying worms, although evidence suggests that it is also induced by living worms and Wolbachia antigens that are excreted or secreted into the surrounding milieu. Inflammatory damage is also exacerbated by secondary bacterial and fungal infections. The initial inflammatory response leads to endothelial and connective tissue proliferation and vessel dilation, which impairs normal lymphatic function and results in lymphedema that is initially reversible. However, worm death results in a granulomatous reaction to released worm and Wolbachia antigens. The infiltration of giant cells as well as plasma cells, eosinophils, and neutrophils can completely occlude the lumen of the lymphatic vessel. Over time, progressive fibrosis and obstruction of lymph flow result in irreversible edema. Although recanalization and collateralization of lymph vessels may occur, lymphatic function remains compromised.  

CLINICAL MANIFESTATIONS

The clinical manifestations of lymphatic filariasis cover a wide spectrum from asymptomatic infection to severe chronic lymphatic obstruction accompanied by lymphedema and enlargement of the affected limb or body part (referred to as elephantiasis). Other common clinical outcomes include acute episodic lymphadenitis (also called filarial fever) and tropical pulmonary eosinophilia. Most infected individuals living in endemic regions are clinically asymptomatic, although microfilariae can be observed in their blood. Despite the absence of a significant inflammatory response, these individuals may nevertheless exhibit dilation of the affected lymphatics on ultrasound, which precedes the onset of clinically apparent disease.

FIGURE 335-5.  Elephantiasis, or chronic lymphedema due to infection with Wuchereria bancrofti. (Courtesy of the Centers for Disease Control and Prevention.)

For unknown reasons, newly exposed individuals may develop acute inflammatory reactions that can rapidly progress to chronic or irreversible changes compared with those born in endemic areas. Severe episodes of lymphadenitis, often with genital involvement, may lead to the relatively rapid development of lymphedema and elephantiasis within a year of presentation. Findings usually resolve quickly if the individual is promptly removed from the endemic area. Microfilariae are usually not detected in these patients.  

Acute Lymphadenitis

Acute episodes of retrograde lymphadenitis occur most commonly in adolescents in endemic areas, often in response to dying adult worms. Painful, erythematous enlargement of an affected lymph node, most commonly inguinal, precedes the onset of lymphangitis and is accompanied by fever and chills. Episodes usually last for about a week, frequently recur, and can be incapacitating. Defervescence is abrupt and associated with desquamation of the overlying skin. In men, inguinal lymphadenitis can be complicated by epididymitis and orchitis. Patients with filarial fevers may be microfilaremic but often are not.  

Elephantiasis

Repeated episodes of lymphadenitis eventually lead to dilation of the lymphatic vessels, resulting in chronic lymphedema over the course of months to years (Fig. 335-5). The extremities, breasts, and genitalia are most commonly affected, although with B. malayi infection, usually only the lower parts of the legs are involved. The edema is initially pitting in nature, but the subcutaneous tissue eventually loses its elasticity, resulting in woody edema with thickening of subcutaneous tissue and hyperkeratosis. Secondary bacterial or fungal infection contributes significantly to the chronic pathologic process of elephantiasis. In bancroftian filariasis, hydrocele is a common manifestation of chronic filariasis in men and can sometimes become massive and debilitating; lymphedema of the vulva is less commonly seen in women. Involvement of the retroperitoneal lymphatics can lead to their rupture to produce intermittent chyluria or chylocele.  

Tropical Pulmonary Eosinophilia

Tropical pulmonary eosinophilia (Chapter 161) develops in a small minority of individuals with filarial infections. The syndrome is most commonly seen

CHAPTER 335  Nematode Infections  

in young men living in southern India, although it also occurs in Pakistan, Sri Lanka, Southeast Asia, and Brazil. Characteristic clinical findings include nocturnal paroxysmal cough, wheeze, and low-grade fever that are accompanied by weight loss and eosinophilia. Levels of total immunoglobulin E (IgE) and antifilarial antibodies are typically high. Chest radiographs may show diffuse interstitial infiltrates or mottled opacities. Without treatment, chronic restrictive lung disease may develop.  

Interestingly, treatment directed against the Wolbachia endosymbiont has been shown to be effective in killing adult W. bancrofti and Brugia worms. Doxycycline, 100 or 200 mg daily for 4 to 8 weeks, reduces female worm fertility with a resulting suppression of microfilaremia for up to a year and reduces the number of live adult worms. Given the duration of treatment, these regimens are not ideal for disease control programs in endemic countries. DEC is highly effective in the treatment of tropical pulmonary eosinophilia. Treatment with a dose of 6 mg/kg/day for 14 to 21 days results in resolution of symptoms within a week, although relapse may occur even after an interval of years. See also Chapter 323.

DIAGNOSIS

Definitive diagnosis usually relies on microscopic examination of a Giemsastained blood smear for microfilariae. Although thick blood smears are relatively insensitive except in cases of high microfilaremia, concentration or filtering techniques can increase the diagnostic yield. Characteristic microfilariae are 250 to 320 µm in length. Collection of blood should be timed according to the known periodicity of the microfilariae. A rapid immunochromatographic card test is available for W. bancrofti (there is no equivalent test for Brugia infections) and has the advantage of not requiring nocturnal collection of blood because it detects circulating antigen of the adult worm and not microfilariae. PCR methods have been developed to detect filarial antigens in blood, but these are not widely available. Serologic detection of antifilarial antibodies is of limited value because of extensive antigenic cross-reactivity with other nematodes. Furthermore, actively infected individuals cannot be distinguished from those previously infected, and those merely exposed but not infected may also have positive serologic test results. Ultrasound examination of the lymphatic vessels of the spermatic cord of men can be used to visualize the “filarial dance sign,” which is pathognomonic for a nest of filarial parasites. Individuals with elephantiasis may be amicrofilaremic. Diagnosis therefore depends on a compatible clinical history and physical examination in the context of the appropriate epidemiology, and it may be supported by a positive antigen test or, in men, by a suggestive scrotal ultrasonogram. It should be distinguished from podoconiosis, a tropical lymphedema that results from long-term barefoot exposure to red-clay soil derived from volcanic rock and that may be a T-cell-mediated inflammatory disease.



Management of lymphatic filariasis differs according to whether the aim is disease control or cure of an individual patient. In endemic areas, annual mass drug administration with a combination of antifilarial drugs can reduce transmission by decreasing the number of microfilariae in the blood available to biting mosquitoes. These programs use different two-drug combinations of single-dose DEC, ivermectin, and albendazole, administered at least annually, although a recent study showed sustained reduction in microfilaremia for 3 years after single-dose coadministration of all three drugs. A6  DEC is administered with albendazole except in areas where onchocerciasis or loiasis is also endemic, in which case ivermectin plus albendazole is used. All individuals with active infection by a lymphatic filarial parasite, whether symptomatic or asymptomatic, should be treated with an antifilarial medication (see Table 335-2). The treatment of choice is DEC (6 mg/kg/day for 12 days). In the United States, DEC is available only through the Centers for Disease Control and Prevention (CDC) drug service (http://www.cdc.gov/laboratory/drugservice/ formulary.html). For patients with high levels of microfilariae in the blood, treatment can be started at a low dose of 50 mg daily and scaled up during the first 3 days to reduce side effects of treatment such as fever, headache, dizziness, nausea, vomiting, rash, myalgias, and arthralgias. These normally resolve after a few days of treatment and can be treated with antipyretics, antihistamines, and, if symptoms are severe, corticosteroids. DEC is both microfilaricidal and partially macrofilaricidal. In individuals who will not be returning to endemic areas, repeated treatments with DEC are often attempted to kill the adult worms instead of just reducing the levels of microfilariae in the blood. Typically, courses of DEC are repeated every 6 to 12 months. Although the adult worm burden is reduced in most treated individuals, all parasites are eliminated in less than one fourth. In men with live adult worms visible by ultrasound in the scrotal lymphatics, serial studies may be performed to monitor the effects of therapy. Unfortunately, lymphedema due to lymphatic filariasis is usually not reversible with DEC treatment, except in the very early stages. Nevertheless, the chronic sequelae of lymphatic filariasis can be limited by prevention of secondary bacterial and fungal infection through meticulous hygiene and prompt treatment of suspected infections with antimicrobials. Limb elevation, physiotherapy, and use of elastic stockings may slow the worsening of lymphedema. Surgery is usually not indicated except for cases of hydrocele.

PREVENTION

Annual mass treatment with single doses of two antifilarial drugs can significantly reduce the prevalence of infection within a community. In some areas, DEC-fortified table salt has been used to reduce the levels of microfilaremia in affected communities to interrupt transmission. Vector control through the use of insecticide-treated bed nets and residual indoor spraying of insecticides may have some efficacy.  

ONCHOCERCIASIS DEFINITION  

Onchocerciasis, or river blindness, caused by the nematode O. volvulus, is transmitted to humans by Simulium blackflies. Infective microfilariae develop into male and female adult worms over a period of several months and live for 9 to 14 years coiled within subcutaneous fibrous nodules (onchocercomas). Adult females measure between 20 and 70 cm in length and remain confined to the nodules; males are only 3 to 5 cm long and freely migrate through the subcutaneous tissues between nodules to inseminate females. Mature female worms produce up to 1500 microfilariae per day, which leave the nodule to migrate primarily through the skin and ocular tissues. Microfilariae live within the host for 12 to 18 months.  

TREATMENT 

2137

EPIDEMIOLOGY

Onchocerciasis is endemic in equatorial Africa, with small foci in Yemen and an area on the border between Brazil and Venezuela. Approximately 18 million people are estimated to be infected, while 270,000 are blind due to this parasite. More than 99% of cases occur in sub-Saharan Africa, with Nigeria being the most highly endemic country. Because Simulium blackflies require fast-flowing, well-oxygenated water for egg laying and reproduction, cases are concentrated around streams and rivers, often in the most fertile farming areas. Blindness caused by O. volvulus results in significant morbidity, long-term disability, and reduced economic productivity. In addition, onchocerciasis has been associated with a reduced life expectancy of at least 10 years compared with uninfected individuals in the same area, an effect that appears to be independent of the blindness that develops.  

PATHOBIOLOGY

The pathologic changes of onchocerciasis are primarily due to an inflammatory reaction to microfilariae, mostly in the skin, eyes, and lymph nodes. Adult worms contained in nodules are relatively isolated from the host immune response. Tissue damage results from a cell-mediated immune response to dying microfilariae, which becomes more pronounced as infection persists. The degree of tissue damage is directly related to the intensity of infection as well as the magnitude of the host response. Sclerosing keratitis, the major cause of blindness, results from an inflammatory reaction to dying intraocular microfilariae that is dependent on helper T-cell type 2 (TH2) cytokines. With time, neovascularization and scarring of the cornea lead to corneal opacification and eventual blindness. In the skin, similar immune responses result in pruritus and angioedema. Ongoing low-grade inflammation in the skin eventually leads to loss of elasticity and atrophy. Chronic inflammatory changes and fibrosis are also seen in lymph nodes. Like the nematodes responsible for lymphatic filariasis, O. volvulus adult worms contain endosymbiotic Wolbachia bacteria that are obligatory for the development, survival, and fertility of these worms. Pro-inflammatory Wolbachia proteins released by dying microfilariae may be responsible for a significant part of the immunopathology associated with onchocerciasis. For example, Wolbachia antigens have been shown to interact

2138

CHAPTER 335  Nematode Infections  

with the innate immune system through a toll-like receptor 2–mediated mechanism.  

TREATMENT  Ivermectin or moxidectin are the treatments of choice for onchocerciasis (see Table 335-2). Administration of a single dose of ivermectin (150 µg/kg) or moxidectin (8 mg [age >12 yrs]) is effective in ameliorating ocular and dermatologic disease by destroying microfilariae and suppressing their release from female worms. Because neither ivermectin nor moxidectin is active against adult worms, treatment must be repeated every 6 to 12 months, probably for at least 10 years in those without further exposure. For unknown reasons, pruritus in lightly infected expatriates may require more aggressive and frequent treatment for the first 2 years. Within 24 hours of treatment, fever and pruritus may occur in reaction to the dying microfilariae or released Wolbachia antigens, especially in those with high pretreatment levels of microfilariae. Another potential agent is moxidectin (8 mg single oral dose), which appears to be as effective, or even more effective, compared with ivermectin. A7  Use of ivermectin or moxidectin in areas where L. loa (see later) is coendemic should be undertaken with caution because treatment may precipitate severe reactions including encephalopathy in those with high L. loa microfilaremia. The LoaScope is a mobile-telephone-based videomicroscope that, with the use of a smartphone coupled to a simple optical device, provides a rapid, pointof-contact, field-friendly, and accurate method to quantify L. loa microfilariae. It has enabled a test-and-not-treat strategy to identify individuals with high blood levels of circulating L. loa microfilariae who are at particular risk for serious adverse events and should be excluded from mass drug administration strategies for the elimination of lymphatic filariasis and onchocerciasis.16 DEC should never be used for treatment of onchocerciasis because of frequent serious reactions to dying microfilariae ranging from urticaria and angioedema to hypotension and death. Although the drug suramin (available from the U.S. CDC drug service) is active against adult O. volvulus worms, because of its excessive toxicity, it is used only in rare situations. Nodulectomy successfully resolves the infection in some cases. Doxycycline, 200 mg/day administered for 4 to 6 weeks, followed by singledose ivermectin, has been shown to deplete Wolbachia endosymbionts from adult worms, resulting in a significant macrofilaricidal effect and suppression of embryogenesis and microfilaria production for up to 18 months in the remaining adult worms. A8  Increasingly, this regimen is recommended for patients with onchocerciasis who have left an endemic area and will not be re-exposed. See also Chapter 323.

CLINICAL MANIFESTATIONS

Onchodermatitis

Onchocerciasis commonly presents with a diffuse papular dermatitis that is intensely pruritic.15 In heavily infected individuals in endemic areas, the pruritus is intractable, leading to scratching and excoriation to the point of bleeding and even suicide. Hypersensitivity reactions, scabies, insect bites, and atopic or contact dermatitis should be considered in the differential diagnosis of the acute papular dermatitis seen with onchocerciasis. The skin of affected areas becomes edematous and thickened, loses its elasticity, and takes on an orange-peel texture. A lichenified dermatitis (referred to as sowda) may occur; it consists of an intensely pruritic eruption limited to one extremity, usually a leg, with hyperpigmented papules and plaques accompanied by edema of the entire limb. Over time, the skin will atrophy and fine wrinkles appear, especially over the buttocks. Pruritus is uncommon at this point. Areas of depigmentation may occur most commonly over the shins, a phenomenon called leopard skin.

Subcutaneous Nodules

Subcutaneous onchocercomas containing adult worms are most often palpable over bony prominences. In Africa, the nodules are most commonly found over the hips and lower limbs; in South America, they are usually located on the head and upper part of the body. Nodules measure between 0.5 and 3 cm in diameter and are freely mobile. In lightly infected individuals, such as expatriates, nodules are usually not detectable.

Ocular Lesions

Initial ocular involvement is characterized by conjunctivitis, excess tearing, and photophobia in response to dying microfilariae. At this point in the course of disease, punctate keratitis or snowflake corneal opacities are present. During 20 to 30 years, this progresses to sclerosing keratitis, neovascularization, and corneal opacification. The anterior chamber of the eye may also be involved, with iritis, iridocyclitis, and secondary glaucoma. Posterior ocular disease can be manifested as chorioretinitis, optic neuritis, and optic atrophy.

Lymphadenopathy

Lymphadenopathy is frequently found in the inguinal and femoral areas in Africa and in the head and neck in South America. Advanced disease in the inguinal region can result in the so-called hanging groin, with elongated atrophic skin containing nontender and fibrotic lymph nodes.  

DIAGNOSIS

Definitive diagnosis has traditionally been made by observing unsheathed motile microfilariae measuring 200 to 300 µm in length that are released from superficial skin snips. To take a skin snip, a thin piece of skin overlying a bone prominence that has been tented up with a needle is sliced with a scalpel blade, or a corneal-scleral punch instrument is used to obtain a small piece of skin. Avoidance of blood contamination is critical so as to avoid confusion with blood-borne microfilariae in cases in which patients are coinfected with other filariases. Typically, six snips are taken, one from over each scapula, iliac crest, and lateral aspect of each calf, and then incubated with warm physiologic saline and examined microscopically for motile microfilariae after incubating for up to 24 hours. PCR amplification of filarial DNA directly from skin snips is far more sensitive than direct visualization but is not widely available. With ocular disease, free microfilariae may be visible by slit lamp examination in the anterior chamber or aqueous humor. Subcutaneous nodules can be sampled or examined by ultrasound to demonstrate the presence of adult worms. Serologic tests are usually positive for antifilarial antibodies but are not specific because of considerable antigenic cross-reactivity with other nematodes. Eosinophilia is a common but inconsistent finding. In the past, the Mazzotti test was used to diagnose onchocerciasis. In this test, a challenge dose of DEC was administered to patients suspected of having onchocerciasis; with O. volvulus infection, an intense pruritic skin reaction would develop within hours. However, in patients with high-intensity infections, the Mazzotti reaction could be severe and even worsen ocular disease, resulting in permanent visual loss. Therefore, this test is no longer recommended, although some instead suggest the application of a small amount of DEC-containing cream to the skin to provoke a localized Mazzotti reaction.



PREVENTION

Regular mass administration of ivermectin to affected communities forms the core of the global eradication strategy for onchocerciasis. Implementation of this program has been made easier because the drug is donated by the manufacturer. In addition to benefiting infected individuals, mass drug administration reduces the microfilariae available to vectors and thus interrupts the transmission cycle. For travelers to endemic areas, use of insect repellent may be beneficial.  

LOIASIS DEFINITION  

Loiasis is caused by infection with the filarial nematode L. loa, otherwise known as the African eye worm. L. loa are transmitted by flies of the genus Chrysops during a blood meal. Adult worms develop during a period of 1 to 4 years and live for up to 17 years. They migrate freely in the subcutaneous tissue, including the subconjunctiva or sclera of the eye. Adult females measure between 40 and 70 mm in length; males are shorter, measuring between 25 and 35 mm. After mating, females release microfilariae into the blood. L. loa microfilariae exhibit a diurnal periodicity coinciding with the feeding habits of Chrysops, with microfilaremia peaking around midday.  

EPIDEMIOLOGY

Loiasis is endemic in the rain forest regions of central and west Africa, with the highest prevalence in Gabon, Cameroon, the Democratic Republic of the Congo, Nigeria, and the Central African Republic.17 Loiasis requires a shorter period of exposure than other filarial infections and can be seen in returning travelers or expatriates who have spent extended periods in Africa.  

PATHOBIOLOGY

Neither adult L. loa worms nor microfilariae have any direct pathologic effects. In a subset of infected individuals, a hypersensitivity response, termed a Calabar swelling, develops to secretions from adult worms or released microfilariae, resulting in recurrent localized angioedema that often precedes the migrating worm. These patients have high serum levels of IgE antibodies and eosinophilia. This reaction is more commonly observed in visitors to endemic areas than

CHAPTER 335  Nematode Infections  



2139

PREVENTION

Weekly chemoprophylaxis with DEC administered as a 300-mg dose is effective in preventing loiasis in long-term residents of endemic areas.  



FIGURE 335-6.  Adult Loa loa worm migrating across the eye (arrow).

in native residents. Unlike other filariae, L. loa does not contain Wolbachia endosymbionts.  

CLINICAL MANIFESTATIONS

Most individuals with loiasis are asymptomatic despite being microfilaremic. Clinical manifestations of infection are more common in long-term visitors to endemic areas than in people native to the regions. Recurrent Calabar swellings are the most common finding in these individuals, who are not usually microfilaremic. They are nonerythematous swellings measuring 5 to 20 cm in diameter that typically occur on the extremities and the face and last for a few days. The onset is often preceded by pruritus and pain. On occasion, adult worms may migrate across the subconjunctiva or sclera of the eye, causing severe pain and inflammation (Fig. 335-6). Rare complications of infection include nephropathy and encephalitis, which usually develop in those with high levels of microfilariae after receiving DEC or ivermectin treatment for other filarial infections. Endomyocardial fibrosis resulting from eosinophilic infiltration of the myocardium has been reported in association with loiasis.  

DIAGNOSIS

Diagnosis depends on microscopic examination of a Giemsa-stained blood film for characteristic sheathed microfilariae. Blood should be collected between 10 am and 2 pm because of the diurnal periodicity of the microfilariae. Because individuals who are not native to endemic areas are usually not microfilaremic, diagnosis relies on a compatible history, clinical findings, peripheral eosinophilia, and elevated antifilarial antibody levels. Adult worms can sometimes be surgically removed while migrating across the eye or through subcutaneous tissues. Calabar swellings must be distinguished from onchocercomas and other causes of angioedema.

TREATMENT  DEC (8 to 10 mg/kg/day for 21 days) is active against both adult L. loa worms and microfilariae (see Table 335-2). Treatment is usually increased from a dose of 50 mg/day on the first day to the full dose on the fourth day to minimize the likelihood of treatment-associated complications, the most serious of which are glomerulonephritis and a potentially fatal encephalopathy. Treatmentassociated complications are more common with high pretreatment microfilarial levels and result from host allergic reactions to dying microfilariae. Antihistamines and corticosteroids may be employed to reduce allergic side effects. Alternatively, apheresis can be used to remove circulating microfilariae before initiation of DEC. Albendazole, which is microfilaricidal but has no activity against adult worms, has been used to reduce microfilaria levels before treatment with DEC. Repeated courses of DEC may be necessary in about half of patients before clinical manifestations completely resolve. Persistent or increasing eosinophilia or levels of antifilarial antibodies 6 months after treatment should also prompt re-evaluation for repeated treatment. Adult worms in the eye may be surgically removed. Ivermectin is microfilaricidal but has no macrofilaricidal effect and may cause toxic encephalopathy in individuals with high microfilaria levels. In areas where onchocerciasis is coendemic, this infection should be ruled out and treated before initiation of DEC to prevent toxicity from dying O. volvulus microfilariae. See also Chapter 323.

LESS COMMON FILARIAL INFECTIONS

Mansonella perstans

M. perstans infection occurs throughout west and central Africa, in northeastern South America, and in parts of the Caribbean. Microfilariae are transmitted by Culicoides midges and develop into adult worms that live in serous body cavities, such as the pleural, pericardial, and peritoneal spaces, as well as in mesenteric and retroperitoneal tissues. Most infections are asymptomatic, although painless conjunctival nodules with eyelid edema have been reported. Transient angioedema and Calabar-like swellings, fever, headache, arthralgias, and neurologic manifestations may also occur. Microfilariae do not exhibit periodicity and can be observed on stained blood films. Eosinophilia is common. M. perstans harbor Wolbachia endosymbionts, and treatment with doxycycline, 200 mg/day for 6 weeks, has been shown to be highly effective in suppressing microfilaremia for up to 3 years, suggesting that the treatment is macrofilaricidal.  

Mansonella ozzardi

Infections with M. ozzardi occur in Central and South America and parts of the Caribbean, especially Haiti. Vectors include Simulium blackflies and midges. Adult worms locate to the peritoneal and thoracic cavities or the lymphatics; microfilariae circulate in the blood without periodicity. Infection usually results in asymptomatic eosinophilia, although arthritis and allergic symptoms such as urticaria and lymphadenopathy may occur in response to dying worms. Administration of ivermectin as a single dose of 200 µg/kg has been reported to provide long-term suppression of microfilaremia and improvement of symptoms. Neither DEC nor the benzimidazoles are effective.  

Mansonella streptocerca

M. streptocerca is endemic in the tropical forest zone of west and central Africa and is transmitted by biting midges. Similar to O. volvulus, adult worms live in the subcutaneous tissues, as do microfilariae. In contrast to onchocerciasis, microfilariae do not invade the eye. Infection is usually asymptomatic, although a pruritic dermatitis with depigmentation similar to onchodermatitis can affect the trunk and upper extremities. Associated axillary or inguinal adenopathy is common. Microfilariae have characteristic hooked tails and can be visualized in skin snips. In areas where onchocerciasis is coendemic, skin specimens must be stained to differentiate M. streptocerca from O. volvulus. DEC is microfilaricidal and macrofilaricidal and is given as 6 mg/kg/day for 12 days. Ivermectin is effective against microfilariae but not adult worms.  

Zoonotic Filarial Infections

A rare accidental filarial infection of humans with the dog heartworm Dirofilaria immitis occurs worldwide. Transmitted by mosquitoes, D. immitis microfilariae cannot reach maturity in humans but embolize to the lung after dying in the right ventricle. Most infections are asymptomatic, but some people experience cough, chest pain, and hemoptysis consistent with lung infarction. Chest radiographs demonstrate typical coin lesions that may be mistaken for carcinoma. Other animal filariae, including Dirofilaria repens of dogs and Dirofilaria tenuis of raccoons, can infect humans and result in subcutaneous nodules that may be migratory. Eosinophilia and antifilarial antibodies are not usually present in zoonotic filarial infections. Surgical removal of lesions is both diagnostic and curative.  

Dracunculiasis

Dracunculiasis is a disfiguring disease caused by the nematode Dracunculus medinensis, or Guinea worm. Although previously found in the Indian subcontinent and Latin America, it is now endemic in only four countries in sub-Saharan Africa (South Sudan, Chad, Mali, and Ethiopia) due to concerted eradication efforts. In 2017, fewer than 50 cases were identified. Transmission to humans occurs through ingestion of copepods, tiny crustacean intermediate hosts that harbor infective larvae. Released larvae penetrate the intestinal wall and migrate to the subcutaneous tissues, where they develop into adult worms. After approximately a year, female worms induce vesicular skin lesions, usually on the lower extremities, that eventually ulcerate. On direct contact with fresh water, the female worm releases thousands of motile larvae that can then complete the transmission cycle by infecting copepods in the water. Adult worms measure up to a meter in length. Fever and allergic symptoms,

including wheezing and urticaria, may precede rupture of the blister or occur upon attempted extraction of the worm. Secondary bacterial infection of the skin lesions is frequent. Although not commonly fatal, dracunculiasis can result in significant disability. Traditionally, emerging worms are extracted by slowly winding a few centimeters of the parasite on a stick each day, taking care not to break it. Surgical removal can also be attempted but may exacerbate allergic symptoms. There is no effective chemotherapy for this infection. Prevention efforts have been highly successful in interrupting transmission and have led to eradication of the parasite from many countries. Strategies include filtering of drinking water through finely woven cloth, education of infected individuals not to enter fresh water, treatment of water sources with larvicides, and provision of safe drinking water from wells.

  Grade A References A1. Moser W, Schindler C, Keiser J. Efficacy of recommended drugs against soil transmitted helminths: systematic review and network meta-analysis. BMJ. 2017;358:1-10. A2. Adegnika AA, Zinsou JF, Issifou S, et al. Randomized, controlled, assessor-blind clinical trial to assess the efficacy of single-versus repeated-dose albendazole to treat Ascaris lumbricoides, Trichuris trichiura, and hookworm infection. Antimicrob Agents Chemother. 2014;58:2535-2540. A3. Moser W, Coulibaly JT, Ali SM, et al. Efficacy and safety of tribendimidine, tribendimidine plus ivermectin, tribendimidine plus oxantel, pamoate, and albendazole plus oxantel pamoate against hookworm and concomitant soil-transmitted helminth infection in Tanzania and Côte d’Ivoire: a randomised, controlled, single-blinded, non-inferiority trial. Lancet Infect Dis. 2017;17:1162-1171. A4. Knopp S, Mohammed KA, Speich B, et al. Albendazole and mebendazole administered alone or in combination with ivermectin against Trichuris trichiura: a randomized controlled trial. Clin Infect Dis. 2010;51:1420-1428. A5. Suputtamongkol Y, Premasathian N, Bhumimuang K, et al. Efficacy and safety of single and double doses of ivermectin versus 7-day high dose albendazole for chronic strongyloidiasis. PLoS Negl Trop Dis. 2011;5:1-7. A6. King CL, Suamani J, Sanuku N, et al. A trial of a triple-drug treatment for lymphatic filariasis. N Engl J Med. 2018;379:1801-1810. A7. Opoku NO, Bakajika DK, Kanza EM, et al. Single dose moxidectin versus ivermectin for Onchocerca volvulus infection in Ghana, Liberia, and the Democratic Republic of the Congo: a randomised, controlled, double-blind phase 3 trial. Lancet. 2018;392:1207-1216. A8. Abegunde AT, Ahuja RM, Okafor NJ. Doxycycline plus ivermectin versus ivermectin alone for treatment of patients with onchocerciasis. Cochrane Database Syst Rev. 2016:CD011146.

GENERAL REFERENCES For the General References and other additional features, please visit Expert Consult at https://expertconsult.inkling.com.

CHAPTER 335  Nematode Infections  

GENERAL REFERENCES 1. Jourdan PM, Lamberton PHL, Fenwick A, Addiss DG. Soil-transmitted helminth infections. Lancet. 2018;391:252-265. 2. Lamberton PH, Jourdan PM. Human ascariasis: diagnostics update. Curr Trop Med Rep. 2015; 2:189-200. 3. Farrell SH, Coffeng LE, Truscott JE, et al. Investigating the effectiveness of current and modified World Health Organization guidelines for the control of soil-transmitted helminth infections. Clin Infect Dis. 2018;66:S253-S259. 4. Ásbjörnsdóttir KH, Means AR, Werkman M, Walson JL. Prospects for elimination of soil-transmitted helminths. Curr Opin Infect Dis. 2017;30:482-488. 5. Loukas A, Hotez PJ, Diemert D, et al. Hookworm infection. Nat Rev Dis Primers. 2016;2:1-18. 6. Krolewiecki A, Nutman TB. Strongyloidiasis: a neglected tropical disease. Infect Dis Clin North Am. 2019;33:135-151. 7. Martinez-Perez A, Roure Diez S, Belhassen-Garcia M, et al. Management of severe strongyloidiasis attended at reference centers in Spain. PLoS Negl Trop Dis. 2018;12:1-12. 8. Henriquez-Camacho C, Gotuzzo E, Echevarria J, et al. Ivermectin versus albendazole or thiabendazole for Strongyloides stercoralis infection. Cochrane Database Syst Rev. 2016;1:CD007745.

2140.e1

9. Repetto SA, Ruybal P, Batalla E, et al. Strongyloidiasis outside endemic areas: longterm parasitological and clinical follow-up after ivermectin treatment. Clin Infect Dis. 2018;66:1558-1565. 10. Ma G, Holland CV, Wang T, et al. Human toxocariasis. Lancet Infect Dis. 2018;18:e14-e24. 11. Rostami A, Gamble HR, Dupouy-Camet J, et al. Meat sources of infection for outbreaks of human trichinellosis. Food Microbiol. 2017;64:65-71. 12. Wilson NO, Hall RL, Montgomery SP, et al. Trichinellosis surveillance—United States, 2008-2012. MMWR Surveill Summ. 2015;64:1-8. 13. Nunes MC, Guimaraes Junior MH, Diamantino AC, et al. Cardiac manifestations of 6 parasitic diseases. Heart. 2017;103:651-658. 14. Barratt J, Chan D, Sandaradura I, et al. Angiostrongylus cantonensis: a review of its distribution, molecular biology and clinical significance as a human pathogen. Parasitology. 2016;143:1087-1118. 15. Showler AJ, Kubofcik J, Ricciardi A, et al. Differences in the clinical and laboratory features of imported onchocerciasis in endemic individuals and temporary residents. Am J Trop Med Hyg. 2019;100:1216-1222. 16. Kamgno J, Pion SD, Chesnais CB, et al. A test-and-not-treat strategy for onchocerciasis in Loa loaendemic areas. N Engl J Med. 2017;377:2044-2052. 17. Whittaker C, Walker M, Pion SDS, et al. The population biology and transmission dynamics of Loa loa. Trends Parasitol. 2018;34:335-350.

2140.e2

CHAPTER 335  Nematode Infections  

REVIEW QUESTIONS 1. A 21-year old woman living in rural Indonesia is in her third trimester of pregnancy and is found to have severe microcytic anemia with a hematocrit of only 19%. She denies vaginal bleeding but consistently has darkened stools, which are positive when tested for occult blood. Blood laboratory testing is also remarkable for eosinophilia and low serum albumin concentration. Which of the following parasitic infections is the most likely to result in this syndrome? A . Enterobiasis B. Amoebiasis C. Diphyllobothrium latum infection D. Hookworm infection E. Strongyloidiasis Answer: D  Intestinal blood loss is the principal clinical manifestation of hookworm infection, which can lead to iron deficiency and microcytic anemia. Pregnant women and children are at greater risk because of their higher iron needs. D. latum infection is associated with development of megaloblastic anemia due to vitamin B12 deficiency and is thus associated with an elevated, rather than low, red blood cell mean corpuscular volume. Infection with Enterobius vermicularis, Entamoeba histolytica, or Strongyloides stercoralis is not usually associated with intestinal blood loss or iron deficiency. 2. A couple has just returned from their honeymoon in the Caribbean complaining of an intensely pruritic rash on their feet, buttocks, and legs. On examination, the rash consists of multiple raised, serpiginous red tracks. Most of their time was spent relaxing on a beach owned by the resort where they were staying. Which is the most likely causative organism? A . Dracunculus medinensis B. Ancylostoma duodenale C. Ancylostoma braziliense D. Strongyloides stercoralis E. Enterobius vermicularis Answer: C  The description of the skin lesions is typical of cutaneous larva migrans. This syndrome is most commonly caused by penetration of the skin by the larvae of the dog or cat hookworms, Ancylostoma caninum or A. braziliense. S. stercoralis may also cause a migrating serpiginous dermatologic rash known as larva currens, but it is found most commonly on the buttocks and lower back because the larvae that cause the tracks originate from the anus. D. medinensis results in a dramatic ulcerative, painful, nonmigratory skin lesion from which the adult female worm emerges to release eggs into the environment. Neither A. duodenale nor E. vermicularis produce rashes. 3. A 60-year-old male kidney transplant recipient from rural Eastern Kentucky is admitted to the hospital with a 14-day history of fever and profuse diarrhea. His medications include prednisone, cyclosporine, mycophil mycophenolate, and lisinopril. Laboratory examination is remarkable for a leukocyte count of 13,000 per μL, of which 25% are eosinophils. In the hospital, blood cultures are repeatedly positive for Escherichia coli. Besides treating the bacteremia with antibacterial therapy, which of the following medications is indicated to treat the most likely underlying predisposing parasitic infection? A . Pyrantel pamoate B. Ivermectin C. Praziquantel D. Mebendazole E. Metronidazole Answer: B  The patient’s clinical picture is suggestive of disseminated strongyloidiasis. Strongyloides stercoralis is endemic in rural parts of Appalachia, and infection can be maintained in a host for decades because of the autoinfection cycle of this parasite. Immunosuppression, particularly with corticosteroids, may lead to hyperinfection and dissemination of larvae to aberrant locations in the body. Enteric bacteria such as E. coli can be carried by the migrating larvae throughout the body, resulting in gram-negative sepsis, meningitis, and abscess formation. Ivermectin is the treatment of choice for strongyloidiasis. Although albendazole is a second-line therapy, mebendazole, also a benzimidazole, has poor oral bioavailability and would therefore not be an ideal choice. Pyrantel pamoate and praziquantel are anthelminthics but have poor activity against S. stercoralis.

4. A young boy who recently emigrated from Guatemala to the United States with his parents is brought by them to the emergency room because of lethargy and severe abdominal pain that has persisted for about 2 days. He is unable to walk because of the pain. On physical exam, he is afebrile; bowel sounds are absent, and there is significant diffuse abdominal tenderness that is more pronounced in the right lower quadrant. Radiographic studies suggest acute intestinal obstruction at the level of the ileum. How did this child become infected with the most likely causative organism? A . Drinking water contaminated with cysts B. Ingesting soil contaminated with embryonated eggs C. Walking barefoot outside, thus permitting larvae in contaminated soil to penetrate the skin D. Swimming in lakes that harbor snails and cercariae E. Eating insufficiently cooked meat infected with cysts Answer: B  Infection with Ascaris lumbricoides causing intestinal obstruction is the most likely diagnosis. This nematode is a soil-transmitted helminth, which to be infective to humans, requires that eggs undergo a phase of development and embryonation in warm, moist soil. Infection results from ingestion of contaminated soil. Drinking contaminated water containing cysts is the means of transmission for protozoan organisms such as Giardia intestinalis and Entamoeba histolytica. Meat is the vehicle for Taenia tapeworm or Trichinella spiralis infections, skin exposure to water containing the intermediate snail host is the means of acquiring schistosomiasis, and skin contact with contaminated soil pertains to the transmission of hookworm or Strongyloides stercoralis. 5. A young boy in Kenya is taken to the doctor because of episodes of blood-streaked diarrhea, fever, and lower abdominal pain. Examination of a smear of his feces under the microscope reveals many oval eggs with bipolar plugs. Which of the following complications might occur if the child is not treated for the causative helminth? A . Obstruction of the duodenum B. Rectal prolapse C. Asthma-like symptoms D. Gram-negative meningitis E. Severe megaloblastic anemia Answer: B  Usually, light infections with the whipworm Trichuris trichiura are asymptomatic. However, heavy infections may produce symptoms that include those of dysentery, as in this child. Failure to treat with an anthelminthic could result in rectal prolapse because of the chronic straining induced by the sensation of rectal fullness caused by the presence of many whipworms in the rectum and colon. Although microcytic anemia may develop due to intestinal blood loss associated with T. trichiura infection, megaloblastic anemia resulting from vitamin B12 deficiency is associated with infection with the tapeworm Diphyllobothrium latum. T. trichiura is small, and the adult worms are located in the large intestine, making obstruction of the duodenum unlikely. Also, whipworm has no pulmonary phase like Ascaris lumbricoides or hookworm and is therefore not associated with respiratory symptoms. Meningitis due to gram-negative bacilli is associated with disseminated strongyloidiasis but not with whipworm infection. 6. A 40-year old woman from southern India has severe lymphedema of her right lower extremity and left breast. She has had this condition for at least a decade, and it is a common condition in her rural village. She now has difficulty going to work and is extremely embarrassed by her deformity to the point at which she tries not to venture outside of her home except at night. What is the most likely diagnosis? A . Lymphatic filariasis B. Loiasis C. Onchocerciasis D. Dracunculiasis E. Podoconiosis

CHAPTER 335  Nematode Infections  

Answer: A  The most likely diagnosis is lymphatic filariasis due to chronic infection with Wuchereria bancrofti. Although loiasis, onchocerciasis, and dracunculiasis all have dermatologic manifestations, none of these result in chronic, irreversible lymphedema. Loa loa is associated with transient and migratory Calabar swellings that are hypersensitivity reactions to migrating adult worms; Onchocerca volvulus infection can result in generalized chronic dermatitis, and adult Dracunculus medinensis worms can rupture through intact skin from their location in the subcutaneous tissues to release eggs into the environment. Podoconiosis can result in chronic lymphedema due to irritant skin exposure to red clay, although this occurs mainly in the lower extremities, almost always starting in the foot. 7. A 28-year old graduate student in physical anthropology has been experiencing unusual swellings over different parts of her body for the past 2 months. The swellings range in size from 5 to 15 cm and generally last for 2 to 4 days before resolving; they are preceded by mild pain and intense pruritus that lasts a couple of hours. The field research for her dissertation is based on observational studies of gorillas in Gabon. A complete blood count is remarkable for very high eosinophilia (4000/μL) as well as an unusual finding on the peripheral blood smear (collected at mid-day) that prompts an urgent call from the laboratory technologist. What is the most likely explanation of the phenomenon seen on the blood smear? A . Microfilaria of Onchocerca volvulus B. Microfilaria of Wuchereria bancrofti C. Promastigote of Trypanosoma brucei gambiense D. Larva of Dracunculus medinensis E. Microfilaria of Loa loa Answer: E  The skin manifestations are most likely Calabar swellings due to migrating adult Loa loa worms in the subcutaneous tissue. These are usually transient and are preceded by pain and pruritus. Loaiasis is endemic in Gabon. O. volvulus is also endemic in this country and may result in subcutaneous nodules, although these are usually neither migratory nor transient in nature. In addition, microfilariae of O. volvulus are visualized after exiting from skin snip specimens, whereas those of L. loa are seen in blood. The microfilariae of W. bancrofti, which is endemic in Africa, are present in the blood, although their periodicity differs from that of L. loa: W. bancrofti microfilariae are best visualized in blood collected at night, whereas those of L. loa are more likely to be seen in a sample collected during the day, preferably between 10 am and 2 pm. T. brucei, the causative agent of African sleeping sickness, does not cause migratory skin swellings and is not associated with eosinophilia, although promastigotes can be observed on a Giemsa-stained blood smear. Finally, D. medinensis is associated with nonmotile skin lesions due to the adult female worm rupturing through subcutaneous tissue to the external environment to release larvae; however, no part of the life cycle occurs in the blood. 8. Within 2 weeks of eating jerky that he made from a cougar that he shot while hunting, an Oklahoma man develops severe vomiting and diarrhea, followed 2 days later by a fever of 38.7° C, throbbing headache, and myalgias. After a few more days of continued symptoms, he seeks medical attention. His white blood cell count at presentation is 17,300/μL, with 25% eosinophils. Which of the following laboratory tests would be most useful in making the correct diagnosis? A . Serum creatinine concentration B. Enzyme-linked immunosorbent assay (ELISA) C. Examination of a Giemsa-stained blood smear using a light microscope D. Serum creatine kinase concentration E. Skin biopsy Answer: D  The most likely diagnosis is trichinellosis, probably due to Trichinella nativa. Infected individuals, especially those with myalgias and muscle tenderness, usually have elevated levels of serum creatine kinase, likely because Trichinella larvae invade striated muscle cells inducing myositis. Unless the myositis is severe and rhabdomyolysis develops, associated renal dysfunction is rare. At this point in the disease process, seroconversion has likely not yet occurred, and therefore, the ELISA for Trichinella would be negative. Larvae are not usually detectable in a stained blood smear or a skin biopsy, although nurse cells which are striated muscle cells containing encysted larvae can be visualized on muscle biopsy specimens.

2140.e3

9. A 7-year old boy who has never traveled outside of North America develops decreased visual acuity in his right eye. Visible leukokoria is evident to his parents, which prompts a visit to their pediatrician, who consults an ophthalmologist. Fundoscopy reveals a single lesion of the retina. Assuming that his ocular condition is caused by a parasitic worm, which is the most likely means that this child became infected? A . Being bitten by an infected black fly B. Eating dirt while playing in a Boston public park C. Swimming in a lake containing snails D. Walking barefoot outside while visiting the family’s summer cabin in rural West Virginia E. Eating ceviche while on a family vacation to Mexico Answer: B  Although several nematode species have been associated with ocular manifestations in humans, Toxocara canis and Toxocara cati are the most common etiologies of ocular larva migrans and are found worldwide. Children who eat dirt due to pica are most at risk. Dogs and cats naturally harbor the adult parasitic worms, but humans—especially children—may become accidentally infected when embryonated eggs are ingested, even though humans are unsuitable hosts for maturation into adult worms. Retinal involvement occurs when excysted larvae penetrate the small intestine and disseminate throughout the body, including the eye. Because of the inflammatory response, visually distinguishing the condition in situ from a retinoblastoma or ocular tuberculosis can be challenging. Onchocerca volvulus, which is transmitted by the bite of a Simulium black fly, can cause ocular lesions, but these are almost exclusively corneal. In addition, onchocerciasis is endemic now mostly in Africa. Skin exposure to fresh water containing certain species of snails is the means of acquiring schistosomiasis, which does not result in retinal manifestations. Nor does Strongyloides stercoralis, which is endemic in certain rural parts of Appalachia and is acquired through skin exposure to soil containing infective larvae. 10. A 35-year old woman who has recently immigrated to the United States from Cameroon presents with chronic atrophic dermatitis and punctate keratitis. Laboratory investigations reveal a peripheral eosinophil count of 3500/μL. Microfilaria are observed in the anterior chambers of both eyes on slit lamp examination. When considering management of this patient, which of the following antiparasitic medications should be contraindicated? A . Ivermectin B. Diethylcarbamazine (DEC) C. Albendazole D. Pyrantel pamoate E. Praziquantel Answer: B  This woman most likely has onchocerciasis due to chronic infection with Onchocerca volvulus, resulting in both the typical skin and ocular manifestations. DEC should never be used for treatment of onchocerciasis because of frequent unacceptable hypersensitivity reactions to dying microfilariae, which can range from urticaria and angioedema to hypotension and death. Ivermectin is the treatment of choice for O. volvulus, whereas albendazole, pyrantel pamoate, and praziquantel are neither effective nor detrimental in this condition.

2140

CHAPTER 336  Antiviral Therapy (Non-HIV)  

  ANTIVIRALS FOR HEPATITIS B

VIRUS INFECTIONS

Acute hepatitis B virus (HBV) infection (Chapter 139) generally does not require antiviral treatment. Currently approved antivirals for chronic hepatitis B (Chapter 140) include six nucleic acid analogues (adefovir, entecavir, lamivudine, telbivudine, tenofovir disoproxil fumarate (TDF), and tenofovir alafenamide [TAF]), as well as two immune modulators (interferon-alfa-2b and pegylated [PEG]-interferon-alfa-2a) (Tables 336-1 to 336-3).1 Treatment may be initiated with any approved antiviral medications, but tenofovir, TAF, entecavir, and PEG-interferon-alfa-2a are generally the preferred agents. Tenofovir or entecavir is preferred for patients with compensated cirrhosis (Chapter 144). The goal of antiviral treatment is to suppress HBV replication and reduce the progression of liver disease and its complications. Patients with chronic hepatitis B (hepatitis B surface antigen [HBsAg] positive for >6 months, detectable serum HBV DNA >20,000 IU/mL, and an alanine aminotransferase [ALT] level more than twice the normal level) should be evaluated for treatment. Patients with clinically decompensated hepatitis B (e.g., icterus or other signs) generally require antiviral treatment. Therapy in hepatitis B e antigen (HBeAg)-positive chronic hepatitis B should be continued until the patient has achieved HBeAg seroconversion and serum HBV DNA is undetectable, followed by at least 6 months of additional treatment after the appearance of anti-HBe. Therapy in HBeAg-negative chronic hepatitis B should continue for at least a year. Patients with decompensated cirrhosis or recurrent hepatitis B after liver transplantation should receive lifelong treatment.

TABLE 336-1 ANTIVIRALS FOR HEPATITIS VIRUS INFECTIONS VIRAL INFECTION Chronic hepatitis B

336  ANTIVIRAL THERAPY (NON-HIV) JOHN H. BEIGEL AND SHYAMASUNDARAN KOTTILIL Although some viral infections are self-limited, others can cause significant morbidity and mortality. Effective therapy is available for many of these infections. This chapter reviews currently available antiviral agents for the treatment of infections caused by viruses other than human immunodeficiency virus (HIV). Not all agents discussed are licensed in all countries. Currently available agents can be classified into those that directly inhibit viral replication at the cellular level (antivirals), those that modify the host response to infection (immunomodulators), and those that directly inactivate viral particles (microbicides/virucides). Antiviral agents can be classified based on their mechanism of action. For example, nucleic acid analogues inhibit viral DNA or RNA synthesis by competing with endogenous nucleic acids and block the viral DNA polymerase or RNA transcriptases. By comparison, protease inhibitors prevent viral replication by binding to the enzymes that cleave viral protein precursors into active proteins. Antiviral strategies that are not covered in this chapter include local destructive measures that destroy both host tissues and virus simultaneously, such as cryotherapy, laser, and podophyllin treatment of warts. Although effective, such measures are useful only for discrete or localized mucocutaneous infections.

Chronic hepatitis C

DRUG

ROUTE

Tenofovir disoproxil fumarate Tenofovir alafenamide Entecavir Naïve virus

PO

300 mg/day

PO

25 mg/day

PO

Lamivudineresistant virus Interferon-alfa-2b

PO

PEG-interferon-alfa-2a Adefovir Lamivudine Telbivudine

SC PO PO PO

0.5 mg daily; optimal duration of therapy unknown 1 mg daily; optimal duration of therapy unknown 6 MU/m2 (up to 10 MU) three times weekly for 16-24 wk 180  µg weekly for 48 wk 10 mg/day 100 mg/day 600 mg/day

Ledipasvir/sofosbuvir

PO

Sofosbuvir/velpatasvir

PO

Sofosbuvir/velpatasvir/ voxilaprevir Daclatasvir/sofosbuvir

PO

Grazoprevir/elbasvir

PO

Glecaprevir/ pibrentasvir Ombitasvir/ paritaprevir/ Ritonavir + dasabuvir

PO

Daclatasvir/asunaprevir Sofosbuvir

PO PO

Simeprevir PEG-interferon-alfa-2a or PEG-interferonalfa-2b plus ribavirin

PO SC SC

SC

PO

PO

PO

USUAL ADULT DOSAGE

90 mg/400 mg once daily for 12-24 wk 400 mg/100 mg/day for 12 wk 400 mg/100 mg/100 mg/day for 12 wk 60 mg/400 mg/day for 12-24 wk 100 mg/50 mg/day for 12-16 wk 300 mg/120 mg/day for 8-16 wk 12.5 mg/75 mg/50 mg (two pills once daily) + dasabuvir 250 mg twice daily for 12-24 wk (+RBV for genotype 1a) 60 mg/400 mg/day for 24 wk 400 mg once daily for 12-24 wk 150 mg once daily for 12 wk 180 µg weekly for 48 wk 1.5  µg/kg weekly for 48 wk 800-1200 mg/day, depending on weight

2140.e4

ABSTRACT

CHAPTER 336  Antiviral Therapy (Non-HIV)  

A number of antiviral agents are available for the treatment of viral infections other than human immunodeficiency virus (HIV). Acute hepatitis B virus (HBV) and hepatitis C virus (HCV) infection generally do not require antiviral treatment. The preferred regimens for chronic hepatitis B include tenofovir disoproxil fumarate, tenofovir alafenamide, entecavir, and PEG-interferon-alfa2a. Ledipasvir/sofosbuvir, sofosbuvir/simeprevir, and ombitasvir/paritaprevir/ ritonavir plus dasabuvir are preferred regimens for chronic HCV genotype 1. Sofosbuvir and weight-based ribavirin are preferred for HCV genotypes 2 and 3. Ledipasvir/sofosbuvir and sofosbuvir/simeprevir are preferred regimens for HCV genotype 4. Sofosbuvir/velpatasvir and ledipasvir/sofosbuvir are considered primary regimens for HCV genotype 5 or 6. Acyclovir or valacyclovir are the preferred treatments of herpes simplex and varicella-zoster virus. Ganciclovir or valganciclovir are the preferred treatments of cytomegalovirus. Oseltamivir is the preferred treatment of influenza. Details about clinical use, dosing, toxicities, and resistance are discussed within the chapter.

KEYWORDS

hepatitis B hepatitis C herpes simplex varicella-zoster cytomegalovirus influenza

CHAPTER 336  Antiviral Therapy (Non-HIV)  

2141

TABLE 336-2 MECHANISMS OF EXCRETION AND THRESHOLDS FOR DOSE ADJUSTMENT MAJOR ROUTE OF ELIMINATION

THRESHOLD FOR ADJUSTMENT IN RENAL INSUFFICIENCY OR FAILURE

ADJUSTMENT FOR HEPATIC FAILURE

Adefovir

Renal

CrCl