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

1,148 167 143MB

English Pages 2664 [4629] Year 2019

Report DMCA / Copyright


Polecaj historie

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

Table of contents :
Front Matter
Copyright Page
Associate Editors
Video Contents
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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
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)
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
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
224 Approach to Women’s Health
225 Contraception
226 Common Medical Problems in Pregnancy
227 Menopause
228 Intimate Partner Violence
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
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
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
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
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
315 Systemic Antifungal Agents
316 Endemic Mycoses
317 Cryptococcosis
318 Candidiasis
319 Aspergillosis
320 Mucormycosis
321 Pneumocystis Pneumonia
322 Mycetoma and Dematiaceous Fungal Infections
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
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
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
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
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
402 Principles of Medical Consultation
403 Preoperative Evaluation
404 Overview of Anesthesia
405 Postoperative Care and Complications
406 Medical Consultation in Psychiatry
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


Organized and Edited by: Min Thant Thaw




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


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: 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


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



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


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



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


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



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



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



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


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



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


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



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


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


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



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


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



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



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



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


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



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


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




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


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 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


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


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





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



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.


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



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.


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.


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.




SYMPTOMS Constitutional Fever

264, 265

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



E-Table 258-1

Poor appetite


Table 123-1

Weight loss

123, 206

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



Figure 207-1

Snoring, sleep disturbances


Table 377-6



Tables 370-1, 370-2

Visual loss, transient

395, 396

Tables 395-2, 396-1

Ear pain


Table 398-3

Hearing loss


Figure 400-1

Ringing in ears (tinnitus)


Figure 400-2



Figure 400-3

Nasal congestion, rhinitis, or sneezing


Loss of smell or taste


Table 399-1

Dry mouth


Table 397-7

Sore throat


Figure 401-2; Table 401-1



Head, Eyes, Ears, Nose, Throat

Cardiopulmonary Chest pain

45, 128

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



Shortness of breath

45, 77

Figure 77-3


45, 56

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


45, 56, 400

Figure 56-1; Table 400-1



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

Cardiac arrest


Figures 57-2, 57-3



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



Tables 77-6, 77-7

Nausea and vomiting


Figure 123-5; Table 123-5

Dysphagia, odynophagia

123, 129

Table 123-1


126, 144

Figure 126-3; Table 126-1


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


128, 131

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

Melena, blood in stool


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


127, 128

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

Fecal incontinence


Figure 136-5

Anal pain



Genitourinary Dysuria

268, 269

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



Table 268-3



Tables 23-1 to 23-3

Renal colic


Figure 117-1

Vaginal discharge


Menstrual irregularities


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

Female infertility

223, 227

Table 223-5

Hot flushes


Table 227-1

Erectile dysfunction


Figure 221-10

Male infertility


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

Scrotal mass


Figure 190-1

Genital ulcers or warts


Table 269-1






Musculoskeletal Neck or back pain


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

Painful joints


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



Table 71-3

Acute limb ischemia


Figure 71-4; Table 71-2


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


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

Abnormal gait


Table 368-2



Tables 375-1 to 375-6

Abnormal bleeding


Table 162-1


407, 412

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


237, 411

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

Abnormal pigmentation


Table 412-2

Alopecia and hirsutism


Tables 413-1, 413-3

Nail disorders


Table 413-4


264, 265

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

Heat illness/hyperthermia


Tables 101-1 to 101-3


7, 101

Tables 101-4 to 101-6


7, 56, 58, 59

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



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


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


Table 395-3

Red eye


Tables 395-4, 395-6

Dilated pupil


Figure 396-4



Table 396-4



Table 396-2



Figure 396-6



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



Table 398-3



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

Oral ulcers and discolorations


Tables 397-1 to 397-4

Salivary gland enlargement


Table 397-6

Neck mass


Figure 181-3



Tables 159-1 to 159-4

Thyroid nodule


Figure 213-5



Figures 213-2, 213-3




SIGNS Vital Signs

Head, Eyes, Ears, Nose, Throat


Breast Breast mass


Lungs Wheezes


Table 77-4

Heart murmur or extra sounds


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

Jugular venous distention


Table 45-6

Carotid pulse abnormalities


Figure 45-4






Abdomen Hepatomegaly


Figure 137-5



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



Table 136-1



Figure 241-1



Figure 45-7






Neurologic Delirium


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

Psychiatric disturbances


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



Tables 376-1 to 376-4


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



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

Suspicious mole


Table 193-1

Nail diseases


Table 413-4



Tables 149-2 to 149-6



Table 157-4



Figure 158-4; Table 158-1



Table 158-3



Table 158-2



Figure 161-1; Table 161-1

Neutropenia   With fever

158 265

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



Table 157-5



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

Prolonged PT or PTT


Figure 162-4


106, 112

Tables 106-2, 112-6

Abnormal liver enzymes


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



Tables 216-1, 216-2



Tables 217-1, 217-2

Electrolyte abnormalities

108, 109

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

Acid-base disturbances


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



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



Figure 232-4; Table 232-6

Hypo- and hyperphosphatemia


Tables 111-2, 111-3

Magnesium deficiency


Table 111-1

Elevated Pco2


Figure 80-2

Solitary pulmonary nodule


Figure 182-2

Pleural effusion


Tables 92-3 to 92-5

ECG abnormalities


Tables 48-2 to 48-5

Skin and Nails



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.


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


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.


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.


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,



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.


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.




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


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


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.


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



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.


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.


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.


CHAPTER 2  Bioethics in the Practice of Medicine  



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.


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


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.


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

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. 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.


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.


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.


CHAPTER 3  Palliative Care  


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 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  


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.


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



CHAPTER 3  Palliative Care  

High Function

Mostly cancer

Death Time


Short period of evident decline


Chronic, consistent with usual role

Mostly heart and lung failure





Long-term limitations with intermittent serious episodes


Mostly frailty and dementia


Chronic, progressive, eventually fatal illness



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.)



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

No Pain 0














































































































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 :



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  






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.


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.


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.


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).


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.


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,


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



Brief Pain Inventory (Short Form) Time:

Date: Name:



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.





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










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










10 Pain as bad as you can imagine


10 Pain as bad as you can imagine


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









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

0 No pain









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

CHAPTER 3  Palliative Care  


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










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









10 Completely interferes









10 Completely interferes









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









10 Completely interferes









10 Completely interferes









10 Completely interferes









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,


CHAPTER 3  Palliative Care  



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?


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


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”


Topic: communicating information about cancer stage


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


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


“Do you have questions about the staging?”


Face the patient Squarely


Adopt an Open body posture


Lean toward the patient


Use Eye contact


Maintain a Relaxed body posture

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

Name the emotion: “You seem worried”


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


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


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


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


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


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


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?”


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


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

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.


6. Center to Advance Palliative Care. 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.


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.


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.



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.


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.



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  


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.


health care disparities health disparities health equity race ethnicity socioeconomic factors



CHAPTER 4  Disparities in Health and Health Care  

100 2015 Hispanic or Latino

Life expectancy (years)

White, not Hispanic Black, not Hispanic 80


White female


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



40 60 80 Life expectancy (years)


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].)







All causes







Heart disease














Chronic lung disease







Cerebrovascular disease







Alzheimer disease







Diabetes mellitus







Influenza & pneumonia














Kidney disease














*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 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  


60 Men


No high school diploma


High school diploma or GED



No high school diploma High school diploma or GED



Some college

Some college

Bachelor’s degree or higher


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


20 Non-Hispanic white1,2 15


Non-Hispanic black1 Hispanic1


0 1999









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: (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


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



Infections, rheumatic heart disease, and nutritional cardiomyopathies

Rural India, sub-Saharan Africa, South America


As above plus hypertensive heart disease and hemorrhagic strokes



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

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.


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.]


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  



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.


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.


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.).


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  


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.


probability statistical analysis decision making Bayes theorem decision analysis prediction rules



CHAPTER 8  Statistical Interpretation of Data  





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.


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.


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.  


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.


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.  









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


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 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)


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).


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.



CHAPTER 8  Statistical Interpretation of Data  


Number of patients


With disease


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.”


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  


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


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.


Probability 1 Outcome






Choice A

1 − (Probability 1) Choice node

Chance node

Probability 2

Choice B

1 − (Probability 2)



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

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.


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.


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


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.


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



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.


health care quality health care value evidence-based health care


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.



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.


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


HEALTH Mortality

Deaths due to colorectal cancer per 100,000 population


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


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.



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 (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






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



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†


Hospitalization ± indefinitely

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

MI, recurrent ischemia



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

Recurrent ischemia




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

Death, MI



1-12 months

300 mg orally initial dose, then 75 mg orally daily

MI, recurrent ischemia

Lipid Lowering Statin Antiplatelet








Anticoagulant Unfractionated heparin


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)


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



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†


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


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

Recurrent ischemia

Aspirin and Clopidogrel or Prasugrel or Ticagrelor



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

Death, MI


≥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


≥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



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


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


MI, recurrent ischemia


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.


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.


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.



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


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



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.


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.


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.



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


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


9 No in-hospital percutaneous coronary intervention


Predicted all-cause mortality from hospital discharge to 6 months


0.50 0.45




0.35 Probability

4 5 6

0.30 0.25 0.20







Mortality risk


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


Total risk score

Findings during hospitalization

(Sum of points) (From plot)

0 70



130 150 170 Total risk score



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.)



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


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


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 









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)





Antigenic/allergic reactions





Systemic fibrinogen depletion


Mild to moderate



Intracerebral hemorrhage





Patency (TIMI 2/3) rate, 90 minutes‡





Lives saved per 100 treated





Cost per dose (approximate U.S. dollars)





*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








0.25 mg/kg IV bolus

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


162-325 mg

81 mg*-325 mg/day


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

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


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.


CHAPTER 67  Infective Endocarditis  

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




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.


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


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


Fungal endocarditis


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


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


Recurrent emboli after appropriate antibiotic therapy


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


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


Increase in vegetation size despite appropriate antimicrobial therapy


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


Persistent fever and leukocytosis with negative blood cultures


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.


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,


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   


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.



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

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.


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.


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.)


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 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.  


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


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  


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.


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


CHAPTER 68  Pericardial Diseases  


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.





CHEST PAIN Character

Pressure-like heavy, squeezing

Sharp, stabbing, occasionally dull

Sharp, stabbing

Change with respiration


Worsened with inspiration

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

Change with position


Worse when supine; improved when sitting up or leaning forward



Minutes (ischemia); hours (infarction)

Hours to days

Hours to days

Response to nitroglycerin


No change

No change

Absent (unless pericarditis is present)

Present in most patients

Pleural friction rub may occur


Localized convex

Widespread concave

Limited to leads III, aVF, and V1

PR segment depression




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.



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.


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 ,








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


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


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


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


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.


CHAPTER 68  Pericardial Diseases  













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



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


Trial of NSAIDs Pain relieved in 20 mm)? No


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.


CHAPTER 68  Pericardial Diseases  


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, 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.  


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.  


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.  


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.


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 ( (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.


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.  


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.  


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







Metoprolol XL

50-200 (1-2)

50 (1)



20-320 (1)

40 (1)


2.5-10 (1)

2.5 (1)


10-80 (1)

10 (1)


2.5-20 (1-2)

2.5 (2)


10-60 (2)

10 (1)

Isradipine CR

2.5-20 (2)

2.5 (2)


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)


20-60 (2)

20 (2)


10-40 (1-2)

10 (2)


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)


100-2400 (2)

200 (2)


5-40 (1)

5 (1)

150-300 (1)

150 (1)



10-80 (1-2)

20 (1)



25-150 (2)

25 (2)



2.5-40 (2)

5 (2)



10-80 (1-2)

20 (2)


1-16 (1-2)

1 (1)


5-80 (1-2)

40 (2)


1-40 (2-3)

1 (2)


7.5-30 (1)

7.5 (1)


1-20 (1)

1 (1)


4-16 (1)

4 (1)

20-120 (2)

20 (2)


5-80 (1-2)

40 (2)

Phenoxybenzamine for pheochromocytoma


2.5-20 (1)

2.5 (1)


1-8 (1)

2 (1)


0.3-1.2 (3)

0.3 (3)

Clonidine patch

0.1-0.6 (weekly)

0.1 (weekly)


2-32 (2)

2 (2)


1-3 (1) (qhs)

1 (1)


250-1000 (2)

250 (2)


0.05-0.25 (1)

0.05 (1)


10-200 (3)

25 (3)


2.5-100 (1)

2.5 (1)


75-300/12.5-25 (1)

150/12.5 (1)


5/50 (1)

5/50 (1)


2.5-5/10-20 (1)

2.5/10 (1)


5-10/20-40 (1)

5/20 (1)


5/20-10/80 (1)

5/20 (1)


5-10/160-320 (1)

5/160 (1)


50-100/25 (1)

50/25 (1)


40-80/12.5-25 (1)

40/12.5 (1)


5-20/6.25-25 (1)

20/6.25 (1)


2.5-10/6.25 (1)

2.5/6.25 (1)


16-32/12.5-25 (1)

16/12.5 (1)


5-10/25 (1-2)

5/25 (1)


600/12.5-25 (1)

600/12.5 (1)


10-20/12.5 (1)

10/12.5 (1)


150-300/12.5-25 (1)

150/12.5 (1)


50-100/12.5-25 (1)

50/12.5 (1)


20-40/12.5 (1)

20/12.5 (1)


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

20/5/12.5 (1)


25/25 ( 1 2 -1)

25/25 (1/2)


40-80/12.5-25 (1)

40/12.5 (1)


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

40/5/12.5 (1)


2-4/180-240 (1)

2/180 (1)


37.5/25 ( 2 -1)

37.5/25 (1/2)


80-160/12.5-25 (1)

160/12.5 (1)


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

160/5/12.5 (1)


40-80 (1)

40 (1)


8-32 (1-2)

8 (1)


400-800 (1-2)

400 (1)


75-300 (1)

150 (1)


25-100 (2)

50 (1)


5-40 (1)

20 (1)


10-80 (1)

40 (1)


80-320 (1-2)

160 (2)

DIURETICS Thiazide-Type Diuretics Indapamide

0.625-2.5 (1)

1.25 (1)


6.25-50 (1)

12.5 (1)


12.5-50 (1)

12.5 (1)


2.5-5 (1)

2.5 (1)

Loop Diuretics Bumetanide

0.5-2 (2)

1 (2)

Ethacrynic acid

25-100 (2)

25 (2)


20-160 (2)

20 (2)


2.5-20 (1-2)

5 (2)

Potassium Sparing Eplerenone

25-100 (1-2)

25 (1)


12.5-100 (1-2)

12.5 (1)


25-100 (1)

37.5 (1)


5-20 (1)

10 (2)


200-800 (2)

200 (2)


25-100 (1)

25 (1)


5-20 (1)

5 (1)


2.5-20 (1)

2.5 (1)


2.5-10 (1)

2.5 (1)


50-450 (2)

50 (2)




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



CHAPTER 70  Arterial Hypertension  



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  






≥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


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.


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.


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.


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

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.


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.


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.


CHAPTER 77  Approach to the Patient with Respiratory Disease  


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



Bronchial breathing

Pneumonia or interstitial lung disease

Fine crackle

Heart failure, interstitial lung disease, alveolar filling disorders

Coarse crackle



Asthma, COPD


Bronchitis, COPD


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.


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).  


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  


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.


cough wheeze dyspnea hemoptysis



CHAPTER 77  Approach to the Patient with Respiratory Disease  







Common cold

Asthma exacerbation


Acute bacterial sinusitis



Heart failure

Sinus radiograph




Exacerbations of COPD

Aspiration syndromes

Methacholine inhalation challenge




Allergic rhinitis

Pulmonary embolism

GERD, esophageal stricture


Exacerbation of bronchiectasis

Modified barium esophagography


Environmental irritant rhinitis

Esophageal pH*




Endobronchial mass/lesion


SUBACUTE COUGH Postinfectious cough Pertussis



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

CHRONIC COUGH Rhinosinus conditions/UACS



Occupational and environmental

Non-asthma eosinophilic bronchitis

Aspiration Bronchogenic carcinoma


Drug-induced cough

Gastroesophageal reflux


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.


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).


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.


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  


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


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




0100 0100 Vital capacity (%)



0 100


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



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


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.



UPPER AIRWAY DISEASES Postnasal drip syndrome

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


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.


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.


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.


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.


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  




Pulmonary embolism

History of risk for thromboembolic disease, positive confirmatory tests.


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.


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


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


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?


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


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


CHAPTER 77  Approach to the Patient with Respiratory Disease  

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


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




Chest CT


Repeat chest radiography in 6-8 weeks


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


Chest CT


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

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.


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.


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.


CHAPTER 78  Imaging in Pulmonary Disease  





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  


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.


chest radiograph CT scanning lung pleura mediastinum


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


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.


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.


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


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 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

CHAPTER 79  Respiratory Testing and Function  


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.



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.


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  


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























Maximal Expiratory Flow, liters/sec

%PRED = percentage of predicted value








–6 –8

Flow Volume Curve Predicted Control Post dilator


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



Maximal Inspiratory Flow, liters/sec

Maximal Expiratory Flow, liters/sec











12 0



3 4 5 Expired Volume, liters




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  


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






















56 7



DLCO (hb adj)





%PRED = percentage of predicted value.

Expiratory Flow, liters/sec




Inspiratory Flow, liters/sec






Expired Volume (L) Inspired




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.


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


Flow Volume Curve PRED mayo Control Post dilator

10 8 6 4 2 0 0



3 4 5 Expired Volume, liters



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























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  


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





























DLCO (hb adj)








%PRED = percentage of predicted value.

Maximal Expiratory Flow, liters/sec


Flow Volume Curve PRED Control Post dilator










3 4 Expired Volume, liters




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.


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























84 100 93

%PRED = percentage of predicted value.

Maximal Expiratory Flow, liters/sec


Flow Volume Curve PRED Control Post dilator

10 8 6 4 2 0 0



3 4 Expired Volume, liters




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  


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

































91 (rest)

73 (exercise)


%PRED = percentage of predicted value.

Maximal Expiratory Flow, liters/sec


Flow Volume Curve Predicted Control Post dilator

10 8 6 4 2 0 0



3 4 Expired Volume, liters




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.


CHAPTER 80  Disorders of Ventilatory Control  



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 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.  


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.  


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.  


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  


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.  


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  


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  


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.


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



CHAPTER 80  Disorders of Ventilatory Control  




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


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  


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.  


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).  


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


CHAPTER 80  Disorders of Ventilatory Control  




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


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.


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

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.


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.


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  




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.  


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.  



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.


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  


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.


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



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.  



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.  


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





>80 >80 >80

No spirometric abnormalities


>80 >70



>60 45-70 30-50


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  




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  


Mechanical Ventilation

Biochemical injury (biotrauma) Epithelium/ interstitium

Cytokines, complement, PGs, LTs, ROS, proteases mφ


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.)


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




























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.


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 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


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

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


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.]


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  




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.  


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.  


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  


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.


shock hypotension hypoperfusion cardiac output vasomotor tone sepsis trauma



CHAPTER 98  Approach to the Patient with 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.



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.


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.


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.


Tachypnea, increased minute ventilation, increased dead space, bronchospasm, hypocapnia with progression to respiratory failure, acute lung injury, and adult respiratory distress syndrome.


Low-flow states may result in abdominal pain, ileus, gastrointestinal bleeding, pancreatitis, acalculous cholecystitis, mesenteric ischemia, and shock liver.


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.


Respiratory alkalosis is the first acid-base abnormality, but metabolic acidosis occurs as shock progresses. Hyperglycemia, hypoglycemia, and hyperkalemia may develop.


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.


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).  


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


Blood pressure





Pulse pressure

Normal or increased




Respiratory rate (per minute)





Urine output (mL/hr)





Mental status

Slightly anxious

Mildly anxious

Anxious, confused

Confused, lethargic

Fluid replacement



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.


CHAPTER 98  Approach to the Patient with Shock  





1-4  µg/kg/min







5-10  µg/kg/min 11-20  µg/kg/min

1-2+ 2-3+

1+ 1+

2+ 2+

2+ 2+

2+ 3+


0.04-0.1 units/min






Septic shock, post– cardiopulmonary bypass shock state, no outcome benefit in sepsis


20-200  µg/min






Vasodilatory shock; best for supraventricular tachycardia

Norepinephrine 1-20  µg/min






First-line vasopressor for septic shock, vasodilatory shock


1-20  µg/min






Refractory shock, shock with bradycardia, anaphylactic shock


1-20  µg/kg/min






Cardiogenic shock, septic shock


37.5-75  µg/kg bolus followed by 0.375-0.75  µg/min






Cardiogenic shock, right-sided heart failure, dilates pulmonary artery; caution in renal failure


“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.


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.  


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

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.


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.


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.


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).



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.  


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  


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.


cardiogenic shock coronary revascularization vasopressors inotropic therapy hemodynamic monitoring IABP mechanical circulatory support


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.  


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


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.  


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  


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 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


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


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.


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


Emergent reperfusion (for Mi/ACS) PCI CABG Fibrinolysis



c o v e r

l Temporary mechanical support* IABP Peripheral VAD ECMO Implantable VAD


i a t i o n

Durable VAD


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. ,


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.


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

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.


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.


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  




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  


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.


septic shock antibiotics fluids norepinephrine vasopressin corticosteroids dobutamine



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


Respiratory (Pao2/Fio2)


Coagulation (platelet count × 103/µL) Liver (bilirubin mg/dL [µmol/L])





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.


CHAPTER 101  Disorders Due to Heat and Cold  



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.



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  


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.


temperature regulation heatstroke heat exhaustion hypothermia freezing and nonfreezing peripheral cold injuries frostbite therapeutic hypothermia and hyperthermia



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


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.  


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.  


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  


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  





Young children or elderly

15-55 years


Chronic illness

Usually healthy




Prevailing weather

Frequent in heat waves




Strenuous exercise

Drug use

Diuretics, antidepressants, anticholinergics, phenothiazines

Ergogenic stimulants or cocaine


Often absent


Acid-base disturbances

Respiratory alkalosis

Lactic acidosis

Acute renal failure


Common (≈15%)



Common (≈25%)


Mildly elevated

Markedly elevated (500-1000 U/L)


Mildly elevated

Markedly elevated













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


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


25 g/dL


0.6-1.2 mEq/L

>1.2 mEq/L††


None measured

>25 mg/dL





15-40  µg/mL

>40 g/mL


10-20  µg/mL

>20 g/mL


≤30 mg/dL

>30 mg/dL


8-20  µg/mL

>20 g/mL

Valproic acid

50-100  µg/mL

>100 g/mL




100 g/24-hr urine††


20 g/L††


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.


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.


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


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


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


Patient stable Is toxin eliminated by kidneys?


Is enhanced elimination desired? Hemodynamic instability may prevent use of extracorporeal modalities

Is toxin removed by extracorporeal device?


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.


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


CHAPTER 102  Acute Poisoning  






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


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


CHAPTER 102  Acute Poisoning  






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


>1 million

1 million

>1000 (mostly due to


>1 million




0.54 mL blood/L urine)



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”)


Drugs: anthraquinones (laxatives), rifampicin Urobilinogenuria


Mepacrine Conjugated bilirubin Phenacetin Riboflavin

Brown or black

Melanin (on standing) Myoglobin (on standing) Alkaptonuria

Green or black

Phenol Lysol


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


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


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.)



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)


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.  


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.


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.  


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.


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


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).  


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 


GFR (mL/min/1.73 m2)


Kidney damage with normal or ↑GFR



Kidney damage with mild or ↓GFR



Moderate ↓GFR



Severe ↓GFR



Kidney failure


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.




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.


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


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.



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.  


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)  


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−,





Lactic acidosis


d-Lactic acidosis


Diabetic ketoacidosis


Starvation ketoacidosis


Alcoholic ketoacidosis

If ethanol is present

Ethylene glycol






5-Oxoprolinuria (acetaminophen)



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  


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.  


Lactic Acidosis  


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.  


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


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.  


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


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.



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


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.


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.


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.  


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


Low renin

Using diuretics?


UCI > 20 mEq/L

Low aldosterone


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 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 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.  


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).


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.



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.  


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



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

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.


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.


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.


CHAPTER 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


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 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  


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.


magnesium hypomagnesemia phosphate hypophosphatemia hyperphosphatemia


CHAPTER 111  Disorders of Magnesium and Phosphorus  

hypomagnesemia with hypercalciuria and nephrocalcinosis; and isolated hypomagnesemia, which is usually associated with hypocalcemia.  


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.


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 =



Severe hypermagnesemia is potentially fatal. Lesser degrees of hypermagnesemia usually respond well to treatment.


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.


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


CHAPTER 130  Acid Peptic Disease  












Age (yr)

60-79 ≥80

1 2




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


Pulse Melena Syncope Altered mental status Comorbidities

100-109 90-99 100 Present Present N/A Hepatic disease Cardiac failure

Blood urea, mmol/L


Hemoglobin, g/dL


INR Albumin


— —

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


• 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.


CHAPTER 131  Approach to the Patient with Diarrhea and Malabsorption  

Chronic Watery Diarrhea

Noninflammatory If high volume or elusive:



Bloody or Watery Stool hemoccult positive Lactoferrin, calprotectin, WBCs


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.



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  



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.


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.


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.  


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


Stool fat test


Carbohydrate malabsorption Evaluation of watery and inflammatory diarrhea (Fig. -3)


Suspicion of pancreatic disease (alcoholic, pancreatic surgery), fecal fat > 30 g/day


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,Chol, vitamin D, A, K, E




ApoB, A MG + FA

Peptides a.a. Di-, tripeptides Oligosaccharides Disaccharides

TG, Chol, vitamins D, A, K, E


Apo proteins Mixed micelle



Bile salts

Pancreatic enzymes


Mucosal Uptake




Phospholipid a.a. Di-, tripeptides


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  


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 (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.  


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).  


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


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).


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).


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.  



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


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  


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