Foot and Ankle Radiology [2 ed.] 1451192835, 9781451192834

Table of contents :
Cover
Contents
SECTION 1: Radiography
CHAPTER 1: Radiologic Physics, Radiobiology, and Radiation Safety
CHAPTER 2: Film—Screen Radiography
CHAPTER 3: Digital Radiography
CHAPTER 4: Positioning Techniques and Terminology
SECTION 2: Radiographic Anatomy
CHAPTER 5: The Normal Foot and Ankle
CHAPTER 6: Normal Variants and Anomalies
CHAPTER 7: Normal Development
CHAPTER 8: Developmental Variants
SECTION 3: Systematic Approach to Bone and Joint Abnormalities
CHAPTER 9: Principles of Radiographic Interpretation
CHAPTER 10: View Selection for the Radiographic Study
CHAPTER 11: Systematic Evaluation of Bone and Joint Abnormalities
SECTION 4: Positional and Developmental Abnormalities
CHAPTER 12: Principles of Biomechanical Radiographic Analysis of the Foot
CHAPTER 13: Foot Segmental Relationships and Bone Morphology
CHAPTER 14: Congenital and Developmental Pediatric Abnormalities
SECTION 5: Bone and Joint Disorders
CHAPTER 15: Fractures and Related Conditions: Fundamentals
CHAPTER 16: Classification of Fractures and Dislocations
CHAPTER 17: Osteonecrosis and Osteochondrosis
CHAPTER 18: Bone Infection
CHAPTER 19: Joint Disease
CHAPTER 20: Tumors and Tumorlike Lesions
CHAPTER 21: Metabolic Diseases, Nutritional Disorders, and Skeletal Dysplasias
CHAPTER 22: The Diabetic Foot
CHAPTER 23: Postoperative Evaluation and Complications
SECTION 6: Special Imaging Procedures
CHAPTER 24: Overview of Special Imaging Studies
CHAPTER 25: Magnetic Resonance Imaging of Foot and Ankle Pathology
CHAPTER 26: Musculoskeletal Ultrasound
Index

Citation preview

Thank you for purchasing this e-book. To receive special offers and news about our latest products, sign up below.

Or visit LWW.com

Foot and Ankle Radiology

Acquisitions Editor: Brian Brown Product Development Editor: David Murphy Editorial Assistant: Lindsay Burgess Production Project Manager: Alicia Jackson Design Coordinator: Elaine Kasmer Manufacturing Coordinator: Beth Welsh Marketing Manager: Dan Dressler Prepress Vendor: S4Carlisle 2nd edition Copyright © 2015 Wolters Kluwer Health 1st edition © Churchill Livingstone 2003 All rights reserved. This book is protected by copyright. No part of this book may be reproduced or transmitted in any form or by any means, including as photocopies or scanned-in or other electronic copies, or utilized by any information storage and retrieval system without written permission from the copyright owner, except for brief quotations embodied in critical articles and reviews. Materials appearing in this book prepared by individuals as part of their official duties as U.S. government employees are not covered by the above-mentioned copyright. To request permission, please contact Wolters Kluwer Health at Two Commerce Square, 2001 Market Street, Philadelphia, PA 19103, via email at [email protected], or via our website at lww.com (products and services). 987654321 Library of Congress Cataloging-in-Publication Data Christman, Robert A., author.

Foot and ankle radiology / Robert A. Christman. — Second edition. p. ; cm. Includes bibliographical references and index. ISBN 978-1-4511-9283-4 (alk. paper) — ISBN 1-4511-9283-5 (alk. paper) eISBN 978-1-4963-0439-1 I. Title. [DNLM: 1. Foot—radiography. 2. Ankle—radiography. 3. Foot Diseases— radiography. WE 880] RC951 617.5’8507572—dc23 2014031879 This work is provided “as is,” and the publisher disclaims any and all warranties, express or implied, including any warranties as to accuracy, comprehensiveness, or currency of the content of this work. This work is no substitute for individual patient assessment based upon healthcare professionals’ examination of each patient and consideration of, among other things, age, weight, gender, current or prior medical conditions, medication history, laboratory data and other factors unique to the patient. The publisher does not provide medical advice or guidance and this work is merely a reference tool. Healthcare professionals, and not the publisher, are solely responsible for the use of this work including all medical judgments and for any resulting diagnosis and treatments. Given continuous, rapid advances in medical science and health information, independent professional verification of medical diagnoses, indications, appropriate pharmaceutical selections and dosages, and treatment options should be made and healthcare professionals should consult a variety of sources. When prescribing medication, healthcare professionals are advised to consult the product information sheet (the manufacturer’s package insert)

accompanying each drug to verify, among other things, conditions of use, warnings and side effects and identify any changes in dosage schedule or contradictions, particularly if the medication to be administered is new, infrequently used or has a narrow therapeutic range. To the maximum extent permitted under applicable law, no responsibility is assumed by the publisher for any injury and/or damage to persons or property, as a matter of products liability, negligence law or otherwise, or from any reference to or use by any person of this work. LWW.com

DEDICATION To my wife, Irene, and my daughter, Jessica, for their love and support To my students—past, present, and future—for their encouragement and inquisitiveness To the Lord God, for all his provisions and blessings

Contributors Albert V. Armstrong, Jr. DPM Associate Professor of Radiology School of Podiatric Medicine Barry University Miami Shores, Florida Staff Podiatrist Department of Surgery Mount Sinai Medical Center Miami Beach, Florida Robert L. Baron, DPM, FACPR, FACFAOM President and Founder, Lower Extremity Imaging Consultants Podiatric Consultant for MetisMD Professor and Chair (Retired), Department of Radiology William M. Scholl College of Podiatric Medicine Rosalind Franklin University of Health Sciences / The Chicago Medical School Private Practice Chicago,Illinois Fellow, American College of Podiatric Radiology

Fellow in Primary Podiatric Medicine, ABPOPPM Fellow in Podiatric Orthopedics, ABPOPPM Philip J. Bresnahan, DPM, FACFAS Assistant Professor Department of Podiatric Medicine Temple University Philadelphia, Pennsylvania Chairman Division of Podiatry, Department of Surgery Grandview Hospital Sellersville, Pennsylvania Jacqueline M. Brill, DPM, FACFAS Assistant Professor Department of Podiatric Surgery Barry University School of Podiatric Medicine Miami Shores, Florida Residency Director Chief of Podiatric Surgery Department of Surgery

Mount Sinai Medical Center Miami Beach, Florida Francis L.S. Chan, DPM Assistant Professor Department of Podatric Medicine, Surgery and Biomechanics College of Podiatric Medicine Western University of Health Sciences Pomona, California Attending Podiatrist Department of Orthopaedics Riverside County Regional Medical Center Moreno Valley, California Robert A. Christman, DPM, EdM Professor with Tenure College of Podiatric Medicine Western University of Health Sciences Pomona, California Past Director of Radiology and Associate Professor, Temple University School of Podiatric Medicine, Philadelphia, Pennsylvania Fellow, American College of Podiatric Radiology (FACPR) Fellow, American College of Foot and Ankle Orthopedics and Medicine

(FACFAOM) Chair, Podiatric Imaging Council, ACFAOM Co-Chair, Radiological Task Force, California Podiatric Medical Association Past Member, Radiation Protection Advisory Committee, Department of Environmental Resources, Commonwealth of Pennsylvania Past Member, Limited Scope of Practice in Radiography Examination Committee, American Registry of Radiologic Technologists (ARRT) Randy E. Cohen, DPM, FACFR Professor and Director Radiology Section, Department of Medical Sciences New York College of Podiatric Medicine New York, New York Podiatrist Department of Surgery Staten Island University Hospital Staten Island, New York Fellow, American College of Podiatric Radiology (FACPR) Corine L. Creech, DPM Resident Podiatric Surgical Residency Program Temple University Hospital

Philadelphia, Pennsylvania Diane M. Koshimune, DPM Assistant Professor Department of Podiatric Medicine, Surgery and Biomechanics Western University of Health Sciences Pomona, California Jonathan Labovitz, DPM Associate Professor Department of Podiatric Medicine, Surgery, and Biomechanics Medical Director Foot and Ankle Center College of Podiatric Medicine Western University of Health Sciences Pomona, California Cristina Marchis-Crisan, MD, DPM Podiatric Physician Volunteer Free Clinics of Southwest Washington Vancouver, Washington Past Assistant Professor Department of Medicine, Surgery, and Biomechanics

College of Podiatric Medicine Western University of Health Sciences Pomona, California Andrew J. Myer, DPM Associate Professor Department of Podiatric Surgery Temple University School of Medicine Philadelphia, Pennsylvania Askone Nouvong, DPM Assistant Professor College of Podiatric Medicine Western University of Health Sciences Pomona, California Associate Clinical Professor Department of Vascular Surgery University of California Los Angeles David Geffen School of Medicine Los Angeles, California Mary C. Oehler, RT (R) (N) (ARRT) Radiology Supervisor

Radiology Department Temple University School of Podiatric Medicine Philadelphia, Pennsylvania Lawrence Steven Osher, B.S., DPM Professor Department of Podiatric Medicine/Radiology Kent State University College of Podiatric Medicine Independence, Ohio Staff Department of Surgery University Hospital East/Richmond Richmond Hts., Ohio Rocco A. Petrozzi, DPM Assistant Professor, Division Head Podiatric Medicine Division Kent State University College of Podiatric Medicine Independence, Ohio Robin C. Ross, DPM Attending Staff Department of Surgery

Eastern Long Island Hospital Greenport, New York Marcos Loreto Sampaio, MD Assistant Professor Department of Radiology University of Ottawa Musculoskeletal Radiologist Department of Diagnostic Imaging The Ottawa Hospital Ottawa, Ontario Canada William H. Sanner, DPM Chairman (Retired) Department of Podiatric Medicine and Surgery Ochsner Clinic Foundation Baton Rouge, LA Mark E. Schweitzer, MD Chief, Diagnostic Imaging Department of Diagnostic Imaging The Ottawa Hospital - General Campus

Ottawa, Ontario Canada Professor and Chair of Radiology Stony Brook University Stony Brook, New York Adnan Sheikh, MD Associate Professor Department of Medical Imaging University of Ottawa Radiologist Department of Medical Imaging The Ottawa Hospital Ottawa, Ontario Canada Anthony J. Short, MPod BAppSc(Pod), FFPM RCPS(Glasg) Visiting Lecturer Podiatry Department School of Clinical Sciences Queensland University of Technology Visiting Podiatrist

Queensland Diabetes & Endocrine Centre Mater Health Services Brisbane, Queensland Australia Frank Spinosa, DPM Past Associate Professor of Radiology Department of Radiology New York College of Podiatric Medicine Attending Department of Surgery Eastern Long Island Hospital Greenport, New York Casmir F. Strugielski, RT (R) BS Retired Associate Professor Department of Radiology Scholl College of Podiatric Medicine Rosalind Franklin University North Chicago, Illinois John Tassone Jr., DPM Associate Professor

Arizona School of Podiatric Medicine Midwestern University Podiatrist Thunderbird Internal Medicine Banner Thunderbird Wound Care Center Glendale, Arizona Jacqueline Truong, DPM, MPH Assistant Professor and Chairman Department of Podiatric Surgery, Medicine and Biomechanics College of Podiatric Medicine Western University of Health Sciences Pomona, California Kendrick A. Whitney, DPM Associate Professor Department of Biomechanics Temple University School of Podiatric Medicine Philadelphia, Pennsylvania Marie L. Williams, DPM, FACFAS Associate Professor Department of Podiatry

Barry University School of Podiatric Medicine Miami Shores, Florida Program Director Department of Podiatry Aventura Hospital and Medical Center Aventura, Florida Fellow, American College of Foot and Ankle Surgeons David E. Williamson District Manager (Retired) Photo Products Department Medical Products Division E.1. DuPont de Nemours & Company, Inc. Wilmington, Delaware

Foreword to the first Edition Dr. Robert Christman has provided us, in his unique text, insights into his personal approach to evaluating radiographs of the foot. How to logically analyze the cardinal radiographic findings seen in a radiograph and its most appropriate list of differential diagnoses is made easy. In his chapter dealing with joint disease, the reader is led step-by-step into constructing practical and concise differential diagnoses. This approach is rare in most texts dealing with this important subject. Dr. Christman’s emphasis on radiographic anatomy and variants are truly appreciated. In most texts this section is only skimmed. Also, sections dealing with radiographic biomechanical analysis correlating radiographic positional finding to foot structure and function as well as fracture and dislocation classifications using actual radiographs are included. An entire section is also dedicated to dealing with the radiologic sciences. Dr. Christman has succeeded in including everything necessary for a podiatric student to learn as well as practice podiatric radiology and radiologic interpretation. The book should serve as a reference for all podiatrists, students, and podiatric assistants. Harvey Lemont, DPM

Preface The goals of Foot and Ankle Radiology, Second Edition, remain the same as of the first edition: to introduce the podiatric medical student to the scope of diagnostic imaging applicable to podiatric medicine, to prepare the podiatric medical student to apply podiatric radiography and radiographic interpretation in practice, and to provide the podiatric practitioner with a comprehensive base of knowledge to make informed decisions. I am honored that the first edition stood the test of time and provided value to those who used it. I am also blessed to have had the opportunity to author and edit this second edition. Significant time was spent editing the entire textbook so that it flowed consistently throughout. Any images in the first edition that poorly demonstrated what was being described were replaced with new ones. There were some comments regarding the first edition that some chapters were hard to read. Hopefully, the reader will find that this edition has corrected that, for the most part. Generally speaking, the six sections remain the same; however, the Special Imaging Procedures section has been moved to the end since the focus of the textbook is on radiology. Also, one chapter, Principles of Radiographic Interpretation, has been moved from the Radiographic Anatomy section into the Systematic Approach to Bone and Joint Abnormalities section where it more aptly belongs. Four new chapters have been added, reflecting their increasing importance in podiatric imaging: Digital Radiography, The Diabetic Foot, Musculoskeletal Ultrasound, and Postoperative Evaluation and Complications. Section 1, Radiography, has been significantly edited and includes a new chapter. Chapter 1, Radiologic Physics, Radiobiology, and Radiation Safety, has been rewritten. Chapter 2, Film–Screen Radiography, combines several of the first edition chapters: Radiography Equipment Considerations and Accessories, Exposure Techniques and Special Considerations, and Film Quality. With the advent of digital radiography, Chapter 3 is an entire chapter

devoted to this subject. Not much has changed regarding positioning techniques (Chapter 4); however, newer terminology has been referenced relating to recent changes made in two major radiographic positioning textbooks (Merrill’s Atlas and Bontrager). Chapter 5, The Normal Foot and Ankle, and Chapter 7, Normal Development, from Section 2, Radiographic Anatomy, did not require much editing because little has changed in this area. However, Chapter 6, Normal Variants and Anomalies, has been expanded and updated. In addition to being more comprehensive, a number of new illustrations have been included as well. Chapter 8, Developmental Variants, has also been updated. Regarding Section 3, Systematic Approach to Bone and Joint Abnormalities, the basic concepts have been further clarified throughout. In addition, Chapter 9, Principles of Radiographic Interpretation, has been expanded to include digital imaging interpretation tools. Chapter 10, View Selection for the Radiographic Study, has been updated to include the Ottawa ankle rules and reference to the American College of Radiology’s appropriateness criteria for musculoskeletal imaging. Chapter 11, Systematic Evaluation of Bone and Joint Abnormalities, has been updated and expanded to include greater illustration of the concepts described. A new topic, the radiology report, has been added, along with a guideline. My former colleagues from Temple University School of Podiatric Medicine, Ken Whitney and Phil Bresnahan, have updated Section 4, Positional and Developmental Abnormalities. No significant content-related changes were made to Chapter 12, Principles of Biomechanical Radiographic Analysis of the Foot. However, several illustrations were added to Chapter 13, Foot Segmental Relationships and Bone Morphology, by Ken’s father, Alan Whitney, who was a former teacher of ours. Existing illustrations have been improved for better clarification of concepts. Chapter 14, Congenital and Developmental Pediatric Abnormalities, has been updated and expanded to include the topic pes cavus that was inadvertently missing from the first edition. Section 5, Bone and Joint Disorders, has been significantly updated and expanded. Diane Koshimune, a colleague at my new institution, Western University of Health Sciences’ College of Podiatric Medicine, reworked and

updated Chapter 15, Fractures and Related Conditions: Fundamentals. Chapter 16, Classification of Fractures and Dislocations, has been entirely rewritten with several other colleagues from WesternU, Jonathan Labovitz, Francis Chan, and Aksone Nouvong, emphasizing the reliability of these classification systems and their ability (or lack of) to predict outcomes. Dr. Cristina Marchis-Crisan provided substantial updating and expansion to three chapters: Chapter 17, Osteonecrosis and Osteochondrosis, Chapter 19, Joint Disease, and Chapter 21, Metabolic Disease, Nutritional Disorders, and Skeletal Dysplasias. The medical description of disease processes has been included, as has the inclusion of special imaging studies and their application. Many tables have been added to help clarify the subject matter; additional images have been included as well. The categorization of and differentiation between joint disorders has also been updated. Chapter 18, Bone Infection, has been updated to include several methods of classification as well as the newer special imaging studies. Chapter 20, Tumors and Tumorlike Lesions, has also been rewritten by my colleague from Kent State University College of Podiatric Medicine, Larry Oshe, including descriptions of additional lesions. Two new chapters have been added to Section 5. The Diabetic Foot, Chapter 22, required a chapter of its own having become a major public health challenge in the Western and emerging worlds. My colleague from “down under,” Tony Short, addresses the following: selecting the appropriate imaging study, general considerations, Charcot neuropathic osteoarthropathy, infection, and peripheral arterial disease. Chapter 23, Postoperative Evaluation and Complications, covers a subject that was overlooked in the first edition. Coincidentally Andy Meyr, a former student and now colleague in academia, was contacted by a radiologist to put something together on this topic for radiologists. I proposed that he submit it as a chapter for this edition. I’m glad he agreed! Dr. Meyr addresses the assessment of deformity correction, serial assessment of osseous healing, and verification of hardware. There have been many advances in diagnostic imaging since the first edition, and, hence, Section 6, Special Imaging Procedures, has been updated and expanded to include these developments and modalities. Chapter 24, Overview of Special Imaging Studies, now has a section on cellular and molecular imaging, including positron emission tomography (PET) and

single-photon emission computed tomography (SPECT). Computed tomography (CT) has been expanded to include the newer generations of spiral or helical CT and multidetector CT. The magnetic resonance imaging (MRI) and ultrasound sections have been expanded to include basic principles and an overview of how these images are obtained. Chapter 25, Magnetic Resonance Imaging of Foot and Ankle Pathology, has been rewritten and includes many images of foot and ankle pathology. I first met Mark Schweitzer many years ago when he was a Fellow in Musculoskeletal Radiology at Thomas Jefferson University Hospital in Philadelphia, Pennsylvania. He has since published numerous articles relating to MRI of the foot and ankle. And, because ultrasound has recently emerged as the “podiatrist’s stethoscope,” with increasing popularity in sports medicine in particular, Chapter 26, Musculoskeletal Ultrasound, has been devoted to obtaining images of anatomic structures and interpretation of related pathology. Al Armstrong, my colleague from Barry University School of Podiatric Medicine, has provided a comprehensive pictorial manual.

preface to the first Edition The purpose of Foot and Ankle Radiology is threefold: to introduce the podiatric medical student to the scope of diagnostic radiology applicable to podiatric medicine, to prepare the podiatric medical student to apply podiatric radiography and radiographic interpretation in practice, and to provide the podiatric practitioner with a comprehensive base of knowledge to make informed decisions. Radiology is an important diagnostic tool useful for the evaluation of foot and ankle pathology. However, because of the potential risks associated with ionizing radiation, those involved in its production must have knowledge of the radiologic sciences in order to provide protection and safety to all involved. The Doctor of Podiatric Medicine is responsible for the practice of proper radiography in the office setting, whether by himself or herself or by an appropriate assistant/technologist. The doctor must also be familiar with all special imaging studies applicable to imaging of the foot and ankle so that they are ordered as warranted. The podiatric physician encounters numerous pathologic conditions radiographically that are either intrinsic to the foot or represent manifestations of extrinsic disease. Therefore the student must not only learn specific radiologic pathology of the foot and ankle, but must acquire an understanding of general diagnostic radiology and pathologic correlation. Furthermore the student must learn how to analyze a radiograph systematically and acquire a basic knowledge of bone radiology to interpret radiographs and establish differential diagnoses. To these ends, this text includes sections on plain film radiography (radiologic science, radiation protection and safety, principles of radiography, and foot and ankle radiography); radiographic anatomy (normal and variant presentations of the adult and developing foot and ankle); systematic approach to bone and joint abnormalities (how to select appropriate views and a using a fundamental process to analyze the radiographic study);

radiographic biomechanical analysis (correlating the clinical and radiographic presentations of both the adult and child); special imaging procedures (emphasizing indications and cross-sectional imaging); and bone and joint disorders (including systematic approaches to interpreting skeletal pathology rather than simply description by disease). Podiatric radiology textbooks have come and gone over the years; at this writing, not one is in print. A scattering of foot and ankle radiology texts written by radiologist or orthopedist is presently available; however, they are geared to the practicing physician and not the student. As a result, they are devoid of sections pertinent to the podiatrist in training, in particular radiologic science and systematic assessment of skeletal abnormalities. Furthermore this textbook devotes serious attention to plain film radiographic anatomy and radiographic biomechanical analysis. An entire chapter is devoted to fracture and dislocation classification systems, classically found not in radiology textbooks but orthopedic texts. Also unique to this text is the inclusion of a chapter on podiatric radiography equipment. Foot and Ankle Radiology will also serve as a valuable reference source to the podiatric assistant, radiologic technologist, and limited license extremity technologist. Also, the radiologist and orthopedist may find this text to be a valuable addition to their library. This text was written with the following goals in mind. Specifically, that the podiatric student and practitioner be able to: • Describe the principles of radiation physics and biology. • Practice appropriate radiation protection and safety.

Acknowledgments I would like to thank Lippincott Williams & Wilkins and Brian Brown, executive editor, for agreeing to take on the publication of this second edition. I would also like to thank Dave Murphy, product manager, for his patience, as it took longer than I expected to complete and submit the manuscript!

Contents Contributors Foreword Preface Preface to the 1st edition Acknowledgments SECTION 1 Radiography CHAPTER 1 Radiologic Physics, Radiobiology, and Radiation Safety   Robert A. Christman CHAPTER 2 Film—Screen Radiography   Robert A. Christman, Mary Oehler, Casimir F. Strugielski, and David E. Williamson CHAPTER 3 Digital Radiography   Robert A. Christman and Mary Oehler CHAPTER 4 Positioning Techniques and Terminology

  Robert L. Baron, Casimir F. Strugielski, and Robert A. Christman SECTION 2 Radiographic Anatomy CHAPTER 5 The Normal Foot and Ankle   Robert A. Christman CHAPTER 6 Normal Variants and Anomalies   Robert A. Christman CHAPTER 7 Normal Development   Robert A. Christman and Jacqueline Truong CHAPTER 8 Developmental Variants   Robert A. Christman and Jacqueline Truong SECTION 3 Systematic Approach to Bone and Joint Abnormalities CHAPTER 9 Principles of Radiographic Interpretation   Robert A. Christman

CHAPTER 10 View Selection for the Radiographic Study   Robert A. Christman CHAPTER 11 Systematic Evaluation of Bone and Joint Abnormalities   Robert A. Christman SECTION 4 Positional and Developmental Abnormalities CHAPTER 12 Principles of Biomechanical Radiographic Analysis of the Foot   William H. Sanner and Kendrick A. Whitney CHAPTER 13 Foot Segmental Relationships and Bone Morphology   William H. Sanner and Kendrick A. Whitney CHAPTER 14 Congenital and Developmental Pediatric Abnormalities   Philip J. Bresnahan SECTION 5 Bone and Joint Disorders CHAPTER 15

Fractures and Related Conditions: Fundamentals   Diane M. Koshimune, Frank A. Spinosa, Robin C. Ross, and Robert A. Christman CHAPTER 16 Classification of Fractures and Dislocations   Jonathan Labovitz, Francis Chan, Aksone Nouvong, and Robert A. Christman CHAPTER 17 Osteonecrosis and Osteochondrosis   Robert A. Christman, Cristina Marchis-Crisan, and Randy E. Cohen CHAPTER 18 Bone Infection   Marie Williams and Robert A. Christman CHAPTER 19 Joint Disease   Cristina Marchis-Crisan and Robert A. Christman CHAPTER 20 Tumors and Tumorlike Lesions   Lawrence Osher, Rocco Petrozzi, and Robert A. Christman CHAPTER 21 Metabolic Diseases, Nutritional Disorders, and Skeletal Dysplasias

  Robert A. Christman, John Tassone Jr, and Cristina Marchis-Crisan CHAPTER 22 The Diabetic Foot   Anthony J. Short CHAPTER 23 Postoperative Evaluation and Complications   Andrew J. Meyr and Corine L. Creech SECTION 6 Special Imaging Procedures CHAPTER 24 Overview of Special Imaging Studies   Robert A. Christman and John Tassone Jr CHAPTER 25 Magnetic Resonance Imaging of Foot and Ankle Pathology   Adnan Sheikh, Marcos Loreto Sampaio, and Mark E. Schweitzer CHAPTER 26 Musculoskeletal Ultrasound   Albert V. Armstrong Jr and Jacqueline M. Brill Index

SECTION 1 Radiography

1 Radiologic Physics, Radiobiology, and Radiation Safety ROBERT A. CHRISTMAN RADIOLOGIC PHYSICS     What is radiation? What are x-rays? How are x-rays produced? How do x-rays interact with matter? Basic Concepts Radiation is defined as energy that is emitted and transferred through space. Examples include radio waves, visible light, and x-rays. Matter is exposed or irradiated when it intercepts and absorbs radiation. Some forms of radiation can remove an orbital electron from an atom with which it interacts; this process is known as ionization, and the radiation is called ionizing radiation. There are two units of radiation measurement: the US customary system and the International System (SI); the latter has been officially adopted by all other countries. Although the United States does not use SI units for the public, all scientific inquiries use the SI system (Box 1-1). The unit for radiation exposure measured in air is the Roentgen or air kerma. The unit for the radiation absorbed dose is the rad or gray (Gy). The unit of effective dose and radiation received by radiation workers is referred to as rem (“roentgen-equivalent-man”) or sievert (Sv). A quantity of radioactive material is referred to as curie (Ci) or becquerel (Bq). BOX 1-1 Radiation Units Traditional Unit Roentgen (R) Rad

SI Unit Air kerma (Gya) Gray (Gyt)

Rem

Sievert (Sv)

Quantity of Exposure Absorbed dose Effective dose equivalent

Curie (Ci)

Becquerel (Bq)

(E) Radioactivity

Radiation can be classified in three ways: as particulate (alpha and beta particles) or nonparticulate (electromagnetic radiation—light, x-rays, radio waves); charged (alpha particles) or uncharged (x-rays); and ionizing (x-rays, gamma rays) or nonionizing (visible light, radio waves). Based on these three classifications, x-rays are nonparticulate, uncharged, and ionizing. Radioactivity is defined as the emission of energy and particles by unstable atoms, radionuclides, to become stable. A radionuclide is an atom with a nucleus that undergoes radioactive decay. Radioisotopes have the same atomic number but a different atomic mass number (neutrons); most are artificially produced. Radioactive decay results in the emission of beta particles (a neutron converts into a proton and an electron-like particle is ejected), alpha particles (which consists of two protons and two neutrons bound together), and gamma rays (which are usually emitted simultaneously with a particle). Gamma rays, however, are a form of electromagnetic (nonparticulate) radiation that is useful in nuclear medicine. Radioactive half-life is the time required for a quantity of radiation to be reduced to one-half of its original value. Seven half-lives are required before a quantity of radioactive material has decayed to less than 1%. When discussing electromagnetic (nonparticulate) radiation, it is important to understand the relationship between the radiation’s wavelength and its energy (Box 1-2). Wavelength and energy are inversely proportional: the longer the wavelength, the lower its energy, and the shorter the wavelength, the greater its energy. Radio waves have the longest wavelength on the electromagnetic spectrum and, therefore, have the lowest energy. Magnetic resonance imaging uses radio waves. Visible light is important in diagnostic radiology because light is necessary in order to view the image on a view box or on a computer display. X-rays, used in both computed tomography and conventional radiography, have very short wavelengths and high energy. Lastly, gamma rays are used in nuclear medicine and have the highest energy. BOX 1-2 The Electromagnetic (EM) Spectrum Applicable to Diagnostic

Imaging

Ionizing radiation is either particulate or electromagnetic. Box 1-3 lists the types of ionizing radiation, where they originate, and whether or not they are charged. Sources of ionizing radiation are either natural or man-made. Natural sources represent 82% of human exposure and include cosmic rays, terrestrial (radon is the largest component), and radionuclides (such as potassium-40). The primary source of man-made ionizing radiation (18% of human exposure) is medical x-rays. Other man-made sources include nuclear medicine and consumer products. BOX 1-3 Types of Ionizing Radiation PARTICULATE (ORIGINATES FROM NUCLEUS, CHARGED) •  Alpha particles    –  4 atomic mass number, positive charge    –  Emitted from nuclei of heavy elements    –  High energy, short range •  Beta particles    –  No mass, positive or negative charge    –  Lower energy and longer range ELECTROMAGNETIC (NO CHARGE, NO MASS) •  x-rays (originate in electron cloud, outside nucleus) •  gamma rays (originate in nucleus of radioisotope)

The photon is the smallest quantity of any type of electromagnetic radiation. It is also referred to as a quantum or bundle of energy. It travels at the speed of light, has no mass, and has no charge. Its frequency and wavelength are inversely related. In particular, it has a wave–particle duality, meaning it may behave as either a wave or a particle. X-rays are similar to light waves by two of three phenomena: x-rays can transmit through matter and can be absorbed by matter, but matter does not reflect them. The most important interaction that the x-ray has with matter is its ability to be partially absorbed, known as attenuation. If x-rays totally penetrated matter, the resultant image would be black. If x-rays were totally absorbed by matter, the resultant image would be white. Partial absorption or attenuation of x-rays by matter is what gives the different shadows or shades of gray that are the actual image itself. Matter that absorbs x-rays will appear white in the image and is referred to as radiopaque; matter that attenuates xrays will be radiolucent and appear darker in the radiograph. Light intensity emitted from a source decreases rapidly as the distance from the source increases. The inverse square law defines this principle, where intensity is inversely proportional to the square of the distance between the object and the source (Box 1-4). Three of the four variables must be known in order to apply the inverse square law. Unlike light, which is identified by wavelength, x-rays are identified by energy in electron volts (eV). Examples of x-ray energies used in medicine include 10 to 20 kVp (dermatology), 30 to 150 kVp (diagnostic imaging), and 200 to 1000 kVp (deep tissue therapy). X-rays travel in straight lines and can travel through a vacuum. Remember that photon energy is directly proportional to photon frequency and is inversely proportional to photon wavelength. Nonionizing radiation has low energy and long wavelength. Ionizing radiation has high energy and short wavelength. BOX 1-4 Inverse Square Law

Example: If x-ray intensity is 1 rad at 2 ft, what would the intensity be at 4 ft?

Therefore, I2 = 0.25 rad Photons interact with matter most easily when matter is approximately the same size as the photon’s wavelength. For example, the television antenna measures in meters, which is the approximate wavelength of radiofrequencies. The wavelength of visible light is measured in micrometers, which is the size of the rods and cones in the eye. The wavelength of x-rays is similar to the size of atoms and electrons. Wilhelm Roentgen discovered x-rays in 1895 while investigating cathode rays (electrons) in a Crookes tube. A plate coated with barium platinocyanide, a fluorescent material, glowed while he experimented in a dark room, which he named “X-light.” He then interposed materials, including his hand, between the Crookes tube and the glowing plate. Roentgen was able to describe the properties of this unknown light in 1 month’s time and published the first medical radiograph in 1896. He received the first Nobel Prize for Physics in 1901. The X-ray Machine Diagnostic lower extremity–specific x-ray units are available either from dealers or directly from the manufacturer. Two types of lower extremity x-ray units are currently available: stationary and mobile. The stationary lower extremity x-ray unit (Figure 1-1A) is more common and connects to a platform known as an orthoposer. The tube head is attached to the orthoposer by an arm or a track. The mobile x-ray unit (Figure 1-1B) has a stand with wheels. The tube head is attached to a sliding track with counterbalance that is mounted to a movable stand instead of an orthoposer. The mobile unit can be moved easily from room to room; however, the wheels must be locked to prevent movement during the exposure. The components of the x-ray machine include the control console, high-

voltage generator, and the x-ray tube head. Control Console The operating or control console of a typical lower extremity x-ray unit either can be a separate box that is mounted inside or outside the x-ray room (Figure 1-1A) or may be incorporated as part of the tube head itself (Figure 1-1B). Three factors are under the control of the limited x-ray machine operator (LXMO): kVp (kilovoltage peak), mA (milliamperage), and exposure time (s). There also is an on/off control. kVp is a unit of electric potential and relates to the energy (quality) of the x-ray beam. A kVp between 55 and 65 is typically used for most foot and ankle applications. One can select between 50 and 70 kVp on most lower extremity x-ray units; higher kVp units are also available. mA represents the tube current and controls the number of photons that are produced. The mA of lower extremity x-ray units is quite low, between 10 and 30 mA. Some units may not have optional tube currents and are preset at the factory. The total quantity or output intensity of x-rays is known as the mAs, that is, milliamperage × time of exposure. A timer controls the length of exposure. Timers may be either synchronous or electronic in nature. Synchronous timers can be set as low as 1/60th of a second; electronic timers are even more accurate, to 1/100th of a second. Whereas mA and/or time manipulate the quantity of x-rays produced, kVp controls the energy or quality of the x-ray beam.

FIGURE 1-1. A: Stationary x-ray unit. The tube head is attached to the orthoposer at a fixed SID. (Courtesy of X-Cel X-Ray Corp., Crystal Lake, IL.) B: Mobile x-ray unit. (Courtesy of MinXray Inc., Northbrook, IL.) An exposure switch is attached to the control panel by a long, coiled cord. It allows the operator to stand a minimum of 6 ft away from the x-ray source. The exposure button operates as a “dead man” switch: the exposure only occurs as the button is depressed and will shut off at the selected time or if the button is released, whichever occurs first. High-Voltage Generator Lower extremity–specific x-ray units plug into a standard 110-V electrical outlet and are shockproof. A transformer is used to convert this relatively low electrical voltage to the high voltage (kilovolts) required by the x-ray tube to produce x-rays. In addition, the current from the wall is alternating and flows in two directions; a rectifier converts the alternating current (AC) into direct current (DC), which is required by the x-ray tube (electrons in the tube can only travel in one direction, from cathode to anode). The electric

current frequency (in the United States) is 60 Hz. The high-voltage generator may be single phase, three phase, or high frequency. With single-phase power generation, the voltage waveform pulsates between zero and its maximum value from 60 to 120 times per second, depending on whether the rectification is half wave or full wave, respectively. (No x-rays are produced at zero.) This is also referred to as 100% voltage ripple (i.e., the variation between the maximum and minimum voltage is 100%). In contrast, the voltage across the x-ray tube is nearly constant and never drops to zero with a three-phase power generation. The resultant voltage ripple is less than 15%. A high-frequency generator is found in newer lower extremity–specific x-ray units; its voltage waveform is nearly constant and has less than 1% ripple, maintaining constant high voltage. They are smaller, less expensive, and more efficient than the singleand three-phase power options. The X-ray Tube Head Components of the x-ray tube head include the supporting arm, x-ray tube, and beam limitation device. The x-ray tube head is held firmly in place by a supporting arm. Any movement of the tube head during the exposure can impair the quality of the final image. Depending on the type of x-ray unit one owns or operates, the arm may originate from an orthoposer or from a mobile stand (Figure 1-1). Arms that originate from an orthoposer are convenient to use because the distance between the x-ray tube and the image receptor, the source-to-image distance (SID), is fixed or limited. (The fixed SID aids in preventing technical errors that can result in poor-quality images.) In contrast, this distance must be determined manually when using a mobile unit; these units may be equipped with a measuring tape or a telescoping antenna-like device for determining the SID. Positioning of the tube head with the mobile unit is not as simple a task as positioning the stationary units with fixed SID. Inconstant SID can have a profound effect on the optical density of the final image if other technical factors are not calculated and adjusted accordingly. According to the NCRP (National Council on Radiation Protection and Measurements), the minimum distance allowable between the patient and the x-ray tube is 12 inches.

The x-ray tube is contained within a protective housing. The protective housing is filled with oil that surrounds the x-ray tube and provides electrical insulation and absorbs heat. The x-ray tube itself is enclosed in glass or metal and maintains a vacuum. There is a small window in the glass or metal tube enclosure that is thinner, which is where the x-rays escape and are directed toward the patient’s area of concern. Accordingly, the housing also has a small window where the x-rays for the study are emitted after exiting the tube enclosure; these x-rays form the useful beam. The cathode is the negative electrode of the x-ray tube; it consists of a filament and a focusing cup (Figure 1-2A). The filament heats up and glows similar to a light bulb filament. However, instead of producing light, the cathode filament produces electrons. The filament is usually made of tungsten and, because of its high melting point, will not melt at high temperatures under normal circumstances. The electrons that are emitted by the filament are then directed toward the anode by the focusing cup. The tube current, or the number of electrons produced by the cathode, is determined by the temperature of the filament and is measured in milliamperes.

FIGURE 1-2. The x-ray tube. A: Cathode (C) with focusing cup. B: Cathode and anode. (Image B is from Figure 1-2 in Martins B: Radiation physics, biology and safety. Chapter 1 in Christman RA: Foot and ankle radiology, 1st edition, Churchill Livingstone, St. Louis, MO, 2003.) The anode is the positive electrode of the x-ray tube (Figure 1-2B). Electrons produced at the cathode are directed toward a small tungsten target on the anode. The area on the target from which x-rays is emitted is known as the focal spot. The focal spot of lower extremity x-ray units is approximately 1 × 1 mm. Because a large amount of heat is generated by this bombardment of

electrons and subsequent formation of x-rays, the anode (except for the target) is made of copper for efficient heat dissipation. Anodes can be either stationary or rotating. General-purpose x-ray units operate at high tube currents (up to 1200 mA) and produce excessive amounts of heat at the target. The targets of these particular units are, therefore, located on a disk that rotates at a high rate. Heat can then be transferred across a larger surface area, thereby reducing the chance of damage to the tungsten target. In contrast, lower extremity–specific x-ray units operate at low tube currents (between 10 and 30 mA). Therefore, the targets of these units are attached to a less-expensive fixed, or stationary, anode. Production of x-rays at the anode is extremely inefficient; less than 1% of the kinetic energy from the projectile electrons is converted to x-rays, and the remainder (99%) is converted to thermal energy (heat). The tube enclosure and protective housing absorb off-focus radiation that does not contribute to the useful x-ray beam. This is called inherent filtration. X-rays that do escape through the housing are referred to as leakage radiation, which can unnecessarily expose the patient and LXMO. As per the NCRP, leakage radiation from the tube housing at 1 m is not to exceed 100 mR/h (1 mGya/h). The useful x-ray beam is also filtered, since it includes low-energy x-ray photons that do not contribute to the diagnostic image. This filter, added by the manufacturer, is known as added filtration and absorbs the low-energy x-ray photons that would otherwise be absorbed by the patient. Therefore, added filtration selectively absorbs low-energy x-rays that have no chance of reaching the image receptor, thereby increasing the active energy of the x-ray beam. It does not alter the quality of the radiographic image. Lower extremity x-ray units typically are supplied with 1.5- to 2.5-mm aluminum filters, meeting or exceeding the required total filtration for machines operating between 50 and 90 kVp. This thin sheet of aluminum is located between the x-ray tube housing and the collimator. Beam Limitation Device Beam limitation devices shape the dimensions of the useful x-ray beam so that only the area of interest is exposed to radiation. This process, known as

collimation, also regulates scatter radiation that can fog the film (conventional radiography), degrade contrast resolution (digital radiography), and be absorbed by the patient. At the very least, the primary beam must be restricted to no more than the size of the image receptor being exposed. In the past, cones and diaphragms were used to limit the beam; variable-aperture collimators are now standard equipment on lower extremity x-ray units and allow the LXMO to further limit the x-ray field size. Variable-aperture collimators use a light source that defines the field of the xray beam, hence the name light beam collimator (Figure 1-3). Depressing a small button activates the light source that remains on for a preset period of time. The tube head can then be positioned so that the x-ray beam is directed to the area in question. The area lit by the light source correlates to the area that will be exposed by the useful beam. The center of the light source is marked with crosshairs or a circle so that the central beam can be accurately positioned. Limiting the size of the x-ray beam is easily and accurately accomplished by adjusting shutters that controls the size of the x-ray beam. Variable-aperture collimators provided on lower extremity x-ray units usually consist of two sets of shutters or plates, oriented at 90° to one another, and are controlled by two knobs. A light source projects onto the patient so that the operator can adjust the aperture size and direct the useful beam over the body part of interest.

FIGURE 1-3. Variable-aperture collimator. (Courtesy of X-Cel X-Ray Corp., Crystal Lake, IL.)

X-ray Production So far you have learned that electrons are generated at the cathode end of the x-ray tube at the filament; the focusing cup helps direct electrons toward the target, which is part of the anode; and the projectile electrons interact with the atoms of the tungsten target, which is where x-rays are produced. X-rays are produced by two mechanisms: characteristic radiation and bremsstrahlung radiation. Characteristic radiation is produced when a projectile electron interacts with an atom’s inner-shell electron and removes that electron, resulting in ionization; x-rays are then produced as the outershell electrons fill the inner shells. Bremsstrahlung radiation is produced when a projectile electron slows down or brakes as it passes by the positively charged nucleus. The electrons change course and lose kinetic energy, resulting in the production of x-rays. The word bremsstrahlung is German in origin and means “braking.” The X-ray Emission Spectrum X-ray quantity is the number of x-rays, also known as output intensity, x-ray intensity, or radiation exposure. X-ray quality refers to the penetrability of the x-ray beam, that is, how much energy it has. An x-ray beam that has an overall high energy is referred to as a “hard” x-ray beam, whereas a beam that has an overall low energy is considered a “soft” x-ray beam. The x-ray emission spectrum refers to the range of x-ray energies present in a quantity of x-rays. X-rays produced by the characteristic radiation mechanism are very discreet in their energies. For example, an x-ray produced after interacting with a tungsten atom by ejecting a K-shell electron that is replaced by an L-shell electron is 57 keV. An x-ray produced after an L-shell electron is replaced by an M-shell electron is only 9 keV. In contrast, x-rays produced by the bremsstrahlung mechanism produce a continuous x-ray energy spectrum. For example, if the kVp setting is at 70, x-rays will be produced between 0 and 70 kVp; therefore, since attenuation of the x-ray beam is responsible for the formation of the diagnostic image, the heterogeneous bremsstrahlung radiation x-rays are going to be much more

valuable than those produced by characteristic radiation. The x-ray emission spectrum is influenced by mA (or mAs), kVp, and added filtration. Increasing mA (or mAs) increases the quantity of the x-ray beam but does not influence its quality. An increase in kVp results in an increase of both the quality and quantity of the x-ray beam. The presence of added filtration results in a decrease of the x-ray beam quantity but an increase in quality. Half-Value Layer The half-value layer (HVL) is a measurement of x-ray quality. The HVL of an x-ray beam is defined as the thickness of absorbing material necessary to reduce the x-ray intensity to half its original value. Recall that added filtration removes the low-energy x-ray photons that would otherwise be absorbed by the patient; this increases beam quality (HVL) but decreases the quantity of x-rays. Even though some very low-energy x-rays are absorbed by the tube housing (inherent filtration), the NCRP recommendations and state regulations require the use of added filters (usually made of aluminum) to harden the beam. The minimum amount of total filtration (inherent plus added) and the recommended HVLs are listed in Box 1-5. BOX 1-5 Minimum Total Filtration and HVLs Recommended by the NCRP kVp 49 50 60 70 71 80 90

Total Filtration (mm Al) 0.5 1.5 1.5 1.5 2.5 2.5 2.5

HVL (mm Al) 0.5 1.2 1.3 1.5 2.1 2.3 2.5

Line-Focus Principle (Stationary Anode) The actual focal spot is the location on the tungsten target where the electrons are directed from the cathode. A smaller actual focal spot will

produce a sharper image; however, because more heat is then concentrated in one particular area, the target could be damaged. The anode surface is tilted so that the x-rays strike a larger area, thereby spreading out the heat. The effective size of the focal spot, as seen by the image receptor, is smaller than the actual focal spot size. This is known as the line-focus principle. It allows a large area for heating while maintaining a small focal spot.

FIGURE 1-4. Heel effect. (From Figure 1-3 in Martins B: Radiation physics, biology and safety. Chapter 1 in Christman RA: Foot and ankle radiology, 1st edition, Churchill Livingstone, St. Louis, MO, 2003.) Heel Effect The useful x-ray beam has a nonuniform angular distribution, with the intensity on the cathode side being greater than on the anode side, an unfortunate consequence of the line-focus principle. This is known as the heel effect (Figure 1-4) and is more pronounced for large field size and

shorter SID, a common situation when using lower extremity–specific x-ray units. One way to exploit this apparent disadvantage is to position thicker parts being examined toward the cathode. Thus, when performing a dorsoplantar view of the foot, the thinner forefoot should be placed under the anode and the thicker rearfoot under the cathode. However, in reality, this is rarely knowingly applied. Tube Failure Since x-ray tubes are expensive, you want to avoid tube failure. Causes include a single excessive exposure, long exposure times, and filament vaporization. Since 99% of the energy of electrons hitting the anode is converted into heat, provisions must be made to dissipate this heat; if not, the anode will melt. The tube rating, given in heat units (HU), is a measure of the maximum load (kVp, mA, and exposure time) that can be applied to an x-ray tube. The manufacturer of the x-ray machine provides a tube-rating chart with the equipment. To prolong the life of an x-ray tube and to avoid permanent damage to it, always operate an x-ray unit so that the tube rating is not exceeded. X-ray Interaction with Matter X-rays interact with matter by five ways: coherent scattering, Compton scattering, photoelectric effect, pair production, and photodisintegration. The first three, coherent scattering, Compton scattering, and photoelectric effect, have application to podiatric radiography; the other two, pair production and photodisintegration, do not apply and hence will not be discussed. Coherent scattering is also known as elastic scatter, classical scatter, and Thompson scatter. As x-ray photons excite the target atom, a secondary x-ray photon is released of equal energy in a different direction. The patient usually absorbs this scatter because it primarily involves low-energy x-ray formation; therefore, this type of x-ray interaction with matter does not contribute to the diagnostic image but contributes to the patient-absorbed dose. Compton scattering occurs when the x-ray photon interacting with matter ejects the outer-shell electron, causing ionization. The original x-ray continues in a different direction with decreased energy. The scattered x-ray

creates a “fog” that impairs the image quality by reducing its contrast. The most important interaction with matter is known as the photoelectric effect. The x-ray photon is totally absorbed and an inner-shell electron is ejected during ionization. Characteristic radiation is then produced as an outer-shell electron fills the vacancy. The good news is that this process is responsible for x-ray beam attenuation and formation of the image; the bad news is that it contributes to patient-absorbed dose. Three types of x-rays are important to image formation: two have a positive influence, one a negative. X-rays that interact with matter by the photoelectric effect and those that pass through the subject and strike the image receptor (known as remnant radiation) form the diagnostic image. Compton scattering causes image fog. Unfortunately, most x-rays interact with matter by Compton scattering; only about 1% of x-rays emitted from the machine directly interacts with the image receptor to form the image. Using a low-kVp setting favors the photoelectric effect, which increases differential absorption and is desired. However, low kVp also increases patient dose. In contrast, high kVp favors Compton scattering. These factors are considered when choosing an appropriate radiographic exposure technique. Differential Absorption Attenuation is defined as the reduction x-ray beam intensity following the absorption and scattering of x-rays by matter. Attenuation depends on the density of matter, the atomic number, and its thickness. Mass density is defined as the quantity of matter per unit volume. Interaction between x-rays and tissue is proportional to the mass density of the tissue. Tissues with a higher effective atomic number, such as bone (13.8), will absorb x-rays to a greater degree than fat (6.3) or soft tissue (7.4), which is why bone appears whiter or radiopaque in the radiographic image. In contrast, fat and soft tissue will appear more black or radiolucent. This knowledge can be used to your advantage. Since cartilage, for example, is not visible in the radiograph, a radiopaque agent such as iodine (53) can be injected into the joint, which outlines the cartilage surface and makes it visible. Substances such as concrete (17) and lead (82) are used as primary and secondary barriers to block the transmission of x-rays.

RADIOBIOLOGY     Why should we be concerned about diagnostic x-ray studies?     How do x-rays cause damage? How can we prevent x-rays from causing damage? Radiobiology is the study of the effects of ionizing radiation on biologic tissue. Wilhelm Roentgen discovered x-rays more than 100 years ago in 1895. Within 1 year after x-rays were discovered, 23 cases of radiodermatitis were reported. Thomas Edison’s assistant died of radiation-related cancer 9 years later. In between the years of 1911 and 1914, three review articles detailed 54 cancer deaths and 198 cases of radiation-induced malignancy. The first official action to reduce radiation exposure to humans occurred in 1921, and the first dose-limiting recommendations were made in 1931. The established maximum permissible dose (MPD) at that time for radiation workers was 50 rem per year. The MPD was then lowered to 30 rem per year in 1936, 15 rem per year in 1948, and 5 rem per year in 1958. As of the year 2011, the whole body exposure for occupational workers remains at 5 rem per year. Ionizing radiation is harmful. The effects of large amounts have been seen in atomic bomb survivors, in victims of the Chernobyl radiation accident, and uranium mine workers, to name a few. However, it is not known whether there are any effects from diagnostic x-ray examinations. Data are not available to indicate if there is a threshold below which no harmful radiation effect will occur. Nor should we think of recommended permissible doses as perfectly safe. Therefore, when obtaining radiographic studies, radiation levels should be kept at the lowest practical level. Human Responses to Ionizing Radiation Stochastic effects are those responses to ionizing radiation where the probability of occurrence or incidence increases with dose. These effects generally do not have a threshold dose and are not observed for many months or years. The term stochastic is a Greek word meaning “chance.” Being exposed to radiation is like buying lottery tickets. One individual could buy a

few hundred lotteries and not win anything, whereas another individual could buy a single ticket and win a few million dollars. In the same manner, one individual may be exposed to a very large dose of radiation and develop no cancer, whereas another individual may be exposed to very little radiation and develop cancer. Just as with lotteries, there is no guarantee of anybody getting or not getting cancer, no matter what the dose. But the risk goes up directly (linear) with the magnitude of the dose. Stochastic effects result from low radiation doses over long time periods. Late human responses to ionizing radiation include leukemia and cancer of bone, lung, and breast. Compared to adults, infant radiation exposure produces up to four times the cancer risk; female infants are nearly double the risk of males. Local tissue damage has been observed in the skin (radiodermatitis) and eyes (cataract). Life-span shortening was a particular risk to radiographers and radiologists in the first half of the 20th century; however, recent studies have shown no relationship. Genetic effects also may occur. The late effects of ionizing radiation exposure are easy to observe but nearly impossible to associate; therefore, risk estimates are used. There are three types of risk estimates: relative risk, excess risk, and absolute risk. Relative risk estimates the late radiation effects in large populations without having any precise knowledge of the radiation dose; this is equivalent to the observed number of cases divided by the expected number of cases. If the incidence of pathology in an irradiated population exceeds that which is expected, then the difference would be excess risk. This is equal to the number of observed cases minus the number of expected cases. The third, absolute risk is the determination of a value from two different dose levels. Deterministic (nonstochastic) effects, in contrast, have a threshold dose, which must be exceeded before the response is observed. Once the threshold is exceeded, the severity of the response is proportional to the dose. Deterministic effects of ionizing radiation on humans are early and occur within minutes or days. An early human response to high-level exposure to ionizing radiation is acute radiation syndrome (“radiation sickness”), which includes death to the hematologic (5+ Sv), gastrointestinal (10+ Sv), and central nervous systems (20+ Sv). Local skin damage results in erythema and

desquamation. The ovaries and testes atrophy and may become sterile. Hematologic depression and cytogenic damage can also be early responses. Human fetal responses to ionizing radiation include prenatal and neonatal death, congenital malformation, childhood malignancy, and diminished growth and development. Effects of Ionizing Radiation on Tissues The effects of radiation on tissues are best described by the law of Bergonie and Tribondeau, which states that the sensitivity of tissues to radiation depends on two factors: proliferative capacity and differentiation. Tissues that are rapidly dividing are more sensitive to radiation than tissues whose cells are dividing slowly or are dormant. Tissues whose cells are differentiating are more sensitive to radiation than are tissues with fully differentiated cells. Based on the law of Bergonie and Tribondeau, bone marrow is considered the most sensitive tissue; lymphocytes are very sensitive, followed by the gastrointestinal tract. The central nervous system is the most resistant to radiation. Physical Factors Affecting Radiosensitivity Four factors affect radiosensitivity: linear energy transfer, relative biologic effectiveness, protraction, and fractionation. Linear energy transfer (LET) is a measure of the amount of energy transferred along the path of radiation. X-rays, gamma rays, and beta particles are low-LET radiations, protons are medium-LET, and alpha particles are high-LET radiations. High-LET radiations usually cause direct damage. The effectiveness of one form of radiation compared to another (using x-rays as the reference radiation) is the relative biologic effectiveness (RBE). It is usually calculated as the ratio of the dose that causes a certain effect for a given radiation, to the dose for the reference radiation that causes the same effect. The RBE depends on various factors, such as the end point, but, in general, higher-LET radiations have higher RBE values. The RBE of x-rays, gamma rays, and beta particles is one (i.e., 1 rad = 1 rem). However, 1 rad of alpha particles is equal to 20 rem. Protraction and fractionation are applied to the treatment of neoplasms. With

protraction, radiation dose is delivered continuously but at a lower dose rate; with fractionation, the same dose rate is delivered in equal fractions, each separated by a period of time. Biologic Factors Affecting Radiosensitivity Various biologic factors can influence the response of cells and tissues to radiation. Biologic tissues are more sensitive to radiation when irradiated in the oxygenated state. The age that is most sensitive to radiation is in utero, which then decreases through adulthood but increases in old age. Males may be slightly more sensitive than females. An irradiated cell is capable of recovering over time if it is not killed before its next division. Chemical agents that reduce radiation-induced damage are called radioprotectors. Molecules with the –SH (sulfhydryl) group, such as cysteine, act as radioprotectors; however, they are only effective in toxic doses and, therefore, not applicable to humans. Substances that enhance the effects of radiation are called radiosensitizers. Examples include vitamin K, methotrexate, and actinomycin D. Protectors and sensitizers are expected to be more effective in situations where the indirect effect is dominant, as with low-LET radiations. Radiation Dose–Response Relationships The radiation dose–response relationship can be linear nonthreshold, linear threshold, nonlinear nonthreshold, or nonlinear threshold. Linear means that the response is directly proportional to the dose, versus nonlinear, where the response is not directly proportional. Threshold means that no response will be produced at a dose below the threshold value; nonthreshold means that any dose, regardless of its size, is expected to produce a response. Generally speaking, the linear nonthreshold dose–response relationship is used to estimate the health effects related to ionizing radiation exposure. Irradiation of Macromolecules High radiation doses tend to kill cells, while low doses will cause damage to deoxyribonucleic acid (DNA, the genetic code). The major effects of ionizing radiation are main-chain scission, cross-linking, and point lesions. The

principal observable effects following DNA irradiation may be cell death, malignant disease, and genetic damage. Direct and Indirect Effect Ionizing radiation can have two effects on molecules that may lead to biologic damage: they are known as the direct and indirect effects of radiation. With the direct effect, ionizing radiation interacts directly with a crucial radiosensitive molecule such as DNA. The direct effect accounts for only a small fraction of damage caused by low-LET radiation, but accounts for most of the damage caused by high-LET radiation. The principal reaction that occurs as ionizing radiation interacts with the body is known as radiolysis of water, which produces free radicals. A free radical is defined as an uncharged molecule containing a single unpaired electron in the outermost electron shell. A free radical migrates from the initial site of ionization to a DNA molecule, transfers its excess energy, disrupts bonds, and causes point lesions; this is referred to as the indirect effect. Free radicals can also produce other poisonous products that are toxic to cells. Because a free radical is highly reactive, very unstable, and has a very short life-span (microseconds), most molecular damage occurs within a few seconds. The indirect effect is dominant for low-LET radiations. The target theory states that, for a cell to die after radiation exposure, its target molecule, that is, DNA, must be irradiated. The target molecule is then inactivated, which can occur through either the direct or indirect effect. The estimated, average annual whole body dose in those of us living in the United States from natural sources of ionizing radiation totals approximately 300 mrem (3 mSv) and is a result of internal radionuclides such as radioactive potassium-40 (K), terrestrial radionuclides (uranium, thorium), cosmic rays (the sun), as well as radon, which is the largest source. The estimated, average annual whole body dose in the United States from man-made sources of ionizing radiation, such as diagnostic x-ray studies (the largest) and consumer products (smoke detectors, etc.) totals 330 mrem (3.3 mSv). Therefore, those living in the United States are exposed on average to approximately 630 mrem (6.3 mSv) of ionizing radiation.

RADIATION PROTECTION AND SAFETY     How can we protect patients and office personnel from x-radiation? According to the NCRP, the specific objectives of radiation protection are (1) to prevent occurrence of deterministic effects by limiting doses to belowthreshold levels for such effects and (2) to limit the risk of stochastic effects, cancers, and genetic effects, giving due consideration to economic factors and to the needs and benefits of society. Any exposure to radiation must be justified, optimized to comply with the ALARA principle (maintain exposure to ionizing radiation at a level “As Low As Reasonably Achievable”), and limited so that individuals are not subjected to unacceptable risks. Radiation exposure must have specific benefits. The dose that individuals receive shall not exceed limits for appropriate circumstances. State and federal agencies license and/or regulate the use of radiation in the United States. National (NCRP) and international groups (ICRP, International Commission on Radiological Protection) provide the scientific basis and recommendations for radiation protection. To reduce patient exposure, the US government, through the Center for Devices and Radiological Health (CDRH), has established regulations for the design and manufacture of x-ray equipment. Thus, both the equipment and its use are regulated. Federal and state agencies conduct inspections to check compliance with the regulations. Protective Barriers Primary barriers provide protection from the primary (useful) x-ray beam; secondary barriers protect from scattered and leakage radiation. Barriers consist of materials that absorb x-rays, such as concrete or lead. Barrier thickness is determined after considering several factors: (1) the distance between the x-ray tube and the barrier, (2) human occupancy in areas other than the procedure room, (3) the number of x-ray examinations performed per week, (4) the maximal kilovoltage and milliamperage output of the x-ray unit, and (5) the direction of the primary beam. A radiation physicist’s

expertise can be quite helpful when designing a particular installation. In Report No. 147 (2004), the NCRP establishes criteria for structural shielding design of rooms or areas where radiation is used. Technique factors (kVp, mAs), amount of usage, location of the x-ray unit, and occupancy of surrounding areas determine the amount of shielding required. It is essential that a radiation survey be conducted after the equipment and shielding have been set up, to determine compliance with regulations. Patient Radiation Exposure When determining whether or not to order a radiographic study or any study that uses ionizing radiation for your patient, be certain to consider the potential benefits versus the potential adverse effects. Reduce unnecessary patient dose by only performing examinations that are justified clinically. Avoid repeat examinations by using the appropriate radiographic technique and positioning the patient appropriately the first time. A technique chart should be attached to the control panel of each x-ray unit. This chart can reduce the number of reexposures resulting from improper technical factors and is required by many states. It should include the mA, kVp, and time of exposure for each positioning technique at the SID recommended by the unit’s manufacturer. A caliper is a measuring device used to determine the thickness of the body part to be studied. A crossbar moves along a scale calibrated in both inches and centimeters (Figure 1-5). The dimensions are to be used with a technique chart to standardize technical factors for radiographic studies. Radiation-Protective Clothing During the radiographic study, the patient should be provided with radiationprotective clothing. Examples include the lead apron, diaper, and gonad shield. The LXMO, assistant, parent, or guardian must wear radiationprotective apparel if he or she must remain in the examination room during an exposure. Personnel and patient protection apparel is designed for safety and comfort. Lead aprons are available in numerous designs, sizes, shapes, and

thicknesses. They are tied to the body by straps and fastened by buckle or Velcro closure. Also, back support and full-wrap aprons exist for special studies. Protective aprons come in both male and female sizes, from small to extra large. It is important to choose the correct size to accommodate all LXMOs. At least two adult-size aprons should be available: one worn by the patient, and the second by the LXMO, if needed. A smaller apron should be available for children.

FIGURE 1-5. Thickness caliper. (Courtesy of Providence Imaging Products, Inc., Providence, RI.) The apron vinyl is impregnated with lead. An apron’s lead-equivalent thickness relates to its x-ray attenuation at a given kilovoltage (kVp) setting. For example, at 50 kVp, aprons with either a 0.25- or 0.50-mm leadequivalent thickness attenuate ≥97% of the x-rays. However, at 75 kVp, the

0.25-mm lead-equivalent thickness apron absorbs only 66% of the x-rays compared to 88% with the apron that offers a 0.50-mm lead-equivalent thickness. The minimum required thickness for protection is 0.25 mm; the normal thickness is 0.5 mm lead. Aprons of higher lead thickness are very heavy and uncomfortable if worn for an extended length of time. Keep in mind that an apron of 0.50-mm leadequivalent thickness weighs nearly twice that of an apron of 0.25-mm leadequivalent thickness. Geriatric patients may have extreme difficulty wearing the heavier apron, which presents a safety hazard for weight-bearing studies. A 0.25-mm lead apron should be available for these patients. To avoid damage to the protective lead lining, the apparel must be stored properly. Lead aprons will crack if care is not taken during handling and storage, especially if aprons are folded. Radiation easily penetrates through the crack to the wearer. Always place a lead apron on an apron hanger after every use to prevent damage to the radiation-protective material. A lead apron hanger functions like a coat hanger. It may be constructed of metal or wood and is designed to support a lead apron while hanging on a rack. The lead apron rack comes in two styles: either it can fully support a lead apron or it will have prongs to hold an apron hanger. It may be wall mounted or mobile. A wall-mounted rack may need extra support if several aprons are to be stored on it. Another apron storage device is a bar or rail; the apron is draped over it (Figure 1-6). However, when using this device, care must be taken to not wrinkle or crease the apron. Whatever hanger device is used, lay the apron smooth; proper storage of a lead apron will extend its useful life. Molded lead gloves are covered with vinyl or resilient leather for flexible use. They come in 0.25- and 0.5-mm lead-equivalent thicknesses. Soft glove liners are provided with lead gloves; they can be removed and washed as needed or desired. A lead glove hanger should be used to support and store the glove when not in use. The glove hanger is available as a separate unit that can be mounted to the wall, or it may be purchased as a combination lead apron/glove hanger unit (Figure 1-6). Gonad shields are useful for children who cannot wear a lead apron during

the exposure. An adult can also use them, especially for non–weight-bearing studies performed on a radiographic table, if an apron does not provide adequate protection. Gonad shields have 0.5 to 1 mm lead equivalency and are held in place by straps. Shield sheets, measuring 18 × 24 in, and triangular-shaped lead diapers with a lead thickness of 0.5 to 1 mm also can be placed over the patient’s reproductive organs. If considering a radiographic study for a patient that is pregnant, make a documented decision whether or not you decide to perform it. If a study is performed, minimize the dose by precise collimation of the x-ray beam, shielding the patient properly, and using a high-kVp technique. The once used 10-day rule is now obsolete. As a precaution, post caution signs and use a patient questionnaire. If a patient is exposed to ionizing radiation and you later find out that she is pregnant, contact the patient’s obstetrician/gynecologist. They will estimate the fetal dose and the stage of gestation when exposed. Most standard radiographic exams deliver between 1 and 5 rad (10–50 mGyt) doses. (Radiographic examination of the foot or ankle delivers an entrance skin dose that is less than 1 rad [1 mGyt].) Therapeutic abortion is not indicated if the fetal dose is less than 10 rad (100 mGyt). If the fetus is exposed to ionizing radiation during the first 2 weeks of pregnancy, which is the most critical time, either a spontaneous abortion will occur or there will be no effect. Radiation exposure during the second to eighth week of pregnancy, which is when major organogenesis occurs, could result in congenital abnormalities such as skeletal deformities or neurologic deficiency. If radiation exposure occurs during the second or third trimester, malignant disease, such as leukemia, could appear during childhood. The genetically significant dose (GSD) is the dose that, if received by every member of the population, would be expected to produce the same total genetic injury as the actual doses received by various individuals. Genetic effects may not manifest for several generations. Dose limits are placed on the whole population (not individuals) due to the fear of genetic effects. Dose Limits

Dose limits (DLs) are prescribed for various organs, the whole body, and various working conditions by the NCRP. DLs imply that, if received annually, the risk of death would be approximately 1 in 10,000. The formula for cumulative whole body occupational exposure is age (in years) × 1 rem. Current DLs are based on a linear nonthreshold dose–response relationship. They are considered the level of radiation exposure acceptable as an occupational hazard. The concept of MPD is now obsolete and has been replaced by DLs.

FIGURE 1-6. Lead apron rack. (Courtesy of Providence Imaging Products, Inc., Providence, RI.) DLs recommended by the NCRP for those who are occupationally exposed are as follows: the annual effective dose is 50 mSv (5 rem); the cumulative annual effective dose is 10 mSv (1 rem) × age; and the equivalent annual dose for skin, hands, and feet is 500 mSv (50 rem). The DL for the pregnant LXMO is 0.5 rem (5 mSv) for the 9 months of gestation. Exposure of the adult public also has DL recommendations by the NCRP. The annual DLs are 1/10th of that for radiation workers: the annual effective dose for infrequent exposure is 5 mSv (500 mrem); the annual effective dose for continuous or frequent exposure is 1 mSv (100 mrem); and the annual

effective dose for skin, hands, and feet is 50 mSv (5 rem). For the child under 18 years of age, the annual exposure effective DL is 1 mSv (100 mrem). Occupational Radiation Exposure Occupational exposure should be reduced however possible. The cardinal principles of radiation control are as follows. (1) Minimize the time of exposure. The less time one spends in a radiation area, the less exposure one will receive. The dose is directly proportional to the time in the radiation area. (2) Maximize the distance from the x-ray source. Radiation exposure falls off inversely as the distance squared. (3) Maximize shielding from the xrays. Any material interspersed between a source of radiation and an individual could protect that individual from radiation, such as a lead apron or mobile barrier. A mobile protective barrier can be used in settings where limited protection is available. The barrier moves on casters and is lead lined. A lead glass window allows the radiographer to view the patient during exposure (see Figure 1-7). The typical size of the mobile barrier is 7 ft high × 30 in wide. And, never stand in the path of the primary beam. All personnel involved in radiography should be monitored to estimate their amount of radiation exposure. Film badges (Figure 1-8) are the most affordable and must be obtained from a certified laboratory. They should be worn outside the apron at the level of the shirt or coat collar. A control badge is placed in an area where there is no ionizing radiation exposure. New film is received from the laboratory monthly. Each month the personnel-monitoring film and the control film are removed from the badges and sent back to the laboratory. The laboratory provides monthly reports, which should be reviewed and filed for future reference; any abnormal exposure reading should be investigated. Motion unsharpness is a detrimental factor affecting radiographic quality. Sandbags provide a quick and safe means of immobilization. A sandbag is placed across or against the extremity, outside the area of interest. The sandbag is heavy enough to restrict and restrain the patient but not to cause harm. More than one sandbag may be needed to achieve the desired positioning and immobilization. The LXMO should never hold a patient to prevent movement; if the patient cannot be restrained mechanically, a parent or guardian should hold the patient and must wear protective apparel.

Nor should the LXMO ever hold the image receptor. A cassette holder can be used to perform non–weight-bearing studies on a stretcher, examination bench, or operating table. It holds a cassette vertically for radiographic exposure and has clamps to accommodate any size cassette (Figure 1-9). The cassette holder is especially useful for performing radiographic studies of a patient confined to a wheelchair. Another application is for performance of weight-bearing ankle studies with a mobile x-ray unit.

FIGURE 1-7. Mobile protective barrier. (Courtesy of Gill Podiatry Supply Co., Middleburg Heights, OH.) The pregnant LXMO should review her previous radiation exposure history and review acceptable practices of radiation protection. She should wear a personnel-monitoring device at the collar and wear a 0.5-mm lead-equivalent apron, not 0.25 mm. An extra personnel-monitoring device can be positioned under the apron at waist level for the fetus. In the practice setting, any new employee should be trained in radiation protection and safety. In-service training should be made available for your staff, as well as counseling during pregnancy.

FIGURE 1-8. Film badge, showing clip on back.

FIGURE 1-9. Cassette holder. (Courtesy of Monee X-Ray Works, Monee, IL.)

Quality Control All newly acquired x-ray units should be tested before use. This examination is referred to as an acceptance test and should not be performed by a manufacturer representative but by someone else. In addition, an x-ray unit should routinely be evaluated periodically (ideally every year) to assess its performance. Parameters tested include the kVp calibration, mA linearity, exposure timer accuracy, collimation alignment, effective focal spot size, exposure reproducibility, and filtration (HVL). For example, the x-ray output intensity must be reproducible and not vary more than 5%; and the x-ray beam and the collimator’s light beam must coincide to within 2% of the SID. Most deficiencies cannot be detected without specialized monitoring devices. A radiation physicist can perform this service. Any deviation from the norm must be corrected. Records should be kept regarding the results of these tests and any corrections made. Protective apparel (lead aprons and gloves) should also be checked regularly for abnormal wear, and x-ray the apparel periodically to check for cracks or leaks. Records of such examinations should be kept on file. REFERENCES   1. Bushong SC. Radiologic Science for Technologists. 10th ed. St Louis, MO: Elsevier; 2013.   2. Carlton RR, Adler AM. Principles of Radiographic Imaging. 4th ed. Clifton Park, NY: Thomson Delmar Learning; 2006.   3. Christman RA. Radiation physics, biology, and safety. In: Weissman SD, ed. Radiology of the Foot. 2nd ed. Baltimore, MD: Williams & Wilkins; 1983.   4. Fosbinder R, Orth D. Essentials of Radiologic Science. Philadelphia, PA: Lippincott Williams & Wilkins; 2012.   5. The Fundamentals of Radiography. 12th ed. Rochester, NY: Eastman Kodak; 1980.

  6. Health Risks from Exposure to Low Levels of Ionizing Radiation: BEIR VII — Phase 2. Washington, DC: National Academies Press; 2006.   7. Martins B. Radiation physics, biology and safety. In: Christman RA, ed. Foot and Ankle Imaging. St Louis, MO: Elsevier; 2003.   8. National Council on Radiation Protection and Measurements (NCRP). Ionizing Radiation Exposure of the Population of the United States. Bethesda, MD: NCRP; 2009. NCRP Report No. 160.   9. National Council on Radiation Protection and Measurements (NCRP). Structural Shielding Design for Medical X-ray Imaging Facilities. Bethesda, MD: NCRP; 2004. NCRP Report No. 147.  10. National Council on Radiation Protection and Measurements (NCRP). Limitation of Exposure to Ionizing Radiation. Bethesda, MD: NCRP; 1993. NCRP Report No. 116.  11. National Council on Radiation Protection and Measurements (NCRP). Implementation of the Principle of As Low As Reasonably Achievable (ALARA) for Medical and Dental Personnel. Bethesda, MD: NCRP; 1990. NCRP Report No. 107.  12. National Council on Radiation Protection and Measurements (NCRP). Radiation Protection for Medical and Allied Health Personnel. Bethesda, MD: NCRP; 1989. NCRP Report No. 105.  13. National Council on Radiation Protection and Measurements (NCRP). Exposure of the U.S. Population from Occupational Radiation. Bethesda, MD: NCRP; 1989. NCRP Report No. 101.  14. National Council on Radiation Protection and Measurements (NCRP). Exposure of the U.S. Population from Diagnostic Medical Radiation. Bethesda, MD: NCRP; 1988. NCRP Report No. 100.  15. National Council on Radiation Protection and Measurements (NCRP). Quality Assurance for Diagnostic Imaging. Bethesda, MD: NCRP; 1988. NCRP Report No. 099.

 16. National Council on Radiation Protection and Measurements (NCRP). Radiation Exposure of the U.S. Population from Consumer Products and Miscellaneous Sources. Bethesda, MD: NCRP; 1987. NCRP Report No. 095.  17. National Council on Radiation Protection and Measurements (NCRP). Exposure of the Population in the United States and Canada from Natural Background Radiation. Bethesda, MD: NCRP; 1987. NCRP Report No. 094.  18. National Council on Radiation Protection and Measurements (NCRP). Ionizing Radiation Exposure of the Population of United States. Bethesda, MD: NCRP; 1987. NCRP Report No. 093.  19. National Council on Radiation Protection and Measurements (NCRP). SI Units in Radiation Protection and Measurements. Bethesda, MD: NCRP; 1984. NCRP Report No. 082.  20. National Council on Radiation Protection and Measurements (NCRP). Medical Radiation Exposure of Pregnant and Potentially Pregnant Women. Bethesda, MD: NCRP; 1977. NCRP Report No. 054.

2 Film–Screen Radiography ROBERT A. CHRISTMAN, MARY OEHLER, CASIMIR F. STRUGIELSKI, AND DAVID E. WILLIAMSON THE IMAGE RECEPTOR For over 100 years, radiography has used some form of film/intensifying screen combination as an image receptor. X-rays, after passing through the patient, interact with the image receptor, and a latent image (an invisible change that represents the object that was radiographed) forms in the film. The radiographic film is processed with chemicals to produce a visible (manifest) image, which is then viewed on a light box, catalogued, and physically stored. The typical image receptor used in film–screen radiography consists of a radiographic film that is sandwiched between two intensifying screens, all protected in a container known as a cassette. Intensifying Screen About 99% of the exposure on radiographic film comes from the light emitted by the intensifying screen. X-rays are extremely inefficient in producing optical density (blackening) of the radiographic film. Because xradiation is absorbed by tissue (and accumulated), any method to reduce dosage and yet provide a usable image is desirable. The intensifying screen is primarily responsible for making radiographic imaging a relatively safe medical diagnostic tool. Intensifying screens are composed of light-emitting phosphors that fluoresce when exposed to x-rays. Because radiographic film is sensitive to light, fewer x-rays are needed to expose the film. This allows the exposure technique to be reduced, which reduces patient exposure to x-radiation. Intensifying screen selection traits include its speed and, more importantly, spectral qualities.

Two types of intensifying screens are available: calcium tungstate and rare earth. Calcium tungstate screens emit light in the blue/blue–violet portion of the visible light spectrum; rare-earth screens primarily emit green light, although a few emit blue light. Green-sensitive (orthochromatic) film must be used with green-emitting intensifying screens and blue-sensitive film with blue-emitting intensifying screens. Incorrect matching of the film and screen will result in a slower film/screen system speed, which means using a longer exposure technique. Rare-earth intensifying screens are faster than calcium tungstate film/screen systems and do not sacrifice resolution. They have, therefore, widely replaced calcium tungstate systems. If they are cared for properly and not accidentally damaged, intensifying screens can last many years. They should be cleaned as recommended by the manufacturer. Screen cleaners often contain an antistatic solution. Do not spray the formula directly onto the screen; spray onto a cloth first. A drop of chemical could stain or damage the screen before it is wiped off. Exercise extreme care not to touch the screen except when cleaning. Scratches, for example, will permanently damage the screen. Radiographic Film In medical radiography, many products have been designed to offer a vast array of speed levels and contrasts appropriate for the body part being radiographed. Generally speaking, the film’s inherent speed depends on the silver halide crystal (grain) size, structure, and sensitivity; the choice of developer; and the time and temperature that the film is processed. It naturally follows that the larger the silver halide crystal, the faster the speed of the film and the less the exposure. Combinations of different-sized crystals are used to create the desired characteristic curve, a measurement that describes the relationship between radiographic exposure and optical density. The layers of an x-ray film (Figure 2-1) and their functions are as follows: •  Overcoat: A protective layer that aids in scratch protection both before and after processing •  Emulsion: The light-sensitive layer, made up of silver halide suspension in a high-grade gelatin support

•  Subcoat: A very thin coating that ensures adherence of the emulsion to the base support •  Base support: The material that forms the structural support onto which the other sensitive and protective layers are coated •  Backing: Another gelatin layer sometimes used to prevent curl, inhibit light piping, and improve image quality or to act as a filter to make the film insensitive to certain wavelengths The two primary types of radiographic film are direct exposure film and screen film. Direct exposure film provides a sharp image of thin body parts having high subject contrast. However, long exposure time is necessary to create an image with appropriate optical density. Direct exposure film was primarily employed to assess the extremities but has been widely replaced with faster high-detail film/screen combinations.

FIGURE 2-1. Cross section of x-ray film. Many types of screen film are available from several manufacturers. This type of film is sensitive to light emitted from an intensifying screen as well as x-rays. Screen film is either single or double emulsion. Single-emulsion film is only coated on one side. It provides a sharper image when used with a single-emulsion, high-detail/fine-grain intensifying screen than does doubleemulsion film used with two screens. Single-emulsion film is not used very often nowadays because of the availability of newer high-detail double-

emulsion film/screen combinations that give optimum image sharpness. Other film characteristics to consider include contrast, speed, and light absorption. After selecting a film type, it must be matched to an appropriate intensifying screen. The choices for film contrast are high, medium, and low. High-contrast film produces more blacks and whites on the radiograph. Medium-contrast film has more shades of gray than does high-contrast film. Low-contrast film demonstrates a long scale of grays, referred to as latitude. Latitude is inversely proportional to contrast. Medium-to-high–contrast film is typically used for extremity radiography. The term film speed refers to the time it takes to respond to an exposure. Faster film speed correlates to less radiation exposure for the patient. However, faster films are not as sharp as slower films. Slower-speed, highdetail (high-resolution) films are primarily employed for extremity studies. Each type of film has specific light absorption characteristics. There are bluesensitive and green-sensitive (also known as orthochromatic) films. A bluesensitive film must be used with an intensifying screen that emits blue light and a green-sensitive film with a screen that emits green light. Mismatched film/screen combinations require increased exposures, thereby increasing the patient dose. Specific guidelines should be followed regarding storage and handling of radiographic film. A relative humidity of 40% to 60% and a constant 70°F temperature are ideal for storage. In a hot, humid environment, the film may stick together. Check the expiration date of each box of film; rotate the film accordingly, using older film first. Expired film should not be used; film quality may be impaired. Film should be stored on its side or end, not flat. Pressure marks (artifacts) may develop if the film is stored flat. Handle films with clean, dry hands. Be careful not to bend or crease the film during handling. The radiographic image is greatly influenced by the type of film and intensifying screen used. Because there are many different types of films and screens, they must be properly matched to one another and to the study (in this case, extremity radiography). Inappropriately matched film/screen combinations can significantly impair the radiographic quality. The wrong

film/screen combination can ultimately lead to misdiagnosis and increase radiation exposure to the patient if a study is repeated.

FIGURE 2-2. Rigid cassette (opened, revealing intensifying screens and rectangular cutouts for identification imprinting). (Courtesy of Providence Imaging Products, Inc., Providence, RI.) Film–Screen Cassettes

Cassettes are lighttight film and screen holders. They swing open on a hinge mechanism. The screens are permanently affixed inside the cassette, one on each side (if two screens are used). The film is “sandwiched” between the two screens; it must be loaded in the darkroom (unless one is using a daylight loading system). Most cassettes are rigid devices (Figure 2-2); major film manufacturers may provide cassettes at little or no cost if you agree to use their film. All cassettes have a backing material to absorb radiation that passes through the cassette. This prevents backscatter that could otherwise impair the image quality. IMAGE QUALITY The goal of producing a radiographic image is to provide the greatest amount of usable information to aid in diagnosis. In medical radiography, there is no tolerance for anything other than the closest to reality that one can achieve. Consistency and faithful recording of anatomical structures with optimum detail are primary goals. Anything less is unacceptable for the interests of both the doctor and the patient. Quality can be described as a level of excellence. There are many adjectives used to clarify it. In medical imaging, one must think of matching the interpreter’s trained eye with the optimum image capturing characteristics of the film. When the image can be visualized to allow the most concise and informative interpretation, one can say that the highest level of radiographic quality has been attained. Many factors contribute to or detract from the endproduct quality. This chapter discusses those factors that the radiographer/limited x-ray machine operator (LXMO) can control and measure, and offers general concepts for assessing radiographic quality. When a film is labeled “high quality,” it usually has low fog, appropriate contrast to display small differences in density, and sufficient speed to overcome blurring caused by motion. This is all embodied in an emulsion that has grain structure fine enough not to impair the definition of the part being radiographed. The radiologic examination is an important adjunct in determining the

etiology (or etiologies) of a patient’s complaint. It is equally important that the finished radiograph be of the highest quality, providing as much information as technically possible. Definitions Radiographic quality refers to the exactness of representation of the anatomic structure on the radiograph within the useful density range. Quality is not easy to objectify and is influenced by radiographic noise and resolution. Radiographic noise is the undesirable fluctuation in the optical density of the image. Radiographic noise has three components: film graininess (size/spacing of silver halide crystals—grains—in the film emulsion; the smaller the better), structure mottle (size/spacing of phosphors in a screen; again, the smaller the better), and quantum mottle (random interaction of xrays with the image receptor; the greater the number of primary beam x-rays that strike the image receptor, the smoother the quantum mottle). Noise is primarily due to scatter radiation. Resolution is the ability to separate and distinguish between two separate objects; this is also referred to as detail. There are two types of resolution: spatial and contrast. Spatial resolution is of great importance when imaging the foot and ankle; it is the ability to distinguish between two objects demonstrating high subject contrast, such as between bone and soft tissue. Contrast resolution best applies to imaging of the abdomen, where organs of similar contrast need to be distinguished. The ability of an x-ray film to respond to an x-ray exposure is the measure of its sensitivity or speed. Fast-image receptors (≥400) have high noise and low resolution; slow-image receptors (50–300) have low noise and high resolution. Therefore, to obtain a high-quality foot or ankle radiograph, meaning one with low noise, high contrast, and high detail, one would use a slow-speed film/screen combination. Factors that affect radiographic quality are related to the x-ray film (density, contrast/latitude, speed, processing time, and temperature), geometrical issues (distortion, magnification), and the subject (motion, and contrast relating to

thickness, density, and atomic number). Just as no two patients are alike, situations and technical factors that produce the image will vary. The fundamental concept is that the finished radiograph should exhibit sufficient radiographic density, acceptable contrast, optimal detail, and minimal distortion. Radiographic Density The radiographic density, also referred to as optical density (OD), is defined as the amount of blackening in a radiograph. Density and adequate differences in density, or contrast, are considered the most important properties of a radiograph. Proper densities and adequate contrast are required to give visibility to the structural detail of a subject. Milliampere-second (mAs) is the primary controlling factor for radiographic density. It is a product of milliampere multiplied by the exposure time expressed in seconds (s). mAs = mA × s 15 mA × 8/60 s = 120/60 = 2 mAs 30 mA × 4/60 s = 120/60 = 2 mAs Both sets of exposure factors (mA and s) yield the same mAs value; therefore, both films should exhibit the same radiographic density, assuming that all other technical factors are unchanged. Also, it is important to remember that as mAs increases, both x-ray quantity and radiographic density increase proportionally, and vice versa. This is known as the law of reciprocity (Figure 2-3). To cause a visible shift in radiographic density, a ±30%-mAs alteration is necessary to produce this effect. This is the minimum change required that will be perceived by the human eye. If adjustment in density is necessary, however, changes in mAs should be made by factors of 2 (either double or half). A film that displays insufficient density and that was originally exposed using 2 mAs, for example, should be repeated at 4 mAs.

A film of questionable radiographic density should be outside the acceptable limits to justify a repeat exposure, either very light or very dark (i.e., the anatomic structures under consideration cannot be adequately distinguished). The appropriate change would be to either double or half the mAs. However, when doing so, change either the mA or the exposure time, not both; changing both will increase the mAs value above the half or double value required.

FIGURE 2-3. As time or mA changes (mAs), a corresponding change in

density (law of reciprocity) occurs. Left image: 4 mAs exposure versus 2 mAs exposure (right image); all other exposure factors (kVp and SID) are unchanged. Other factors influencing radiographic density are kilovoltage, distance, film/screen combination, and film processing. Kilovoltage Kilovoltage alters both the quality (energy) of the x-ray beam and the quantity of x-ray photons produced. A 15% increase in kVp will cause the same change in radiographic density as doubling the mAs, and a 15% decrease in kVp will result in a change in density similar to decreasing the mAs by half. This is referred to as the 15% rule. A 4% to 8% change in kVp is necessary in order to produce a visible difference in the radiograph. This will require a change in at least 3 to 7 kVp (the kVp range used with lower extremity–specific x-ray units is 50–70 kVp). Distance Changes in distance between the x-ray tube and the film, also referred to as the source-to-image distance (SID), will alter the number of photons striking the film (quantity of radiation). This can significantly alter the radiographic density. If the tube is too far away from the film, an insufficient number of x-ray photons strike the film, producing a light film (decreased radiographic density). And, if the tube is too close to the film, a greater number of x-ray photons, than are required, strike the film, producing a very dark film (increased radiographic density). This is known as the inverse square law and is expressed by the following formula:

where I1 = original intensity, I2 = new intensity, d1 = original distance, and d2 = new distance. For example: The original exposure intensity is 2 mAs at an SID of 28 in.

What would the radiation intensity be at the film if the SID were increased to 36 in?

Therefore, by increasing the SID from 28 to 36 in, the x-ray intensity at the film has been reduced to 1.2 mAs, nearly half of the original 2 mAs, which results in a film with decreased radiographic density. In order to determine the mAs needed at the 36-in SID to produce the same radiographic density as the original film (2 mAs and 28-in SID), use the following formula:

This type of scenario will arise when performing radiography with variable SID, such as with a portable/mobile x-ray unit. Film/Screen Combinations

Years ago, the speed of film/screen systems was divided into three groups: Slow (detail) Average (medium, par) Fast

Double the average screen speed Baseline Half the average screen speed

The speed of the film/screen system can also be expressed in numeric terms, developed by the manufacturers, referred to as the relative speed (RS) value (Figure 2-4). This refers to the sensitivity or response that the intensifying screen phosphors have to x-ray photons. The higher the RS, the faster the film/screen system speed; therefore, less radiation is needed to produce an image. However, there is a compromise: higher film/screen system speed reduces image sharpness (detail). Since older calcium tungstate screen phosphors have mostly been replaced by rare-earth phosphors in imaging technology, the compromise between detail and speed has diminished. A balance between speed (sensitivity to x-ray photons) and resolution (detail) has been accomplished. Should it become necessary to use or change factors because you are employing different speed systems, the following formula can be used:

where mAs1 = old mAs, mAs2 = new mAs, RS1 = old relative speed, and RS2 = new relative speed. For example: If there was a change from a slower 200 RS system, using 2 mAs, to a faster 400 RS system, what mAs would be necessary to produce a film with the same radiographic density as the original system (assuming all other technical factors remain unchanged)?

Therefore, doubling the speed of the film/screen system resulted in a reduction of mAs by half.

FIGURE 2-4. RS values of film/screen systems.

FIGURE 2-5. Single-screen system. Position of notches (arrow and arrowheads) aids in proper placement of emulsion against intensifying screen. Film/screen systems generally have two intensifying screens, one on either side of the film, which has emulsion on both sides. A single screen may be employed to image the extremities. This system improves the sharpness of

detail. However, when using a single-screen imaging cassette, the correct film type (single-sided emulsion) must be used. Care must be taken when loading the film into the cassette. The film emulsion, generally the dull side, must face the single screen. This can be accomplished by making sure that the notch of the film is placed in the correct position in the cassette to have the film emulsion side face the screen surface (Figure 2-5). If switching from a dual-screen system to a single-screen system, the mAs would need to be doubled to obtain the same radiographic density as with the double-screen system. Compensation Filter The foot thickness is not the same in the forefoot as in the rearfoot. As a result, in the anteroposterior/dorsoplantar foot view, the radiographic density varies considerably between the toes and the tarsus. If an exposure technique is selected to obtain an ideal film density for the toes, the tarsus will be underexposed (decreased radiographic density). In contrast, a technique chosen to image the tarsus will overexpose the toes, which may appear too dark (increased radiographic density). The unequal radiographic density between the forefoot and the rearfoot can be balanced by using device known as a compensation filter. This type of filtration can improve the visibility of bone and soft tissue and lower patient exposure. The compensation filter system (Figure 2-6) is easily attached to most conventional collimators. The plate is wedge shaped and made of a radiolucent lead–plastic material; it does not alter image quality or contrast. Patient, film, and tube head positioning as well as light beam collimation are performed before attaching the filter. Attention must be paid to initial installment of the filter holder and placement of the filter; the thick portion of the wedge must be positioned at the toe end of the light field so that the thinner portion of the wedge approaches the midfoot. To be certain the filter is positioned over the forefoot properly, and light beam collimation should also be performed after attaching the filter. Use the same technical factors that would be used to image the tarsus only. The LXMO must remember to remove the filter before performing other positioning techniques.

FIGURE 2-6. Compensation filter system. (Courtesy of Nuclear Associates, Carle Place, NY.) After using the compensation filter, the toes are more visible in the resultant film and the radiographic density is similar to the tarsus. Also, if the area of concern is limited to the digits, a collimated view of the toes alone is preferable; the adjustment of technical factors obviates the need for the compensation filter. Effects of Film Processing When proper temperature, replenishment rate, and sequence of film processing are maintained, processing should not be an influencing factor concerning radiographic density. However, the temperature of the developer solution is a crucial factor. Manufacturers have compiled recommended temperature and time charts for each specific type of film. If the film is immersed in the developer shorter than the required length of time, it will exhibit decreased radiographic density; increased radiographic density results if the film is immersed longer than the required time. If, for some reason, the developer temperature increases, the time that the film is immersed in the developer solution should decrease; if not, increased radiographic density will result. Likewise, if solution temperature decreases, the film immersion time should increase. (Generally speaking, one should not deviate from the film manufacturer’s recommendations.) Adequate replenishment rates must be maintained to have proper activity level of the processing chemicals. Improper replenishment, either too high or too low, will affect the radiographic density. To ensure proper activity of the processing solutions, new chemicals should be prepared every 1 to 2 months. Automatic processors should be monitored regularly for correct temperature levels, replenishment rates, proper roller transport function, proper water circulation, and sufficient dryer blower function. Even if not required by state regulations, sensitometry and densitometry, also known as sensidensitometry, should be performed daily on your processor in the morning prior to performing any studies. Sensidensitometry confirms whether or not the

processor is working properly, that is, that the chemicals are in the appropriate concentration and temperature to achieve a radiographic image of optimal quality. Sensitometry and Densitometry Sensitometry is the study of a film’s response to exposure and processing. The graphic basis of sensitometry—the H&D (Hurter and Driffield) curve— is an important tool for measuring the relationships between radiographic exposure and density. Radiographic density is the result of a deposit of metallic silver remaining on a film after exposure and processing. Densitometry measures the amount of blackening after shining a light through the film. The more the blackening, the more light is absorbed and the less is transmitted. Densities are read by a specialized light meter called a densitometer, and the results are measured in logarithms. Applied to sensitometry, density is defined as the common logarithm of the ratio of the amount of light striking one side of the film, compared to the amount of light coming out of the other side. When 10 units of light are shined at a film and only 1 unit comes out the other side, the film is said to have a density of 1; there is a 10:1 ratio. When the ratio is 100:1, the density is 2, and when the ratio is 1000:1, the density is 3, and so on. To produce an H&D (characteristic) curve, the film is first exposed to a calibrated step wedge that has a series of equal-increment step changes. The film is processed, and each step of density is read and plotted on special graph paper. Figure 2-7 is a characteristic curve of an x-ray film exposed with intensifying screens and processed in x-ray chemistry. The portion of the curve designated as the “toe” demonstrates the response of the emulsion to relatively small amounts of radiant energy. With an increase in exposure, the density builds slowly until the “straight-line” portion of the film is achieved. Along this straight-line portion the density increases uniformly with the logarithm of the exposure until the shoulder of the curve is reached. In the shoulder region, additional exposure results in smaller increases in density to a point where additional exposure does not produce any greater density. In fact, if sufficient exposure is given, the density will actually decrease. Sensitometrically, the term contrast refers to the slope or steepness of the characteristic curve of the film; contrast is a generic term. There are different

contrasts depending on where they are “read” on the characteristic curve, as follows.

FIGURE 2-7. Characteristic curve of x-ray film. Gradient is the contrast of the film at a given density. When a straight line is drawn tangent to the characteristic curve at a particular density, this line

forms a slope that is the gradient (or contrast) at this density (Figure 2-8). Average gradient is found most often in radiography because it is more useful to have a single number to indicate the effective contrast of a film. This number, known as the average gradient, is determined by drawing a straight line between two selected densities on the characteristic curve. The two densities used for this measurement are those most often at the thresholds of where the information in the radiograph falls—from 0.25 to 2.00 (Figure 29). Another term used to describe density is gamma, the slope of the straight-line portion of the characteristic curve. This term is seldom used in radiography, because the curves of radiographic films have relatively short straight-line portions. Also, this portion of the curve may not coincide with the density range that is most useful in radiology. If gamma is a reference, it is best to determine exactly which of the “contrasts” is being referred to. Exposure is defined as intensity multiplied by time. This can be expressed in either absolute exposure units (ergs per square centimeter of x-radiation) or relative units. Relative exposure is much more convenient and equally useful to us. In radiography, we refer to exposure in terms of milliampere-second (mAs). In general, when the mAs is doubled, the exposure is doubled, kVp remaining constant. In plotting a characteristic curve, density is plotted against the log of relative exposure. If the kilovoltage remains constant, the ratios of the exposures reaching the film through two different regions of the subject are always the same, regardless of the values of the milliamperage, time, or SID. On a logarithmic scale, the same distance on the exposure scale, regardless of the absolute values, will always separate any two exposures whose ratio is constant. For example, two exposures, one of which is twice the other, will always be separated by 0.3 on the logarithmic scale (the logarithm of 2 is 0.3).

FIGURE 2-8. Calculation of gradient. It has been determined that the contrast of a film is indicated by the shape of the characteristic curve. Speed is indicated by the location of the curve along

the exposure axis. The faster a film is, the more it will lie toward the left of the graph. The faster it is, the less relative exposure it will take to reach the speed point density. The separation of two films plotted on the same graph may be measured at the speed point to determine the speed difference (Figure 2-10). The convenience of using relative exposures also applies to speed. The speed of one film can be expressed on a relative basis to another when one is made the standard of comparison. The reference film can be assigned any arbitrary speed value, such as 100. If another film only requires half as much exposure to reach the same density as the reference film, then the faster film will have a RS of 200. A density of 1 above base plus fog has been designated as the density for computing film speeds. This density was chosen because it represents the average of the useful density range (0.25–2.00). However, a density of 0.25 above base plus fog appears to be the approximate density most often viewed in the diagnostic region of a typical radiograph. For practical purposes, RSs are often calculated at this lower density.

FIGURE 2-9. Determination of average gradient.

FIGURE 2-10. Differences of film contrast and speed result in changes of curve shape and position, respectively. Contrast

Contrast is primarily responsible for allowing the visibility of detail. There are two types of contrast: radiographic and subject. Radiographic contrast is the variation or difference in densities that allows one to discern between two adjacent densities within the image. It is influenced by the type of intensifying screen, the film’s density and characteristic curve, and processing. Subject contrast is the result of attenuation differences as the xrays pass through a body part; it is affected by tissue thickness, type, atomic number, and density.

FIGURE 2-11. Example of contrast scale. Aluminum step wedge: The left image was produced using 50 kVp. It demonstrates higher contrast (short gray scale, smaller number of steps). The right image was produced with 70

kVp and demonstrates lower contrast (long gray scale, larger number of steps). The two scales of contrast, long and short, are differentiated in Box 2-1 and Figure 2-11. A short scale of contrast is most appropriate for radiography of the foot and ankle, which primarily concerns bone anatomy (trabeculations, cortex) and distinguishing the bone margins from adjacent soft tissue. Abdominal radiography would require a longer scale of contrast to visualize the different soft tissue structures that have more similar densities. BOX 2-1 Characteristics of Short-Scale and Long-Scale Contrast Short-Scale Contrast High contrast Limited number of gray shades Increased contrast Lower level of kilovoltage Limited areas of visibility

Long-Scale Contrast Low contrast Increased number of gray shades Decreased contrast Higher level of kilovoltage Thicker areas more visible

Kilovoltage is the primary controlling factor in determining the type or scale of contrast produced. Generally speaking, higher kVp levels will produce low contrast and low kVp will produce high contrast. To produce a visible or noticeable contrast change, the kVp should be changed by 15%, usually about ±7 kVp; remember to adjust the mAs accordingly to maintain the same radiographic density (Box 2-2). Contrast will not be affected by changes in mAs or SID alone unless they result in a radiographic density that falls outside the diagnostic range. In cases involving poor radiographic density, a contrast mask (Figure 2-12) may be useful in eliminating adjacent densities for better evaluation. Otherwise, the study should be repeated.

FIGURE 2-12. Contrast mask, made from black construction paper with 4.5cm-diameter viewing area, placed on radiograph at base of first metatarsal– cuneiform joint left foot. BOX 2-2 15% Rule To lengthen the scale of contrast, ↓ mAs by ½ and ↑ kVp 15% To shorten the scale of contrast, ↑ mAs by 2 and ↓ kVp 15% Spatial Resolution Spatial resolution (detail, sharpness, definition) is the recorded accuracy of the structures imaged. The image structures may appear either sharp (crisp) or not sharp (fuzzy) depending on the degree of detail recorded. This property also involves how well two closely placed objects are perceived or seen as two distinct objects. This differentiation becomes crucial when dealing with minute structures. Recorded detail may be influenced by geometry, material, film/screen contact, and motion. Geometry The focal spot is the area on the anode where electrons strike the target. A larger focal-spot size will result in decreased detail; a smaller one will increase detail. The focal spot size is fixed on lower extremity specific units and cannot be changed. The object-to-image distance (OID) is the distance between the body part being examined and the image receptor (film). The greater the distance the object is away from the film, the less detail or sharpness will be recorded (blurred image). Increasing the OID results in magnification. The SID is the distance between the x-ray tube and the image receptor. Considered separately, a longer SID produces less geometric blurring than a shorter SID and, therefore, a sharper image. There is a trade-off, however; a longer SID requires increased mAs to obtain an image of similar radiographic density (recall the inverse square law).

Material (Films/Screens) Slower-speed film/screen systems produce a sharper recorded image. Sharpness is related to the intensifying screen’s crystal size and phosphor layer thickness (Box 2-3). BOX 2-3 Intensifying Screens and Their Attributes Slow (Fine Detail) Screen Small crystal size Thin phosphor layer Greater detail

Fast Screen Large crystal size Thick phosphor layer Less detail

Film/Screen Contact Uniform film/screen contact inside the cassette must be maintained over the entire film surface. If not, a localized area of blurring may occur. The area of blurring will appear in the same area of the image when the same cassette is used for other studies. The LXMO can perform a test if inadequate film/screen contact is suspected. Place a fine wire mesh screen (the type employed in windows or doors) on the cassette in question. Make sure the screen is in direct, flush contact with the entire cassette front. Expose the cassette using standard technical factors for an average-sized foot with the dorsoplantar projection. Develop the film normally. Check the processed image for irregularities in sharpness and density. The screen should be replaced if localized areas of blurring are seen consistent with those seen in other images. To help in identifying this and other problems (artifacts), each screen can be numbered to better identify the problem cassette (Figure 2-13). Motion Any motion that occurs during the x-ray exposure, either the patient or the xray tube, can result in image blurring. Of the two types, patient motion is more common. It may be a voluntary motion, which is under direct control of the patient, or an involuntary motion that is not under direct control of the patient (usually related to a neurologic disorder). Pediatric patients may have

difficulty staying still during an exposure. Reduction of exposure time, immobilization of the body part being examined, or, in some cases, sedation may be indicated to limit patient motion blur.

FIGURE 2-13. Cassette with ID number. Rub-on number 6 placed on intensifying screen to better identify problem cassette. The number will show up on exposed x-ray film. In order to reduce exposure time to decrease the effect of potential patient motion, decrease the exposure time by half and increase kilovoltage by 15% (the 15% rule) to maintain the same radiographic density. For example: The highest mA setting (15 mA), 60 kVp, and an exposure time of 16/60 second are currently in use. What would the new kVp setting be if you decreased the time by half to 8/60 second and wanted to maintain the same radiographic density?

New kVp = (60 kVp × 15%) + 60 kVp = 69 kVp The two films should exhibit equal radiographic densities. The 15% rule can be applied to pediatric or geriatric patients, or to individuals whom you determine may not hold still for the duration of the x-ray exposure. Vibration or drift of the x-ray tube housing may occur if the locking devices are not tightened properly or if the arm attached to the tube head needs adjustment. Check the equipment manual for instructions or call the manufacturer if this is suspected. Periodic equipment inspections should be performed to check that the locking mechanisms are functioning properly. Distortion Distortion is the misrepresentation of size, shape, or positional relationships of recorded structures. Size distortion, known as magnification, occurs as the OID increases. The amount of magnification, or the magnification factor (mF), can be determined by dividing the SID by the source-to-object distance (SOD). The SOD is determined by subtracting the OID from the SID. For example: If the existing SID is 28 in and the OID is increased from 0 to 6 in, what would be the magnification factor? SOD = SID × OID = 28 × 6 = 22 in

In this example, the magnification will be 27%, that is, the image will be 127% of the object size. This applies to each structure such that the magnification is uniform throughout the OID plane. Magnification should be used sparingly since it does result in blurring (less sharpness) of the image. Blurring from magnification can be reduced if the SID is increased to 40 in; however, most lower extremity units are fixed at an SID of 24 or 28 in. Even when the object (or foot) is placed directly against the image receptor (OID = 0 in), some bones may be one or more centimeters

away from the film. Therefore, some magnification of internal structures is inherent and unavoidable. Shape distortion is the unequal magnification of the anatomical structures being imaged, displacing structures from their actual position. Elongation projects the object so that the image appears longer; this occurs when the tube head is angled relative to the object or image receptor (Figure 2-14). Foreshortening projects the object so that the image appears shorter than normal; this occurs when the part is improperly aligned with the tube head and image receptor. It is important to have the area of interest, the central ray (CR) of the x-ray beam, and film in proper alignment to avoid shape distortion, or positional changes. Ideally the tube should be perpendicular to the image receptor and the body part being examined. Careful attention to part placement, film placement, and tube head angulation (degree and direction) will help minimize distortion. Be aware that various degrees of distortion may be already present in the radiograph with some positioning techniques. Distortion is occasionally acceptable when attempting to enhance visualization of difficult areas (consider the Harris–Beath and Broden views). Shape distortion can occur by three mechanisms. The first is by manipulating the tube head angle so that the central beam is not perpendicular to the film and/or body part. Greater distortions occur with greater tube angles. Compare non–weight-bearing oblique positions of the foot (0° tube head angle) with the weight-bearing counterparts (45°) (Figure 2-15). The second way distortion is created is by part rotation or angulation, so that the plane of the body part is not parallel to the film. Finally, shape distortion occurs if the xray beam is not centered over the middle of the body part to be examined. Technique Guides Another valuable tool toward achieving optimal radiographs is to develop and employ a technique chart. This chart is required by many state licensing organizations. It must be posted near each piece of x-ray–generating unit that the user operates. The guides must include the following: the patient’s size

versus technique factors, type and size of film/screen, and SID. A technique chart may also include the x-ray room number, type of x-ray unit, and a list of positioning techniques that includes information such as the exposure factors, tube head position, and film placement. The first consideration is the technique or exposure system. There are various types of exposure systems; the two applied in podiatric radiology are fixed kilovoltage and variable kilovoltage systems. Fixed Kilovoltage With a fixed kilovoltage system, the kVp is held constant while the mAs is varied per the part thickness. An optimal kVp is used that produces appropriate contrast, typically 60 kVp for the foot and ankle. Generally speaking, the kVp setting for a fixed kVp technique is higher than that used with variable-kVp techniques.

FIGURE 2-14. A: Dorsoplantar foot view at 0° tube angle. Metallic wires represent tibial (U-shaped) and fibular (circular-shaped) sesamoids (arrowheads). B: Tube angle 15°. Sesamoid markers appear to move distally. The first metatarsal–cuneiform, second metatarsal–cuneiform, and medial and intermediate cuneiform–navicular joints are now visible (arrows). C: Tube angle 30°. Note the distal sesamoid marker position (arrow) and

elongation of the metatarsals. D: Tube angle 45°. Again, the sesamoids are seen more distal (long arrows). Also note the loss of joint space at all metatarsophalangeal joints and in the midfoot (short arrows).

FIGURE 2-15. Distortion caused by tube head position. There is unequal

magnification of all bones in the weight-bearing medial oblique view (right image) compared to the non–weight-bearing oblique view (left image). Advantages Decreases patient dose Uniform radiographic contrast Lower x-ray tube heating Lengthens exposure latitude Extends tube life Decreases exposure time

Disadvantages Lower contrast levels Produces more scattered radiation Small incremental changes not possible      

Variable Kilovoltage With a variable kilovoltage system, the kVp varies, depending on the part thickness; only one mAs value is used for that specific part, regardless of its size. Advantages Small incremental changes possible Higher contrast levels

Disadvantages Varying contrast levels Varying radiation dose to patient

A typical variable-kVp chart is set up by measuring the thickness of the area in question and selecting the appropriate kilovoltage for the varying part thicknesses (Boxes 2-4 and 2-5). BOX 2-4 Average Adult Part Thicknesses Foot Dorsoplantar: 6–8 cm Lateral–medial: 7–9 cm

Ankle Anteroposterior: 8–10 cm Lateral–medial: 6–9 cm

BOX 2-5 Suggested kVp Range Small extremities Medium extremities Large extremities

50–60 kVp 55–65 kVp 60–70 kVp

To determine the kilovoltage for a particular area, use one of the following: (2 kVp × cm part thickness) + 40 kVp = New kVp (2 kVp × cm part thickness) + 50 kVp = New kVp For example: If the dorsoplantar foot thickness (dorsal navicular area to plantar aspect) were 8 cm, what kVp setting would you select? (2 kVp × 8 cm) + 40 kVp = 16 kVp + 40 kVp = 56 kVp No amount of mAs will compensate for insufficient kilovoltage. It is very important, regardless of the type of technique chart employed, that the part is measured properly when using a caliper device (Figure 2-16). Used properly, the longer side of the caliper should correspond to the central ray. Place the device so that the longer side mimics the central ray path through the desired anatomical area, at the thickest part for that particular projection. (For the dorsoplantar foot technique, measure at the base of second metatarsal; for toe techniques, measure at the metatarsophalangeal joints proximally.) Be certain that the two shorter arms are placed so that they remain parallel to each other when obtaining the proper centimeter measurement. Record these calculations in the patient’s medical record for future reference. Other Exposure Technique Considerations Children The pediatric patient may pose a problem for the operator. In the pediatric patient, the tissues have a greater percentage of water, which causes more scattering of the remnant radiation. Consider using a lower kVp technique, which can decrease the amount of scattering. Many facilities group children into age groups (three or four) to the age of 12 while varying the mAs (Boxes 2-6 and 2-7). Also consider that pediatric patients at a similar age may vary in size because of their development. Children younger than 5 years may need a parent or guardian present in the room to assist in the examination process. This person must be provided with a lead apron and must wear it during the x-ray examination. Radiographic

studies may be performed by “simulated weight bearing” for those infants unable to stand (weight bearing). The parent holds the child under the arms to simulate weight bearing. The younger infant may have to be examined while supine. The key is using the quickest exposure time possible to eliminate motion artifacts. Proper instructions must be given to parents/guardians as to how to hold the child.

FIGURE 2-16. Caliper device. This device is used to measure anatomic regions for proper factor setting in conjunction with technique guide (centimeters and inches are listed along the vertical arm). In this example the flexible arm (arrowhead) is positioned over the talonavicular joint. The distance between this location and the bottom of the foot (arrow) is read along the side of the longer, vertical arm. BOX 2-6 Pediatric kVp Technique Birth–2.5 y 2.6–6 y 7–12 y

0.3× adult mAs 0.5× adult mAs 0.75× adult mAs

BOX 2-7 Pediatric mAs Technique Birth–1 y 1–3 y 3–7 y 7–12 y

0.25× adult mAs 0.5× adult mAs 0.70× adult mAs 0.9× adult mAs

The LXMO should also explain the procedure to the assistant (parent or guardian) and to those children who may understand. Be patient and understanding, especially with infants and children with disabilities. If using restraining devices, ensure that they will not interfere with the projection. Geriatrics The geriatric patient may require alterations to the technical factors. For example, kVp can be decreased 6% to 8% for patients with known or suspected osteoporosis. For patients with poor muscle tone, the mAs can be decreased 25% to 30%. The LXMO must also consider other geriatric conditions, especially if they have a decline in auditory senses and vision, poor balance and/or flexibility, or inability to sit still for the study. Use the best combination of mA and time for the shortest exposure time possible, thereby minimizing exposure. Use clear, concise instructions for positioning. To decrease the chance of injury, assist the patient.

Some instances require immobilizing or holding the foot or ankle for required projections/positions. The person who is holding the patient should wear a lead apron and gloves. Before completing the radiographic study, check the first film of multiple-projection studies for quality. Pathologic Considerations Different pathologic conditions may warrant changes in technique. These changes are either increases (“additive” pathology) or decreases (“destructive” pathology) in mAs. The additive pathologies are those that increase either the volume or the density of the soft tissues or osseous structures. This warrants an increase in the amount of x-radiation to properly penetrate the area and expose the film. The destructive pathologies are those that decrease either the volume or the density of the soft tissues or osseous structures. Because of this effect, the amount of radiation must be decreased to adequately expose the film. Box 2-8 lists pathologies that may be encountered in the foot and ankle. BOX 2-8 Skeletal System Additive Paget disease Acromegaly Osteopetrosis  

Destructive Active osteomyelitis Gout Osteoporosis Carcinoma

The changes should be in either mAs or kVp. The minimum change in mAs is 30%. Likewise, the change in kVp should be 5% to 8%. These changes provide for a visible alteration in the image density/contrast. There is no set value for each change. The change in technique will depend on the patient’s condition. Cast Radiography It may be necessary to perform radiographic studies of patients who have casts because of postoperative procedures, had trauma, or have other indications for immobilization. The cast may either be wet or dry. A wet cast tends to be more radiopaque than a dry cast. A general guideline is presented

in Box 2-9. Measure the contralateral part to determine normal the mAs value, consult the technique chart, and make the appropriate adjustment increases. BOX 2-9 Cast Radiography Plastica Wet: 3× normal mAs or +15 kVp Dry: 2× normal mAs or +10 kVp Based on normal cast thickness

Fiberglassa Wet: 2× mAs or +10 kVp Dry: 1.5× mAs  

aThickness of cast is important. Some areas may have greater amounts than

surrounding area(s). Magnification Magnification is a useful adjunct when viewing subtle findings in a radiograph. Aside from using a high-quality 10× handheld loupe with fine (detail) screens and film, lower extremity specific x-ray units are not equipped to perform high-quality magnification procedures known as optical and radiographic magnification, which are discussed in Chapter 24, Overview of Special Imaging Studies. However, some basic concepts will be discussed. Optical magnification uses extremely fine grain film and a long SID. The image is then viewed with a high-quality magnifying glass. Radiographic magnification uses a very small focal spot size, with a long SID and high-detail film. Magnification is achieved by increasing the OID; however, recall that a larger OID (high mF) reduces image sharpness. Also, because the object being studied is closer to the x-ray tube, this technique increases the radiation dose to the tissue. Recall the formula for determining the mF:

SOD = SID − OID

Soft Tissue In certain clinical presentations, it may be desirable to enhance visibility of soft tissue structures. Examples include soft tissue masses, articular disorders, Achilles tendon pathology, and embedded foreign bodies (e.g., metal, glass, wood chips, plastics). This can be accomplished by using a high-contrast film. To achieve high contrast, decrease the kVp used for the examination. This can be done by decreasing kVp by 15% and doubling the mAs value. Another method is by decreasing the kVp by an appropriate percentage (between 4 and 8 percent). It is important to remember that as kVp decreases, radiographic density increases and the bones will not be as visible. Soft tissue technique should only be used in selective cases. Because of the lower kVp technique, patient absorbed dose increases. When properly exposed, soft tissue radiographs serve as adjuncts to conventional diagnostic films. PROCESSING AND DARKROOM EQUIPMENT AND ACCESSORIES A darkroom is used for several purposes: film storage, loading, processing, and duplication. Applicable equipment and furniture include a film storage box, workbench, automatic processor unit, and duplicator. The room must be properly lit with a white light while anyone is cleaning the work area and equipment and replacing or replenishing processing chemicals. An exhaust fan or other means of ventilation is necessary to prevent a buildup of noxious fumes in the darkroom. Any doors providing access to the darkroom must be lighttight. The door should have a lock on the inside to prevent unauthorized access during safelight conditions. Safelights or light filters are used during film handling and processing. The workbench area must be kept clean and free from dust and spills. Careless handling of unprocessed film results in artifacts and poor-quality radiographs. Safelight It is necessary to match safelight filtration to the spectral sensitivity of the film so that accidental fogging does not occur. Even then, the best practice is to place safelights as far away from the work surfaces as possible. Also, use a

bulb of the lowest wattage possible to provide enough illumination to know where things are and yet have time to handle films safely. Exposed films are much more sensitive to fogging than are raw films prior to exposure. As discussed previously, there are two types of film and each has its own spectral response to light: one is blue sensitive, and the other is green sensitive. A red safelight filter must be used with green-sensitive (orthochromatic) film. Either an amber or a red safelight filter can be used with blue-sensitive film. However, light passing through the amber filter will fog green-sensitive film. Different types of safelight units are available. Generally speaking, a 15-W (or lower) bulb is used in the safelight lamp fixture with the filter placed in front of the safelight bulb. Units can be purchased that either screw into an existing light fixture or plug into an electrical outlet. The latter is mounted to the wall or ceiling. Fluorescent lights can also be adapted with a slip-on polycarbonate filter sleeve. The safelight must be located at least 4 ft from wherever the film is to be handled. Handle the film quickly and carefully, especially before processing the undeveloped, exposed film. Examine safelight filters closely on a periodic basis. Any filter that has a crack or split should be replaced immediately. Film Storage A film bin is a lighttight storage container for unexposed x-ray film. An economical unit is compact and sits on the countertop. For larger storage needs, a floor-standing cabinet is standard fare. This type of film bin has a hinged front door that tilts outward from the top; it is best positioned under the darkroom workbench. Film bins with locking systems are available that can be opened only in safelight conditions. Film Carry Case The transport of undeveloped, exposed films to a processing area away from the darkroom necessitates the use of a lightproof carrier. It looks like a briefcase and holds any size film up to and including 14 × 17 in. An economical carrying case is made of plastic.

Automatic Film Processors There are two major designs for automatic film processors: full size and compact. They have functional systems for temperature control, transport, circulation, replenishment, and drying. The larger automatic film processor is designed to process all sizes of films up to and including 14 × 17-in films. They have processing capacities of up to 250 sheets or more per hour, depending on the size of film used. The unit may occupy from 4 to 5 ft2. The full-size automatic film processor is usually installed so that the unprocessed film is loaded into the processor in the darkroom and the processed film will exit on the other side of the wall into a radiograph-viewing room. Large film processors are suitable for hospitals, large clinics, and high-volume film processing areas. Compact and economical, automatic cold water processors are also available. They have standby controls that reduce water and power consumption. The time cycles on the processor can vary from 90 seconds to 3 minutes. Thirty-second processors are available for use in emergency and surgical areas. The chemicals in 30-second processors are very concentrated, and the developer and fixer temperatures are very high. Compact film processors require less than 3 ft2 of space and sit on a stand, cart, or tabletop. They are designed for private offices and are ideal for lowvolume film processing. Several different compact processors are available. Some can be installed with a through-the-wall kit so that the unprocessed film is loaded in the darkroom and the processed film exits into a viewing room. However, most compact film processors are designed for darkroom use only. The typical unit has front film feed and top return of the processed film. Others have top feed and front delivery. Some compact film processors have a lighttight feed tray that allows the operator to open the exit door and leave the darkroom during film feed. The processing capacity of compact processors can range from 30 films per day to 125 films per hour. Compact processors are available that accept 10 × 12-in films, sufficient for office settings performing only foot and ankle studies. Like the larger processors, there is often the need for a plumber to install the compact processor. Access to cold water may be all that is necessary.

Plastic replenisher tanks for the developer and fixer are connected to automatic processor units by dispenser tubes. It is important that the tanks are accurately labeled as “developer” and “fixer.” Tanks for large processor units may hold between 14 and 55 gal. These units are usually cleaned and replenished by a maintenance company. Replenishing systems for compact processors either can be small plastic tanks that hold up to 8 gal or can be plastic 1-gal bottles. Film-processing chemicals for compact automatic film processors are supplied in two forms: either as a concentrate that must be mixed with water or in a premixed, ready-to-use bottle. A daylight loading system can be used with general-purpose automatic processors, as well as with some compact, low-volume automatic processors. However, it must be used with daylight cassettes and a daylight identification printer (unless radiopaque labeling tape is used during the exposure). A daylight loader eliminates the need for a darkroom. The film is automatically unloaded through a slot at one end of the cassette and fed into the processor. Some daylight loaders can reload the cassette with an unexposed film before returning it to the radiographer. The daylight film loader may be either built into a daylight automatic processing system or externally attached to a conventional automatic processor. The daylight system is designed for film processing only. Use of a total daylight loading system virtually eliminates film-handling artifacts. It also is an extremely efficient process, although relatively expensive compared to conventional processing equipment and accessories. FILM PROCESSING The manual processing steps include wetting, development, stop bath, fixing, washing, and drying. The purpose of each processing step is as follows: •  Wetting: swelling of emulsion to permit subsequent chemical penetration •  Development: production of a manifest image from the latent image •  Stop bath: terminate development and remove excess chemical from emulsion •  Fixing: removal of remaining silver halide from emulsion and hardening

of gelatin •  Washing: removal of excess chemicals •  Drying: removal of water and preparation of radiograph for viewing Some of these steps are combined for automatic processing. Wetting is included with development, whereas the stop bath is included with fixing. When remnant radiation and light from the intensifying screens strike the film emulsion, an invisible change immediately occurs in the silver halide crystals representing the object that was radiographed. This invisible representation is referred to as the latent image. Going from latent to visible image requires the process of development. Development The principal action of the developer is to change silver ions of the exposed silver halide crystals into metallic silver, and to concentrate this metallic silver in the region of the sensitivity speck. The primary component is hydroquinone, a reducing agent that produces black tones slowly. Another reducing agent phenidone produces shades of gray quickly. The reduction to metallic silver occurs in total darkness or under safelight conditions. While in contact with the developer, prior to contact with the fixer, the image is unstable and, if exposed to light, will be obliterated; the radiograph will become a uniform dense blackness or maximum density. Controlling time and temperature in the developing process is the only way to ensure accurate and consistent sensitometric results. Time and temperature control in the other areas of the image production are not as crucial as in the developing stage. The remaining chemicals in the developer help to swell the gelatin, produce alkalinity, control pH and oxidation, maintain balance among developer components, control emulsion swelling, remove metallic impurities, and stabilize the developing agent. Fixing

The fixing bath is another water vehicle solution with a variety of dissolved chemicals. During fixation, the film is treated so that the image will remain permanent, thereby producing a film of archival quality that does not deteriorate with age, but remains in its original state. The primary component is a clearing agent, thiosulfate, also referred to as hypo. Its principle purpose is to dissolve the unexposed, undeveloped silver halide crystals from the emulsion making it easier to wash out. Remaining functions of the fixer include neutralizing the developer and stopping its action, hardening and shrinking the emulsion, maintaining chemical balance and proper pH, removing aluminum ions, and dissolving other components. Once fixation is complete, the remaining fixer chemicals are washed off the film. If the film is cleared or fixed sufficiently but not washed thoroughly, over time there may be darkening of the image; this is known as hyporetention. It slowly oxidizes the film, causing the image to discolor to brown over a long period of time. The image will also fade and exhibit poor archival quality. Lastly, the film travels through the drying cycle, making the film suitable for prolonged storage. If fixation was incomplete, the film will not dry properly and come out of the dryer tacky with an opalescent appearance. This commonly leads to early archival breakdown and a brown discoloration known as dichroic stain, the result of residual developer. Recovery Systems It is strongly recommended, for environmental safety reasons, that silver be recovered from the processor outflow before being discharged into a common drain. It may even be required at the local or state level (check with the applicable environmental protection agency). This safety measure can be achieved with a silver recovery unit. Developer and fixer can also be recycled. Processing Quality Control Faulty processing produces films that are of poor quality and sometimes unreadable. This results in either potentially missed diagnoses or, if the study is repeated, increased radiation exposure to the patient. These scenarios are

preventable in many instances if the processor is monitored daily. The goal is to maintain high-quality film processing and produce optimal diagnostic images. Uniform radiographic quality can be maintained by performing a simple procedure at the beginning of each working day. This procedure uses the methods of sensitometry and densitometry. Processor quality control should be performed before processing any films that day. Equipment required for this procedure includes a box of control film, a sensitometer, a densitometer, and control charts. The control film should not be from the same box used for patient radiography; a separate, fresh box of film should be used for this purpose. The control film need not be 10 × 12 inches in size; to contain cost, a smaller size can be used. The sensitometer is an instrument that exposes the control film to light. It precisely controls the intensity and duration of film exposure. Light is projected through a neutral density step tablet that modulates the exposure received by the control film. A 21-step density wedge is typically used. The individual performing this test must know the spectral sensitivity of the film being used (blue or green) before sensitometry is performed. The light source is set accordingly, usually by a switch on the instrument. Once exposed, the control film is then run through the warmed-up automatic processor. A densitometer is used to measure optical density, that is, light transmitted through the processed film. The control film is placed in the densitometer, and the light passing through the exposed area on the film is measured. Density values can then be plotted to obtain a curve of film optical density versus log relative exposure; this relationship is called as a characteristic curve or the H&D curve (discussed above). Measured parameters, including base plus fog, speed index, and contrast index, are plotted daily on a control chart. The developer temperature is also recorded daily and plotted on the control chart. Sensitometers and densitometers are available that are small, portable, and battery operated. Other processor quality-control activities include cleaning and maintenance. The transport and crossover racks, including rollers, should be removed and cleaned weekly or more frequently if sludge and debris build up. (Cleaning is

necessary if particles are found on the processed film.) Routine assessment of all mechanical, moving parts should be observed weekly as part of a scheduled maintenance plan. And, replacement of parts before their failure (preventive maintenance) is necessary to prevent processor downtime. Darkroom cleanliness should also be monitored. The room should be cleaned regularly and tasks and date documented. FILM IDENTIFICATION AND STORAGE All radiographic films must be properly identified for medical and legal purposes. Identification should include the physician’s name or facility where the radiographic study was performed, address (city and state), date of study, and the patient’s name and age or birth date. Identification handwritten on or taped to an already processed radiograph is not admissible in a court of law. Identification must be incorporated into the radiograph before processing. This may be performed with a film identification printer, radiographic labeling tape, or marker set. Film Identification Printer Radiographic film is sensitive to light as well as x-rays. The identification printer uses a light source and 3 × 5-in index card to identify a radiograph. It also requires using a cassette that has been prepared by the manufacturer especially for this procedure. The cassette has a rectangular lead blocker located in a corner or along an edge. That portion of the film beneath the lead blocker remains unexposed after the radiographic study. Before the study, patient information is typed or written onto the identification card. This card, ordered from a printer, includes standard information identifying the doctor’s office or facility where the study is performed and provides a template for adding additional information, including the date, the patient’s name, and birth date or age. It is the easiest and the most professional way to perform film identification. The following is an overview of the process. First, fill out the index card with the proper information. Perform the radiographic study. Be certain that the body part is not over the cassette’s rectangular lead blocker. (A sticker usually identifies the area.) In the darkroom, insert the identification card into

the identification printer’s slot, remove the film from the cassette, and slide the rectangular lead block area under the card. (Guides on the identification printer can be adjusted to aid in proper placement of the card and film.) The film will meet resistance and stop when aligned with the card. Press down on the exposure plate or button, and release when the light exposure is complete. The identification printer comes with an adjustment for light intensity and exposure. When the process is complete, the information will be displayed on the film in a rectangular area.

FIGURE 2-17. A: Radiopaque labeling tape holder. B: Writing information on labeling tape. C: Application of tape onto holder. (A: Courtesy of Medical I.D. Systems Inc., Grand Rapids, MI; B and C: courtesy of X-Rite, Inc., Grandville, MI.) The procedure just described applies to the identification printer used in a darkroom. A daylight identification printer can also be used; the rectangular identification area is exposed while still inside the cassette. On its backside, the cassette has a rectangular slot that slides open inside the daylight printer for exposure. Radiopaque Labeling Tape Radiopaque labeling tape contains a semisoft strip of lead or tungsten mixture onto which is written or typed the pertinent information. It can be purchased in precut strips or as a long roll. The protective backing is peeled away from the adhesive tape that can then be placed on the cassette before the exposure. The label tape should be positioned such that it will not be superimposed on the body part being studied. The patient’s information then becomes part of the radiograph.

Identification labeling tape is usually combined with a holder device (Figure 2-17). The holder contains a density filter that blocks excessive radiation so that the information is visible radiographically. Choose a density filter that corresponds to the mAs and kVp settings typically used. The holder can be customized with the name and address of the doctor or facility along its top and bottom edges. Some holders have right (R) and left (L) indicators built in. The label tape with pertinent information is centered on top of the holder and is then placed on the cassette while positioning the patient, before exposure. Right and Left Identification Markers All extremity radiographs must be correctly identified as either right or left. This labeling must also be a permanent part of the radiograph. The capital letters R and L are used to identify right and left extremities (Figure 2-18). The markers vary in size and are made of lead. Stainless steel “clip” markers are also available. This U-shaped device has an R and L punched out on either arm of the clip. The marker slides over the cassette’s edge; the applicable letter is positioned over the front of the cassette. This device is especially useful for vertically positioned cassettes. The R or L will become a permanent part of the radiograph during the exposure. FILM-VIEWING TOOLS View boxes (illuminators), spotlights, and magnifying glasses comprise the required viewing tools. Further information regarding use of these instruments are discussed in Chapter 9, Principles of Radiographic Interpretation. Radiograph Illuminators Bright and even illumination is essential to properly view a radiograph. Holding the radiograph up to a lamp or an overhead light fixture does not provide proper illumination. The viewing area on the illuminator should ideally be the size of the radiograph. Any ambient light that escapes from the margins of a film placed on a larger view box will impair visualization of the image. Illuminators should be selected based on the largest-size film used.

Ideally, differently sized view boxes should be available for each different size of radiograph. Fortunately, an office dedicated to performing only foot and ankle studies will use only one size of film.

FIGURE 2-18. Left and right identification letters. (Courtesy of Providence Imaging Products, Inc., Providence, RI.)

FIGURE 2-19. This 12-by-24–in illuminator can either be wall mounted or set on a desk or countertop. (Courtesy of Wolf X-Ray Corp., West Hempstead, NY.) Illuminators are available with 8 × 10-, 10 × 12-, and 11 × 14-in viewing areas. An illuminator may have a single viewing area for a particular size of film, or a bank of multiple illuminators positioned side-by-side or in a double tier. It is strongly recommended that viewing areas of bank illuminators be separated by reflectors and have individual on/off switches. Surfacemounted, recessed wall-mounted, and desktop illuminators are available (Figure 2-19). Illuminators use either Circline or straight fluorescent tubes. A daylight or cool white fluorescent bulb provides the light source. Use the same type of bulb (cool white or daylight) in an illuminator with multiple bulbs. The two types have different light hues/color temperatures. Film retainers are found

along the top of an illuminator. Their purpose is to grip the film and hold it in place. The grip mechanism varies; try each and choose one that “feels” right to you. Optional film-activated microswitches can be ordered; the bulb turns on as a film is placed under the retaining clip. The viewing panel is a white, radiolucent acrylic sheet. It is easily removed to service the light bulbs and fixtures. The viewing panel must be kept clean and free of dirt, marks, and scratches. Magnifiers Magnifiers are useful aids for radiographic interpretation. They can be handheld or mounted on a moving arm attached to a view box. Spotlight Illuminators A bright spotlight increases the illumination in a small area of the film and enhances visibility. It is not meant to be a replacement for technically overexposed or overdeveloped films. The typical spotlight is equipped with a standard incandescent lamp that should not exceed 75 W (check the manufacturer’s recommendation). Films will easily be damaged with higherwattage bulbs. An attached rheostat foot pedal allows the viewer to control gradients of brightness (Figure 2-20). On more expensive units, an adjustable shutter can limit the viewing area to a smaller desired area of interest, and intensification can be adjusted accordingly. Use the light intermittently to avoid burning the film, which can occur quite quickly. A reflector surface surrounds the bulb; both must be kept clean.

FIGURE 2-20. Spotlight illuminator with rheostat foot pedal. (Courtesy of Wolf X-Ray Corp., West Hempstead, NY.) FILM LABELING, FILING, AND DUPLICATION Film Envelopes/Preservers Film preservers are constructed of heavy brown Kraft stock. The envelopes can be ordered with a preprinted template for patient identification, which may be either ruled or unruled. They come in many sizes, styles, and colors.

X-ray File Cabinets Radiographs should be stored in a cabinet with firm, preferably metal, reinforced shelving. Wood shelves may warp under the weight of many files. Open and stackable film storage units are available, as are enclosed filing cabinets with pullout drawers. Steel drawer filing cabinets specific for storing radiographs are also available. Filing System Radiographs can be filed with the medical chart or separately. If separate, it should be filed according to the system used for the medical file or chart. It is much easier to retrieve radiographs with matching case numbers or names. Film Label Tape If a patient has or will have numerous radiographic studies, it is best to colorcode the films and index the code on the film envelope. A color-coded tape system will expedite retrieval of studies: performed on a specific date. The date is logged on the front of the patient’s film envelope, and a piece of colored label tape (approximately 1 inch in length) is placed adjacent to the date. A piece of the same color tape is then placed across the top edge of the film or films created on that date. Films are stored with the top edge of the film (with the tape) at the top or open end of the envelope. Logbook Many states require that an x-ray study logbook be maintained. Information should include the date of the examination, patient name and file number (if applicable), techniques performed, purpose of the study, technique (mA, kVp, SID, and time), type of film/screen system, and the name of any person holding the patient during an exposure. All information should be recorded immediately following the study. Radiographic Film Duplicators The radiograph is a vital component of a patient’s medical history. Occasionally a patient, physician, insurance carrier, or attorney may request

copies of radiographs. Or you may desire to send a study to another physician for consultation. Never let an original film leave the office unless a copy has been made first. It is strongly recommended to keep the original and send the copy. Duplicating film, a direct reversal film, is used to copy radiographs. The emulsion is only coated on one side of the film. It is formulated to react to light sources with a high ultraviolet light output, such as a BLB (black light blue) fluorescent lamp, included with many commercially available duplicators. There are many types of film duplicators to choose from. Large, selfcontained units include a BBL fluorescent lamp and operate in one of two fashions: either the films are placed onto a glass plate (known as a box-type contact printer) or they are fed through a processor-like machine with a rotating cylinder. Depending on the unit, there will be controls for the light intensity or the time of exposure. Exposure times are typically on the order of several seconds. An economical method of copying radiographs employs a cassette-like device known as a printing frame. It has a rigid frame with a clear glass or plastic plate front that uses overhead light fixtures or a view box as the light source. Radiographs are easily copied, but certain guidelines must be followed. The duplicating film should be the same size as the original radiograph. A specific order and placement of the two films is required: the original radiograph is positioned closest to the light source, and the emulsion side of the duplicating film is placed against the original radiograph. (Duplicating films have a notch in one corner that aids in film placement.) The two films (original and duplicating) are then placed in the copying device and covered tightly (if applicable). The copying process requires that both the original and the duplicating films be held in close contact with one another. Poor film-to-film contact will result in blurring of the duplicate image. Variable light or timer controls allow a dark or light original film to be enhanced. The light intensity or time is increased to make a duplicate radiograph appear less dense than the original and is decreased to make it darker (increased film density). ARTIFACTS

An artifact is unwanted information that has been recorded on the film. It is not caused by the superimposition of anatomy in the primary x-ray beam. Artifacts often can be “read around,” but when more than an annoyance, they can and often do result in misinterpretation.

FIGURE 2-21. Kink mark (arrow)—severe enough to cause sensitization. Radiographic film has reached such a high standard of quality and uniformity that they are not the source of artifacts alone; in most cases, faulty storage, handling, or processing cause artifacts. Physical artifacts include marks or blemishes on the film caused by transport, rough handling, scratches, kinks, and any direct physical damage to the emulsion surface. An example of severe handling is the kink mark (Figure 2-21). It may look like a fingernail mark, but it is not! Processing artifacts may be more difficult to determine, because they are usually caused by the presence or absence of chemicals or a physical or chemical reaction in or on the film emulsion. Deposits on the film, stains, and

fog manifest such defects. Static Static is caused by a discharge of electricity. The light from a spark of electricity exposes the film. If the electrical charge is slight, the result is smudge static; if it is severe, the result is tree static (Figure 2-22). Static is less a problem when the relative humidity is between 40% and 60%. It often occurs when the atmosphere is extremely dry, such as in temperate zones during the winter when the heat is on. If static snaps from your hand to a doorknob, you will be sure to see it on your films. To minimize static, a humidifier can be used to replace the moisture.

FIGURE 2-22. Tree static artifact.

FIGURE 2-23. Edge fog. Film bin or storage box exposed to white light along film edge (arrowheads). Normally not seen on a radiograph because maximum exposure is given to edge of film. Fog Fog is unwanted density that fills in the toe density, reducing the contrast of the fine detail in that area. Fog can be caused by radiation, white light (Figure 2-23), overexposure to safelights, age of film, and chemical reaction. Although it is difficult to determine the exact cause of fog after the fact, some general guidelines may help reduce it. Keep cassettes out of the exposure room. Keep them as far away from radiation as possible. Radiation fog creates an overall gray appearance, and, because it is penetrating, will record objects between the source and the film. If too high, a kilovoltage level is used for the part being radiographed, and then the secondary and scattered radiation will be recorded as fog. White light fog results in much greater densities and is not as uniform unless the entire film is exposed to the beam of light. Safelight fog is usually caused by excessive illumination, a short distance between the safelight and the film, an improper filter, or long safelight exposure time. Often these causes are combined. By the process of elimination, this kind of fog can often be minimized. Note that after an initial exposure to radiation and the light from the screens, films are more sensitive to subsequent exposure to safelight. It is best to keep exposed films from as little safelight exposure as possible. Often, merely the placement of the safelight over the feed tray of the processor or over the hand tank is enough to cause safelight fogging. Relocating the safelight will eliminate the problem. When fresh, film has a base plus fog range from 0.15 to 0.20 normally. As film ages, fog increases. Once the fog reaches 0.25 and higher in density, information is being lost. Film will remain at a lower fog level throughout the period prior to the expiration date if it is stored at a constant temperature with relative humidity of 40% to 60%. If it is stored longer, or exposed to high heat levels and high humidity, the fog will rise to objectionable levels. Film fog has also been associated with cold storage. Old chemistry or high temperature can result in chemical fogging. If the

unexposed portion of a film, which is usually clear and bright, is fogged, the source may be chemical. Stain Chemical drops on the film prior to development or on the intensifying screen prior to exposure can result in unwanted artifact (Figure 2-24). Fixer stain is a frequent culprit. Keep intensifying screens at a distance from all chemicals; inspect them regularly, and clean or replace as needed. Hyporetention results in yellowish discoloration of the entire film.

FIGURE 2-24. Stain. A: Drop of fixer (arrow) came in contact with film prior to development. B: Fixer stain on screen.

FIGURE 2-25. Scratches. A: Deep gouge from rough handling during processing (arrow). Also note stain artifacts. B: Scratches caused by film being in contact with rough surface, probably in developer, with another film.

FIGURE 2-26. Double exposure. Lateral views of the left and right foot are superimposed on one another. Scratches As mentioned earlier, rough handling of film before and during processing will result in unwanted artifacts. Figure 2-25 demonstrates some examples. Technical Error Technical errors occur during the radiographic exposure and frequently require repeat films. Errors include poor patient positioning, patient motion, and improper exposure technique. Examples include inadequate exposure and double exposure (Figure 2-26). REFERENCES   1. Basic X-ray Machine Operator Study Guide. Tallahassee, FL: Florida Department of Health and Rehabilitative Services; 1986.   2. Block IH. Wedge-shaped filters for radiography of the foot. J Am Podiatry Assoc. 1968;58:182.

  3. Bushong SC. Radiologic Science for Technologists: Physics, Biology and Protection. 3rd ed. St Louis, MO: Mosby; 1984.   4. Bushong SC. Radiologic Science for Technologists. 10th ed. St Louis, MO: Elsevier; 2013.   5. Camasta CA, Pontious J, Boyd RB. Quantifying magnification in pedal radiographs. J Am Podiatr Med Assoc. 1991;81(10):545.   6. Carlton RR, Adler AM. Principles of Radiographic Imaging: An Art and a Science. Albany, NY: Delmar; 1992.   7. Carlton RR, Adler AM. Principles of Radiographic Imaging: An Art and a Science. 4th ed. Clifton Park, NY: Thomson Delmar Learning; 2006.   8. Chamberlain WE, Henny GC. Medical Physics. St Louis, MO: Mosby; 1944.   9. Christensen EE, Curry TS, Dowdey JE. An Introduction to the Physics of Diagnostic Radiology. 2nd ed. Philadelphia, PA: Lea & Febiger; 1978.  10. Christman RA, Zulli LP. Radiologic aspects of aging in the foot. Clin Podiatr Med Surg. 1993;10:97.  11. Cullinan JE, Cullinan A. Illustrated Guide to X-ray Technics. 2nd ed. Philadelphia, PA: JB Lippincott; 1980.  12. Curry TC, Dowdey JE, Murry RC. Christensen’s Physics of Diagnostic Radiology. 4th ed. Philadelphia, PA: Lea & Febiger; 1990.  13. DeVos DC. Basic Principles of Radiographic Exposure. Philadelphia, PA: Lea & Febiger; 1990.  14. Eastman T. Radiographic Fundamentals and Technique Guide. St Louis, MO: Mosby; 1979.  15. Fischer E. Low kilovolt radiography. In: Resnick D, ed. Bone and Joint Imaging. Philadelphia, PA: WB Saunders; 1989.

 16. Fosbinder R, Orth D. Essentials of Radiologic Science. Baltimore, MD: Walters Kluwer Health/Lippincott Williams & Wilkins; 2012.  17. Fuchs AW. The Science of Radiology. Springfield, IL: Charles C Thomas; 1933.  18. Geissberger H. Wedge-shaped filters for improved radiography of the thoracic vertebrae and the foot. Med Radiogr Photogr. 1966;42:6.  19. Genant HK, Doi K, Mall JC. Optical versus radiographic magnification for fine-detail skeletal radiography. Invest Radiol. 1975;10:160.  20. Genant HK, Doi K. High-resolution skeletal radiography: image quality and clinical applications. Curr Probl Diagn Radiol. 1978;7:3.  21. Genant HK, Resnick D. Magnification radiography. In: Resnick D, ed. Bone and Joint Imaging. Philadelphia, PA: WB Saunders; 1989.  22. Groenendyk DJ. Densitometers and sensitometers in QC. Radiol Technol. 1994;65(4):249.  23. Health Sciences Market Division. The Fundamentals of Radiography. 12th ed. Rochester, NY: Eastman Kodak; 1980. Kodak Publication NO-M118.  24. Heinlein CW. Radiographic technique for infants and children. Radiol Technol. 1966;38:25.  25. Jennings MB, Cohen RE, Marino JP, et al. Evaluation of compensation filters in pedal radiographs. J Am Podiatr Med Assoc. 1999;89(4):169–173.  26. John DHO. Radiographic Processing in Medicine & Industry. New York, NY: Focal Press; 1967.  27. Josephs RL. Technique for duplicating x-rays. J Am Podiatry Assoc. 1974;64:794.  28. Kath K. Pocket Reference to Radiographic Exposure Techniques. St Louis, MO: Mosby; 1993.

 29. Kodak X- OMAT Duplicating Film Technic Guide. Rochester, NY: Eastman Kodak; 1982.  30. Lawrence DJ. A simple method of processor control. Med Radiogr Photogr. 1973;49:2.  31. Malott JC, Fodor J. The Art and Science of Medical Radiography. St Louis, MO: Mosby; 1993.  32. McKinney WEJ. Radiographic Processing and Quality Control. Philadelphia, PA: JB Lippincott; 1988.  33. Mees CEK, James TH. The Theory of Photographic Process. 3rd ed. New York, NY: Macmillan; 1966.  34. Miller TH, Brummitt W. This Is Photography. New York, NY: Garden City Publishing; 1945.  35. Ngo C, Yaghmai I. The value of immersion hand radiography in soft tissue changes of musculoskeletal disorders. Skeletal Radiol. 1988;17:259.  36. Patterson CVS. Roentgenography: Fluoroscopic and Intensifying Screens, Medical Physics. Vol 2. St Louis, MO: Mosby; 1950.  37. Patterson Intensifying Screens. Wilmington, DE: E. I. du Pont de Nemours & Co.; 1950.  38. Processor Quality Assurance: An Informational Guide to Monitoring Xray and Cine Film Processors. Grandville, MI: X-Rite; 1990.  39. Sensitometric Properties of X-ray Films. Rochester, NY: Eastman Kodak; 1974.  40. Spencer RB, Bradley MB. Radiographic reproduction. Podiatry Today. February, 1990:45.  41. Sprawls P Jr. The Physical Principles of Diagnostic Radiology. Baltimore, MD: University Park Press; 1977.

 42. Sprawls P. Physical Principles of Medical Imaging. Rockville, MD: Aspen; 1987.  43. Sundaram M. The clinical value of direct magnification radiography in orthopedics. Skeletal Radiol. 1978;3:85.  44. Weiss A. A technique for demonstrating fine detail in bones of the hands. Clin Radiol. 1972;23:185.

3 Digital Radiography ROBERT A. CHRISTMAN AND MARY OEHLER X-rays interact with an image receptor and form the latent image (an invisible change that represents the object that was radiographed) after passing through the patient. The visible (manifest) radiographic image is produced following processing of the latent image. Conventional radiography has, for more than 100 years, employed as an image receptor the film–intensifying screen combination. The exposed film must be processed with chemicals to produce the manifest image, which is then viewed on a light box, cataloged, and physically stored. With the advent of filmless or digital radiography (DR) in the late 20th century, an electronic detector has replaced the film–screen image receptor. The latent image formed on the electronic detector is measured and converted into an analog electrical signal that is then converted into a digital signal, which is processed by a computer, creating a digital image that is viewed on a monitor. These images are then stored electronically. No matter what image acquisition system is used to process the image, however, they all depend on the same x-ray machine. Furthermore, the four quality factors that the limited x-ray machine operator (LXMO) controls before the x-ray beam strikes the image receptor, radiographic density, contrast, recorded detail, and distortion, apply to DR as well. These factors are discussed in detail in Chapter 2. Attempts have been made to categorize the different types of DR. They have been categorized by form factor (cassette-based vs. no cassette), image acquisition time (immediate vs. longer), x-ray signal conversion (indirect vs. direct), and linkage to computer (indirect vs. direct). We have found it useful to categorize by the type of detector system, but will include reference to the other categorizations when appropriate. DIGITAL RADIOGRAPHY DETECTOR SYSTEMS

Four DR detector systems are available for podiatric use: the photostimulable storage phosphor (PSP) detector, charge-coupled device (CCD) detector, complementary metal-oxide semiconductor (CMOS) detector, and flat-panel detector (FPD). Each has three components: the capture element, coupling element, and collection element. The x-ray may be captured by a photostimulable phosphor, scintillation phosphor, or a photoconductor. The signal generated by the capture element is transferred or coupled to the collecting element via an optical system, photodiode, or a photoconductor. The element that collects the x-ray–generated signal may be a photodetector, CCD camera, thin-film transistor (TFT) array, or CMOS. The collection element sends an electric signal to an analog-to-digital converter (ADC), which then goes to a computer that has appropriate software installed. Computer algorithms convert the digital data into a digital image that is then viewed on a monitor. The DR field is growing and changing rapidly. The descriptions of the various technologies below are generic, since there is constant development of new and improved devices. However, the emphasis is on application to podiatric radiology. Photostimulable Storage Phosphor Detectors Also known as computed radiography (CR), the PSP detector is widely used in podiatric settings because it has been available the longest. The initial investment also is less expensive compared to the remaining DR systems, which may also require retrofitting of existing x-ray equipment (in particular, the orthoposer). Although several newer DR detector systems have become available, CR will more than likely prevail for several years to come. The capture element is a photostimulable storage phosphor, barium fluorohalide with europium. The PSPs are incorporated into a storage phosphor screen. Whereas x-ray film and intensifying screens are housed in a cassette, the PSP screen is housed in what is called an imaging plate (IP). The IP is not linked directly to a computer. The PSP electrons are energized when exposed to x-rays and form the latent image. (At least half of these electrons return to their ground state

immediately and release light when doing so, but the remaining electrons may stay in an excited state for several hours. Furthermore, in 8 hours the latent image will lose approximately 25% of its energy. Therefore, the IP should be processed soon after exposure.) When the IP is placed into a PSP reader, the phosphor screen is removed and scanned by a laser beam; this stimulates many of the remaining excited electrons to their ground state and the PSPs emit light, referred to as photostimulable luminescence (PSL). The light is collected and directed to a photodetector via an optical system. The photodetector measures the light signal, then amplifies the light and sends it as an electrical signal to an ADC. The x-rays, therefore, are indirectly converted into an electrical signal. Lastly, the phosphor plate is exposed to a bright light that erases the latent image and any remaining excited electrons so that the IP can be used again. The PSP IP is always “on,” meaning that it is sensitive to scatter radiation; therefore, it should not be stored in a radiography room. Since it is also sensitive to background radiation, if the IP is not used for a period of time (e.g., over the weekend), it should be erased again before reusing it. Of all the forms of DR, CR is the only one to functionally emulate the film– screen paradigm; similarly, CR shares some of its limitations. The image receptor is “cassette-based” (the IP), which has to physically be transferred to a “processor” (the PSP reader), which will likely be in another room. The IP can be easily damaged, especially if dropped while transporting the cassette to and from the IP reader. The phosphor screen can be processed in two ways: after “manually” removing it from the IP, or the IP is placed into a processor that “automatically” reveals the phosphor screen for processing (Figure 3-1). In either case, the time to “process” the image by the PSP reader takes approximately 1 minute. These are a few of the limitations of CR that have motivated manufacturers to develop FPD technology. On the other hand, because of its mobility, the cassette-based system allows for a multitude of positioning techniques. Charge-Coupled Device Detector The CCD detector is the same light sensor that is used in most digital cameras. The detector is not housed in a cassette and requires modification of the orthoposer because the CCD detector can be rather large and bulky

(Figure 3-2). It uses a phosphor scintillator, either cesium iodide (CsI) or gadolinium oxysulfide (GdOS), which emits light when exposed to x-rays. Light from the scintillator is transferred to a CCD by an optical system (lenses or fiber optic tapers) that demagnifies the light. After collecting the light, the CCD detector produces an electrical signal that is then converted into a digital signal by an ADC. Similar to CR, the x-rays are converted indirectly into an electrical signal; however, unlike CR, the CCD detector is directly linked to a computer. The image acquisition time is 20 to 30 seconds. Some consider CCD DR an “interim” technology until FPDs become more readily available. Though it is an improvement over CR, eliminating the cassette with separate PSP reader and reducing the time for image development, the optical collecting element reduces image quality during demagnification. Also, the production cost of CCDs is high, and they are susceptible to radiation damage. Flat-Panel Detector A flat-panel detector, often integrated into the orthoposer, is used for direct and indirect conversion DR; however, it is totally different in design and structure from the PSP device. It consists of an x-ray absorber material (photoconductor or scintillator) that is coupled to a TFT array. The FPD converts the absorbed x-rays into an electrical signal directly (direct conversion DR) or indirectly via light (indirect conversion DR). Similar to CR and CCD-based DR, a computer then processes the electrical signal from the collection element that has been converted into a digital signal by an ADC. (FPD systems have also been referred to as active matrix flat-panel imager (AMFPI) systems.)

FIGURE 3-1. “Automatic” PSP reader. Cassette is being fed into reader from the right. This reader is a tabletop unit that extends to the left, behind the monitor.

FIGURE 3-2. CCD detector system with retrofit orthoposer base. This detector unit can be moved to accommodate weight-bearing dorsoplantar and lateral positioning techniques. (Courtesy of 20/20 Imaging, LLC, Crystal Lake, IL.) The direct conversion DR system uses the photoconductor amorphous selenium (a-Se) instead of a scintillator to absorb the x-rays. The a-Se is coupled to a TFT array so that the x-rays are directly converted into an electrical signal. The intermediate scintillation layer is not necessary and is eliminated; the a-Se serves as both the capture and coupling elements by converting x-rays directly to charge. Similar to CCD-based DR, the indirect conversion DR system uses a phosphor scintillator (CsI or GdOS) that emits light when exposed to x-rays. But that is where the similarity ends. Light from the indirect conversion DR phosphor scintillator is transferred to a TFT array by an amorphous silicon

(a-Si) photodiode array, an electronic element that converts light into charge. After collecting the light, the TFT array produces an electrical signal. Complementary Metal-Oxide Semiconductor Detector The CMOS system has had problems that slowed the introduction of this technology into medicine. In particular, the presence of electronic noise has hindered development; however, there has been some success with small detector sizes, and there is at least one CMOS detector available for podiatric use. Like the CCD detector, the orthoposer must be modified. CMOS systems use a scintillating phosphor element, such as CsI that emits light when struck by x-rays. The CMOS has photodiodes and storage capacitors built in to them, though the light may also be coupled optically to the CMOS. The light is converted into electrons by sensors and stored in the capacitors. The most popular semiconductor material is silicon. Because semiconductors alone do not conduct electricity very well, impurities (also known as dopants) are added to them. The “doped” semiconductors then become highly conductive and act as transistors that amplify the charge that is sent to the ADC. CHARACTERISTICS OF THE DIGITAL IMAGE An ADC converts the electrical (analog) signal from the DR detector into binary numbers (the digital signal). These binary digits are referred to as bits. (A “word” consisting of 8 bits is a byte, a term used to indicate storage capacity.) Bits are discrete numbers (0 or 1) that are processed by a computer and transformed into a two-dimensional array of rows and columns of numbers referred to as a matrix. The number of picture elements (pixels) in these rows and columns determines the matrix size. (For example, a 512-by512 matrix contains 512 pixels in each row and column.) Therefore, a larger matrix contains a greater number of pixels than a smaller matrix. The smallest unit in a digital image is the pixel. Each pixel has three numbers associated with it: two numbers define its specific location in the matrix (along the x- and y-axes); the third numeric value represents the brightness or intensity of the pixel at that location. The pixel is a two-dimensional representation that corresponds to a three-dimensional volume of tissue,

called a voxel. Field of view (FOV) describes the part of the body being imaged. The size of a pixel (which equals FOV/matrix size) is directly related to spatial resolution. The pixel bit depth, that is, the number of bits (or pieces of information) within a pixel, refers to the number of shades of gray that a pixel can produce in the image. (Bit depth is also known as the z-axis of a pixel.) DR systems typically use 8, 10, or 12 bit depth. Because the binary number system uses base 2, if the bit depth is represented by the letter k, each pixel will have 2k gray levels. For example, if the bit depth of a pixel is 10, then the number of shades of gray that can be produced by the pixel is 210, which equals 1024 shades of gray. Bit depth is directly related to contrast resolution. Therefore, the appearance of the digital image, especially its spatial resolution and contrast resolution, will be affected by the matrix size, pixel size, and bit depth. Spatial Resolution Spatial resolution (also referred to as detail, sharpness, or definition) is the ability of a system to image the small details of high-contrast objects and to accurately display these objects in two dimensions. It is directly related to pixel size (in the x- and y-axes): the smaller the pixel, the higher the spatial resolution. (In film–screen radiography, resolution is determined by the sizes of the film crystal and intensifying screen phosphor.) So, if the FOV is fixed, increasing the matrix size will decrease the pixel size, resulting in higher spatial resolution. Also, if the FOV is decreased, and the matrix and pixel size remains unchanged, then spatial resolution will increase. On the flip side, images with higher spatial resolution require larger file sizes. Line pairs per millimeter (lp/mm) is the measurement for spatial resolution, also referred to as spatial frequency. This can be equated to the number of parallel black lines, equally separated from one another by the same width of the line, in a set distance on a white background. Since it takes two pixels to form a line pair, spatial resolution is limited to the size of the pixel, a disadvantage of DR. Spatial resolution is better at high spatial frequency (high lp/mm), which is required, for example, when evaluating a bone’s trabecular pattern. A limitation of DR compared to film–screen radiography

is its lower spatial resolution (4–6 lp/mm vs. 8–10 lp/mm, respectively). At 4 lp/mm, the pixel size would be 125 µm. For perspective, the approximate spatial resolution for bone scintigraphy is 0.1 lp/mm, for CT and MRI is 1.5 lp/mm, and for mammography is 15 lp/mm. Modulation transfer function (MTF) describes the ability of an imaging system to accurately reproduce an object’s structural detail (sometimes referred to as fidelity or trueness of the image). MTF is a measure of an imaging system’s ability to preserve signal contrast as a function of spatial frequency, that is, the image-to-object ratio. The MTF of a perfect imaging system, for example, where the object is equal to the image, is 1 or 100%. However, as spatial frequency increases, MTF decreases because blurring occurs and contrast decreases. Objects demonstrating low spatial frequency are easier to image than those with high spatial frequency. Therefore, at low spatial frequencies DR has a higher MTF. Contrast Resolution Contrast resolution (also known as gray scale) is the smallest density change between two tissues that is detectable. This directly relates to an imaging system’s ability to display a range of grays from black to white. The number of shades of gray that an imaging system can reproduce is known as its dynamic range. A high-contrast image has fewer shades of gray than a low-contrast image and is referred to as having short-scale contrast (also called narrow latitude). Conversely, a low-contrast image contains many shades of gray (long scale, wide latitude). The bit depth of a pixel identifies the dynamic range (or latitude) of an imaging system. For example, a 14-bit dynamic range, such as that used in DR, demonstrates 16,384 (214) distinct gray values. (The 12-bit dynamic range of CT and MRI has 4096 (212) distinct values for gray scale.). Even though the human eye can only distinguish 30 shades of gray simultaneously, all shades of gray are available and accessible with postprocessing, which allows more information to be extracted from the image. Contrast resolution is limited by noise, which is anything that interferes with image formation. Noise (quantum mottle) increases as Compton-scattered xrays increase, and digital image receptors are more sensitive to it than film–

screen receptors. (Antiscatter grids are not required for thin body part radiography, such as the extremities, which generate less scatter.) Collimation regulates scatter radiation that can fog the film (film–screen radiography), degrade contrast resolution (DR), and be absorbed by the patient. Signal refers to the x-rays that exit the patient and correspond to anatomical attenuation by the subject. A higher signal-to-noise ratio (SNR) correlates to higher contrast resolution. (Recall from Chapter 2 that Compton scattering can be reduced by lowering the kVp; however, lower kVp increases the photoelectric effect, which increases the patient-absorbed dose. Raising the mAs can also increase the SNR; however, this also increases the patientabsorbed dose. Just as in film–screen radiography, the appropriate radiographic technique should be chosen that offers a high-quality image while the exposure dose is as low as reasonably achievable [ALARA].) CR systems may require higher exposure techniques in order to increase its SNR. The SNR of a CCD DR system is further reduced by its demagnification of optical collecting element. Detective Quantum Efficiency Detective quantum efficiency (DQE) is a measure of detector-image quality performance, that is, how efficiently the digital image is formed from x-rays. DQE represents the percentage of remnant x-rays that are absorbed by the image detector; ideally, it would be 100%. Generally speaking, the DQE of film–screen, CR, and CCD-based detector systems is less than that of indirect and direct DR systems, and the DQE of a flat-panel system is double that of a CR system (60% vs. 30%, respectively). Exposure Technique and Patient Dose Considerations Compared to film–screen radiography, a lower exposure technique should be possible with DR. However, a significant reduction in exposure technique (mAs) will result in more image noise (quantum mottle) and a lower SNR. Therefore, less care may be taken by the LXMO to adjust exposures appropriately. And, if one were to err, it is better to choose a technique resulting in overexposure. As a result, higher exposure dose may result; this is referred to as “dose creep” or “exposure creep.” The use of radiographic technique charts for all positioning techniques posted for each x-ray detector system should prevent this from occurring.

A good digital image can be achieved even with poor exposure technique. This is because of the wider dynamic range of the digital detector. It is rare that a digital image is so poor due to exposure technique that it would need to be repeated. Contrast resolution is easily influenced by overexposure or underexposure in film–screen radiography, but, regardless of the exposure, it is preserved in digital imaging. As a result, kVp is not as important in formation of the digital image as it is with radiographic film. In film–screen radiography, kVp controls contrast and mAs controls radiographic density; this is not as important in DR. VIEWING THE DIGITAL IMAGE Two types of high-resolution monitors are currently in use: the cathode ray tube (CRT) and the liquid crystal display (LCD). The CRT monitor was popular because of its higher resolution, brightness, viewing angle, and low cost. However, the LCD takes up less space, produces little or no glare, consumes less energy, and produces less heat; these factors have led to it becoming the primary monitor in the market and on store shelves. The flat-panel LCD monitor that a digital image is displayed on should be viewed straight on because, as with many television monitors, contrast and light from the image may not be viewed as well from an angle. For acceptable contrast, the display’s luminance ratio (LR) should be a minimum of 250; 350 is preferable. The pixel pitch (the space of pixel structures) determines the maximum spatial frequency of the image. Pixel pitch should be around 0.200 mm but no larger than 0.210 mm. An appropriate display size is 21 in viewed in portrait presentation, not in wide-screen (landscape). The screen resolution should be at least 1600 × 1200 (2k). Be aware of the reduced image resolution and potential diagnostic limitations of viewing digital images on a display that is only 1280 × 1024 (1k). In an attempt to improve the appearance of a digital image, it can be manipulated after processing. Postprocessing the digital image is performed manually. Examples include window and level, magnification, annotation, and image flip (Figure 3-3). Multiple images can also be viewed simultaneously.

FIGURE 3-3. A: Example of a menu bar from imaging software that offers image manipulation tools. B: An image with text and arrow annotations.

The brightness (or density) of the image is controlled by window level. Lowering the window level makes the image brighter (or decreases the image’s density); increasing the window level increases the image density, making it blacker overall. The choice of appropriate image brightness is based on the visibility of proper anatomical densities in the area of concern. Window width controls the contrast or shades of gray in the image. Narrowing the window width increases the image contrast, resulting in shortscale contrast; increasing window width decreases contrast. Since foot and ankle radiography is primarily concerned with imaging the bony anatomy, a narrow window width is preferred. Small areas of an image or the entire image may be magnified (Figure 3-4). Images can be turned or flipped to match the orientation of other images. Text or graphics, such as arrows or circles, can be added to the image through annotation. Axes can be drawn and angles measured as well. Digital Image Artifacts Errors in the digital system can result in fogging (due to oversensitivity of the IP) and quantum mottle (noise caused by inadequate exposure technique, i.e., insufficient mAs). Image receptor artifacts will result from pixels that are not working properly (defective) and scratches or dirt on the IP. Also, if the IP is not completely erased from a previous image, ghost artifacts (also called phantom images) may appear. Cracks in the IP will also be seen in the image, secondary to wear and aging; in this case, the IP should be replaced.

FIGURE 3-4. Spot magnification. If a TFT flat-panel receptor is immediately reused, it may demonstrate what is known as image lag. Also referred to as “memory effect,” some of the image may persist as some excitable electrons, still remaining after exposure is complete, are slowly releasing energy over a period of time. Quality Control In addition to the quality control measures that are performed on an x-ray machine, the luminance of display monitors should be measured and monitored for uniformity with a photometer. Additional electronic test patterns are also used to assess the quality of the monitor display. ADVANTAGES OF DIGITAL RADIOGRAPHY

The primary advantage of DR over film–screen radiography is the wide dynamic range response of the digital imaging receptor. The response of a digital imaging receptor to x-rays is linear, unlike the characteristic curve of a film–screen system, which has a limited “straight-line” range (Figure 3-5). Therefore, the digital imaging receptor has greater exposure latitude over film–screen. Recall that the characteristic curve of film–screen has a toe and a shoulder, where no significant contrasting densities are formed. The toe corresponds to underexposed areas within an image; the shoulder region of the curve corresponds to areas of overexposure. In DR, however, the computer can enhance the underexposed areas, and the overexposed areas can be brought down into the visible density range. Therefore, less radiation exposure is necessary to produce an image of similar density to conventional film–screen radiography, and the digital image is more forgiving for overexposure and underexposure. Repeating a radiographic study due to poor exposure technique, which reexposes the patient, is rarely necessary. Images processed and created from film–screen radiography are permanent, and quality cannot be changed. Characteristics of the digital image, however, can be enhanced after processing. The ability to change and optimize the contrast is of great value, as is the ability to enhance visibility of detail. Also, digital images can be soft copied and duplicated (digitally) without loss of image quality. (Hard copies can also be produced if a laser imager/printer is added to the system.)

FIGURE 3-5. Characteristic curve of a film–screen system, which has a limited “straight-line” portion versus digital system, which is linear. Existing radiographic film images can be converted into digital format by use

of a film digitizer. There are two types of digitizers: laser and CCD. Laser film digitizers are the gold standard, but they are more expensive than the CCD technology. The CCD digitizer, however, is slower. Other advantages of DR, depending on the detector system used, include the ability to create an image very quickly (in some cases the image can be previewed instantly) and efficiently by bypassing chemical processing. (The time to process a CR image is not much less than processing a radiographic film automatically.) Workflow is greatly improved: images can be digitally transferred and viewed at multiple locations simultaneously. Physical storage space is immensely reduced, and images are rapidly stored on or retrieved from an electronic storage device. Another large motivator to adopt DR over film–screen radiography is its potential to reduce the costs associated with processing, managing, and storing films. MANAGEMENT OF THE DIGITAL IMAGE The setup for a radiology information system (RIS) in the office setting may simply include the image acquisition device (digital image receptor) and a computer to process, manipulate, and store the image. If, however, you want to link your RIS to an existing electronic medical record (EMR) system and/or want to distribute images to other workstations, you may need to use a picture archiving and communications system (PACS). The PACS system uses a standardized communication protocol known as DICOM (digital imaging and communications in medicine) to manage and transfer the digital images. Most stand-alone RISs are DICOM compatible. (However, be aware of potential DICOM compatibility issues between vendors.) Textual information is communicated by a standardized format known as HL-7 (Health Level Seven). A stand-alone RIS system using only one image acquisition device and a single local area network (LAN) is known as miniPACS. REFERENCES   1. American Association of Physicists in Medicine (AAPM) Task Group 18. Assessment of display performance for medical imaging systems.

http://www.aapm.org/pubs/reports/OR_03.pdf. Accessed January 24, 2014.   2. Andriole KP, Ruckdeschel TG, Flynn MJ, et al. ACR–AAPM–SIIM practice guideline for digital radiography. J Digit Imaging. 2013;26:​26–37.   3. American College of Radiology. ACR practice guideline for communication of diagnostic imaging findings. Resolution 11; 2010.   4. Berthel A, Bonin T, Cadilhon S, et al. Digital radiography: description and user’s guide. DIR 2007—International Symposium on Digital Industrial Radiology and Computed Tomography; June 25–27, 2007; Lyon, France.   5. Bushong S. Radiologic Science for Technologists. 10th ed. St. Louis, MO: Elsevier Mosby; 2013.   6. Carlton RR, Adler AM. Principles of Radiographic Imaging: An Art and a Science. 4th ed. Clifton Park, NY: Thomson Delmar Learning; 2006.   7. Carter C, Veale B. Digital Radiography and PACS. 2nd ed. Maryland Heights, MO: Elsevier Mosby; 2014.   8. Dougherty G. Digital Image Processing for Medical Applications. Cambridge, UK: Cambridge University Press; 2009.   9. Fosbinder R, Orth D. Essentials of Radiologic Science. Baltimore, MD: Walters Kluwer Health/Lippincott Williams & Wilkins; 2012.  10. Introduction to Digital Radiography: The Role of Digital Radiography in Medical Imaging. Rochester, NY: Eastman Kodak Company; 2000. M1412 (Cat No. 183 6998).  11. Konstantinidis AC, Szafraniec MB, Speller RD, et al. The Dexela 2923 CMOS X-ray detector: a flat panel detector based on CMOS active pixel sensors for medical imaging applications. Nucl Instrum Methods Phys Res A. 2012;689:12–21.  12. Krupinski EA, Williams MB, Andriole K, et al. Digital radiography image quality: image processing and display. J Am Coll Radiol. 2007;4:389– 400.

 13. Lehnert T, Naguib NN, Korkusuz H, et al. Image-quality perception as a function of dose in digital radiography. AJR Am J Roentgenol. 2011;197:1399–1403.  14. American College of Radiology. Practice Guidelines and Technical Standards. Reston, VA: American College of Radiology; 2005:5–9.  15. Ranger NT. Assessment of detective quantum efficiency: intercomparison of a recently introduced international standard with prior methods. Radiology. 2007;243(3):785–795.  16. Samei E. Performance of digital radiographic detectors: quantification and assessment methods. In: Advances in Digital Radiography: RSNA Categorical Course in Diagnostic Radiology Physics; 2003:37–47. Oak Brook, IL  17. Seeram E. Digital Radiography: An Introduction. Clifton Park, NY: Delmar Cengage Learning; 2011.  18. Seibert JA. Digital radiography: the bottom line comparison of CR and DR technology. Appl Radiol. 2009;5:21–28.  19. Vaño E. Transition from screen-film to digital radiography: evolution of patient radiation doses at projection radiography. Radiology. 2007;243(2):461–466.  20. Williams M, Krupinski EA, Strauss KJ, et al. Digital radiography image quality: image acquisition. J Am Coll Radiol. 2007;4(6):371–388.

4 Positioning Techniques and Terminology ROBERT L. BARON, CASIMIR F. STRUGIELSKI, AND ROBERT A. CHRISTMAN Radiography has played, and continues to play, an integral role in assessing foot and ankle disorders. Initially, views must be selected that will best image the part in question (see Chapter 10, View Selection for the Radiographic Study). The objective is to select positioning techniques that will yield the most information while minimizing radiation exposure. A firm knowledge of normal radiographic anatomy is a prerequisite. A diversity of positioning techniques are available for foot and ankle radiography. The limited x-ray machine operator (LXMO) must clearly understand specific terminology used to describe positioning techniques before attempting to perform them. Unfortunately, discrepancy exists regarding use of the terms view, position, and projection. Even the word lateral can have two meanings. The following discussion of these terms, we hope, will clarify their application, especially as to the naming of positioning techniques. USE OF THE TERMS VIEWS, PROJECTIONS, AND POSITIONS Numerous techniques have been described for radiography of the foot and ankle. They range from non–weight bearing to weight bearing and have every possible aspect of the extremity positioned against the image receptor. Oblique foot techniques can be especially confounding; they may be performed non–weight bearing, with the plantomedial, plantolateral, dorsomedial, or dorsolateral aspect of the foot positioned against the image receptor (IR), or weight bearing, with the central x-ray beam directed at 45° toward either the dorsomedial or dorsolateral aspect of the foot. The result is confusion not only over to which positioning techniques to choose for a particular study but also over what to name them.

The Predicament Several factors contribute to the current disarray regarding terminology for positioning techniques and radiographs. They include whether positioning techniques are named as positions or projections and how people use the terms projection, position, and view. Positioning techniques have historically been named by two methods: (1) according to the path or projection of the x-ray beam through the body part and (2) by that surface of the body positioned closer to the IR. The former classically pertains to examination of the coronal anatomic plane (e.g., anteroposterior and posteroanterior projections). Position, in contrast, is used to describe oblique and lateral techniques.1,2 Initially, positional terms alone were sufficient to specify oblique and lateral techniques. However, as specialized positioning techniques evolved, projectional terms were being used in conjunction with, and even in place of, positional terms to describe oblique and lateral techniques. Use of the term “view” further confounds the predicament. The term “view” should pertain only to the final image or radiograph. It should not be used to describe a positioning technique or projection.3 Doctors usually attach the term view to whatever positioning technique was performed when naming the radiographic image; that is, the term view is used synonymously with projection and position.4 For example, an anteroposterior projection is an anteroposterior view. However, one publication states that the term view means exactly the opposite of projection.5 By this definition, the image produced by an anteroposterior projection should be called a posteroanterior view. Taken a step further, a view was named either by that part of the body closest to the IR or by where the x-ray beam exits.6 The authors who distinguish between the terms view and projection by definition, however, fail to address the issue in practice. Figures illustrating positioning techniques in those texts are named as projections, but the radiographic images accompanying the technical descriptions are either named as a projection or not named at all! The term view is omitted intentionally. Another positioning textbook does not define how the term view is applied except that it is reserved for discussing the

radiograph, not the technique.3 Discrepancies abound in the medical and technical literature regarding the names of positioning techniques and resultant images. An issue of Clinics in Podiatric Medicine and Surgery dedicated to radiology of the foot and ankle demonstrates this inconsistency. One article in this issue stresses routine use of the term projection for describing both technique and radiograph7; the projection is based on the body surface that the x-ray beam first enters. In contrast, another article consistently uses the term view when describing positioning techniques8; oblique positions are named by the surface nearest to the IR, yet the dorsoplantar view is named by the direction of the x-ray beam. This confusion appears to be rooted in the separation of a radiographic study into technical and diagnostic components. The technical component is threedimensional, and, therefore, appropriate terminology is needed to adequately describe the procedure. The radiograph, however, is only two-dimensional. It is not necessary to consider that the body part is being visualized from the film side, as suggested by Merrill.4 The radiograph looks the same no matter how the film is placed on the viewbox. In fact, anteroposterior and posteroanterior chest radiographs are customarily viewed as if the doctor were facing the patient (e.g., the patient’s right is on the viewer’s left).9 This practice has no regard for the patient’s position relative to the film. A final source of confusion pertains to usage of the term lateral. The term lateral can be defined in two ways. (1) It can, generally speaking, refer to any side of the body; in this case, both sides of the foot are considered lateral. Lateral positions of the foot simply correspond to side positions of the body relative to the IR. The term does not differentiate between true lateral and medial sides. Therefore, medial refers to the center or midline of the foot or ankle.10 (2) The second definition of lateral corresponds to true lateral, that is, that aspect of the foot farther from the body’s midline, the latter being located between both extremities and separating them into right and left.11 In this case, medial corresponds to that side of an extremity that is closer to the midline of the torso. In radiography, the designation “lateral position” is not specific and corresponds to the first definition. The modifier “lateromedial” or “mediolateral projection” must be added to specify the precise positioning

technique. The second definition (true medial and lateral) is applied when describing projections. The lateral foot and ankle positions were originally performed with the patient not bearing weight and with the true lateral aspect of the extremity positioned against the IR. More specific terminology was not necessary to further identify the technique performed. The weight-bearing lateral foot and ankle techniques were developed later. The true medial aspect of the foot or ankle was positioned against the IR, not the lateral. However, instead of naming it a lateral position/lateromedial projection, technicians named it by that aspect of the extremity closer to the x-ray tube (i.e., by the surface the xray beam enters first). The two techniques were differentiated as being either weight bearing or recumbent.12 The most recent editions of two radiographic positioning textbooks have made some changes (again!) to terms and tube head angles used for standard foot- and ankle-positioning techniques.5,13 These newer terms will be referenced in the descriptions of relevant positioning techniques as “other names.” The Solution One goal of this textbook is to standardize the terminology applied to positioning techniques and radiographic images of the foot and ankle. This can be easily accomplished without grossly changing the terminology or routines already being practiced. Definitions for the terms projection, position, view, and positioning technique are followed by specific applications.    Position: Pertains to that aspect of the body closest to the IR. It is used to name the oblique and lateral (not the true lateral but a side) positioning techniques. A directional term (projection) and other adjectives (e.g., weight bearing or non–weight bearing) should be used with the oblique and lateral positions to further define the positioning technique.    Projection: The direction that the x-ray beam travels through the body. This direction is described as being anteroposterior or posteroanterior, dorsoplantar or plantodorsal, or lateromedial or mediolateral. (The latter

terms refer to true lateral and medial.) The projection is used in addition to a position term to further describe a particular oblique or lateral position. The term projection is used to describe a positioning technique; it does not refer to the radiographic image.    View: Pertains to the radiographic image only. The terminology used to describe the positioning technique will simply be applied to the image, but the word view will replace the term position or projection. For example, the technique for a dorsoplantar projection produces a dorsoplantar view, the technique for a lateral position (lateromedial projection) produces a lateral (or lateromedial) view, and the technique for a medial oblique position produces a medial oblique view. View is not the opposite of projection. Additional terms should be included when appropriate, such as weight bearing or non– weight bearing.    Positioning Technique: The actual method of performing the study, including the position of the patient, tube head, and IR and the projection of the x-ray beam. Application As noted earlier, numerous positioning techniques are possible for imaging the foot and ankle. Each is discussed in detail in the following sections. The terminology will adhere to the definitions just noted. Some positioning techniques have been named after the authors who first described them. Their names are included with the description of the positioning technique. Positioning techniques that produce images of the ankle and foot in the coronal and transverse planes, respectively, should be named by projections alone. Ankle techniques are anteroposterior projections, and foot techniques are typically dorsoplantar projections. The descriptor weight bearing or non– weight bearing should precede the name. The radiographic image has the same name except the word view replaces the term projection. An example of a positioning technique and corresponding view is the weight-bearing dorsoplantar projection and weight-bearing dorsoplantar view. Descriptions of lateral positioning techniques, in contrast, require two designations: a position term accompanied by a projection term. The term

describing the projection of the x-ray beam follows the position term. The designation weight bearing or non–weight bearing precedes the term lateral position; for example, “weight-bearing lateral position (lateromedial projection).” The four lateral positioning techniques, including the weight-bearing mediolateral and lateromedial projections and the non–weight-bearing mediolateral and lateromedial projections, produce views that look similar to one another. The resultant images or views can all be named lateral views; the projection terms need not be included (especially if the full positioning technique has been previously described). Oblique positioning techniques may or may not require additional position and projection terms. Non–weight-bearing oblique positioning techniques should include a term designating that aspect of the extremity closest to the IR. An example is the non–weight-bearing medial oblique position. Weightbearing oblique positions (which are not advocated for the reasons listed in the section Positioning Considerations) do not need the position term plantar added to the title. The term weight bearing implies that the plantar surface of the foot is closest to the IR; including it would be redundant. However, a projection term is required to designate the direction of the x-ray beam through the extremity. The two weight-bearing oblique positioning techniques are either mediolateral or lateromedial oblique projections. Radiographic images are named the same except the term view replaces the words position or projection. An example of an oblique positioning technique and corresponding view is the non–weight-bearing medial oblique position and the non–weight-bearing medial oblique view. POSITIONING CONSIDERATIONS Weight-bearing oblique foot-positioning techniques with the tube head angled at 45° have been advocated by some doctors of podiatric medicine. One reason for this practice has been to standardize performance of the technique. Weight-bearing techniques alone, however, do not necessarily standardize the resultant view. It has been shown that foot positioning (supination and pronation) influences the radiographic positional relationships of the bones during weight bearing.14 Another attempt to

standardize positioning techniques is by positioning the extremity in its angle and base of gait.15 This technique is used so that biomechanical measurements of the foot are standardized and reproducible.16 Biomechanical measurements, however, are only performed on dorsoplantar and lateral radiographs, not on oblique views. Magnification (size distortion) and shape distortion of the image result from weight-bearing oblique positioning techniques. If true magnification is desired, a magniposer can be used to increase the object-to-image distance (OID); the geometric distortion is much less than that obtained by the weightbearing oblique study. Occasionally, distortion of the object may be desirable in an attempt to better visualize a particular pathology. If so, the distorted oblique projection should be performed adjunctively as a special technique. A non–weight-bearing oblique position, with the foot tilted 45° and the tube head directed perpendicular to the IR, should be performed initially. Non– weight-bearing oblique foot positions can be accurately performed with a radiolucent foam wedge positioning aid. The foot should be positioned perpendicular to the leg as the sole of the foot rests against the wedge. The only other time that the distorted weight-bearing oblique positioning technique should be performed is if the patient cannot position his or her extremity properly for the non–weight-bearing oblique study. On a final note, the non–weight-bearing oblique requires a lesser exposure technique and can be performed on half of a 10 × 12-in image receptor. The weight-bearing oblique may require the use of a separate image receptor for each extremity if performing a bilateral study. Positioning techniques for imaging the foot and ankle should be relatively simple to perform by both technician and patient. The type of x-ray unit being used, however, may already impose limitations. Many techniques can be performed with lower extremity–specific x-ray units. Orthoposer-mounted units offer convenience (the source-to-image distance [SID] is fixed) and safety (the patient can hold on to an attached railing, and the weight of the tube head holds the orthoposer unit steady); they are best for performing weight-bearing studies. However, it is difficult, if not impossible, to perform non–weight-bearing ankle studies with these particular units. A mobile x-ray unit, in contrast, can serve two purposes: (1) non–weight-bearing studies can be performed with the patient lying on an examination chair or table and (2)

weight-bearing studies can be performed on an orthoposer platform placed adjacent to the chair or table. Greater technical expertise, however, is necessary to position the tube head properly with a mobile unit. Extremitypositioning aids are useful for performing certain techniques, such as positioning blocks and the weight-bearing sesamoid axial projection. Positioning blocks or wedges are useful aids to position the foot or ankle for non–weight-bearing oblique techniques. They are made of a radiolucent material and are resilient, firm, durable, and washable. Positioning blocks will not slip when positioned properly under the patient. Wedges are available in a variety of shapes and sizes; a 45° triangular wedge accommodates most oblique positioning techniques. A device called the axial poser is valuable for performing the weight-bearing axial sesamoid-positioning technique. It is sold in pairs, for the left and right foot, and is made of radiolucent plastic. The front and back of the device are angled superiorly to elevate the digits and rearfoot. This positioning causes the metatarsal heads to be positioned inferiorly relative to the remainder of the foot. After obtaining standard views (or based on the clinical location of a finding), additional, nonstandard positioning techniques may be chosen to more clearly view an area of concern. For example, Osher et al.17 have shown the value of oblique forefoot axial modifications. (Other articles proposing specialized techniques are listed in the section Suggested Readings.) OVERVIEW OF POSITIONING TECHNIQUES AND TERMINIOLOGY The following is an outline of the positioning techniques that can be performed on the foot and ankle. They are categorized as foot-, toe-, sesamoid-, tarsal-, and ankle-positioning techniques. Each category is further subdivided into dorsoplantar (plantodorsal) or anteroposterior (posteroanterior) projections, oblique positions, and lateral positions, when applicable. I.

Foot-Positioning Techniques

     A. Dorsoplantar (DP) projections         1. Weight bearing         2. Non–weight bearing            a. Foot flat            b. Forefoot angled 15°      B. Oblique positions         1. Non–weight bearing            a. Medial            b. Lateral         2. Weight bearing            a. Lateromedial oblique projection            b. Mediolateral oblique projection      C. Lateral positions (may be performed weight bearing or non–weight bearing)         1. Lateromedial projection         2. Mediolateral projection II.

Toe-Positioning Techniques

     A. Dorsoplantar projection (weight bearing or non–weight bearing)      B. Oblique positions         1. Non–weight bearing            a. Medial oblique

           b. Lateral oblique         2. Weight bearing            a. Lateromedial oblique projection            b. Mediolateral oblique projection      C. Lateral positions (may be performed weight bearing or non–weight bearing)         1. Lateromedial projection         2. Mediolateral projection III.

Sesamoid-Positioning Techniques

     A. Posteroanterior sesamoid axial projections         1. Weight-bearing axialposer         2. Lewis method      B. Anteroposterior sesamoid axial projection (Holly method)      C. Lateromedial tangential projection (Causton method) IV. Tarsal-Positioning Techniques      A. Dorsoplantar calcaneal axial projection      B. Plantodorsal calcaneal axial projection      C. Harris–Beath      D. Broden      E. Isherwood V.

Ankle-Positioning Techniques (may be performed weight bearing or

non–weight bearing)      A. Anteroposterior (AP) projection      B. Mortise position      C. Oblique positions         1. Medial (or internal) oblique         2. Lateral (or external) oblique      D. Lateral positions         1. Lateromedial projection         2. Mediolateral projection Terminology  1. Angle of gait: The angle formed between the feet and line of progression while walking (approximately 10°–15° of abduction in the normal individual).  2. Base of gait: The distance between both medial malleoli while walking (approximately 2 in).  3. Midline of foot: The imaginary line that enters the center of the heel and exits through the second digit, thereby dividing the foot into two halves.  4. Central ray (CR): The most direct beam of radiation from the tube.  5. Dorsum (dorsal): In reference to the top of the foot.  6. Plantar: Bottom (sole) of foot.  7. Lateral (two definitions):     a.  Away from the torso’s midline (outer side of the extremity).

    b.  Away from the midline of the extremity (inner or outer side of the extremity).  8. Medial: Inner side of the extremity (toward the torso’s midline).  9. Extension: The process of straightening the joint. 10. Flexion: The process of bending the joint in an angle. 11. Supine: Lying face up. 12. Prone: Lying face down. 13. Specific body position: Describes the placement of the foot or ankle relative to the image receptor. Non–weight-bearing oblique positions are named by that aspect of the extremity nearest the image receptor. 14. Positioning: Manner in which the tube head, IR, patient, and central ray are placed to obtain a radiographic image of a particular body part. 15. Projection: The direction that the primary x-ray beam travels through the body part. Used to describe anteroposterior (dorsoplantar), lateral, and weight-bearing foot oblique positioning techniques. For example, the x-ray beam enters the dorsal aspect of the foot and exits the plantar aspect in the dorsoplantar projection. 16. Tube head angulation: The number of degrees the tube head is set from vertical (0°). 17. View: This term refers to the image in the radiograph. It is not a positioning term.3 18. Oblique: The condition when the plane of a body part is neither perpendicular nor parallel to the IR. 19. Axial: The long axis of a structure or body part. 20. Image receptor (IR): A device that receives the remnant x-rays after passing through the paient and stores the latent image. In plain film

radiography it houses the film and intensifying screens. Also referred to as the imaging plate (IP). 21. Orthoposer: Platform used to perform weight-bearing studies. GENERAL CONSIDERATIONS Most views of the foot and ankle can be obtained with either a weightbearing or a non–weight-bearing positioning technique. The patient being considered for weight-bearing positioning techniques should be ambulatory, able to walk without assistance. If not, partial weight bearing or non–weight bearing may be preferable. The added weight of a lead apron or shield to a patient whose mobility is questionable poses a considerable risk for injury if an unsteady patient is positioned on an orthoposer for a weight-bearing study. Partial or non–weight bearing should be considered for the patient who is not surefooted or uses an ambulation assistance device (e.g., walker, cane, or crutches). Patients who have recently had a cast put on or who have undergone a surgical procedure may not be stable enough to allow full weight bearing. Non–weight-bearing positioning techniques are necessary for individuals who are confined to a wheelchair, have recently undergone an extensive surgical procedure, or have acute pain (e.g., secondary to recent trauma). Positioning techniques can be modified as needed. This is especially true for oblique positions if you are trying to view a specific anatomic landmark or location. If a lesion marker is placed on the skin, only dorsoplantar and lateral positioning techniques should be performed. Marker position cannot be accurately assessed with oblique views. Furthermore, the tube head must be positioned vertically (0°) for the dorsoplantar projection. Any other tube head angulation will alter the position of the marker relative to the internal foot structures. This also applies to the location of radiopaque foreign objects. PREPARATION FOR THE RADIOGRAPHIC STUDY Preparation of the radiography room expedites patient turnaround and minimizes the chance for mistakes. Several aspects of the study should be

considered and performed before the patient is brought into the room. Initially, the desired views must be selected; positioning techniques are chosen that will result in the desired images. The LXMO can then gather all applicable items pertaining to the study: image receptor, positioning aids, and identification markers, for example. Standard technical factors (mA, kVp, time, and SID) should be posted by each unit’s control panel for all positioning techniques. Any modifications should be determined in advance. Tube head and image receptor placement, identification markers, and control panel adjustments can be ready for the first positioning technique before the patient enters the radiography room. Once the patient is in the radiography room, he or she must be given precise instructions as to what to do and what is expected for each of the positioning techniques. The patient must clearly understand these instructions so that the necessity for repeat studies is minimized. Image blur caused by patient movement is probably the most common reason for repeating any particular study. Clear instructions must also be given to the parent or guardian who remains in the room and is helping to maintain an infant’s or young child’s position. Occasionally, a patient, particularly a trauma or geriatric patient, may express discomfort during positioning of the extremity. If it becomes unreasonable to perform a certain positioning technique, an alternative position or projection or even a non–weight-bearing technique should be considered. A final check should be made by the LXMO to make sure that all technical parameters are correct: patient positioning, identification markers, and technical factors. This should be done for each positioning technique. Error regarding any of these parameters can result in having to repeat a study. Lead aprons must be provided for and used by each patient who is being examined. This also applies to the infant or child patient. An apron and lead gloves must be provided to any individual who remains in the room during the exposure to assist in positioning. ANGLE AND BASE OF GAIT POSITIONING Weight-bearing dorsoplantar and lateral foot radiographs have been advocated for decades, as early as 1943.18 The weight-bearing attitude is felt

to create an anatomic image that is most feasible for assessing normal versus pathological biomechanical conditions under the stresses and strain of body weight.19 Shortly after Sgarlato15 described the kinesiologic and structural relationships of the angle of gait, Hlavac14 demonstrated that foot position has a profound influence on the radiographic relationships and forms of osseous structures and angular biomechanical measurements. Weight-bearing foot radiography with the foot positioned in angle and base of gait is considered standard technique in podiatric practice for dorsoplantar and lateral foot radiographs and has since been advocated to the orthopedic community.20 The purpose of using angle and base of gait is to standardize the weightbearing radiographic technique in order to visualize positional relationships of the foot bones in a midstance attitude.16 The LXMO must first study the relationship of the foot to the line of progression in the patient’s gait cycle. The degree of abduction or adduction of the foot from the midline (line of progression) is known as the angle of gait (Figure 4-1). The separation or distance between both heels during the gait cycle is referred to as the base of gait (Figure 4-2). Because both heels are not on the ground at the same time during gait (as they are in foot radiography), base of gait is an attempt to maintain a normal midstance relationship of the tibia to the ground.16 The LXMO must then position the patient’s foot relative to the IR and central beam such that angle and base of gait is reproduced. When using an orthoposer, the image receptor and tube are first positioned for the desired technique (dorsoplantar or lateral) and the patient’s stance is adjusted to place the foot in proper position. Figures 4-3 and 4-4 show examples of dorsoplantar and lateral positioning techniques, respectively, in angle and base of gait.

FIGURE 4-1. Angle of gait. FOOT-POSITIONING TECHNIQUES Note: Weight-bearing techniques can also be performed as partial weight bearing. The only difference in positioning is that the patient places most of the weight on the opposite foot while partially bearing weight on the foot being studied. If there is any question whether or not a patient can safely step onto or stand on the orthoposer, perform the study non–weight bearing, with the patient seated.

FIGURE 4-2. Base of gait.

FIGURE 4-3. Weight-bearing dorsoplantar positioning technique in angle and base of gait. The patient’s right foot is positioned on the right half of the image receptor. The left foot is then positioned such that both feet are in angle and base of gait. This usually means that the opposite foot will partly be off the edge of the image receptor. Ideally, another block of material of the same thickness of the image receptor should be placed against the side of the image receptor; this ensures patient comfort and full weight bearing without any compensation by the patient. This example demonstrates the use of a soft sheet of foam. Dorsoplantar Projections  1. Weight-bearing dorsoplantar projection (Figure 4-5)    Other names: AP weight-bearing projection,13 AP axial projection,

weight-bearing method.5    IR placement: Flat (horizontal) on orthoposer; long side of image receptor parallel to long side of orthoposer.    Tube head angulation: 15° from vertical, directed posteriorly.    Foot position: Flat on image receptor with foot’s midline parallel to long side of image receptor. The opposite foot is placed in angle and base of gait.    Central ray direction: Second metatarsocuneiform joint (or base of 3rd metatarsal).5    Other considerations: It has been advocated that dorsoplantar projections be performed with the tube head angled 5°, 10°, or even 20° posteriorly, depending on metatarsal declination. On the average, metatarsal declination is 15° relative to the plane of support. A 15° tube head angulation should be performed initially for dorsoplantar projections; tube head angulation at 5°, 10°, or 20° should be performed only as an adjunctive study, if needed. A standardized tube head angulation eliminates another variable when interpreting the resultant view. If the tube head is angled at different degrees for each patient, it will be difficult to appreciate the form and position of foot bones relative to normal expectations; in other words, is a bone’s shape, size, or position viewed in the radiograph distorted by the nonstandard tube head angulation?

FIGURE 4-4. Weight-bearing lateral positioning technique in angle and base of gait. The patient’s right foot is positioned against the image receptor. The left foot is then positioned such that both feet are in angle and base of gait.

FIGURE 4-5. Dorsoplantar projection. A: Weight-bearing dorsoplantar projection. B: Dorsoplantar view.  2. Non–weight-bearing dorsoplantar projection    Other names: AP projection (non–weight bearing).5    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 15° from vertical, directed posteriorly. Bontrager and Lampignano13 recommend 10° as the standard angulation, but 15° for a high arch, 5° for a low arch, and 0° for evaluating a foreign body.

   Foot position: Flat on image receptor with foot’s midline parallel to long side of image receptor. The knee must be flexed so that the foot fully purchases the image receptor. A sandbag should be placed along the front edge of the image receptor to prevent it from sliding.    Central ray direction: Second metatarsocuneiform joint.    Other considerations:    (a) To perform this technique with an orthoposer, the patient must be seated, facing the orthoposer. The long axis of the image receptor is perpendicular to the long side of the orthoposer. The patient places the sole of the foot in contact with the image receptor. The tube head is directed posteriorly toward the second metatarsocuneiform joint at 15° from vertical.    (b) A variation of this technique angles the tube head 0° (vertical). Either the foot can remain flat on the image receptor or a 15° radiolucent wedge can be placed beneath the forefoot so that the metatarsals are parallel to the IR. Oblique Positions  1. Non–weight-bearing medial oblique position (Figure 4-6)    Other names: AP oblique projection, medial rotation.5,13    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: Patient lies supine and flexes knee. The sole of the foot is placed flat on the image receptor. The knee should be flexed such that the lower leg is nearly perpendicular to the foot. The foot’s midline should be parallel to the long side of the image receptor. The knee is then turned inward so that the medial side of the foot is closer to the image receptor and the sole forms a 45° angulation with the image receptor. A 45° radiolucent positioning wedge can be placed under the sole of the foot to aid in positioning. (A variation of this technique positions the foot so that it is angulated 30°.)

   Central ray direction: Third metatarsocuneiform joint.    Other considerations: To perform this technique with an orthoposer, the patient must be seated, facing the orthoposer. The long side of the image receptor is perpendicular to the long side of the orthoposer. The tube head, foot, and central ray are positioned or directed as just noted.  2. Non–weight-bearing lateral oblique position (Figure 4-7)    Other names: AP oblique projection, lateral rotation.13    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: Patient lies supine and flexes knee. The sole of the foot is placed flat on the image receptor. The knee should be flexed such that the leg is nearly perpendicular to the foot. The foot’s midline should be parallel to the long side of the image receptor. The knee is then turned out so that the lateral side of the foot is closer to the image receptor and the sole forms a 45° angulation with the image receptor. A 45° radiolucent positioning wedge can be placed under the sole of the foot to aid in positioning. (A variation of this technique positions the foot so that it is angulated 30°.)    Central ray direction: First metatarsocuneiform joint.    Other considerations: To perform this technique with an orthoposer, the patient must be seated, facing the orthoposer. The long axis of the image receptor is perpendicular to the long side of the orthoposer. The tube head, foot, and central ray are positioned or directed as noted earlier.

FIGURE 4-6. Non–weight-bearing medial oblique position. A: Non–weightbearing medial oblique position. B: Non–weight-bearing medial oblique view.

FIGURE 4-7. Non–weight-bearing lateral oblique position. A: Non–weightbearing lateral oblique position. B: Non–weight-bearing lateral oblique view.  3. Weight-bearing lateromedial oblique projection    IR placement: Flat (horizontal) on orthoposer; long axis of image receptor parallel to front end of orthoposer.

   Tube head angulation: 45° from vertical.    Foot position: Lateral aspect of foot faces the tube head. The patient stands on the image receptor so that the foot’s midline is parallel to the long side of the image receptor. The lateral aspect of the foot should be positioned along the long side of the image receptor if dividing it into halves for two views. The 45° angulation of the tube head results in image distortion (unequal magnification), and the foot, therefore, takes up a larger portion of the IR than it does with the non–weight-bearing oblique position.    Central ray direction: Fourth metatarsocuboid joint.  4. Weight-bearing mediolateral oblique projection    IR placement: Flat (horizontal) on orthoposer; long axis of image receptor parallel to front end of orthoposer.    Tube head angulation: 45° from vertical.    Foot position: Medial aspect of foot faces the tube head. The patient places the foot on the image receptor so that its midline is parallel to the long side of the image receptor. The medial aspect of the foot should be positioned along the long side of the image receptor if dividing into halves for two views. The opposite foot must be positioned behind and to the side of the foot under study; the patient is instructed to hold onto a support rail to limit movement caused by imbalance and as a safety precaution. The 45° tube head angulation results in image distortion, and the foot, therefore, takes up a larger portion of the IR than it does with the non–weight-bearing oblique position.    Central ray direction: First metatarsocuneiform joint. Lateral Positions (Figure 4-8)  1. Weight-bearing lateromedial projection    Other names: Lateral weight-bearing projection (lateromedial; “longitudinal arch”).13

   IR placement: Upright (vertical) in slot or well of orthoposer.    Tube head angulation: 90° from vertical.    Foot position: Patient stands on the orthoposer with the medial aspect of the forefoot positioned against the image receptor and the midline of the foot parallel to the image receptor. The tube head must be precisely positioned at 90°. If not, the plantar aspect of the foot may become superimposed by the orthoposer and not visible. (If this occurs, have the patient stand on ¼- or ½in felt blocks; the opposite foot should also be placed on top of felt.) The opposite foot is placed in angle and base of gait behind the image receptor.    Central ray direction: Lateral cuneiform/cuboid.  2. Non–weight-bearing lateromedial projection    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: For a right-foot study, the patient lies on the left body side and vice versa. The medial aspect of the foot is positioned against the image receptor. The foot must be nearly perpendicular to the lower leg.    Central ray direction: Lateral cuneiform/cuboid.    Other considerations: To perform this technique with an orthoposer, the patient must be seated, facing the orthoposer. The image receptor stands vertical in the well of the orthoposer. The medial aspect of the forefoot is positioned against the image receptor with the midline of the foot parallel to the image receptor. The tube head must be precisely positioned at 90°. If not, the plantar aspect of the foot may become superimposed by the orthoposer and not visible. (If this occurs, have the patient stand on ¼- or ½-in felt blocks; the opposite foot should also be placed on top of felt.) The opposite foot is placed in angle and base of gait behind the image receptor.  3. Non–weight-bearing mediolateral projection    IR placement: Flat (horizontal) on radiographic table.

   Tube head angulation: 0° (vertical).    Foot position: For a right-foot study, the patient lies on the right body side and for the left foot, on the left body side. The lateral aspect of the foot is positioned against the image receptor. The foot should be nearly perpendicular to the lower leg and the sole of the foot should be perpendicular to the image receptor.    Central ray direction: Medial cuneiform.    Other considerations: To perform this technique with an orthoposer, the patient must be seated, facing the orthoposer. The image receptor lies flat on the orthoposer. The lateral aspect of the foot is positioned against the image receptor so that the sole of the foot is perpendicular to the image receptor. The tube head angulation and direction of the central ray are as noted earlier.

FIGURE 4-8. Lateral position. A: Weight-bearing lateromedial projection. B: Lateral view. C: Non–weight-bearing mediolateral projection. TOE-POSITIONING TECHNIQUES Dorsoplantar Projections Weight-bearing and non–weight-bearing dorsoplantar projections These projections of the toes are performed in the same manner as the weight-bearing and non–weight-bearing dorsoplantar projections of the foot with the following exceptions:    The tube head angulation is 0° (vertical).    The central ray is directed toward the second digit proximal phalanx if

viewing all digits or the digit in question if only one digit. Oblique Positions Non–weight-bearing medial and lateral oblique positions and weight-bearing lateromedial and mediolateral oblique projections The oblique positioning techniques of the toes are performed in the same manner as the oblique positions of the foot with the following exception:    The central ray is directed toward the third digit proximal phalanx if viewing all digits or the digit in question if only one digit. Lateral Positions The following descriptions pertain to standard IR sizes (10 × 12 in or 8 × 10 in). However, an option is to use dental film.21 The unexposed dental film is placed vertically behind the toe in question. Occasionally, nonstandard positioning of the foot (e.g., turned slightly inward or outward) may enhance visibility of the toe in question in the radiograph.  1. Weight-bearing lateromedial projection (Figure 4-9)    IR placement: Upright (vertical) in slot or well of orthoposer.    Tube head angulation: 90° from vertical.    Foot position: Patient stands on the orthoposer with the medial aspect of the hallux positioned against the image receptor. Place a roll of gauze beneath the digit of concern to raise it above the remaining digits. (If there is a history of trauma, place the roll of gauze beneath the remaining toes to raise them above the digit under study.) An 8-ft length of tube gauze can also be used to isolate a digit. The toe is positioned at the midpoint of the tube gauze; the patient is instructed to pull the affected or adjacent digit(s) up or down (this depends on which toe is to be isolated and whether or not there are digital deformities) with the two ends of tube gauze.    Central ray direction: Toe in question.

 2. Non–weight-bearing lateromedial projection    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: For a right-foot study, the patient lies on the left body side and vice versa. The medial aspect of the hallux is positioned against the image receptor. The affected toe can be further isolated with tube gauze as described for the weight-bearing lateromedial projection.

FIGURE 4-9. Lateral toe positions. A: Weight-bearing lateromedial projection of the hallux (hallux elevated). B: Lateral view of the hallux. C: Weight-bearing lateromedial projection of the hallux (lesser toes elevated).    Central ray direction: Toe in question.    Other considerations: To perform this technique with an orthoposer, the patient must be seated, facing the orthoposer. The image receptor stands vertical in the well of the orthoposer. The medial aspect of the hallux is positioned against the image receptor. The central ray is directed as noted earlier.  3. Non–weight-bearing mediolateral projection    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: For a right-foot study, the patient lies on the right body side and for the left foot, on the left body side. The lateral aspect of the fifth toe is positioned next to the image receptor. The affected toe can be isolated with tube gauze as described for the weight-bearing lateromedial projection.    Central ray direction: Toe in question.    Other considerations: To perform this technique with an orthoposer, the patient must be seated, facing the orthoposer. The image receptor lies flat (horizontal) on the orthoposer. The lateral aspect of the foot is positioned against the image receptor so that the sole of the foot is perpendicular to the image receptor. The tube head angulation and direction of the central ray are as noted earlier. SESAMOID-POSITIONING TECHNIQUES Posteroanterior Sesamoid Axial Projections Note: The lead apron should be reversed so that it is placed over the patient’s back.

 1. Weight-bearing axialposer method22,23 (Figure 4-10)    IR placement: Upright (vertical) in slot or well of orthoposer.    Tube head angulation: 90° from vertical.    Foot position: The front of the axialposer positioning device is placed against the image receptor so that the toes will be closest to the IR and the heel farthest. The patient stands on the axialposer (heel and toes are elevated so that the metatarsophalangeal joints are the most inferiorly positioned anatomic structures). (The tube head is behind the patient.)    Central ray direction: Center of back of axialposer device.    Other considerations: If an axialposer is not available, the following variation can be performed. IR placement and tube head angulation are as just described. A ¼-in felt pad is placed under the ball of the foot. The digits (toes) are placed against the image receptor, and dorsiflexion is maintained. The patient then elevates the heel. The central ray is aimed at the plantar surface of the metatarsal head, usually at the third metatarsal head surface. The lead apron is worn in reverse. If the patient has difficulty in maintaining the dorsiflexion of the digits against the image receptor, a 3-in roll of gauze can be placed beneath the toes to provide support.

FIGURE 4-10. Weight-bearing sesamoid axial projections. A: Weightbearing axialposer method (precut Styrofoam device). B: Weight-bearing axialposer method (adjustable axialposer device). C: Weight-bearing sesamoid axial projection without axialposer device. D: Weight-bearing sesamoid axial view.  2. Lewis method24 (Figure 4-11A)    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: Patient lies prone on the radiographic table. The knee should rest on a foam cushion. The toes are forcibly dorsiflexed as they are positioned against the IR.    Central ray direction: Ball of foot.

Anteroposterior Sesamoid Axial Projection (Holly Method25) (Figure 4-11B)    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: Patient lies supine on the radiographic table. The heel is placed against the IR, and the foot is upright and slightly plantarflexed. The toes are forcibly dorsiflexed by the patient with a long strip of tube gauze placed around the toes.    Central ray direction: Ball of foot. Lateromedial Tangential Projection (Causton Method26) (Figure 4-11C)    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 40° from vertical, directed posteriorly.    Foot position: Patient lies on the radiographic table on unaffected body side with knees flexed. The medial aspect of the affected foot is placed against the IR; the sole is perpendicular to the image receptor.    Central ray direction: Sesamoids. TARSAL-POSITIONING TECHNIQUES Dorsoplantar Calcaneal Axial Projection (Figure 4-12A,B)    Note: The lead apron should be reversed so that it is placed over the patient’s back.    IR placement: Flat (horizontal) on orthoposer.    Tube head angulation: 25° from vertical, directed anteriorly.    Foot position: Patient stands on the image receptor. (The tube head is behind the patient.) The heel is placed so that its posterior aspect is about 1 to 1.5 in away from the edge of the image receptor closest to the x-ray tube. The

patient can be instructed to flex the knees slightly into a “ski jump” configuration so that the foot is dorsiflexed at the ankle joint.    Central ray direction: Posterior aspect of foot between the Achilles tendon insertion and the ankle joint. Other considerations:  1. One variation of this positioning technique is known as the posterior tangential projection.27 The IR placement, foot position, and central ray position are the same as just described except that the patient stands erect (no additional dorsiflexion is needed at the ankle joint). The tube head is positioned at 45° from vertical and directed toward the midpoint between the inferior aspects of the malleoli. The resultant view shows the middle and posterior articular facets in addition to the calcaneal body.  2. Another variation is known as the calcaneal apophysis or superoinferior calcaneus projection.28 It is used to assess the posterior calcaneal surface (Achilles tendon enthesis) in the adult and the apophysis in the child. IR placement and foot position are the same as described earlier except that the patient is instructed to place both hands on the support rails and lean forward while the heels still purchase the image receptor. The heels are parallel to each other on the IR and the lower limbs are bent slightly forward. The tube head angulation is 0° (vertical). Some modification of tube head angulation may be necessary to adequately visualize the Achilles tendon enthesis and to see different aspects of the apophysis. The central ray is directed toward the posterior aspect of the heel.

FIGURE 4-11. Non–weight-bearing sesamoid axial projections (see text). A: Lewis method. B: Holly method. C: Causton method.

FIGURE 4-12. A: Dorsoplantar calcaneal axial projection. B: Calcaneal axial view. C: Plantodorsal calcaneal axial projection. D: Calcaneal axial (Harris–Beath) view (45°). The middle (open arrow) and posterior (solid arrow) talocalcaneal joints are visible. Plantodorsal Calcaneal Axial Projection (Figure 4-12C)    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 40° to 45° from vertical, directed posteriorly.    Foot position: Patient lies supine on the radiographic table. The back (posterior aspect) of the heel is placed against the image receptor, and the toes point upward. The lateral and plantar aspects of the foot should be perpendicular to the image receptor. The patient then forcibly dorsiflexes the forefoot with a long strip of tube gauze placed around the ball of the foot and grasped by both hands at its ends.    Central ray direction: Center of heel. Harris–Beath Calcaneal Axial Projection Note: The lead apron should be reversed so that it is placed over the patient’s back. The resultant image is also known as the “coalition position”13 (Figure 412D). Lilienfeld29 first described this for assessment of talocalcaneal coalition. It has been further described and used by Harris and Beath,30 Coventry,31 and Vaughan and Segal.32 The positioning is similar to the dorsoplantar calcaneal axial projection except for the tube angulation.    IR placement: Flat (horizontal) on orthoposer.    Tube head angulation: Depends on the angle that the posterior subtalar facet forms with the weight-bearing plantar surface. This can be measured on a standard weight-bearing lateral projection. The usual tube angle is at 40°, but can vary anywhere from 35° to 45°.

   Foot position: Patient stands on the image receptor. (The tube head is behind the patient.) The heel is placed so that the posterior aspect of the heel is about 1 to 1.5 in away from the edge of the image receptor closest to the xray tube. The patient is instructed to flex the knees slightly into a “ski jump” configuration.    Central ray direction: Center of heel. Broden Method33 Multiple positioning techniques are performed to assess different aspects of the posterior subtalar joint:  1. Medial oblique positions (Figure 4-13)    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: Four positioning techniques are performed at 10°, 20°, 30°, and 40° from vertical, directed posteriorly and cephalad.    Foot position: Patient lies supine on the radiographic table. The back (posterior aspect) of the heel is placed against the image receptor, and the toes point upward. (The posterior aspect of the heel is placed about 1 to 1.5 in away from the edge of the image receptor closest to the x-ray tube.) A 45° positioning wedge is placed along the medial aspect of the foot and ankle. The leg and foot are turned inward and rest on the wedge. The patient dorsiflexes the foot so that it is perpendicular to the lower leg. This can be maintained with a long strip of tube gauze placed around the ball of the foot and grasped by the patient with both hands at its ends.    Central ray direction: Between the fibular malleolus and fifth metatarsal base.  2. Lateral oblique positions (Figure 4-14)    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: Two positioning techniques are performed at 15°

and 18° from vertical, directed posteriorly and cephalad.    Foot position: Patient lies supine on the radiographic table. The back (posterior aspect) of the heel is placed against the image receptor, and the toes point upward. (The posterior aspect of the heel is placed about 1 to 1.5 in away from the edge of the image receptor closest to the x-ray tube.) A 45° positioning wedge is placed along the lateral aspect of the foot and ankle. The leg and foot are turned outward and rest on the wedge. The patient dorsiflexes the foot so that it is perpendicular to the lower leg. This can be maintained with a long strip of tube gauze placed around the ball of the foot and grasped by the patient with both hands at its ends.    Central ray direction: Between the tibial malleolus and navicular tuberosity. Isherwood Method34,35 (Figure 4-15) Three positioning techniques are performed to assess the anterior, middle, and posterior subtalar joints:  1. Medial oblique position (to assess the anterior articulation)    Note: This positioning technique is the same as the non–weight-bearing medial oblique foot position except for the central ray direction.    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: Patient lies on the unaffected side and flexes knee. The knee of the affected extremity sits on the table and the medial side of the foot lies against the image receptor; the foot is held perpendicular to the lower leg. A 45° positioning wedge is placed beside the sole of the foot; the knee is lifted off the table so that the medial side of the foot stays against the image receptor and the sole of the foot rests against the wedge, forming a 45° angulation with the image receptor.    Central ray direction: Between the fibular malleolus and cuboid.

 2. Medial oblique axial position (to assess the middle articulation)    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 10° from vertical, directed posteriorly and superiorly.    Foot position: Patient sits supine on the radiographic table. The back (posterior aspect) of the heel is placed against the image receptor and the toes point upward. The foot is held perpendicular to the lower leg. A 30° positioning wedge is placed beside the medial aspect of the foot. The knee is slightly flexed and the extremity turned internally so that the back of the heel remains against the image receptor and the medial side of the foot and ankle rests against the wedge, forming a 30° angle with the image receptor. A pillow or sandbag should be placed under the knee for support. The patient is instructed to dorsiflex and invert the foot. This can be maintained with a long strip of tube gauze placed around the ball of the foot and grasped by the patient with both hands at its ends.    Central ray direction: Between the fibular malleolus and cuboid.  3. Lateral oblique axial position (to assess the posterior articulation)    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 10° from vertical, directed posteriorly and superiorly.    Foot position: Patient sits supine on the radiographic table. The back (posterior aspect) of the heel is placed against the image receptor, and the toes point upward. The foot is held perpendicular to the lower leg. A 30° positioning wedge is placed beside the lateral aspect of the foot. The knee is slightly flexed and the extremity turned externally so that the back of the heel remains against the image receptor and the lateral side of the foot and ankle rests against the wedge, forming a 30° angulation with the image receptor. A pillow or sandbag should be placed under the knee for support. The patient is instructed to dorsiflex and evert the foot. This can be maintained with a long strip of tube gauze placed around the ball of the foot and grasped by the

patient with both hands at its ends.    Central ray direction: Between the tibial malleolus and navicular tuberosity.

FIGURE 4-13. Broden method: medial oblique position. A: Medial oblique position. B: Medial oblique view at 10°. C: Medial oblique view at 20°. D: Medial oblique view at 30°. E: Medial oblique view at 40°. ANKLE-POSITIONING TECHNIQUES Anteroposterior Projections

 1. Weight-bearing anteroposterior projection (Figure 4-16)    IR placement: Upright (vertical) in slot of orthoposer; long axis of image receptor perpendicular to orthoposer surface.    Tube head angulation: 90° from vertical.    Foot position: Back of heel is placed against the image receptor. The foot’s midline axis is perpendicular to the image receptor.    Central ray direction: Center of ankle joint.

FIGURE 4-14. Broden method: lateral oblique position. A: Lateral oblique position. B: Lateral oblique view at 15°. C: Lateral oblique view at 18°.

FIGURE 4-15. Isherwood method. A: Medial oblique position. B: Medial oblique view. C: Medial oblique axial position. D: Medial oblique axial view. The middle (arrows) and posterior talocalcaneal joints are visible. E: Lateral oblique axial position. F: Lateral oblique axial view. The posterior talocalcaneal joint is visible (arrow) and viewed in profile.

FIGURE 4-16. Anteroposterior projection. A: Weight-bearing anteroposterior projection. B: Anteroposterior view.  2. Non–weight-bearing anteroposterior projection    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: Patient lies supine or sits on radiographic table. The back of the heel is placed against the image receptor, and the toes point upward. The knee is nearly straight; a small pillow or sandbag can be placed under the knee for patient comfort. The foot’s midline axis is perpendicular to the image receptor. The foot should be held nearly perpendicular to the leg; a sandbag can be propped against the sole of the foot to aid in positioning.    Central ray direction: Center of ankle joint. Mortise Position  1. Weight-bearing mortise position (Figure 4-17)    IR placement: Upright (vertical) in slot of orthoposer; long axis of image receptor perpendicular to orthoposer surface.

   Tube head angulation: 90° from vertical.    Foot position: Back of heel is placed against the image receptor. The ankle joint axis (an imaginary line running through both malleoli) is aligned so that it lies parallel to the image receptor. This is best achieved by placing an index finger on each malleolus and visualizing the axis through the ankle. A slight internal rotation of the extremity is required, anywhere from 5° to 15°. After positioning the extremity, recheck the malleolar axis; excessive pronation causes exaggerated internal rotation of the talus and lower leg on weight bearing, resulting in an oblique view of the ankle.    Central ray direction: Center of ankle joint.

FIGURE 4-17. Mortise position. A: Weight-bearing mortise position. B: Mortise view.  2. Non–weight-bearing mortise position    Other names: AP mortise projection 15° to 20° medial rotation.13    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: Patient lies supine or sits on radiographic table. The back of the heel is placed against the image receptor. The knee is nearly straight; a small pillow or sandbag can be placed under the knee for patient comfort. The ankle joint axis (an imaginary line running through both malleoli) is aligned so that it lies parallel to the image receptor, or rotate the limb until the medial and lateral malleoli are equidistant from the image receptor.36 This is best achieved by placing an index finger on each malleolus and visualizing the axis through the ankle. A slight internal rotation of the extremity is required, anywhere from 5° to 15°. The foot should be held nearly perpendicular to the leg; a sandbag can be propped against the sole of the foot to aid in positioning.    Central ray direction: Center of ankle joint.

Oblique Positions  1. Weight-bearing medial (internal) oblique position (Figure 4-18)    IR placement: Upright (vertical) in slot of orthoposer; long axis of image receptor is perpendicular to orthoposer surface.    Tube head angulation: 90° from vertical.    Foot position: Back of heel is placed against the image receptor. The lower extremity (foot and leg) is internally rotated 45° so that the medial surface of the extremity faces the image receptor. A 45° positioning wedge can be used to assist in positioning. (A variation of this technique positions the foot and leg so that they are internally rotated 30°.)    Central ray direction: Center of ankle.  2. Non–weight-bearing medial (internal) oblique position    Other names: AP oblique projection 45° medial rotation.13    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: Patient lies supine or sits on radiographic table. The back of the heel is placed against the image receptor. The knee is nearly straight; a small pillow or sandbag can be placed under the knee for patient comfort. A 45° positioning wedge is placed along the medial aspect of the foot and ankle. The lower extremity (foot and leg) is internally rotated 45° so that the medial surface rests against the wedge. The foot should be held nearly perpendicular to the leg; a sandbag can be propped against the sole of the foot to aid in positioning. (A variation of this technique positions the foot and leg so that they are internally rotated 30°.)    Central ray direction: Center of ankle.  3. Weight-bearing lateral (external) oblique position (Figure 4-19)

   IR placement: Upright (vertical) in slot of orthoposer; long axis of image receptor is perpendicular to orthoposer surface.    Tube head angulation: 90° from vertical.    Foot position: Back of heel is placed against the image receptor. The lower extremity (foot and leg) is externally rotated 45° so that the lateral surface of the extremity faces the image receptor. A 45° positioning wedge can be used to assist in positioning. (A variation of this technique positions the foot and leg so that they are externally rotated 30°.)    Central ray direction: Center of ankle.

FIGURE 4-18. Medial (internal) oblique position. A: Weight-bearing internal oblique position. B: Internal oblique view.

FIGURE 4-19. Lateral (external) oblique position. A: Weight-bearing external oblique position. B: External oblique view.  4. Non–weight-bearing lateral (external) oblique position    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: Patient lies supine or sits on radiographic table. The back of the heel is placed against the image receptor. The knee is nearly straight; a small pillow or sandbag can be placed under the knee for patient comfort. A 45° positioning wedge is placed along the lateral aspect of the foot and ankle. The lower extremity (foot and leg) is externally rotated 45° so that the lateral

surface rests against the wedge. The foot should be held nearly perpendicular to the leg; a sandbag can be propped against the sole of the foot to aid in positioning. (A variation of this technique positions the foot and leg so that they are externally rotated 30°.)    Central ray direction: Center of ankle. Lateral Positions (Figure 4-20)  1. Weight-bearing lateromedial projection    IR placement: Upright (vertical) in slot or well of orthoposer.    Tube head angulation: 90° from vertical.    Foot position: Patient stands on the orthoposer with the medial aspect of the ankle positioned against the image receptor so that the ankle joint axis (an imaginary line running through both malleoli) is perpendicular to the image receptor. This is best achieved by placing an index finger on each malleolus and visualizing the malleolar axis through the ankle. The heel may have to be pulled away from the image receptor a small distance to achieve a perpendicular position. The foot should be nearly perpendicular to the lower leg.    Central ray direction: Center of ankle.  2. Non–weight-bearing lateromedial projection    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: For a right-ankle study, the patient lies on the left body side and vice versa. The medial aspect of the ankle is positioned against the image receptor. The foot should be held nearly perpendicular to the leg; a sandbag can be propped against the sole of the foot to aid in positioning.    Central ray direction: Center of ankle.

   Other considerations: To perform this technique with an orthoposer, the patient must be seated, facing the orthoposer. The image receptor stands vertical in the well of the orthoposer. The medial aspect of the ankle is placed against the image receptor so that the ankle joint axis lies perpendicular to the image receptor. The tube head is angled at 90° (horizontal), and the central ray is directed at the center of the ankle.  3. Non–weight-bearing mediolateral projection    IR placement: Flat (horizontal) on radiographic table.    Tube head angulation: 0° (vertical).    Foot position: For a right-foot study, the patient lies on the right body side and for the left foot, on the left body side. The lateral aspect of the foot is positioned against the image receptor. The foot should be nearly perpendicular to the lower leg; a sandbag can be propped against the sole of the foot to aid in positioning.    Central ray direction: Center of ankle.

FIGURE 4-20. Lateral ankle positions. A: Weight-bearing lateromedial projection. B: Lateral view. C: Non–weight-bearing mediolateral projection. REFERENCES

  1. Meschan I. Roentgen Signs in Diagnostic Imaging. Vol 2, 2nd ed. Philadelphia, PA: WB Saunders; 1985.   2. Juhl JH. Paul & Juhl’s Essentials of Roentgen Interpretation. 4th ed. Philadelphia, PA: Harper & Row; 1981.   3. Bontrager KL, Anthony BT. Textbook of Radiographic Positioning and Related Anatomy. Denver, CO: Multi-Media Publishing; 1982.   4. Merrill V. Atlas of Roentgenographic Positions. 3rd ed. St Louis, MO: Mosby; 1967.   5. Frank ED, Long BW, Smith BJ. Merrill’s Atlas of Radiographic Positioning and Procedures. Vol 1–3, 12th ed. St Louis, MO: Mosby; 2011.   6. Weissman SD. Radiology of the Foot. 2nd ed. Baltimore, MD: Williams & Wilkins; 1989.   7. Weissman S. Standard radiographic techniques for the foot and ankle. Clin Podiatr Med Surg. 1988;5(4):767.   8. Kaschak TJ, Laine W. Surgical radiology. Clin Podiatr Med Surg. 1988;5(4):797.   9. Squire LF, Novelline RA. Fundamentals of Radiology. 4th ed. Cambridge, MA: Harvard University Press; 1988.  10. Greenfield GB, Cooper SJ. A Manual of Radiographic Positioning. Philadelphia, PA: Lippincott; 1973.  11. Dorland’s Illustrated Medical Dictionary. 27th ed. Philadelphia, PA: WB Saunders; 1988.  12. Gamble FO, Yale I. Clinical Foot Roentgenology. 2nd ed. Huntington, NY: Robert E Krieger Publishing; 1975.  13. Bontrager KL, Lampignano J. Textbook of Radiographic Positioning and Related Anatomy. 7th ed. St Louis, MO: Mosby; 2009.

 14. Hlavac HF. Differences in x-ray findings with varied positioning of the foot. J Am Podiatr Med Assoc. 1967;57:465.  15. Sgarlato TE. The angle of gait. J Am Podiatr Med Assoc. 1965;55:645.  16. Ruch JA. Significance and use of radiographs in the angle and base of gait. PAL Perspectives. 1980;1(2):1.  17. Osher LS, DeMore M, Atway S, et al. Extended pedal imaging via modifications of the traditional forefoot axial radiographic study. J Am Podiatr Med Assoc. 2008;98(3):171.  18. Gamble FO. A special approach to foot radiography. Radiogr Clin Photogr. 1943;19(3):78.  19. Gamble FO. Applied Foot Roentgenology. Baltimore, MD: Williams & Wilkins; 1957.  20. Brand PW, Coleman WC. A standard for dorsal-plantar and lateral radiographic projections of the feet. Orthopedics. 1987;10(1):117.  21. Blass BC, Imanuel HM, Marcus S. Technique for radiographic digital isolation. J Am Podiatr Med Assoc. 1974;64:870.  22. Downey MA, Dorothy WL. A radiographic technique to demonstrate the plantar aspect of the forefoot in stance. J Am Podiatr Med Assoc. 1969;59:140.  23. Fuson SM, Blau K, Beilman BA. A new sponge axial poser. J Am Podiatr Med Assoc. 1979;69:681.  24. Lewis RW. Non-routine views in roentgen examination of the extremities. Surg Gynecol Obstet. 1938;69:38.  25. Holly EW. Radiography of the tarsal sesamoid bones. Med Radiogr Photogr. 1955;31:73.  26. Causton J. Projection of sesamoid bones in the region of the first metatarsophalangeal joint. Radiology. 1943;9:39.

 27. Kleiger B, Mankin HJ. A roentgenographic study of the development of the calcaneus by means of the posterior tangential view. J Bone Joint Surg Am. 1961;43:961.  28. Helal B, Wilson D. The Foot. Vol 1. New York, NY: Churchill Livingstone; 1988.  29. Lilienfeld L. Anordnung der normalisierten rontgenaufnahmen des menschlichen korpers. ed 4. Berlin: Urban & Schwarzenberg; 1927. Cited by: Ballinger PW. Merrill’s Atlas of Radiographic Positions and Radiologic Procedures. 5th ed. St Louis, MO: Mosby; 1982.  30. Harris RI, Beath T. Etiology of peroneal spastic flatfoot. J Bone Joint Surg Br. 1948;30:624.  31. Coventry MB. Flatfoot with special consideration of tarsal coalition. Minn Med. 1950;33:1091–1097. Cited by: Ballinger PW. Merrill’s Atlas of Radiographic Positions and Radiologic Procedures. 5th ed. St Louis, MO: Mosby; 1982.  32. Vaughan WH, Segal G. Tarsal coalition, with special reference to roentgenographic interpretation. Radiology. 1953;60:855–863. Cited by: Ballinger PW. Merrill’s Atlas of Radiographic Positions and Radiologic Procedures. 5th ed. St Louis, MO: Mosby; 1982.  33. Broden B. Roentgen examination of the subtaloid joint in fractures of the calcaneus. Acta Radiol. 1949;31:85.  34. Isherwood I. A radiological approach to the subtalar joint. J Bone Joint Surg Br. 1961;43:566.  35. Pinsky MJ. The Isherwood views. J Am Podiatr Med Assoc. 1979;69:200.  36. Montagne J, Chevrot A, Galmiche JM. Atlas of Foot Radiology. New York, NY: Masson Publishing; 1981. SUGGESTED READINGS

Anthonsen W. An oblique projection for roentgen examination of the talocalcanean joint, particularly regarding intra-articular fracture of the calcaneus. Acta Radiol. 1943;24:306. Baron RL, Knight BL, Strugielski C. X-ray positioning. Pt III: Specialty radiographs. Podiatry Today. September, 1989:25. Baron RL, Strugielski C. X-ray positioning. Pt I: Weight bearing projections. Podiatric Staff. February, 1989:65. Baron RL, Strugielski C. X-ray positioning. Pt II: Non–weight bearing and partial weight bearing projections. Podiatric Staff. March, 1989:41. Cahoon JB. Radiography of the foot. Radiogr Clin Photogr. 1946;22:2. Cobey JC. Posterior roentgenogram of the foot. Clin Orthop. 1976;118:202. Dreeban S, Thomas PBM, Noble PC, et al. A new method for radiography of weight bearing metatarsal heads. Clin Orthop. 1987;224:260. Gabriel GR, Burger ES. Angular plates for fixed angle radiography. J Am Podiatr Med Assoc. 1967;57:1. Goff CW. Weight bearing x-rays of the feet. Am J Orthopedic Surg. 1968;10(1):13. Holly EW. Oblique anteroposterior projection for medial bones of foot. Med Radiogr Photogr. 1955;31:118. Inchaustegui N. Modified lateral view of the distal foot and hand for the evaluation of phalanges. Rev Interam Radiol. 1981;6:61. Johnson JE, Lamdan R, Granberry WF, et al. Hindfoot coronal alignment: a modified radiographic method. Foot Ankle Int. 1999;20(12):18. Keim HA, Ritchie GW. Weight bearing roentgenograms in the evaluation of foot deformities. Clin Orthop. 1970;70:133. Kirby KA, Loendorf AJ, Gregorio R. Anterior axial projection of the foot. J

Am Podiatr Med Assoc. 1988;78:159. Mendicino RW, Catanzariti AR, John S, et al. Long leg calcaneal axial and hindfoot alignment radiographic views for frontal plane assessment. J Am Podiatr Med Assoc. 2008;98(1):75–78. Min W, Sanders R. The use of the mortise view of the ankle to determine hindfoot alignment: technique tip. Foot Ankle Int. 2010;31(9):823. Saltzman CL, El-Khoury GY. The hindfoot alignment view. Foot Ankle Int. 1995;16(9):572–576. Samuelson KM, Harrison R, Freidman MAR. A roentgenographic technique to evaluate and document hindfoot position. Foot Ankle Int. 1985;1:286. Warren AG. Radiographic examination of the feet. Lepr Rev. 1973;44:131.

SECTION 2 Radiographic Anatomy

5 The Normal Foot and Ankle ROBERT A. CHRISTMAN A common pitfall when interpreting foot and ankle radiographs, made by both student and practitioner alike, is mistaking normal anatomic shadows for pathology. To truly appreciate abnormalities of position, form, density, and architecture, the interpreter must have a working knowledge and appreciation of normal radiographic anatomy. In the foot and ankle, this requires correlation of normal three-dimensional osteology (the bone specimen) to two-dimensional radiographic anatomy (the radiographic image). The reader is, therefore, encouraged to compare the radiographic image to a foot skeleton when interpreting radiographic studies. The following collection of illustrations, tracings of “rectus” foot images, and their respective radiographs should be used as a general reference guide. Slight variations in the appearance of each bone occur depending on the position of the foot relative to the image receptor. For example, a pronated or supinated foot affects the overall radiographic appearance of the entire foot as well as each individual bone (Figure 5-1). Therefore, it is important to concentrate on the correlation of gross anatomy to the radiograph, not to memorize the appearance of each bone per se. The illustrations that follow include a variety of radiographic views (Figures 5-2 to 5-12). Sarrafian’s Anatomy of the Foot and Ankle1 was used as the reference for the terminology in the labels. Occasionally, however, I used the best generic term to label a radiographic “shadow” that Sarrafian had not specifically identified. The label key is given in Box 5-1. I have spent much of my career studying foot and ankle bone specimens (normal three-dimensional osteology) and correlating them to twodimensional radiographs.2–8 The original manuscript for this chapter

included a complete reference guide to radiographic anatomy of the foot and ankle. But, because of its large size and numerous images, it was not included in this textbook. The manuscript, however, has been accepted for publication in five parts and should be in print by the time this second edition becomes available.9–13 The project correlates detailed radiographic anatomy of the entire adult foot and ankle (two-dimensional) to osteology (threedimensional). Images of each foot and distal leg bone (“front” and “back” perspectives) are presented alongside a corresponding radiographic image for comparison. It should serve as a baseline (“normal”) that future researchers can use as well as a reference that both students and practitioners can use for comparison when interpreting radiographs and distinguishing abnormal findings from normal.

FIGURE 5-1. The effect of pronation and supination on the position and form of bones (all images of the same foot): A: Pronated; B: Rectus; C: Supinated.

FIGURE 5-2. A and B: Dorsoplantar (DP) foot view.

FIGURE 5-3. A and B: Lateral foot view.

FIGURE 5-4. A and B: Medial oblique foot view.

FIGURE 5-5. A and B: Lateral oblique foot view.

FIGURE 5-6. A and B: Sesamoid axial view.

FIGURE 5-7. A and B: Calcaneal axial view.

FIGURE 5-8. A and B: Anteroposterior (AP) ankle view.

FIGURE 5-9. A and B: Mortise ankle view.

FIGURE 5-10. A and B: Medial oblique ankle view.

FIGURE 5-11. A and B: Lateral oblique ankle view.

FIGURE 5-12. A and B: Lateral ankle view. BOX 5-1 Label Key 1C 2C 3C 5MT ABLS ACTM ALET

First cuneiform (medial cuneiform) Second cuneiform (intermediate cuneiform) Third cuneiform (lateral cuneiform) Tuberosity of fifth metatarsal base Anterior border of distal tibial lateral surface Anterior colliculus (of tibial malleolus) Anterolateral extension of trochlear

AMET ASMA ASMP ATC ATT BC BPB BPC C Ca Cu DP F FF FM FS G GAC INCJ ILR LNCJ LPT LTAS LTC LTM LTS

surface (talus) Anteromedial extension of trochlear surface (talus) Articular surface for medial malleolus, anterior margin Articular surface for medial malleolus, posterior margin Anterior tuberosity (anterior tubercle) Anterior tibial tubercle Beak of cuboid Bony projection, distal phalanx base (varies in size) Bursal projection, posterior calcaneus Crista (crest of metatarsal head) Calcaneus Cuboid Distal phalanx Fibula Fibular (digital) fossa Fibular malleolus Fibular sesamoid Groove separating tubercle and articular surface Great apophysis (anterior process of calcaneus) Intermediate naviculocuneiform joint Interosseous ligament rugosity Lateral naviculocuneiform joint Lateral process (talus) Lateral trochlear articular surface (hallux proximal phalanx) Lateral tuberosity (lateral tubercle) of calcaneus Lateral tubercle (for metatarsophalangeal ligaments) Lateral trochlear surface (first

MCJ1 MCJ2 MCJI MCJS MNCJ MP MPA MTAS MTC MTM MTS N NT PBLS PCTM PEL PEM PL PLT PMT PP PR PTF PTT

metatarsal) First metatarsocuneiform joint Second metatarsocuneiform joint Inferior aspect of first metatarsocuneiform joint Superior aspect of first metatarsocuneiform joint Medial naviculocuneiform joint Middle phalanx Medial and lateral margins of plantar apex of bone Medial trochlear articular surface (hallux proximal phalanx) Medial tuberosity (medial tubercle) of calcaneus Medial tubercle (for metatarsophalangeal ligaments) Medial trochlear surface (first met) Navicular Tuberosity of navicular Posterior border of distal tibial lateral surface Posterior colliculus (of tibial malleolus) Proximal extension of metatarsal head articular surface, laterally Proximal extension of metatarsal head articular surface, medially Tubercle for insertion of peroneus longus tendon Posterolateral tubercle (talus) (trigonal process) Posteromedial tubercle (talus) Proximal phalanx Phalangeal ridge Posterior tubercle (fibula) Posterior tibial tubercle

RP RTE ST STC Ta TC TCJA TCJM TCJP Ti TS UT

Remnant of physis Retrotrochlear eminence Sustentaculum tali Sinus tarsi/tarsal canal Talus Tuberosity of cuboid Talocalcaneal joint, anterior Talocalcaneal joint, middle Talocalcaneal joint, posterior Tibia Tibial sesamoid Ungual tuberosity (tuft of distal phalanx)

REFERENCES   1. Kelikian AS. Sarrafian’s Anatomy of the Foot and Ankle: Descriptive, Topographic, Functional. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2011.   2. Christman RA. Radiographic anatomy of the calcaneus. I. Inferior surface. J Am Podiatr Med Assoc. 1987;77(10):549–553.   3. Christman RA. Radiographic anatomy of the calcaneus. II. Posterior surface. J Am Podiatr Med Assoc. 1987;77(11):581–585.   4. Christman RA. Radiographic anatomy of the calcaneus. III. Superior surface. J Am Podiatr Med Assoc. 1987;77(12):633–637.   5. Christman RA. Radiographic anatomy of the calcaneus. IV. Lateral and medial surfaces. J Am Podiatr Med Assoc. 1988;78(1):11–14.   6. Christman RA, Ly P. Radiographic anatomy of the first metatarsal. J Am Podiatr Med Assoc. 1990;80(4):177–203.   7. Ruiz JR, Christman RA, Hillstrom HJ. Anatomical considerations of the peroneal tubercle. J Am Podiatr Med Assoc. 1993;83(10):563.

  8. Morgan A, Kim PS, Christman RA. Radiographic anatomy of the talus. J Am Podiatr Med Assoc. 2003;93(6):449.   9. Christman RA. Radiographic anatomy of the foot and ankle. Part 1: The distal leg. J Am Podiatr Med Assoc. 2014;104(4):402.  10. Christman RA. Radiographic anatomy of the foot and ankle. Part 2: The greater tarsus. J Am Podiatr Med Assoc. 2014;104(5).  11. Christman RA. Radiographic anatomy of the foot and ankle. Part 3: The lesser tarsus. J Am Podiatr Med Assoc. 2014;104(6).  12. Christman RA. Radiographic anatomy of the foot and ankle. Part 4: The metatarsals. J Am Podiatr Med Assoc. 2015;105(1).  13. Christman RA. Radiographic anatomy of the foot and ankle. Part 5: The toes. J Am Podiatr Med Assoc. 2015;105(2).

6 Normal Variants and Anomalies ROBERT A. CHRISTMAN Thirty bones compose the foot and ankle complex. (This number includes the distal tibia and fibula and the two sesamoids at the first metatarsophalangeal joint.) The previous chapter described the expected radiographic appearance of each bone in the many views available. However, variations in their appearance are not uncommon. Variations typically are incidental findings and asymptomatic, although symptomatology or pathology may occur secondary to a skeletal variation. More common examples of the latter include the accessory navicular and tarsal coalition. Several words are frequently used when referring to variations of the skeleton. Their definitions are as follows (from Dorland’s Illustrated Medical Dictionary1):    Accessory: supplementary to another similar and generally important thing.    Anomaly: marked deviation from the normal standard, especially as a result of congenital defects.    Os: bone; a general term that is qualified by the appropriate adjective to designate a specific type of bony structure or a specific segment of the skeleton.    Ossicle: a small bone.    Partite: having parts or divisions.    Sesamoid: a small, nodular bone embedded in a tendon or joint capsule.    Supernumerary: in excess of the regular or normal number.

   Synostosis: the osseous union of two bones that are normally distinct.    Variant: something that differs in some characteristic from the class to which it belongs, as a variant of a disease, trait, and so forth. The terms partite, supernumerary, and accessory bone in the following discourse are differentiated thus: Partite can pertain to either a normally existing bone or an accessory bone that is subdivided; supernumerary pertains to anomalous duplication of a normally existing bone; and accessory bone refers to those ossicles (not anomalous duplication) that are found in addition to the normally existing bones. Also note that although all sesamoid bones are ossicles, not all ossicles are sesamoid bones. Variations can be expressed in several ways (Box 6-1). Examples of variations involving normally existing bones include variants of number, position, form (shape and size), density, and architecture. Accessory ossicles, in addition to the existing 30 bones, are not uncommon. Variants of number include partite bones, supernumerary bones, and absence of bones. Bipartite sesamoids, for example, are common. Although rare, bipartite medial cuneiform and navicular bones may be encountered. Supernumerary bones include duplication of phalanges, metatarsals, and tarsal bones. BOX 6-1 Expression of Variants in the Foot Variations Involving Normally Existing Bones Variants of number Variants of position Variants of form Variants of density Variants of architecture Accessory ossicles

The axis of each bone has a characteristic position relative to the axes of adjacent bones. In the two-dimensional radiograph, this position can vary considerably among patients and depends heavily on foot type and the degree of pronation or supination during weight-bearing stance in angle and base of gait.2 Variations of position can predispose for future pathology. Most variations of existing bones are variants of form. Variants of form concern the girth, length, and contour of a bone. Synostosis/coalition is also a variation of form. Other variants of form can be attributed to the position of the foot or ankle bones relative to each other and/or the image receptor. For example, the navicular bone frequently appears rectangular in a pes cavus foot but wedge shaped in a pes planus foot. Varying forms may also be encountered in weight-bearing studies secondary to pronation and supination (internal and external leg rotation, respectively; refer to Figure 5-1). Angle and base of gait positioning provide a means for minimizing the technical variation that may occur (described in Chapter 4). Positional changes are predictable, however, and the experienced interpreter can recognize these alterations and correlate them to normal radiographic anatomy. Variants of density can mimic abnormal or pathologic processes. More commonly, this appears as a relative radiolucency or decreased density. Occasionally, focal areas of increased density are seen. Each bone has its own characteristic shadows. These shadows represent cortex, trabeculae, and superimposed osseous landmarks in the adult skeleton and constitute the architecture of the bone. The outer margin of the bone is its external architecture, and the remainder is its internal architecture. Most architectural variants relate to the appearance of cortical bone and trabeculae. Variation of an osseous landmark’s superimposed shadow may correlate to variant size, shape, or position of that landmark. Variations of the existing 30 bones are not the only variants one may encounter. Numerous accessory bones are commonly found in the foot and ankle. One may be an isolated finding or multiple ossicles can occur throughout both extremities. Appreciation of normal radiographic anatomy and the summation of shadows concept (Chapters 5 and 9, respectively) are the first steps of image

interpretation. The interpreter must then become familiar with the numerous variations of normal. Many of these variants appear unilaterally and can mimic fractures or other pathology. Bilateral studies, therefore, are not always useful for their differentiation and can be misleading. Furthermore, even if they are bilateral, they may be asymmetric in appearance. The purpose of this chapter is to present the gamut of radiologic variants that one may encounter in the adult foot and ankle. VARIANTS OF NUMBER Partite Sesamoids at the First Metatarsophalangeal Joint It is common to see partite tibial and fibular sesamoids; they are typically bipartite. Partitioning may involve one or both sesamoids and be unilateral or bilateral. And, when bilateral, symmetry is not the rule.3 For example, if only the tibial sesamoid is partite in one foot, only the fibular sesamoid may be partite in the opposite foot. The tibial sesamoid is more commonly partite than the fibular sesamoid.4 The presentations (partitioning, shape, size, and number) are extremely variable and follow no specific patterns. For example, a bipartition may divide the sesamoid into equal or unequal halves; the division can be transverse (most common, Figure 6-1A), oblique, or, least common, longitudinal (Figure 6-1B). It can be difficult to differentiate a fractured sesamoid from a partite sesamoid,5 especially in the dorsoplantar (DP) view. The presence of jagged edges alone is not a useful distinguishing feature, because both the bipartite and the fractured sesamoid can appear to have jagged edges. For example, a partite sesamoid that is complicated by degenerative joint disease has spurs that give the appearance of a jagged edge, similar to that of a fracture. More commonly, the prominent primary trabeculations that are normally seen in the first metatarsal distal metaphysis are superimposed on the sesamoids; these shadows can exaggerate the perception of a jagged fracture line. Nor can fracture be determined by the amount of separation between the two segments. It is not uncommon to see apparent “distraction” in the asymptomatic, untraumatized bipartition. Fracture is best differentiated from variant bipartition with the sesamoid isolated, that is, with no superimposition on the metatarsal head (this can only be accomplished with the sesamoid

axial and lateral oblique or modified non–weight-bearing lateral—or Causton —views). Certainly correlation with clinical history is important in these instances. Bone scintigraphy and magnetic resonance imaging (MRI) may be valuable adjunctive studies for further differentiation of partition versus fracture versus symptomatic partition. The sesamoids are initially evaluated with the dorsoplantar and sesamoid axial views. The lateral oblique view is valuable for assessing tibial sesamoid pathology (Figure 6-1C).6 The tibial and fibular sesamoids are superimposed on the first and second metatarsals, respectively, in the medial oblique view; this view generally does not provide any additional information apart from the dorsoplantar, lateral oblique, and axial views. The weight-bearing lateral view is useless for imaging the sesamoids; they are superimposed on each other in addition to other metatarsal and phalangeal bones.

FIGURE 6-1. Bipartite tibial sesamoid. A: Transverse partition. B: Longitudinal partition. C: Lateral oblique view for evaluating tibial sesamoid.

FIGURE 6-2. Bipartite medial cuneiform. A: Complete partition (arrowheads). B: Incomplete partition posteriorly (arrow).

Bipartite Medial Cuneiform (Os Cuneiforme I Bipartitum) In rare instances, the medial cuneiform presents as two bones in the adult skeleton, a variant known as the bipartite medial cuneiform. They are more frequently found bilaterally and predominate in males.7 The partition classically divides the bone into upper and lower halves and is best seen in the lateral view (Figure 6-2A). The classic bipartition, when present, is easily overlooked. It is fully superimposed on the remaining cuneiform bones and easily mistaken for other bone shadows. However, the transverse joint space identified in the center of the medial cuneiform is its characteristic radiologic feature. This arthrodial joint space typically is complete from anterior to posterior. Midfoot/arch pain may be associated with this entity.8,9 A variation may be encountered where the bipartition is incomplete. The medial cuneiform may be incompletely divided into two segments at its anterior and/or posterior margins but is “fused” centrally. Indentations are seen along the articular surfaces (Figure 6-2B). Bipartite Navicular (Müller–Weiss Syndrome) The few cases that have been reported as the bipartite navicular mimic a disorder referred to as Müller–Weiss syndrome. Interestingly, there is minimal to no overlap between these two entities when reported in the literature; in other words, reports of the bipartite navicular rarely mention Müller–Weiss syndrome, and vice versa. Are they the same, or not? Müller–Weiss syndrome, which appears to occur spontaneously in the adult navicular, is of unknown etiology. The deformed navicular resembles osteonecrosis on MRI; however, most histopathologic studies have not been able to provide conclusive evidence of osteonecrosis.10 This syndrome is usually bilateral and occurs predominantly in females in their fourth to sixth decades.11 Bipartite navicular is also of unknown etiology, though some feel it may be a congenital anomaly. Maceira and Rochera12 have presented a large cohort of cases; they propose

that Müller–Weiss syndrome is related to delayed ossification of the navicular that becomes deformed secondary to abnormal force distribution on the bone. They have grouped the radiographic presentations of increasing deformity into five stages for didactic purposes, which, as Brailsford13 noted in 1939, are chronic and progressive. Generally speaking, stage 1 is a normal radiograph, with the talar axis bisecting the navicular and medial cuneiform bones; between stages 2 and 4, the talar axis moves superiorly then inferiorly, and the navicular bone fragments and subluxates off the talar head; in stage 5, there is greater subluxation, even dislocation of the navicular fragments such that the lateral and intermediate cuneiforms appear to articulate with the talar head. Ultimately, the progression of deformity results in osteoarthritis of the talonavicular joint.14 Classically, the division or partition separates the bone into a smaller superolateral segment and a larger inferomedial segment.15 Bipartite navicular, which appears similar to (or, is the same as) the “stage 3” Müller– Weiss syndrome, appears as follows: in the dorsoplantar view, the larger navicular segment has been described as being is shaped like a wedge15 or comma12 with its base medially and apex pointing laterally; the smaller segment is superimposed on the lateral cuneiform and cuboid (Figure 6-3A). In the lateral view, the smaller segment is wedge-shaped, positioned along the dorsal aspect of the larger segment; its apex is directed inferiorly (Figure 6-3B).16 The smaller bipartite segment appears closely associated, possibly articulating, with the intermediate cuneiform superoposteriorly. Computed tomography (CT) is valuable for confirming the navicular’s bipartite nature.17 Supernumerary Bones Anomalous duplication of a bone is rare (Figure 6-4). The phalanges are most frequently affected. Metatarsal duplication is usually incomplete or dwarfed unless accompanied by tarsal duplication. Absence of Normally Existing Bones

On rare occasion one or several normally existing bones may be absent. The absence of a single bone may not be obvious clinically and is identified incidentally with radiographs; primary examples include absence of the tibial or fibular sesamoid18 and absence of a lesser toe middle phalanx (Figure 65A,B). Absence of a sesamoid and aphalangia may be either bilateral or unilateral. Neither is associated with symptomatology. In contrast, anomalous absence of multiple bones can present with gross deformity. A classic example is the so-called claw or lobster foot (Figure 6-5C).

FIGURE 6-3. Bipartite navicular. A: DP view: comma-shaped larger medial segment; the arrow identifies the smaller superolateral segment. Also note the medial subluxation of navicular relative to talar head. B: Lateral view: partition is not clearly seen (arrowhead); odd presentation of the lateral cuneiform-navicular articulation (straight arrow).

FIGURE 6-4. Supernumerary bones. A: Duplicate distal and middle phalanges, second toe (a unilateral finding). B: Duplicate hallux (a unilateral finding). C: Duplicate fifth metatarsal and toe (a bilateral finding). VARIANTS OF POSITION Each bone has a characteristic position in the foot and ankle. This is best assessed in the dorsoplantar and lateral views with the foot bearing weight in its angle and base of gait. The axis of any particular bone lies at a particular position relative to other bones. All three anatomic planes should be considered when evaluating position. Unfortunately, with two-dimensional radiographs it is difficult to fully appreciate position in all planes. The frontal plane position of most foot bones cannot be directly evaluated radiographically; assessment, therefore, requires logical analysis and reasoning while looking at both the dorsoplantar (transverse plane) and lateral (sagittal plane) views.

FIGURE 6-5. Absent normally existing bones. A: Tibial sesamoid (a bilateral finding). B: Middle phalanges (a unilateral finding). C: Cuneiforms, metatarsals, and toes (a bilateral finding). Position variants predispose the patient to future symptomatology and pathology. Normal and abnormal position of foot bones, including foot structure, and their relationships to biomechanics, are discussed in Section 4 (Chapters 12 and 13). VARIANTS OF FORM Distal Phalanx Variations of the ungual tuberosity may be encountered in the dorsoplantar view. Its entire outer margin may be irregular or spiculated (Figure 6-6), simulating the reaction associated with inflammatory processes (e.g., chronic nail infection or psoriatic arthritis). Another variation is absence of the medial or lateral margin; absence of both margins simulates the whittling of bone associated with forefoot neuropathic arthropathy. Occasionally, the entire ungual tuberosity may be absent or appear atrophic. In a lesser toe, for example, the distal phalanx may appear triangular.

FIGURE 6-6. Varying presentation of the ungual tuberosity. Multiple distal indentations (arrowheads). (Also note the enlarged basal inferomedial tubercle, the superimposed sclerotic anterior aspect of the flexor tuberosity, and an os interphalangeus.) Spur formation is occasionally seen along the posteroinferior, medial, and lateral margins of the ungual tuberosity. It will be seen in the dorsoplantar and lateral views. The lateral interosseous ligaments insert along these margins.19,20 A spur may be seen superiorly in the lateral view; this “mechanical” subungual spur may be related to hallux limitus.21 Occasionally, variation of the length or girth of a distal phalanx is encountered. The phalanx may be increased or decreased in length or girth. Increased girth can many times be attributed to the flexor tuberosity that may be enlarged or hypertrophied (Figure 6-7A,B). This is best identified in the isolated hallux lateral view. A wavy contour that mimics a periosteal reaction is infrequently seen along the shaft’s medial and lateral margins (Figure 67C). “Whiskering” (small spurs along the diaphysis) may also be seen as a normal variation that is suggestive of psoriatic arthritis. Variations of the hallucal distal phalanx base may also be encountered. The size of the basal inferomedial tubercle, for example, varies considerably, ranging from absent to large radiographically (Figures 6-8 and 6-6, respectively). A recent investigation of this bony excrescence proposes that it probably represents a reaction of normal bone to repeated forces occurring during gait.22 The lateral interosseous ligaments that run between the base and ungual tuberosity originate from the inferomedial and inferolateral basal tubercles. It infrequently calcifies, and spur formation may be noted at the entheses. Occasionally, variation of the lesser toe distal phalanx base is seen. The base commonly flares medially and laterally in the dorsoplantar view. Often it appears quite pronounced relative to the narrow shaft. In rare instances, the lateral aspect of the hallux distal phalanx base appears absent.

FIGURE 6-7. Varying presentations of the distal phalanx shaft. A: Enlarged flexor tuberosity (lateral view). B: Enlarged flexor tuberosity (dorsoplantar view, a bilateral finding). C: Wavy medial and lateral margins mimicking periosteal reaction (a bilateral finding). Middle Phalanx The shape of the lesser toe middle phalanx may be either square or rectangular. The margins of a rectangular middle phalanx may appear either flat or indented (concave). The head of a middle phalanx occasionally is angulated laterally relative to its shaft, resulting in angulation of its respective distal phalanx (Figure 6-9). Proximal Phalanx The proximal phalanx is a short, tubular bone. Variations in its length and girth are frequently encountered. For example, the phalanx may be shorter in length and wider in girth. This variant more frequently affects the lesser toes (Figure 6-10), although it also can involve the hallux. In contrast, a decreased diaphyseal girth may be seen. One example is the spool-shaped proximal phalanx.23 The medullary canal may even appear obliterated in exaggerated cases. Occasionally, a proximal phalanx head is enlarged. This finding primarily involves the fifth digit and usually accompanies the clinical presence of an adductovarus contracture with a heloma durum overlying the hypertrophied head. A small exostosis may be seen along the head’s superolateral aspect. A remnant of the phalangeal cleft (a variant of normal development) may be identified in a similar location. Rarely, the hallux proximal phalanx head is enlarged. A wavy contour and associated increased girth are commonly encountered along the medial and lateral margins of a lesser toe’s diaphysis (Figure 611A); this corresponds to the diaphyseal ridge, which may be misinterpreted as a periosteal reaction. Its size varies considerably, and it may or may not be seen on all lesser digit proximal phalanges of the same patient. Two structures insert along the diaphyseal ridge: the wing of the extensor hood apparatus and the fibrous flexor sheath. Rotation of the digit obscures its

visualization. A different type of cortical thickening and irregular contour is infrequently seen along the medial and/or lateral margins of the hallux proximal phalangeal distal diametaphysis (Figure 6-11B).

FIGURE 6-8. Varying presentation of the distal phalanx base: Absent tubercles (a bilateral finding). Compare to the enlarged tubercle in Figure 6-6. Sesamoids The size of a sesamoid bone is fairly constant, although infrequently an enlarged or atrophic sesamoid may be encountered. The two sesamoids are equal in size in approximately 50% of the time; the tibial sesamoid is larger in 35%, the fibular in 15%.4 They typically have smooth and regular margins. Occasionally, however, small protuberances resembling spurs may be seen along their posterolateral margins in the absence of any degenerative process or enthesopathy elsewhere. Rarely, a sesamoid may be much smaller than normally expected and when compared to its mate (Figure 6-12). Coalition of the two sesamoids has been reported.24 Metatarsals Metatarsals have a characteristic length pattern in the dorsoplantar view (Figure 6-13A). Typically, the anterior end of the second metatarsal is most distal, followed by the first and third metatarsals (the anterior ends of the latter two bones are nearly at the same position distally). The anterior end of the fourth metatarsal is more proximal relative to the third metatarsal, and the fifth is most proximal. Many variations are encountered regarding the lengths of each metatarsal relative to one another, some appearing longer and others shorter than the normal pattern described earlier (Figure 6-13B). The short fourth metatarsal is typically anomalous.

FIGURE 6-9. Varying presentation of the middle phalanx. Lateral aspect of second toe middle phalanx is significantly shorter than the medial aspect, resulting in angulation of the distal phalanx laterally. Note also the square middle phalanges of the third and fourth toes.

FIGURE 6-10. Short proximal phalanx third and fourth toes (a bilateral finding).

FIGURE 6-11. Varying presentations of the proximal phalanx shaft. A: Phalangeal ridge (arrowheads, second, third, and fourth toes). B: Increased girth along the anterior aspect of the medial shaft (arrow) simulating a periosteal reaction. Present bilaterally.

FIGURE 6-12. Varying presentation of the sesamoids: Smaller tibial sesamoid (bilateral). The first metatarsal head can have varying presentations. The shape of the distal articular surface may be round (most common), be flat, or demonstrate a central ridge (Figure 6-14A,B). The medial and superior aspects of the first metatarsal head vary in size and can be quite enlarged and hypertrophied. An enlarged tubercle may be seen medially at the diametaphyseal margin. The superolateral tubercle is occasionally seen along the superior aspect of the first metatarsal neck in the lateral view (Figure 6-14C). Regarding the lesser metatarsals, the head may (rarely) be flat distally with associated increased girth. The tubercles along the superomedial and superolateral aspects of their heads can be hypertrophied and prominent, visible in the medial oblique and dorsoplantar views, respectively. The neck’s girth is consequently increased laterally when the superolateral tubercle is enlarged (Figure 6-15A), and may simulate an old, healed fracture. The anatomic groove located between each tubercle and the anterior articular surface may be exaggerated in radiographic appearance (Figure 6-15B). The posteroinferior articular extensions of the lesser metatarsal heads may be pronounced in size; this finding, although superimposed, will be identified in the dorsoplantar view (Figure 6-15C). A lesser metatarsal head may bow or angulate medially relative to the shaft, which is typically symmetrical and bilateral. Variations of the girth, contour, and form of a metatarsal shaft are encountered. The girth, for example, may be increased (Figure 6-16A) or decreased. Infrequently, a metatarsal shaft is overtubulated (over constricted) and appears extremely narrow (Figure 6-16B). Anatomic variations have been described that would correspond to this radiographic presentation.25 Increased girth of a metatarsal shaft may be secondary to Wolff’s law (increased stresses result in bone formation and remodeling). An example is the enlarged second metatarsal shaft associated with a short first metatarsal (also known as the Morton’s foot).26 Occasionally, the cortex along the inferolateral aspect of the first metatarsal shaft proximally appears thickened. This site correlates anatomically to the insertion of the peroneus longus

tendon. Lateral bowing of a metatarsal, especially the fifth, will be also seen. Increased girth of the fifth metatarsal midshaft medially may accompany this finding and is symmetric bilaterally (Figure 6-16C).

FIGURE 6-13. Variation of metatarsal length. A: Normal. B: Short second metatarsal.

FIGURE 6-14. Varying presentations of the first metatarsal head. A: Flat. B:

Combined flat/ridge. (For comparison, an example of the round metatarsal head is seen in Figure 6-13B.) C: Prominent lateral tubercle (arrow). Along the margins of the metatarsal bases variations may be seen that could be mistaken for pathology. The medial and lateral aspects of the lesser metatarsal bases normally are irregular; this finding is exaggerated by a deep groove located along the inferior and anterior aspects of the articular facet for the adjacent metatarsal. It is especially evident along the lateral aspects of the second-, third-, and fourth metatarsal bases (Figure 6-17A). A pronounced articulation is occasionally seen between the bases of the first and second metatarsals (Figure 6-17B); the articulations between the lesser metatarsal bases are rarely visible. A tubercle may be found, along the first metatarsal base medially, which provides insertion for the tibialis anterior tendon (Figure 6-17C). Another tubercle might be seen superiorly. The size of the fifth metatarsal tuberosity is variable. It may appear absent, elongated posteriorly, or project laterally.

FIGURE 6-15. Varying presentations of the lesser metatarsal heads. A: Enlarged superolateral tubercle (arrowhead), second metatarsal (a bilateral finding). B: Exaggerated notch between superolateral tubercle and anterior articular surface (arrowhead), second metatarsal (a bilateral finding). C: Enlarged and elongated plantar-medial condyle (arrowheads), fourth metatarsal.

FIGURE 6-16. Metatarsal variants of form. A: Increased girth (first metatarsal). B: Decreased girth third metatarsal, increased girth second metatarsal. C: Fifth metatarsal bowing with cortical thickening (arrowheads). A variant of density, radiolucency of the fifth metatarsal head medially, is also seen. Variant articular surface shapes can also be encountered in the DP view. The cuneiform articular surface of the first metatarsal base may be flat or even concave. Occasionally, the lateral cuneiform articular surface of the third metatarsal is also concave, but, in most cases, it cannot be clearly identified because adjacent structures are superimposed. Cuneiforms, Cuboid, and Navicular Variability regarding the shape or form of the medial cuneiform is primarily positional in nature, but its form along the medial surface can vary. (The complete and incomplete bipartite medial cuneiforms are discussed earlier in the sub-section Variants of Number.) One specific variant finding not related

to position is a tubercle along the medial surface seen in the dorsoplantar view (Figure 6-18). The size of this tubercle varies, from small to large, and its position is inferomedial and typically situated closer to the first metatarsal articular surface, although it can also appear more posteriorly. It probably provides a gliding surface or “guide” for the tibialis anterior tendon, similar in function to the peroneal tubercle for the peroneal tendon.

FIGURE 6-17. Varying presentations of the metatarsal base. A: Exaggerated, irregular intermetatarsal surfaces (arrowheads) between bases and proximal shafts (a bilateral finding). B: Articulation between first- and second metatarsal bases (arrowhead) (a bilateral finding). C: Tubercle for tibialis anterior tendon insertion (arrow) (a bilateral finding).

FIGURE 6-18. Tubercle along the medial surface of the medial cuneiform (arrowhead). Superimposition of the midfoot bones is normally seen in the dorsoplantar view. This could lead to the misdiagnosis of a cubocuneiform coalition. Although trabeculations appear continuous between the two bones, the superimposed shadows of their articular margins can still be identified. This latter finding would not be seen if it were a true coalition. The cuboid has few variations to note. Its combined articular surface for the fourth and fifth metatarsals is typically flat or nearly flat; infrequently, it has a prominent triangular ridge centrally separating the two articular surfaces, best seen in the medial oblique view (Figure 6-19). The anteromedial corner of the cuboid frequently juts medially toward the cuneiform, simulating an osteophyte or spur. Most variants of navicular form are identified along its medial margin in the dorsoplantar or lateral oblique views. The navicular occasionally appears wedge-shaped (its lateral half is narrow relative to the larger medial half in the dorsoplantar view), although it more frequently is somewhat rectangular. The tuberosity may be large, small, or absent in some individuals. When the tuberosity is elongated, it appears to wrap around the talar head. The elongated tuberosity may be related to a fused accessory navicular ossification center (see sub-section Accessory Navicular). A large tubercle is infrequently seen along the navicular medial surface anterior to the tuberosity (Figure 6-20). In yet other patients, the tuberosity may have an anomalous location medially; it classically is situated posteromedially. The medial margin of the navicular’s medial cuneiform articular surface may be pronounced and extend anteriorly. A similar spurlike extension may be seen in the lateral view projecting anteriorly along the intermediate cuneiform articular surface. Variant form of the navicular simulating a bipartite navicular may be associated with a pronounced superoanterior margin. Finally, a tubercle may be visible in the medial oblique view for the cuboid articular surface.

FIGURE 6-19. Cuboid variant of form: Ridge separating fourth- and fifth metatarsal articulating surfaces (arrowhead).

FIGURE 6-20. Navicular variants of form: Tubercle along medial surface anterior to the tuberosity (arrow). Talus The talus is best isolated in the lateral view. Its form is fairly consistent; however, occasionally it varies. The neck, for example, may be short or elongated. Less frequently, the body may appear flattened and the head/neck enlarged. Rarely, the head appears flattened. Spurs are frequently seen along the superior aspect of the talar head or neck. Anatomically, a small ridge is normally present along the superior surface at the junction of the talar head and neck that parallels the navicular articular

surface; the talonavicular and talotibial joint capsules and ligaments insert here.27 This ridge may be enlarged as a variation of normal (Figure 6-21). A spur is frequently identified on the talar head that is continuous with the navicular articular surface; this is an osteophyte, a feature of talonavicular joint osteoarthritis, and is found in association with limited range of tarsal joint motion. However, its formation may be related to a prior injury.

FIGURE 6-21. Prominent ridge along the talar head/neck superiorly (arrowhead).

FIGURE 6-22. Variant of talar posterolateral process: Prominent trigonal process. The talar posterolateral process is continuous with the tibial articular surface of the talar dome in the lateral view and presents as a small protuberance. The process may be elongated and extend posteriorly. This has also been referred to as the trigonal process or Stieda’s process and probably represents a fused accessory ossification center (Figure 6-22) (see sub-section Os Trigonum). The talar dome normally has a semicircular outline in the lateral view. If the ankle joint axis is not positioned perpendicular to the x-ray image receptor, however, the medial and lateral shoulders of the dome are not aligned and give the appearance of a flattened talar dome (Figure 6-23). This appearance usually is positional in nature and is seen with a supinated/cavus foot in the weight-bearing lateral view.

FIGURE 6-23. The appearance of this “flattened” talar dome is purely positional in nature. Note that the fibular malleolus (F) is posterior in position relative to the tibial malleolus (T). Calcaneus The form of the calcaneus is fairly constant. However, occasionally, the overall shape appears somewhat rectangular and/or expanded, with distortion of its normal distinctive curves and lines. For example, the calcaneal length may appear elongated or foreshortened, the latter being the result of an enlarged posterior segment (Figure 6-24). Localized hypertrophy of the bursal projection (the posterosuperior aspect of the calcaneus) may or may not be associated with clinical symptomatology (Haglund disease and/or retrocalcaneal bursitis). The inferior surface rarely appears perfectly flat (Figure 6-25). The anterior process may be enlarged and extend superiorly in

the lateral view in the presence or absence of a calcaneonavicular coalition (Figure 6-26). A projection of bone is rarely seen along the medial aspect of the calcaneus in the ankle anteroposterior view. This variation articulates with an extension from the talar medial process and has been called the “assimilated os sustentaculum tali” because of its location. Anatomically, it is found along the posterior aspect of the sustentaculum tali. It may be identified as a bony palpable protuberance clinically and has been associated with symptomatology.28 This probably represents an extra-articular talocalcaneal coalition (see discussion of synostosis and coalition later in this chapter). The tubercles and tuberosities are occasionally enlarged and prominent. An enlarged peroneal tubercle or retrotrochlear eminence is infrequently seen along the lateral calcaneal body in the dorsoplantar and calcaneal axial views (Figures 6-27A,B).29 The anterior tuberosity may also be enlarged (Figure 627C). Occasionally, the lateral tubercle may appear hook-shaped with a foramen, probably the result of variant development (Figure 6-27D). Radiographically, the middle and anterior articular surfaces for the talar head appear continuous (this is only appreciated with the medial oblique view). Variations include a separate articular surface for the anterior subtalar joint and absence of an articulation completely (Figure 6-28). The superior surface of the sustentaculum tali typically is flat in the lateral view; rarely, it appears curved. Also, the posterior aspect of the sustentaculum should not be continuous or articulate with the posteromedial talar process in the lateral view. Continuity is indicative of osseous coalition, articulation, or fibrocartilaginous union (see discussion of tarsal coalitions later in this chapter). Distal Tibia and Fibula Form variants of the distal tibia or fibula are infrequent. For example, the tibial and fibular malleoli are occasionally elongated (Figure 6-29A). An extended tibial malleolus anterior colliculus may be related to ossification of the deltoid ligament at its enthesis. Elongation of the fibular malleolus may be accompanied by a hooklike extension at its inferior tip. Another variant of

form is the presence of an enlarged tubercle or process along the medial aspect of the tibial malleolus (Figure 6-29B) or the posterior aspect of either the tibial posterior malleolus or the fibular malleolus (Figure 6-29C,D).

FIGURE 6-24. Variation of calcaneal form/length. A: Elongated with normal contours. B: Short with enlarged posterior segment.

FIGURE 6-25. Flattened inferior calcaneal surface (arrowheads).

FIGURE 6-26. Variation of the anterior process: Enlarged and projecting superiorly (arrowheads), simulating an “anteater” calcaneus. COALITION (SYNOSTOSIS/SYNCHONDROSIS/SYNDESMOSIS) Union between two bones may be osseous, cartilaginous, or fibrous. Several terms have been used to describe these enigmas. The term synostosis, defined earlier, pertains to osseous union between two normally distinct bones. The term coalition is used more loosely and can pertain to osseous, cartilaginous (synchondrosis), or fibrous (syndesmosis) union; it is frequently applied to anomalous union of tarsal bones. The term ankylosis refers to consolidation

of a joint because of disease, injury, or surgical procedure. (e.g., Fibrous or bony ankylosis may be seen as an end-stage presentation of inflammatory joint disease.) And fusion is defined as the operative formation of an ankylosis.1 Synostosis of a lesser toe distal interphalangeal joint is the most common synostosis seen in the foot and is almost always an incidental finding. Partial synostosis or bridging may be incidentally seen between metatarsal bones or at metatarsocuneiform joints. Tarsal coalitions, in contrast, are frequently associated with clinical symptomatology. This most likely is a result of faulty biomechanics and of limited range of motion between the affected bones. Superimposition of two adjacent bones may mimic a synostosis. This can easily be distinguished by visually tracing the outer, articular margin of each bone. If their outlines can be identified, there is no synostosis. It is not recommended to use “continuity of the trabeculae across a joint” in determining whether or not a synostosis exists. This finding is often misleading when bones are superimposed on one another. Interphalangeal Joint Synostosis Proximal interphalangeal joint synostosis rarely occurs. However, synostosis of the lesser toe distal interphalangeal joint is frequently encountered in the foot (Figure 6-30). The fifth toe is most commonly affected. In decreasing order of frequency, distal interphalangeal joint synostoses of the fourth, third, and second digit occasionally are seen. Generally they are found bilaterally; and, fourth toe distal interphalangeal joint (DIPJ) synostosis will be accompanied with fifth toe DIPJ synostosis, third toe DIPJ synostosis will be accompanied with fourth and fifth toe DIPJ synostosis, and so on. A lesser digit that is clinically contracted (e.g., hammer toe or mallet toe) radiographically appears as if it lacks a joint space, and simulates a synostosis. However, if the articular margins of the distal phalanx base and middle phalanx head can be visually traced and identified, then there is no synostosis. They are superimposed on each other and will collectively have an increased density relative to the remainder of each bone. The radiodensity of a synostosis is homogeneous and similar to the middle and distal phalanges; no outline for the base or head can be visualized.

Intersesamoid Coalition A rare instance of synostosis between the first metatarsophalangeal joint tibial and fibular sesamoids was reported by Saxby et al.24 in 1992. Intermetatarsal Coalition Partial coalition between two metatarsals (rare) may or may not be entirely osseous and is frequently associated with other anomalies. Metatarsocuneiform Coalition It is extremely rare to see synostosis between metatarsal and tarsal bones, especially the first cuneometatarsal joint.30 When present, they may be partial or complete (Figure 6-31). The incomplete can be fibrocartilaginous and, when so, will appear as osteoarthritis. Tarsal Coalition Synostosis may be seen between any two tarsal bones, but the most commonly encountered coalitions are calcaneonavicular and talocalcaneal.31 Talonavicular, calcaneocuboid, intercuneiform, cubonavicular, and cubocuneiform coalitions are rarely seen (Figure 6-32).

FIGURE 6-27. Variant calcaneal tubercles and tuberosities. A: Enlarged and prominent peroneal tubercle (arrowhead), dorsoplantar view. B: Large peroneal tubercle (arrowhead), axial view (different patient from A). C: Large anterior tuberosity (arrow) (lateral view). Also note the variant increased density that runs between the medial tubercle inferiorly and the Achilles tendon enthesis posteriorly. D: Incomplete hook-shaped lateral tuberosity (arrow) and resultant foramen. In Asian countries, the isolated naviculocuneiform coalition is reported to be relatively common.32 The medial naviculocuneiform coalition, diagnosed

by CT, was plantar (incomplete) and histologically comprised of fibrocartilaginous tissue in a large group of patients; the dorsal half of the joint was normal. Radiographically, the predominant finding is osteoarthritis, which is best seen in the dorsoplantar and lateral oblique views (Figure 633A,B). An incomplete synostosis may also be seen (Figure 6-33C,D). The calcaneonavicular coalition (also known as calcaneonavicular bar) is clearly visible in the medial oblique view (Figure 6-34A,B) and the diagnosis is usually straightforward. The calcaneonavicular coalition can also be recognized in the lateral view, though superimposed, as an extension or elongation of the anterior calcaneal process superiorly (Figure 6-34C); this finding has been referred to as the “anteater nose sign.”33,34 In the literature, this has also been referred to as the “TLAP” or “too long anterior process” of the calcaneus.35 The appearance of the calcaneonavicular coalition in the medial oblique view varies, depending on the type of union. The superomedial aspect of the anterior calcaneal beak and inferolateral aspect of the navicular are continuous as one bony structure if there is osseous coalition or synostosis (Figure 6-34A). However, the fibrous or cartilaginous calcaneonavicular coalition is more frequently encountered. In this case, the two bones are in close apposition and appear to articulate with one another. The apposing margin of each bone may be quite irregular and sclerotic, resembling degenerative arthritis (Figure 6-34B). In contrast, the two bones are occasionally found in close anatomic relationship to one another, yet no obvious articulation or marginal sclerosis is identified. Although this latter presentation represents a variant form, it is questionable whether or not it may be a true calcaneonavicular coalition.

FIGURE 6-28. Subtalar joint variation. A: Absent anterior subtalar joint (arrowheads). B: Separate facet for anterior subtalar joint (arrowheads).

FIGURE 6-29. Distal tibial and fibular variants of form. A: Elongated fibular malleolus (arrowhead). B: Tubercle along medial aspect of tibial malleolus (arrowhead), a bilateral and symmetrical finding. C: Small tubercle along posterior surface of distal tibia (arrowhead), found bilaterally. D: Prominence of the posterior aspect of the distal fibula (arrow), a bilateral finding.

FIGURE 6-30. Interphalangeal joint synostosis: Third-, fourth-, and fifth toe DIPJs.

FIGURE 6-31. Intermediate cuneiform–metatarsal coalition. This finding was bilateral and symmetrical.

FIGURE 6-32. Rare intertarsal synostoses. Talonavicular coalition: A: Lateral view. B: Medial oblique view. Lateral cuneiform–cuboid coalition: C: Medial oblique view; there also is a fibrocartilaginous calcaneonavicular coalition.

FIGURE 6-33. Incomplete medial cuneiform–navicular coalition. Fibrocartilaginous (arrow): A: DP view. B: Lateral oblique view. Synostosis (in a different patient): C: DP view. D: Lateral oblique view. Lysack and Fenton36 have proposed four types of calcaneonavicular morphology that may be encountered in the medial oblique view: (1) the normal presentation where there is a wide gap between the two bones and no suggestion of articulation or union; (2) a short gap between the two bones, similar to an articulation, with widening and flattening of the calcaneus, and the adjacent margins are smooth and well defined (a synchondrosis); (3) a

narrow articulation between the two bones that demonstrates irregular, rough, and poorly defined margins (syndesmosis); and (4) osseous union (synostosis) between the two bones. (See Figures 6-28A,B for examples of types 1 and 2; types 3 and 4 are demonstrated in Figures 6-34B,A, respectively.) It has also been suggested that the os calcaneus secundarius, found at the same site, is related to the formation of the calcaneonavicular coalition.37

FIGURE 6-34. Calcaneonavicular coalition. A: Osseous (arrowheads)

(medial oblique view). B: Fibrocartilaginous (arrowheads) (medial oblique view). C: “Anteater” calcaneus (arrowheads) (lateral view). Based on three-dimensional anatomy using CT, Vidyadhar et al.38 classified calcaneonavicular coalition into four types (the cuboid form—its “medially extending prominence”—is described from a plantar perspective): Type I, the forme fruste, either the cortical tip of the calcaneus was irregular or an ossicle was present between the calcaneus and navicular, accompanied by slight blunting of the cuboid; Type II, the syndesmosis, narrowing of the nonossified gap between the calcaneus and navicular bones accompanied by further blunting of the cuboid extension; Type III, the synchondrosis, accompanied by distinct “squaring off” of the cuboid bone; and Type IV, the synostosis, accompanied by an absent (“no longer visible”) cuboid medial prominence. In their series of 37 patients demonstrating 69 coalitions (32 were bilateral), the distribution for Types I through IV was 28%, 23%, 45%, and 4%, respectively. Identification of a calcaneonavicular coalition is best visualized if the foot is oblique 45° relative to the x-ray image receptor (the medial oblique view). If less than 45° is attained, there is superimposition of the anterior calcaneus and navicular, mimicking a calcaneonavicular coalition. Proper foot position can be determined by examining the relationship of the cuboid and lateral cuneiform in the radiograph. If the two bones are separate and distinct (such that the cuboid is wholly isolated), the oblique position was performed properly. If not, a distinct space is not seen between the two bones; this should not be misinterpreted as coalition.

FIGURE 6-35. Middle talocalcaneal coalition (obliquely oriented) calcaneal axial view. A: Osseous (arrowhead). B: Fibrocartilaginous (arrowhead). The talocalcaneal coalition receives the most attention in the literature of all the tarsal coalitions. Coalition of the middle talocalcaneal joint region is the most frequently encountered of the three possible talocalcaneal coalitions, anterior, middle, and posterior. Anatomically, the middle talocalcaneal coalition is either extra-articular (located between the posterior aspect of the sustentaculum tali and the talar body) and/or intra-articular.18 It may either be complete, nearly complete, incomplete, or rudimentary.39 Varying radiographic findings, therefore, can be attributed to variant presentations.40 Radiographically, the complete middle talocalcaneal coalition is best appreciated in the axial (Harris–Beath) view.41 The joint space is obliterated if osseous coalition is present (Figure 6-35A). If fibrocartilaginous, the

middle talocalcaneal joint will be obliquely oriented (at approximately 45°), accompanied by gross joint space narrowing and subchondral sclerosis (Figure 6-35B). The os sustentaculi may rarely be identified in association with middle talocalcaneal coalition and, like the os calcaneus secundarius, has been suggested to contribute to the formation of coalition.42 Although it is not obvious in the lateral foot view, specific radiographic findings may be identified, leading to its diagnosis. Normally the shadows of the sustentaculum tali and the talar posteromedial process are separate and distinct. In contrast, if osseous middle talocalcaneal coalition is present, the inferior margins of these structures are continuous with one another (Figure 6-36). This finding has been coined the “C sign.”43 Absence of the C sign, however, does not negate diagnosis of coalition, especially in patients that have smaller lesions or are not yet skeletally mature.44 If fibrocartilaginous, one will see an articulation between these two anatomic landmarks that is irregular and sclerotic, mimicking a joint with osteoarthritis. In either case, associated findings include a rounded or flattened talar lateral process inferiorly (it normally is pointed). A spur or “beak” along the superior aspect of the talar neck (at the ridge) has also been associated with talocalcaneal coalition.27 But this latter finding is not a reliable indicator of coalition, because it can be seen when coalition is not present and vice versa. If necessary, CT and MRI can be used to further characterize the coalition’s location, character, and size, especially for preoperative assessment.31

FIGURE 6-36. Middle talocalcaneal coalition: “C sign” (continuity between sustentaculum and talar posteromedial process, marked by arrowheads); associated findings are also seen (rounded lateral talar process and talar head beak). Anterior and posterior talocalcaneal coalitions are rare. The anterior talocalcaneal coalition is identified in the medial oblique view. As already noted with the calcaneonavicular coalition, however, anterior talocalcaneal coalition is mimicked by superimposition of the two bones if the foot is not positioned properly. The posterior talocalcaneal coalition can be seen in the calcaneal axial view; however, CT may be necessary to make the diagnosis. VARIANTS OF DENSITY Radiolucencies Varying degrees of radiolucency may be encountered at several anatomic

locations. They are found in the absence of any clinical symptomatology, appear bilaterally in most instances, and are variations of normal radiographic anatomy. For example, a central radiolucency may be seen in the hallux proximal phalanx diaphysis that mimics a geographic, solitary lytic lesion (Figure 6-37A). Another example of a variant radiolucency is found in the medial aspect of the navicular in the dorsoplantar view, adjacent to the tuberosity (Figure 6-37B). It typically is bilateral and symmetric. A larger area of radiolucency is occasionally encountered in the body of the calcaneus in the lateral view (Figure 6-37C). The arrangement of trabeculae in this area is such that three patterns of stress trabeculae surround an oval or triangular central radiolucency, often referred to as the neutral triangle. Occasionally, the margins of this radiolucency are well defined, simulating a unicameral bone cyst. Other sites where variant radiolucency is encountered include the metatarsal heads and the distal fibula. The medial aspect of the fifth metatarsal head is commonly radiolucent (Figure 6-37D). It simulates inflammatory joint disease, especially early rheumatoid arthritis. However, the subchondral bone plate is well defined and continuous when viewed with magnification. A similar radiolucency may be seen in the medial aspect of the remaining lesser metatarsal heads.

FIGURE 6-37. Variant decreased densities. A: Hallux proximal phalanx diaphysis. B: Medial aspect of navicular. C: Calcaneal neutral triangle. D: Fifth metatarsal head medially and the geographic decreased density in the second metatarsal head (a bilateral presentation).

FIGURE 6-38. Variant “punched-out” radiolucent densities (arrowheads). A: Second and third toe proximal phalanx distal diametaphyses. B: Calcaneal body, central location. C: Calcaneal lateral tuberosity, small. Small, well-defined, and round “punched-out” radiolucencies are occasionally seen. It is not uncommon to see a tiny, round lucency in the distal metaphysis of the lesser toe proximal phalanx that is central in location (Figure 6-38A). A similar finding may be encountered in the calcaneal body, in the “neutral triangle” (Figure 6-38B). Another round radiolucency is infrequently found adjacent to the calcaneal lateral tuberosity in the medial oblique view (Figure 6-38C); this has been referred to in the literature as a nutrient foramen,45 although it could also represent a developmental defect or

incomplete ossification of the apophyseal growth center (compare to Figure 6-27D). Increased Densities A common variant of increased density appears as a solitary, geographic increased (cortical) density found in cancellous bone (Figure 6-39). This tumorlike lesion might be found in any foot bone and has been referred to as a bone island (also known as enostosis and endosteoma). Bone islands are incidental findings and are not associated with clinical symptomatology. A superimposed accessory sesamoid bone may simulate a bone island. Some increased radiodensities correspond to osteologic landmarks. For example, a curvilinear sclerosis may be seen in the hallux distal phalanx base in the dorsoplantar view that corresponds to the anterior margin of the flexor tuberosity (see Figure 6-6). Another sclerotic density is commonly seen in the body of the calcaneus in the lateral view, superior to the medial tuberosity (Figure 6-39C). This curvilinear density corresponds to the inferior margin of the lateral tuberosity; its position and outline vary among patients. Occasionally, the lower one-third of the posterior calcaneus is sclerotic relative to the body (Figures 6-27C and 6-39C). This sclerosis extends from the Achilles tendon enthesis to the medial and/or lateral tubercle inferiorly. It appears to represent accentuation of the stress trabeculae as well as superimposition of the margin of the convex posteroinferior calcaneus.

FIGURE 6-39. Variant increased densities in cancellous bone: Bone island. A: Hallux distal phalanx (arrowhead). B: Second metatarsal head (arrowhead). C: Central calcaneus (arrowhead). The semicircular increased density (arrow) in the medial tubercle is the superimposed lateral tubercle; it continues to the Achilles tendon enthesis posterosuperiorly as prominent trabeculations.

FIGURE 6-40. “Gun-barrel” presentation of distal phalanx due to positioning. A severely contracted digit may be positioned such that its phalangeal shaft is parallel to the central x-ray beam in the dorsoplantar view. The resultant image may appear as an intense, radiodense circle, corresponding to the diaphyseal cortex. This has been coined the “gun barrel sign” because it mimics the appearance seen when looking down the barrel of a gun (Figure 6-40). Superimposed soft tissue structures may present interesting findings. For example, an infrequent finding is the appearance of a well-defined increased density in the anteroposterior ankle view that runs vertically through the center of the ankle joint; its width spans at least half the width of the talar dome and widens superiorly. This shadow corresponds to the superimposed Achilles tendon (Figure 6-41).

FIGURE 6-41. Achilles tendon shadow (outlined by arrowheads) superimposed on distal tibia and talus in anteroposterior ankle view. VARIANTS OF ARCHITECTURE Cortical Bone The cortex normally is radiopaque and homogeneous in density. Both the subperiosteal and endosteal surfaces should be smooth, continuous, and well defined. The combined thicknesses of the mid-diaphyseal cortices (medial and lateral) should equal 50% to 75% of the bone width. It is not uncommon, however, to see variations in the appearance of the cortex. Uniform increased thickness of the cortices may give the appearance of a “disappearing” medullary canal (Figure 6-42A). Also, the contour of a lesser metatarsal’s external margin may be irregular or wavy (but nonuniform) along its subperiosteal surface that can involve one or multiple bones (Figure 6-42B); this finding is predominantly located along the proximal one-half diaphysis and is an incidental finding. The differential diagnosis might include hypertrophic osteoarthropathy and melorheostosis.

FIGURE 6-42. Variant architecture: Cortical bone. A: Symmetrical cortical thickening, periosteal and endosteal, second and third metatarsals. B: Asymmetrical and wavy periosteal, cortical thickening (arrowheads) of the second, third, and fourth metatarsals. Cancellous Bone The trabecular pattern in the distal metaphysis of the first metatarsal is commonly coarse in appearance (Figure 6-43A). These thickened trabeculae are normal and (if a solitary finding) do not represent osteopenia. The superimposition of these trabeculations on the sesamoids can simulate a fracture of the latter. A rare finding is a mosaiclike pattern of trabeculations, which is a generalized finding especially noticeable in the first metatarsal. A similar pattern may involve the phalanges (Figure 6-43B). Transversely oriented trabeculae are occasionally encountered in the

medullary portion of tubular bones (Figure 6-43C). These may be related to bone bars (reinforcement lines) in older patients but may present in younger adults. ACCESSORY OSSICLES Numerous accessory ossicles are present in the foot. Although their anatomic locations are fairly consistent, the morphology of these ossicles can be quite complex and diverse. It is not uncommon to find them unilaterally. When present bilaterally, their size, shape, or number is frequently asymmetric. These presentations certainly can confound interpretation of radiographs, especially when they are found in areas associated with clinical symptomatology. It is, therefore, important that the interpreter be able to recognize the typical and atypical presentations of these enigmas. Os Interphalangeus The os interphalangeus, also referred to as the interphalangeal “sesamoid,” is classically found along the inferior aspect of the hallux interphalangeal joint. It is rare to see this ossicle at the interphalangeal joints of the lesser toes. The position of the os interphalangeus is either central (more common, Figure 644A) or eccentric (Figure 6-44B); it may be round or oval in shape. Rarely, an ossicle may be encountered along the superior aspect of an interphalangeal joint (Figure 6-44C). The centrally located os interphalangeus has long been considered a sesamoid bone because of its location in the plantar capsule that attached to the flexor tendon.16 (McCarthy et al.46 identified its location as in the joint capsule and the flexor hallucis capsularis interphalangeus; however, identification of this muscle has not been reproduced in the literature.47) Recent studies have reported that the ossicle is not a true sesamoid bone: it is intra-articular, positioned along the dorsal surface of the joint capsule that is separated from the tendon by a bursa; when the ossicle is absent, the bursa is not present and the flexor tendon is attached to the capsule.47,48 It also may appear as a pair (Figure 6-44D). The eccentric os interphalangeus appears to have a different genesis from that

of the centrally located ossicle. During development, the basal epiphysis of the distal phalanx occasionally has multiple ossification centers. A segment of this ossification center may remain separate into adulthood and persist as the eccentric os interphalangeus. A defect in the adjacent phalangeal base is frequently observed that corresponds to the size and shape of the unfused ossification center in this case. Rarely, the ossicle is adjacent to a defect along the proximal phalanx head. The eccentric os interphalangeus may also be the sequella of an old, unhealed fracture (i.e., nonunion). However, the majority of these patients do not recall any history of trauma, and many times the radiographic finding is bilateral in presentation.

FIGURE 6-43. Variant architecture: Cancellous bone. A: First metatarsal distal metaphysis. B: Mosaiclike presentation in phalanges. C: Transverse trabeculations (arrows) in first metatarsal diaphysis.

FIGURE 6-44. Os interphalangeus. A: Central location (arrowhead), large (DP view). B: Eccentric position (arrowhead), laterally. C: Rare superior position (arrowhead) isolated in lateral view. D: A pair (arrows). The os interphalangeus can be identified in either the dorsoplantar or isolated lateral view of the hallux. Its transverse-plane position is best determined with the dorsoplantar view. It is superimposed on the proximal phalanx head and appears as a fairly well-defined oval of increased density. The os interphalangeus can be clearly identified with the lateral view if the hallux is isolated from the lesser digits. The lateral or medial eccentric os interphalangeus can be isolated with the medial oblique or lateral oblique view, respectively.

Accessory Sesamoids A pair of sesamoid bones is consistently found at the first metatarsophalangeal joint. This occurs almost without exception. In addition, sesamoid bones may also be found along the inferior aspects of any lesser metatarsophalangeal joint, in varying combinations, and are known as accessory sesamoids (Figure 6-45A). For example, accessory sesamoids may appear at one, two, three, or all four lesser metatarsophalangeal joints as a single entity or in pairs. They appear circular or oval in shape and vary in size; rarely, they are bipartite. They frequently are bilateral, but may be asymmetrically distributed. Accessory sesamoids are best isolated in the sesamoid axial view (Figure 645B). A solitary sesamoid is usually only medial, superimposed on the medial aspect of the metatarsal head in the dorsoplantar view, and seen along the inferomedial aspect of the metatarsophalangeal joint in the axial view. Os Intermetatarseum (Os Intermetatarsale I) The os intermetatarseum is situated superiorly between the first- and second metatarsal bases. Case et al.49 have described three types: freestanding, articulating, and fused. The freestanding os intermetatarseum does not articulate with the first or second metatarsal or the medial cuneiform (Figure 6-46A,B), as does the articulating type. The fused form is most rare and has been described united with either the first or second metatarsal or the medial cuneiform (Figure 6-46C). The os intermetatarseum is best seen in the dorsoplantar view. It can occasionally be seen in the lateral view superiorly, although it typically is superimposed on the first metatarsal base. This ossicle may be round, oval, kidney-shaped, or linear, or may even resemble a rudimentary metatarsal. Its size also varies. Calcification of the perforating branch between the dorsal and plantar metatarsal arteries may simulate an os intermetatarseum. An os intermetatarseum is rarely encountered at the second lesser metatarsocuneiform joint (also known as the os cuneometatarsale II dorsale50) (Figure 6-46D). The os intermetatarseum is infrequently the cause of foot symptomatology.51–54

A sesamoid has been reported in the extensor hallucis brevis tendon at the level of the first metatarsocuneiform joint.55 The lateral radiograph of this lesion looks similar to that of the os intermetatarseum, although the location of the former is slightly more superior. Os Cuneometatarsale I Tibiale The os cuneometatarsale I tibiale is located adjacent to the first metatarsocuneiform joint medially (Figure 6-47).56 The os cuneometatarsale I tibiale plantare is a different ossicle that is located inferior to the joint.5 When present, it may be seen with the lateral view; however, due to its small size and superimposition by other structures, it might be difficult to visualize. Tibialis Anterior Tendon Sesamoid A sesamoid may infrequently be found in the tibialis anterior tendon near its insertion onto the first metatarsal base (Figure 6-48). It is located adjacent to either the first metatarsocuneiform joint or the medial cuneiform57 and is best seen in the dorsoplantar and, if located adjacent to the insertion site, lateral oblique views. More specifically, O’Rahilly56 describes this sesamoid as being adjacent to the facet along the cuneiform anteromedially for the tendon. This sesamoid may be either round, oval, or linear in shape. It may also be bipartite.

FIGURE 6-45. Accessory sesamoids. A: Dorsoplantar view (arrowheads identify multiple accessory sesamoids: one at the second, third, and fourth metatarsophalangeal joints; a pair is seen at the fifth metatarsophalangeal joint). B: Sesamoid axial view, different patient.

FIGURE 6-46. Os intermetatarseum (arrowheads). A: Long, linear. B: Large, triangular shape isolated in lateral view. C: Fused form. D: Rare position at second metatarsocuneiform joint. Os Paracuneiforme (Os Cuneonaviculare Mediale) This rare ossicle has been described or illustrated as being located either medial to the medial cuneiform or to the naviculocuneiform 1 joint (Figure 649).56,58 Os Intercuneiforme

This very rare ossicle has been reported along the posterosuperior aspect of the medial cuneiform.59 Dwight60 and Geist61 describe this ossicle as being small and wedge-shaped, located between the medial and intermediate cuneiforms (Figure 6-50). Os Uncinatum (Os Unci) The os uncinatum ossicle was identified inferior to the lateral cuneiform, adjacent to the third metatarsal.56 It may be free or appear as a hooklike prominence off the lateral cuneiform. Os Cuboid Zimmer observed an ossicle between the anterior aspects of the cuboid and lateral cuneiform, adjacent to the fourth metatarsal base (Figure 6-51).62 He questioningly names it “os cuboid?” in the figure caption, for lack of a better term. This rare entity appears to provide an articular surface for the fourth metatarsal base medially.

FIGURE 6-47. Os cuneometatarsale I tibiale (arrowhead).

FIGURE 6-48. Tibialis anterior tendon sesamoid (arrowhead): Adjacent to medial surface of medial cuneiform. Os Cuboideum Secundarium This extremely rare ossicle is found in the plantar aspect of the foot between the navicular and lateral cuneiform posteroinferiorly; proximally it articulates with the talus and calcaneus.63 It may also have contact with the navicular. Only three have been reported in recent literature.64–66 Even though the original description refers to its position between the navicular and cuboid inferiorly,63 the cases reported by Gaulke and Schmitz64 and Kaufman and Stacy65 show the ossicle directly beneath the cuboid and not near the navicular.

FIGURE 6-49. Os Paracuneiforme (arrow).

FIGURE 6-50. Os intercuneiforme (arrow) (bilateral and symmetrical). Os Vesalianum There has been debate in the literature as to whether or not the small ossicle (or “sesamoid”) located at the tip of the fifth metatarsal tuberosity briefly mentioned by Vesalius67 in his 1543 text was truly accessory or

supernumerary and not the remnant of prior trauma or simply the normally occurring apophysis.56,68–70 In a paper by Burman and Lapidus,71 their “impression from the study of the literature is that there is no bone that can be called os vesalianum.” In 1928, Holland70 stated that the apophysis “is a good example of the bone which has so often been wrongly described as the Bone of Vesalius.” For example, Geist61 shows two cases of children with open physes, and he identifies the apophysis as the os Vesalii.

FIGURE 6-51. Os cuboid (arrowhead). Holland70 describes three different ossicles that can occur at this site: the true epiphysis (or apophysis) seen only in childhood; a separate ossification that persists into adulthood and makes up nearly the entire tuberosity; and a small ossicle at the tip of the existing tuberosity. He proposes two options: either all three ossicles be referred to as the “Bone of Vesalius” or the term be restricted to a small ossicle at the tip of the tuberosity. The literature rarely ever refers to the true apophysis as an os vesalianum; therefore, the persistent apophysis should not be referred to as the os vesalianum either. Kohler and Zimmer state that the persistent apophysis of the fifth metatarsal tuberosity “does not represent a true os vesalianum,”62 an opinion also echoed by Birkner.50 Furthermore, Sarrafian states that the os vesalianum be differentiated from four entities: the ossifying apophysis, fracture, the ununited apophysis, and the sesamoid within the peroneus longus tendon (os peroneum).16 He further notes, “The os vesalianum is located just proximal to the tip of the well-developed tuberosity of the fifth metatarsal.” Based on this survey of the literature, the ossicle located at the posterior tip of the fifth metatarsal tuberosity is known as the os vesalianum (Figure 6-52). The presence of this ossicle is quite rare, and it typically presents as a small, rounded calcific density. It is best seen in the medial oblique view. Ossification of the peroneus brevis tendon and old, unhealed (nonunion) avulsion fracture of the tuberosity tip may look similar to the os vesalianum. Persistent Fifth Metatarsal Apophysis The fifth metatarsal tuberosity apophyseal ossification center may remain separate into adulthood (Figure 6-53). This is known as a persistent apophysis and is occasionally misidentified as the os vesalianum.72 It is large in size and appears to articulate with the metatarsal base. It is clearly identified with the medial oblique view and is frequently bilateral and symmetrical.

FIGURE 6-52. Os vesalianum (arrowhead) (lateral view).

FIGURE 6-53. Persistent fifth metatarsal apophysis (arrowhead). Os Peroneum The os peroneum is a sesamoid bone found in the peroneus longus tendon. It varies not only in size, but also in number. It commonly is partite. The os peroneum is classically situated beside the cuboid bone just proximal to where the tendon runs along the peroneal sulcus, but its position varies considerably. It is best isolated in the medial oblique view (Figure 6-54). The os peroneum generally is superimposed on the cuboid in the lateral and dorsoplantar views. However, it may be identified in the lateral view if its position anatomically is more distal in the tendon; at this location, it articulates with the anterior aspect of the cuboid’s inferior tuberosity. The os peroneum may infrequently be found at a more proximal location, adjacent to the calcaneocuboid joint or anterior calcaneus. This latter entity may easily be misinterpreted as an avulsion fracture. Location of the os peroneum adjacent to the calcaneus as opposed to the cuboid may be variation73 or may indicate a ruptured peroneal tendon with posterior sesamoid displacement.74 Recent literature has reported friction syndrome complicated by fatigue fracture,75 tendon rupture through a bipartition,76 and degenerative joint disease77 of the os peroneum.

FIGURE 6-54. Os peroneum (arrowhead): Typical size and location adjacent to cuboid. Os Infranaviculare An ossicle uncommonly found along the superior aspect of the intermediate cuneiform–navicular joint is the os infranaviculare (Figure 6-55A).62 It has also been referred to as the os naviculocuneiforme I dorsale and os paracuneiforme I.78 It is best identified in the lateral view. Another ossicle is seen infrequently along the superomedial aspect of the medial cuneiform–navicular joint in the medial oblique view (Figure 6-55B). This is near the location of the os intercuneiforme (see Figure 6-50). Kim and Roh79 recently reported a large ossicle located superior to the medial cuneiform that articulates with the navicular, medial, and intermediate cuneiforms. Though its location was similar to that of the os intercuneiforme, they named it as os infranaviculare because of its direct attachment to the navicular. Os Supranaviculare (Os Talonaviculare Dorsale) The os supranaviculare is seen along the superior aspect of the talonavicular joint in the lateral view (Figure 6-56). This ossicle has a multitude of configurations. It may appear as an entirely separate ossicle, as a continuation of the articular subchondral bone, or even as an attachment to the navicular. As with many of the accessory ossicles, it may be impossible to differentiate the os supranaviculare from an old, nonunion fracture. An ossicle may infrequently be identified in the medial oblique view along the superomedial aspect of the talonavicular, possibly representing a variant position of the os supranaviculare. Ingalls and Wissman80 have, by CT, demonstrated in two cases that the os supranaviculare sits in a small depression or “cortical notch.” They discuss the association of an incomplete fracture line (stress fracture) that originated from this notch and suggest further study.

Os Supratalare The accessory ossicle located along the superior surface of the talar head is known as the os supratalare (Figure 6-57). It typically is located over the ridge along the talar head/neck but may be seen distally over the head. It easily can simulate an old, nonunion avulsion fracture and is only identified in the lateral view. Calcification of the talonavicular ligament may occasionally appear in the same location and mimic the appearance of an os supratalare.

FIGURE 6-55. Os infranaviculare. A: DP view (arrowhead). B: Medial oblique view (arrow) (different patient).

FIGURE 6-56. Os supranaviculare (arrowhead). Accessory Navicular (Os Tibiale Externum, Os Naviculare) An ossicle of varying size, shape, and position may be found adjacent to the navicular tuberosity. It is best identified in the lateral oblique view. Melamed81 recently suggested a modification of the Harris–Beath method to visualize the accessory navicular by rotating the toes inward. The ossicle that is found adjacent to the navicular tuberosity may either represent a sesamoid in the posterior tibial tendon or an accessory ossification center for the navicular tuberosity. The sesamoid characteristically is round, small, and located at a distance at least 3 mm from the navicular tuberosity.82 This is the true os tibiale externum.83 The accessory ossification center, in contrast, is larger, characteristically triangular or heart-shaped, and in close apposition to the tuberosity. Cartilage or fibrocartilage may attach it to,84 or

it may articulate with, the tuberosity, containing true synovial tissue.85 Occasionally, the accessory ossification center is fused to the tuberosity. This latter instance has been referred to as a “wraparound” navicular, cornuate or cornuted navicular, gorilloid navicular, and the Kidner foot type. With so many varied presentations, one can see why there has been discrepancy in the literature as to the definition and identification of this entity.

FIGURE 6-57. Os supratalare (arrowhead). Based on these three distinct types of ossicles identified adjacent to the navicular tuberosity, Lawson84 has classified them as follows:    Accessory navicular type I: sesamoid in the tendon (Figure 6-58A)    Accessory navicular type II: articulating accessory ossification center

(Figure 6-58B)    Accessory navicular type III: fused accessory ossification center (Figure 6-58C)

FIGURE 6-58. Accessory navicular. A: Type I: Bipartite sesamoid (arrowheads) in tendon. B: Type II (arrow). C: Type III (arrowhead).

FIGURE 6-59. Type II accessory navicular subtypes. A: Type IIA. B: Type IIB. This classification system best distinguishes between the varying forms of this enigma. It has also been correlated to their cross-sectional imaging

appearances on CT and MRI.86 The term os tibiale externum should be used cautiously and applied only to the sesamoid entity (type I).87 The accessory navicular type I may be partite. The partition could be either transverse or longitudinal. Sella and Lawson88 modified this classification by dividing the type II accessory navicular into two separate entities based on two criteria in the lateral view: (1) the accessory navicular position; and (2) the SOT (synchondrosis–ossicle–talar) angle. The two axes forming the SOT angle are (1) a line running through the synchondrosis and (2) a line that connects the inferior surface of the navicular to the inferior tip of the lateral talar process. In Type IIa, the TOC angle averages 56° (range 50°–70°) (Figure 6-59A). In Type IIb, the accessory navicular sits in a more inferior position and the SOT angle averages 21.5° (range 10°–35°) (Figure 6-59B). Huang et al.89 further subdivided each type into three subtypes; however, they only show noncropped images (making it difficult to see the accessory bone), and there are no accompanying descriptions. The articulation between the accessory navicular type II and navicular may be quite irregular and sclerotic. These findings most likely represent degenerative joint disease and can simulate the appearance of a hypertrophic nonunion fracture. Of all types, the Type II accessory navicular has been most associated with the most symptomatology, 70% according to Chiu et al.90 Bone scintigraphy has been advocated for distinguishing between symptomatic and asymptomatic accessory navicular.91 MRI is also valuable for assessing the symptomatic type II.92,93 Os Calcaneus Secundarius The os calcaneus secundarius is best seen in the medial oblique view, adjacent to and in close apposition to the calcaneal anterior process along its superomedial surface (Figure 6-60). Anatomically, the adjacent calcaneus has a crescent-shaped notch in the anterior calcaneal facet for the ossicle,94 which may or may not be appreciated radiographically. The ossicle is located centrally between the anterior calcaneus, talar head, cuboid, and navicular and appears to articulate with them, especially when larger in size. It is not

clearly seen in the lateral view because it is superimposed on the calcaneus and talus. The os calcaneus secundarius can be mistaken for a fracture of the anterior calcaneal process.95 MRI may be valuable for differentiating fracture from the variant ossicle.96 In other cases, it may be that the ossicle itself is symptomatic; bone scintigraphy has also been used to support this.97

FIGURE 6-60. Os calcaneus secundarius (arrow). Os Sustentaculi This rare ossicle, also known as the os sustentaculum tali, has been identified anatomically along the posterior aspect of the sustentaculum tali (Figure 661). It is best visualized with the anteroposterior ankle and calcaneal axial (Harris–Beath) views.98 Another rare condition that simulates this ossicle is known as the “articulatio talocalcanea”99 or assimilated os sustentaculum tali,62 which represents an accessory joint between enlargements of the sustentaculum tali (posteriorly) and adjacent talus (medial tubercle).28 Talocalcaneal coalition has been attributed to fusion of the os sustentaculi with the talus and calcaneus.41,42 MRI has been used to evaluate the symptomatic os sustentaculi.100

FIGURE 6-61. Os sustentaculum tali (arrowhead). Os Trochleare Calcanei Rarely, an ossicle is found along the lateral aspect of the peroneal tubercle (or trochlea), adjacent to the peroneal tendons (Figure 6-62). It also has been referred to as the accessory calcaneus,101,102 os calcaneus accessorius, and os talocalaneare laterale.5,56,69 It may be seen in the dorsoplantar foot, anteroposterior ankle, or calcaneal axial views along the lateral surface of the anterior calcaneus, adjacent to the peroneal tubercle. Larger presentations also position the ossicle adjacent to the talocalcaneal joint. It may appear similar to a talus secundarius or an os subfibulare, which are located above the level of the talocalcaneal joint in the ankle AP view.102

FIGURE 6-62. Os trochleare calcanei (arrow).

FIGURE 6-63. Os subcalcis (arrow). Os Talus Secundarius Rarely reported in the literature, the os talus secundarius is found along the lateral surface of the talus in the anteroposterior ankle view.103,104 It may be attached to the talus by a synchondrosis or synostosis. If large enough, it may also articulate with the calcaneus and fibula. Os Subcalcis

Another rare ossicle, the os subcalcis, may be seen along the inferior aspect of the calcaneal tuberosity. This probably represents an unfused ossification center for the calcaneal apophysis. When it is located beneath the medial tuberosity, it is seen in the lateral view. However, when closely associated with the lateral tuberosity, it is clearly identified with the medial oblique view (Figure 6-63). Os Aponeurosis Plantaris An ossicle or ossicles located in the plantar fascia have been referred to as the os aponeurosis plantaris (Figure 6-64).32 This entity may or may not be related to trauma (myositis ossificans). Os Trigonum An accessory ossification center can be found along the posterior aspect of the talar posterolateral process (Figure 6-65). When fused to the talus, the elongated posterior process is known as the trigonal (Stieda’s) process. If it remains unfused and separate, it is known as the os trigonum. In either case, its inferior surface typically articulates with the calcaneus105 or the posterior subtalar joint.106 The os trigonum may have a fibrous, fibrocartilaginous, or cartilaginous attachment to the talus.16,107 An articulation may be identified between it and the posterior talus. Occasionally, it may exhibit findings suggesting osteoarthritis. The size of this ossicle ranges from small to large. It is best seen in the lateral view but is infrequently viewed in the medial oblique view (the leg must be perpendicular to the foot to prevent superimposition by the fibula). Paulos et al.108 recommend a subtalar oblique technique performed at 30° to help differentiate between os trigonum and acute fracture.

FIGURE 6-64. Os aponeurosis plantaris (arrowheads).

FIGURE 6-65. Os trigonum (arrowhead). Bone scintigraphy may be a valuable diagnostic study for differentiation between symptomatic and asymptomatic os trigonum.84,109 Focal intense uptake suggests degenerative disease and/or unhealed fracture. CT110 and MRI,111 the technique of choice, have been used to assess the symptomatic os trigonum or to differentiate the trigonal process from fracture. Dynamic fluoroscopy and ultrasound have also been used.106 Talus Partitus (Bipartite Talus) Schreiber et al.112 report a case that divides the talar body into anterior and posterior segments; they believe that it may be a secondary ossification center for the talus.112 It resembles a large os trigonum and talar body fracture.5 It

has been reported with talar fracture.113 Differential diagnosis for the os trigonum includes the talus partitus.114

FIGURE 6-66. Os accessorium supracalcaneum (arrow) posterior to an os trigonum.

Os Accessorium Supracalcaneum The os accessorium supracalcaneum is an extremely rare ossicle. It is found along the superior surface of the posterior calcaneus. It can be mistaken for the os trigonum; however, the os supracalcaneum is not in direct apposition to the posterior talar process. The os supracalcaneum, when present, may be large and easily recognized in the lateral view. It has been reported to be attached to the Achilles tendon and have a bony bridge with the calcaneus.115 Occasionally, it is also found with an os trigonum just anterior to it (Figure 666). Os Talotibiale An accessory ossicle infrequently seen along the anterior aspect of the ankle joint116,117 in the lateral view (Figure 6-67) is the os talotibiale. It is found in close apposition to the most anteroinferior aspect of the tibia. It can easily be mistaken as a loose osseous body. However, the latter entity is usually associated with other degenerative joint findings. Rarely, an ossicle may present along the posterior aspect of the ankle joint. This unnamed ossicle, like the os talotibiale, is closely apposed to the tibia. Os Subtibiale The os subtibiale is an ossicle found just inferior to the tibial malleolus (Figure 6-68). It varies considerably in size and shape. Coral118 states that the true os subtibiale is distinct and separate from the anterior and posterior colliculi of the medial malleolus, and can be mistaken for the following conditions: unfused malleolar accessory ossification center, loose osseous body, acute or nonunion avulsion fracture, and deltoid ligament ossification. Reports of the os subtibiale misinterpreted as fracture has been reported,119,120 and it also has been related to posterior tibial tendon dysfunction.121 The cases reported by Coral118,120 are directly apposed to the posterior colliculus of the tibial malleolus. In contrast, malleolar avulsion fractures involve the anterior colliculus. This relationship is best appreciated in the lateral view. The os subtibiale and avulsion fracture may be impossible to differentiate in the AP and mortise ankle views alone.

FIGURE 6-67. Os talotibiale (arrowhead).

FIGURE 6-68. Os subtibiale (arrow); probably an unfused malleolar accessory ossification center. Bellapianta et al.122 recently reported large, bilateral os subtibiale associated with middle talocalcaneal coalition. However, since both ossicles appear to encompass both colliculi and are more anteriorly positioned, these are more than likely examples of the unfused malleolar accessory ossification center. Coral118 believes that small ossicles inferior to the tip of the anterior colliculus are also unfused secondary ossification centers, if not related to prior injury. Os Subfibulare Several ossicles may be found adjacent and inferior to the fibular malleolus; they are sometimes all generically referred to as the os subfibulare. However, Coughlin5 states that the true os subfibulare is positioned posterior to the fibular malleolus and should be differentiated as such from the unfused apophyseal ossification center that is located anteriorly, and the unfused avulsion fracture123 versus unfused accessory ossification124 that may be seen inferiorly (Figure 6-69). Champagne et al.125 found the posteriorly positioned os subfibulare encased within both peroneal tendons; they agree with Coughlin that the term os subfibulare be reserved for the posteriorly positioned ossicle, but also that it be considered a sesamoid. The os subfibulare is seen in both the AP and mortise ankle views. It is not readily visible in the lateral view but can appear as a fairly well-defined increased density superimposed on the talus. One can determine the anterior versus posterior position with the lateral view, and may see the smaller ossicle inferior to the tip in some instances. A large accessory ossification center for the fibular malleolus may be encountered in the same location as the os subfibulare. It is directly related to ankle joint instability because of insertion of the calcaneofibular and/or anterior talocalcaneal ligaments and abnormal movement between the accessory ossification center and distal fibula.126

FIGURE 6-69. Os subfibulare (arrow); probably an unfused apophyseal ossification center. A: Anteroposterior view. B: Lateral view, identified along the anteroinferior margin of the fibular malleolus.

FIGURE 6-70. Os retinacula (arrow). Os Retinaculi An ossicle may be seen along the lateral aspect of the fibular malleolus near the peroneal retinaculum (Figure 6-70); this ossicle is known as the os retinacula.127

Accessory Talus (Talus Accessorium) Feeney et al.128 reported a large ossicle along the dorsomedial aspect of the talus in a 7-year-old female who also had an accessory toe removed from the first metatarsal base. The extra tarsal ossicle was believed to be a duplicate talus.

FIGURE 6-71. Dorsoplantar view of accessory ossicles. (See Box 6-2 for labels.) SUMMARY Figures 6-71 through 6-74 illustrate the accessory ossicles in the views that they would most likely be seen.

FIGURE 6-72. Lateral view of accessory ossicles. (See Box 6-2 for labels.)

FIGURE 6-73. Medial oblique view of accessory ossicles. (See Box 6-2 for labels.)

FIGURE 6-74. Anteroposterior view of accessory ossicles. (See Box 6-2 for labels.) BOX 6-2 Labels Used in Figures 6-71 through 6-74. an as oap oc ocm1tp ocm1t ocm2d

Accessory navicular Accessory sesamoid Os aponeurosis plantaris Os cuboid Os cuneometatarsale I tibiale Plantare Os cuneometatarsale I tibiale Os cuneometatarsale II dorsale

ocs oic oim oin oip op opc or osbt osbf osc osn ospc ost osta ot otc ott ov p5ma tats

Os calcaneus secundarius Os intercuneiforme Os intermetatarseum Os infranaviculare Os interphalangeus Os peroneum Os paracuneiforme Os retinacula Os subtibiale Os subfibulare Os subcalcis Os supranaviculare Os supracalcaneum Os sustentaculum tali Os supratalare Os trigonum Os trochleare calcanei Os talotibiale Os vesalianum Persistent 5th metatarsal apophysis Tibialis anterior tendon sesamoid

REFERENCES

  1. Dorland’s Illustrated Medical Dictionary. 31st ed. Philadelphia, PA: Saunders; 2007.   2. Hlavac HF. Differences in x-ray findings with varied positioning of the foot. J Am Podiatry Assoc. 1967;57(10):465.   3. lnge GAL, Ferguson AB. Surgery of the sesamoid bones of the great toe. Arch Surg. 1933;27:466–488.   4. Karadaglis D, Grace D. Morphology of the hallux sesamoids. Foot Ankle Surg. 2003;9:165.   5. Coughlin MJ. Sesamoids and accessory bones of the foot. In: Coughlin MJ, Mann RA, Saltzman CA, ed. Surgery of the Foot & Ankle. 8th ed. St Louis, MO: Mosby; 2006:531–610.   6. Anwara R, Anjumb SN, Nicholl JE. Sesamoids of the foot. Curr Orthop. 2005;19:40.   7. Burnett SE, Case DT. Bipartite medial cuneiform: new frequencies from skeletal collections and a meta-analysis of previous cases. Homo. 2011;62:109.   8. Panu A, Konin G, Saboeiro G, et al. Symptomatic bipartite medial cuneiform treated with fluoroscopic and ultrasound-guided injections. HSS J. 2014;10:92.   9. Azurza K, Sakellariou A. ‘Osteosynthesis’ of a symptomatic bipartite medial cuneiform. Foot Ankle Int. 2001;22:499.  10. Mohiuddin T, Jennison T, Damany D. Müller–Weiss disease: review of current knowledge. Foot Ankle Surg. 2014;20(2):79–84.  11. Haller J, Sartoris DJ, Resnick D, et al. Spontaneous osteonecrosis of the tarsal navicular in adults: imaging findings. AJR Am J Roentgenol. 1988;151:355.  12. Maceira E, Rochera R. Müller–Weiss disease: clinical and biomechanical features. Foot Ankle Clin. 2004;9:105.

 13. Brailsford JF. Osteochondritis of the adult tarsal navicular. J Bone Joint Surg. 1939;21(1):111.  14. Tanaka Y, Takakura Y, Omokawa S, et al. Crank-shaped arthrodesis for a flatfoot with a bipartite navicular: a case report. Foot Ankle Int. 2006;27:734.  15. Sizensky JA, Marks RM. Imaging of the navicular. Foot Ankle Clin. 2004;9:181.  16. Kelikian AS. Sarrafian’s Anatomy of the Foot and Ankle. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2011.  17. Shawdon A, Kiss ZS, Fuller P. The bipartite tarsal navicular bone: radiographic and computed tomography findings. Australas Radiol. 1995;39:192.  18. Day F, Jones PC, Gilbert CL. Congenital absence of the tibial sesamoid. J Am Podiatr Med Assoc. 2002;92:153.  19. Winter WG, Iwerson LJ, Johnson ED. Lateral supporting ligament of the distal phalanx. Foot Ankle. 1989;9(6):310.  20. Shrewsbury M, Johnson RK. The fascia of the distal phalanx. J Bone Joint Surg Am. 1975;57A(6):784.  21. Lemont H, Christman RA. Subungual exostosis and nail disease and radiologic aspects. In: Scher RK, Daniel CR, ed. Nails: Therapy, Diagnosis, and Surgery. 2nd ed. Philadelphia, PA: Saunders; 1997.  22. Lee M, Hodler J, Haghighi P, et al. Bone excrescence at the medial base of the distal phalanx of the first toe: normal variant, reactive change, or neoplasia? Skeletal Radiol. 1992;21:161.  23. Sutro CJ, Sutro WH. Spool-shaped proximal pedal phalanges. Bull Hosp Jt Dis. 1986;46:122.  24. Saxby T, Vandemark RM, Hall RL. Coalition of the hallux sesamoids: a

case report. Foot Ankle. 1992;13(6):355.  25. Bhargava KN, Malaviya G. Variations in the second metatarsal bone of man. Indian J Med Sci. 1961;15:100.  26. Morton DJ. Metatarsus atavicus. J Bone Joint Surg. 1927;9:531.  27. Resnick D. Talar ridges, osteophytes, and beaks: a radiologic commentary. Radiology. 1984;151:329.  28. Bloom RA, Libson E, Lax E, et al. The assimilated os sustentaculi. Skeletal Radiol. 1986;15:455.  29. Ruiz JR, Christman RA, Hillstrom HJ. Anatomical considerations of the peroneal tubercle. J Am Podiatr Med Assoc. 1993;83(10):563.  30. Takakura Y, Nakata H. Isolated first cuneometatarsai coalition: a case report. Foot Ankle Int. 1999;20(12):815.  31. Lemley F, Berlet G, Hill K, et al. Current concepts review: tarsal coalition. Foot Ankle Int. 2006;27(12):1163.  32. Kumai T, Tanaka Y, Takakura Y, et al. Isolated first naviculocuneiform joint coalition. Foot Ankle Int. 1996;17:635.  33. Oestreich AE, Mize WA, Crawford AH, et al. The “anteater nose”: a direct sign of calcaneonavicular coalition on the lateral radiograph. J Pediatr Orthop. 1987;7:709.  34. Chapman VM. The anteater nose sign. Radiology. 2007;245(2):604.  35. Pouliquen JC, Duranthon LD, Glorion C, et al. The too-long anterior process calcaneus: a report of 39 cases in 25 children and adolescents. J Pediatr Orthop. 1998;18:333.  36. Lysack JT, Fenton PV. Variations in calcaneonavicular morphology demonstrated with radiography. Radiology. 2004;230:493.  37. Pfitzner W. Beiträge zur Kenntniss des menschlichen

Extremitätenskelets. VII. Die Variationen im Aufbau des Fusskelets. In: Schwalbe G, ed. Morphologischen Arbeiten. Vol 6. Jena: Fischer; 1896. Cited by: Webster FS, Roberts WM. Tarsal anomalies and peroneal spastic flatfoot. JAMA. 1951;146:1099.  38. Vidyadhar VU, Chambers RC, Mubarak SJ. Analysis of calcaneonavicular coalitions using multi-planar three-dimensional computed tomography. J Child Orthop. 2008;2:301.  39. Harris RI. Follow-up notes on articles previously published in the journal. J Bone Joint Surg Am. 1965;47A(8):1657.  40. Linklater J, Hayter L, Vu D, et al. Anatomy of the subtalar joint and imaging of talo-calcaneal coalition. Skeletal Radiol. 2009;38:437.  41. Harris RI, Beath T. Etiology of peroneal spastic flatfoot. J Bone Joint Surg Br. 1948;30B:624.  42. Webster FS, Roberts WM. Tarsal anomalies and peroneal spastic flatfoot. JAMA. 1951;146:1099.  43. Lateur LM, Van Hoe LR, Van Ghillewe KV, et al. Subtalar coalition: diagnosis with the C sign on lateral ankle radiographs of the ankle. Radiology. 1994;193:847.  44. Taniguchi A, Tanaka Y, Kadono K, et al. C Sign for diagnosis of talocalcaneal coalition. Radiology. 2003;228:501.  45. Magilner AD. Normal roentgen variant: calcaneal nutrient canal. Radiology. 1980;136:10.  46. McCarthy DJ, Reed T, Abell N. The hallucal interphalangeal sesamoid. J Am Podiatr Med Assoc. 1986;76(6):311.  47. Davies MB, Dalal S. Gross anatomy of the interphalangeal joint of the great toe: implications for excision of plantar capsule accessory ossicles. Clin Anat. 2005;18:239.  48. Suwannahoy P, Srisuwan T, Pattamapaspong N, et al. Intra-articular

ossicle in interphalangeal joint of the great toe and clinical implication. Surg Radiol Anat. 2012;34:39.  49. Case DT, Ossenberg NS, Burnett SE. Os intermetatarseum: a heritable accessory bone of the human foot. Am J Phys Anthropol. 1998;107(2):199.  50. Birkner R. Normal Radiologic Patterns and Variances of the Human Skeleton. Baltimore, MD: Urban & Schwarzenberg; 1978.  51. Reichmeister JP. The painful os intermetatarseum: a brief review and case reports. Clin Orthop. 1980;153:201.  52. Nakasa T, Fukuhara K, Adachi N, et al. Painful os intermetatarseum in athletes: report of four cases and review of the literature. Arch Orthop Trauma Surg. 2007;127:261.  53. Kose IC, Hizal M, Bulut EG, et al. Bilateral fused os intermetatarseum presenting as dorsal foot pain: a case report. Surg Radiol Anat. http://dx.doi.org/10.1007/s00276-013-1203-4. Accessed February 16, 2014.  54. Giordano M. Compression of the medial branch of the deep peroneal nerve, relieved by excision of an os intermetatarseum. A case report. Acta Orthop Belg. 2003;69(6):568.  55. Johnston FE, Lee RH. Anomalous accessory sesamoids associated with saddle deformity. Curr Podiatr Med. Jan/Feb 1991:13.  56. O’Rahilly R. A survey of carpal and tarsal anomalies. J Bone Joint Surg Am. 1953;35A:626.  57. Haliburton RA, Butt EG, Barber JR. The tibialis anterior sesamoid. Can J Surg. 1961;4:480.  58. Morrison AB. The os paracuneiforme. Some observations on an example removed at operation. J Bone Joint Surg. 1953;35B(2):254.  59. Prescher A. Some remarks on, and a new case of the rare os intercuneiforme (Dwight). Ann Anat. 1997;179:317.

 60. Dwight T. Variations of the Bones of the Hands and Feet. Philadelphia, PA: Lippincott; 1907.  61. Geist ES. Supernumerary bones of the foot—a roentgen study of one hundred normal individuals. Am Orthopedic Surg. 1914;12:403.  62. Schmidt H, Freyschmidt J. Kohler/Zimmer Borderlands of Normal and Early Pathologic Findings in Skeletal Radiography. 4th ed. New York, NY: Thieme Medical Publishers; 1993.  63. Dwight T. Description of a free cuboides sucundarium, with remarks on that element and on the calcaneus secundarius. Anat Anz. 1910;37:218.  64. Gaulke R, Schmitz HG. Free os cuboideum secundarium: a case report. J Foot Ankle Surg. 2003;42(4):230.  65. Kauffmann G, Stacy GS. Os cuboideum secundarium: a rare accessory ossicle with the potential to mimic a mass on magnetic resonance imaging. Skeletal Radiol. 2014;43:95.  66. Logan PM, Connell DG, Janzen DL. Painful os cuboideum secundarium: cross-sectional imaging findings. J Am Podiatr Med Assoc. 1996;86:123.  67. Vesalius A. De Humani Corporis Fabrica (“On the Fabric of the Human Body”). Book One (The things that sustain and support the entire body, and what braces and attaches them all. [the bones and the ligaments that interconnect them]). Chapter 33 (On the Bones of the Foot); 1543. http://vesalius.northwestern.edu/. Accessed June 7, 2014.  68. Bizarro AH. On sesamoid and supernumerary bones of the limbs. J Anat. 1920;55:256–268.  69. Trolle D. Accessory Bones of the Human Foot. Copenhagen, Denmark: Einar Munksgaard, 1948.  70. Holland CT. The accessory bones of the foot with notes on a few other conditions. In: The Robert Jones Birthday Volume. A Collection of Surgical

Essays. London, England: Oxford University Press; 1928:157–182,.  71. Burman MS, Lapidus PW. The functional disturbances caused by the inconstant bones and sesamoids of the foot. Arch Surg. 1931;22:936.  72. Shands AR. The accessory bones of the foot. South Med Surg. May 1931:326.  73. Bloom RA. The infracalcaneal os peroneum. Acta Anat. 1991;140:34.  74. Thompson FM, Patterson AH. Rupture of the peroneus longus tendon. Report of three cases. J Bone Joint Surg Am. 1989;71A:293.  75. Bashir WA, Lewis S, Cullen N, et al. Os peroneum friction syndrome complicated by sesamoid fatigue fracture: a new radiological diagnosis? Skeletal Radiol. 2009;38:181.  76. Blitz NM, Nemes KK. Bilateral peroneus longus tendon rupture through a bipartite os peroneum. J Foot Ankle Surg. 2007;46(4):270.  77. Muehleman C, Williams J, Bareither Ml. A radiologic and histologic study of the os peroneum: prevalence, morphology, and relationship to degenerative joint disease of the foot and ankle in a cadaveric sample. Clin Anat. 2009;22:747.  78. O’Rahilly R. Developmental deviation in the carpus and tarsus. Clin Orthop. 1957;10:9.  79. Kim JK, Roh KJ. Symptomatic os infranaviculare. Clin Orthop Surg. 2013;5(2):152.  80. Ingalls J, Wissman R. The os supranaviculare and navicular stress fractures. Skeletal Radiol. 2011;40:937.  81. Melamed EA. Intoeing Harris view for accessory navicular visualization and hindfoot alignment: technique tip. Foot Ankle Int. 2010;31(12):1122.  82. Bareither DJ, Muehleman CM, Feldman NJ. Os tibiale externum or

sesamoid in the tendon of tibialis posterior. J Foot Ankle Surg. 1995;34(5):429.  83. Du Vries HL. Surgery of the Foot. 4th ed. Mann RA, ed. St Louis, MO: Mosby; 1978:105.  84. Lawson JP. Symptomatic radiographic variants in extremities. Radiology. 1985;157:625.  85. Lemont H, Travisano VL, Lyman J. Accessory navicular. Appearance of a synovial joint. J Am Podiatry Assoc. 1981;71(8):423.  86. Perdikakis E, Grigoraki E, Karantanas A. Os naviculare: the multiossicle configuration of a normal variant. Skeletal Radiol. 2011;40:85.  87. Lawson JP, Ogden JA, Sella E, et al. The painful accessory navicular. Skeletal Radiol. 1984;12:250.  88. Sella E, Lawson J. Biomechanics of the accessory navicular synchondrosis. Foot Ankle. 1987;8:156.  89. Huang J, Zhang Y, Ma X. Accessory navicular bone incidence in Chinese patients: a retrospective analysis of x-rays following trauma or progressive pain onset. Surg Radiol Anat. 2013;36:167.  90. Chiu NT, Jou IM, Lee BF, et al. Symptomatic and asymptomatic accessory navicular bones: findings of Tc-99m MDP bone scintigraphy. Clin Radiol. 2000;55:353.  91. Grogan DP, Gasser SI, Ogden JA. The painful accessory navicular: a clinical and histopathological study. Foot Ankle. 1989;10:164.  92. Mosel LD, Kat E, Voyvodic F. Imaging of the symptomatic type II accessory navicular bone. Australas Radiol. 2004;48:267.  93. Issever AS, Minden K, Eshed I, et al. Accessory navicular bone: when ankle pain does not originate from the ankle. Clin Rheumatol. 2007;26:2143.  94. Mann RW. Calcaneus secundarius: description and frequency in six

skeletal samples. Am J Phys Anthropol. 1990;81:17.  95. Ersen O, Akyildiz F, Ozyurek S, et al. Os calcaneus secundarius mimicking fracture. BMJ Case Rep. 2013. doi:10.1136/bcr-2013-009954. Published online. Accessed February 22, 2014.  96. Kurklu M, Kose O, Yurttas Y, et al. Anterosuperior calcaneal process fracture or OS calcaneus secundarius? Am J Phys Med Rehabil. 2010;89:522.  97. Stauss J, Connolly LP, Perez-Rossello J, et al. Skeletal scintigraphy of possible os calcaneus secundarius. Clin Nucl Med. 2003;28:424.  98. Bencardino J, Rosenberg ZS, Beltran J, et al. Os sustentaculi: depiction on MR images. Skeletal Radiol. 1997;26:505.  99. March HC, London RI. The os sustentaculi. Am J Roentgenol Radium Ther Nucl Med. 1956;76:1114. 100. Mellado JM, Salvado E, Camins A, et al. Painful os sustentaculi: imaging findings of another symptomatic skeletal variant. Skeletal Radiol. 2002;31:53. 101. Baghla DPS, Shariff S, Bashir WA. Acquired cavo-varus deformity caused by an accessory calcaneus: a case report and literature review. Skeletal Radiol. 2010;39:193. 102. Krause J, Rouse A. Accessory calcaneus: a case report and literature review. Foot Ankle Int. 1995;16(10):646. 103. Viana SL, Fernandes JL, Mendonca JL, et al. Painful os talus secundarius: a case report and imaging findings. Foot Ankle Int. 2007;28:624. 104. Hahn JE. Surgical excision of an os talus secundarius: a case report. J Foot Surg. 1978;17:129. 105. Stieda L. Ueber secundare fusswurzelknochen. Arch Physiol Wissensch Med. 1869:108. Cited by: Chao W. Os trigonum. Foot Ankle Clin. 2004;9:787.

106. Song AJ, Del Giudice M, Lazarus ML, et al. Radiologic case study. Os trigonum syndrome. Orthopedics. 2013;36:5. 107. Chao W. Os trigonum. Foot Ankle Clin. 2004;9:787. 108. Paulos LE, Johnson CL, Noyes FR. Posterior compartment fractures of the ankle. A commonly missed athletic injury. Am J Sports Med. 1983;11:439. 109. Giuffrida AY, Lin SS, Abidi N, et al. Pseudo os trigonum sign: missed posteromedial talar facet fracture. Foot Ankle Int. 2003;24:642. 110. Sopov V, Liberson A, Groshar D. Bone scintigraphic findings of os trigonum: a prospective study of 100 soldiers on active duty. Foot Ankle Int. 2000;21:822. 111. Wakeley CJ, Johnson DP, Watt I. The value of MR imaging in the diagnosis of the os trigonum syndrome. Skeletal Radiol. 1996;25:133. 112. Schreiber A, Differding P, Zollinger H. Talus partitus. J Bone Joint Surg Br. 1985;67B(3):430. 113. Serrato J, Bennett JT. Adolescent fracture of the talus associated with talus partitus. J Surg Orthop Adv. 2013;22(2):164. 114. Weinstein SL, Bonfiglio M. Unusual accessory (bipartite) talus simulating fracture. J Bone Joint Surg Am. 1975;57A(8):1161. 115. Milgrom C, Kaplan L, Lax E. Case report 341: os accessorium supracalcaneum of the left hind foot (also present, but to a lesser extent, on the right). Skeletal Radiol. 1986;15:150. 116. Tsuruta T, Shiokawa Y, Kato A, et al. Radiological study of the accessory skeletal elements in the foot and ankle (author’s transl). Nihon Seikeigeka Gakkai Zasshi. 1981;55:357. 117. Nwawka OK, Hayashi D, Diaz LE, et al. Sesamoids and accessory ossicles of the foot: anatomical variability and related pathology. Insights Imaging. 2013;4:581.

118. Coral A. The radiology of skeletal elements in the subtibial region: incidence and significance. Skeletal Radiol. 1987;16:298. 119. Madhuri V, Poonnoose PM, Lurstep W. Accessory os subtibiale: a case report of misdiagnosed fracture. Foot Ankle Online J. 2009;2(6):3. 120. Coral A. Os subtibiale mistaken for a recent fracture. Br Med J. 1986;292:1571. 121. Park HG, Sim JA, Koh YH. Posterior tibial tendon dysfunction secondary to os subtibiale impingement: a case report. Foot Ankle Int. 2005;26:184. 122. Bellapianta JM, Andrews JR, Ostrander RV. Bilateral os subtibiale and talocalcaneal coalitions in a college soccer player: a case report. J Foot Ankle Surg. 2011;50:462. 123. Berg EE. The symptomatic os subfibulare. Avulsion fracture of the fibula associated with recurrent instability of the ankle. J Bone Joint Surg Am. 1991;73(8):1251. 124. Kono T, Ochi M, Takao M, et al. Symptomatic os subfibulare caused by accessory ossification: a case report. Clin Orthop Relat Res. 2002;399:197. 125. Champagne IM, Cook DL, Kestner SC, et al. Os subfibulare. Investigation of an accessory bone. J Am Podiatr Med Assoc. 1999;89:520. 126. Karlsson J, Lansinger O. Separate centre of ossification of the lateral malleolus with instability of the ankle joint. Arch Orthop Trauma Surg. 1990;109:291. 127. Gruber W. Uber den Fortsatz des Seitenhockers-Processus tuberositatis lateralis-des Metatarsale V und sein Auftreten als Epiphyse. Arch Anat Physiol Wissenschaft Med. 1875:48–58. Cited by: Coughlin MJ. Sesamoids and accessory bones of the foot. In: Coughlin MJ, Mann RA, Saltzman CA, ed. Surgery of the Foot & Ankle. 8th ed. St. Louis, MO: Mosby; 2006:531– 610.

128. Feeney MS, Devitt AT, Stephens MM. Duplication of the medial column presenting as a fixed equinus deformity: a case report. Foot Ankle Int. 1998;19(2):120. SUGGESTED READINGS General References on Variants Bierman MI. The supernumerary pedal bones. AJR. 1922;9:404. Bizarro AH. On sesamoid and supernumerary bones of the limbs. J Anat. 1921;55:256. Hark FW. Congenital anomalies of the tarsal bones. Clin Orthop. 1960;16:21. Helal B. The accessory ossicles and sesamoids. In: Helal B, Wilson D, ed. The Foot. Vol 1. New York, NY: Churchill-Livingstone; 1988:567. Holland CT. On rarer ossifications seen during x-ray examinations. J Anat. 1921;55:235. Kleinberg S. Supernumerary bones of the foot. Ann Surg. 1917;65:499. Pirie AH. Extra bones in the wrist and ankle found by roentgen rays. AJR. 1921;8:572. Partite Sesamoids Dobas DC, Silvers MD. The frequency of partite sesamoids of the first metatarsophalangeal joint. J Am Podiatry Assoc. 1977;67(12):880. Feldman F, Pochaczevsky R, Hecht H. The case of the wandering sesamoid and other sesamoid afflictions. Radiology. 1970;96:275. Frankel JP, Harrington J. Symptomatic bipartite sesamoids. J Foot Surg. 1990;29(4):318. Golding C. The sesamoids of the hallux. J Bone Joint Surg Br. 1960;42B(4):840.

Hubay CA. Sesamoid bones of the hands and feet. AJR. 1949;61(4):493. Inge GAL, Ferguson AB. Surgery of the sesamoid bones of the great toe. Arch Surg. 1933;27:466. Leonard MH. The sesamoids of the great toe: the pedal polemic. Clin Orthop. 1960;16:295. Scranton PE, Rutkowski R. Anatomic variations in the first ray. II. Disorders of the sesamoids. Clin Orthop. 1980;151:256. Walling AK, Ogden JA. Case report 666. Skeletal Radiol. 1991;20:233. Weil LS, Hill M. Bipartite tibial sesamoid and hallux abducto valgus deformity: a previously unreported correlation. J Foot Surg. 1992;31(2):104. Bipartite Medial Cuneiform Barclay M. A case of duplication of the internal cuneiform bone of the foot (cuneiforme bipartitum). J Anat. 1932;67:175. Dellacorte MP, Lin PJ, Grisafi PJ. Bilateral bipartite medial cuneiform. J Am Podiatr Med Assoc. 1992;82:475. Bipartite Navicular de Fine Licht E. On bipartite os naviculare pedis. Acta Radiol. 1941;22:377. Waugh W. Structural deformities of the outer third of the adult tarsal navicular. Proc R Soc Med. 1956;49:965–967. Wiley JJ, Brown DE. The bipartite tarsal scaphoid. J Bone Joint Surg Br. 1981;63B:583. Supernumerary Bones Acker I. Residuum of a supernumerary phalanx 30 years after surgery. J Am Podiatr Assoc. 1966;56:124.

Cole M. Three uncommon anomalies. J Am Podiatr Assoc. 1970;60:400. Gold AG, Katz M, Comerford JS. Distal accessory phalanx of the foot. J Am Podiatr Med Assoc. 1990;80(6):323. Rao BR. Supernumerary toe arising from the medial cuneiform. J Bone Joint Surg Am. 1979;61A:306. Wishnie PA, London E, Porat S. Complete duplication of an accessory first ray. J Foot Surg. 1990;29(5):471. Absence of Bone Carroll BW, Greenberg DC, Simpson RR. The two-phalanged fifth toe: development, occurrence and relation to heloma durum. J Am Podiatr Assoc. 1978;68(9):641. Zinsmeister BJ, Edelman R. Congenital absence of the tibial sesamoid: a report of two cases. J Foot Surg. 1985;24(4):266. Variants of Form Berenter JS, Goldman FD. Surgical approach for enlarged peroneal tubercles. J Am Podiatr Med Assoc. 1989;79(9):451. Bisceglia CF, Sirota AD, Dull DD. An unusual case of hypertrophied peroneal tubercles. J Am Podiatr Assoc. 1983;73(9):481. Scranton PE. Pathologic anatomic variations in the sesamoids. Foot Ankle. 1981;1(6):321. Techner LM, DeCarlo RL. Peroneal tubercle osteochondroma. J Foot Surg. 1992;31(3):234. Synostosis/Coalition Agostinelli JR. Tarsal coalition and its relation to peroneal spastic flatfoot. J Am Podiatr Med Assoc. 1986;76(2):76.

Bonk JH, Tozzi MA. Congenital talonavicular synostosis. J Am Podiatr Med Assoc. 1989;79(4):186. Bullitt JB. Variations of the bones of the foot. Fusion of the talus and navicular, bilateral and congenital. AJR. 1928;20:548. Carson CW, Ginsburg WW, Cohen MD, et al. Tarsal coalition: an unusual cause of foot pain: clinical spectrum and treatment in 129 patients. Semin Arthritis Rheum. 1991;20(6):367. Cavallaro DC, Hadden HR. An unusual case of tarsal coalition. J Am Podiatr Assoc. 1978;68(2):71. Cohen AH, Laugner TE, Pupp GR. Calcaneonavicular bar resection. J Am Podiatr Med Assoc. 1993;83(1):10. Downey MS. Tarsal coalitions: a surgical classification. J Am Podiatr Med Assoc. 1991;81(4):187. Frost RA, Fagan JP. Bilateral talonavicular and calcaneocuboid joint coalition. J Am Podiatr Med Assoc. 1995;85(6):339. Green MR, Yanklowitz B. Asymptomatic naviculocuneiform synostosis with a ganglion cyst. J Foot Surg. 1992;31(3):272. Hart DJ, Hart TJ. Iatrogenic metatarsal coalition: a postoperative complication of adjacent V-osteotomy. J Foot Surg. 1985;24(3):205. Jack EA. Bone anomalies of the tarsus in relation to “peroneal spastic flatfoot.” J Bone Joint Surg Br. 1954;36B:530. Kashuk KB, Hanft JR, Schabler JA, et al. An unusual intermetatarsal coalition. J Am Podiatr Med Assoc. 1991;81(7):384. Keenleyside A, Mann RW. Unilateral ectrodactyly, metatarsal synostosis, and hypoplasia in an Eskimo. J Am Podiatr Med Assoc. 1991;81(1):18. Kim SH. The C sign. Radiology. 2002;223:756–757.

Kumar SJ, Guille JT, Lee MS, et al. Osseous and non-osseous coalition of the middle facet of the talocalcaneal joint. J Bone Joint Surg Am. 1992;74A(4):529. Lapidus PW. Congenital fusion of the bones of the foot: with a report of congenital astragaloscaphoid fusion. J Bone Joint Surg. 1932;14:888. Newman JS, Newberg AH. Congenital tarsal coalition: multimodality evaluation with emphasis on CT and MR imaging. Radiographics. 2000;20:321–332. Pachuda NM, Lasday SD, Jay RM. Tarsal coalition: etiology, diagnosis, and treatment. J Foot Surg. 1990;29(5):474. Page JC. Peroneal spastic flatfoot and tarsal coalitions. J Am Podiatr Med Assoc. 1987;77(1):29. Palladino SJ, Schiller L, Johnson JD. Cubonavicular coalition. J Am Podiatr Med Assoc. 1991;81(5):262. Pensieri SL, Jay RM, Schoenhaus HD, et al. Bilateral congenital calcaneocuboid synostosis and subtalar joint coalition. J Am Podiatr Med Assoc. 1985;75(8):406. Percy EC, Mann DL. Tarsal coalition: a review of the literature and presentation of 13 cases. Foot Ankle. 1988;9(1):40. Perlman MD, Wertheimer SJ. Tarsal coalitions. J Foot Surg. 1986;25(1):58. Person V, Lembach L. Six cases of tarsal coalition in children aged 4 to 12 years. J Am Podiatr Med Assoc. 1985;75(6):320. Pontious J, Hillstrom HJ, Monahan T, et al. Talonavicular coalition: objective gait analysis. J Am Podiatr Med Assoc. 1993;83(7):379. Pouliquen JC, Duranthon LD, Glorion C, et al. The too-long anterior process calcaneus: a report of 39 cases in 25 children and adolescents. J Pediatr Orthop. 1998;18(3):333–336.

Sakellariou A, Sallomi D, Janzen DL, et al. Talocalcaneal coalition: diagnosis with the C-sign on lateral radiographs of the ankle. J Bone Joint Surg Br. 2000;82:574–578. Salomao O, Napoli MMM, de Carvalho AE, et al. Talocalcaneal coalition: diagnosis and surgical management. Foot Ankle. 1992;13(5):251. Schlefman BS, Ruch JA. Diagnosis of subtalar joint coalition. J Am Podiatr Assoc. 1982;72(4):166. Takakura Y, Sugimoto K, Tanaka Y, et al. Symptomatic talocalcaneal coalition. Clin Orthop. 1991;269:249. Venning P. Variation of the digital skeleton of the foot. Clin Orthop. 1960;16:26. Wiles S, Palladino SJ, Stavosky JW. Naviculocuneiform coalition. J Am Podiatr Med Assoc. 1988;78(7):355. Accessory Bones Helal B. The accessory ossicles and sesamoids. In: Helal B, Wilson D, ed. The Foot. New York, NY: Churchill-Livingstone; 1988. Os Interphalangeus Genakos JJ. Clinical sign consistent with the hallucal interphalangeal sesamoid. J Am Podiatr Med Assoc. 1993;83(12):696. Os Intermetatarseum Henderson RS. Os intermetatarseum and a possible relationship to hallux valgus. J Bone Joint Surg Br. 1963;45B:117. Scarlet JJ, Gunther R, Katz J, et al. Os intermetatarseum-one. J Am Podiatr Assoc. 1978;68(6):431. Os Supranaviculare

Miller GA, Black JR. Symptomatic os supranaviculare. J Am Podiatr Med Assoc. 1990;80(5):248. Pacini AJ. Anomalies of the pedal scaphoid. Am J Electrother Radiol. 1921;39(6):217. Pirie AH. A normal ossicle in the foot frequently diagnosed as a fracture. Arch Radiol Electrother. 1919;24:93. Accessory Navicular Anspach WE, Wright EB. The divided navicular of the foot. Radiology. 1937;29:725. Chater EH. Foot pain and the accessory navicular bone. Ir J Med Sci. 1962;442:471. Fredrick LA, Beall DP, Ly JQ, et al. The symptomatic accessory navicular bone: a report and discussion of the clinical presentation. Curr Probl Diagn Radiol. 2005;34:47. Smith TR. Management of dancers with symptomatic accessory navicular: 2 case reports. J Orthop Sports Phys Ther. 2012;42(5):465. Wood WA, Spencer AM. Incidence of os tibiale externum in clinical pes planus. J Am Podiatr Assoc. 1970;60(7):276. Zadek I, Gold AM. The accessory tarsal scaphoid. J Bone Joint Surg Am. 1948;30A(4):957. Os Calcaneus Secundarius Herrmann NP. An unusual example of a calcaneus secundarius. J Am Podiatr Med Assoc. 1992;82(12):623. Mann RW. Calcaneus secundarius: variation of a common accessory ossicle. J Am Podiatr Med Assoc. 1989;79(8):363. Accessory Calcaneus

Heller AG. Accessory calcaneus. J Am Podiatr Med Assoc. 1961;51:275. Os Trigonum Blake RL, Lallas PJ, Ferguson H. The os trigonum syndrome. J Am Podiatr Med Assoc. 1992;82(3):154. Brodsky AE, Khalil MA. Talar compression syndrome. Foot Ankle. 1987;7:338. Grogan DP, Walling AK, Ogden JA. Anatomy of the os trigonum. J Pediatr Orthop. 1990;10(5):618. Hamilton WG. Stenosing tenosynovitis of the flexor hallucis longus tendon and posterior impingement upon the os trigonum in ballet dancers. Foot Ankle. 1982;3:74. Mann RW, Owsley DW. Os trigonum: variation of a common accessory ossicle of the talus. J Am Podiatr Med Assoc. 1990;80(10):536. Marotta JJ, Micheli LJ. Os trigonum impingement in dancers. Am J Sports Med. 1992;20(5):533. Martin BF. Posterior triangle pain: the os trigonum. J Foot Surg. 1989;28(4):312. Reinherz RP. The significance of the os trigonum. J Foot Surg. 1979;18:61. Wenig JA. Os trigonum syndrome. J Am Podiatr Med Assoc. 1990;80(5):278. Os Supracalcaneum Milgrom C, Kaplan L, Lax E. Case report 341: quiz. Skeletal Radiol. 1986;15:150. Os Subfibulare Bowlus TH, Korman SF, DeSilvio M, et al. Accessory os subfibulare avulsion secondary to the inversion ankle injury. J Am Podiatr Assoc.

1980;70(6):302. Mancuso JE, Hutchison PW, Abramow SP, et al. Accessory ossicle of the lateral malleolus. J Foot Surg. 1991;30(1):52.

7 Normal Development ROBERT A. CHRISTMAN AND JACQUELINE TRUONG The radiographic presentation, or ossification, of the pediatric foot and ankle varies considerably from patient to patient. It depends on the age and sex of the individual, and variation is frequently encountered. Ossification center time of appearance can vary not only between individuals but also within the same individual with respect to bones in the same extremity. The normal expected learning curve, therefore, is much longer than that for radiographic anatomy of the adult foot and ankle, especially since one sees far fewer pediatric radiographic studies than the adult. Evaluation of the pediatric skeleton requires observation of all the following:  1. Presence of the ossification centers  2. Orderly appearance of the ossification centers  3. Form (size and shape) of visible ossification centers  4. Relationships of one ossification center to another Several authors have listed the time of appearance of primary and secondary ossification centers; Tables 7-1 through 7-4 summarize many of them.1–3 In some instances, you can see a range of appearance time, and these times may vary among authors. The appearance time of ossification centers is earlier for females than for males. The sequence of fetal (primary) foot bone ossification centers is listed in Box 7-1.4 The time in which these primary centers completely ossify is listed in Table 7-5. Some accessory ossification centers have also been included: the os trigonum, and the tips of the medial and lateral malleoli. Sarrafian also lists a secondary ossification center for the posterior border of the talus that unites with the talus at 12.9 ± 1.3 years (male) and 9.8 ± 1.3 years (female); it appears the same time as the os

trigonum ossification center.3 TABLE 7-1

  Time of Appearance of Primary Ossification Centers: Male

TABLE 7-2

 

Time of Appearance of Primary Ossification Centers: Female

TABLE 7-3

 

Time of Appearance of Secondary Ossification Centers: Male

A complete description of the developing foot’s radiographic anatomy at different ages is nearly impracticable as the range of ossification varies considerably within individual children and between children of the same age, as shown in Tables 7-1 through 7-4. Instead, composite radiographs based primarily on the orderly appearance of primary and secondary ossification centers, after reviewing numerous sets of children radiographs at varying stages of development, are illustrated in Figures 7-1 through 7-19. These examples adhere to the skeletal maturity indicators described by Hoerr et al.5 TABLE 7-4

 

Time of Appearance of Secondary Ossification Centers: Female

BOX 7-1 Sequence of Ossification of Fetal Foot Sequence 1st Hallux distal phalanx and second, third, and fourth metatarsals 2nd Second through fifth toe distal phalanges and the first and fifth metatarsals 3rd Hallux and second toe proximal phalanges 4th Second, third, and fourth toe proximal phalanges 5th Second, third, and fourth toe middle phalanges

6th Fifth toe middle phalanx 7th Calcaneus 8th Talus 9th Cuboid Compiled from Arey LB. Developmental Anatomy. 7th ed. Philadelphia, PA; Saunders; 1965:104. TABLE 7-5   Completion of Ossification   Male (y) Female (y) Hallux distal phalanx basal epiphysis 13.6–16 11.3–13.4 Fifth toe proximal phalanx basal epiphysis 14.7–16.7 12.9–15.1 Fifth metatarsal head epiphysis 14.8–16.7 13–15.3 Medial cuneiform 13.8–15.6 11.2–13.3 Lateral cuneiform 13.8–15.6 11–13.5 Navicular 13.8–15.6 11.1–13.3 Calcaneal apophysis 14.5–16.6 12.8–15.2 Distal tibial epiphysis 15.3–17.2 13.8–15.9 Distal fibular epiphysis 15.4–17.4 14–16.2 Compiled from Acheson RM. Maturation of the skeleton. In: Falkner F, ed. Human Development. Philadelphia, PA: Saunders; 1966.

FIGURE 7-1. 3 months old. A: Dorsoplantar view. B: Lateral view. The following primary ossification centers are visible: most phalanges, all metatarsals, the cuboid, talus, calcaneus, distal tibia, and fibula. The distal tibial and fibular secondary ossification centers are also visible.

FIGURE 7-2. 5 months old. A: Lateral view. B: Dorsoplantar view. The lateral cuneiform ossification center is now visible.

FIGURE 7-3. 10 months old. A: Dorsoplantar view. B: Lateral view. The hallux distal phalanx secondary ossification center (basal epiphysis) is visible.

FIGURE 7-4. 1 year 5 months old. A: Dorsoplantar view. B: Lateral view. The medial cuneiform primary ossification and the lesser toe proximal phalangeal secondary ossification centers (basal epiphyses) are now becoming visible.

FIGURE 7-5. 1 year 8 months old. A: Dorsoplantar view. B: Lateral view. The hallux proximal phalanx and first metatarsal secondary ossification centers (basal epiphyses) are barely visible.

FIGURE 7-6. 2 years old. A: Dorsoplantar view. B: Lateral view. The intermediate cuneiform ossification center is visible.

FIGURE 7-7. 2 years 3 months old. A: Dorsoplantar view. B: Lateral view. The navicular ossification center is visible.

FIGURE 7-8. 2 years 6 months old. A: Dorsoplantar view. B: Lateral view. The lesser metatarsal secondary ossification centers (distal epiphyses) are beginning to ossify. The first metatarsal distal epiphysis (sometimes referred to as the pseudoepiphysis), when present, may also be seen.

FIGURE 7-9. 3 years old. A: Dorsoplantar view. B: Lateral view. The middle phalanx secondary ossification centers (basal epiphyses) are visible.

FIGURE 7-10. 4 years 5 months old. A: Dorsoplantar view. B: Lateral view. Continued modeling of all visible ossification centers.

FIGURE 7-11. 4 years 8 months old. A: Dorsoplantar view. B: Lateral view. The lesser toe distal phalanx basal epiphyses are visible. The posterior calcaneus becomes roughened.

FIGURE 7-12. 6 years old. A: Dorsoplantar view. B: Lateral view. The posterior calcaneus (metaphysis) becomes jagged.

FIGURE 7-13. 6 years 6 months old. A: Dorsoplantar view. B: Lateral view. Early ossification of calcaneal apophysis (secondary epiphysis) is visible.

FIGURE 7-14. 7 years old. A: Dorsoplantar view. B: Lateral view. Continued ossification of calcaneal apophysis.

FIGURE 7-15. 8 years old. A: Dorsoplantar view. B: Lateral view. Calcaneal apophysis appears sclerotic relative to calcaneal body.

FIGURE 7-16. 9 years old. Lateral view. Ossification of os trigonum.

FIGURE 7-17. 11 years old. A: Dorsoplantar view. B: Lateral view. Fifth metatarsal apophysis (basal epiphysis/tuberosity) is visible, as are the sesamoids.

FIGURE 7-18. 13 years old. A: Lateral view. B: Dorsoplantar view. Continued ossification of all centers, especially calcaneal apophysis.

FIGURE 7-19. 15 years 6 months old. A: Dorsoplantar view. B: Lateral view. Physes are closing at multiple sites. REFERENCES   1. Garn SM, Rohmann CG, Silverman FN. Radiographic standards for postnatal ossification and tooth calcification. Med Radiogr Photogr. 1967;43(2):45.   2. Acheson RM. Maturation of the skeleton. In: Falkner F, ed. Human Development. Philadelphia, PA: Saunders; 1966.   3. Kelikian AS. Sarrafian’s Anatomy of the Foot and Ankle. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2011.   4. Arey LB. Developmental Anatomy. 7th ed. Philadelphia, PA: Saunders; 1965:104.   5. Hoerr NL, Pyle SI, Francis CC. Radiographic Atlas of Skeletal Development of the Foot and Ankle. Springfield, IL: Charles C Thomas; 1962. SUGGESTED READINGS Acheson RM, Vicinus JH, Fowler GB. Studies in the reliability of assessing skeletal maturity from x-rays. III. The methods contrasted. Hum Biol. 1966;38:204. Bareither D. Prenatal development of the foot and ankle. J Am Podiatr Med Assoc. 1995;85(12):753. Blass BC. A preliminary study of the fetal skeleton of the human foot. J Am Podiatr Assoc. 1973;63(1):12. Cheng X, Wang Y, Qu H, et al. Ossification processes and perichondral ossification groove of Ranvier: a morphological study in developing human calcaneus and talus. Foot Ankle. 1995;16(1):7.

Chung T, Jaramillo D. Normal maturing distal tibia and fibula: changes with age at MR imaging. Radiology. 1995;194:227. Ferguson AB, Gingrich RM. The normal and the abnormal calcaneal apophysis and tarsal navicular. Clin Orthop. 1957;10:87. Gardner E, Gray DJ, O’Rahilly R. The prenatal development of the skeleton and joints of the human foot. J Bone Joint Surg Am. 1959;41A(5):847. Gould N, Moreland M, Alvarez R, et al. Development of the child’s arch. Foot Ankle. 1989;9(5):241. Graham CB. Assessment of bone maturation—methods and pitfalls. Radiol Clin North Am. 1972;10:185. Harris EJ. The relationship of the ossification centers of the talus and calcaneus to the developing bone. J Am Podiatr Assoc. 1976;66(2):76. Hensinger RN. Standards in Pediatric Orthopedics. Tables, Charts, and Graphs Illustrating Growth. New York, NY: Raven Press; 1986. Hubbard AM, Meyer JS, Davidson RS, et al. Relationship between the ossification center and cartilaginous anlage in the normal hindfoot in children: study with MR imaging. AJR. 1993;161:849. Kahn SL, Gaskin CM, Sharp VL, et al. Radiographic Atlas of Skeletal Maturation. New York, NY: Thieme; 2011. Keats TE. Atlas of Roentgenographic Measurement. 6th ed. St Louis, MO: Mosby; 1990. Keats TE, Smith TH. An Atlas of Normal Developmental Roentgen Anatomy. 2nd ed. Chicago, IL: Mosby; 1988. Kuhns LR, Poznanski AK. Radiological assessment of maturity and size of the newborn infant. CRC Crit Rev Diagn Imaging. 1980;12(3):245. Meschan I. Roentgen Signs in Diagnostic Imaging. Vol 2, 2nd ed. Philadelphia, PA: Saunders; 1985.

O’Rahilly R, Gardner E, Gray DJ. The skeletal development of the foot. Clin Orthop. 1960;16:7. O’Rahilly R, Gray DJ, Gardner E. Chondrification in the hands and feet of staged human embryos. Contrib Embryol. 1957;36:185. Senior HD. The chondrification of the human hand and foot skeleton. Anat Rec. 1929;42:35. Vilaseca RR, Ribes ER. The growth of the first metatarsal. Foot Ankle. 1980;1(2):117.

8 Developmental Variants ROBERT A. CHRISTMAN AND JACQUELINE TRUONG The number of variations in the adult foot may appear overwhelming, but at least they are of one developmental age group, that is, mature. In contrast, consider all the variations that might occur at all different developmental stages. And the radiographic appearance of one child’s foot at 5 years of age may be considerably different from another child’s at the same chronologic age. Fortunately, the variants of development are not that numerous and are primarily seen during the earlier stages of development; the reader should, however, refer to Chapter 7 for the varying radiographic presentations seen during skeletal development. Developmental variants may easily be mistaken for pathology. The more frequent variants are multiple primary or secondary ossification centers, the presence of accessory ossification centers at sites other than the expected locations, and variants of form. A jagged metaphysis adjacent to the physis may suggest abnormality but is usually a normal variation. Accessory ossicles may be seen, but usually only later in skeletal development. Also, do not be surprised to see the initial appearance of an ossification center much earlier or even later than expected. Confusion exists regarding the terms epiphysis and apophysis. An epiphysis is defined as “the expanded articular end of a long bone, developed from a secondary ossification center…”1 In contrast, an apophysis is “any outgrowth or swelling, especially a bony outgrowth that has never been entirely separated from the bone of which it forms a part, such as a process, tubercle, or tuberosity.”1

FIGURE 8-1. Multiple ossification centers, medial cuneiform, in a 5-yearold male. MULTIPLE OSSIFICATION CENTERS Both primary and secondary (epiphyseal/apophyseal) ossification centers can radiographically appear to ossify from multiple sites. These findings typically are asymptomatic and have been associated with a heterogeneous group of disorders known as the osteochondroses2 (see Chapter 17). Generally

speaking, the radiographic picture of multiple centers of ossification is a variant of normal and occurs frequently. It is rare to see the primary centers of tubular bones ossify from multiple areas, but the irregular bones of the midtarsus are often affected. The more commonly affected midtarsal bones are the medial cuneiform and navicular (Figures 8-1 and 8-2, respectively). Depending on the stage of development, appearance can vary considerably. The presence of multiple navicular ossification centers alone is not diagnostic of the osteochondrosis known as Kohler disease even though they both may appear flattened and asymmetric when first ossified.3 In fact, in most cases, radiographs of these same patients years later show normal ossification although the contour of the adult bone may be altered (Figure 8-3).4

FIGURE 8-2. Multiple ossification centers, navicular, in a 7-year-old male (medial oblique view).

FIGURE 8-3. Variant navicular ossification can easily be misdiagnosed as Kohler disease. A: Sclerotic fragmented appearance at 4 years of age (asymptomatic). B: Normal navicular ossification at 8 years of age in the same patient. Rarely, the calcaneus may appear to ossify from two centers (Figure 8-4); this has been called the bifid calcaneus, calcaneus bipartitus, and duplicate calcaneus.5–7 Characteristically, the calcaneus in this situation develops from a larger posterior and smaller anterior segment.8 It was first described by Sever9 in 1930. Cormier-Daire et al.10 believe that the duplicate calcaneus may reflect a more generalized developmental defect. The persistence of two separate primary ossification centers into adulthood is known as a bipartition. The precursor to bipartition may be seen in the developing skeleton; it affects the first metatarsophalangeal sesamoids, albeit later in development,11 and rarely the medial cuneiform, navicular, and talus12,13 bones. Secondary ossification centers (epiphyses and apophyses) frequently appear to ossify from multiple sites14 and are most evident in the metatarsals and calcaneus. Other sites affected include the hallux proximal phalanx basal epiphysis and the distal tibial epiphysis (Figure 8-5). Partitioning of the basal and apophyseal secondary ossification centers, in particular, may be mistaken for fracture. In general, multiple secondary ossification centers seem always to coalesce into one center, although the entire center may remain separate

from its adjacent primary ossification center. A metatarsal epiphysis may ossify from two, sometimes three centers (Figure 8-6). However, not all metatarsal epiphyseal centers need to be partite in the same foot or even bilaterally. Furthermore, their size and shape may not be symmetric. Segmentation of the hallux proximal phalanx epiphysis is known as a bipartite basal epiphysis and can easily mimic a fracture (Figure 8-7). Typically, the bipartition is off center, the medial segment being the larger of the two. It may or may not present bilaterally.

FIGURE 8-4. Multiple ossification centers, calcaneus. A: Smaller, separate ossification center anteriorly (arrow). B: One year later, the two ossification centers are united.

FIGURE 8-5. Segmentation (bipartition, arrow) of the distal tibial epiphysis, lateral view (a bilateral finding).

FIGURE 8-6. Multiple ossification centers, metatarsals. A: Bipartite first, second, and third metatarsals (5-year-old male). B: Bipartite fifth metatarsal tuberosity epiphysis.

FIGURE 8-7. Bipartite basal epiphysis, hallux proximal phalanx (a unilateral finding in this patient). The calcaneal apophysis characteristically appears with multiple ossification centers radiographically (Figure 8-8). These centers are often irregular in outline and easily mimic fracture (“Sever disease”). However, this apparent “fragmentation” of the apophysis is entirely normal—often appearing to be more sclerotic than the rest of the bone and irregular.3 At first ossification begins posteroinferiorly; later ossification extends inferiorly to include the tuberosities and superiorly the bursal projection. At times an ossification center for the lateral tuberosity may be very large and suggests the development of an os subcalcis (Figure 8-9).

FIGURE 8-8. Multiple ossification centers, calcaneal epiphysis. A: Five-

year-old female. B: Note the irregular appearance of the adjacent metaphysis. C: Twelve-year-old male. Note the separate ossification of the bursal projection (posterosuperior aspect, arrow) mimicking a fracture. D: Nineyear-old female. Medial oblique view demonstrating ossification of lateral tuberosity. Multiple ossification centers, calcaneal apophysis. E: Thirteenyear-old male. Lateral oblique view demonstrating ossification of medial tuberosity.

FIGURE 8-9. Separate, large ossification center for the lateral tuberosity in a 14-year-old male (medial oblique view). This center eventually united with the parent bone 2 years later. The fifth metatarsal tuberosity apophysis appears between 6 and 15 years of

age as a thin and fragile shell-like bone lateral to the base of the fifth metatarsal (Figure 8-10A).3 Irregular segmentation of this apophysis is frequent (Figure 8-10B) and commonly misinterpreted as fracture. PSEUDOEPIPHYSES Secondary ossification centers may appear to present at locations in addition to the normally expected epiphyseal sites. Typical sites in the foot include the head of the first metatarsal and the bases of the second through fourth metatarsals.15 The terms pseudoepiphysis and supernumerary epiphysis have been used to describe these variants in the forefoot. Lachman16 differentiates between the two types radiographically as follows: Pseudoepiphysis refers to “those epiphyses that show a bony bridge connecting them with the shaft,” and supernumerary epiphysis presents as a “truly independent center.” However, these two types more likely represent the same entity at varying stages of development. Furthermore, discrepancy exists regarding the histology of this enigma. Ogden et al.17 have demonstrated that this is not a true epiphysis but an extension of the metaphysis (hence, pseudoepiphysis). In contrast, Vilaseca and Ribes18 claim that it histologically and anatomically resembles all other physes. Another study has shown that this entity is actually quite common.19

FIGURE 8-10. A: Early ossification of the fifth metatarsal apophysis (arrow). B: Irregular, multiple ossification centers for the fifth metatarsal tuberosity epiphysis. Radiographically, the first metatarsal pseudoepiphysis appears to develop similar to other secondary ossification centers; however, as ossification increases, superimposition and early closure of the physis give the appearance of incomplete development (Figure 8-11). Hypertrophy of the adjacent distal metaphysis may be noticed laterally and superiorly (Figure 812). Pseudoepiphyses may also be seen along the bases of the second, third, and fourth metatarsals. Initially the developing bases appear irregular; separate

ossification centers may or may not be easily visualized, again depending on the amount of superimposition and the stage of development (Figure 8-13). ACCESSORY OSSIFICATION CENTERS An ossification center may be seen for any of the accessory ossicles mentioned in Chapter 6. Typically, such centers appear during the later stages of development. Examples include the accessory sesamoid, os intermetatarseum, and os supranaviculare. Separate centers of ossification for the posterior talar tubercles typically appear between the ages of 8 and 11 years. During development, these centers may unite with the parent bone. Occasionally, the ossification center lateral to the flexor hallucis longus tendon can appear elongated after uniting with the main body of the talus—known as a trigonal process or Stieda’s process.20 If it remains separate, it is known as the os trigonum (Figure 8-14). According to Dale et al.,21 24% of all talar fractures occur at the lateral process making it important to note the normal position and appearance of this ossicle. The os trigonum is found in 14% of males and 18% of females.3

FIGURE 8-11. First metatarsal distal epiphysis (pseudoepiphysis) at varying stages of development. A: Two-year-old male. B: Five-year-old male. C: Six-year-old female.

FIGURE 8-12. Hypertrophy of first metatarsal distal metaphysis adjacent to distal epiphysis (arrows). A: Lateral hypertrophy (dorsoplantar view). B: Superior hypertrophy (lateral view). In about 2% of the population,4 the center of ossification for the tuberosity of the navicular persists as a separate ossicle and is called an accessory navicular. Its presence can weaken the longitudinal dynamic arch support of the foot by altering the insertion point of the posterior tibialis tendon medially. In some individuals, this can cause symptomatic pes planovalgus deformity. The os subcalcis is extremely rare to see in the adult. Although the lateral tuberosity ossifies later in development as part of the calcaneal apophysis (see Figures 8-8D and 8-9), a large, separate ossification in a young child has been observed before the calcaneal apophysis had ossified. This example appears similar to the reverse calcaneal spur (discussed later) except that the outline of a superimposed ossicle can be visualized. Turhan et al.22 reported an accessory ossification center along the plantar calcaneal surface bilaterally, near the anterior tubercle, in a patient with multiple coalitions. Another rare ossicle has been reported along the lateral aspect of the calcaneus, adjacent to the trochlear process. This large ossicle is known as the accessory calcaneus, but has also been referred to as the os trochleare and

calcaneum accessorium.23 If large enough, it may articulate with the talus.24

FIGURE 8-13. Supernumerary epiphyses at the bases of the second, third, and fourth metatarsals. A: Rarely, epiphyseal ossification is represented by irregular metaphyseal margins and superimposed epiphyses (2-year-old female). B: Epiphyses and metaphyses well defined later in development (5year-old male).

FIGURE 8-14. Os trigonum (arrow). Early ossification (9-year-old male). Accessory ossification centers have been reported in the distal tibial and fibular malleoli.25 Though they appear radiologically to be separate entities from the main ossification centers (Figure 8-15), anatomically they are not.26 Histologically, the epiphyseal accessory center of ossification for the medial malleolus is continuous with the adjacent distal tibial epiphysis.27 In his review of 100 children radiographs between 6 and 12 years of age, Powell28 found the incidence of a separate tibial malleolus ossification center in 20 children (20%: 13% bilateral and 7% unilateral; boys only slightly more than girls) and a separate fibular malleolus center in 1 child. Selby29 found that, in a group of 151 children 6 to 12 years of age, there were 45 accessory ossification centers (30%) for the tip of the medial malleolus (15 in boys and 30 in girls); 41 of these accessory centers joined with the tibia. Persistence of a tibial or fibular malleolus accessory ossification center is known as an os subtibiale or os subfibulare, respectively, and may simulate fractures.

FIGURE 8-15. Accessory ossification center (arrow) for the tibial malleolus (7-year-old female). ABSENT OSSIFICATION CENTERS Absence of a primary ossification center, especially tarsal or metatarsal, results in an anomaly (see Chapter 6). However, it is not uncommon to have absence of secondary ossification centers in some phalanges of the lesser toes.3 Absence of the middle phalangeal basal epiphysis is most common (Figure 8-16). Absence of the distal phalanx basal epiphysis can accompany synostosis.

TARSAL COALITION Obvious tarsal coalition is not usually seen radiographically until the second decade, although the talonavicular coalition can be seen earlier.30 Coalitions are discussed and illustrated in Chapter 6. VARIANTS OF FORM Phalanges Clefts are frequently seen in the heads of the phalanges. They mimic the appearance of a supernumerary epiphysis, except that they are only at the margins of the phalanx and do not completely cross it. The cleft seen in the hallux proximal phalanx head is more often found only on its lateral side, although it may be seen both medially and laterally (Figure 8-17). It may either run transversely or obliquely. Clefts also are seen in the lesser toe middle and proximal phalanges. Those in the middle phalanges more resemble that seen in the hallux (Figure 8-18); lesser toe proximal phalangeal clefts are subtle in appearance. Epiphyses of the lesser toe phalanges occasionally are triangular in shape due to a delay in ossification that results in irregularity of the distal end.3 They have also been called bell- or cone-shaped epiphyses. The proximal, middle, or distal phalangeal basal epiphyses may be affected (Figure 8-19).

FIGURE 8-16. Absent phalangeal basal epiphysis. Middle phalanx epiphyses of all lesser toes are absent (5-year-old male).

FIGURE 8-17. Cleft (arrow) in the hallux proximal phalanx positioned laterally and oriented transversely (10-year-old male).

FIGURE 8-18. Middle phalangeal clefts (arrows) second toe (medial and lateral) (10-year-old female). A rare variant of form that affects small tubular bones is known as the longitudinal epiphyseal bracket.31 Jones32 originally suggested the term “delta phalanx” to describe this deformity in the middle phalanx of a finger: “[T]he bone is triangular in shape and has a continuous epiphysis running from the proximal to the distal end along the shortened side.” Although this entity has been associated with anomalies such as polydactyly,33,34 it may present solely as varus congenitus in the hallux.35 An example of the longitudinal epiphyseal bracket in a second toe middle phalanx is shown in Figure 8-20. In this case, the epiphysis is seen laterally and communicates with what could be interpreted as a cleft in the head. The lateral side of this

bone is short relative to the medial side, resulting in lateral angulation of the distal phalanx at the interphalangeal joint. This may be related to the adult skeletal variant shown in Figure 6-9 on page 85.

FIGURE 8-19. Triangular (bell- or cone-shaped) epiphyses. A: Distal phalanges (10-year-old male). B: Proximal phalanges (5-year-old male). Other phalangeal variants include an “exostosis” originating from the distal metaphysis of the proximal phalanx (Figure 8-21), short phalanx secondary to early physeal closure (Figure 8-22), absent ungual tuberosity (Figure 8-23), and circular radiolucency in the shaft of a proximal phalanx (Figure 8-24). Metatarsals As noted earlier, hypertrophy along the first metatarsal distal metaphysis has been associated with the supernumerary epiphysis. The first metatarsal basal epiphysis can vary in its form, especially at its junction with the metaphysis (Figure 8-25). Irregular outline of the proximal ends of the second through fourth metatarsals is not uncommon and may indicate early ossification of supernumerary epiphyses (Figure 8-26). Early physeal closure causes short metatarsal length.

Tarsal Bones The metaphysis adjacent to the posterior calcaneal apophysis frequently appears jagged (Figure 8-27). This finding is entirely normal and should not be mistaken for pathology. It typically is symmetric in presentation bilaterally. Another unusual variation is known as the reverse calcaneal spur. This infrequent entity is seen very early in development (within the first 2 years) and later disappears. It has been described in both boys and girls.36,37 The spur may either point posteriorly, anteriorly, or inferiorly (Figure 8-28). Other calcaneal variants include a “hole” or geographic lucency in the neutral triangle and an unusual shape overall (Figure 8-29). Variant shape of the talus may also be seen (Figure 8-30).

FIGURE 8-20. Longitudinal epiphyseal bracketing (arrow) second digit middle phalanx, a bilateral finding (11-year-old male).

FIGURE 8-21. Hypertrophy (arrows) of the margins of the proximal phalangeal distal metaphyses.

FIGURE 8-22. Short second, third, and fourth toe proximal phalanges caused by premature physeal closure (10-year-old female).

FIGURE 8-23. Absent ungual tuberosities (5-year-old female).

FIGURE 8-24. Circular radiolucencies (arrows) in the diaphyses of the second through fourth toe proximal phalanges, a bilateral finding (10-yearold female). Distal Tibia and Fibula The distal tibial epiphysis is very irregular in form. In both the anteroposterior (or mortise) and lateral ankle views, the physis appears widened medially and anteriorly, respectively, and somewhat undulating (Figure 8-31). This “hump” has been described by Kump38 and has since been referred to as “Kump’s hump”39; Peterson40 and Love et al.25 refer to it as “Poland’s hump,” which may be related to a description by Poland41 in a publication from 1898. Similar irregularities of the lateral distal fibular epiphysis can be seen in the anteroposterior or mortise ankle views.3 These common findings should not be misinterpreted as pathologic. Occasionally, a

lucent defect or rarefaction is seen along the lateral aspect of the distal fibular metaphysis (Figure 8-32) in the anteroposterior or mortise ankle views.3 This finding may or may not present bilaterally and can be misdiagnosed as a destructive bone lesion.4

FIGURE 8-25. Irregular outline (arrows) of first metatarsal basal physis and adjacent metaphysis and epiphysis (12-year-old male). A: Dorsoplantar view. B: Lateral view.

FIGURE 8-26. The bases of the second through fourth metatarsals are quite irregular (arrows), possibly initial ossification of supernumerary epiphyses (2-year-old male).

FIGURE 8-27. Jagged posterior calcaneal metaphysis in a 10-year-old male.

FIGURE 8-28. Reverse calcaneal spur (arrowhead) (6-month-old male).

FIGURE 8-29. Odd calcaneal form. A: Posterior half is very large relative to anterior segment. B: Prominent superior aspects of anterior and posterior articular margins with resultant exaggerated concavity between posterior articular surface and bursal projection. A bilateral finding in this 8-year-old male. C: Large and elongated posterior section with uncharacteristic flattening of superior and inferior surfaces, present bilaterally.

FIGURE 8-30. Elongated talar head and neck relative to the body in a 5year-old female.

FIGURE 8-31. Irregular outline (arrows) of distal tibial physis in a 9-yearold female (aka Kump’s hump or Poland’s hump). A: Anteroposterior view. B: Lateral view.

FIGURE 8-32. Lucent defect (arrows) at distal fibular metaphysis. A: Sixyear-old male. B: Thirteen-year-old male. REFERENCES   1. Dorland’s Illustrated Medical Dictionary. 31st ed. Philadelphia, PA: Saunders; 2007.   2. Resnick D. Bone and Joint Imaging. 2nd ed. Philadelphia, PA: Saunders; 1996.   3. McCarthy JJ, Drennan JC. Drennan’s the Child’s Foot & Ankle. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2010.   4. Tachdjian MO. The Child’s Foot. Philadelphia, PA: Saunders; 1985.   5. Schmidt H, Freyschmidt J. Kohler/Zimmer Borderlands of Normal and Early Pathologic Findings in Skeletal Radiology. 4th ed. New York, NY: Thieme Medical Publishers; 1993.

  6. Szaboky GT, Anderson JJ, Wiltsie RA. Bifid Os Calcis. Clin Orthop. 1970;68:136.   7. Uzel M, Cetinus E, Gumusalan Y, et al. Atypical bifid os calcis: a case report. Foot Ankle Int. 2007;28(12):1295.   8. Ogden JA. Anomalous multifocal ossification of the os calcis. Clin Orthop. 1982;162:113.   9. Sever JW. Bifid os calcis. Surg Gynecol Obstet. 1930;50:1012.  10. Cormier-Daire V, Savarirayan R, Unger S, et al. “Duplicate calcaneus”: a rare developmental defect observed in several skeletal dysplasias. Pediatr Radiol. 2001;31:38.  11. Feldman F, Pochaczevsky R, Hecht H. The case of the wandering sesamoid and other sesamoid afflictions. Radiology. 1970;96:275.  12. Schreiber A, Differding P, Zollinger H. Talus partitus. J Bone Joint Surg Br. 1985;67B(3):430.  13. Weinstein SL, Bonfiglio M. Unusual accessory (bipartite) talus simulating fracture. J Bone Joint Surg Am. 1975;57A(8):1161.  14. Roche AF, Sunderland S. Multiple ossification centres in the epiphyses of the long bones of the human hand and foot. J Bone Joint Surg Br. 1959;41B(2):375.  15. Nakashima T, Furukawa H. A rare case of complete proximal epiphyses (so-called pseudoepiphyses) of the metacarpal and metatarsal bones in the human. Ann Anat. 1997;179:549.  16. Lachman E. Pseudo-epiphyses in hand and foot. Am J Roentgenol Radium Ther Nucl Med. 1953;70(1):149.  17. Ogden JA, Ganey TM, Light TR, et al. Ossification and pseudoepiphysis formation in the “nonepiphyseal” end of bones of the hands and feet. Skeletal Radiol. 1994;23:3.

 18. Vilaseca RR, Ribes ER. The growth of the first metatarsal bone. Foot Ankle. 1980;1(2):117.  19. Mathis SK, Frame BA, Smith CE. Distal first metatarsal epiphysis: a common pediatric variant. J Am Podiatr Med Assoc. 1989;70(8):375.  20. Chao W. Os trigonum. Foot Ankle Clin North Am. 2004;9:787–796.  21. Dale JD, Ha AS, Chew FS. Update on talar fracture patterns: a large level I trauma center study. Am J Roentgenol. 2013;201(5):1087.  22. Turhan AU, Dinc H, Aydõn H, et al. An accessory ossification centre in the calcaneus with talonavicular and second metatarsocuneiform coalitions. Eur Radiol. 1999;9:481.  23. Krause J, Rouse A. Accessory calcaneus: a case report and literature review. Foot Ankle Int. 1995;16(10):646.  24. Baghla DPS, Shariff S, Bashir WA. Acquired cavo-varus deformity caused by an accessory calcaneus: a case report and literature review. Skeletal Radiol. 2010;39:193.  25. Love SM, Ganey T, Ogden JA. Postnatal epiphyseal development: the distal tibia and fibula. J Pediatr Orthop. 1990;10:298.  26. Mandalia V, Shivshanker V. Accessory ossicle or intraepiphyseal fracture of lateral malleolus: are we familiar with these? Emerg Med J. 2005;22:149.  27. Stanitski CL, Micheli LJ. Observations on symptomatic medial malleolus ossification centers. J Pediatr Orthop. 1993;13(2):164.  28. Powell HDW. Extra centre of ossification for the medial malleolus in children. J Bone Joint Surg Br. 1961;43B(1):107.  29. Selby S. Separate centers of ossification of the tip of the lateral malleolus. Am J Roentgenol Radium Ther Nucl Med. 1961;86:496.  30. Person V, Lembach L. Six cases of tarsal coalition in children aged 4 to

12 years. J Am Podiatr Med Assoc. 1985;75(6):320.  31. Light TR, Ogden JA. The longitudinal epiphyseal bracket: implications for surgical correction. J Pediatr Orthop. 1981;1:299.  32. Jones GB. Delta phalanx. J Bone Joint Surg Br. 1964;46B(2):226.  33. Watson HK, Boyes JH. Congenital angular deformity of the digits. J Bone Joint Surg Am. 1967;49A(2):333.  34. Jaeger M, Refior HJ. The congenital triangular deformity of the tubular bones of hand and foot. Clin Orthop. 1971;81:139.  35. Neil MJ. Bilateral delta phalanx of the proximal phalanges of the great toes. J Bone Joint Surg Br. 1984;66B(1):77.  36. Robinson HM. Symmetrical reversed plantar calcaneal spurs in children. Radiology. 1976;119:187.  37. van Wiechen PJ. Reversed calcaneal spurs in children. Skeletal Radiol. 1987;16:17.  38. Kump WL. Vertical fractures of the distal epiphysis. Am J Roentgenol Radium Ther Nucl Med. 1966;97(3):676.  39. Keats TE. An Atlas of Normal Roentgen Variants That May Simulate Disease. 8th ed. St. Louis, MO: Mosby; 2004.  40. Peterson HA. Epiphyseal Growth Plate Fractures. New York, NY: Springer; 2007:274.  41. Poland J. Traumatic Separation of the Epiphyses. London, England: Smith, Elder & Co.; 1898.

SECTION 3 Systematic Approach to Bone and Joint Abnormalities

9 Principles of Radiographic Interpretation ROBERT A. CHRISTMAN Radiographic interpretation is an art form. Effective evaluation requires selecting appropriate views for the radiographic study (Chapter 10), use of the proper viewing tools, a consistent means for viewing digital images on a monitor or placing films on a view box, understanding basic concepts of image formation, and using a systematic approach for image assessment (Chapter 11). This chapter focuses on the fundamentals of radiographic image formation and evaluation. X-RAY FILM–VIEWING TOOLS The minimum requirements for an x-ray film–viewing room include a view box, magnifying glass, and spotlight. View Box A view box should be used whose size is consistent with the size of film being viewed. For example, if 10- × 12-in films are used, then the view box size should be 10 × 12 in or multiples thereof. If the lit viewing area is larger than the film, extraneous light escapes around the sides of the film and the user perceives less useful visual information from the radiograph. The radiograph appears more dense (blacker) if extraneous light surrounds the film.1 To illustrate, place a high-contrast dorsoplantar foot radiograph on a fully lit view box that accommodates two or more films. Notice how parts of the image that are relatively darker, such as the toes, are barely visible. Then turn off or cover all remaining view box lights except that being used to view the radiograph. Finally, take the x-ray folder and cover up the entire foot except the toes. Notice how the toes become more visible with each step. Therefore, not only should the size of the view box correspond to the film size but also each single film–viewing area should have its own on/off power switch and there should be reflectors that divide and separate each 10- × 12-

in viewing area so that there is no crossover illumination. Overhead lighting, lamps, and sunlight can also impair the viewer’s ability to interpret radiographs by producing glare and surface reflections. Therefore, all other light sources should be turned off or subdued when viewing films. Cool white or daylight fluorescent light bulbs are used as the view box light source. Cool white bulbs are preferable; daylight bulbs are “bluer” and do not appear to provide as much light through the illuminator surface and film. If the view box uses more than one light bulb, be sure to use the same type of bulb in all fixtures. Colored (green) fluorescent bulbs are also available; one manufacturer claims they are less straining to the eyes. This is fine if you do not mind looking at a green radiograph! (Personally, I favor the black and white/shades of gray variety.) A white, transparent illuminator surface diffuses light of uniform brightness for viewing the radiograph. This surface must be kept clean and free of scratches. Dirt or other foreign matter on the illuminator surface will appear as artifact or may mimic pathology in the radiograph. Magnifying Glass Every viewing area should have an accessible magnifying glass. Glasses come in many shapes, sizes, and powers. Invest in a precision-cut piece of glass. Inexpensive, bargain-brand magnifiers tend to distort the image because they are improperly manufactured. A small handheld magnifying glass suffices. Spotlight A spotlight generally uses an incandescent bulb for its light source. The bulb wattage is typically 60 or 75 W. The device can be purchased with or without a pedal or dial for variable adjustment of light intensity. When viewing a film over a spotlight, be extremely careful not to damage the film. The proper method for viewing the film with this light source is to slowly but constantly move and/or rotate the film. A film will warp if it is held over the high-intensity light source too long. (Consider the following

analogy: If your hand is held still over a lit match, the skin will burn. However, continually moving your hand in a circular motion over the flame will prevent the skin from burning.) The emulsion may eventually crack in the damaged area, ruining the image. Another way of knowing if the film is burning is to continually hold a finger against the film near the area being highlighted. If your finger feels intense heat, then you know the film is being damaged. Burning x-ray film also emits an unpleasant smell. This damage, by the way, is permanent and cannot be reversed. DIGITAL IMAGE–VIEWING TOOLS Digital images, whether viewed at a PACS (picture archiving and communications system) workstation or on a personal computer, are best displayed on a high-resolution monitor. Two types of monitors, the cathode ray tube (CRT) and the liquid crystal display (LCD), are used. Images of the foot and ankle are best viewed on a portrait monitor with a screen resolution of at least 1600 × 1200 (2k). The LCD flat panel monitor should be viewed straight on, as is recommended for most LCD television monitors. Ambient light, including overhead lighting, lamps, and sunlight, can produce glare and surface reflections, especially on CRT monitors, that will impair the viewer’s ability to interpret radiographs. Therefore, all other light sources should be turned off or subdued when viewing images. A CD/DVD drive is necessary for viewing studies performed outside your practice setting that may be brought in by the patient. Additional DICOM software is necessary to view and manipulate the images on a personal computer. CDs provided by outside imaging centers typically include software to “play” or view the images, but do not allow the end user to “record” or copy the images. The provided software is a limited version of the full software, but allows the end user to manipulate the image’s brightness (window level), contrast (window width), extraneous light (“shuttering”), orientation, and size; in addition, the image can be magnified and measurements made.2 IMAGE PLACEMENT How should radiographic images be viewed? The conventional method

entails positioning the image so that the patient is facing the viewer. For example, the patient’s left ankle is seen on the viewer’s right. And, by this method, the anteroposterior/dorsoplantar foot view would be placed so that the toes are at the bottom, rearfoot at the top. Most people do not view foot images in this manner, however. The pedal image is commonly positioned (and reproduced in textbooks or journals) so that the patient’s right foot is on the viewer’s right and the toes are at the top. Merrill’s atlas mentions the exception as to how foot images are viewed, with the toes pointed up.3 Although the foot may appear more visually acceptable in this position, it does not simulate how the patient’s foot would be examined clinically for correlation to the radiograph. Whatever method is employed, the important factor is consistency; that is, position the image consistently in the same way (however that may be) each time it is evaluated. Keep in mind that the radiographic image is two-dimensional and does not have depth. The image will not be altered no matter how the image is viewed. Oblique images of the foot are best viewed in the same fashion that an anteroposterior/dorsoplantar foot view is positioned. There are three reasons for this: First, this image orientation simulates an anteroposterior/dorsoplantar presentation except that the foot is rotated along its coronal plane. Second, if you are viewing a bilateral oblique study with both feet on the same image, it is much easier to compare symmetry between the feet. And, lastly, the image placed in this position fits best on a view box sized for 10- × 12-in films or a portrait monitor. Lateral views of the foot are certainly best analyzed with the sole of the foot positioned horizontally. Bilateral lateral foot views on the same image should additionally be analyzed with the feet vertical and sole to sole, if possible, to compare for symmetry. Ankle views should be positioned by conventional means as already noted, that is, as if the patient were facing you. The patient’s left extremity is on the viewer’s right, and vice versa. BASIC CONCEPTS OF IMAGE FORMATION The radiographic image is a two-dimensional representation of a threedimensional (3D) body part. Individuals vary in their ability to apply spatial

relationships to radiography. To some it comes easily; others need to develop and/or strengthen this visual skill. Spatial aptitude/skills tests are available on the internet that can measure your ability to mentally manipulate shapes in two dimensions or to visualize three-dimensional objects presented as twodimensional pictures; these sites also offer study guides to strengthen these skills. An understanding of certain fundamental principles is necessary before attempting to interpret radiographs. They are as follows: First, any substance has a characteristic radiographic appearance that depends on its atomic number, thickness, and form; and, second, the image is a summation of anatomic shadows. Let us first briefly review how the radiographic image is formed. The useful x-ray beam comes into contact with the body part (in this case, the foot). The x-rays either travel through the foot and exit directly with no interaction or are absorbed (and/or scattered) by it. Those that directly exit the foot come into contact with the x-ray image receptor. The x-ray photons (and light photons emitted from a phosphor screen) form or are converted into the radiographic image (depending on the type of image receptor system used). The x-ray beam is composed of individual x-ray photons with varying degrees of energy. Higher-energy x-ray photons are able to penetrate matter more readily than lower-energy photons. Therefore, higher-energy photons stand a greater chance of reaching the film; the foot absorbs the lower-energy photons. X-rays interact with matter in different ways (see Chapter 1). X-rays may be scattered in different directions (coherent scattering and Compton scattering), or atoms can absorb them (photoelectric effect). Scattered x-rays that exit the patient can fog the image and reduce contrast and are of no diagnostic value. Whether an x-ray photon is absorbed or not depends on two factors: the energy of the photon (as already discussed) and the atomic number of the substance. The difference between those x-rays absorbed by the foot and those that penetrate it directly is known as differential absorption. Objects with greater atomic numbers absorb x-rays more readily than do those with lower atomic numbers. Therefore, the element lead, with an atomic number of 82, absorbs more x-ray photons than bone, made of calcium, which has an

atomic number of 20. Another factor to consider is the thickness of the material. The thicker any particular substance is, all other factors remaining unchanged, the more xrays it absorbs. This process is known as attenuation. The term radiopaque applies to those substances that absorb x-rays; representative areas appear white in the radiographic image. The term radiolucent refers to substances that permit x-ray penetration more readily; representative areas appear dark or black in the image.4 When dealing with shades of gray, as in radiography, radiodensities are discussed in relative terms. Bone is radiopaque (“whiter”) relative to muscle, but radiolucent (“blacker”) compared to a heavy metal such as lead. How does all this relate to formation of the radiographic image? Because radiopaque objects absorb more x-ray photons than those that are radiolucent, fewer x-rays reach the image receptor. Therefore, a radiopaque object appears “whiter.” Figure 9-1 lists matters found in the foot and their relative radiodensities.

FIGURE 9-1. Relative radiodensities of matter seen in the foot. A thicker foot appears more radiopaque (“whiter”) than a thinner foot, assuming all other factors, such as radiographic technique, remains the same when exposing the two feet. Also, areas containing fluid (e.g., edema) add density to that part of the image. Next we need to consider the form of the object being viewed radiographically. Interpreting radiographs requires imagination and logical analysis. This is especially true if the object in question is not parallel to the x-ray image receptor or has a complex form. (You are encouraged to read the introductory chapter of Fundamentals of Radiography, by Lucy Frank Squire.5 He presents an excellent discourse on this subject matter.) If a solid, flat object such as a quarter lies parallel to the x-ray image receptor and the x-ray beam is perpendicular to the object and the image receptor, the

shape of the quarter will be recognized true to form, that is, circular. However, if this same object is turned obliquely or vertically, its shape will become oblong or rectangular (Figure 9-2). Mentally visualize how anatomic structures change and move relative to one another from any angle. The interpreter must eliminate any preconceived opinions about an image, which is a skill that takes time and repetition to acquire. There are several perceptual problems that may be encountered. For example, staring at a part of the image that seems complex does not improve visibility of the image. In contrast, scanning the image improves the ability of retinal cone and rod cells to view dim objects, which are best viewed peripherally; and rod cells function better when image information keeps changing.6 Also, viewing the image at the same distance limits perception; therefore, when viewing potentially complex areas in the image, view it at varying distances. Let us now consider complex objects. As an exercise, close your eyes and try to imagine what an egg would look like radiographically. Next imagine a conch shell with and without spines. Check your imaginative pictures with Figures 9-3 and 9-4. An x-ray image is not a picture per se, but a collection of shadows. Think of these shadows being superimposed on one another layer by layer. The egg, for example, is not a homogeneous density. The outer shell is made of calcium; the inner fluids and pocket of air are less radiodense. The resultant image is a summation of the inner substances superimposed on the outer shell, which is viewed, roughly speaking, as a flat sheet. Note the radiopaque periphery or margin of the egg. As a rule, the curved portion of an object will appear radiopaque if it is perpendicular to the image receptor and parallel to the x-ray beam. It will appear relatively radiolucent if it is parallel to the image receptor and perpendicular to the x-ray beam, as in the center of the egg. Did you imagine the conch shell to appear the way it does in Figure 9-4? The shell is physically the same density throughout, yet many shadows of differing densities are appreciated radiographically. Apply the concept just described regarding the curved object and its position relative to the image receptor and x-ray beam to the conch shell. Tubular objects are commonly encountered in a foot radiograph. The characteristic image of a tubular object, such as the metatarsal, is shown in Figure 9-5. The periphery of the metatarsal is radiopaque. This is because the

curved margin of the shaft is perpendicular to the image receptor and parallel to the x-ray beam. The center consists of a less dense material, the bone marrow; cortical bone is superimposed on the marrow but is nearly parallel to the image receptor and perpendicular to the x-ray beam. Therefore, the center of the tube appears relatively lucent.

FIGURE 9-2. Radiograph of a familiar object (a quarter). A: Lying flat. B: Oblique. C: On its edge.

FIGURE 9-3. Radiograph of an infertile emu egg. An object with complex form can look different in many ways simply because of the way it is positioned relative to the x-ray image receptor and beam (Figure 9-6). Interpreting foot radiographs is particularly challenging because multiple bones are superimposed on one another, especially in the tarsal region. You will eventually become familiar with the radiographic appearance of each bone. Always think in layers: Shadows are layers of objects superimposed on one another. The boundary or outline of an object generally appears as a well-defined shadow on the radiograph. Separate each shadow mentally, subtracting everything else. This concept is extremely important when distinguishing normal radiographic anatomy from abnormal. Finally, think in terms of three dimensions. To mentally formulate a threedimensional picture, two views that are perpendicular to one another are necessary. This is because radiographs can only demonstrate two of the three imaging perspectives, anterior–posterior, medial–lateral, and superior– inferior. The most common examples are anteroposterior (dorsoplantar) and lateral views of the part in question. This is best illustrated by the fracture in Figure 9-7. To summarize, interpreting a radiograph is an exercise of imagination and reasoning. The formation of the image depends on the object’s atomic numbers and thicknesses. The image is a summation of shadows superimposed on one another. So think in layers when viewing a radiograph. The form of an object may be easily recognized if it is parallel to the image receptor and perpendicular to the x-ray beam; however, a familiar object may look quite unfamiliar if it is positioned differently relative to the image receptor and x-ray beam. Finally, think in terms of three dimensions, always correlating the two-dimensional radiographic shadows to the threedimensional object.

FIGURE 9-4. Radiograph of a conch shell. A: With spines. B: Without spines.

FIGURE 9-5. Radiograph of a common tubular object found in the foot: the first metatarsal bone.

FIGURE 9-6. A and B: Radiographs of a bottle of glue (Barge cement) and applicator brush in two views perpendicular to one another.

FIGURE 9-7. Views of a distal fibular fracture. A: Anteroposterior. B: Lateral. Because the fracture is displaced only in the sagittal plane, it is nearly imperceptible in the anteroposterior view but fairly obvious in the lateral view. REFERENCES   1. Fuchs AW. Principles of Radiographic Exposure and Processing. 2nd ed. Springfield, IL: Charles C Thomas; 1958.   2. Carter C, Veale B. Digital R adiography and PACS. 2nd ed. St Louis, MO: Mosby; 2014.   3. Frank ED, Long BW, Smith BJ. Merrill’s Atlas of Radiographic

Positioning and Procedures: 3 -Volume Set. 12th ed. St. Louis, MO: Mosby; 2011.   4. Dorland’s Illustrated Medical Dictionary. 31st ed. Philadelphia, PA: Saunders; 2007.   5. Novelline RA. Squire’s Fundamentals of Radiography. 6th ed. Cambridge, MA: Harvard University Press; 2004.   6. Carlton RR, Adler AM. Principles of Radiographic Imaging: An Art and Science. 5th ed. Delmar, CA: Cengage Learning; 2013:220–227;chap 14.

10 View Selection for the Radiographic Study ROBERT A. CHRISTMAN Careful thought and consideration are necessary to select the positioning techniques that optimally demonstrate the area in question. One must be able to predict, before the study is ordered, how each bone will appear in every view. This is not easy. In addition to the differing appearances of the foot in each view, the appearance of the bones in a pronated foot, for example, is quite different from those in a supinated foot. Competency, therefore, requires continual and repetitive review of radiographs, paying special attention to the form, position, density, and architecture of each bone. Selecting positioning techniques for a foot or ankle study should not be performed as a routine. However, it has become common practice to perform a standard set of positioning techniques for these studies; emphasis is not placed on the specific clinical area of symptomatology but on the views that show the most bones with the least amount of superimposition. They include the dorsoplantar, lateral, and medial oblique foot views and the anteroposterior, mortise, and lateral ankle views. Alternative views that may be more appropriate are not initially considered in this scenario. Because of this approach, some pathologic conditions may not be recognized. It is not uncommon for certain practices or procedures to continue for decades without being questioned. This, unfortunately, does not make such procedures correct. One example is the routine bilateral radiographic foot or ankle study. Another is the number of positioning techniques performed. A third example concerns the technical aspects of a weight-bearing foot study. These are discussed in turn. Overutilization of radiographic studies is also briefly examined. The last half of this chapter concentrates on which views best demonstrate each bone and specific osseous landmarks. A method is presented that simplifies the process by which views are selected.

BILATERAL VERSUS UNILATERAL STUDY No written rule states that an extremity study should routinely be performed bilaterally. Some practitioners, however, habitually obtain views of both extremities for every radiographic study performed. The reasoning behind this practice has been to provide a comparison view of the opposite extremity, a baseline for future reference, or a means for biomechanical (orthomensurative) examination. Use of comparison views should decrease as the practitioner’s level of experience increases. The more familiar one becomes with normal radiographic anatomy, the less necessary are comparison views. Use of a reference standard, such as that provided in the radiographic anatomy section of this text, may, in many instances, be sufficient to replace the need for a comparison view of the opposite extremity. Furthermore, short tubular bones, such as the metatarsal and phalanx, rarely require comparison views of the opposite extremity; the remaining tubular bones of the same extremity serve as useful standards for comparison. To become familiar with the multiple superimposed shadows in the midfoot, rearfoot, and ankle regions, one must study these areas carefully and correlate them to the gross anatomic specimen routinely, not just when searching for a specific pathology. Bilateral studies have also been advocated for the pediatric radiographic study, especially for epiphyseal injuries.1 However, irregular ossification, multiple ossification centers, and accessory ossification centers are frequently encountered as variations of normal development; their differentiation from abnormality may be impossible, even with comparison views of the opposite extremity. Comparison views of the pediatric foot or ankle may actually be misleading more often than they are beneficial. For example, developmental variants are frequently unilateral. The absence of a variant in the opposite extremity encourages the misdiagnosis of fracture or other pathology.2 Studies have shown that pediatric radiologists do not consider bilateral studies routinely necessary.3,4 The need for bilateral studies is obvious for patients who exhibit symptomatology in both extremities. They are also valuable for assessing joint disease. However, it is illogical to order and perform bilateral

radiographic studies routinely for every patient. Direct trauma, for example, does not require radiographic examination of the opposite extremity if the fracture is obvious. Comparison views of the opposite, asymptomatic extremity may be useful when a questionable lesion or finding is present in the affected extremity. In most instances, the questionable finding only appears in one view. A comparison study of the opposite extremity can, therefore, be limited to the view in question. Other supplementary comparison views offer no additional diagnostic information and expose the patient to additional potentially harmful radiation. If you were the patient, would you appreciate the physician ordering radiographic studies that were not going to contribute to the diagnosis or treatment of your condition? BOX 10-1 Potential Indications for Radiographic Study of the Foot or Ankle (MINTCAP) Metabolic, endocrine, nutritional disorders Infection Neoplasm Trauma Congenital/developmental Arthritis Positional deformity A radiographic study should not be ordered or performed until after an adequate history and physical examination has been completed on the patient and then only if clinically indicated. Baseline radiographic studies without clinical indication should, therefore, not be performed and are discouraged. Nor should radiography for biomechanical examination be performed routinely. The foot should first be analyzed clinically; radiographs should only be obtained if the examination suggests that an osseous structural or

positional abnormality exists and could be contributing to the presenting concern and if the outcome of the study could affect the treatment rendered. Potential indications for radiographic study of the foot or ankle are listed in Box 10-1. NUMBER OF VIEWS How many views are necessary for evaluating the foot or ankle? This depends entirely on the provisional diagnosis (or diagnoses) and the area of concern. For example, when evaluating the first metatarsophalangeal joint for degenerative joint disease, only dorsoplantar and lateral views are necessary to make the diagnosis. An oblique view ordinarily does not offer any additional information in the diagnosis of osteoarthritis unless the physician is looking for a more specific condition or lesion, such as a loose osseous body. Controversy exists regarding the number of views necessary to assess ankle trauma. The practice of obtaining three views is advocated widely in the literature.5–7 These views include the anteroposterior, lateral, and oblique (typically the mortise). However, others suggest that the number of ankle views be limited to two (anteroposterior and lateral) or even eliminated in the absence of soft tissue swelling.8–11 A rational, analytic approach for selecting a radiographic study is to first obtain two or three views that best demonstrate the area in question. Typically this would include dorsoplantar and lateral views; an oblique or axial view may be indicated if the area of concern is best seen in one. If your diagnosis is confirmed, no other views are necessary. If a unilateral study is inconclusive or a questionable lesion is seen, additional views should then be obtained of the same extremity. If comparison views of the opposite extremity are warranted, select only those views that were questionable from the initial study. Of course, conditions that can demonstrate generalized radiographic findings, such as inflammatory joint disease, require examination of other regions and/or extremities. Because the distribution of radiographic findings and patterns of joint involvement are important aspects for the evaluation of joint disease, bilateral dorsoplantar and lateral views are advocated.

OVERUTILIZATION/APPROPRIATENESS Overutilization of radiography has been defined as “excessive irradiation per unit of diagnostic information, therapeutic impact, or health income”; it can include excessive radiation per image, excessive images per examination, or excessive examinations per patient.12 Regarding foot radiography, excessive radiation per image can be minimized by limiting the size of the x-ray beam to the area in question (also known as collimation), using appropriate image receptors, and shielding the patient with a lead apron; excessive images per examination can be limited by initially obtaining only the basic views to assess the area in question and by preventing repeat examinations due to poor positioning and exposure technique; and excessive examinations per patient can be reduced by depending more on clinical findings than relying on radiographs for follow-up examination, by educating the demanding patient that a radiographic study may not be indicated, and by imagining that the cost of the study is being funded by the patient, not reimbursed by insurance. Examples have already been briefly addressed citing areas of potential overuse of radiographic studies, including bilateral comparison studies, the number of views selected for any particular study, and whether the study should be performed at all. Several factors affect one’s decision to obtain radiographic studies (Box 102).13,14 A primary factor is the training and experience of the practitioner. Someone trained to order bilateral comparison views for every patient would certainly perpetuate this activity in practice. Experience, addressed earlier, should also have an effect: The number of studies and/or views ordered should decrease as practical experience increases. Another factor addresses the patient evaluation: Radiographs should never replace a thorough history and physical examination. Studies have suggested that ankle fractures can be discriminated from nonfractures by considering specific clinical variables.15,16 These criteria (Box 10-3), referred to as the Ottawa Ankle Rules (OAR), were designed to minimize false-negative results. The OAR boasts nearly 100% sensitivity and has since been reproduced and validated in different types of medical settings; it has reduced the cost of treating ankle injuries by reducing the necessity of radiographic studies.17–20 In essence, the need for radiographs of the ankle in adults with acute ankle injury is reduced at least 30%. A modification to the OAR (the “Buffalo rules”) (Box

10-3) has shown reduction of greater than 50% while maintaining the high sensitivity.21 Boutis et al.22,23 have validated what is referred to as the “Low Risk Ankle Rule” to safely reduce radiography in children with acute ankle injuries (Box 10-3). The Low Risk Ankle Rule boasts 100% sensitivity for fractures of the distal fibula and has the potential to reduce radiographic studies by nearly 60%.24,25 BOX 10-2 Factors Affecting Radiographic Decision Making Physician training and experience Patient examination Medicolegal justification Reassurance Patient insistence BOX 10-3 Ankle Trauma Rules Radiography indicated if:

Ottawa Ankle Rules (adults)

• Tenderness over the inferior or posterior aspect of either malleolus, including the distal 6 cm • Inability to bear weight (four steps taken independently, even if limping) at the time of injury and at the time of evaluation Radiography indicated if:

• Tenderness over the midportion or crest of the bone from the tip of each Modified Ottawa Ankle Rules (adults) malleolus to 6 cm proximal (aka “Buffalo Rules”)

• Inability to bear weight (four steps

taken independently, even if limping) at the time of injury and at the time of evaluation Radiography may not be indicated to further exclude high-risk ankle injury if tenderness and swelling isolated to: Low Risk Ankle Rule (children)

• Distal fibula and/or • Adjacent lateral ligaments distal to the tibial anterior joint line

Diagnostic imaging criteria have been developed by the American College of Radiology (ACR), which are evidence-based guidelines for making appropriate imaging decisions and enhancing quality of care.26 More specifically, they have developed musculoskeletal imaging guidelines for acute foot trauma, chronic foot pain, suspected ankle fracture, chronic ankle pain, and suspected foot osteomyelitis in patients with diabetes mellitus.27 Also relevant to the foot practitioner are guidelines for imaging soft tissue masses and primary bone tumors. Nonclinical factors may significantly affect the practitioner’s decision to perform a radiographic study. Radiographs are routinely ordered by many for patients presenting with a history of trauma, with the intent of providing medicolegal documentation. Long13 poses this dilemma: “If the justification for ordering a radiograph is to protect the physician, one wonders what number of those radiographs are clinically necessary.” Radiographic studies are also ordered for reassurance, especially after orthopedic surgical procedures and fractures. Generally speaking, most follow-up radiographic examinations only require reassessment at 3- to 4-week intervals, unless clinical history or physical evaluation warrants otherwise. Examples of the latter include postreduction, reinjury, and infection. Another nonclinical factor includes the patient’s insistence. The practitioner should be prepared to educate the patient regarding the determination for radiographs or against their inclusion in the diagnostic workup. The goal to controlling overutilization is learning diagnostic restraint.28

Occasionally we need to rely on our clinical wisdom; a condition does not always have to be “ruled in or out” by performing one more test, in this case, the radiographic study. Will the outcome of the study affect the treatment instituted? We also must remember that there are possible dangers associated with low levels of ionizing radiation, including diagnostic x-ray studies.29,30 Ask yourself if the radiographic study really is necessary.31,32 Finally, determine whether or not another diagnostic technique may be more appropriate to assess the underlying problem.33 TECHNICAL CONSIDERATIONS Bilateral studies should be performed as individual studies of each extremity. This limits and directs the x-ray beam to the part under study. X-ray beam limitation (collimation) reduces scatter radiation that otherwise may be absorbed by the patient. For example, bilateral dorsoplantar views of the feet should be performed so that individual exposures of each foot are obtained. The useful x-ray beam can then be collimated to the individual foot (a 5- × 12-in area for a 10- × 12-in image receptor). Exposure of both feet together requires collimation to an area that is much larger (10 × 12 in for a 10- × 12in image receptor). Scatter radiation increases when collimating to a larger area; this low-energy radiation is easily absorbed by the patient, increasing their dose of ionizing radiation, and can impair the quality of the image. Image quality can also be affected by the direction of the central x-ray beam. A central x-ray beam that is directed to the individual foot or area in question reduces geometric blurring and distortion, resulting in an image that is truer in size, shape, and position. A basic principle in radiography is to have the x-ray central beam directed perpendicular to both the subject and the image receptor. Oblique views of the feet should, therefore, not be performed weight bearing. Significant distortion of the image results when the central x-ray beam is directed at a 45° angulation to the image receptor (Figure 10-1). Oblique foot views should be performed non–weight bearing and with the foot turned 45° in the coronal plane so that the x-ray beam is directed perpendicular to both the foot and the image receptor. This positioning technique minimizes image distortion.

FIGURE 10-1. Weight-bearing versus non–weight-bearing oblique view. A: Distorted weight-bearing oblique view with tube head directed at 45°. B: Non–weight-bearing oblique view with tube head directed perpendicular to image receptor and foot turned 45°. A compromise in positioning technique is made with the weight-bearing dorsoplantar view. The central x-ray beam, unfortunately, cannot be directed perpendicular to both the foot and the image receptor. Because the tube head physically cannot be positioned perpendicular to the image receptor with the patient standing and because the metatarsals are declined approximately 15°

(when averaged together), the tube head is directed at a 15° angle from perpendicular so that it is perpendicular to the osseous structures. Minimal image distortion occurs in this view. Occasionally a question arises regarding which side of the foot should be positioned closer to the image receptor. Generally speaking, that aspect of the body farthest from the image receptor appears magnified and less sharp. This is especially true in the chest, which has considerable depth. The size of the heart appears larger in an anteroposterior view than that in a posteroanterior view. However, the foot is not very thick (relative to the chest), and size of the foot bones is not a critical aspect in their assessment. The width of the average foot is approximately 3 in. Magnification of that side farthest from the image receptor is only between 1.08 and 1.14, depending on the sourceto-image distance (40 and 24 in, respectively). (The magnification factor is determined by dividing the source-to-image receptor distance by the sourceto-object distance.34) SELECTION OF POSITIONING TECHNIQUES Before studying the following paradigm, it is important to understand the concept of marginal (or tangential) surfaces. A tangent is defined as a line that intersects a curved surface at a single point of intersection35 (Figure 102). The outermost aspect or margin of a tubular bone, for example, seen in a two-dimensional radiograph represents the three-dimensional anatomic surface that is tangent to the primary x-ray beam. Unfortunately, bones are irregularly shaped and occasionally have flat surfaces. Furthermore, more than one point or surface may be tangent to the x-ray beam in the same plane. For these reasons, I use the word margin, not tangent, to describe the outermost aspect or outline of a bone seen in the two-dimensional radiograph.

FIGURE 10-2. Illustration of a line tangent to a circle.

BOX 10-4 Marginal Bone Surfaces Seen in Each Foot View Foot View Dorsoplantar Lateral Medial oblique Lateral oblique Sesamoid axial Calcaneal axial

Marginal Surfaces Medial, lateral, anterior, and posterior Superior, inferior, anterior, and posterior Superomedial, inferolateral, anterior, and posterior Superolateral, inferomedial, anterior, and posterior Superior, inferior, medial, and lateral Medial, lateral, superior, and inferior

BOX 10-5 Marginal Bone Surfaces Seen in Each Ankle View Ankle View Anteroposterior Mortise Medial oblique Lateral oblique Lateral

Marginal Surfaces Medial, lateral, superior, and inferior Same as anteroposterior, although slightly rotated Anteromedial, posterolateral, superior, and inferior Posteromedial, anterolateral, superior, and inferior Anterior, posterior, superior, and inferior

Boxes 10-4 and 10-5 list the bone margins that are seen in each standard foot and ankle view. The marginal surfaces for most views are straightforward and obvious; however, the oblique views are confusing. Use a foot skeleton to correlate the bone position for each positioning technique to the radiographic image. The general principle of marginal surfaces can be applied to three scenarios: the selection of an appropriate positioning technique, interpretation of a precise anatomic location in the radiograph, and precise clinical correlation to the radiograph. For example, a patient has an ingrown toenail along the superomedial aspect of the hallux distal phalanx and you suspect an underlying subungual exostosis. What is the one view that will image this area tangentially (Figure 10-3)? In essence, the subungual exostosis is being viewed in profile (from the side) with the tangential view (medial oblique) and en face (face on) with the lateral oblique view. Another example: A patient experienced an inversion sprain and has focal pain on palpation of the superolateral aspect of the calcaneal anterior process. What is the one view that will image this area tangentially (Figure 10-4)?

Selecting the positioning technique depends on the anatomic site in question. Appropriate views are recommended below that optimally demonstrate each bone. The “basic study” consists of those views that best isolate or allow visibility of the area in question. Applicable adjunctive views are also included. However, the best view may not necessarily be the one that isolates a bone with little or no superimposition of other structures. Positioning techniques should also be selected based on the marginal surface to be examined radiographically, when appropriate.

FIGURE 10-3. Lesion (x) located superomedially on distal phalanx is isolated and viewed tangentially with the medial oblique view. A: Dorsoplantar view. B: Medial oblique view. C: Lateral oblique view. D: Lateral view. When standard positioning techniques do not reveal suspected pathology, one should consider performing modifications of the standard positioning techniques before expensive cross-sectional imaging studies are ordered. For example, oblique views of the foot can be performed at varying degrees if a specific location or lesion may be viewed better.36 Or additional, nonstandard positioning techniques can be chosen to more clearly view an area of concern. For example, Osher et al.37 have shown the value of oblique forefoot axial modifications.

Hallux The hallux (Figure 10-5) is best seen in the dorsoplantar view. It is also visible in both oblique views, if the hallux is separated from the adjacent second toe. Unfortunately, the hallux is fully superimposed on the remaining lesser digits in the weight-bearing lateral view, although the distal phalanx can be isolated if attention is paid to elevating the hallux prior to exposure. A lateral view of the isolated hallux may be better achieved with a non–weightbearing positioning technique. Lesser Digits The lesser digits (Figure 10-5) are best seen in the dorsoplantar and both oblique views. To minimize superimposition, the digit or digits in question should be separated from the remaining adjacent digits. In the lateral view, the lesser digits are superimposed on each other. However, hammer toe deformity and/or elevation of the second digit may provide partial visibility of the phalanges. The fourth and fifth toes are seen best in the oblique views. These digits are typically adductovarus in position, so a dorsoplantar view of these digits shows them obliquely oriented. A medial oblique view of an adductovarus digit demonstrates a true dorsoplantar perspective of the toe, and a lateral oblique view shows the digit from a lateral perspective.

FIGURE 10-4. Lesion (arrows) located superolaterally on calcaneus is isolated and viewed tangentially with the lateral oblique view. A: Dorsoplantar view. B: Medial oblique view. C: Lateral oblique view. D: Lateral view. Sesamoids The sesamoids (Figure 10-6) are isolated only in the axial view. Although the entire outline of each ossicle is seen in the dorsoplantar view, they are fully superimposed on the first metatarsal. The tibial and fibular sesamoids are superimposed on the first and second metatarsals, respectively, in the medial

oblique view. Superimposition also occurs in the lateral oblique view; however, approximately a third of the tibial sesamoid is isolated and viewed clearly, its inferomedial aspect. The lateral view should not be considered when performing a radiographic study for examination of the sesamoids alone. The sesamoids are fully superimposed on each other as well as on the lesser metatarsophalangeal joint structures in this view. First Metatarsal Optimal views for examining the first metatarsal and metatarsophalangeal joint (Figure 10-6) are the dorsoplantar and lateral views. The first metatarsal is isolated almost entirely in the dorsoplantar view; minimal superimposition of the proximal phalanx base and medial cuneiform is found at its anterior and posterior margins, respectively. Although the first metatarsal is superimposed on the remaining metatarsals in the lateral view, its entire outline can be traced. This should not hinder evaluating the first metatarsal. The metatarsosesamoid articulation is best seen in the axial view. The medial oblique view demonstrates a different aspect of the phalangeal articular surface from that seen in the dorsoplantar and lateral views. The superomedial and inferolateral surfaces are viewed along the margins. Examples of osseous landmarks and pathology identified in this view include the medial tubercle, inferolateral articular surface for the proximal phalangeal base, bunion hyperostosis and accompanying degenerative cysts, insertion site of the peroneus longus tendon, and gouty erosions. Although this view shifts the fibular sesamoid laterally (relatively speaking) so that it is not superimposed on the first metatarsal, it is now superimposed on the second metatarsal.

FIGURE 10-5. Hallux and lesser digits. A: Dorsoplantar view. B: Lateral view. C: Medial oblique view. D: Lateral oblique view. The following landmarks can be identified in the lateral oblique view: the lateral tubercle, inferomedial articular surface for the proximal phalangeal base, insertion site for the tibialis anterior tendon, as well as other structures along the inferomedial and superolateral surfaces of the first metatarsal. Osteophyte proliferation along the superolateral aspect of the first metatarsophalangeal joint is isolated in this view. Second through Fourth Metatarsals The second, third, and fourth metatarsals (Figure 10-6) are best seen in the medial oblique view, because the metatarsal bases are positioned obliquely in the arch of the midfoot. The metatarsal heads are occasionally viewed in the

dorsoplantar perspective but are typically everted in this view. The inferolateral articular surface and medial tubercle are isolated at the head’s margins. The metatarsal heads are viewed in the dorsoplantar perspective to slightly inverted in the dorsoplantar view. When inverted, the inferomedial articular surface and lateral tubercle are visible. The bases are rotated and, therefore, obliquely viewed and partially superimposed. The lateral oblique view is not very useful for evaluating the second through fourth metatarsals. However, it does provide visibility of a different aspect of the metatarsal head and neck. This can be valuable when evaluating subtle fractures or other pathology to these areas. All three metatarsals are fully superimposed in the lateral view and cannot be easily visualized. This view, however, can be used to evaluate the positional relationships of the metatarsal heads in the sagittal plane. Obviously, excellent-quality images are necessary for this application. A smooth parabola or transition should be noted, with each metatarsal being only slightly superior to the next from lateral to medial. This view probably represents a more accurate reflection of the positioning of the metatarsal heads than does the sesamoid axial view with its overextension of the toes at the metatarsophalangeal joints. The inferior articular aspect of each metatarsal head is isolated in the axial view.

FIGURE 10-6. Metatarsals and sesamoids (l, lateral tubercle; m, medial tubercle; pl, peroneus longus insertion; ta, tibialis anterior insertion). A: Dorsoplantar view. B: Lateral view. C: Medial oblique view. D: Lateral oblique view. E: Sesamoid axial view.

Fifth Metatarsal The entire fifth metatarsal (Figure 10-6) is isolated in the medial oblique view. The tuberosity and metatarsocuboid articulation can be seen. The fifth metatarsal is clearly visible in the dorsoplantar view except for the proximal articulation. Also, the tuberosity is partially superimposed on the base. The fifth metatarsal can be seen in the lateral view even though there is some superimposition on the fourth metatarsal. The tuberosity is viewed clearly, but the metatarsocuboid joint cannot be appreciated. The lateral oblique view is not useful for evaluating the fifth metatarsal except for pathology involving the head or neck. Cuneiforms The cuneiforms (Figure 10-7) are superimposed on themselves and on neighboring bones in most views. The dorsoplantar view best isolates the medial cuneiform. It can also be seen in the medial oblique view, but it is superimposed on the intermediate cuneiform. Although it is fully superimposed on the remaining cuneiforms in the lateral view, the entire outline of the medial cuneiform can be traced. Most of the medial cuneiform is superimposed in the lateral oblique view; only its inferomedial aspect is isolated. The intermediate cuneiform can best be isolated in a modified medial oblique view. Depending on the foot structure (cavus, rectus, or planus), the foot should be positioned such that the sole of the foot is angled somewhere between 5° and 20° relative to the image receptor. The joint between the intermediate and lateral cuneiforms can be isolated with a 20° medial oblique position; the joint between the intermediate and medial cuneiforms becomes isolated with a 10° lateral oblique position.36 Most of the intermediate cuneiform can be seen in the dorsoplantar view. Only the superoposterior aspect is isolated in the lateral view. The lateral oblique view is not useful for evaluating the intermediate cuneiform.

FIGURE 10-7. Cuneiforms, cuboid, and navicular (t, navicular tuberosity). A: Dorsoplantar view. B: Lateral view. C: Medial oblique view. D: Lateral oblique view. The lateral cuneiform is best isolated radiographically in the medial oblique view. This bone is partially superimposed in the dorsoplantar view. It is fully superimposed on neighboring bones in the lateral oblique and lateral views. Cuboid The cuboid (Figure 10-7) is fully visible in the medial oblique view. It can also be evaluated in the dorsoplantar and lateral views, although there is some superimposition. The outline of the cuboid is barely visible in the lateral

oblique view; it is fully superimposed on multiple structures and cannot be assessed. Navicular The navicular (Figure 10-7) can be assessed in all foot views except the axial views. Evaluation is best accomplished using both the dorsoplantar and lateral views. The tuberosity can be partially seen in the dorsoplantar view. Although the tuberosity is fully superimposed in the lateral view, its outline can still be identified; this view can prove quite valuable when evaluating tuberosity fractures. The navicular tuberosity is best seen in the lateral oblique view. The dorsomedial aspect of the navicular is isolated in the medial oblique view. Talus The entire talus (Figure 10-8) can be identified only in the lateral view. Superimposition prevents adequate visibility in the oblique views. However, the anterior articular facet for the calcaneus is isolated in the medial oblique view. Occasionally the posterolateral trigonal process can be identified with this view, although it is superimposed. Only the talar head and neck are visible in the dorsoplantar view. Calcaneus The entire calcaneus (Figure 10-8) can be seen in the lateral and medial oblique views. The medial tuberosity is isolated in the lateral oblique and lateral views. The sustentaculum tali is visible in the lateral, axial, medial oblique, and lateral oblique views. The middle and posterior talar articular surfaces can be seen in the lateral and axial views. The lateral tuberosity, anterior talar articular facet, and bursal projection are isolated in the medial oblique view. Only the lateral and anterior aspects of the anterior calcaneus are visible in the dorsoplantar view.

FIGURE 10-8. Talus and calcaneus (a, anterior talocalcaneal joint; bp, bursal projection; lt, lateral tuberosity; m, middle talocalcaneal joint; mt, medial tuberosity; p, posterior talocalcaneal joint; st, sustentaculum tali). A: Dorsoplantar view. B: Lateral view. C: Medial oblique view. D: Lateral oblique view. E: Calcaneal axial view.

Distal Tibia The tibial malleolus (Figure 10-9) can be identified in all ankle views, although it is best seen in the anteroposterior view. The lateral view is useful for distinguishing between the anterior and posterior colliculi and differentiating between avulsion fractures of the anterior colliculus versus the os subtibiale. The articular surfaces are seen best in the anteroposterior, mortise, and lateral views. Distal Fibula The fibular malleolus (Figure 10-9) is isolated in the mortise view. It is also visible, with only slight superimposition, in the anteroposterior and medial oblique views. Although the malleolus is fully superimposed on the tibia in the lateral view, its entire outline can still be seen. The lateral view is extremely valuable for evaluation of distal fibular fractures. The articulation between the distal fibula and the talus is best viewed in the mortise view.

FIGURE 10-9. Distal tibia and fibula (a, anterior colliculus; p, posterior colliculus; ptc, posterior talocalcaneal joint). A: Anteroposterior view. B: Lateral view. C: Mortise view. D: Medial oblique view. E: Lateral oblique view. REFERENCES   1. Oloff J. Radiology of the foot in pediatrics. In: Weissman SD, ed. Radiology of the Foot. 2nd ed. Baltimore, MD: Williams & Wilkins; 1989.   2. Ozonoff MB. Pediatric Orthopedic Radiology. Philadelphia, PA:

Saunders; 1979.   3. McCauley RGK, Schwartz AM, Leonidas JC, et al. Comparison views in extremity injury in children: an efficacy study. Radiology. 1979;131:95.   4. Merten DF. Comparison radiographs in extremity injuries of childhood: current application in radiological practice. Radiology. 1978;126:209.   5. Weissman BNW, Sledge CB. Orthopedic Radiology. Philadelphia, PA: Saunders; 1986.   6. Berquist TH. Fractures/dislocations. In: Berquist TH, ed. Imaging of the Foot and Ankle. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2011.   7. Conrad JJ, Tannin AH. Trauma to the ankle. In: Jahss MH, ed. Disorders of the Foot. Philadelphia, PA: Saunders; 1982.   8. Garfield JS. Is radiological examination of the twisted ankle necessary? Lancet. 1960;2:1167.   9. de Lacey GJ, Bradbrooke S. Rationalizing requests for x-ray examination of acute ankle injuries. Br Med J. 1979;1:1597.  10. Cockshott WP, Jenkin JK, Pui M. Limiting the use of routine radiography for acute ankle injuries. Can Med Assoc J. 1983;129:129.  11. Wallis MG. Are three views necessary to examine acute ankle injuries? Clin Radiol. 1989;40:424.  12. Abrams HL. The “overutilization” of x-rays. N Engl J Med. 1979;300:1213.  13. Long AE. Radiographic decision-making by the emergency physician. Emerg Med Clin North Am. 1985;3(3):437.  14. Hall FM. Overutilization of radiological examinations. Radiology. 1976;120:443.

 15. Diehr P, Highley R, Dehkordi F, et al. Prediction of fracture in patients with acute musculoskeletal ankle trauma. Med Decis Making. 1988;8:40.  16. Stiell IG, Greenberg GH, McKnight RD, et al. Decision rules for the use of radiography in acute ankle injuries. JAMA. 1993;269:1127.  17. Stiell IG, McKnight RD, Greenberg GH, et al. Implementation of the Ottawa ankle rules. JAMA. 1994;271:827–832.  18. Pigman EC, Klug RK, Sanford S, et al. Evaluation of the Ottawa clinical decision rules for the use of radiography in acute ankle and midfoot injuries in the emergency department: an independent site assessment. Ann Emerg Med. 1994;24:41–45.  19. McBride KL. Validation of the Ottawa ankle rules. Experience at a community hospital. Can Fam Physician. 1997;43:459–65.  20. Pijnenburg AC, Glas AS, De Roos MA, et al. Radiography in acute ankle injuries: the Ottawa ankle rules versus local diagnostic decision rules. Ann Emerg Med. 2002;39:599.  21. Leddy JL, Smohnski RJ, Lawrence J, et al. Prospective evaluation of the Ottawa ankle rules in a university sports medicine center. With a modification to increase specificity for identifying malleolar fractures. Am J Sports Med. 1998;26:158–165.  22. Boutis K, Willan AR, Babyn P, et al. Randomized, controlled trial of a removable brace versus casting in children with low risk ankle fractures. Pediatrics. 2007;119:e1256–e1263.  23. Boutis K, Constantine E, Schuh S, et al. Pediatric emergency physician opinions on ankle radiograph clinical decision rules. Acad Emerg Med. 2010;17:709–717.  24. Boutis K, Komar L, Jaramillo D, et al. Sensitivity of a clinical examination to predict the need for radiography in children with ankle injuries: a prospective study. Lancet. 2001;358:2118–2121.

 25. Boutis K, Grootendorst P, Willan A, et al. Effect of the low risk ankle rule on the frequency of radiography in children with ankle injuries. Can Med Assoc J. 2013;185: E731.  26. American College of Radiology. http://www.acr.org/QualitySafety/Appropriateness-Criteria. Accessed February 7, 2014.  27. American College of Radiology. http://www.acr.org/QualitySafety/Appropriateness-Criteria/Diagnostic/Musculoskeletal-Imaging. Accessed February 7, 2014.  28. Reuben DB. Learning diagnostic restraint. N Engl J Med. 1984;310:591.  29. Boice JD. The danger of x-rays: real or apparent? N Engl J Med. 1986;315:828.  30. Upton AC. The biological effects of low-level ionizing radiation. Sci Am. 1982;246:41.  31. Rigler LG. Is this radiograph really necessary? Radiology. 1976;120:449.  32. McClenahan JL. Wasted x-rays. Radiology. 1970;96:453.  33. Palmer PES, Cockshott WP. The appropriate use of diagnostic imaging. JAMA. 1984;252:2753.  34. Bushong SC. Radiologic Science for Technologists. 10th ed. St Louis, MO: Elsevier; 2013.  35. Funk and Wagnall’s New Comprehensive International Dictionary of the English Language. Newark, NJ: Publisher’s International Press; 1982.  36. Montagne J, Chevrot A, Galmiche JM. Atlas of Foot Radiology. New York, NY: Masson; 1981.  37. Osher LS, DeMore M, Atway S, et al. Extended pedal imaging via modifications of the traditional forefoot axial radiographic study. J Am

Podiatr Med Assoc. 2008;98(3):171.

11 Systematic Evaluation of Bone and Joint Abnormalities ROBERT A. CHRISTMAN The radiologic thought process entails three processes: Detect, Describe, and Differential diagnosis (the 3 Ds). The interpreter must first recognize or detect an abnormal finding, describe the abnormal finding by using appropriate terminology, and, finally, determine a list of differential diagnoses based on the findings. This is no different from evaluating a new patient clinically; during the physical examination the clinician is looking for abnormal findings; if an abnormality is detected (e.g., a skin lesion), it must be described appropriately (papule, macule, scale, etc.); differential diagnoses are then determined (psoriasis, tinea, etc.). The purpose of this chapter is twofold: (1) to illustrate detectable abnormal findings with appropriate terminology to describe them and (2) to provide a systematic method to evaluate a radiographic study and report the findings. DETECT To detect an abnormal finding, one must be able to predict the normal and variant expectations radiographically for each and every view. If not, normal and variant features of a structure will be misinterpreted as abnormal. Abnormalities are recognized and appreciated more readily if judged against the normal expected appearance of each structure. At the outset of image evaluation, consider all radiographic shadows normal anatomy until proven otherwise. If a shadow cannot be explained as normal radiographic anatomy, one can conclude that it is either variant or abnormal. Familiarity with the normal and variant expectations takes time and requires looking at numerous radiographic studies. When doing so, consider all anatomical structures in every available view. Avoid the habit of immediately honing in on the area of question. Not only may you see other pathology but also as you analyze all anatomic structures, you will become more and more

familiar with the normal and variant presentations. This, in turn, will greatly enhance your interpretation skills. Do not rely on solely memorizing each and every pathologic picture; varying radiographic presentations are seen in many disorders. To complicate matters, since bone has a limited response to disease processes, different pathologies may have similar radiographic presentations. The challenge is to recognize that an abnormality exists. A common pitfall when interpreting foot and ankle radiographs is mistaking normal anatomic shadows for pathology. It is, therefore, strongly recommended to keep at hand a loosely articulated set of foot and ankle bones when viewing radiographs of the same. (Familiarity with normal and variant radiographic anatomy cannot be emphasized enough.) DESCRIBE Anatomical structures display four primary radiographic features (Box 11-1). Every bone in a foot or ankle radiograph has a characteristic (1) position relative to other bones, (2) form in each view, (3) combination of densities, and (4) architecture. Therefore, if a bone is abnormal, it will display an abnormality of position, form, density, or architecture. This same terminology also applies to individual lesions found in bone. The few soft tissue structures encountered, such as the Achilles tendon and plantar fascia, have a characteristic position, form, and density in the radiographic image; however, most findings detected in the soft tissues are abnormalities of density. BOX 11-1 Primary Features in Anatomical Structures Position Form Density Architecture

Position Appropriate terminology used to describe position abnormalities is listed in Box 11-2. BOX 11-2 Terminology Used to Describe Positional Relationships Apposition Displacement Alignment Angulation Distraction

FIGURE 11-1. Apposition between two articular surfaces (x represents the medial and lateral margins of the first metatarsal base articular surface; o represents the medial and lateral margins of the adjacent medial cuneiform articular surface). A: 100% apposition (normal) between the first metatarsal base and the medial cuneiform. B: Partial (or ≈50%) apposition. C: 0% apposition. In B and C, the metatarsal is displaced laterally (arrow) relative to the cuneiform. All normally occurring bones articulate with at least one other bone. The degree of contact (touching) between their two articular surfaces is referred to as apposition; there normally is 100% apposition between two bones at a joint%1 (Figure 11-1A). (An exception is the talonavicular joint, which also articulates with the plantar calcaneonavicular [spring] ligament.) When there is less than 100% apposition, the finding can be described as partial or no

apposition (Figure 11-1B,C). Preferably, the amount of apposition would be described as follows: There is 50% apposition between the first metatarsal and medial cuneiform. The diagnosis for partial apposition is subluxation; 0% apposition represents dislocation. If there is less than 100% apposition between two bones, then there must have been shifting of position or displacement of the distal bone in some direction relative to the proximal bone (Figure 11-1B,C). For example: The first metatarsal is displaced laterally relative to the medial cuneiform. Alignment conveys the degree of parallelism of two bones.%2 Angulation is a lack of alignment or parallelism of two bones and can be described by the “tilt” of the distal segment relative to the proximal segment (Figure 11-2). (If this is difficult to ascertain, draw two lines, one parallel to the end of each bone; the two lines should cross if angulation is present. The apex formed by the two lines indicates the direction of angulation%3 [Figure 11-2B].) For example: The hallux proximal phalanx is angulated laterally relative to the first metatarsal. The term distraction refers to when two bones are “pulled apart” from each other. There is a separation between the joint surfaces, which occurs without displacement and without ligament rupture.%4 Radiographically this will appear as an increased joint space (Figure 11-3). The same position terminology described earlier is used when describing a fracture. Is there 100% apposition between the two fragments? If not, what direction is the distal segment displaced? Are the two fracture fragments aligned anatomically? If not, what direction is the distal segment angulated? A fracture must be described in two perpendicular planes to fully appreciate its three-dimensional position (Figure 11-4). The axis of each bone has a characteristic position relative to neighboring bones; this is most applicable to the planar views, that is, dorsoplantar (or anteroposterior, if the ankle) and lateral. The angulation between two bones is assessed and compared to “normal” expectations, either by examining the nonangular positional relationships between bones (described later) or by measuring angles between axes. For example, in the dorsoplantar view the

talocalcaneal angle (formed by the talar and calcaneal axes) is normally ≈21°; when the foot is pronated the angle increases (>24°), and it decreases (2 mm on injured foot • >1 mm compared to uninjured contralateral foot Loss of alignment between Dorsoplantar

• Medial base of the second metatarsal and the medial border of the middle cuneiform • Lateral base of the first metatarsal and the lateral border of the medial cuneiform “Fleck sign”—small avulsion fracture from the medial base of the second metatarsal caused by the Lisfranc ligament

Medial oblique (30°)

Loss of alignment between the medial base of the fourth metatarsal and the medial border of the cuboid Loss of alignment of first and second metatarsals superior margin with corresponding cuneiforms

Lateral view

Flattening of the longitudinal arch and/or loss of alignment between the plantar aspect of the fifth metatarsal and the medial cuneiform

Adapted from Eleftheriou KI, Rosenfeld PF, Calder JDF. Lisfranc injuries: an update. Knee Surg Sports Traumatol Arthrosc. 2013;21:1434–1446; Foster SC, Foster RR. Lisfranc’s tarsometatarsal fracture-dislocation. Radiology. 1976;120(1):79–83.

FIGURE 16-16. Posttraumatic arthritis secondary to Hardcastle type A lateralizing Lisfranc injury. To assist with the diagnostic injury when equivocal radiographs are obtained, stress abduction views can be performed; this requires anesthesia because the technique causes pain. This test exacerbates the joint instability and diastasis due to the abductory force applied, thus recreating the fracture–dislocation. There is a 10% to 20% false-positive rate, which is likely because there is no standardized stress abduction technique; therefore, doubt in the validity and reliability of the test exists.43 However, CT scans provide enhanced images with less importance on patient positioning and without requiring painful manipulation of the acutely injured foot.42,50,53 The simplicity of patient positioning to optimize visualization and the lack of inducing further pain are the primary reasons that many experts prefer multiplanar reconstruction CT images for Lisfranc injuries, especially when clinical or radiographic assessment is difficult (Table 16-4). CT imaging is also easy to perform while providing enhanced anatomic detail free of superimposed structures in the midtarsus, thus improving sensitivity and diagnostic accuracy. Radiography has missed subtle 1 mm subluxations and most 2 mm subluxations, all of which were detected using CT imaging.54 Haapamaki et al.42 reported that radiographs are 25% to 33% sensitive for diagnosing midfoot fractures and 25% sensitive for diagnosing Lisfranc fracture–dislocations, with 24% false negatives. In addition, 35% of feet had occult fractures in other areas of the foot that were missed with radiography. CT images also show the “fleck sign” and better delineate the extent of the fracture–dislocation, dorsal “step-off,” and other occult fractures and associated injuries, such as chondral and periarticular abnormalities. CT images also provide significantly more information helpful for preoperative planning.39,55–57 Leenen and van der Werken58 developed a classification system using CT images to assist in determining the appropriate treatment (Table 16-6). Grade I injuries were virtually nondisplaced, requiring immobilization, whereas grade II lesions were displaced less than 50% of the metatarsal diaphyseal width. These injuries were treated with open reduction internal fixation (ORIF), should closed reduction under general anesthesia

fail. Meanwhile, ORIF was done for all grade III injuries, which represented those with total displacement. However, some experts prefer MRI since it also enhances diagnostic accuracy compared to radiography and is better able to depict a diastasis compared to CT imaging.43,48,59,60 MRI is recommended for the subtle abnormality when ligamentous injury is suspected (Table 16-4). Enhanced anatomical detail is clearly evident in MRI, as this imaging technique is superior in depicting ligament abnormalities in the foot.60 More specifically, MRI is ideal for visualizing the Lisfranc ligament and injuries of the complex joint structure.48,60,61 Optimal visualization of the tarsometatarsal joint ligaments occurs in distinct anatomic planes (Table 16-7). However, MRI also allows for identification of non- or minimally displaced injuries and small avulsion fractures, which may be identified by the presence of bone marrow edema, when radiographs and CT images appear normal.56 Sensitivity is significantly greater than radiography, with MRI identifying disruption or a grade II sprain of the plantar bundle of Lisfranc ligament, which has a positive predictive value of 94% and a sensitivity and specificity for diagnosing tarsometatarsal joint instability of 94% and 75%, respectively.43 The most well-known classification schemes are based on the congruency and displacement of the metatarsal bases. Quenu and Kuss62 first classified the tarsometatarsal joint fracture–dislocation in 1909 (Figure 16-18). This scheme, based on frontal plane dislocations, classified the injury into homolateral, partial, and divergent dislocations. Using the Quenu and Kuss framework, Hardcastle et al.41 stratified the injury into a new classification to help guide treatment. This classification was based on dislocations occurring in all planes and consisted of total incongruity, partial incongruity, and divergent injuries. Partial incongruity injuries are subdivided into medial or lateral patterns. Divergent injuries were described as having partial or total incongruity, with the possibility of sagittal plane displacement.41 The most widely accepted classification is the Hardcastle classification with Myerson’s modification, which attempts to assist with treatment planning (Table 16-6, Figure 16-19). While neither of the earlier classification systems aid in treatment planning or prognosis, since they are based on the mechanism of

injury, Myerson et al.63 modified the Hardcastle classification in an attempt to assist with treatment planning and provide prognostic value (Figure 1620). However, the prognostic ability has been questioned due to a weak correlation with outcomes and low interobserver and intraobserver reliability.63 Overall, these classifications provide a good descriptive tool for Lisfranc fracture–dislocations, but the guidance in treatment and forecast of outcomes remain limited. TABLE 16-6   Lisfranc Fracture–Dislocation Classifications

FIGURE 16-17. Divergent Lisfranc fracture–dislocation with lateral cuneiform, navicular, and cuboid fractures (arrows indicate lateral dislocation of the metatarsals). (From Figure 19-11 in Spinosa FA. Classification of fractures and dislocations. In: Christman RA. Foot and Ankle Radiology. 1st ed. St. Louis, MO: Churchill Livingstone; 2003:chap 19.) CUNEIFORM FRACTURE Cuneiform fracture is a rare injury, accounting for 4.2% of tarsal fractures, with isolated cuneiform fractures representing 1.7% of tarsal fractures. Most often, cuneiform fracture is associated with tarsometatarsal joint fracture– dislocation and has been observed with midtarsal joint dislocation. The medial cuneiform is the most commonly injured, and dislocation of any

cuneiform is rare. This may be due to the confined alignment of the midtarsal bones since, unlike the articulations of the middle and lateral cuneiforms, the medial cuneiform lacks an articulation medially. Anatomic Plane for Optimal Visualization of TABLE 16-7   Tarsometatarsal Joint Fracture–Dislocation on Magnetic Resonance Imaging Anatomic Plane for Optimal Anatomic Structure Visualization Oblique axial plane best Lisfranc ligament Intermetatarsal ligament Tarsometatarsal ligament Osseous alignment

Can be viewed in axial, sagittal and coronal planes Coronal images (almost exclusively) Sagittal images Sagittal images for dorsal alignment Oblique axial plane for transverse alignment

Adapted from Foster SC, Foster RR. Lisfranc’s tarsometatarsal fracturedislocation. Radiology. 1976;120(1):79–83; Preidler KW, Yang Y, Brossom J, et al. Tarsometatarsal joint: anatomic details on MR images. Radiology. 1996;199(3):733–736. Cuneiform fractures are classified as avulsion fractures, body fractures, or dislocations/subluxations. Avulsion fractures are seen on the medial side of the medial cuneiform, secondary to traction applied by the anterior tibial tendon (Figure 16-21). Body fractures are typically crush injuries from direct force, and dislocations/subluxations are typically associated with Lisfranc fracture–dislocations. In the case of isolated cuneiform fracture, direct trauma is the most common mechanism of injury. More specifically, direct force accounts for two-thirds of the six reported cases of isolated medial cuneiform fracture, with the remaining cases caused by axial or rotational forces.64 The indirect mechanism is more prevalent in cases causing fractures associated with Lisfranc fracture–dislocations secondary to the forces applied throughout the

midfoot. These same mechanisms are implicated in isolated middle and lateral cuneiform fractures. Clinically, the isolated cuneiform fracture presents with pain and some degree of edema and antalgic gait. Ecchymosis can be present and likely indicates direct trauma. Other findings include pain with transverse plane and frontal plane motion and pain when resisting the tibialis anterior muscle if the medial cuneiform is involved. Dorsoplantar, lateral, and medial oblique radiographs are the initially obtained when this injury is suspected. (Three views are necessary to improve diagnostic accuracy, since only two views lack a tangential appreciation of the midtarsus that reduces the osseous superimposition.) Overlap of the irregularly shaped midtarsal bones is implicated as the cause of 50% of isolated medial cuneiform fractures being misdiagnosed.64,65 CT imaging may be necessary to better understand the fracture pattern in some cases. The complexity of the midtarsal area, the nondisplaced nature of this fracture, and the presence of subtle fractures in the presence of obvious accompanying fractures impede accurate radiographic evaluation. Availability-based heuristic error also increases the difficulty in diagnosing isolated cuneiform fractures since identifying these rare occurrences is challenging when observed so infrequently.66 Bone scintigraphy, CT, and MRI are other imaging techniques that can be used to improve diagnostic accuracy in equivocal cases. When performing bone scintigraphy, it should be obtained 3 to 5 days after the injury to demonstrate focal uptake; prior to 3 days, a diffuse uptake can present, representing hyperemia or traumatic synovitis.67 CT can make a definitive diagnosis independent of time after injury and is the modality of choice.64,65 CT affords an accurate assessment of the osseous structures, including cortical integrity and small amounts of displacement. However, if a soft tissue injury, stress fracture, or compression injury of the cuneiform is suspected, MRI is more advantageous because of the enhanced soft tissue detail and the ability to detect bone marrow edema and trabecular compression. Both CT and MRI can identify other fractures that must be ruled out since isolated cuneiform fractures should be a diagnosis of exclusion.

FIGURE 16-18. Schematic representation of Quenu–Kuss Lisfranc fracture– dislocation classification.62 (From Figure 19-9 in Spinosa FA. Classification of fractures and dislocations. In: Christman RA. Foot and Ankle Radiology. 1st ed. St. Louis, MO: Churchill Livingstone; 2003:chap 19.)

FIGURE 16-19. Schematic representation of Hardcastle classification with Myerson’s modifications.41,63 A: Type A. B: Type B1. C: Type B2. D: Type C1. E: Type C2.

FIGURE 16-20. An example of Hardcastle type B and Myerson type B1. A bipartite cuneiform can also lead to misdiagnosis. Bipartition is an uncommon developmental variant with an incidence between 0.3% and 2.4%.68 The distinguishing radiographic feature of bipartite medial cuneiform is a horizontally oriented articulation in the sagittal plane (see Figure 6-2 in Chapter 6), which is a rare orientation for a medial cuneiform fracture; fracture is usually observed in the coronal plane. The plantar bipartite cuneiform segment is larger than the dorsal segment, while combined they may be larger than the normal cuneiform size. The first metatarsal articulating surface is also larger than expected when a bipartition is present. Furthermore, the bipartite cuneiform has smooth, well-defined margins. In contrast, a fractured medial cuneiform has an irregular, jagged cleavage between fragments. On MRI, the well-defined space between the bipartite segments, the horizontal lines superior and inferior to the dorsal and plantar segments, and the vertical joint space between the cuneiform segments and first metatarsal base create a unique configuration termed the “E” sign.68 Fractures typically create marrow edema visible on T2-weighted images, while bipartition does not present with marrow edema since the normal biomechanics are not typically altered and a fractured synchondrosis is rare.68

FIGURE 16-21. Medial cuneiform avulsion fractures. A: Posteromedial, with the arrow indicating the fracture fragment along the medial side of the medial cuneiform. B: Anterosuperior. CUBOID FRACTURE Cuboid fracture is also a rare injury, accounting for 5% of tarsal fractures.69 Rarely an isolated fracture, this injury frequently occurs with another lateral column fracture or tarsometatarsal joint fracture–dislocation. Cuboid fractures are classified as avulsion fracture, body fracture, fracture– dislocation, or stress fracture. Avulsion fracture accounts for two-thirds of cuboid fractures. Traction from the plantar calcaneocuboid ligament results most often in the avulsion fragment; some avulsion fractures result from cuboid adduction on the calcaneus, which places excessive tension on the lateral calcaneocuboid ligament. The avulsed fragment is typically observed along the lateral border of the cuboid (Figure 16-22). Body fractures occur when the foot strikes the ground in a plantarflexed position with rotational motion applied to the forefoot. This indirect force causes a simple crescent shaped, intra-articular fracture fragment. The most common mechanism causing a body fracture is a high-energy abductory force crushing the cuboid between the lateral metatarsals and calcaneus. This nutcracker fracture commonly presents with a navicular fracture. Since cuboid fracture frequently occurs associated with other fractures, a thorough evaluation of the entire foot is necessary, including radiography; the medial oblique view, in particular, is critical to isolating the cortical margins of the cuboid, best determine the fracture line, and evaluate the anatomic relationship between all of the tarsal bones.69 Accurate assessment of potential displacement is essential; length of the cuboid is critical to reestablishing the length of the lateral column, which significantly impacts the stability and function of the foot.70,71 The significant amount of cancellous bone in the cuboid may limit the visibility of cuboid fracture radiographically. It has been suggested in pediatric cases that repeat

radiographs be performed 2 weeks after the injury if a fracture is suspected despite normal initial radiographs; at that time a sclerotic line may become visible.16 CT scanning is best reserved for surgical planning, if necessary; bone scintigraphy is most beneficial when considering a cuboid stress fracture.

FIGURE 16-22. Lateral cuboid avulsion fracture, including a fifth metatarsal distal diaphyseal spiral fracture.

STRESS FRACTURE Stress fractures are common fractures in the foot and ankle, possibly accounting for 10% of all sports injuries. The metatarsals are the most common location, accounting for 95% of stress fractures in the foot, and may present bilaterally. In one study of military recruits, 84% of stress fractures were in the metatarsals, calcaneus, and tibia.72 The second metatarsal is involved most often, followed by the third metatarsal. The fourth and fifth metatarsals account for 13% of all metatarsal stress fractures. The fractures are usually diaphyseal; the second most common location is the neck. However, first metatarsal stress fractures are typically located at the base. Despite metatarsal stress fractures involving 80% of injuries in minimalist running, fibular stress fractures are most frequent in athletes.72,73 Stress fractures are classified as fatigue fractures or insufficiency fractures. Fatigue fractures occur when abnormal stress is placed on normal bone. The increased load from muscle activity overwhelms the normal metabolic equilibrium of bone resorption and remodeling, causing the fracture. This is commonly described as an overuse injury and is seen in athletes and military recruits. Insufficiency fractures involve normal stresses applied to bone with deficient structural integrity. This frequently occurs in patients with osteoporosis, or with medications or other illnesses that weaken the osseous structure, such as steroids and rheumatoid arthritis. Risk factors for stress fractures are related to weight bearing, bone vascularity, training regimen, and the equipment used. Dietary habits, metabolic state, menstrual patterns, anatomic alignment, and overall fitness level also have an influence on the risk of developing stress fractures. In addition to the significant risk when menstrual disturbances are present, females have a relative risk of stress fractures of 0.95 to 3.5 times greater than males.74 This may be secondary to skeletal mass, a lower moment of inertia, and differences in bone density. The diagnosis is based on clinical symptoms of insidious pain onset over a 2to 3-week period. Frequently there is also a recent increase in activity level. Tenderness overlying the affected bone is the hallmark sign, and there may be localized edema and increased pain with percussion of the bone involved.

Stress fractures of the second and fifth metatarsals, sesamoids, navicular, and anterior tibial diaphysis are high risk for developing nonunions. Radiography is the most useful imaging test for stress fractures. It is readily available, inexpensive, and can preclude additional tests if findings are present. However, diagnosing stress fractures requires a high level of suspicion, since the fracture may not be apparent on radiographs. Table 16-8 depicts the four basic radiographic patterns of stress fracture.72 Metatarsal stress fractures are typically type I and progress to the type III pattern, which is most commonly seen (Figure 16-7). In comparison to metatarsal stress fractures, calcaneal stress fractures present with a type II pattern with focal sclerosis, likely related to weight-bearing forces (Figure 16-23). TABLE 16-8   Patterns of Stress Fractures in the Foot Stress Fracture Pattern Presentation Fracture line demonstrable with no evidence Type I of endosteal bone or periosteal reaction Focal bone sclerosis manifesting as trabecular Type II condensation with cancellous bone and endosteal callus formation Type III Periosteal reaction and external callus Type IV Mixed combination of above patterns Adapted from Wilson E, Katz F. Stress fractures: an analysis of 250 consecutive cases. Radiology. 1969;92:481–486. Radiographic patterns of stress fracture may merely represent the phases of osseous healing. The appearance of a stress fracture is dependent on the time radiographs are obtained following the injury and the type of bone. Radiographic findings lag behind clinical symptoms by 2 to 3 weeks; some stress fractures may never become evident. Wilson and Katz72 noted that initial radiographs of 250 military recruits were negative in 24% of calcaneal stress fractures and in 17% of all cases. The earliest finding of a cortical bone stress fracture, such as in the metatarsal, may be a subtle radiolucency or cortical break secondary to resorption; a compensatory periosteal reaction often follows (Figure 16-7).

The late findings include thickening and sclerosis of endosteal and periosteal bone, weeks to months after the onset of symptoms. In contrast, cancellous bone stress fracture demonstrates a focal sclerotic band perpendicular to the trabeculae.75 At times, this sclerotic line is not visible for up to 24 weeks after the injury.66 If a stress fracture is suspected despite nondiagnostic radiographs, the study should be repeated 2 to 3 weeks later. However, if there is urgency to determine the presence of a stress fracture, bone scintigraphy or MRI can be performed.75 Bone scintigraphy is particularly useful for early detection since it has a sensitivity of nearly 100%, which is more than twofold greater than radiographs. In fact, findings have been observed on technetium bone scans prior to clinical evidence of the fracture.74,76 Stress fractures were identified on initial radiographs in 42% of cases while on bone scintigraphy in 95.8% of cases.77 However, the specificity of bone scintigraphy is significantly lower than radiography for localization and characterization of fracture lines. There is also potential confusion with other pathology that exhibits increased focal uptake, such as bone tumor and osteomyelitis.66,74 When a stress fracture is observed, there is a sharply marginated or fusiform area of increased uptake, involving one cortex or the entire width of the bone. Increased uptake is present in all three phases of the study.78 In contrast to stress fractures, stress reaction has a positive focal uptake during the first two (blood and delayed pool) phases only.

FIGURE 16-23. Stress fracture, calcaneus (type II: ill-defined increased density in cancellous bone, arrows). Ultrasonography is sensitive for occult fracture of more superficial bones, such as the metatarsals. A hypoechoic band, cortical step-off, and periosteal reaction can be indicative of a fracture. A hyperechoic area can be indicative of bone callus formation consistent with remodeling during fracture healing. MRI can be cost prohibitive but is becoming the preferred diagnostic imaging modality for stress fracture. Its sensitivity is at least as high as bone scintigraphy and specificity is much higher. The greatest visualization of MRI findings occurs when the study is performed within 3 weeks of symptoms onset. MRI is highly sensitive for endosteal marrow edema and periosteal edema, which are the earliest findings of stress fractures. These findings are observed as increased signal intensity on fluid filled sequences and decreased signal intensity on T1-weighted images.74,75,79 Muscle edema,

cortical thickening, and hypointense fracture lines affecting the cortical or cancellous bone may also be evident. CT imaging has also been used for evaluating stress fractures. It can be useful for distinguishing between stress fracture and stress reaction; however, MRI remains more sensitive and specific in this scenario. Unlike CT, MRI does not expose the patient to ionizing radiation, so it is now more commonly used for diagnosing stress fractures when faced with equivocal radiographs. However, CT is considered the gold standard for evaluating navicular stress fracture. In this case, CT assists with understanding fracture characteristics, comminution, displacement, and healing.6,80,81 CT has proved to be more accurate in the detection of navicular stress fracture than MRI. However, all stress reactions were identified using MRI while none were observed on CT images.82 A classification based on CT appearance was developed to guide treatment and forecast healing time since nonsurgical versus surgical treatment remains controversial. This classification relies on the location of the fracture line and by the presence of risk factors associated with poorer prognoses (Table 16-9).83 It is important to keep in mind that this classification has not been validated and others have noted flaws in the conclusions regarding the treatment recommendations.84 The navicular stress fracture is located within the zone of maximum shear stress, which is in the central one-third of the body just lateral to the center of the talar articular surface. The fracture line is linear in the sagittal plane or slightly oblique from dorsal–medial to plantar–lateral (Figure 16-24). Radiographically, the fracture appears as a sclerotic line along the proximal articular border in the dorsoplantar view; however, its visualization may be difficult.75 A review of 55 cases revealed that 11% of navicular stress fractures were accurately diagnosed using CT after being missed on radiographs initially.85 The fracture is typically incomplete, thus contained within the dorsal 5 mm of the navicular.66,81 CT in the anatomic anteroposterior plane with the foot supinated and forefoot elevated will provide the best opportunity for identifying the fracture.66 TABLE 16-9  

Computed Tomography–Based Classification of Navicular

Type I II III Modifiera    A    C    S

Stress Fractures Description

Treatment Conservative or ORIF in an Dorsal cortical fracture athlete Dorsal fracture propagates into Conservative or ORIF in navicular body active/athletic patient Fracture penetrates second cortex (plantar, medial, or Most often ORIF lateral) Description Notes Avascular necrosis Most often seen in type III Cystic degeneration Most often seen in type I Sclerosis of fracture lines Most often seen in type II

aA modifier can be assigned to any type within the classification.

Calcaneal stress fracture is easily overlooked secondary to the cancellous nature of the bone. When present, a dense, sclerotic band is observed in the posterior calcaneal body. This is a common injury with initial weight bearing after periods of immobilization. Tibial stress fractures are located along the anterior aspect of the diaphysis. Radiographically, in the lateral view, the anterior cortex appears thickened and contains a lucent line running through it.

FIGURE 16-24. Stress fracture, navicular. A: Nondisplaced. B: Displaced, with an accessory navicular. NAVICULAR FRACTURE Navicular fractures are rare, making up 0.3% of fractures in one study.72 Several classifications have been published to describe navicular fractures.86– 88 Eichenholtz provided a more simplistic classification, which identifies the cortical avulsion fracture, tuberosity fracture, and body fracture as the three main fracture types. Stress fracture of the navicular, which was previously discussed, was later identified as a fourth type of navicular fracture.89 Navicular fractures most commonly occur in patients between 30 and 60 years of age and are two times more prevalent in females than in males. Patients who suffer navicular fractures often have other associated foot fractures 47% to 50% of the time.90–92 Cortical avulsion fracture accounts for 47% of navicular fractures in the Eichenholtz classification (Figure 16-25).91 This injury usually results from an inversion twist or sprain with the foot in plantarflexion. Dorsal lip fractures result from dorsal capsule or dorsal talonavicular ligament avulsion. These fractures are best seen on the lateral foot radiograph.

FIGURE 16-25. Dorsal avulsion fracture of navicular. Navicular tuberosity fracture occurs 24% of the time (Figure 16-26).91 This injury results from MVA, fall, sprain, and twisting injury.92 The mechanism of injury involves forced eversion of the foot, resulting in avulsion of the tuberosity.89,91,93,94 However, there is some controversy as to which anatomic structure actually causes the avulsion fracture. Some authors believe the culprit to be the posterior tibial tendon, while others believe it to be the spring ligament.89,91,93 Anterior calcaneal process fractures have been observed in 50% of patients with navicular tuberosity avulsion fracture.90,92 Dorsoplantar and lateral radiographic views are commonly used to diagnose this fracture; the lateral oblique view might provide additional information.94 The diagnosis becomes more challenging when radiographs reveal an accessory navicular; furthermore, the accessory navicular itself could fracture or become avulsed.91,93,94 Contralateral films might be helpful, but 50% to 89% of patients with an accessory navicular have a bilateral presentation.95– 97 Table 16-10 delineates the three types of accessory navicular (see Figures 6-58 and 6-59 in Chapter 6).

FIGURE 16-26. Avulsion fracture of the navicular tuberosity (arrow). (From Figure 19-12 in Spinosa FA. Classification of fractures and dislocations. In: Christman RA. Foot and Ankle Radiology. 1st ed. St. Louis, MO: Churchill Livingstone; 2003:chap 19.) TABLE 16-10   The Three Types of Accessory Navicular Accessory Navicular Type Description Sesamoid bone within the posterior tibialis

1 2 3

tendon Arises from secondary center of ossification, connected to main body of navicular via fibrocartilaginous bridge or hyaline cartilage Arises from secondary center of ossification with bony bridge to main body of navicular

Avulsion fracture of an accessory navicular is thought to occur mainly in type 2, with fracture occurring through the cartilaginous connection.89,93,97 Studies have advocated the use of bone scintigraphy for fracture confirmation.98,99 However, a more recent study, evaluating asymptomatic and symptomatic accessory navicular using bone scintigraphy, achieved a sensitivity of 100% and a specificity of 50%.97 Half of the asymptomatic accessory navicular ossicles demonstrated increased focal uptake on bone scintigraphy.97 In this situation, MRI has been shown to provide more anatomic detail, but comparative sensitivities and specificities have not been established.100 Navicular body fractures, excluding stress fractures, result from higher energy mechanisms of injury and account for 29% of navicular fractures.71,91,101 Consensus on the exact mechanism of injury for navicular body fractures is lacking, and there are several proposed theories.86,91,102,103 The most common modes of injury were MVA, motorcycle accidents, falls, and crush injury.91,101,102 Using radiography to diagnose intra-articular navicular body fractures, Sangeorzan et al.101 published a classification system based on direction of fracture line, direction of forefoot displacement, and pattern of hindfoot joint disruption (Table 16-11). Type 2 fractures were found to be most common, occurring in 50% of patients, with dorsal–medial talonavicular joint subluxation being the most frequent pattern. This classification system may also provide prognostic information.102 As the complexity of injury progressed from type 1 to type 3, the ability to achieve anatomic reduction decreased and functional outcomes worsened.101 Secondary to the rarity of this injury and paucity of literature on this classification, the prognostic value

of this classification system has not been confirmed. Dorsoplantar, lateral, and medial oblique radiographic views are recommended for initial diagnosis of navicular body fractures.71,89,101,102,104 However, 12% to 20% of navicular fractures are missed or fracture details are underrepresented when using radiographs alone.91,102 If additional CT imaging is not pursued, a type 3 injury may not be fully identified radiographically and, therefore, not be treated appropriately.102 Therefore, CT remains the imaging modality of choice for navicular body fractures.89,102,104 TALAR FRACTURE Talar fracture accounts for 1% to 1.3% of all fractures, and 3% to 6% of all foot fractures.105–108 Fracture of the talar body and neck tend to be the most common types, accounting for more than 60% of talar fractures.107,109 A less common location is fracture of the lateral talar process.107,110 MVA, falls from a height, and ground level falls or sprains are common mechanisms of talar fracture.107,111,112 Historically, talar fracture has also been documented to result from aviation accidents.113 TABLE 16-11  

Classification of Intra-articular Navicular Body Fractures Developed by Sangeorzan et al.101

Talar Neck Fracture Talar neck fracture occurs most frequently from MVA and fall from a height

(Figure 16-27).105,114 Males are affected more than females at a 6:1 ratio.112 Talar neck fracture is presumed to result from forced dorsiflexion, as the talar neck is pressed against the anterior tibia.112,115 This injury tends to cause a vertical fracture through the neck and can deviate from the frontal plane and extend into the talar body. The widely accepted Hawkins classification for talar neck fracture, developed in 1970, involves three groups based on the radiographic fracture pattern.112 In 1978, Canale and Kelly added a fourth injury pattern to the Hawkins classification system (Table 16-12, Figure 1628).116

FIGURE 16-27. Talar neck fracture with medial displacement of talar head and medial dislocation of subtalar joint. (From Figure 19-20 in Spinosa FA. Classification of fractures and dislocations. In: Christman RA. Foot and Ankle Radiology. 1st ed. St. Louis, MO: Churchill Livingstone; 2003:chap 19.) This classification does not differentiate between talar fractures extending exclusively through the neck and neck fractures involving part of the talar body. Utilization of this classification system for talar fractures extending into the body has therefore been mixed.117,118 Though the Hawkins classification may not strictly define talar neck fractures, it does provide an

understanding of the injury progression. Additionally, it provides prognostic information for both isolated talar neck fracture and mixed talar neck and body fractures. Two cases of talar neck fracture have been reported where the injury pattern did not follow the Hawkins classification. In these reports, talonavicular joint dislocation occurred with and without subtalar joint subluxation or dislocation, while the ankle joint remained intact.119,120 The mechanisms of the injury were falls from a height and MVA, which do not differ from the common talar neck fracture mechanisms that follow the injury progression outlined by Hawkins. Routine foot and ankle views and Canale views continue to be well-accepted radiographic studies for diagnosis, treatment planning, intraoperative evaluation, and postoperative evaluation.118,121,122 CT provides more detailed imaging of the fracture fragments and alignment of the affected joints. Therefore, some authors recommend obtaining CT scans preoperatively.123 Talar Arterial Supply As the number of peritalar joint dislocations increase following the Hawkins classification, the severity of vascular compromise to the talus also increases. Three main arteries provide arterial supply to the talus, and they have been thoroughly studied and documented.124–126 Using gadolinium-enhanced MRI, Miller et al.124 demonstrated that the talus receives 47% of its arterial supply via the posterior tibial artery, 36.2% via the anterior tibial artery, and 16.9% via the peroneal artery. These findings were consistent with previous qualitative findings of talar arterial supply.125,126 TABLE 16-12   Hawkins112 Classification of Talar Neck Fractures Group Injury 1 Vertical talar neck fracture, nondisplaced 2 Vertical talar neck fracture, subluxated/dislocated STJ Vertical talar neck fracture, subluxated/dislocated 3 STJ, ankle joint Vertical talar neck fracture, subluxated/dislocated

4a

STJ, ankle joint, TNJ

aCanale and Kelly modification.116

STJ, subtalar joint; TNJ, talonavicular joint.

FIGURE 16-28. Schematic representation of the Hawkins classification.112 (From Figure 19-21 in Spinosa FA. Classification of fractures and dislocations. In: Christman RA. Foot and Ankle Radiology. 1st ed. St. Louis, MO: Churchill Livingstone; 2003:chap 19.) In the Hawkins group 1 injury, arterial supply at the neck of the talus is disrupted but osteonecrosis occurs infrequently. This is due to the intricate and variable arterial anastomoses of the talus, especially the contributions by the artery to the sinus tarsi, the artery to the tarsal canal, and the deltoid artery.124,126 The artery to the tarsal canal and the deltoid artery are

consistently branches of the posterior tibial artery; the sinus tarsi artery can arise from either the anterior tibial artery or the peroneal artery.125,126 The occurrence of talar osteonecrosis in Hawkins groups 2 through 4 injuries is dependent on the severity of vascular disruption on the remaining talar arterial supply.114,126,127 Literature from the 1970s showed higher rates of talar osteonecrosis compared to more recent studies.112,114,115,116,127 This difference is most likely due to better understanding of the injury and its complications, timely treatment initiation, improvements in surgical technique, and implants. Osteonecrosis of the Talus Osteonecrosis, also referred to as avascular necrosis, is a devastating sequella of talar neck fractures, as it may lead to talar dome collapse and end-stage ankle arthritis.114,127,128 Vallier et al.127 found radiographic evidence of talar osteonecrosis at an average of 19 weeks after injury, while all radiographs showed signs of osteonecrosis within 10 months after the injury. Subsequent talar dome collapse occurred in 63% of the osteonecrotic tali at an average of 39 weeks, while 37% of the osteonecrotic tali showed signs of revascularization at an average of 35 weeks. Revascularization may take up to 36 months.123 In order to limit osteonecrosis, the emphasis during treatment should be placed on immediate reduction of displacement to preserve the remaining tenuous arterial supply to the talus (Figure 1629).116,127,129

FIGURE 16-29. Talar neck fracture with subtalar joint subluxation. For Hawkins group 1 injury, several studies have found osteonecrosis to occur from 6% to 17%.110,117 However, more recent studies from level 1 trauma centers found osteonecrosis rates of 0% in this injury group.114,127 These studies support the majority of reports leading to the commonly accepted rate of osteonecrosis occurrence of 0% for Hawkins group 1 injuries.112,113,130 For Hawkins group 2 injuries, studies prior to the year 2000 found talar

osteonecrosis to occur in 16% to 71% of cases.122 Later studies report osteonecrosis occurring in 25% to 39% of cases.114,127 For Hawkins group 3 injuries, osteonecrosis was found to occur in 33% to 100% prior to the year 2000, compared to 42% to 64% in more recent studies.114,127 Osteonecrosis occurring in 50% to 100% of Hawkins group 4 injuries remains unchanged.116,127 Comminuted and open talar neck fracture are the only two independent risk factors associated with developing osteonecrosis secondary to talar neck fracture.127 At the time the classification system was developed, Hawkins described a radiographic finding that provided insight on possible talar osteonecrosis development. Known as the Hawkins sign, a positive finding occurs when subchondral lucency along the talar dome is visualized radiographically in ankle views at weeks 6 to 8 after the injury (Figure 16-30A).112,115,131 A positive sign is thought to signify the intact vasculature required for bone resorption during the healing process, which is also known as reactive hyperemia.112,115,131 Hawkins also postulated that disuse atrophy contributes to the talar subchondral lucency. MRI may provide further insight to the extent of talar osteonecrosis.124 Recent studies evaluated sensitivity and specificity of the Hawkins sign and its ability to detect potential occurrence of osteonecrosis, which becomes sclerotic radiographically (Figure 16-30B).1,131,132 Sensitivity of 67% to 100% and specificity of 37% to 86% were found. Furthermore, positive and negative predictive values were found to be 31.3% to 100% and 63% to 100%, respectively. Even with the wide range of results, authors of these studies all conclude that a positive Hawkins sign reliably rules out the development of osteonecrosis.117,131,132

FIGURE 16-30. A: Hawkin sign, indicated by the subchondral radiolucency (arrows) parallel to the talar dome cortical outline. B: Osteonecrosis of talar body following talar neck fracture. Interestingly, much attention has been focused on talar osteonecrosis; however, in recent studies, the most common complication was found to be posttraumatic arthritis of the peritalar joints.114,118,121,127 The incidence of posttraumatic arthritis of these joints is 48% to 100% in talar neck fractures (Figure 16-31).114,118,127,131 Vallier et al.127 observed that 54% of talar neck fractures resulted in posttraumatic arthritis and 62% developed endstage posttraumatic arthritis with complete joint space loss. Incorporating prognostic information on posttraumatic arthritis into the specific groups of the Hawkins classification would be more valuable. Talar Osteochondral Defect An osteochondral defect or osteochondritis dissecans (OCD) of the talar dome is defined as an injury with varying degrees of cartilage and subchondral bone separation or impaction.133 With advances in imaging modalities, talar OCD can now be visualized with more detail. These advances in diagnostic imaging should not replace radiography as the initial imaging modality.

FIGURE 16-31. Posttraumatic ankle arthritis resulted secondary to talar neck and body fractures. Collapse of the talar dome is seen in both cases, secondary to osteonecrosis. New classification systems have been developed utilizing MRI, CT, and arthroscopy.134–140 A unanimous classification system that provides pathologic staging, treatment recommendation, prognostic information, and ease of use has yet to be developed. The Berndt and Harty classification, published in 1959, remains the most commonly used system, characterizing and providing a general pathologic course of talar OCD using radiography (Table 16-13, Figure 16-32).133 In 1989, Anderson et al. added stage IIa to the Berndt and Harty classification system to account for subchondral cysts without a transchondral fracture.137 In 1993, Loomer contributed the same diagnosis as Anderson, labeling it stage V.141 Some publications have favored the use of the stage V modification over the stage IIa designation.139 Interestingly, histologic studies have shown that, regardless of injury stage, the hyaline cartilage remains viable even after the talar OCD injury, as long as the lesion remains exposed to nutrients within synovial fluid. However, the bone connected to the talar OCD undergoes necrosis when intraosseous vascular supply is interrupted.115,133,137 With the use of CT and MRI, visualization and sensitivity for diagnosis for talar osteochondral lesions have been enhanced. Several studies have found that 30% to 43% of talar OCD cannot be visualized radiographically but are visible on MRI.135,137,142 Unlike radiography, MRI also allows for visualization of isolated cartilaginous injuries. Stages I, IIa, and V are most often missed radiographically.105,107,137,143 Hepple et al.135 later developed an MRI classification system that mirrors the radiography-based Berndt and Harty classification system (Table 16-14). Furthermore, stage 1 talar dome compression fractures were found to be the most commonly missed talar fracture. Thirty-one percent of stage 1 compression fractures are missed radiographically, which provide 74% to 78% sensitivity for stage 1 diagnosis.105,107 Some authors disagree on the ideal imaging modality for talar OCD.144,145 While MRI and CT are both acceptable modalities, MRI provides visualization of cartilage and bone marrow edema, which are not

provided by CT. However, CT provides clearer visualization of osseous cortical margins.144,145 Verhagen et al.145 found that, with 3.0T MRI becoming more available, MRI will most likely become the study of choice as it provides insight on the cartilage layer and exposes patients to less radiation. 133

Classification System of Talar TABLE 16-13   Berndt and Harty Osteochondral Lesions with Modifications Stage Berndt and Harty Modifications Small subchondral   1   compression fracture Incomplete osteochondral   2   fracture 2aa—formation of     subchondral cyst Complete osteochondral   3   fracture, nondisplaced   4 Complete fracture, displaced       5b—radiolucent fibrous defect aAnderson modification.137 bLoomer modification.15

Berndt and Harty133 identified talar OCD in two specific areas of the talar dome. Specific mechanisms of injury for each location were determined through experimentation. OCD of the anterolateral one-third of the talus occurs when the talar dome is impacted against the lateral malleolus with excessive inversion while the ankle joint in dorsiflexion. In contrast, injuries to the posteromedial one-third of the talus occur when the talar dome is impacted against the posterior malleolus with excessive plantarflexion and inversion (Figure 16-32). Canale and Belding143 provided further insight when they retrospectively found that medial lesions tend to be deeper and “cup-shaped,” while lateral lesions tend to be shallower and “wafer-shaped.” Aside from the aforementioned mechanisms consistent with ankle sprains

that can produce isolated talar OCD injuries, ankle fracture mechanisms can also cause talar OCD. During ORIF of ankle fractures, talar OCD have been visualized in 27.9% to 73.2% of closed ankle fractures.146 –149 Patients who suffered both injuries experienced decreased range of motion and worse subjective outcomes.146 The number of malleolar fractures has not been shown to correlate with occurrence of talar OCD.147,148 Other Talar Fractures Talar head fracture accounts for 3% to 10% of talar fractures.107,110,113,123,130 The mechanism of injury is thought to result from a sudden dorsiflexion force against the talus while the foot is in a plantarflexed position.123 These injuries have resulted from fall, MVA, and aviation accident.110,113,130 These fractures are difficult to visualize radiographically and are therefore easily missed.106,123 Treatment planning, deciding between conservative and surgical treatment, is dependent on the talonavicular joint alignment and the percentage of joint involvement.106,123 Talar body fracture incidence has been reported as 15% to 20% of all talar fractures.106 Of these fractures, the most commonly isolated talar body fractures are compression of the talar dome and fractures of the lateral process and posterior tubercle.107,150,151 Talar body fractures have been classified into six types (Table 16-15). In Sneppen and Buhl’s150 patient population, falls from height, direct trauma, and MVA were the mechanisms of injury. Osteonecrosis was found only in shearing and crush fractures, occurring at a rate of 38% and 75% of these injuries, respectively. This amounted to a total of 16% of talar body fractures developing osteonecrosis, whereas 55% of all talar body fractures developed posttraumatic arthritis, which is the most common complication. Of these talar body fractures, 75% of the posterior tubercle injuries developed posttraumatic arthritis, despite only 9% of the posterior tubercle injuries were displaced (Figure 16-33).151 McCrory and Bladin developed a classification system, similar to the Hawkins classification, for lateral talar process fractures in 1996 that has been used by some authors (Table 16-16).152,153 These injuries are relatively

rare, accounting for approximately 1% of all ankle injuries in the 1970s.153,154 This fracture can result from MVA, fall from height, ankle sprain, and snowboarding.106,107,150–155 In snowboarders, lateral process fracture tends to occur to the leading foot and in snowboarders wearing softshell boots (Figure 16-34).152,153 There is agreement that the mechanism of injury is thought to be due to dorsiflexion with impaction.111,152,153 However, some authors think other adjunctive motions are a part of the mechanism, such as inversion, while other authors have found external rotation and occasional eversion to play a role, especially in snowboarders.111,125,153 Posttraumatic arthritis develops in 11% to 15% of lateral talar process fractures.151 This injury is often missed and misdiagnosed as a severe ankle sprain. Radiographs, especially ankle mortise and Broden views, can be used for initial evaluation. Since there is poor radiographic visualization of this fracture, CT is recommended if radiographs are negative and a high suspicion of a fracture persists.151,153,155

FIGURE 16-32. Talar dome lesions (OCD, arrows) (Berndt and Harty classification). A: Stage I, subchondral compression. B: Stage II, incomplete. C: Stage III, complete. D: Stage IV, complete and displaced. 135 Classification System of Talar Osteochondral

TABLE 16-14   Hepple

Stage   1   2a   2b   3   4   5

Defects Based on MRI Studies Description Articular cartilage damage, not visible on radiographs Cartilage injury with underlying subchondral fracture and bony edema Stage 2a without bony edema Detached cartilage and subchondral fracture, nondisplaced Detached cartilage and subchondral fracture, displaced Subchondral cyst

CALCANEAL FRACTURE Calcaneal fractures account for 2% to 3% of all fractures in the body.156,157 It is the most commonly fractured tarsal bone, accounting for 60% of tarsal injuries.88 Intra-articular fractures make up approximately 75% to 85% of calcaneal fractures, while the remaining fractures are extra-articular.157,158 Males sustain this injury four to five times more than females.159,160 Calcaneal fracture most commonly results from fall from a height, with MVA the second most common mechanism of injury.88,160–164 With these highenergy mechanisms, spinal fractures have been observed in patients sustaining calcaneal fractures. Spinal fractures are prevalent in 3% to 12% of calcaneal fractures, with L1–L3 vertebrae most commonly affected.88,156,160,165 TABLE 16-15   Sneppen151 Classification of Talar Body Fractures Type Talar Body Fracture Types   A Compression fracture   B Coronal plane shearing fracture   C Sagittal plane shearing fracture   D Posterior tubercle fracture   E Lateral tubercle fracture   F Crush fracture

FIGURE 16-33. Fracture (arrows) of the posteromedial talar process (aka Shepherd fracture). Intra-articular Calcaneal Fracture In 1948, Palmer164 described two types of intra-articular calcaneal fracture. Essex-Lopresti165 later refined Palmer’s observations and provided a classification system (Table 16-17, Figure 16-35). He postulated that intraarticular calcaneal fracture results from the lateral talar process being driven axially into the calcaneus at the crucial angle of Gissane.165 This impact produces the primary fracture line extending plantarly from the sinus tarsi. The force then travels through the bone to produce two types of intra-articular fractures (Table 16-17). Soeur and Remy166 published a classification system, which considered the number of articular bone fragments seen radiographically. Zwipp et al.167 was the first to use CT to develop a calcaneal fracture classification system. Sanders163 also used CT to evaluate intra-articular fractures, later developing

a classification system (Table 16-18). This classification system is commonly used, as it provides information on ease of reduction and prognosis.163,168,169 Coronal and axial views, which provide visibility of the wide posterior talar facet, are used to classify the fracture type. Anatomic reduction was found to be more achievable with fewer intraarticular fracture fragments. In one study, anatomic reduction was achieved in 86% of type II fractures, 60% of type III (Figure 16-36) and 0% of type IV fractures. Outcomes also followed a similar pattern, as excellent and good outcomes were achieved in 73% of type II fractures, 70% of type III fractures, and 9% of type IV fractures. Contrary to the anticipated outcomes, 48% of type II fractures and 23% of type III fractures required subtalar joint arthrodesis.161–163 This finding was explained by the fact that this classification system does not grade the degree of chondral injury. Similarly, this also explains why anatomic reduction of the osseous fragments may not always guarantee satisfactory outcome.161–163 Though the classification may not be able to differentiate the need for subsequent subtalar joint arthrodesis in type II and III injuries, type IV injuries are 5.5 times more likely to require subtalar joint arthrodesis.170 153

Classification of Lateral Process TABLE 16-16   McCrory and Cladin Fractures of the Talus Lateral Talar Process Fracture Type Types Avulsion chip fracture of talofibular   1 ligament (no joint involvement) Large lateral talar process fracture,   2a nondisplaced Large lateral talar process fracture,   2b displaced Comminuted lateral talar process   3 fracture

FIGURE 16-34. Lateral talar process fracture (arrow). Several authors have tested the reliability and reproducibility of calcaneal fracture classification systems.168,171–174 Though no classification system is perfect, the Sanders classification system has consistently performed as one of the more reliable and reproducible systems. It provides fair-to-moderate interobserver reliability and intraobserver reproducibility; when the Sanders classification is applied without the use of specific fracture line subtyping, the reliability and reproducibility improve, maintaining a moderate level more consistently. With its prognostic value, the Sanders classification is a wellaccepted intra-articular calcaneal fracture classification.161– 163,168,169,171,172,174,175

TABLE 16-17   Essex-Lopresti165 Intra-articular Classification Type Fracture Secondary fracture line runs   I Tongue type directly to posterior aspect of calcaneal tuberosity Secondary fracture line exits  II Joint depression posterior-superior to posterior facet

FIGURE 16-35. Schematic representation of Essex-Lopresti tongue-type calcaneal fracture. A: nondisplaced; the gray lines along the superior surface of the calcaneus indicate a normal Böhler angle. B: Displaced; the gray lines along the superior surface of the calcaneus indicate an abnormal (negative) Böhler angle. (Images modified from Figure 19-31 in Spinosa FA. Classification of fractures and dislocations. In: Christman RA. Foot and Ankle Radiology. 1st ed. St. Louis, MO: Churchill Livingstone; 2003:chap 19.) Radiography is routinely used for making the initial diagnosis of calcaneal fractures. Dorsoplantar, lateral, and calcaneal axial foot views are typically ordered.176 Böhler angle and Gissane angle are used to evaluate the height of the calcaneus and the position of the articular facets, respectively.177,178 Böhler angle has been shown to be useful in forecasting outcomes, since an angle of less than 0° on presentation is 10 times more likely to require a subtalar joint arthrodesis after initial ORIF.170 Similar results have been reported where patients who presented with a negative Böhler angle (Figure 16-37) had significantly worse functional outcomes at 2 years after the injury.179 The Sanders163 Classification for Intra-articular Calcaneal TABLE 16-18   Fractures Type I II III IV

Fracture Nondisplaced intra-articular calcaneal fracture Two fracture fragments (1 fracture line): A/B/C Three fracture fragments (2 fracture line): AB/BC/AC Four fracture fragments (3 fracture lines)—highly comminuted: ABC

FIGURE 16-36. Sanders Classification. A: Lateral ankle radiograph of acute injury demonstrates abnormal Böhler angle. B: Coronal plane CT view demonstrates multiple fracture lines at the distal aspect of the posterior calcaneal facet. Sagittal (C) and coronal (D) plane CT images demonstrate significant joint depression injury. E: Broden view visualizing the posterior calcaneal facet 4 weeks postoperatively. A more specialized radiographic technique, called the Broden views, is ordered to visualize the posterior and anterior subtalar joint facets.176,180 These radiographic views are useful intraoperatively, as the Broden views provide a clearer image of subtalar joint reduction than the other foot views.176,180 CT is usually ordered secondarily and is better tolerated by a symptomatic patient because it does not require the patient to put their foot into a dorsiflexed position.181 CT provides greater detail than radiographs and assists in preoperative planning; and, if applicable, the Sanders classification can be applied.161–163,181 Extra-articular Calcaneal Fracture Extra-articular calcaneal fracture makes up approximately 15% to 25% of calcaneal fractures.157,158 Patients who suffer them are usually in their thirties.172,182 Several extra-articular calcaneal fracture classifications have been published.165,178,183. The Rowe classification is the most commonly used, primarily serving as a descriptive tool of both extra-articular and intraarticular calcaneal fractures (Table 16-19, Figure 16-38).165,178,183

FIGURE 16-37. Lateral view demonstrating abnormal Bohler angle. Anterior Process Fracture The most common extra-articular calcaneal fracture is the anterior calcaneal process fracture. This type of fracture has accounted for 3% to 38% of all calcaneal fractures.172,182 The main mechanism of injury is commonly described as an ankle sprain; the bifurcate ligament avulses the anterior process as the calcaneus plantarflexes and inverts.184,185 A less common

mechanism of injury is forceful abduction of the foot, resulting in compression of the calcaneocuboid joint.186 This mechanism of injury can cause more injury to the plantar portion of the anterior process than the avulsion mechanism.186 Fracture of the anterior process can be seen on dorsoplantar, lateral hindfoot, and oblique views of the foot (Figure 16-39). Degan et al.185 published a classification system in 1982 for this fracture (Table 16-20). In this study, type III fractures accounted for 83% of nonunions, which required subsequent surgical excision. The fracture type, delayed diagnosis, and resultant nonunion were determined to provide prognostic information of this injury. To our knowledge, additional study evaluating prognostic value of the aforementioned factors has not been performed. TABLE 16-19   Rowe183 Classification of Calcaneal Fractures Type Fracture Prevalence (%) I A    Tuberosity fracture 21   B    Sustentaculum tali fracture     C    Anterior process fracture   II A    “Beak” fracture 3.8 B    Avulsion fracture at Achilles     tendon insertion Oblique fracture without STJ III 19.5 involvement IV Intra-articular fracture involving STJ 24.7 Intra-articular fracture involving STJ V with central depression and 31 comminution STJ, subtalar joint.

FIGURE 16-38. Rowe calcaneal fracture classification.183 A: Type I. B: Type II. C: Type III. D: Type IV. E: Type V. (Images modified from Figure 19-32 in Spinosa FA. Classification of fractures and dislocations. In: Christman RA. Foot and Ankle Radiology. 1st ed. St. Louis, MO: Churchill Livingstone; 2003:chap 19; From Rowe CR, Sakellarides HT, Freeman PA, et al. Fractures of the os calcis. JAMA. 1963;184(12):920.)

FIGURE 16-39. Anterior calcaneal process avulsion fractures (arrows). A: Dorsomedial (medial oblique view). B: Dorsolateral (lateral oblique view). Degan185 Classification for Extra-articular Fractures of the TABLE 16-20   Anterior Calcaneal Process Type Anterior Process Fracture Nondisplaced, usually involves just the tip of anterior I process II Displaced, extra-articular Displaced, intra-articular involvement of III calcaneocuboid joint Tuberosity Avulsion Fracture Tuberosity avulsion fracture makes up 2% to 2.6% of calcaneal fractures.172 Falling and landing on the forefoot with simultaneous triceps surae contraction, and tripping and direct blow to the calcaneal tuberosity are the commonly described mechanisms of injury.172 Osteoporosis has been documented to be a risk factor.187–190 Lateral hindfoot radiographs are useful for diagnosing tuberosity avulsion fracture, and MRI is valuable for evaluating the extent of Achilles tendon involvement for treatment

planning.172 Classification systems have been developed to aid in the description and treatment guidance, but these have not been validated and are not commonly used (Figure 16-40).172,191 Sustentaculum Tali Fracture Secondary to the thick cortical bone, isolated sustentaculum fracture rarely occurs; instead, it presents with other lower extremity injury such as subtalar dislocation and talar fracture.192–194 Isolated sustentaculum tali fractures account for 0.3% to 2.6% of calcaneal fractures.158,165,195,196 This fracture usually after falling from a height with the foot in a supinated position on impact.192

FIGURE 16-40. Calcaneal tuberosity avulsion fracture. When discussing sustentaculum tali fracture, disagreement on articular involvement inevitably arises.165,172,182,183,192,196 Historically, the fracture line in an isolated sustentaculum tali fracture is thought to pass between the middle and posterior facet of the calcaneus. Consequently, it has been classified as an extra-articular fracture.172,182 More recent publications using CT have shown the fracture line coursing through the middle articular facet and causing subluxation of the subtalar joint. Therefore, some authors

disagree with the extra-articular fracture classification.192,196 Due to the low incidence of these fractures, an ideal classification has yet to be developed.158,172,182,195 ANKLE FRACTURE There are approximately five million ankle injuries seen in the emergency department annually.197 Ankle fracture accounts for 9% of fractures; only proximal femoral fracture presents more frequently in the lower extremity.198,199 Ankle injury is common in the general population, especially in middle-aged to older women and young men.200 The incidence of ankle fracture is estimated to be 107 to 184 per 100,000 people. The peak age range is 15 to 24 years for men and 65 to 75 years for women. Over the past three decades, there has been a threefold increase in the incidence among elderly females.201 It is also common in the pediatric population; it makes up 5% of all pediatric fractures and 15% of all physeal injuries, with the peak incidence between the ages of 8 and 15 years.202 The most common causes of these fractures are twisting and falls, followed by sports injuries. Ankle fracture constitutes 21% of all sports-related injuries.203 Diagnostic imaging techniques that can be used to evaluate these injuries include radiography, CT, MRI, radionuclide bone scanning, and ultrasonography.204 Anatomy The ankle joint is a complex hinge-type joint. It has the smallest surface area of the major weight-bearing joints. During ambulation and other activities, the stress across the ankle joint varies from 1.25 to 5.5 times the body weight, which is twice the stress in the hip and knee.197 The tibia and fibula articulate at the proximal and distal tibiofibular joints; an interosseous membrane connects the two bones proximal to the ankle joint. The three ligament complexes stabilizing the ankle joint are the lateral collateral ligament complex, the syndesmotic ligament complex, and the medial collateral

ligament complex (deltoid ligament).205 The lateral collateral ligament complex is the most commonly injured ligament in patients with ankle trauma. The three components of the lateral complex are the anterior talofibular (ATFL), posterior talofibular (PTFL), and calcaneofibular (CFL) ligaments.205 The syndesmotic ligament complex includes the anterior-inferior tibiofibular (AITFL), posterior-inferior tibiofibular (PITFL), and transverse tibiofibular ligaments, as well as the interosseous membrane. The AITFL is one of the most commonly injured ligaments in the ankle, attaching slightly proximal to the talofibular ligament above the level of the tibiotalar joint line.205 The medial collateral ligament complex lies deep to the medial flexor tendons. The complex has four components that are divided into superficial and deep layers. The superficial layer is comprised of the tibionavicular and the tibiocalcaneal ligaments. The anterior and posterior tibiotalar ligaments make up the deep layer of the deltoid ligament. The medial collateral complex stabilizes against valgus forces, anterior and lateral talar excursion, and rotatory forces.205 Pott206 described the first ankle fracture classification system in 1768, after sustaining one himself in 1756. The classification was based on clinical findings, which was divided into number of malleoli involved. A first-degree Pott fracture was a fracture involving one malleolus. A second-degree Pott fracture was a bimalleolar ankle fracture and a third-degree fracture was a trimalleolar injury. Though easy to use with intraobserver reliability, it does not differentiate between stable and unstable injuries.207 However, Ashhurst and Bromer208 were credited with the first published classification on ankle fractures in a logical fashion in 1922.208 This classification system was further developed and refined by Lauge-Hansen using fractures produced experimentally in combination with clinical and roentgenologic examinations.209 Lauge-Hansen Classification

In 1950, Lauge-Hansen classified ankle injuries based on the position of the foot and the direction of the deforming force at the time of the injury (Table 16-21).209 The foot position may be pronated (everted) or supinated (inverted). While in one of these positions, the deforming force may cause the talus to displace or rotate in relation to the ankle mortise in abduction, adduction, or external rotation.205 Based on the findings, the four groups of ankle fractures identified were supination-adduction (SAD), supinationeversion (supination-external rotation [SER]), pronation-abduction (PAB), and pronation-eversion (pronation-external rotation [PER]). The first term of this dual designation refers to the position of the foot at the time of injury and the second refers to the direction of the pathologic force being applied to the talus.210 Each main group is further divided into stages bases on the degree of severity.211 The stage of each group is defined by the injury occurring, whether it involves failure resulting in a fracture or ligamentous disruption. The Lauge-Hansen classification was initially proposed to guide fracture reduction through an understanding of the mechanism of injury. Knowledge of the precise mechanism of an ankle fracture results in predictable fracture patterns aiding in treatment planning for closed or open reduction of the fracture. The mechanism of injury is determined by the radiographic appearance of the fracture. Anteroposterior, mortise, and lateral ankle radiographs are necessary to classify an injury using the Lauge-Hansen classification system.205 Supination-Adduction The SAD mechanism accounts for 20% of ankle fractures.212,213 The fracture pattern results from the foot being in a fixed supinated position while the ankle translates from a lateral to medial position.212,214–216 The talus adducts within the ankle mortise, causing traction of the lateral malleolus and a compressive force on the medial ankle structures. There are two stages to this fracture group. Ankle radiographs are useful for identifying SAD stage II fracture as well as identifying ankle joint subluxation and dislocation. Stage I SAD results in a rupture of the lateral collateral ankle ligament, an avulsion fracture of the tip of the lateral malleolus, or a transverse fracture of the fibula at the level of the ankle joint. Increased force can produce a stage II

SAD injury, resulting in an oblique-to-vertical fracture of the medial malleolus. The medial fracture is caused by the impact of the talus on the medial malleolus (Figure 16-41). TABLE 16-21   Lauge-Hansen209 Classification of Ankle Fractures Mechanism Stage Description Transverse avulsion fracture of the fibula at or below the Supination-adduction I ankle joint level or rupture of the lateral collateral ligaments Stage I plus an oblique-to  II vertical fracture of the medial malleolus Rupture of the anteriorSupination-eversion inferior tibiofibular ligament (supination-external I within its substance or at the rotation) ligamentous–osseous interface Stage I plus spiral-oblique   II fracture of the fibula beginning at the ankle joint Stage II plus rupture of the posterior-inferior tibiofibular   III ligament or an avulsion fracture of the posterior malleolus Stage III plus an avulsion fracture of the medial   IV malleolus or rupture of the deltoid ligament Pronation-abduction

I

 

II

Transverse avulsion fracture of the medial malleolus or rupture of the deltoid ligament Stage I plus rupture of the anterior and posterior-inferior tibiofibular ligaments within their substance or at their ligamentous–osseous interface

 

III

Pronationeversion(pronationexternal rotation)

I

 

II

 

III

 

IV

Stage II plus a short oblique fracture of the fibula with the fracture line beginning immediately proximal to the ankle joint with an occasional butterfly fragment Transverse avulsion fracture of the medial malleolus or rupture of the deltoid ligament Stage I plus rupture of the anterior-inferior tibiofibular ligament and interosseous membrane Stage II plus high spiraloblique fibular fracture with the proximal extent of the fracture depending on how far proximally the interosseous ligament rupture has progressed Stage III plus rupture of the posterior interior tibiofibular ligament or avulsion fracture of the posterior malleolus

Supination-Eversion or Supination-External Rotation The SER mechanism is most common, accounting for 40% to 70% of all ankle fractures. The fracture occurs when a supinated foot is subjected to internal rotation of the leg (external rotation of the foot). In the supinated foot, the medial (deltoid) ligament is relaxed while the lateral ligaments are taut. During this mechanism, the lateral structures are under stress and, as the force continues, the talus impacts the fibula; also, as the medial force continues, the medial structures are compromised.215

FIGURE 16-41. SAD injury with dislocated talus, significant medial displacement of transverse fibular fracture and vertical medial malleolar

fracture. The SER fracture mechanism has four stages. In stage I SER, the talus rotates laterally and pushes the lateral malleolus posteriorly. This results in failure of the AITFL, either within the substance of the ligament or as an osseous failure from the tibia or lateral malleolus at the AITFL attachment. A small flake of bone avulsed off the anterior tubercle of the tibia is known as a Chaput or Tillaux fracture. If the avulsion is from the anterior tubercle of the fibula, it is referred to as a Wagstaffe fracture. This injury can be confused with a classic inversion ankle sprain because the radiographic study is unremarkable and there is anterior ankle pain clinically. However, the pain associated with this SER fracture is often along the anterior distal tibiofibular area rather than at the ATFL ligament. At this stage the injury is stable.212,214,216,217 If the force continues, the talus continues rotating laterally, producing a spiral-oblique fracture of the fibula beginning at the level of the ankle joint anteriorly, then progressing laterally, posteriorly, and superiorly. This progression causes a posterior spike at the proximal extent of the fibular fracture. Stage II SER is the most common ankle fracture, regardless of the mechanism, and is considered a stable fracture. To classify the injury as a stage II SER fracture, rupture of the AITFL and spiral fracture of the fibula must be present.212,214,216 Stage III SER injury is caused by lateral rotation of the talus in the ankle mortise, which results in failure of the PITFL, either within the ligament itself or at the site of attachment to the tibia or fibula. An avulsion fracture of the posterior tibia (posterior malleolus) is also referred to as a Volkmann fracture. The size of the fracture ranges from a flake to less than a quarter or greater than one quarter of the articular surface. Treatment is dependent on the amount of the posterior malleolus articular surface involved. Stage IV SER injury occurs if the force is great enough and the talus continues to rotate laterally until either a deltoid ligament rupture or a transverse fracture of the medial malleolus occurs (Figure 16-42). The deltoid ligament rupture frequently appears radiographically as widening of the medial gutter of the ankle joint. An increased medial clear space of greater

than 4.0 mm is considered diagnostic. Schuberth et al.218 have found, on the contrary with arthroscopic evaluation, that the deep deltoid ligaments on average were intact up to 6.6 mm.218

FIGURE 16-42. SER injury pattern, stage 4. Nevertheless, ankle joint alignment is of utmost importance. Ankle radiography, including anteroposterior, lateral, and mortise views, is essential for evaluating ankle joint dislocation or its susceptibility to dislocate. An occult deltoid ligament injury may not always be perceived on non–weightbearing ankle radiographs. Eversion stress radiography can confirm deltoid ligament tear; however, patients do not tolerate manual stress radiography as well as gravity stress radiography (Figure 16-43).219 The patient pain rating on a visual analog scale was 3.45 and 6.14, respectively, in favor of gravity stress radiography. In addition, diagnostic sensitivity also favors gravity stress radiography at 55% compared to 45% for manual stress radiography.219 Anecdotally, gravity stress radiography can be useful at diagnosing syndesmotic disruption not only for SER mechanisms but also for other Lauge-Hansen injury patterns. Pronation-Abduction The PAB mechanism of ankle fracture is nearly the reverse of the SAD mechanism. The foot is in a pronated position when the talus abducts in the ankle mortise, which results in traction of the medial ankle complex and compression laterally (Figure 16-44). A stage I PAB injury results in either a rupture of the deltoid ligament or a transverse fracture of the medial malleolus below the level of the syndesmosis. Stage II PAB injury is characterized by rupture of both the AITFL and PITFL. A Tillaux-Chaput or Wagstaffe avulsion fracture or a fracture of the posterior tibia is possible. It is important to recognize that a stage II PAB injury can occur without any fracture and is unstable. This fracture pattern is often misdiagnosed as a high ankle sprain. Stage III PAB injury occurs as the talus continues to rotate and produces a fracture of the fibula, generally below the level of the syndesmosis. The fibular fracture tends to be a short oblique fracture that is visualized on the anteroposterior and lateral ankle views. A lateral butterfly fragment or spike

will be seen in anteroposterior view, while a transverse fracture line will be seen in the lateral view.212,214

FIGURE 16-43. Gravity stress view of the ankle requires the patient to be in a lateral decubitus position on the injured side, with the distal half of leg hanging off of the radiography exam table. A mortise ankle view, with the tibia internally rotated approximately 10°, is taken.216

FIGURE 16-44. Pronation-abduction injury stage 3, with medial malleolar avulsion fracture (curved arrow) and short oblique lateral malleolar fracture (straight arrow) visualized. On an anteroposterior ankle radiograph, widening of the medial ankle mortise correlates to the presence of a deltoid ligament tear. If there is a medial malleolus fracture, it has a transverse orientation. The presence of a short oblique fibular fracture is indicative of a stage III PAB injury. By deduction, the posterior tibiofibular ligament is likely ruptured despite absence of a posterior fracture on the lateral view.216 Pronation-Eversion or Pronation-External Rotation The hallmark of the PER mechanism is a high fibular fracture above the syndesmosis. The foot is in a pronated position and there is subsequent internal rotation of the lower leg or external rotation of the foot. When the foot is pronated, the deltoid ligament is taught, causing injury to the medial ankle structures. As the deforming force continues, a spiral fracture of the fibula and a posterior malleolus fracture occur. The PER mechanism has four stages. Stage I PER injury initially results in stress along the medial ankle structures leading to either deltoid ligament rupture or transverse fracture of the medial malleolus. Rupture of the deltoid ligament may be occult or appear as medial widening of the ankle mortise. In stage II PER, the talus continues to rotate laterally, producing a rupture of the anterior syndesmosis that involves the AITFL, extending proximally into the interosseous membrane. This can cause a Tillaux-Chaput or Wagstaff fracture. As the talus continues rotating laterally, it impacts the fibula, producing a stage III PER fracture. Stage III PER injury is defined by a spiral or oblique fracture of the fibula above the syndesmosis. The fracture is located at the lateral superior extent of the interosseous membrane rupture (Figure 16-45). The fibular fracture may be as superior as the fibular neck, which is referred to as a Maisonneuve fracture. Though Maisonneuve’s fracture is not common, representing less than 25% of all PER injuries, it is often missed. Long-leg radiographs should be performed to evaluate a high fibular neck fracture. This type of fibular

fracture rarely needs open reduction or internal fixation, due to the large amount of surrounding muscle mass that provide protection and stability.212,216 A stage IV PER injury involves the posterior ankle structures; the force produces a PITFL rupture or a large fracture of the posterior malleolus. A critical evaluation of the posterior malleolar fracture size is essential for treatment planning and optimization of outcomes.217 However, McDaniel and Wilson220 found in their study that estimates of posterior malleolar fractures are consistently 5% to 15% smaller from radiographic estimates when compared to intraoperative measurements. Considering all ankle fractures, this mechanism is considered the most unstable. Pronation-Dorsiflexion This rare injury, accounting for less than 0.5% of ankle fractures, was not a part of the original Lauge-Hansen classification. However, since the mechanism of injury is determined by radiographic findings, it is frequently included with the classification system. The pronation-dorsiflexion (PDF) fracture occurs when axial compression forces impaction of the talus into the ankle mortise on a pronated foot. Initially in the stage I PDF injury, a transverse or slightly oblique fracture of the medial malleolus occurs. As the force continues, the second fracture is an intra-articular anterior lip fracture of the tibia; this stage II finding is one of the key fractures when classifying this injury pattern. A stage III PDF injury is characterized by a supramalleolar fibular fracture. Stage IV PDF injury adds a transverse posterior tibial fracture. The posterior fracture connects with the anterior lip fracture, resulting in comminution of the distal tibial metaphysis. The stage IV PDF fracture may be overlooked on the lateral radiograph; therefore, when positioning the patient, slight external leg rotation is beneficial if this injury pattern is suspected.

FIGURE 16-45. PER injury pattern, stage 4 (arrow indicates the proximal fibular fracture/periosteal reaction just distal to the fibular neck.). Despite the complexity of the Lauge-Hansen classification, the following rules simplify the multiple factors involved.205 First, it is important to

understand that injuries occur in a predictable sequence and stages cannot be skipped. All injuries within the classification sequence that occur in the stages prior to the cessation of the force are present. The final stage that occurs represents the stage of the fracture. Injuries occurring in stages beyond the point the force ceases are not included. For example, a SER stage III injury includes the fractures of an SER stage I, stage II, and stage III and is, therefore, called an SER stage III injury. The second rule that establishes an efficient approach to classifying the fractures is to first identify the presence and type of fibular fracture seen on radiographs. Finally, in the absence of a fibular fracture, identification of a medial malleolar fracture or widening of the medial ankle mortise is most helpful when classifying an ankle fracture. When evaluating for ankle fracture, other soft tissue and osseous conditions should be considered, such as ankle impingement syndrome, ankle sprain, talar dome OCD, metatarsal fracture, and metatarsalgia. The most common of the associated metatarsal fractures (such as the fifth) can frequently be observed on anteroposterior and lateral ankle views, in addition to foot views. Danis-Weber The second most common ankle fracture classification utilized is the DanisWeber system, likely due to its ease of use (Table 16-22). In 1949, Danis introduced a pathologic-anatomical classification system designed for operative treatment. It was later modified by Weber and adopted by the Arbeitsgemeinschaft fur Osteosynthesisfragen (AO) group, which translates to the Association for the Study of Internal Fixation.221 This system is based on the location of the fibular fracture compared to the ankle syndesmosis, since the lateral structures are considered the main stabilizing structures of the ankle. The fibular fracture predicts the integrity of the tibiofibular syndesmosis and need for operative corrections. There are certain instances in which the Lauge-Hansen and Danis-Weber classifications overlap. However, a direct translation of Lauge-Hansen to Danis-Weber is not always possible, since there are disagreements in regards to the mechanism of injury. There are three types in the Danis-Weber classification system.212,213,222 Type A is a transverse fracture of the distal fibula, parallel or below the level of the tibiotalar articulation (syndesmosis) (Figure 16-46). Type A is

synonymous with the Lauge-Hansen SAD injury. Either a rupture of the lateral collateral ankle ligament or an avulsion or transverse fracture of the lateral malleolus can occur. If the force is increased, it may result in an oblique-to-vertical fracture of the medial malleolus. In general, the syndesmosis and deltoid ligament are considered intact. TABLE 16-22   Danis-Weber221 Classification of Ankle Fractures

Type B is an injury that results in a spiral or oblique fibular fracture, beginning at the level of the syndesmosis (Figure 16-42). This is similar to the Lauge-Hansen PAB and SER injuries. There may be syndesmosis rupture, transverse fracture of the medial malleolus, or deltoid ligament rupture. Additionally, the AITFL may be ruptured. Fracture of the posterior malleolus may occur. This accounts for the most common mechanism of all ankle fractures.212,213,222 Type C fracture occurs above the level of the syndesmosis (Figure 16-45). This type corresponds to the Lauge-Hansen PER injuries. Either deltoid ligament rupture or medial malleolus fracture is associated with this injury. There is always rupture of the syndesmosis ligament and interosseous membrane extending to the level of the fibular fracture. Similar to type B, a Volkmann fracture may occur. This mechanism is the least common of all ankle fractures.212,213,222 Many authors have recognized the inability of any classification to account for all fracture patterns. To assess the interobserver variation and the ability of Lauge-Hansen, AO, and Broos-Bisshop classification systems to encompass all possible fracture patterns, radiographs of 293 patients with 294 malleolar fractures were studied. The overall percentage of unclassified fracture patterns was 0.7% with the Broos-Bisshop system, 10% with the Lauge-Hansen system, and 8.7% with the AO system. Low interobserver reliability was observed with all three classification systems.223

FIGURE 16-46. Type A Weber. Lateral malleolar fracture distal to the ankle syndesmosis. Others have supported these findings, with one such study demonstrating the highest rate of unclassified fractures (10%) using the Lauge-Hansen system.213 The reliability of the Lauge-Hansen classification was also determined to be poor regarding intraobserver reproducibility (64%–82%).224 Rodriguez et al.225 assessed whether the Lauge-Hansen and AO ankle fracture classifications radiographically correlated with in vivo injuries based on observed mechanism of injury. Thirty injury videos and corresponding radiographs were collected. They found that the Lauge-Hansen and AO

systems were 65% and 81% consistent in predicting fracture patterns, respectively, thus concluding that the AO classification correlated more closely with in vivo injuries.225 Michelson et al. attempted to reproduce Lauge-Hansen’s findings using biomechanical techniques in a cadaveric model for the commonly observed SER mechanism. Though they had limitations positioning the foot and distal leg, the study showed that the SER testing configuration did not result in the typical SER fibular fracture pattern, and no ankle had an oblique fibular fracture. Additionally, only 7 of 30 AITFL and 4 of 30 PITFL injuries were reproducible, reducing the predictive value of radiographically silent soft tissue injury.222 A retrospective cohort study of 59 patients examined the ability of the LaugeHansen classification to predict ligamentous injury using x-ray and MRI. Of the 49 fractures that could be classified according to the classification, 53% had ligamentous failure and fracture patterns that did not coincide with predictions. In addition, they found concomitant complete ligamentous rupture and corresponding malleolar fracture in over 65% of the cases. The result of their study found poor prediction of soft tissue injury using MRI based on Lauge-Hansen’s mechanism, leading to the conclusion that the Lauge-Hansen classification should not be used as the only guide for treatment of rotational ankle fracture.226 Although there is no classification system that incorporates all fracture patterns or has a high interobserver and intraobserver reliability, the most accepted and utilized classifications remain the Lauge-Hansen and DanisWeber systems. Radiographic Approach to Ankle Fracture In general, avulsion fractures are usually transverse while impaction fractures are often oblique or spiral in appearance. The standard radiographic ankle study should include anteroposterior, mortise, and lateral views, which are required to classify fractures using the Lauge-Hansen system and to plan treatment. On the anteroposterior view, the tibiotalar joint is partially overlapped by the fibula. The lateral view should include the base of the fifth

metatarsal to rule out proximal fracture. The mortise view is obtained by positioning the ankle approximately 15° to 20° internally rotated; this aligns the medial and lateral malleoli parallel to the image receptor, producing an un-obscured view of the talar dome and plafond. When patients have proximal leg tenderness or medial clear space widening with no obvious fibular fracture, a full-length radiograph of the tibia and fibula must be obtained to rule out a Maisonneuve fracture. CT and MRI are rarely indicated with exception of triplane ankle and pilon fractures.207,215,216 TIBIAL PILON FRACTURE A pilon fracture of the tibia involves the horizontal articular surface of the distal tibia and extends proximally. In 1911, Destot compared the distal tibia to a pestle and termed distal tibial fractures that extended into the ankle joint as “pilon fractures.”227 In 1950, Bonin described the same fracture as a “plafond fracture,” as the fracture occurred at the roof of the ankle joint. Currently, these terms are used interchangeably. The pilon represents 5% to 10% of tibial fractures, but less than 1% of lower extremity fractures.228,229 Pilon fractures are often due to an axial load across the ankle resulting in a comminuted, intra-articular distal tibial fracture. There are two primary mechanisms of injury that can result in a tibial plafond fracture. Lower energy fractures are often from a fall or sport injury such as skiing, which leads to a rotational fracture of the distal tibia. Higher energy injuries are often the result of axial loading where the talus is driven into the distal tibia; it is associated with a fall from height or MVA. The impaction causes an implosion of the distal tibial articular surface, resulting in comminution of the metaphyseal bone. The foot position at the moment of impact determines the area of the tibial articular surface most affected. A vertical compression force while the foot is in a plantigrade position will lead to central depression of the tibia. If the same force is applied to a plantarflexed or dorsiflexed foot, the injury will be a posterior or anterior malleolar fracture, respectively.229,230 234 and AO/OTA235 Pilon Fracture TABLE 16-23   Ruedi–Allgower Classifications Classification Type Description

Ruedi–Allgower classification

I

 

II

 

III

AO/OTA classification

43-A

 

43-A1

 

43-A2

 

43-A3

 

43-B

 

43-B1

 

43-B2

 

43-B3

 

43-C

No comminution or displacement of intraarticular fragments Some displacement but no comminution or impaction of the joint surface Significant displacement with comminution and/or impaction of the joint surface Extra-articular fracture of the tibial plafond Simple fracture Wedge fracture with contact between the main fragments preserved Comminuted fracture with no contact between the main fragments Intra-articular fracture involving only part of the tibial plafond joint surface Simple split fracture Depressed split fracture Local comminution of articular surface preserving one cortical wall Intra-articular fracture involving only part of the joint surface

 

43-C1

 

43-C2

 

43-C3

Split fracture involving the metaphysis (V, Y, or T fracture) Metaphyseal impaction without dissociation of the articular surface Depressed, comminuted fracture of the articular surface

Bonin and Böhler were one of the first to analyze and classify pilon fractures.231,232 Destot classified pilon fractures by dividing them into four main subgroups, based on anatomic location. The subgroups stratified the fractures as posterior marginal, anterior marginal, explosion, and supraarticular.227 This classification was neither descriptive nor comprehensive enough to guide treatment or predict outcomes. There are several classification systems that have been developed defining a worse prognosis for fractures created by higher energy forces.230,233 Although there are no universally accepted classification systems, the two main classification systems used are Ruedi–Allgower and the AO/OTA classification (Table 1623).234,235 The most well known is the Ruedi–Allgower classification, while the AO/OTA was developed more recently to unify the multiple classification systems in the literature.230 The Ruedi–Allgower and AO classification systems are based on radiographic findings; another classification system, by Topliss et al.236, is based on radiography and CT imaging. The Ruedi–Allgower classification system is based on the degree of displacement, comminution, and impaction of the fracture fragment.234 It divides the fracture into three subtypes. Type I is a tibial plafond fracture, absent of displacement or comminution (nondisplaced-split). Type II has displacement, but remains absent of comminution or impaction (displacedsplit). Type III presents with significant displacement with impacted and/or comminuted fragments of the distal tibial articular surface (displacedcomminuted). Type III has been noted to be the most prevalent, comprising

of 47% of pilon injuries; type II accounts for 28% and type I 25% of pilon fractures.222 The Ruedi–Allgower classification system has demonstrated fair interobserver reliability in various studies.237,238 One study demonstrated that the greatest difficulty involved identifying the fragments of the tibial articular surface on radiographs, which led to the conclusion that the number and displacement of small fragments may be the cause of poor performance on reliability analyses.1 As many as 10% to 30% of pilon fractures are open fractures, which are then most often classified using the Gustilo and Anderson classification. Gustilo and Anderson239 developed their three-tier classification based on retrospective and prospective analyses of 1025 open fractures. The classification accounts for local soft tissue injury, contamination, and the severity of the fracture pattern. The type III group was later subdivided by Gustilo et al.,240 based on severity, into subgroups A, B, and C; the type IIIC fracture accounted for any open fracture with a vascular injury (Table 16-24). The AO/OTA classification system is a generic classification system designed to standardize fracture classifications; it provides a more descriptive system that can be applied to any bone in the body. The AO system is based on the degree of involvement of the articular surface. The first part of the classification defines the location of the bone involved, followed by subgroups that are based on the degree of continuity between the metaphysis and epiphysis. Type A fractures are extra-articular, type B fractures are partially articular, and type C fractures are complete articular fractures. Each of these three subgroups is then further divided into three stages based on the degree of comminution. Stage 1 represents no comminution of the metaphysis or the epiphysis. Stage 2 fractures involve comminution of the metaphysis, sparing the epiphysis. Stage 3 fractures exhibit comminution of both the metaphysis and the epiphysis. Partial comminution type B pilon fractures are most common; 58% of type B fractures have an intact fibula, while nearly 70% of all pilon fractures present with a fractured fibula and are type C fractures. Over 50% of the type C fractures were further classified as stage 3 injuries, due to the severe comminution present.241 Interobserver reliability of the AO classification for pilon fractures has been determined to be moderate at 0.57 for the fracture type while determination of the

subgroups was insufficient.237 The Topliss classification uses both radiography and CT. Three main groups of fractures were identified, based on the main orientation of the fracture lines. These were defined as the coronal family, sagittal family, and the comminuted group that encompasses fractures that did not fall within the other two families. The coronal family, found in older patients after lower energy trauma, occurred most often (56%) and typically caused valgus angulation. The sagittal family fracture, found in younger patients after higher energy injuries, was prevalent in one-third of the cases and caused varus angulation. The comminuted group was prevalent in only 6% of cases.236 TABLE 16-24   Gustilo and Anderson239 Classification of Open Fractures Type Description of Soft Tissue Description of Bone Fracture with clean wound 1 cm II

Surrounding tissue with minor or no Fracture instability is signs of contusion moderate or severe No dead muscle Extensive soft tissue damage

Frequently with compromised vascularity with or without wound contamination Type III IIIA   Often from high-energy subtype trauma; Adequate soft tissue coverage 237 modification of fractured bone IIIB   Extensive soft tissue loss with periosteal stripping and bone   exposure; Often with massive contamination   IIIC   Arterial injury requiring repair III

Fracture pattern is complex with marked instability

To date, pilon fracture classification systems continue to have interobserver and intraobserver agreement only, even with the help of CT scans.242 Advanced imaging modalities such as CT and MRI have not enhanced intraobserver reliability when compared to using radiographs exclusively. Radiographic Approach to Pilon Fracture Pilon fracture is best diagnosed using radiography (Figure 16-47). Anteroposterior and lateral views, along with a thorough examination for concomitant injury, is essential. Full-length tibial and fibular orthogonal views are also suggested for this purpose. CT scans displaying axial, sagittal, and coronal reconstructions provide details not always apparent radiographically, which may change surgical planning.236,243 Topliss et al.236 reported on the anatomical features of 126 consecutive tibial pilon fractures using CT scans and radiographs. CT images revealed lesions that may have been missed with other examinations, such as tubercle de Chaput avulsion fracture and syndesmotic rupture.236 Additional information from CT has been reported in 83% of cases, resulting in changes in surgical planning in 64% of cases. Additionally, surgeons reported subjectively that surgical time was reduced based on the information from the CT scans, with a better understanding of the fracture pattern.243 In a two-stage operative practice, CT should be completed after applying the external fixation. In general, MRI is not required in the treatment of tibial pilon fractures. PHYSEAL INJURY During skeletal growth, physeal injuries can result in premature growth arrest, deformed growth, malalignment, or arthritis. The result of physeal injuries may vary from minimal to severe disruption of normal anatomy. Though the mechanisms of injury are similar in children and adults, the injury patterns are unalike, due to ligaments being stronger than the immature bone in children.244–246 Physeal injuries occur more commonly in boys than in girls, and are usually acquired during athletic activities.247

FIGURE 16-47. Pilon fracture. The epiphyseal plate consists of four zones: the resting zone, zone of proliferation, zone of hypertrophy, and zone of calcification.245 The zone of hypertrophy is the weakest due to relatively less collagen and calcification compared to the other zones. Consequently, separation through this layer is most at risk.248 This difference in composition contributes to specific injury patterns in the ankle during certain periods of growth. Pediatric ankle injuries are the second most common injury in children between the ages of 10 and 15, comprising 9% to 18% of all pediatric physeal injuries.249–251 Different areas of the distal tibial physis ossify in a sequential manner. Starting approximately at the age of 14 years in girls and 16 years in boys, growth plate closure occurs over approximately 18 months.252,253 The distal tibial physis first ossifies centrally, progresses medially, and is finalized laterally.254 Due to this maturation sequence, specific fracture patterns present, which are termed transitional fractures.255–257 The fracture pattern is dependent on the amount of calcification as the physeal plate closes, thus changing the weakest area of the physeal plate. Examples of transitional fractures are the Tillaux and triplane ankle fractures.246,258 TABLE 16-25   Salter–Harris245 Classification System Salter–Harris Type Fracture Pattern Fracture line through entire physis at the level I of the zone of hypertrophy Fracture line through part of physis, exiting II the metaphysisa Fracture line through part of physis, exiting III the epiphysis Fracture line through the epiphysis, physis, IV and metaphysis V Compression fracture of the physis aIn type II fractures, the triangular metaphyseal fracture fragment is known

as the Thurston Holland fragment.

FIGURE 16-48. Schematic representation of the Salter–Harris classification of epiphyseal plate fractures. (From Figure 19-36 in Spinosa FA. Classification of fractures and dislocations. In: Christman RA. Foot and Ankle Radiology. 1st ed. St. Louis, MO: Churchill Livingstone; 2003:chap 19.) Epiphyseal Ankle Injury Classification (Dias and Tachdjian258) Type Stage Classification of Injury Separation of distal fibular epiphyseal plate with Supination-inversion I possible fracture into the metaphysis Intra-articular fracture TABLE 16-26  

 

II

Supination-plantarflexion

I

Supination-external rotation

I

 

II

Pronation-eversion-external rotation

I

through the tibial epiphysis with possible fracture into the metaphysis Separation of the tibial epiphyseal plate, usually with a metaphyseal fragment Separation of the tibial epiphyseal plate with a long spiral fracture of the distal tibia Separation of the tibial epiphyseal plate with a long spiral fracture of the distal tibia and a spiral fracture of the fibula above the growth plate Separation of the tibial epiphyseal plate with a metaphyseal fragment and a short oblique fracture in the diaphysis of the fibula

From Table 19-4 in Spinosa FA. Classification of fractures and dislocations. In: Christman RA. Foot and Ankle Radiology. 1st ed. St. Louis, MO: Churchill Livingstone; 2003:chap 19. Multiple authors have published classification systems to provide a better understanding and a communication tool for physeal injuries.245,248,252,259,260 The classification system published by Salter and Harris245 has become the most popular and has been utilized in classification of physeal injuries throughout the body261 (Table 16-25, Figure 16-48). Examples are shown in Figure 16-49. In 1978, Dias and Tachdjian258 devised a classification system specifically for pediatric ankle fractures, which was modeled after the adult ankle fracture classification by Lauge-Hansen (Table 16-26). The interobserver and intraoberserver reliability of these two

classification systems have not been evaluated as thoroughly as other fracture classification systems. The Salter–Harris classification system was found to have substantial agreement in intraobserver reliability and moderate interobserver reliability.262 In 82.8% of the cases reviewed, the additional information from CT imaging to plain radiographs did not change treatment decision making.262 The Ottawa Ankle Rules has been shown to be a useful tool in indicating necessity for radiographic evaluation in children.263,264 Meta-analysis found Ottawa Ankle Rules to be more applicable in use of children older than 5 years, due to the requirement of assessing ambulatory ability within the Ottawa Ankle Rules.262 The sensitivity and missed fracture rate found in the meta-analysis, when using Ottawa Ankle Rules to children older than 5 years, was found to be 98.5% and 1.2%, respectively. It is estimated that there would be a 24.8% reduction in radiographs when using Ottawa Ankle Rules in evaluating childhood injuries.262

FIGURE 16-49. Examples of physeal injuries. A: Separation of the hallux proximal phalanx epiphysis from the metaphysis (Salter–Harris type I). B: Short separation of the epiphysis from the metaphysis laterally with a large triangular metaphyseal fragment (Salter–Harris type II). C: Epiphysis is separated from the metaphysis medially and centrally with a small epiphyseal fragment (Salter–Harris type III). REFERENCES   1. Dirschl DR, Cannada LK. Classification of fractures. In: Rockwood & Green’s Fractures in Adults. 6th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2006:chap 2.   2. Martin JS, Marsh JL, Bonar SK, et al. Assessment of the AO/ASIF fracture classification for the distal tibia. J Orthop Trauma. 1997;11(7):477– 483.   3. Armagan OE, Shereff MJ. Injuries to the toes and metatarsals. Orthop Clin North Am. 2001;32(1):1–10.   4. Fife D, Barancik JL. Northeastern Ohio trauma study III: incidence of fractures. Ann Emerg Med. 1985;14:244–248.   5. Daly N. Fractures and dislocations of the digits. Clin Podiatr Med Surg. 1996;13(2):309–326.   6. Anderson LD. Injuries of the forefoot. Clin Orthop Relat Res. 1977;122:18–27.   7. Park DY, Han KJ, Han SH, et al. Barefoot stubbing injuries to the great toe in children: a new classification by injury mechanism. J Orthop Trauma. 2013;27:651–655.   8. Venegas L, Rainierri JJ, Rzonca EC. Fracture of the fifth digit. J Am Podiatr Med Assoc. 1995;85(3):166–168.   9. Giannestras NJ, Sammarco J. Fractures and dislocations in the foot. In: Rockwood CA, Green DP, eds. Fractures. Philadelphia, PA: Lippincott;

1975:1400–1489.  10. McBryde AM, Anderson RB. Sesamoid foot problems in the athlete. Clin Sports Med. 1988;7:41–60.  11. Sanders TG, Rathur SK. Imaging of painful conditions of the hallucal sesamoid complex and plantar capsular structures of the first metatarsophalangeal joint. Radiol Clin North Am. 2008;46:1079–1092.  12. Giannikas AC, Papachristou G, Papavasiliou N, et al. Dorsal dislocation of the first metatarso-phalangeal joint: a report of four cases. J Bone Joint Surg Br. 1975;57(3):384–386.  13. Jahss MH. Traumatic dislocation of the first metatarsophalangeal joint. Foot Ankle. 1980;1:15–21.  14. Brunet JA. Pathomechanics of complex dislocations of the first metatarsophalangeal joint. Clin Orthop Relat Res. 1996;332:126–131.  15. Salamon PB, Gelberman RH, Huffer JM. Dorsal dislocation of the metatarsophalangeal joint of the great toe: a case report. J Bone Joint Surg Am. 1974;56(5):1073–1075.  16. Ribbons WJ, Natarajan R, Alavala S. Pediatric foot fractures. Clin Orthop Relat Res. 2005;432:107–115.  17. Herrera-Soto JA. Fractures of the fifth metatarsal in children. J Pediatr Orthop. 2007;27(4):427–431.  18. Rammelt S, Heineck J, Zwipp H. Metatarsal fractures. Injury. 2004;35:S-B77–S-B86.  19. Torriani M, Thomas BJ, Bredella MA, et al. MRI of metatarsal head subchondral fractures in patients with forefoot pain. AJR Am J Roentgenol. 2008;190(3):570–575.  20. Buddecke DE, Polk MA, Barp EA. Metatarsal fractures. Clin Podiatr Med Surg. 2010;27:601–624.

 21. Zwitser EW, Breederveld RS. Fractures of the fifth metatarsal: diagnosis and treatment. Injury. 2010;41:555–562.  22. Petrisor BA, Ekrol I, Court-Brown C. The epidemiology of metatarsal fractures. Foot Ankle Int. 2006;27:172–174.  23. Laurich LJ, Witt CS, Zielsdarf LM. Treatment of fractures of the fifth metatarsal bone. J Foot Surg. 1983;22:207–211.  24. Richli ER, Rosenthal DI. Avulsion fracture of the fifth metatarsal: experimental study of pathomechanics. AJR Am J Roentgenol. 1984;143: 889–891.  25. Theodorou DJ, Theodorou SJ, Kakitsubata Y, et.al. Fractures of the proximal portion of fifth metatarsal bone: anatomic and imaging evidence of a pathogenesis of avulsion of the plantar aponeurosis and the short peroneal muscle tendon. Radiology. 2003;226:857–865.  26. Stewart IM. Jones’s fracture: fracture of base of fifth metatarsal. Clin Orthop. 1960;16:190–198.  27. Pao DG, Keats TE, Dussault RG. Avulsion fracture of the base of the fifth metatarsal not seen on conventional radiography of the foot: the need for additional projection. AJR Am J Roentgenol. 2000;175:549–552.  28. Jones R. Fracture of the base of the fifth metatarsal bone by indirect violence. Ann Surg. 1902;35(6):697–700.  29. Chuckpaiwong B, Queen R, Easley ME, et al. Distinguishing Jones and proximal diaphyseal fractures of the fifth metatarsal. Clin Orthop Relat Res. 2008;466:1966–1970.  30. Lawrence SJ, Botte MJ. Jones fractures and related fractures of the proximal fifth metatarsal. Foot Ankle. 1993;14(6):358–365.  31. Dameron TB Jr. Fractures and anatomic variations of the proximal portion of the fifth metatarsal. J Bone Joint Surg Am. 1975;57:788–792.  32. Quill GE Jr. Fractures of the proximal fifth metatarsal. Orthop Clin

North Am. 1995;26:353–361.  33. Kavanaugh JH, Brower TD, Mann RV, et al. The Jones fracture revisited. J Bone Joint Surg Am. 1978;60:776–782.  34. Torg JS, Balduini FC, Zelko RR, et al. Fractures of the base of the fifth metatarsal distal to the tuberosity: classification and guidelines for nonsurgical and surgical management. J Bone Joint Surg Am. 1984;66:209–214.  35. Englanoff G, Anglin D, Hutson HR. Lisfranc fracture-dislocation: a frequently missed diagnosis in the emergency department. Ann Emerg Med. 1995;26(2):229–233.  36. Saab M. Lisfranc fracture-dislocation: an easily overlooked injury in the emergency department. Eur J Emerg Med. 2005;12:143–146.  37. Norfray JF, Geline RA, Steinberg RI, et al. Subtleties of Lisfranc fracture-dislocations. AJR Am J Roentgenol. 1981;137:1151–1156.  38. Peicha G, Labovitz J, Seibert FJ, et al. The anatomy of the joint as a risk factor for Lisfranc dislocation and fracture-dislocation. J Bone Joint Surg Br. 2002;84(7):981–985.  39. Gotha HE, Lareau CR, Fellars TA. Diagnosis and management of Lisfranc injuries and metatarsal fractures. R I Med J. 2013;96(5):33–36.  40. Davies MS, Saxby TS. Intercuneiform instability and the “gap” sign. Foot Ankle Int. 1999;20(9):606–609.  41. Hardcastle PH, Reschauer R, Kutscha-Lissberg E, et al. Injuries to the tarsometatarsal joint: incidence, classification, and treatment. J Bone Joint Surg Br. 1982;64:349–356.  42. Haapamaki V, Kiuru MJ, Koskinen S. Lisfranc fracture-dislocation in patients with multiple trauma: diagnosis with multi-detector computed tomography. Foot Ankle Int. 2004;25(9):614–619.  43. Raikin SM, Elias I, Dheer S, et al. Prediction of mid-foot instability in the subtle Lisfranc injury: comparison of magnetic resonance imaging with

intra-operative findings. J Bone Joint Surg Am. 2009;91(4):892–899.  44. Nunley JA, Vertullo CJ. Classification, investigation, and management of midfoot sprains: Lisfranc injuries in the athlete. Am J Sports Med. 2002;30(6):871–878.  45. Shapiro MS, Wascher DC, Finerman GA. Rupture of Lisfranc’s ligament in athletes. Am J Sports Med. 1994;22:687–691.  46. Foster SC, Foster RR. Lisfranc’s tarsometatarsal fracture-dislocation. Radiology. 1976;120(1):79–83.  47. Van Rijn J, Dorleijn DMJ, Boetes B, et al. Missing the Lisfranc fracture: a case report and review of the literature. J Foot Ankle Surg. 2012;51:270–274.  48. Potter HG, Deland JT, Gusmer PB, et al. Magnetic resonance imaging of the Lisfranc ligament of the foot. Foot Ankle Int. 1998;19:438–446.  49. Vuori J, Aro HT. Lisfranc joint injuries: trauma mechanisms and associated injuries. J Trauma. 1993;35(1):40–45.  50. Gupta RT, Wadhwa RP, Learch TJ, et al. Lisfranc injury: imaging findings for this important but often-missed diagnosis. Curr Probl Diagn Radiol. 2008:3:115–126.  51. Schenk H Jr, Heckman D. Fractures and dislocations of the forefoot: operative and non-operative treatment. J Am Acad Orthop Surg. 1995;3:70– 78.  52. Sherief TI, Mucci B, Greiss M. Lisfranc injury: how frequently does it get missed? And how can we improve? Injury. 2007;38:856–860.  53. Goiney RC, Connell DG, Nichols DM. CT evaluation of tarsometatarsal fracture-dislocation injuries. AJR Am J Roentgenol. 1985;144:985–990.  54. Lu J, Ebraheim NA, Skie M, et al. Radiographic and computed tomographic evaluation of Lisfranc dislocation: a cadaver study. Foot Ankle Int. 1997;18:351–355.

 55. Marin-Pena OR, Recio FV, Gomez, TS, et al. Fourteen years follow up after Lisfranc fracture-dislocation: functional and radiological results. Injury. 2012;43(suppl 2):S79–S82.  56. Eleftheriou KI, Rosenfeld PF, Calder JDF. Lisfranc injuries: an update. Knee Surg Sports Traumatol Arthrosc. 2013;21:1434–1446.  57. Arntz CT, Veith RG, Hansen ST Jr. Fractures and fracture-dislocations of the tarsometatarsal joint. J Bone Joint Surg Am. 1988;70:173–181.  58. Leenen LPH, van der Werken C. Fracture-dislocation of the tarsometatarsal joint, a combined anatomical and computed tomographic study. Injury. 1992;23(1):51–55.  59. Peicha G, Preidler KW, Seibert FJ, et al. Acute hyperflexion trauma of the foot: a clinical comparison between conventional radiography, CT, and MRI. Suom Ortoped Traumatol. 1999;22:19–26.  60. Preidler KW, Yang Y, Brossom J, et al. Tarsometatarsal joint: anatomic details on MR images. Radiology. 1996;199(3):733–736.  61. Preidler KW, Brossom J, Daenen B, et al. MR imaging of the tarsometatarsal joint: analysis of injuries in eleven patients. AJR Am J Roentgenol. 1996;167:1217–1222.  62. Quenu E, Kuss G. Etude sur les luxations du metatarse (luxations metatarsotarsiennes) du diastasis entre le 1er et le 2e metatarsien. Rev Chir. 1909;39:281–336, 720–91, 1093–134.  63. Myerson MS, Fisher RT, Burgess AR, et al. Fracture dislocations of the tarsometatarsal joints: end results correlated with pathology and treatment. Foot Ankle. 1986;6(5):225–242.  64. Guler F, Baz AB, Turan A, et al. Isolated medial cuneiform fractures: report of two cases and review of the literature. Foot Ankle Spec. 2011;4(5):306–309.  65. Olson RC, Mendicino SS, Rockett MS. Isolated medial cuneiform

fracture: review of the literature and report of two cases. Foot Ankle Int. 2000;21(2):150–153.  66. Santi M, Sartoris DJ. Diagnostic imaging approach to stress fractures of the foot. J Foot Surg. 1991;30(1):85–97.  67. Sartoris DJ, Resnick D. Computed tomography of podiatric disorders: a review. J Foot Surg. 1986;25(5):394–403.  68. Elias I, Dheer S, Zoga AC, et al. Magnetic resonance imaging findings in bipartite medial cuneiform—a potential pitfall in diagnosis of midfoot injuries: a case series. J Med Case Rep. 2008;2:272. doi:10.1186/1752-19472-272  69. Grivas TB, Vasiliadis ED, Koufopoulos G, et al. Midfoot fractures. Clin Podiatr Med Surg. 2006;23:323–341.  70. Sangeorzan BJ, Mayo KA, Hansen ST. Intraarticular fractures of the foot: talus and lesser tarsals. Clin Orthop Relat Res. 1993;292:135–141.  71. Hunter JC, Sangeorzan BJ. A nutcracker fracture: cuboid fracture with an associated avulsion fracture of the tarsal navicular. Am J Roentengol. 1996;166:888.  72. Wilson E, Katz F. Stress fractures: an analysis of 250 consecutive cases. Radiology. 1969;92:481–486.  73. Salzler MJ, Blumen EM, Noonan S, et al. Injuries involved in minimalist runners. Foot Ankle Int. 2012;33(4):262–267.  74. Brukner P, Bennell K. Stress fractures in female athletes. Sports Med. 1997;24(6):419–429.  75. Shindle MK, Endo Y, Warren RF, et al. Stress fractures about the tibia, foot, and ankle. J Am Acad Orthop Surg. 2012;20(3):167–176.  76. Zwas ST, Elkanovitch R, Frank G. Interpretation and classification of bone scintigraphic findings in stress fractures. J Nucl Med. 1987;28:452–457.

 77. Taunton JE, Clement DB, Webber D. Lower extremity stress fracture in athletes. Phys Sports Med. 1981;9:77–83.  78. Rufuani HD, Holder HE, Espinola ED, et al. Three-phase radionuclide bone imaging in sports medicine. Radiology. 1985;156:187–196.  79. Terrell PN, Davies AM. Magnetic resonance imaging of fatigue fractures of long bones of the lower limb. Br J Radiol. 1994;67(796):332– 338.  80. Lee S, Anderson RB. Stress fractures of the tarsal navicular. Foot Ankle Clin North Am. 2004;9:85–104.  81. Mann JA, Pedowitz DL. Evaluation and treatment of navicular stress fractures, including nonunions, revision surgery, and persistent pain after treatment. Foot Ankle Clin North Am. 2009;14:187–204.  82. Burne SG, Mahoney CM, Forster BB, et al. Tarsal navicular stress injury: long-term outcome and clinicoradiological correlation using both computed tomography and magnetic resonance imaging. Am J Sports Med. 2005;33:1875–1881.  83. Saxena A, Fullem B, Hannaford D. Results of treatment of 22 navicular stress fractures and a new proposed radiographic classification system. J Foot Ankle Surg. 2000;39(2):96–103.  84. Torg J. Letters to the editor: author’s response. Am J Sports Med. 2010;38(10):NP3–NP5.  85. Kiss Z, Kahn K, Fuller P. Stress fractures of the tarsal navicular bone: CT findings in 55 cases. Am J Radiol. 1992;160:111–115.  86. Main BJ, Jowett RL. Injuries of the mid-tarsal joint. J Bone Joint Surg Br. 1975;57(1):89–97.  87. Makwana NK, van Liefland MR. Injuries of the midfoot. Curr Orthop. 2005;19:231–242.  88. Cave EF. Fracture of the os calcis—the problem in general. Clin Orthop

Relat Res. 1963;30:64–66.  89. Sizensky JA, Marks RM. Imaging of the navicular. Foot Ankle Clin North Am. 2004;9:181–209.  90. Davis CA, Lubowitz J, Thordarson DB. Midtarsal fracture-subluxation. Clin Orthop Relat Res. 1993;292:264–268.  91. Eichenholtz SN, Levine DB. Fractures of the tarsal navicular bone. Clin Orthop Relat Res. 1964;34:142–157.  92. Howie CR, Hooper G, Hughes SPF. Occult midtarsal subluxation. Clin Orthop Relat Res. 1986;209:206–209.  93. Chen YJ, Shih HN, Huang, TJ, et al. Posterior tibial tendon tear combined with a fracture of the accessory navicular: a new subclassification? J Trauma. 1995;39(5):993–996.  94. Dodsworth K, Hutchison PW, Gudeon A. Avulsion fracture of the navicular tuberosity. J Am Podiatr Med Assoc. 1990;80(5):274–277.  95. Geist ES. The accessory scaphoid bone. J Bone Joint Surg Am. 1925;7:570–574.  96. Mygind HB. The accessory tarsal scaphoid. Acta Orthop Scand. 1953;23:142–151.  97. Chiu NT, Jou IM, Lee BF, et al. Symptomatic and asymptomatic accessory navicular bones: findings of Tc-99m MDP bone scintigraphy. Clin Radiol. 2000;55:353–355.  98. Romanowski CAJ, Barrington NA. The accessory navicular—an important cause of medial foot pain. Clin Radiol. 1992;46:261–264.  99. Shah S, Achong DM. The painful accessory navicular bone. Scintigraphic and radiographic correlation. Clin Nucl Med. 1999;24:​125–142. 100. Miller TT, Staron RB, Feldman F, et al. The symptomatic accessory tarsal navicular bone: assessment with MR imaging. Radiology.

1995;195:849–853. 101. Sangeorzan BJ, Benirschke SK, Mosca V, et al. Displaced intraarticular fractures of the tarsal navicular. J Bone Joint Surg Am. 1989;71(10):1504–1510. 102. Rockett MS, Brage ME. Navicular body fractures: computerized tomography findings and mechanism of injury. J Foot Ankle Surg. 1997;36(3):185–191. 103. Nadeau P, Templeton J. Vertical fracture dislocation of the tarsal navicular. J Trauma. 1976;16(8):669–671. 104. DiGiovanni CW. Fractures of the navicular. Foot Ankle Clin North Am. 2004;9:25–63. 105. Haapamaki V, Kiuru MJ, Koskinen SK. Ankle and foot Injuries: analysis of MDCT findings. AJR Am J Roentgenol. 2004;183:615–622. 106. Adelaar RS. The treatment of complex fractures of the talus. Orthop Clin North Am. 1989;20(4):691–707. 107. Dale JD, Ha AS, Chew FS. Update on talar fracture patterns: a large Level I trauma center study. AJR Am J Roentgenol. 2013;201:1087–1092. 108. Santavirta S, Seitsalo S, Kiviluoto O, et al. Fractures of the talus. J Trauma. 1984;24(11):986–989. 109. Raikin SM. Stage VI: massive osteochondral defects of the talus. Foot Ankle Clin North Am. 2004;9:737–744. 110. Pennal GE. Fractures of the talus. Clin Orthop. 1963;30:53–63. 111. Hawkins LG. Fracture of the lateral process of the talus. J Bone Joint Surg Am. 1965;47:1170–1175. 112. Hawkins LG. Fractures of the neck of the talus. J Bone Joint Surg Am. 1970. 52(5):991–1002.

113. Coltart WD. Aviator’s astragalus. J Bone Joint Surg Br. 1952;34(4):545–566. 114. Elgafy H, Ebraheim NA, Tile M, et al. Fractures of the talus: experience of two level 1 trauma centers. Foot Ankle Int. 2000;21:1023–1029. 115. Canale ST. Fractures of the neck of the talus. Orthopedics. 1990;13:1105–1115. 116. Canale ST, Kelly FB Jr. Fractures of the neck of the talus: long-term evaluation of seventy-one cases. J Bone Joint Surg Am. 1978;60(2):143–156. 117. Inokuchi S, Ogawa K, Usami N. Classification of fractures of the talus: clear differentiation between neck and body fractures. Foot Ankle Int. 1996;17(12):748–750. 118. Lindvall E, Haidukewych G, DiPasquale T, et al. Open reduction and stable fixation of isolated, displaced talar neck and body fractures. J Bone Joint Surg Am. 2004;86:2229–2234. 119. Kapoor SK, Patra SR. Talar neck fracture associated with talonavicular dislocation and fracture of posteromedial process of talus. J Foot Ankle Surg. 2011;50(6):758–761. 120. Veerappa LA, Gopalkrishna C. Talar neck fracture with talar head dislocation and intact ankle and subtalar joints—a rare case report. Foot (Edinb). 2010;20(1):39–41. 121. Vallier H, Nork S, Benirschke S, et al. Treatment of talar body fractures. J Bone Joint Surg Am. 2003;85:1716–1724. 122. Metzger MJ, Levin JS, Clancy JT. Talar neck fractures and rates of avascular necrosis. J Foot Ankle Surg. 1999;38(2):154–162. 123. Fortin P, Balazsy J. Talus fractures: evaluation and treatment. J Am Assoc Orthop Surg. 2001;9:114–127. 124. Miller AN, Prasarn ML, Dyke JP, et al. Quantitative assessment of the vascularity of the talus with gadolinium-enhanced magnetic resonance

imaging. J Bone Joint Surg Am. 2011;93(12):1116–1121. 125. Mulfinger GL, Trueta J. The blood supply of the talus. J Bone Joint Surg Br. 1970;52:160–167. 126. Gelberman RH, Mortensen WW. The arterial anatomy of the talus. Foot Ankle. 1983;4(2):64–72. 127. Vallier H, Nork S, Barel D, et al. Talar neck fractures: results and outcomes. J Bone Joint Surg Am. 2004;86:1616–1624. 128. Adelaar RS, Marion JR. Avascular necrosis of the talus. Orthop Clin North Am. 2004;35:383–395. 129. Berlet GC, Lee TH, Massa EG. Talar neck fractures. Orthop Clin North Am. 2001;32(1):53–64. 130. Kenwright J, Taylor RG. Major injuries of the talus. J Bone Joint Surg Br. 1970;52:36–48. 131. Tezval M, Dumont C, Sturmer KM. Prognostic reliability of the Hawkins sign in fractures of the talus. J Orthop Trauma. 2007;21(8): 538– 543. 132. Rodriguez-Paz S, Munoz-Vives JM, Froufe-Siota MA. Is the Hawkins sign able to predict necrosis in fractures of the neck of the astragalus? [in Spanish]. Rev Esp Cir Ortop Traumatol. 2013;57(6):403–408. 133. Berndt A, Harty M. Transchondral fractures (osteochondritis dissecans) of the talus. J Bone Joint Surg Am. 1959;41(6):988–1020. 134. Ferkel R, Scaglione N, Del Pizzo W. Arthroscopic treatment of osteochondral lesion of the talus: technique and results. Orthop Trans. 1990;14:172–176. 135. Hepple S, Winston I, Glew D. Osteochondral lesions of the talus: a revised classification. Foot Ankle Int. 1999;20(12):789–793. 136. Taranow WS, Bisignani GA, Towers JD, et al. Retrograde drilling of

osteochondral lesions of the medial talar dome. Foot Ankle Int. 1999;20:474– 470. 137. Anderson IF, Crichton KJ, Grattan-Smith T, et al. Osteochondral fractures of the dome of the talus. J Bone Joint Surg Am. 1989;71(8):1143– 1152. 138. Griffith JF, Lau DTY, Yeung DKW, et al. High-resolution MR imaging of talar osteochondral lesions with new classification. Skeletal Radiol. 2012;31:387–399. 139. Scranton PE Jr, Frey CC, Feder KS. Outcome of osteochondral autograft transplanatation for type-V cystic osteochondral lesions of the talus. J Bone Joint Surg Br. 2006;88(5):614–619. 140. Pritsch M, Horoshovski H, Farine I. Arthroscopic treatment of osteochondral lesions of the talus. J Bone Joint Surg Am. 1986;68:862–865. 141. Loomer R, Fisher C, Lloyd-Smith R, et al. Osteochondral lesions of the talus. Am J Sports Med. 1993;21(1):13–19. 142. Flick AB, Gould N. Osteochondritis dissecans of the talus (transchondral fractures of the talus): review of the literature and new surgical approach for medial dome lesions. Foot Ankle. 1985;5(4):165–185. 143. Canale ST, Belding RH. Osteochondral lesions of the talus. J Bone Joint Surg Am. 1980;62(1):97–102. 144. Roach R. Osteochondral lesions of the talus. J Am Podiatr Med Assoc. 2003;93(4):307–311. 145. Verhagen RAW, Maas M, Dijkgraaf MGW, et al. Prospective study on diagnostic strategies in osteochondral lesions of the talus. J Bone Joint Surg Br. 2005;87:41–46. 146. Lantz BA, McAndrew M, Scioli M, et al. The effect of concomitant chondral injuries accompanying operatively reduced malleolar fractures. J Orthop Trauma. 1991;5(2):125–128.

147. Aktas S, Kocaoglu B, Gereli A, et al. Incidence of chondral lesions of talar dome in ankle fracture types. Foot Ankle Int. 2008;29(3):287–292. 148. Sorrento DL, Mlodzienski A. Incidence of lateral talar dome lesions in SER IV ankle fractures. J Foot Ankle Surg. 2000;39(6):354–358. 149. Takao M, Ochi M, Uchio Y, et al. Osteochondral lesions of the talar dome associated with trauma. Arthroscopy. 2003;19(10):1061–1067. 150. Sneppen O, Buhl O. Fracture of the talus: a study of its genesis and morphology based upon cases with associated ankle fracture. Acta Orthop Scand. 1974;45:307–320. 151. Sneppen O, Christensen B, Krogsoe O, et al. Fracture of the body of the talus. Acta Orthop Scand. 1977;48(3):317–324. 152. Valderrabano V, Perren T, Ryf C, et al. Snowboarder’s talus fracture: treatment outcome of 20 cases after 3.5 years. Am J Sports Med. 2005;33:871–880. 153. McCrory P, Bladin C. Fractures of the lateral process of the talus: a clinical review. “Snowboarder’s ankle.” Clin J Sport Med. 1996;6(2):124– 128. 154. Mukherjee SK, Pringle RM, Baxter AD. Fracture of the lateral process of the talus: a report of thirteen cases. J Bone Joint Surg Br. 1974;56(2):263– 273. 155. Heckman J, MacLean M. Fractures of the lateral process of the talus. Clin Orthop 1985;199:108–113. 156. Schofield RO. Fractures of the os calcis. J Bone Joint Surg. 1936; 18(3):556–580. 157. Slatis P, Kiviluoto O, Santavirta S, et al. Fractures of the calcaneum. J Trauma. 1979;19(12):939–943. 158. Lance EM, Carey EJ, Wade PA. Fractures of the os calcis: a follow-up study. J Trauma 1964;4:15–56.

159. Maskill JD, Bohay DR, Anderson JF. Calcaneus fractures: a review article. Foot Ankle Clin North Am. 2005;10:463–489. 160. Lindsay WRN, Dewar FP. Fractures of the os calcis. Am J Surg. 1958;95:555–576. 161. Sanders R. Displaced intra-articular fractures of the calcaneus. J Bone Joint Surg Am 2000; 82: 225–250. 162. Sanders R. Intra-articular fractures of the calcaneus: present state of the art. J Orthop Trauma. 1992;6:252–265. 163. Sanders R, Fortin P, DiPasquale A, et al. Operative treatment in 120 displaced intra-articular calcaneal fractures: results using a prognostic computed tomographic scan classification. Clin Orthop. 1993;290:87–95. 164. Palmer I. The mechanism and treatment of fractures of the calcaneus. J Bone Joint Surg Am. 1948;30:2–8. 165. Essex-Lopresti P. Fractures of the os calcis: the mechanism, reduction technique, and results in fractures of the os calcis. Br J Surg. 1952;39:395– 419. 166. Soeur R, Remy R. Fractures of the calcaneus with displacement of the thalamic portion. J Bone Joint Surg Br. 1975;57:413–421. 167. Zwipp H, Tscherne H, Thermann H, et al. Osteosynthesis of displaced intra-articular fractures of the calcaneus: results in 123 cases. Clin Orthop. 1993;290:76–86. 168. Rubino R, Valderrabano V, Sutter PM, et al. Prognostic value of four classifications of calcaneal fractures. Foot Ankle Int. 2009;30:229–238. 169. Furey A, Stone C, Squire D, et al. Os calcis fractures: analysis of interobserver variability in using Sanders classification. J Foot Ankle Surg. 2003;42(1):21–23. 170. Csizy M, Buckley R, Tough S, et al. Displaced intra-articular calcaneal

fractures: variables predicting late subtalar fusion. J Orthop Trauma. 2003;17(2):106–112. 171. Sayed-Noor AS, Agren PH, Wretenberg P. Interobserver reliability and intraobserver reproducibility of three radiological classification systems for intra-articular calcaneal fractures. Foot Ankle Int. 2011;32(9):861–866. 172. Schepers T, van Lieshout EMM, Ginai AZ, et al. Calcaneal fracture classification: a comparative study. J Foot Ankle Surg. 2009;48(2):156–162. 173. Marsh J, Saltzman C, Iverson M, et al. Major open trauma of the talus. J Orthop Trauma. 1995;9:371–376. 174. Patel R, VanBergeyk A, Pinney S. Are displaced talar neck fractures surgical emergencies? A survey of orthopaedic trauma experts. Foot Ankle Int. 2005;26:378–381. 175. Lauder AJ, Inda DJ, Bott AM, et al. Interobserver and intraobserver reliability of two classification systems for intra-articular calcaneal fractures. Foot Ankle Int. 2006;27(4):251–255. 176. Koval KJ, Sanders R. The radiologic evaluation of calcaneal fractures. Clin Orthop Relat Res. 1993;290:41–46. 177. Gissane W. News notes: proceedings of the British Orthopedic Association. J Bone Joint Surg. 1947;29:254–255. 178. Bohler L. Diagnosis, pathology, and treatment of fractures of the os calcis. J Bone Joint Surg. 1931;13:75–89. 179. Loucks C, Buckley R. Bohler’s angle: correlation with outcome in displaced intra-articular calcaneal fractures. J Orthop Trauma. 1999;13(8): 554–558. 180. Broden B. Roentgen examination of the subtaloid joint in fractures of the calcaneus. Acta Radiol. 1949;31:85–91. 181. Lowrie IG, Finlay DB, Brenkel IJ, et al. Computerised tomographic assessment of the subtalar joint in calcaneal fractures. J Bone Joint Surg Br.

1988;70(2):247–250. 182. Germann CA, Perron AD, Miller MD, et al. Orthopedic pitfalls in the ED: calcaneal fractures. Am J Emerg Med. 2004;22(7):607–611. 183. Rowe CR, Sakellarides HT, Freeman PA, et al. Fractures of os calcis: a long term follow-up study of one hundred forty-six patients. JAMA. 1963;184:920–923. 184. Trnka HJ, Zettl R, Ritschl P. Fracture of the anterior superior process of the calcaneus: an often misdiagnosed fracture. Arch Orthop Trauma Surg. 1998;117(4–5):300–302. 185. Degan TJ, Morrey BF, Braun DP. Surgical excision for anterior-process fractures of the calcaneus. J Bone Joint Surg Am. 1982;64(4):519–524. 186. Hunt DD. Compression fracture of the anterior articular surface of the calcaneus. J Bone Joint Surg Am. 1970;52(8):1637–1642. 187. Lowy M. Avulsion fractures of the calcaneus. J Bone Joint Surg Br. 1969;51(3):494–497. 188. Protheroe K. Avulsion fractures of the calcaneus. J Bone Joint Surg Br. 1969;51(1):118–122. 189. Squires B, Allen PE, Livingstone J, et al. Fractures of the tuberosity of the calcaneus. J Bone Joint Surg Br. 2001;83(1):55–61. 190. Wren TA, Yerby SA, Beaupre GS, et al. Influence of bone mineral density, age, and strain rate on the failure mode of human Achilles tendons. Clin Biomech. 2001;16(6):529–534. 191. Beavis RC, Rourke K, Court-Brown C. Avulsion fracture of the calcaneal tuberosity: a case report and literature review. Foot Ankle Int. 2008;29(8):863–866. 192. Dürr C, Zwipp H, Rammelt S. Fracture of the sustentaculum tali. Oper Orthop Traumatol. 2013;25:569–578.

193. Rammelt S, Zwipp H. Calcaneus fractures: facts, controversies and recent developments. Injury. 2004;35:443–461. 194. Gatha M, Pedersen B, Buckley R. Fractures of the sustentaculum tali of the calcaneus: a case report. Foot Ankle Int. 2008;29(2):237–240. 195. Lance EM, Carey EJ Jr, Wade PA. Fractures of the os calcis: treatment by early mobilization. Clin Orthop Relat Res. 1963;30:76–90. 196. Rocca GJD, Nork SE, Carei DP, et al. Fractures of the sustentaculum tali: injury characteristics and surgical technique for reduction. Foot Ankle Int. 2009;30(11):1037–1041. 197. Mandi DM. Ankle fractures. Clin Podiatr Med Surg. 2012;29(2):​155– 186. 198. Clare MP. A rational approach to ankle fractures. Foot Ankle Clin. 2008;13(4):593–610. 199. Court-Brown CM, Caesar B. Epidemiology of adult fractures: a review. Injury. 2006;37(8):691–697. 200. Lin CW, Moseley AM, Refshauge KM. Effects of rehabilitation after ankle fracture: a cochrane systematic review. Eur J Phys Rehabil Med. 2009;45(3):431–441. 201. Haraguchi N, Armiger RS. A new interpretation of the mechanism of ankle fracture. J Bone Joint Surg Am. 2009;91(4):821–829. 202. Davis AW, Alexander IJ. Problematic fractures and dislocations in the foot and ankle of athletes. Clin Sports Med. 1990;9(1):163–181. 203. Gehrmann RM, Rajan S, Patel DV, et al. Athletes’ ankle injuries: diagnosis and management. Am J Orthop. 2005;34(11):551–561. 204. Scott AM. Diagnosis and treatment of ankle fractures. Radiol Technol. 2010;81(5):457–475. 205. Okanobo H, Khurana B, Sheehan S, et al. Simplified diagnostic

algorithm for Lauge-Hansen classification of ankle injuries. Radiographics. 2012;32(2):E71–E84. 206. Pott P. Some Few General Remarks on Fractures and Dislocations. London, England: Hawes, Clarke, Collins; 1768. 207. Bugler KE, Watson CD, Hardie AR, et al. The treatment of unstable fractures of the ankle using the Acumed fibular nail: development of a technique. J Bone Joint Surg Br. 2012;94(8):1107–1112. 208. Ashhurst APC, Bromer RS. Classification and mechanism of fractures of the leg bones involving the ankle: based on a study of three hundred cases from Episcopal Hospital. Arch Surg. 1922;4:51–129. 209. Lauge-Hansen N. Fractures of the ankle. II: combined experimentalsurgical and experimental-roentgenologic investigations. Arch Surg. 1950;60(5):957–985. 210. Perlman MD, Leveille D, Gale B. Traumatic classifications of the foot and ankle. J Foot Surg. 1989;28(6):551–585. 211. Yde J. The Lauge Hansen classification of malleolar fractures. Acta Orthop Scand. 1980;51(1):181–192. 212. Budny AM, Young BA. Analysis of radiographic classifications for rotational ankle fractures. Clin Podiatr Med Surg. 2008;25(2):139–152. 213. Lindsjo U. Classification of ankle fractures: the Lauge-Hansen or AO system? Clin Orthop Relat Res. 1985;(199):12–16. 214. Arimoto HK, Forrester DM. Classification of ankle fractures: an algorithm. AJR Am J Roentgenol. 1980;135(5):1057–1063. 215. Koulouris G, Morrison WB. Foot and ankle disorders: radiographic signs. Semin Roentgenol. 2005;40(4):358–379. 216. Russo A, Reginelli A, Zappia M, et al. Ankle fracture: radiographic approach according to the Lauge-Hansen classification. Musculoskelet Surg. 2013;97(suppl 2):S155–S160.

217. Perlman M, Leveille D, DeLeonibus J, et al. Inversion lateral ankle trauma: differential diagnosis, review of the literature, and prospective study. J Foot Surg. 1987;26(2):95–135. 218. Schuberth JM, Collman DR, Rush SM, et al. Deltoid ligament integrity in lateral malleolar fractures: a comparative analysis of arthroscopic and radiographic assessments. J Foot Ankle Surg. 2004;43(1):20–29. 219. Schock HJ, Pinzur M, Manion L, et al. The use of gravity or manualstress radiographs in the assessment of supination-external rotation fractures of the ankle. J Bone Joint Surg Br. 2007;89(8):1055–1059. 220. McDaniel WJ, Wilson FC. Trimalleolar fractures of the ankle—an end result study. Clin Orthop Relat Res. 1977;122:37–45. 221. Danis, R. Les fractures malleolaires [in French]. In: Danis, R. ed. Theorie et pratique de l’osteosynthese. Paris, France: Masson et Cie;1949:133–165. 222. Mandracchia VJ, Evans RD, Nelson SC, et al. Pilon fractures of the distal tibia. Clin Podiatr Med Surg. 1999;16(4):743–767. 223. Alexandropoulos C, Tsourvakas, S, Papachristos J, et al. Ankle fracture classification: an evaluation of three classification systems: Lauge-Hansen, A.O. and Broos-Bisschop. Acta Orthop Belg. 2010;76(4):521–525. 224. Nielsen JO, Dons-Jensen H, Sørensen HT, et al. Lauge-Hansen classification of malleolar fractures: an assessment of the reproducibility in 118 cases. Acta Orthop Scand. 1990;61(5):385–387. 225. Rodriguez EK, Kwon JY, Herder LM, et al. Correlation of AO and Lauge-Hansen classification systems for ankle fractures to the mechanism of injury. Foot Ankle Int. 2013;34(11):1516–1520. 226. Gardner MJ, Demetrakopoulos D, Brigss SM, et al. The ability of the Lauge-Hansen classification to predict ligament injury and mechanism in ankle fractures: an MRI study. J Orthop Trauma. 2006;20(4):267–272.

227. Destot E. Traumatismes du Pied et Rayons X: Malleoles, Astragale, Calcaneum, Avant Pied. Paris, France: Masson; 1911:1–10. 228. Ruedi T. Fractures of the lower end of the tibia into the ankle joint: results 9 years after open reduction and internal fixation. Injury. 1973;5(2):130–134. 229. Ruedi T, Matter P, Allgöwer M. Intra-articular fractures of the distal tibial end. Helv Chir Acta. 1968;35(5):556–582. 230. Muller FJ, Nerlich M. Tibial pilon fractures. Acta Chir Orthop Traumatol Cech. 2010;77(4):266–276. 231. Bonin JG. Injuries to the Ankle. London, England: William Heinemann; 1950:248–260. 232. Bohler L. Die Technik der Knochenbruchbehandlung. 13th ed. Vienna, Austria: Maudrich; 1951. 233. Chen SH, Wu PH, Lee YS. Long-term results of pilon fractures. Arch Orthop Trauma Surg. 2007;127(1):55–60. 234. Reudi T, Allgower M. The operative treatment of intra-articular fractures of the lower end of the tibia. Clin Orthop. 1979;138:105–110. 235. Orthopedic Trauma Association. Fracture and dislocation compendium. J Orthop Trauma 1996;10(Suppl1)1–55. 236. Topliss CJ, Jackson M, Atkins RM. Anatomy of pilon fractures of the distal tibia. J Bone Joint Surg Br. 2005;87(5):692–697. 237. Swiontkowski MF, Sands AK, Agel J, et al. Interobserver variation in the AO/OTA fracture classification system for pilon fractures: is there a problem? J Orthop Trauma. 1197;11(7):467–470. 238. Dirschl DR, Adams GL. A critical assessment of methods to improve reliability in the classification of fractures, using fractures of the tibial plafond as a model. J Orthop Trauma. 1997;11:471–476.

239. Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analyses. J Bone Joint Surg Am. 1976;58:​453–458. 240. Gustilo JT, Mendoza RM, Williams DN. Problems in the management of type III (severe) open fractures: a new classification of type III open fractures. J Trauma. 1984;24(8):742–746. 241. Luk PC, Charlton TP, Lee J, et al. Ipsilateral Intact Fibula as a predictor of tibial plafond fracture pattern and severity. Foot Ankle Int. 2013;34(10):1421–1426. 242. Ramappa M, Bajwa A, Singh A, et al. Interobserver and intraobserver variations in tibial pilon fracture classification systems. Foot (Edinb). 2010;20(2–3):61–63. 243. Tornetta P III, Gorup J. Axial computed tomography of pilon fractures. Clin Orthop Relat Res. 1996;(323):273–276. 244. Carothers CO, Crenshaw AH. Clinical significance of a classification of epiphyseal injuries at the ankle. Am J Surg. 1955;89:879–889. 245. Salter RB, Harris WR. Injuries involving the epiphyseal plate. J Bone Joint Surg Am. 1963;45:587–622. 246. Wuerz TH, Gurd DP. Pediatric physeal ankle fracture. J Am Acad Orthop Surg. 2013;21:234–244. 247. Spiegel P, Cooperman D, Laros G. Epipyseal fractures of the distal ends of the tibia and fibula. J Bone Joint Surg Am. 1978;60:1046–1050. 248. Aitken AP. Fractures of the epiphyses. Clin Orthop Relat Res. 1965;41:19–23. 249. King J, Diefendorf D, Apthorp J, et al. Analysis of 429 fractures in 189 battered children. J Pediatr Orthop. 1988;8(5):585–589. 250. Peterson HA, Madhok R, Benson JT, et al. Physeal fractures, part 1. Epidemiology in Olmsted County, Minnesota, 1979–1988. J Pediatr Orthop.

1994;14(4):423–430. 251. Mann DC, Rajmaira S. Distribution of physeal and nonphyseal fractures in 2,650 long-bone fractures in children aged 0–16 years. J Pediatr Orthop. 1990;10(6):713–716. 252. Bishop PA. Fractures and epiphyseal separation fractures of the ankle. AJR Am J Roentgenol. 1932;28:49. 253. Johnson EW Jr, Fahl JC. Fractures involving the distal epiphysis of the tibia and fibula in children. Am J Surg. 1957;93(5):778–781. 254. El-Karef E, Sadek HI, Nairn DS, et al. Triplane fracture of the distal tibia. Injury. 2000;31(9):729–736. 255. Zaricznyj B, Shattuck LJ, Mast TA, et al. Sports-related injuries in school age children. Am J Sports Med. 1980;8:318–24. 256. Marmor L. An unusual fracture of the tibial epiphysis. Clin Orthop Relat Res. 1970;73:132–135. 257. Cooperman DR, Spiegel PG, Laros GS. Tibial fractures involving the ankle in children: the so-called triplane epiphyseal fracture. J Bone Joint Surg Am. 1978;60:1040–1046. 258. Dias LS, Tachdjian MO. Physeal injuries of the ankle in children: classification. Clin Orthop Relat Res. 1978;136:230–233. 259. Crenshaw AH. Injuries of the distal tibial epiphysis. Clin Orthop. 1965;41:98–107 260. Peterson HA. Epiphyseal Growth Plate Fractures. Berlin, Germany: Springer; 2007. 261. Polyzois VD, Vasiliadis E, Zgonis T, et al. Pediatric fractures of the foot and ankle. Clin Podiatr Med Surg. 2006;23:241–255. 262. Thawrani D, Kuester V, Gabos PG, et al. Reliability and necessity of computerized tomography in distal tibial physeal injuries. J Pediatr Orthop.

2011;31(7):745–750. 263. Dowling S, Spooner CH, Liang Y, et al. Accuracy of Ottawa Ankle Rules to exclude fractures of the ankle and midfoot in children: a metaanalysis. Acad Emerg Med. 2009;16(4):277–287. 264. Plint AC, Bulloch B, Osmon MH, et al. Validation of the Ottawa Ankle Rules in children with ankle injuries. Acad Emerg Med. 1999;6(10):1005– 1009. SUGGESTED READINGS Bhattacharya R, Vassan UT, Finn P, et al. Sanders classification of fractures of the os calcis: an analysis of inter- and intra-observer variability. J Bone Joint Surg Br. 2005;87(2):205–208. Biehl WC III, Morgan JM, Wagner FW Jr, et al. Neuropathic calcaneal tuberosity avulsion fractures. Clin Orthop Relat Res. 1993;296:8–13. Bremner AE, Warrick CK. Fractures of the calcaneus. J Fac Radiol. 1951;2(3):235–241. Copeland CL, Kanat IO. A new classification for traumatic dislocation of the first metatarsophalangeal joint: type IIC. J Foot Surg. 1991;30:234–237. de Souza LJ, Rutledge E. Grouping of intra-articular calcaneal fractures relative to treatment options. Clin Orthop. 2004;420:261–267. Faciszewski T, Burks RT, Manaster BJ. Subtle injuries of the Lisfranc joint. J Bone Joint Surg Am. 1990;72A:1519–1522. Gellman M. Fractures of the anterior process of the calcaneus. J Bone Joint Surg Am. 1951;33(2):382–386. Golano P, Farinas O, Saenz I. The anatomy of the navicular and periarticular structures. Foot Ankle Clin North Am. 2004;9:1–23. Haliburton RA, Sullivan CR, Kelly PJ, et al. The extra-osseous and intraosseous blood supply of the talus. J Bone Joint Surg Am. 1958;40(5):1115–

1120. Higgins TF, Baumgaertner MR. Diagnosis and treatment of fractures of the talus: a comprehensive review of the literature. Foot Ankle Int. 1999;20:595– 605. Humphrey CA, Dirschl DR, Ellis TJ. Interobserver reliability of a CT-based fracture classification system. J Orthop Trauma. 2005;19(9):616–622. Jahss MH. Disorders of the hallux and first ray. In: Disorders of the Foot and Ankle: Medical and Surgical Management. 2nd ed. Philadelphia, PA: Saunders, 1991:1124–1129:chap 39. Janzen DL, Connell DG, Munk PL, et al. Intraarticular fractures of the calcaneus: value of CT findings in determining prognosis. AJR Am J Roentgenol. 1992;158:1271–1274. Kalia V, Fishman EK, Carrino JA, et al. Epidemiology, imaging, and treatment of Lisfranc fracture-dislocation revisited. Skeletal Radiol. 2012;41:129–136. Labovitz JM, Schweitzer ME. Occult osseous Injuries after ankle sprains: incidence, location, pattern and age. Foot Ankle Int. 1998;19(10):661–667. Letournel E. Open reduction and internal fixation of calcaneal fractures. Clin Orthop. 1993;290:60–67. Lee SM, Huh SW, Chung JW, et al. Avulsion fracture of the calcaneal tuberosity: classification and its characteristics. Clin Orthop Surg. 2012;4(2):134–138. Marks RM, Antoniades S, Myerson MS. Injury to the sustentaculum tali. Foot. 1996;6:182–187. Mayich DJ, Mayich MS, Daniels TR. Effective detection and management of low-velocity Lisfranc injuries in the emergency setting. Can Fam Physician. 2012;58(11):1199–1204. McBryde AM. Stress fractures in athletes. J Sports Med. 1975;3:312–317.

Michelson J, Solocoff D, Waldman B, et al. Ankle fractures: the LaugeHansen classification revisited. Clin Orthop Relat Res. 1997;(345): 198–205. Robinson DE, Winson IG, Harries WJ, et al. Arthroscopic treatment of osteochondral lesions of the talus. J Bone Joint Surg Br. 2003;85:989–993. Szyszkowitz R, Reschauer R, Seggl W. Eight-five talus fractures treated by ORIF with five to eight years follow-up study of 69 patients. Clin Orthop. 1996;199:97–107. Thordarson D, Triffon MJ, Terk MR. Magnetic resonance imaging to detect avascular necrosis after open reduction and internal fixation of talar neck fractures. Foot Ankle Int. 1996;17:742–747. Timpone V, Tall M, Puckett A. Intermetatarsal fat pad sign: radiographic aid to diagnosis of occult tarsometatarsal joint injuries. AJR Am J Roentgenol. 2009;192(1):W36–W37. Torg JS, Pavlov H, Cooley LH, et al. Stress fractures of the tarsal navicular. J Bone Joint Surg Am. 1982;64(5):700–712. Ugolini PA, Raikin SM. The accessory navicular. Foot Ankle Clin North Am. 2004;9:165–180. Verhagen RAW, Maas M, Dijkgraaf MGW, et al. Prospective study on diagnostic strategies in osteochondral lesions of the talus. J Bone Joint Surg Br. 2005;87:41–46. Warrick CK, Bremner AE. Fractures of the calcaneum, with an atlas illustrating the various types of fracture. J Bone Joint Surg Br. 1953;35(1):33–45. Williams T, Barba N, Noailles T, et al. Total talar fracture—inter- and intraobserver reproducibility of two classification systems (Hawkins and AO) for central talar fractures. Orthop Trauma Surg Res. 2012;98:S56–S65.

17 Osteonecrosis and Osteochondrosis ROBERT A. CHRISTMAN, CRISTINA MARCHIS-CRISAN, AND RANDY E. COHEN Osteonecrosis and osteochondrosis are two radiographically similar disorders associated with multiple etiologies. Overlap between these disorders often confuses the novice. Osteonecrosis can affect either epiphyseal or nonepiphyseal bone and demonstrates distinct pathologic features that can occur in all age groups. Trauma is probably the most common etiology of osteonecrosis in the foot. Osteochondrosis, however, represents a group of disorders with similar radiographic features affecting epiphyses and apophyses only; they have been associated with three underlying etiologies: osteonecrosis, trauma without underlying osteonecrosis, and variant ossification.1 Osteochondral injury is often discussed in relation to osteonecrosis and osteochondrosis; the latter has been considered a juvenile form of osteochondral injury.2 OSTEONECROSIS Osteonecrosis, or bone death, results from bone ischemia. Several terms have been used synonymously with osteonecrosis: ischemic necrosis, avascular necrosis, aseptic necrosis, and bone infarct. Osteonecrosis is the death of bone tissue due to a lack of blood supply.3 This may be due to interruption of the arterial supply or occlusion of the venous drainage, resulting in stasis and oxygen deprivation.4 Osteonecrosis is defined as massive death of bone and bone marrow, occurring as the most predominant or only abnormality.5 Causes may be local, related to pathology causing major stress to bone, such as infection, fracture, tumor, and severe osteoarthritis. It has also been reported postoperatively, for example, at the first metatarsal head following osteotomy for bunion correction.6 Systemic causes include alcohol abuse,

corticosteroids, pancreatitis, systemic lupus erythematosus, caisson disease, Gaucher disease, gout, sickle cell disease, and hyperlipidemia; about 25% of the cases are iatrogenic.5,7–9 Etiologic theories that involve alterations of lipid metabolism include fat emboli causing vascular occlusion, sludging and hemorrhage, and a systemic buildup of lipid in the marrow that compromises vascular space. The elevated cholesterol content in the necrotic tissues may contribute to cell death by altering membrane metabolism.10 Bone death is followed by attempts at repair, which include revascularization, resorption of dead bone, and reossification. All osteonecroses undergo this sequence of events.11 Since dead bone has no vascular supply, its density does not change radiographically. It is the healing process that causes the subsequent radiographic changes.12 Initially, before the healing process has begun, the necrotic bone appears the same density as the viable surrounding bone. No appreciable changes have yet occurred radiographically. If hyperemia is present, the surrounding viable bone becomes osteopenic. The necrotic bone does not change density. However, the necrotic segment relatively appears more radiodense than the osteopenic bone surrounding it. Healing varies radiographically depending on the anatomic site and vascular status of the area. In general, as revascularization and resorption occur around the area of necrosis, a radiolucent rim appears. Fibroblastic ingrowth within the necrotic bone may appear as irregular radiolucencies, which can give the radiographic appearance of fragmentation. New bone being laid down on necrotic trabeculae causes an absolute increased bone density and appears more radiodense than normal bone. If the dead trabeculae are reabsorbed, the bone regains its normal density.4,11–13 Osteonecrosis occurs most commonly within the epiphysis during skeletal development and the metaphyseal/diaphyseal marrow cavity of an adult long tubular bone. The femoral and humeral heads are common epiphyseal sites, but the lesser metatarsal heads are also affected.7 Osteonecrosis also involves irregular bones such as the talus and calcaneus, which varies in etiology, size of area affected, and age predilection.

Osteonecrosis can affect any bone in the foot or ankle, especially in chronically ill children on long-term steroid therapy. Common sites affected include the talar dome, navicular, and second metatarsal distal epiphysis. MRI (magnetic resonance imaging) is often used to identify and characterize osteonecrosis in its early stages when radiographs are normal and to assess progression over time.14 Radiographically, diametaphyseal osteonecrosis (commonly referred to as bone infarct) differs significantly from epiphyseal osteonecrosis. Diametaphyseal infarct shows nothing radiographically for weeks or months. As necrotic bone is reabsorbed and repair occurs, an area of serpiginous calcification and ossification surrounds a lucent region (Figure 17-1). The entire area often calcifies (Figure 17-2).11 Epiphyseal osteonecrosis (commonly referred to as ischemic or avascular necrosis), in contrast, affects the weight-bearing capacity of the joint and results in concomitant changes of the articulating cartilage and underlying subchondral bone. The radiographic appearance of epiphyseal osteonecrosis varies depending on its stage of repair. Initially radiographs are normal. Early findings include osteopenia mixed with ill-defined sclerosis (Figure 17-3). A radiolucent fracture line that parallels the subchondral bone plate, referred to as the crescent line (Figure 17-4), follows. The weakened subchondral bone collapses, with subsequent deformity of the epiphysis (Figure 17-5). Osteoarthritis develops secondarily (Figure 17-6).

FIGURE 17-1. Bone infarct/diametaphyseal osteonecrosis, distal tibia. Classic appearance, a large, serpiginous calcification surrounding an area of relative lucency. A: Mortise view. B: Lateral view.

FIGURE 17-2. Bone infarct/diametaphyseal osteonecrosis, distal tibia. Nearly solid calcification. A: Anteroposterior view. B: Lateral view. Osteonecrosis of irregular bone may be radiographically similar to the appearance of bone infarct in diametaphyseal bone (Figure 17-7). However, posttraumatic osteonecrosis in certain bones, such as the talus and navicular, eventually demonstrate sclerosis following normal radiographs (Figure 17-8). In the traumatized talus, with patent vascularity, Hawkins15 described a linear radiolucency that parallels the articular surface of the talar dome, now referred to as Hawkins sign. This finding may be found in other

bones as well (Figure 17-9).

FIGURE 17-3. Epiphyseal osteonecrosis, second metatarsal. Centralized sclerosis mixed with ill-defined rarefaction.

FIGURE 17-4. Epiphyseal osteonecrosis, fourth metatarsal. Collapse of subchondral bone; an ill-defined curvilinear radiolucency parallels the subchondral surface, the “crescent” sign (arrows). Dysbaric osteonecrosis (caisson disease), which occurs in people who work with compressed air (high-pressure environments), was first described by Bornstein and Plate.16 Bone infarction occurs during decompression; at lower atmospheric pressure, nitrogen leaves solution and forms bubbles in its gaseous state. These nitrogen bubbles can occlude vessels and are thought to cause the characteristic bilateral osseous lesions.4 Histopathologic changes suggest that a disturbance of blood supply caused by platelet aggregation, red cell sludging, and thrombosis associated with intravascular air bubble formation is related to the occurrences of osteonecrosis in divers.17 Radiography demonstrates large bilateral and symmetrical serpentine lesions. The lesion is initially radiolucent with a calcific or ossific rim and usually calcifies over time.11 Juxta-articular lesions may progress to fragmentation and structural failure of the subchondral plate, resulting in symptomatology and osteoarthritis.18

FIGURE 17-5. Epiphyseal osteonecrosis, third metatarsal. Flattened subchondral surface, fragmentation of epiphysis resulting in large, central loose body.

FIGURE 17-6. Old, healed epiphyseal osteonecrosis, second metatarsal, and secondary osteoarthritis. The second metatarsal head and proximal phalangeal base are flattened and deformed with increased girth, along with osteophyte formation and uneven joint space narrowing.

FIGURE 17-7. Osteonecrosis of diametaphyseal and irregular bones (arrows). A: First metatarsal. B: Calcaneus. Note the serpiginous calcification similar to that seen in the diametaphyseal infarct.

FIGURE 17-8. Healed osteonecrosis of an irregular bone, the navicular. It is sclerotic and deformed.

FIGURE 17-9. The concept of Hawkins sign applies to other bones as well. Note the linear radiolucency that parallels the outer margin of every bone in the medial oblique view. This is a result of acute osteopenia, and depicts a patent blood supply to bone.

Gaucher disease is a metabolic disorder in which cerebrosides are deposited in the cells of the reticuloendothelial system. The skeletal changes are a result of the infiltration of the marrow with cerebrosides that occlude the vessels and produce osteonecrosis.19 Early radiographic changes of femoral head osteonecrosis include osteopenia of the distal femur that progressively expands the cortex distally to a contour resembling an Erlenmeyer flask.19,20 The hemoglobinopathies are subdivided into two groups. The first group, sickle cell anemia, results from an inherited structural alteration in one of the globin chains. The second group, the thalassemias, results from inherited defects in the rate of synthesis of one or more of the globin chains.21 The fundamental abnormality in sickle cell disease resides in the abnormal structure of the globin portion of the hemoglobin molecule. Sickle cell hemoglobin (HbS) is much less soluble than normal adult hemoglobin (HbA) in the deoxygenated state. The insoluble hemoglobin molecules aggregate in erythrocytes, which distorts the shape of the red cells and produces the characteristic sickle cell deformity.22 Sickle cell disease and the other hemoglobinopathies can cause either epiphyseal osteonecrosis or metaphyseal and diaphyseal infarcts.22 Sludging of sickled erythrocytes within the sinusoidal vascular bed results in occlusion.11 The painful vasoocclusive crisis is probably caused by infarction of bone marrow. Signs of acute long bone infarction in children are common and can resemble those of acute bacterial osteomyelitis. Acute long bone infarction is at least 50 times more common than bacterial osteomyelitis in sickle cell disease.23 In older patients, infarcts can also appear in flat bones. Radiographs do not show the infarcts for several weeks. Lesions in long bones often affect the femoral and humeral heads.22 The thalassemias are a group of anemias that are hereditary and manifest hypochromic microcytic anemia with a decreased synthesis of one or more of the constituents of hemoglobin.21 Cooley anemia is the homozygous form of the disease inherited from both parents, and the children rarely survive past adolescence. The small bones, including the metatarsals and phalanges, appear rectangular, due to loss of the normal concavity of the shaft and cortical thinning.23 Trabecular resorption occurs, and the remaining

trabeculae appear coarsened11 (Figure 17-10).

FIGURE 17-10. Thalassemia major. OSTEOCHONDROSIS Osteochondrosis is defined based upon the radiographic appearance of an ossification center (epiphysis or apophysis) that is sclerotic and fragmented (Figure 17-11).1 It also can appear irregular and smaller than the expected size.12 Epiphysoid bones that can appear sclerotic and fragmented, such as the navicular, are also included, which form from a cartilaginous precursor by ossification from the center outwards, like an epiphysis.24 Though Koulouris and Morrison25 believe the term osteochondrosis, which was originally believed to be avascular necrosis, is anachronistic, it still is in common use today. Some of the so-called osteochondroses are normal variations, others

are true osteonecroses, and some are growth disturbances with no evidence of necrosis.12 Table 17-1 lists the named osteochondroses in the lower extremity, based upon the author who (in most cases) first described it in the literature. Osteochondroses have also been classified based upon anatomical site of involvement: articular (Freiberg), nonarticular (Sever), or physeal (Blount).26 The nonarticular osteochondroses are further divided into the following sites: (1) tendon attachment, (2) ligament attachment, and (3) the site of impact26,27; Some osteochondroses could easily fit into two if not all three of these subdivisions.28 OSTEOCHONDRAL INJURY The terms osteocartilaginous body, joint mouse, transchondral fracture, osteochondral lesion, osteochondral defect, osteochondrosis dissecans, and osteochondritis dissecans have all been used to refer to osteochondral injury.29 The term osteochondral injury is preferred because it encompasses both traumatic and nontraumatic causes (such as repetitive microtrauma, primary ischemic event, genetics, and corticosteroid use).9,14,29–33 It has been observed in two age groups, the juvenile with open physes (between 5 and 15 years of age) and adults and older adolescents with closed physes (between 15 and 50 years of age).34 The juvenile group has at times been referred to as osteochondrosis2; it has also been referred to as localized osteonecrosis in adults because the underlying subchondral bone may become necrotic.11

FIGURE 17-11. Osteochondrosis of the calcaneal apophysis demonstrating characteristic features of fragmentation (arrows) and sclerosis (arrowhead). TABLE 17-1   Named Osteochondroses in the Lower Extremity1,12,28,166 Eponymous Location Legg–Calve–Perthes Osgood–Schlatter Blount Liffert and Arkin Diaz; Moucheta Sever Köhler Buschkea Iselin

Capital femoral epiphysis Tibial tubercle Posteromedial aspect of proximal tibial physis (involving metaphysis and epiphysis) Distal tibial epiphysis Talar body Calcaneal apophysis Navicular Tarsal cuneiforms Apophysis at base of fifth metatarsal

Freiberg Renander; Trevesa Thieman

Metatarsal head, most frequently the second and third Tibial sesamoid Base of proximal phalanx hallux

aThough listed by Breck,166 no direct references to these authors could be

found in the literature. Osteochondral injury occurs most commonly in the knee; the most common site in the foot is the talar dome.34 Most osteochondral injuries appear to be related to trauma, which could have been either acute or chronic and repetitive.25 The characteristic radiographic picture is a “button” of subchondral bone sitting in a radiolucent defect (Figure 17-12).11,35 Bone scintigraphy, arthrography, computed tomography (CT), and especially MRI have all been shown useful in identifying and staging osteochondral lesions.2,30,36,37

FIGURE 17-12. Osteochondral injury of the talar dome demonstrating a “button” of subchondral bone sitting in a well-defined lucent defect (arrows). PRESENTATIONS OF OSTEONECROSIS, OSTEOCHONDROSIS, AND OSTEOCHONDRAL INJURY IN INDIVIDUAL BONES Femur

Osteonecrosis of the femoral head epiphysis is known as Legg–Calvé– Perthes disease, based upon three separate descriptions of a similar abnormality affecting the hip joint published in the early 1900s. It is a true osteonecrosis that occurs between infancy and age 16 with the greatest incidence at age 5; it occurs much more frequently in boys than girls but is rare in black individuals.11 The diagnosis of capital femoral epiphysis osteonecrosis relies on imaging methods, including radiography, CT, scintigraphy, and MRI.38 The radiographic features and stages of epiphyseal osteonecrosis in general are based upon the research and descriptions of capital femoral epiphysis osteonecrosis. It is characterized by necrosis and followed by a regenerative process that is variable, depending on the patient’s age, the adequacy of treatment, and the rapidity with which treatment is instituted.39 Radiographic findings consist of initial joint space swelling with lateral displacement of the hip.40 In this early phase, mottled, scattered radiodense areas may be mixed with osteopenia. Subchondral fracture shows up radiographically as a radiolucent line and is called the crescent or rim sign38; this is visible only before collapse of the articular cartilage and attached necrotic bone.41 Later, after flattening of the femoral head, joint space narrowing occurs and secondary osteoarthritis results.42 In the initial stages, when radiographs are normal, MRI is the method of choice for the diagnosis of osteonecrosis, as it is more sensitive than radiography or CT. Literature shows the sensitivity of MRI to be 97% in differentiating osteonecrosis from normal hip, and 85% in differentiating femoral head osteonecrosis from other hip disorders.43 MRI is accurate in detecting osteonecrosis before clinical or radiographic changes become evident. Gadolinium-enhanced spin echo sequences and fat-suppressed images have improved MRI specificity even more (75%–100%). MRI is considered the “gold standard” noninvasive diagnostic method for osteonecrosis and is indicated when differential diagnosis with other hip disorders is difficult using other methods.38 Tc-99 bone scintigraphy is very useful in the early stages when the radiographs are normal, and helps in mapping the extent of the disease when

multiple sites of osteonecrosis are present. Proximal Tibia Osteochondrosis of the tibial tubercle, or Osgood–Schlatter disease, is not osteonecrosis. It is considered to be a traction apophysitis, due to repetitive strain or pull, where the patellar tendon attaches to the secondary ossification center of the tibial tuberosity.35,44 This condition is more frequent in boys between the ages of 11 and 15 who are involved in jumping and running activities, just as the ossification center of the proximal tibial tubercle appears radiographically.11,45,46 Pain, swelling, and tenderness over the tibial tubercle are noted.12 Pain is exacerbated by physical activity such as running, going up or down steps, or squatting, which can result in traumatic avulsion of the quadriceps tendon or a tendonitis of the tendon with heterotopic new bone formation.12,47,48 The normal tibial tubercle is often fragmented and sclerotic; more than one ossification center is frequently evident (Figure 1713). In patients with Osgood–Schlatter disease, lateral radiographs may show anterior soft tissue swelling and irregularity of the apophysis, with separation from the tibial tuberosity in early stages. Widening or fragmentation of the tibial tuberosity, with the presence of bony ossicles at the patellar tendon insertion, is noted in later stages (Figure 17-14). MRI can demonstrate details not visible on radiographs and can help differentiate Osgood–Schlatter disease from acute tibial apophyseal fracture, infection, and tumor.44,49,50 Hirano50 demonstrated five stages in the progression of this disease on MRI: normal, early, progressive, terminal, and healing. A normal MRI study does not necessarily rule out the diagnosis, as it might very well mean that the patient is in an early stage of the disease. Ultrasound may reveal pretibial edema, thickening of the patellar tendon, and fragmentation of the ossification center.49,51 Blount disease, also known as tibia vara and osteochondrosis deformans tibiae, is also not a true osteonecrosis; it is a growth disturbance of the posteromedial part of the proximal tibial physis. Growth of the epiphyseal cartilage is arrested, resulting in varus medial torsion of the tibia and a degree of flexion of the diaphysis on the upper tibial epiphysis.52 Blount53 described

two types of the disease, infantile and adolescent. The infantile form occurs between ages 1 and 3 and is much more common and severe. The adolescent form occurs between ages 6 and 13, and is rare in comparison.54 In contrast to Osgood–Schlatter disease, Blount disease is common in black children.55 Radiographically, the medial part of the epiphysis is poorly developed and shows a beak on the posteromedial aspect of the metaphysis; these changes are progressive (Figure 17-15).52 The sharply angular appearance of infantile Blount disease differs from the gradual curve in physiologic bowed legs. The nature and severity of the osseous changes are highly variable.7

FIGURE 17-13. Variant ossification and segmentation of the tibial tubercle ossification center mimicking Osgood–Schlatter disease. This patient was asymptomatic at that location.

FIGURE 17-14. Osgood–Schlatter disease. Prominent tibial tubercle with fragmentation associated with positive clinical symptomatology, consistent with osteochondrosis of the tibial tubercle.

Distal Tibia Siffert and Arkin56 were the first to report a case of avascular necrosis of the distal tibial epiphysis. It resulted from a crush injury to the lateral half of the epiphysis. Radiographic findings included sclerosis and fragmentation, and portions of the bone were histologically positive for osteonecrosis. Osteochondral injury of the distal tibial plafond has been rarely described in the literature, but may not be as rare as the number of cases reported. The etiology, though unknown,57 is probably traumatic, as in the talar dome.58 It is seen better in the anteroposterior (AP) ankle view than the lateral view (Figure 17-16). There does not appear to be any location preference along the tibial plafond as there is on the talar dome. The imaging characteristics are similar to those seen in the talar dome, described next.

FIGURE 17-15. Blount disease. There is severe varus deformity of the left leg. Talus Diaz first described talar body osteonecrosis in 1928.59 Its etiology is most commonly posttraumatic with secondary disruption of the talar body’s blood supply.9 Osteonecrosis of the talar dome is a well-recognized complication of talar neck fractures, occurring in 30%–50% of such injuries.14,60 Radiography is the initial imaging study of choice. The necrotic area of bone may appear sclerotic relative to surrounding osteopenia earlier in the disease (Figure 17-17). As repair progresses, the area of osteonecrosis becomes more sclerotic as new bone is laid down upon necrotic trabeculae. A radiolucent rim eventually surrounds the sclerosis. The talar dome may collapse and fragment in more severe cases (Figure 17-18).9 CT has been used to confirm the radiographic findings.61 MRI is the most sensitive technique for detecting talar osteonecrosis. Bone scintigraphy is not commonly used in clinical practice and is less sensitive than MRI in diagnosing symptomatic avascular necrosis.9,62 Avascular necrosis of the talus has occurred after a variety of surgical procedures, including medial and lateral release of congenital clubfoot, tibiotalar-calcaneal arthrodesis, triple arthrodesis, and talonavicular arthrodesis, which may disrupt talar vascularization.6,63,64 When evaluating patients with persistent pain and disability postoperatively, artifacts from metallic implants may obscure characteristic imaging signs of osteonecrosis.

FIGURE 17-16. Osteochondral injury of the tibial plafond. There is a large defect in the superomedial corner of the tibial plafond (closed arrow); the open arrow points to an ossicle that may have originated from here.

Radiography and CT are the mainstay imaging modalities of the postoperative ankle and foot. MRI and bone scintigraphy play an important complementary role in differentiating avascular necrosis from infection, nonunion, occult fracture, and secondary osteoarthritis.6 Thordarson65 has noticed that, after open reduction internal fixation of talar neck fractures, the use of MRI in less than 6 weeks postoperatively had a high false negative rate, a finding confirmed by other authors as well. He noted that it might take 3 to 6 weeks before osteonecrosis is seen on MRI. Radiographs in this postop period may show a negative Hawkins sign, indicative of osteonecrosis, while MRI might be negative.9,66

FIGURE 17-17. Osteonecrosis of the talus, early. Lateral view demonstrates fracture of the talar neck and diffuse osteopenia; the talar body appears relatively increased in density. Hawkins sign, characterized by subchondral osteopenia on AP or mortise radiographs (Figure 17-19), classically begins at the medial subchondral bone of the talar dome and progresses laterally.15,67–69 Typically, it results from the resorption of subchondral bone in the setting of disuse and a sufficient vascular supply. Thus, Hawkins sign, indicating partial revascularization, is a reliable indicator of talar viability and its presence serves as an early negative predictor of osteonecrosis. However, the absence of Hawkins sign does not confirm osteonecrosis, because it has greater sensitivity than specificity.9,68 Detection of Hawkins sign 4 to 8 weeks after fracture or surgical intervention

is suggestive of revascularization of the relevant portion of the talar body.6,9 A partial Hawkins sign is more commonly observed in the medial talus, which is indicative of incomplete osteonecrosis; this indicates the susceptibility of the lateral talar dome or inferior articular surface of the body to osteonecrosis.6,61 Complete revascularization after surgery may take between 6 months and 3 years. During this time, fractures may heal as the progressive sclerosis and cysts of osteonecrosis either resolve or lead to osseous collapse.6 CT is particularly useful in the assessment of healing or evolution of osteonecrosis after triple arthrodesis.6

FIGURE 17-18. Osteonecrosis of the talus, late and severe deformity. A: Anteroposterior view. B: Lateral view.

FIGURE 17-19. Hawkins sign. Note the linear radiolucency that parallels the articular surface of the talar dome (arrows). This is a result of acute osteopenia, and depicts a patent blood supply to the talar body.

FIGURE 17-20. Osteochondral injury, talar dome, superomedially (arrows). A: Mortise view. B: Lateral view. Osteochondral injury of the talar dome has also been referred to as osteochondritis dissecans (a confusing term as it implies inflammation), osteochondral defect, transchondral fracture, talar dome fracture, osteochondral fracture, and avascular necrosis of the talus.6,14,70 It frequently occurs between 20 and 30 years of age, with a slight preponderance in males, and is mostly unilateral, with only about 10% of the cases being bilateral finding.70 According to Van Buecken, osteochondral injury is noted as a complication in 6.5% of ankle sprains.70,71 It is fairly common and occurs in either the upper medial or lateral aspect of the dome72 (Figure 17-20). Flick and Gould studied hundreds of OCLs and found that 98% of the lateral dome lesions and 70% of the medial dome lesions were associated with a history of trauma.29,33 Medial lesions are most common in the posterior third of the dome and may have a deep cup or crater appearance. Lateral lesions are often situated in the middle third of the dome and are shallower. Central dome lesions are uncommon.14,37,73 It is suggested that, for the initial evaluation of acute ankle injury, radiography should be the first choice for diagnosis and temporal observation of talar osteochondral injury.61,70 Even if radiographs cannot detect early and

less advanced disease, it is recommended to be obtained in all cases of suspected or known osteonecrosis.9 If any lesions are detected, further details should be obtained via CT. If radiographs are normal, but the patient continues to have persistent pain, an MRI study may be warranted.70,74 Radiography, using standard weight-bearing views (AP, lateral, oblique), remains the preferred first line imaging study when suspicion of osteochondral injury arises, in part to rule out fracture.29 Canale and Kelly67 used a special positioning technique in order to detect any “offset” or varus deformity of the talar head and neck; the image is performed as a non– weight-bearing (or semi–weight-bearing) dorsoplantar (DP) foot view with the ankle joint maximally plantarflexed and the foot pronated 15°, while the x-ray beam is angled at 15° from vertical. Posteromedial lesions are best visualized in the mortise view of the ankle in plantar flexion.70,75 The main disadvantage of radiography in assessing osteochondral injury is that it is unable to assess the state of the cartilage.29,76 CT is useful for assessing in greater detail the bony injury, including the site, size, shape, and extent of fragment displacement. MRI, is becoming the gold standard for diagnosis of osteochondral injury because it is able to detect early bone bruise, cartilage damage or other soft tissue injury; MRI also correlates well with arthroscopic findings.29,70,77–79 MRI is the most sensitive technique for detecting talar osteochondral injury and can be used when it is strongly suspected clinically despite normal radiographs61; however, the signal patterns in the talus may overestimate the severity of the bone injury.29,80 MRI is also performed to evaluate stability of the osteochondral fragment. The radiographic classification system developed by Berndt and Harty81 in 1959 is still in current use; Anderson et al.30 and Loomer et al.76 later modified it. (The reader is referred to Chapter 16 for further discussion of talar osteochondral injury and the Berndt and Harty classification.) There are several other grading systems used to assess fragment stability of talar dome osteochondral injury based on arthroscopic, MRI, and CT findings.29,77,82–85

Calcaneus Osteochondrosis of the calcaneal apophysis is known as Sever disease.86 In 1912, Sever reported heel pain in adolescent boys with the radiographic picture of a sclerotic apophysis with fragmentation.87 For decades it was believed that these findings were diagnostic of calcaneal apophysitis. Osteochondrosis of the calcaneal apophysis, however, is now generally considered to be a variation of normal development. The calcaneal apophysis is an accessory ossification center and appears radiographically at 4 to 7 years in girls and 4 to 10 years in boys; it fuses at an average of 16 years of age.88,89 The center of ossification is oriented perpendicular to the axis of the calcaneal tuberosity. The Achilles tendon inserts on its posterior aspect and the plantar fascia originates at its inferior extent (the calcaneal tubercle); superiorly it forms the bursal projection.49 The normal calcaneal apophysis is sclerotic, relative to the calcaneal body, and often appears fragmented because of the presence of multiple ossification centers (Figure 17-11).90–92 It is also normal during development to see a jagged or serrated appearance along the margin of the adjacent metaphysis (Figure 17-21).93 Radiographically, it has been shown that increased density of the calcaneal apophysis is normal and attributed to weight bearing because it is absent in children without normal weight bearing.91,92 There is no consensus on the radiographic appearance of calcaneal apophysitis in the relevant literature, and no signs have been accepted as pathognomonic for the diagnosis of Sever disease.86,94 Furthermore, there is no study on the interobserver and intraobserver reliability of any of the radiographic findings regarding Sever disease, and neither their sensitivity or specificity are known.86,94 However, Perhamre et al.92 concluded that a high degree of fragmentation was significantly more frequent in patients with calcaneal apophysitis than in the control group. As such, the use of radiographs as a routine screening procedure does not seem to be cost-effective and there is the risk of ionizing radiation. Therefore, radiography should be considered only to aid in differential diagnosis when considering other pathologic conditions in recalcitrant cases.86,89,94–96

Heel pain in the patient with an unfused apophysis may represent apophysitis, especially in boys 10 to 12 years of age.49,89–92 However, growth disturbances and bone density changes are not typical.49 No radiographic criteria have been established for the diagnosis of Sever disease.91,92 The diagnosis of Sever disease is clinical (via the highly sensitive and specific one heel standing test and heel squeezing test).92 Ogden et al.93 have indicated that Sever disease may result from repetitive compression injury to the actively remodeling trabecular metaphysis, rather than being a traction apophysitis. They used the microfracture, hemorrhage, and edema visible on MRI to sustain their conclusions. MRI in Sever disease shows bone edema in the metaphysis adjacent to the apophysis, suggesting metaphyseal stress fracture; the adjacent apophysis was normal. The metaphyseal bone edema resolved after immobilization.93,97 Navicular Osteochondrosis of the navicular bone is known as Köhler disease; it is an abnormality of endochondral ossification.98 True atraumatic osteonecrosis of the navicular is extremely uncommon. It occurs more frequently in males than females, typically between the ages of 3 and 7. This condition is selflimiting, with unilateral involvement present 75% to 80% of the time, and has an excellent prognosis.7,14,99 According to Ozonoff,13 the diagnosis can only be made by showing that a normal navicular was present before the development of sclerosis and fragmentation, or by showing progressive bone resorption followed by progressive bone repair on serial radiographic studies. Much more common is an anatomic variation, delayed appearance of the navicular ossification center.100 In this case, the navicular radiographically appears sclerotic, fragmented, and narrow in its AP diameter with a hazy outer border (Figure 17-22). Ferguson found irregular ossification of the navicular in about one-third of normal children.100 Although navicular osteochondrosis may present bilaterally, contralateral radiographs are helpful for comparison.97 The differential diagnosis includes acute or stress navicular fracture.97 Even though the navicular in the developing skeleton is not an epiphysis or apophysis, Köhler disease is considered an osteochondrosis

because it is an epiphysoid bone that forms from a cartilaginous precursor by ossification from the center outwards, like an epiphysis.24

FIGURE 17-21. Osteochondrosis of the calcaneal apophysis. The appearance of sclerosis and fragmentation are variations of normal. The adjacent serrated metaphysis is normal development.

FIGURE 17-22. A: Variant ossification and segmentation of the navicular ossification center mimicking Köhler disease. B: Opposite foot, same patient. McCauley and Kahn101 used bone scintigraphy to evaluate a possible Köhler disease. He found decreased uptake of the symptomatic navicular, which he argued supported a diagnosis of true osteonecrosis. However, Brower12 suggested that the decreased uptake was probably caused by delayed ossification at the growth center. She goes on to argue that, because a true osteonecrosis must involve a hypervascular state, to support the diagnosis the scan should be hot at some point. Weston demonstrated an increased focal uptake of scanning agent in a true Köhler disease.102 Whereas true osteonecrosis requires immobilization to prevent collapse of the navicular, the anatomic variation generally ossifies normally regardless of treatment. Mueller–Weiss syndrome is an idiopathic, painful deformity of the adult navicular bone. Although it resembles osteonecrosis on MRI, this not been supported by histopathologic studies in the literature. It also has been suggested that it is the sequella of undiagnosed navicular stress fracture.103 This syndrome occurs predominantly in females in their 4th to 6th decades and is usually bilateral.104

Maceira and Rochera105 propose that Müller–Weiss syndrome is related to delayed ossification of the navicular that becomes deformed secondary to abnormal force distribution on the bone. They have grouped the radiographic presentations of increasing deformity into five stages, which are chronic and progressive (Table 17-2).106 The “3rd stage” of Müller–Weiss syndrome appears similar to the bipartite navicular; the division or partition separates the bone into a smaller superolateral segment and a larger inferomedial segment107 in the DP view, the larger navicular segment is shaped like a wedge107 or comma.105 Its base is positioned medially and apex points laterally; the smaller segment is superimposed on the lateral cuneiform and cuboid (Figure 17-23A). In the lateral view, the smaller segment is shaped like a wedge and positioned along the dorsal aspect of the larger segment; its apex is directed inferiorly (Figure 17-23B).22 The smaller bipartite segment appears to articulate with the intermediate cuneiform superoposteriorly. Osteochondral injury of the navicular is rarely reported in the literature.58,108 The few cases reported were in patients ranging in age from 19 to 42; its etiology is unknown.109 The lesion is found along the concave proximal articular surface for the talus, in its central one-third. Radiographically, the lesion is seen in a localized depression or lucency along the subchondral bone plate, surrounded by sclerosis (Figure 17-24). Both CT and MRI can be used to confirm the lesion and determine its extent. TABLE 17-2   Stage 1

2

Mueller–Weiss Syndrome: Five Progressive Deformity Stages in Lateral View as per Maceira and Rochera105 Radiographic Findings Normal radiograph despite abnormal MRI Talar axis superior to first metatarsal; navicular appears to subluxate inferiorly relative to talar head; mild compression of navicular Talar axis returns parallel to first metatarsal; space between talar head

3

4

5

and cuneiforms is greatly reduced; the navicular bone fragments and begins to subluxate off talar head superiorly; lowering of longitudinal arch Talar axis inferior to first metatarsal; further compression of navicular; greater navicular subluxation superiorly; collapse of longitudinal arch Extrusion of navicular, dislocation of navicular fragments such that lateral and intermediate cuneiforms appear to articulate with talar head; rearfoot equinus attitude; ultimately, progression of deformity results in osteoarthritis of talonavicular joint

FIGURE 17-23. Müller–Weiss syndrome, stage 5. A: Dorsoplantar view. B: Lateral view. Recent studies have found an increased occurrence of navicular osteonecrosis after traumatic injuries ranging from avulsion fracture to comminuted fracture and dislocation. Fracture of the navicular body, usually due to highenergy trauma including motor vehicle accident, is most likely to develop osteonecrosis, as the center of the bone has only centripetal intraosseous flow.107,110 Cuneiforms Osteochondrosis of a cuneiform has been referred to as Buschke disease; however, the literature points to Buchman (1933) first reporting osteochondrosis of the medial cuneiform, Lewin (1929) the intermediate cuneiform, and Wagner (1928) the lateral cuneiform.111–113 The reversible nature of the irregular size, shape, and trabecular density of the cuneiforms, regardless of treatment, is a strong indication that this disorder is probably not a true osteonecrosis and is nothing more than a normal variation in endochondral ossification.112,114 (Figure 17-25). Repetitive microtrauma is often cited as the cause.92,115

FIGURE 17-24. Osteochondral injury of the navicular proximal articular surface (arrow). Because of its benign, short duration (2–8 months),116 self-limiting, and sometimes asymptomatic nature, it is possible that this transient disorder of the ossification process is easily missed.112 Only 18 cases of idiopathic cuneiform osteochondrosis have been found in the literature so far, between ages of 4 and 6 years,112 more common in girls: 1 in the lateral cuneiform bone,116 5 in the intermediate cuneiform bone,114,115,117,118 and 12 in the medial cuneiform bone.111,113,117–125 The radiologic characteristics are similar in all the cases; only the duration of the clinical symptoms (pain, local tenderness, antalgic gait) vary. In the case described by Garcia-Mata,112

osteochondrosis of the intermediate cuneiform bone was associated with a limp in a 4-year-old child. In this case, radiographically, an increase in density of the second cuneiform and smaller size of the bone were noted, which resolved in about 5 months. Loss of signal may be noted with T1weighted MR images.115,118 Conservative treatment including rest, or limitation of physical exercises, and NSAIDs help in complete recovery of the size, shape, and density of the involved cuneiform, with resolving symptoms.112,114,116,118,124

FIGURE 17-25. Osteochondrosis of the medial cuneiform ossification center demonstrating variant ossification, with sclerosis and segmentation. Fifth Metatarsal Osteochondrosis of the apophysis at the base of the fifth metatarsal is referred to as Iselin disease. The term also has been used to refer to apophysitis of the styloid process of the fifth metatarsal, which, like calcaneal apophysitis, is a clinical diagnosis. The apophysis of the proximal fifth metatarsal appears

radiographically at about age 10 in girls and age 12 in boys, with fusion 2 years later.49,126 The secondary ossification center is within the insertion site of the peroneus brevis on the dorsolateral aspect of the tuberosity.49,127 Radiographically, the secondary center of ossification is best seen in the medial oblique view.128 It appears as a small, shell-shaped fleck of bone, slightly oblique to the long axis of the metatarsal shaft along the plantarlateral aspect of tuberosity.49,97,127–129 An irregular shaped apophysis may be normal. Although the apophysis can appear fragmented, this is a normal finding, not pathologic (Figure 17-26). Osteochondrosis of the fifth metatarsal apophysis has the typical radiographic appearance of fragmentation and sclerosis. In bilateral studies, the apophysis may appear slightly enlarged in symptomatic patients,128 with slight separation of the chondro-osseous junction.97

FIGURE 17-26. Osteochondrosis of the fifth metatarsal basal apophysis ossification center, medial oblique view. Enlargement of the apophysis with fragmentation and widening of the chondro-osseous junction, seen best with the medial oblique view, confirm the diagnosis in a young athlete presenting clinically with painful enlargement of the fifth metatarsal base.49,126,128 If radiographs are normal despite high clinical suspicion, a Tc-99 bone scan will often show increased uptake over the apophysis.130,131 Bony union across the apophyseal line will occur in most patients,126 and symptoms usually resolve when the apophysis unites to the metaphysis.49 It may be difficult to differentiate an avulsion fracture from the apophysis radiographically.28,49 However, the apophysis is oriented slightly oblique to the long axis of the metatarsal shaft and does not incorporate the joint space, while tuberosity fractures are oriented more transverse, and can be intraarticular.49 Second, Third, and Fourth Metatarsals Freiberg disease (originally described as an “infraction”132) is a true osteonecrosis of the lesser metatarsal head seen most commonly in the second metatarsal. It is much more prevalent in females than males (5:1) and occurs most commonly between the ages of 12 and 18 during adolescent growth spurts.59,133 Freiberg disease undergoes a radiographic sequence of events consistent with osteonecrosis; a subchondral bone fatigue fracture may occur secondary to repetitive mechanical stress (trauma) resulting in ischemic bone, although the exact pathophysiology remains unclear.14,90,134 Braddock135 showed that, in this age group, the weakest area in the metatarsal is the epiphyseal area, which is the area most likely to fracture. Clinically, pain, swelling, and tenderness about the metatarsal head area are noted, with decreased range of motion at the metatarsophalangeal joint.35 Smillie136 described the macroscopic appearance of Freiberg disease as five progressive stages (Table 17-3).137 The radiographic presentation of Freiberg

disease also has varying presentations depending on its stage of healing (Figure 17-27). The radiographic study, at a minimum, should include DP and medial oblique views. Early signs of osteonecrosis include ill-defined subchondral sclerosis mixed with osteopenia and flattening of the metatarsal head (Figure 17-3).97 The earliest finding may be joint space widening; however, a flattened metatarsal head associated with joint space widening but no subchondral sclerosis is a variation of normal.133 A fracture in the subarticular necrotic bone presents as a crescent-shaped radiodensity (Figure 17-4).41 Loose body formation (Figure 17-5) and secondary thickening of the metatarsal shaft can also occur.97,138 In later stages, the shaft and neck of the metatarsal increase in girth. The base of the proximal phalanx often widens and molds to the abnormally shaped metatarsal head.11 The end result is marginal osteophytosis and joint space narrowing from damage to the articular cartilage.59,90,137 The classic radiographic picture of Freiberg disease in the adult represents an old, healed osteonecrosis with secondary osteoarthritis (Figure 17-6). Table 17-4 summarizes the radiographic progression of Freiberg disease. Gauthier and Elbaz134 proposed another classification system based on the degree of osteonecrosis and potential for subsequent healing (Table 17-5). Smillie136 Classification of Freiberg Disease with TABLE 17-3   Radiologic Correlation133 Radiographic Stage Macroscopic Features136 Correlation133 Fissure fracture in ischemic epiphysis (not visible I radiographically; detectable Joint space widening only with technetium bone scan or MRI) Resorption of cancellous bone on proximal side of lesion; dorsal metatarsal head begins Flattening of II to collapse into the epiphysis; metatarsal head alteration of the articular surface

III

IV

V

Further collapse of metatarsal head with still intact plantar Subchondral collapse articular cartilage; sinking of centrally the main portion of the articular surface leaving bony projections on each side; small dorsal exostosis Collapse on entire metatarsal head, including plantar aspect results in a large loose body; Multiple loose bodies bony projections fracture and at joint periphery fold over the central loose body Severe arthrosis; marked flattening and deformity of Degenerative joint metatarsal head; loose body disease reduced in size; thickened and dense metatarsal shaft

MRI is particularly useful early in the disease process to quantify osteonecrosis, as foot radiographs may be normal.14,137 MRI displays findings similar to that described in other bones for osteonecrosis.14,133 Osteochondral injury can also involve the lesser metatarsal head in adults (Figure 17-28); it can result in an osteoarthritic joint that mimics end stage Freiberg disease. The etiology is probably similar to that of Freiberg disease, a combination of mechanical stress, subchondral fracture, vascular injury, and subsequent osteonecrosis.59

FIGURE 17-27. Radiographic progression of Freiberg disease in one patient. A: Head begins to flatten, with ill-defined sclerosis with mixed osteopenia. B: Metatarsal head is flat, and there is an irregular fracture line that parallels to subchondral bone plate (crescent sign). C: Diffuse sclerosis, still ill defined; the subchondral surface has sunken into the epiphysis. D: Remodeling continues, the metatarsal head is becoming enlarged. E: Early closure of the physis; sclerosis is now well defined. TABLE 17-4   Radiographic Progression of Freiberg Disease Radiographic Findings Normal radiograph Joint space widening Flattening of epiphyseal articular surface; ill-defined sclerosis mixed with periarticular osteopenia; subchondral fracture (“crescent” sign) Flattened epiphysis as subchondral bone collapses with progressive central depression; joint space preserved; subchondral rarefaction bounded by sclerosis Loose bodies, bone deformity (widening); sclerosis

Osteophyte, gross deformity, joint space narrowing Torriani et al.139 observed that MRI is able to detect subchondral fracture prior to radiographs. Their findings also suggest that MRI is able to differentiate between early and late stages of metatarsal head osteonecrosis, correlating to Smillie’s five-stage classification of Freiberg disease. First Metatarsal First metatarsal head osteonecrosis is uncommon. Most reports of it follow correctional surgery for hallux valgus140,141; it may rarely occur after isolated cheilectomy for repair of hallux rigidus.6,142 The location of the osteotomy and extent of soft tissue resection may contribute to the development of osteonecrosis.6,110 Symptoms may not manifest until osteoarthrosis has already developed, and patients usually present with pain and limited range of motion. TABLE 17-5   Stage 0 I II III IV

Stages of Freiberg Disease according to Gauthier and Elbaz134 Degree of Osteonecrosis and Potential for Subsequent Healing Epiphyseal fracture: healing now can maintain normal joint Osteonecrosis: healing now can maintain normal joint Mild metatarsal head flattening: restoration is still possible Fragmentation and loose bodies: irreparable damage Advanced arthrosis

Meier and Kenzora143 proposed a radiographic classification system for osteonecrosis of the first metatarsal head (Table 17-6) The radiographic findings are dependent upon the time that the osteonecrosis is evaluated; also, the presence of a stage I finding does not necessarily mean that the osteonecrosis will progress to stages II or III.144 When radiographs are

normal but osteonecrosis is suspected clinically, bone scintigraphy or MRI is valuable for confirming suspicion.145 Postsurgically, distinction of radiographic signs of osteonecrosis from other processes, including osteotomy and bone resection, thermal damage, hyperemia, altered mechanics, and osteotomy instability, is difficult. Increased radiodensity, mild osseous resorption, and small subchondral cyst formation may be seen as part of the normal healing process following first metatarsal osteotomy. Sequential evolving lucency in the subchondral bone and eventual collapse are highly suggestive of osteonecrosis.6

FIGURE 17-28. Osteochondral injury of the third metatarsal head in an adult. Note the severe collapse of subchondral bone and mixed sclerosis and rarefaction. TABLE 17-6   Stage I: Precollapse     II: Collapse   III: Arthritis  

Radiographic Classification of First Metatarsal Head Osteonecrosis Time of Radiographic Findings Evaluation Ia: Early Normal radiograph Ib: Intermediate Acute osteopenia in surrounding bone Ic: Late Sclerosis IIa: Early Mild subchondral bone collapse Severe subchondral bone collapse and IIb: Late fragmentation Osteophyte, geode, joint space IIIa: Early narrowing Joint space obliteration; sclerosis; IIIb: Late deformity

Modified from Meier PJ, Kenzora JE. The risks and benefits of distal first metatarsal osteotomies. Foot Ankle. 1985;6:7. CT is commonly used to assess for osteotomy complications including malunion in dorsiflexion, nonunion, and avascular necrosis. Small field of view MRI is ideally suited to imaging cartilage due to the excellent resolution and contrast discrimination, assisting in the distinction of osteonecrosis of the metatarsal from osteoarthritis.6,146

FIGURE 17-29. Osteochondral injury of the first metatarsal head. The arrow identifies a lucent subchondral defect.

FIGURE 17-30. Osteonecrosis of the fibular sesamoid. A: Dorsoplantar view. B: Sesamoid axial view. Findings include significant fragmentation, deformity, and sclerosis with mixed rarefaction. Osteochondral injury has also been reported along the first metatarsal head articular surface (Figure 17-29). It is considered to be a precursor to hallux rigidus.147,148 There also is a high prevalence of osteochondral lesions in patients undergoing surgical hallux valgus correction.149 Sesamoids (First Metatarsophalangeal Joint) Renander150 was the first to report osteonecrosis of a first metatarsal sesamoid, which he described as an osteochondropathy. Osteonecrosis of the hallucal sesamoid bones is a relatively rare disorder, with unclear prevalence.151,152 This true osteonecrosis occurs most commonly in young adult women (second and third decades of life),153,154 with a precipitating history of minor forefoot trauma.150–152,155,156 There is no agreement on which sesamoid is most frequently affected. Some authors claim that it afflicts both equally157–159; some argue that greater incidence is in the tibial

sesamoid160 while others report a higher rate of fibular injury.153,161 According to Jahss,162 radiographic evidence of sesamoid osteonecrosis may not be evident until 9 to 12 months after the initial onset of symptoms; however, Ogata et al.163 reported four cases that all appeared within 6 months. When visible, findings include fragmentation, mottling (acute osteopenia), irregular shape, and geode formation, followed by sclerosis, collapse, flattening and widening (Figure 17-30).151,153,163 Radiographs may be difficult to interpret due to the superimposition of other bone structures in most views.153 The sesamoid axial view is best for isolating the sesamoids and usually shows fragmentation of the affected bone.164,165 The three-phase bone scan is useful in diagnosing true osteonecrosis, when diagnosis is unclear; however, MRI is a more specific modality to diagnose sesamoid disorders.164 At present, MRI is the most sensitive noninvasive diagnostic method available, enabling differentiation from other sesamoid pathologies.153,166 REFERENCES   1. Resnick D, Kransdorf M. Bone and Joint Imaging. 3rd ed. Philadelphia, PA: WB Saunders; 2004.   2. Frost A, Roach R. Osteochondral injuries of the foot and ankle. Sports Med Arthrosc Rev. 2009;17(2):87.   3. Stedman’s Medical Dictionary. 25th ed. Baltimore, MD: Williams & Wilkins; 1990.   4. Solomon, L. Mechanisms of idiopathic osteonecrosis. Orthop Clin North Am. 1985;16:655.   5. Lafforgue P. Pathophysiology and natural history of avascular necrosis of bone. Joint Bone Spine. 2006;73:500.   6. Buchan CA, Pearce DH, Lau J, et al. Imaging of postoperative avascular necrosis of the ankle and foot. Semin Musculoskelet Radiol.

2012;16:192.   7. Resnick D, Niwayama G. Diagnosis of Bone and Joint Disorders. Philadelphia, PA: WB Saunders; 1981.   8. Boettcher WG, Bonfiglio M, Hamilton HH, et al. Nontraumatic necrosis of the femoral head. J Bone Joint Surg Am. 1970;52:312.   9. Chiodo CP, Herbst SA. Osteonecrosis of the talus. Foot Ankle Clin. 2004;9:745.  10. Boskey AL, Raggio CL, Bullough PG, et al. Changes in the bone tissue lipids in persons with steroid and alcohol induced osteonecrosis. Clin Orthop. 1983;172:289.  11. Edeiken J, Dalinka M, Karasick D. Edeiken’s Roentgen Diagnosis of Disease of Bone. 4th ed. Baltimore, MD: William & Wilkins; 1990.  12. Brower AC. The osteochondroses. Orthop Clin North Am. 1983;14:99.  13. Ozonoff MB. Pediatric Orthopedic Radiology. Philadelphia, PA: WB Saunders; 1979.  14. Iver RS, Thapa MM. MR imaging of the paediatric foot and ankle. Pediatr Radiol. 2013;43(1):S107.  15. Hawkins LG. Fractures of the neck of the talus. J Bone Joint Surg Am. 1970;52:991.  16. Bornstein A, Plate E. Über chronische gelenkveränderungen, entstandendurch presslufterkrankung. Fortschr Geb Roentgenstr Nuklearmed. 1911;18:197.  17. Kawashima M, Torisu T, Hayashi K, et al. Pathological review of osteonecrosis in divers. Clin Orthop. 1978;130:107.  18. Heard JL, Schneider CS. Radiographic findings in commercial divers. Clin Orthop. 1978;130:179.

 19. Amstutz HC. The hip in Gaucher’s disease. Clin Orthop. 1973;90:83.  20. Rourke JA, Weslin JD. Gaucher’s disease. AJR Am J Roentgenol. 1965;94:621.  21. Williams WJ, Beutler E, Erslev AJ, et al. Hematology. New York, NY: McGraw-Hill; 1977.  22. Piggs LW. Bone and joint lesions in sickle cell disease. Clin Orthop. 1967;52:119.  23. Kelley K, Buchanan GR. Acute infarction of long bones in children with sickle cell anemia. J Pediatr. 1982;101:170.  24. Revell PA. Pathology of Bone. Berlin, Germany: Springer-Verlag; 1986.  25. Koulouris G, Morrison WB. Foot and ankle disorders: radiographic signs. Semin Roentgenol. 2005;40:358.  26. Siffert RS. Classification of the osteochondroses. Clin Orthop. 1981;158:10.  27. Katz JF. Nonarticular osteochondroses. Clin Orthop Relat Res. 1981;158:70.  28. Ralph BG, Barrett J, Kenyhercz C, et al. Iselin’s disease: a case presentation of nonunion and review of the differential diagnosis. J Foot Ankle Surg. 1999;38(6):409.  29. O’Loughlin PF, Heyworth BE, Kennedy JG. Current concepts in the diagnosis and treatment of osteochondral lesions of the ankle. Am J Sports Med. 2010;38:392.  30. Anderson IF, Crichton KJ, Grattan-Smith, et al. Osteochondral fractures of the dome of the talus. J Bone Joint Surg Am. 1989;71:1143.  31. Pettine KA, Morrey BF. Osteochondral fractures of the talus: a longterm follow-up. J Bone Joint Surg Br. 1989;69:89.

 32. Pritsch M, Horoshovski H, Farine L. Arthroscopic treatment of osteochondral lesions of the talus. J Bone Joint Surg Am. 1986;68:862.  33. Flick AB, Gould N. Osteochondritis dissecans of the talus (transchondral fractures of the talus): review of the literature and new surgical approach for medial dome lesions. Foot Ankle. 1985;5(4):165.  34. Schenck RC, Goodnight JM. Current concepts review: osteochondritis dissecans. J Bone Joint Surg Am. 1996;78(3):439.  35. Tachdjian MO. Pediatric Orthopedics. Philadelphia, PA: WB Saunders; 1990.  36. Davies AM, Cassar-Pullicino VN. Demonstration of osteochondritis dissecans of the talus by coronal computed tomographic arthrography. BJR. 1989;62:1050.  37. DeSmet AA, Fisher DR, Burnstein MI, et al. Value of MR imaging in stating osteochondral lesions of the talus (osteochondritis dissecans): results in 14 patients. AJR Am J Roentgenol. 1990;154:555.  38. Pavelka K. Osteonecrosis. Baillieres Best Pract Res Clin Rheumatol. 2000;14(2):399.  39. Jacobs BW. Early recognition of osteochondrosis of capital epiphysis of femur. JAMA. 1960;172:527.  40. Caffey J. The early roentgenographic changes in essential coxa plana: their significance in pathogenesis. AJR Am J Roentgenol. 1968;103:620.  41. Kenzora JE, Glimcher MJ. Pathogenesis of idiopathic osteonecrosis: the ubiquitous crescent sign. Orthop Clin North Am. 1985;16:681.  42. Lee GC, Khoury V, Steinberg D, et al. How do radiologists evaluate osteonecrosis? Skeletal Radiol. 2014;43:607.  43. Glickstein MF, Benk DL, Schiebler Ml, et al. Avascular necrosis versus other diseases of the hip: sensitivity of MRI imaging. Radiology. 1988;169:213.

 44. Gholve PA, Screr DM, Khakharia S, et al. Osgood Schlatter syndrome. Curr Opin Pediatr. 2007;19:44.  45. Venes D. Taber’s Cyclopedic Medical Dictionary. 8th ed. Philadelphia, PA: FA Davis; 2001.  46. Prentice WE. Arnheim’s Principles of Athletic Training: A CompetencyBased Approach. 12th ed. New York, NY: McGraw-Hill; 2006.  47. Ogden JA, Southwick WD. Osgood Schlatter’s disease and tibial tubercle development. Clin Orthop. 1976;116:180.  48. Bowers KD Jr. Patellar tendon avulsion as a complication of Osgood Schlatter’s disease. Am J Sports Med. 1981;9:356.  49. Frush TJ, Linderfield TN. Peri-epiphyseal and overuse injuries in adolescent athletes. Sports Health. 2009;1(3):201.  50. Hirano A, Fukubayashi T, Ishii T, et al. Magnetic resonance imaging of Osgood-Schlatter disease: the course of the disease. Skeletal Radiol. 2002;31:334.  51. Blankstein A, Cohen I, Heim M, et al. Ultrasonography as a diagnostic modality in Osgood-Schlatter disease: a clinical study and review of the literature. Arch Orthop Trauma Surg. 2001;121:536.  52. Golding JSR, McNeil-Smith JDG. Observations on the etiology of tibia vara. J Bone Joint Surg Br. 1963;45:320.  53. Blount WP. Tibia vara. J Bone Joint Surg. 1937;19:1.  54. Langenskiöld A. Tibia vara: osteochondrosis deformans tibiae. Clin Orthop. 1981;158:77.  55. Bathfield CA, Beighton PH. Blount disease. Clin Orthop. 1978;135:29.  56. Siffert RS, Arkin AM. Post-traumatic aseptic necrosis of the distal tibial epiphysis: report of a case. J Bone Joint Surg Am. 1950;32:691.

 57. Schuh A, Hönle W, Schuh R. Osteochondrosis dissecans of the medial malleolus: a case report. Arch Orthop Trauma Surg. 2009;129:1133.  58. Bui-Mansfield LT, Lenchik L, Rogers LF. Osteochondritis dissecans of the tarsal navicular bone: imaging findings in four patients. J Comput Assist Tomogr. 2000;24:744.  59. Yochum TR, Rowe LJ. Essentials of Skeletal Radiology. 3rd ed. Baltimore, MD: Lippincott Williams & Wilkins; 2005.  60. Miller AN, Prasarn MI, Dyke JP, et al. Quantitative assessment of the vascularity of the talus with gadolinium-enhanced magnetic resonance imaging. J Bone Joint Surg Am. 2010;93:116.  61. Pearce DH, Mongiardi CN, Fornasier VL, et al. Avascular necrosis of the talus: a pictorial review. Radiographics. 2004;25:399.  62. Sierra A, Ptochen EJ, Moore J, et al. High-field magnetic-resonance imaging of aseptic necrosis of the talus. J Bone Joint Surg Am. 1986;68:927.  63. Jones CK, Nunley JA. Osteonecrosis of the lateral aspect of the talar dome after triple arthrodesis: a report of three cases. J Bone Joint Surg Am. 1999;81(8):1165.  64. Hermus JPS. Osteonecrosis of the talus after talonavicular arthrodesis: a case report and review of the literature. J Foot Ankle Surg. 2011;50(3):343.  65. Thordarson DB, Triffon MJ, Terk MR. Magnetic resonance imaging to detect avascular necrosis after open reduction and internal fixation of talar neck fractures. Foot Ankle Int. 1996;17:742.  66. Henderson R. Postraumatic necrosis of the talus: the Hawkins sign versus magnetic resonance imaging. J Orthop Trauma. 1991;5:96.  67. Canale ST, Kelly FB Jr. Fractures of the neck of the talus: long-term evaluation of seventy-one cases. J Bone Joint Surg Am. 1978;60:143.  68. Tehranzadeh J, Stuffman E, Ross SDK. Partial Hawkins sign in fractures of the talus: a report of three cases. AJR Am J Roentgenol.

2003;181:1559.  69. Babu N, Schuberth JM. Partial avascular necrosis after talar neck fracture. Foot Ankle Int. 2010;31(9):777.  70. Chew KTL, Tay E, Wong YS. Osteochondral lesions of the talus. Ann Acad Med Singapore. 2008;37:63.  71. Van Buecken K, Barrack RL, Alexander AH, et al. Arthroscopic treatment of transchondral talar dome fractures. Am J Sports Med. 1989;17:350.  72. O’Farrel TA, Costello BG. Osteochondritis dissecans of the talus. J Bone Joint Surg Br. 1982;64:494.  73. Letts M, Davidson D, Ahmer A. Osteochondritis dissecans of the talus in children. J Pediatr Orthop. 2003;23:617.  74. Stone JW. Osteochondral lesions of the talar dome. J Am Acad Orthop Surg. 1996;4:63.  75. Verhagen RA, Maas M, Dijkgraaf MG, et al. Prospective study on diagnostic strategies in osteochondral lesions of the talus: is MRI superior to helical CT? J Bone Joint Surg Br. 2005;87:41.  76. Loomer R, Fisher C, Lloyd-Smith R, et al. Osteochondral lesions of the talus. Am J Sports Med. 1993;21(1):13.  77. Dipaola J, Nelson DW, Colville MR. Characterizing MR Lesions by magnetic resonance imaging. Arthroscopy. 1991;7:101.  78. Mintz DN, Tashjian GS, Connell DA, et al. Osteochondral lesions of the talus: a new magnetic resonance grading system with arthroscopic correlation. Arthroscopy. 2003;19(4):353.  79. Ferkel RD, Flannigan BD, Elkins BS. Magnetic resonance imaging of the foot and ankle: correlation of normal anatomy with pathologic conditions. Foot Ankle. 1991;11(5):289.

 80. Urquhart MW, Mont MA, Michelson JD, et al. Osteonecrosis of the talus: treatment by hindfoot fusion. Foot Ankle. 1996;17:275.  81. Berndt Al, Harty M. Transchondral fractures (osteochondritis dissecans) of the talus. J Bone Joint Surg Am. 1959;41:988.  82. Clanton TO, DeLee JC. Osteochondritis dissecans: history, pathophysiology and current treatment concepts. Clin Orthop Relat Res. 1982;167:50.  83. Hepple S, Winson IG, Glew D. Osteochondral lesions of the talus: a revised classification. Foot Ankle Int. 1999;20(12):789.  84. Taranow WS, Bisignani GA, Towers JD, et al. Retrograde drilling of osteochondral lesions of the medial talar dome. Foot Ankle Int. 1999;20(8):474.  85. Elias I, Jung JW, Raikin SM, et al. Osteochondral lesions of the talus: change in MRI findings over time in talar lesions without operative intervention and implications for staging systems. Foot Ankle Int. 2006;27(3):157.  86. Scharfbillig RW, Jones S, Scutter SD. Sever’s disease: what does the literature really tell us? J Am Podiatr Med Assoc. 2008;98:212.  87. Sever JW. Apophysitis of the os calcis. NY State J Med. 1912;95:1025.  88. Heneghan MA, Wallace T. Heel pain due to retrocalcaneal bursitisradiographic diagnosis (with a historical footnote on Sever’s disease). Pediatr Radiol. 1985;15:119.  89. Micheli LJ, Ireland ML. Prevention and management of calcaneal apophysitis in children: an overuse syndrome. J Pediatr Orthop. 1987;7:34.  90. Forrester DM, Kricun ME, Kerr R. Imaging of the Foot and Ankle. Rockville, MD: Aspen; 1988.  91. Shopfner CE, Coin CG. Effect of weight bearing on the appearance and development of the secondary calcaneal epiphysis. Radiology. 1966;86:201.

 92. Perhamre S, Lazowska D, Papageorgiou S, et al. Sever’s injury: a clinical diagnosis. J Am Podiatr Med Assoc. 2013;103(5):361.  93. Ogden JA, Ganey TM, Hill JD, et al. Sever’s injury: a stress fracture of the immature calcaneal metaphysis. J Pediatr Orthop. 2004;24:488.  94. Kose O. Do we really need radiographic assessment of the diagnosis of non-specific heel pain (calcaneal apophysitis) in children? Skeletal Radiol. 2010;39:359.  95. Madden CC, Mellion MB. Sever’s disease and other causes of heel pain in adolescents. Clin Sports Med. 1992;11:713.  96. Bartold S. Heel pain in young athletes. Aust Podiatr. 1993;27:103.  97. Gillespie H. Osteochondroses and apophyseal injuries of the foot in the young athlete. Curr Sports Med Rep. 2010;9(5):265.  98. Patel CV. The foot and ankle: MR imaging of uniquely pediatric disorders. Magn Reson Imaging Clin N Am. 2009;17:539.  99. Borges JL, Guile JT, Bowen JR. Kohler’s bone disease of the tarsal navicular. J Pediatr Orthop. 1995;15:596. 100. Ferguson AB, Gingrich RM. The normal and the abnormal calcaneal apophysis and tarsal navicular. Clin Orthop. 1957;10:87. 101. McCauley MB, Kahn PC. Osteochondritis of the tarsal navicular. Radiology. 1978;22:332. 102. Weston WJ. Köhler’s disease of the tarsal scaphoid. Aust Radiol. 1978;22:332. 103. Mohiuddin T, Jennison T, Damany D. Müller–Weiss disease: review of current knowledge. Foot Ankle Surg. 2014;20:79. 104. Haller J, Sartoris DJ, Resnick D, et al. Spontaneous osteonecrosis of the tarsal navicular in adults: imaging findings. AJR Am J Roentgenol.

1988;151:355. 105. Maceira E, Rochera R. Müller–Weiss disease: clinical and biomechanical features. Foot Ankle Clin. 2004;9:105. 106. Brailsford JF. Osteochondritis of the adult tarsal navicular. J Bone Joint Surg. 1939;21(1):111. 107. Sizensky JA, Marks RM. Imaging of the navicular. Foot Ankle Clin. 2004;9(1):181. 108. Ozturk E, Sirvanci M, Mutlu H, et al. Osteochondritis dissecans of the tarsal navicular. Foot Ankle Int. 2008;29:442. 109. Beil FT, Bruns J, Habermann CR, et al. Osteochondritis dissecans of the tarsal navicular bone: a case report. J Am Podiatr Med Assoc. 2012;102(4):338. 110. DiGiovanni CW, Patel A, Calfee R, et al. Osteonecrosis in the foot. J Am Acad Orthop Surg. 2007;15(4):208. 111. Buchman J. Osteochondritis of the internal cuneiform. J Bone Joint Surg Am. 1933;15:225. 112. Garcia-Mata S. Avascular necrosis of the intermediate cuneiform bone in a child: a very rare cause of limp in a child. A variant of the normality? J Pediatr Orthop. 2013;22:255. 113. O’Donoghue AF, Donohue ES, Zimmerman WW. Bilateral osteochondritis of the tarsal navicular and the first cuneiform: a case report. J Bone Joint Surg Am. 1948;30:780. 114. Leeson MC, Weiner DS. Osteochondrosis of the tarsal cuneiforms. Clin Orthop. 1985;196:260. 115. Kose O, Demiralp B, Oto M, et al. An unusual cause of foot pain in a child: osteochondrosis of the intermediate cuneiform. J Foot Ankle Surg. 2009;48(4):474.

116. Mubarak SJ. Osteochondrosis of the lateral cuneiform: another cause of a limp in a child: a case report. J Bone Joint Surg Am. 1992;74:285. 117. Hicks BTG. Case report—osteochondritis of the tarsal second cuneiform bone. Br J Radiol. 1953;26:214. 118. Watmough PJ, Tselentakis G, Day JB. Avascular necrosis of the intermediate cuneiform bone. J Pediatr Orthop B. 2003;12:402. 119. Chew JT, Low CK, Mitra AK. A case report on bilateral avascular necrosis of the medial cuneiform. Ann Acad Med Singapore. 1995;24:453. 120. Gago BM, Garcia CH, Chaves Gomez-Orihuela JM, et al. Osteochondritis of the first cuneiform: a case report of a bilateral case. Rev Orthop Traumatol. 1988;32:139. 121. Haboush EJ. Bilateral disease of the internal cuneiform bone with an associated disease of the right scaphoid bone (Koehler’s). JAMA. 1933;100:41. 122. Meilstrup DB. Osteochondritis of the internal cuneiform, bilateral: case report. AJR Am J Roentgenol. 1947;58:329. 123. Scarfi G, Landi F. A rare case of bilateral osteochondrosis of the first cuneiform bone. Arch Putti Chir Organi Mov. 1983;33:447. 124. Vaquero Martin J, Vicente-Herrera E, Pereiro De Lamo J, et al. Osteochondritis of the medial cuneiform. J Pediatr Orthop B. 1999;8:69. 125. Zimberg J, Levitt JCA, Brahim F. Osteochondrosis of the medial cuneiform: a case report. J Am Podiatr Med Assoc. 1985;75(10):538. 126. Canale ST, Williams KD. Iselin’s disease. J Pediatr Orthop. 1992;12:90. 127. Dameron TB Jr. Fractures and anatomical variations of the proximal portion of the fifth metatarsal. J Bone Joint Surg Am. 1975;57:788. 128. Canale ST, Richardson DR. Osteochondroses and related problems of

the foot and ankle. In: DeLee JC, Drez D Jr, Miller MD, eds. DeLee and Drez’s Orthopaedic Sports Medicine: Principles and Practice. 3rd ed. Philadelphia, PA: WB Saunders; 2010:2142. 129. Wilson JC, Rodenberg RE. Apophysitis of the lower extremities. Contemporary Pediatrics. 2011. http://contemporarypediatrics.modernmedicine.com/contemporarypediatrics/news/modernmedicine/modern-medicine-featurearticles/apophysitis-lower-extre?id=&sk=&date. Accessed on February 2, 2014. 130. Lehman RC, Gregg JR, Torg E. Iselin’s disease. Am J Sports Med. 1986;14:494. 131. Anderson LD. Injuries of the forefoot. Clin Orthop. 1977;122:18. 132. Freiberg AH. Infraction of the second metatarsal bone-a typical injury. Surg Gynecol Obstet. 1914;19:191. 133. Carmont MR, Rees RJ, Blundell CM. Current concepts review: Freiberg’s disease. Foot Ankle Int. 2009;30:167. 134. Gauthier G, Elbaz R. Freiberg’s infraction: a subchondral bone fatigue fracture. Clin Orthop. 1979;142:93. 135. Braddock GTF. Experimental epiphyseal injury and Freiberg’s disease. J Bone Joint Surg Br. 1959;41:154. 136. Smillie IS. Treatment of Freiberg’s infarction. Proc R Soc Med. 1967;60:29. 137. Ahmad J, Raikin SM. Osteonecrosis of the second metatarsal head, navicular and talus. Curr Opin Orthop. 2006;17:103. 138. Sproul J, Klaaren H, Mannarino F. Surgical treatment of Freiberg’s infraction in athletes. Am J Sports Med. 1993;21:381. 139. Torriani M, Thomas BJ, Bredella MA. MRI of metatarsal head subchondral fractures in patients with forefoot pain. AJR Am J Roentgenol.

2008;190:570–575. 140. Malal JJG, Shaw-Dunn J, Kumar CS. Blood supply to the first metatarsal head and vessels at risk with a chevron osteotomy. J Bone Joint Surg Am. 2007;89(9):2018. 141. Prasad MG, Shankar NS. Clinical results of Keller’s arthroplasty. Foot. 1998;8:223. 142. Brosky TA II, Menke CRD, Xenos D. Reconstruction of the first metatarsophalangeal joint following post-cheilectomy avascular necrosis of the first metatarsal head: a case report. J Foot Ankle Surg. 2009;48(1):61. 143. Meier PJ, Kenzora JE. The risks and benefits of distal first metatarsal osteotomies. Foot Ankle. 1985;6:7. 144. Banks AS. Avascular necrosis of the first metatarsal head. J Am Podiatr Med Assoc. 1999;89(9):441. 145. Edwards WH. Avascular necrosis of the first metatarsal head. Foot Ankle Clin. 2005;10:117. 146. Shortt CP. Magnetic resonance imaging of the midfoot and forefoot: normal variants and pitfalls. Magn Reson Imaging Clin N Am. 2010;18(4):707. 147. Street CC, Shereff MJ. Traumatic osteochondral defect of the first metatarsal head: a case report. Am J Orthop (Belle Mead NJ). 1999;28(10):584–586. 148. Kim YS, Park EH, Lee HJ, et al. Clinical comparison of the osteochondral autograft transfer system and subchondral drilling in osteochondral defects of the first metatarsal head. Am J Sports Med. 2012;40:1824. 149. Jastifer JR, Coughlin MJ, Doty JF, et al. Osteochondral lesions in surgically treated hallux valgus. Foot Ankle Int. 2014;35(7):643–649. 150. Renander A. Two cases of typical osteochondropathy of the medial

sesamoid bone of the first metatarsal. Acta Radiol. 1924;3(6):521. 151. Williams G, Kenyon P, Fischer B, et al. An atypical presentation of hallucial sesamoid avascular necrosis: a case report. J Foot Ankle Surg. 2009;48 (2):203. 152. Keating S, Fisher D, Keating D. Avascular necrosis of an accessory sesamoid of the foot: a case report. J Am Podiatr Med Assoc. 1987;77(11):612. 153. Pinto RR, Freitas D, Massada M, et al. Hallux sesamoid osteonecrosis associated to ballet: literature review. Rev Port Ortop Traum. 2010;18(4):429. 154. Wuelker N, Wirth CJ. The great toe sesamoids. Foot Ankle Surg. 1996;2:167. 155. Toussirot E, Jeunet L, Michel F, et al. Avascular necrosis of the hallucal sesamoids update with reference to two case-reports. Joint Bone Spine. 2003;70(4):307. 156. Fleischli J, Cheleuitte E. Avascular necrosis of the hallucal sesamoids. J Foot Ankle Surg. 1995;34(4):358. 157. Sammarco HG. Dance injuries. Contemp Orthop. 1984;8:15. 158. Jahss MS. Disorders of the Foot and Ankle: Medical and Surgical Management. 2nd ed. Philadelphia, PA: WB Saunders; 1991:1062–1075. 159. Leventen EO. Sesamoid disorders and treatment: an update. Clin Orthop. 1991;269:236. 160. Kliman ME, Fross AE, Pritzker KP, et al. Osteochondritis of the hallux sesamoid bones. Foot Ankle. 1983;3:220. 161. Karasick D. Disorders of the hallux sesamoid complex. Skeletal Radiol. 1998;27:411. 162. Jahss MH. The sesamoids of the hallux. Clin Orthop Relat Res.

1981;157:88–97. 163. Ogata K, Sugioka Y, Urano Y, et al. Idiopathic osteonecrosis of the first metatarsal sesamoid. Skeletal Radiol. 1986;15:141. 164. Cohen BE. Hallux sesamoid disorders. Foot Ankle Clin N Am. 2009;14:91. 165. Grace DL. Sesamoid problems. Foot Ankle Clin. 2000;5(3):609. 166. Garrido IM, Bosch MN, González MS, et al. Osteochondritis of the hallux sesamoid bones. Foot Ankle Surg. 2008;14:175. 167. Breck LW. An Atlas of the Osteochondroses. Springfield, IL: CC Thomas; 1971. SUGGESTED READINGS Ashman CJ, Klecker RJ, Yu JS. Forefoot pain involving the metatarsal region: differential with MR Imaging. Radiographics. 2005;21(6): 1425– 1440. Baastrup CI. Os vesalianum tarsi and fracture of tuberositas ossis metatarsi V. Acta Radiol. 1922;1:334–351. Baravarian B, Zgonis T, Lowery C. Use of Cobb procedure in the treatment of posterior tibial tendon dysfunction. Clin Podiatr Med Surg. 2002;19:371– 389. Beaman FD, Bancroft LW, Peterson JJ, et al. Imaging characteristic of bone graft materials. Radiographics. 2006;26(2):373–388. Bucholz RW, Heckman JD, Court-Brown CM, et al. Rockwood and Green’s Fractures in Adults. 6th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2006. Cheng MS, Ferkel RD, Applegate GR. Osteochondral lesions of the talus: a radiologic and surgical comparison. Presented at the Annual Meeting of the Academy of Orthopedic Surgeons; February 16–21, 1995; New Orleans, LA.

Choi YS, Lee KT, Kang HS, et al. MI iImaging findings of painful type II navicular bone: correlation with surgical and pathologic studies. Korean J. 2004;5(4):274–279. Chowchuen P, Resnick D. Stress fractures of the metatarsal heads. Skeletal Radiol. 1998;27:22–25. Conti S, Mickelson J, Jahss M. Clinical significance of magnetic resonance imaging in preoperative planning for reconstruction of posterior tibial tendon ruptures. Foot Ankle. 1992;13:208–214. Delanois RE, Mont MA. Atraumatic osteonecrosis of the talus. J Bone Joint Surg Am. 1998;80A(4):529–536. Demeyere N, De Maeseneer M, Osteaux. Quiz case: symptomatic type II accessory navicular. Eur J Radiol. 2001;37:60–63. Duran-Stanton MA. Foot pain for 5 years, and an abnormal radiograph. Diagnostic Imaging Review. JAAPA. 2009;22(11):58–59. Ferkel RD, Flannigan BD, Elkins BS. Magnetic resonance imaging of the foot and ankle: correlation of normal anatomy with pathologic conditions. Foot Ankle. 1991;11(5):289–305. Geideman WM, Johnson JE. Posterior tibial tendon dysfunction. J Orthop Sports Phys Ther. 2000;20(2):68–77. Hain SF, Fogelman I. Nuclear medicine studies in metabolic bone disease. Semin Musculoskelet Radiol. 2002;6(4):323–329. Kvist MH, Heinonen OJ. Calcaneal apophysitis (Sever’s disease) a common cause of heel pain in young athletes. Scand J Med Sci Sports. 1991;1:235. Lawson JP, Ogden JA, Sella E, et al. The painful accessory navicular. Skeletal Radiol. 1984;13:250–262. Lewin P. Apophysitis of the os calcis. Surg Gynecol Obstet. 1925;41:578. Lusse S, Claassen H, Gehrke T, et al. Evaluation of water content by spatially

resolved transverse relaxation times of human articular cartilage. Magn Reson Imaging. 2000;18(4):423–430. Mazieres B, Arlet J, Boussaton M, et al. Assessment of intramedullary pressure versus bone scintigraphy in the diagnosis of osteonecrosis of femoral head. In: Arlet J, Mazieres B, eds. Bone Circulation and Bone Necroses. Berlin, Germany: Springer-Verlag; 1990:264–266. McCrea JD. Pediatric Orthopedics of the Lower Extremity. Mount Kisco, NY: Futura; 1985:249–251. Mesgarzadeh M, Sapega AA, Bonakdarpour A, et al. Osteochondritis dissecans: analysis of mechanical stability with radiography, scintigraphy, and MR imaging. Radiology. 1987;165:775. Miller TT, Staron RB, Feldman F, et al. The symptomatic accessory tarsal navicular bone: assessment with MR imaging. Radiology. 1995;195:849–853. Mink JH. Tendons. In: Deutsch AL, Mink JH, Kerr R, et al. MRI of the Foot and Ankle. New York, NY: Raven Press; 1992;135–172. Mont Ma, Ulrich SD, Seyler JM, et al. Bone scanning of limited value for diagnosis of symptomatic oligofocal and multifocal osteonecrosis. J Rheumatol. 2008;35(8):1629–1634. Mosel LD, Kat E, Voyvodic F. Imaging of the symptomatic type II accessory navicular bone. Australas Radiol. 2004;48:267–271. Mosher TJ. MRI of osteochondral injuries of the knee and ankle in the athlete. Clin Sports Med. 2006;2006:25:843–866. Mostofi SB. Fracture Classifications in Clinical Practice. 2nd ed. London, England: Springer; 2013. Mubarak SJ, Carroll NC. Familial osteochondritis dissecans of the knee. Clin Orthop Relat Res. 1979;140:131–136. Ogden JA, Southwick WO. Osgood –—Schlatter’s disease and tibial tuberosity development. Clin Orthop Relat Res. 1976;116:180–189.

Pick MP. Familial osteochondritis dissecans. J Bone Joint Surg Br. 1955;37(1):142–145. Rhoad RC, Davidson RS, Heyman S, et al. Pretreatment bone scan in SCFE: a predictor of ischemia and avascular necrosis. J Pediatr Orthop. 1999;19:164–168. Roden S, Tillegard P, Unanderscharin L. Osteochondritis dissecans and similar lesions of the talus: a report of fifty-five cases with special reference to etiology and treatment. Acta Orthop Scand. 1953;23(1):51–66. Rosenberg ZS, Beltran J, Bencardino JT. From the RSNA refresher courses. Radiology Society of North America. MR Imaging of the ankle and foot. Radiographics. 2000;20:S153–S179. Rosenblat M, Bauchot G. Sever’s disease, a new therapeutic approach in a series of 68 athletes. J Traumatol Sport. 1994;11:90. Sangeorzan BJ, Bernischke SK, Mosca V, et al. Displaced intra-articular fractures of the tarsal navicular. J Bone Joint Surg Am. 1989;71:1504–1510. Santopietro FJ. Foot and foot-related injuries in the young athlete. Clin Sports Med. 1988;7:563. Schwartz B, Jay RM, Schoenhaus HB. Apophysitis of the fifth metatarsal base. JAMA. 1991;81:128. Shafa MH, Fernandez- Uloa M, Rost RC. Diagnosis of aseptic necrosis of the talus by bone scintigraphy. Clin Nucl Med. 1983;2:50–53. Stanittski CL. Combating overuse injuries: a focus on children and adolescents. Phys Sports Med. 1993;21:87. Stavinoha RR, Scott W. Osteonecrosis of the tarsal navicular in two adolescent soccer players. Clin J Sport Med. 1998;8:136–138. Sugimoto H, Okubo RS, Ohsawa T. Chemical shift and the double-line sign in MRI of early femoral avascular necrosis. J Comput Assist Tomogr.

1992;16:727–730. Tax HR. Podopediatrics. 2nd ed. Baltimore, MD: Williams and Wilkins; 1985:201–202. Torg JS, Balduini FC, Zelko RR. Fractures of the base of the fifth metatarsal distal to tuberosity: classification and guidelines for non-surgical and surgical management. J Bone Joint Surg Am. 1984:66(2):209–214. Urman M, Ammann W, Sisler J, et al. The role of bone scintigraphy in the evaluation of the talar dome fractures. J Nucl Med. 1991;32:2241–2244. Volpon JB, de Carvalho FG. Calcaneal apophysitis: a quantitative radiographic evaluation of the secondary ossification center. Arch Orthop Trauma Surg. 2002;122:338–341.

18 Bone Infection MARIE WILLIAMS AND ROBERT A. CHRISTMAN The radiographic presentation of infection depends on several factors, including the age of patient, route of infection, type of organism involved, anatomic location, and nature of the disease process. Early diagnosis of infection, especially involving bone, allows appropriate treatment and diminishes the risk of long-term sequelae and complications. To thoroughly understand the etiology of bone infection, the clinician must first have a basic understanding of bone pathophysiology and how infecting organisms invade bone and joints. When the vascular supply to bone is disrupted, physiologic changes take place within the Haversian and Volkmann systems affecting osteoblastic and osteoclastic function, which are responsible for normal bone function. Correlating clinical findings with radiographic findings can give the practitioner a greater comprehension of the disease process. DEFINITION OF TERMS Infectious periostitis is the term used for infection that invades the periosteum only and does not involve the cortex and bone marrow. With infectious (or suppurative) periostitis the changes are subtle and may be identified by a periosteal reaction (Figure 18-1). As the infection penetrates into the cortex but does not invade medullary bone, the term infectious osteitis is used (Figure 18-2). Once the infection involves both cortex and bone marrow, the more accurate term is osteomyelitis (Figure 18-3).1 It can be difficult to differentiate infectious osteitis from osteomyelitis radiographically, because a lag time of approximately 10–14 days separates the clinical presentation from the visible radiographic findings.1–3 That it because a reduction of bone density between 30% and 50% must occur before radiographic evidence is noted.1,3–6 For this reason, bone scintigraphy

has been valuable in early diagnosis of osteomyelitis if no other underlying pathology is present. Radiographically, the fascial planes of the plantar and dorsal midfoot and tarsus normally can be identified in the lateral view. As soft tissues become infiltrated with an infecting organism, the fascial planes disappear, which may be the initial radiographic sign of impending bone infection and should alert the practitioner to institute early treatment. Occasionally gas or air is seen in the affected area (Figures 18-4). Soft tissue infection of the digits appears radiographically as an increased soft tissue density and volume (Figure 18-5), which, unfortunately, is not specific for infection. If inappropriately treated or left untreated, however, soft tissue infection can lead to bone and/or joint infection.1–3,5,6 The term septic arthritis is used to describe infection of a joint. Joint infection erodes cartilage and decreases joint mobility.1,7 Radiographically, early septic arthritis appears similar to acute gouty arthritis, that is, significant increase in soft tissue volume and density with or without joint space widening or juxta-articular rarefaction (Figure 18-6). However, infection leads quickly to subchondral resorption and osteolysis. Septic arthritis can later lead to bony ankylosis (Figure 18-7). CLASSIFICATIONS OF OSTEOMYELITIS Osteomyelitis has been classified according to several methods: its clinical presentation and onset, the transmission route of the infecting organism, patient age, its anatomic type, and the infecting organism.8,9

FIGURE 18-1. Infectious periostitis. A subtle, linear periosteal reaction (arrows) can be seen along the lateral aspect of the third toe proximal phalanx. Note the associated soft tissue increase in volume and density affecting the third and fourth toes.

FIGURE 18-2. Infectious osteitis. Erosion (arrow) can be identified along the lateral aspect of the fifth toe proximal phalanx distal diametaphysis. No rarefaction or other lysis is seen radiographically that would suggest medullary involvement.

Clinical Presentation and Onset The clinical stages of osteomyelitis may be acute, subacute, or chronic (Table 18-1).10 Acute osteomyelitis is clinically identified by an insidious onset of pain localized to bone, with soft tissue swelling and erythema of the involved area. A patient may develop fever, malaise, irritability, and marked increase in pain. During the initial stage of infection, it is difficult to ascertain the infectious process radiographically. Soft tissue swelling and obliteration of fascial planes may be the only radiographic signs (Figure 18-8A).11 The term rarefaction is used to describe localized loss of bone density. It is one of the earliest findings of osteomyelitis (Figure 18-8B).1,2,4,6 Despite the fact that periostitis is considered one of the “classic” findings of osteomyelitis,12 with the exception of hematogenous osteomyelitis, periostitis is often absent, especially in the toes. It may, however, be seen with less virulent infectious processes (Figure 18-1). The term osteolysis applies to a more destructive process or resorption of bone and usually follows localized rarefaction (Figure 18-8C).

FIGURE 18-3. Osteomyelitis, fourth toe proximal phalanx. Infection

involves the entire bone, resulting in loss of its form. The term subacute osteomyelitis is used to describe a well-defined lytic lesion in bone that is caused by an infectious organism of low virulence.1,3 The walled-off abscess in bone known as Brodie abscess was first described by Sir Benjamin Brodie in 1832.13 A Brodie (bone) abscess can radiographically be differentiated from a benign bone cyst: the former appears as a geographic lytic area of bone with a dense sclerotic rim that fades peripherally (Figure 18-9).4 In contrast, the bone cyst will have a thin, well-defined sclerotic margin surrounding geographic destruction. The bone abscess may vary in size, from 1 to 4 cm in diameter, and is commonly found in metaphyseal bone.2

FIGURE 18-4. Soft tissue infection, gas gangrene. A: Dorsoplantar view. Air/gas is seen throughout the soft tissues of the forefoot. B: The lateral view shows loss of fascial planes dorsally as well as air/gas (arrowheads) in the soft tissues.

FIGURE 18-5. Soft tissue infection, second toe. A: Increased soft tissue volume and density affects the entire digit. B: Four weeks later, gas is evident (arrows). Chronic osteomyelitis develops when an acute infection is inappropriately treated or therapy has been inadequate. It has been defined as an infection that has been present for at least 6 weeks.14 During the chronic process, viable new bone is laid down around dead bone, developing a periosteal envelope that can surround the entire shaft. This covering is called an involucrum (Figure 18-10).1,3,6 Therefore, involucrum surrounds dead bone and may give the remodeled bone an irregular or jagged outline. Longstanding chronic osteomyelitis results in significant bone deformity (Figure 18-11). Devitalized bone that is detached from the surrounding bone and necrotic due to the infection is known as a sequestrum.15 It may appear sclerotic relative to the viable surrounding area if the latter is osteopenic (Figure 18-11). Sequestrum is one of the most important findings in the assessment of chronic osteomyelitis. The presence of necrotic bone represents active infection during the chronic stage of the disease.1,3,6 The term cloaca refers to a defect in the cortex that allows pus and nonviable

bone (sequestrum) to be expelled from the bone (Figure 18-12). Similarly, fistula or sinus tract is an opening that allows the sequestrum or pus to move through the soft tissues.

FIGURE 18-6. Septic arthritis, hallux interphalangeal joint. The primary finding is subchondral resorption (arrows). The joint space is slightly increased.

FIGURE 18-7. Ankylosis secondary to septic arthritis, first metatarsophalangeal joint. Significant joint space narrowing and subchondral resorption are seen along both sides of the first metatarsophalangeal joint. Bony union is occurring across the articulation. TABLE 18-1   Classification of Osteomyelitis Based on Clinical Stages Clinical Signs Radiographic Signs Acute Osteomyelitis No radiographic findings initially; early findings may include periostitis, Insidious onset with pain, malaise, and rarefaction, and increased soft tissue fever volume and density; eventually osteolysis Cellulitis Obliteration of fascial planes Subacute Osteomyelitis Low-grade pain with no systemic Well-defined lytic lesion in bone with signs; occasional pain around the dense sclerotic rim usually 1 to 4 cm affected area in diameter (Brodie abscess) Chronic Osteomyelitis Pain localized to the affected area, with soft tissue swelling and localized Bone malformation with involucrum, cellulitis; ulceration and draining; cloaca, and sequestrum sinus may be present

FIGURE 18-8. Progression of osteomyelitis, fifth toe proximal phalanx. A: Only finding is increased soft tissue volume and density. B: Significant rarefaction (arrow) is seen in the distal half of the proximal phalanx. Notice that the form of the phalanx is still intact. C: Osteolysis. Sclerosing osteomyelitis of Garré is a chronic form of osteomyelitis caused by organisms of low virulence. The disease is nonsuppurative and usually affects a single bone. Radiographically the bone appears sclerotic and shows marked cortical thickening with very little evidence of a draining sinus or osteolysis. This entity must be differentiated from Ewing sarcoma, which is a more destructive lesion showing scalloping of the periosteum.1,3,6,16 Route of Transmission Many authors have classified bone infection according to the route of contamination, which is important radiographically, because the findings may appear differently depending on the route.1,2,6,17 Different routes of contamination that cause osteomyelitis include osteomyelitis secondary to contiguous soft tissue; puncture wounds or implantation devices such as pins, wires, screws, or staples from surgery; postoperative wound infections; and hematogenous osteomyelitis.1 The Waldvogel classification is most familiar, and divides osteomyelitis into two modes of transmission: direct extension

and hematogenous.17 Osteomyelitis secondary to contiguous soft tissue or direct extension to bone is more common than hematogenous osteomyelitis in the lower extremity and foot.2 Direct extension osteomyelitis is caused by an organism invading bone from the outside via a portal of entry.4,6 As a result, infection first contacts the periosteum, and resides between it and the cortex (infectious periostitis). If left untreated, the infectious process erodes the underlying cortex (infectious osteitis). After the periosteum is invaded and the cortex is destroyed, the infection will reach the marrow cavity and is now a full-blown osteomyelitis.

FIGURE 18-9. Brodie abscess, first metatarsal. This is an 8-year-old boy, 11 months after metatarsus adductus surgery with a painful first metatarsal. A geographic, lytic lesion lies at the medial aspect of the first metatarsal proximal metaphysis (arrowhead). This lesion represents a bone abscess. Note that the epiphysis is spared and the lesion has a sclerotic rim that fades peripherally away from the epiphyseal plate. Initial radiographic findings associated with direct extension osteomyelitis, whether it be from contiguous soft tissue infection, puncture wound, or surgery, are increased soft tissue volume and density and obliteration of fascial planes. (The soft tissue radiographic abnormalities in hematogenous osteomyelitis occur later in its course.) As the infection invades the periosteum, it lifts the cortex and can stimulate formation of a periosteal reaction. If left untreated, it invades the cortex. As the infection traverses the Haversian and Volkmann canals, cortical erosion and rarefaction occur. Pus can infiltrate the medullary bone vascular supply as the bone marrow becomes affected; circulation to bone becomes sluggish, and osteolysis occurs.

FIGURE 18-10. Chronic osteomyelitis, involucrum. The distal tibia is grossly deformed; the collar of remodeled bone surrounding the infected bone is known as an involucrum. Osteomyelitis is difficult to differentiate from normal healing bone in the postoperative patient because a proliferative periosteal reaction may also occur in the latter (Figure 18-13). Early diagnosis is very important, however, to prevent an acute osteomyelitis or infectious periostitis from becoming a chronic problem. Other diagnostic modalities, such as white blood cell– labeled bone scintigraphy or magnetic resonance imaging (MRI), may help differentiate normal bone healing from infection. Increased osseous and/or cartilaginous destruction with radiolucency around an implant or prosthesis should alert the clinician to possible bone infection.1 A subtype of osteomyelitis from a contiguous focus of infection is whether or not there is vascular insufficiency.18,19 This is particularly important because of the fact that conditions such as diabetes fall in this category. The most common age group affected is that between 50 and 70 years of age. Osteomyelitis in this subtype usually develops from a localized infection or ulceration of the skin. Radiography of a patient with severe peripheral vascular disease and osteomyelitis may show minimal findings except for osteopenia caused by poor perfusion to bone in general. Osteomyelitis can be difficult to differentiate from Charcot neuropathic osteoarthropathy (Figure 18-14).20,21 Diabetic osteomyelitis is often associated with an ulcer or skin infection. Radiographs show increased soft tissue volume and density with or without gas or air in the tissue, depending on the type of organism present. Gasproducing organisms such as Clostridium and Bacteroides can lead to soft tissue necrosis and ultimately to gas gangrene. Obliteration of fascial planes is a radiographic sign associated with cellulitis. Periosteal new bone formation, rarefaction, subchondral resorption, and osteolysis can occur, as with exogenous osteomyelitis. Radiographic evaluation of osteomyelitis in the diabetic with underlying peripheral neuropathy and/or vascular disease can be confusing, because the radiographic features overlap (Figure 18-15). Furthermore, osteomyelitis and neuropathic osteoarthropathy can coexist

(Figure 18-16). Nuclear medicine and MRI play an important role in differentiating osteomyelitis from Charcot neuropathic osteoarthropathy and can be used if the question arises. The radiographic picture of Charcot neuropathic osteoarthropathy is described in Chapter 19; the diabetic foot is the focus of Chapter 22. Hematogenous osteomyelitis is an infection in which the bone is infected by blood-borne organisms that are deposited in medullary and metaphyseal regions of bone. It gives a somewhat different appearance from that of direct extension osteomyelitis. Metaphyseal bone is most often affected; it is more vascular and the sluggish, venous sinusoidal blood flow provides a good medium for bacterial growth, enhancing localization of the organisms in the metaphysis and marrow.22,23 This is especially true in children, in whom nutrient arteries are relatively large, and the branches to the marrow, cortex, and metaphysis are small end arteries or capillaries that are relatively stagnant.22 Intraosseous pressure rapidly increases with infection, leading to marked necrosis and demineralization. If the infection persists, exudate continues to expand and spread through the Volkmann and Haversian systems of bone. Blood flow is increasingly disrupted, leading to advanced necrosis, resorption, and possibly sclerosis. As the infection advances, it crosses the cortex, with ensuing periosteal proliferation. Subperiosteal abscess formation and subsequent irritation may lead to new bone formation or involucrum, if chronic.4,7,22 Subperiosteal proliferation may be noted as early as 5 to 7 days in children, whereas in adults it may appear in 10 to 14 days; it is usually a more common finding in children because the periosteum is loosely attached. In addition, large areas of dead bone (sequestra) may be surrounded by granulation tissue and walled-off from adjacent viable bone.

FIGURE 18-11. Chronic osteomyelitis, distal leg. A: Anteroposterior view. B: Lateral view. After many years, the periosteal remodeling (involucrum, white arrow) yields severe bone deformity. The black arrow identifies a probable sequestrum, oval-shaped and sclerotic. CT may be necessary to

document its separation within the infected bone.

FIGURE 18-12. Chronic, active osteomyelitis. Note the well-defined lytic lesion within the base of the fifth metatarsal. Also note the extensive periosteal thickening and remodeling (arrowheads), representing an involucrum. The involucrum surrounds or walls off the sequestra within the cyst like lesion. A small sequestrum (straight arrow) is exiting through a cloaca (curved arrow).

FIGURE 18-13. Normal postoperative periostitis. A periosteal reaction is seen along the lateral aspect (open arrows) of the first metatarsal proximal metadiaphysis. This radiograph was performed 13 weeks after a base wedge osteotomy with screw fixation. The screw was removed because it loosened. The periosteal reaction could represent infection, but this example proved to be an uninfected periostitis secondary to normal healing.

FIGURE 18-14. Osteomyelitis versus neuropathic osteopathy, second metatarsal head. This diabetic patient previously had amputation of the second toe because of severe vascular compromise. Subchondral resorption is now evident along the distal aspect of the metatarsal head. In most cases, diabetic osteopathy cannot be easily differentiated from osteomyelitis radiographically.

FIGURE 18-15. A diabetic patient with severe peripheral vascular disease. In the proper clinical setting, soft tissue edema, rarefaction, osteolysis, erosion, and periosteal new bone production all suggest osteomyelitis.

FIGURE 18-16. Mixed long-standing osteomyelitis and neuropathic osteopathy in a diabetic patient. Exuberant remodeling involves the first four metatarsals and adjacent proximal phalanges. The geographic, lytic lesion

surrounded by diffuse sclerosis in the fourth metatarsal distal diaphysis is a bone (Brodie) abscess (arrow). Patient Age Metaphyseal vascular anatomy is age dependent and subsequently influences the radiographic presentation of hematogenous osteomyelitis (Figure 18-17). Trueta22 has described three distinct patterns of osteomyelitis based on metaphyseal blood supply in infants, children, and adults. An infantile pattern is described in children less than 1 year of age. Metaphyseal vessels penetrate the epiphyseal growth plate in the infant to supply the epiphysis. Subsequently, metaphyseal infection may cross the growth plate, invade contiguous joint spaces, and result in sepsis. Neonatal osteomyelitis rapidly transgresses the growth plate and leads to septic arthritis, which can destroy the joint as well as bone.5,22,17 Involvement of the growth plate may lead to a decrease in the length of the affected limb. When the epiphyseal plate is spared, hyperemia may accelerate growth rate, with early maturation and closure of the growth plate.6 The juvenile pattern affects children from 1 year of age to closure of the physis. No vascular penetration of the growth plate occurs during this period; acting as a “barrier,” it confines the infection to the metaphyseal region (Figure 18-18). Infection may spread laterally, however, perforating the cortex and elevating the loosely adhered periosteum. The adult pattern is evident after growth plate closure. Infection can penetrate into subchondral bone and joint sepsis is more likely. Because the periosteum adheres more firmly to underlying bone in the adult, less periosteal reaction is seen than in the child. Anatomic Type Cierny et al.24 proposed a classification system that considers the bones anatomic nature, the host quality, as well as treatment and prognostic factors. This system, however, may not be as useful for the toes or other small bones.25 There are four anatomic stage types: medullary, superficial,

localized, and diffuse osteomyelitis. There is soft tissue compromise in both the medullary and superficial locations. The nidus of medullary osteomyelitis is endosteal; however, superficial osteomyelitis has a contiguous focus on the bone’s surface. Localized osteomyelitis involves the full thickness of bone and includes the periosteum, cortex, and medullary canal; its etiology is typically direct extension. Cierny et al.24 state that localized osteomyelitis is often a combination of medullary and superficial findings. Diffuse osteomyelitis is defined as a “permeative, circumferential, or through and through disease of hard and soft tissue.”24

FIGURE 18-17. Normal vascular patterns of a tubular bone in the child, infant, and the adult. A: In the child, the capillaries of the metaphysis turn sharply, without violating the open growth plate. B: In the infant, some metaphyseal vessels may penetrate the open growth plate, ramifying in the epiphysis. C: In the adult, with closure of the growth plate, a vascular connection between metaphysis and epiphysis can be recognized. (Courtesy of Resnick D. Diagnosis of Bone and Joint Disorders. Philadelphia, PA: Saunders; 1981:2048 [Figure 60-2].) The Infecting Organisms Staphylococcus aureus is the most common infectious agent in all age groups.8,26 Organisms most commonly causing acute hematogenous

osteomyelitis usually stem from a single organism such as Staphylococcus or Streptococcus. In young adults Staphylococcus aureus is considered the primary infecting organism, whereas in the elderly gram-negative rods are most commonly isolated. Pseudomonas aeruginosa and methicillin-resistant Staphylococcus are frequently found in the intravenous (IV) drug abuser. Osteomyelitis secondary to contiguous soft tissue or by direct extension depends on the mechanism of transmission. For example, a puncture wound is most commonly affected by Staphylococcus or Pseudomonas, whereas an animal bite from a dog or a cat can cause an infection from Pasteurella multocida. Pseudomonas is also seen in nosocomial infections.26 Diabetics tend to contract polymicrobial wound infections leading to osteomyelitis. Aerobes as well as anaerobes must be considered. The primary infecting organism is most commonly isolated by bone biopsy. A culture from the sinus tract or ulcer is usually affected with mixed flora and can be unreliable for proper diagnosis and treatment of the bone infection.27 Chronic osteomyelitis can be caused by many organisms, including bacteria, fungi, mycobacteria, and the spirochete Treponema pallidum (which causes syphilis). Osteomyelitis secondary to Mycobacterium tuberculosis can be identified radiographically as well as clinically because of the chronicity of the disease. Clinically the patient has pain, stiffness, mild-to-moderate swelling, and erythema of the affected part. The soft tissues are affected by a granulomatous reaction leading to severe soft tissue destruction. The soft tissue destruction is characterized by mononuclear cell infiltrates, giant cell inflammatory infiltrates, fibroblast proliferation, mild edema, and small vessel congestion.28,29 Radiographic findings of osteomyelitis secondary to M. tuberculosis include soft tissue changes early on followed by bone destruction. Bone destruction is usually evident, because of the chronic nature of the disease. Bone changes include subchondral osteopenia followed by irregular areas of destruction with minimal marginal sclerosis. Periosteal reaction is characteristically minimal. Joint changes may occur, and spaces between joints may narrow. Damage to epiphyseal plates can cause growth abnormalities in children. Sequestrum is much less common with M. tuberculosis than in osteomyelitis

from a bacterial origin.30,31 Fungal osteomyelitis may mimic bacterial or tuberculosis infection of bone both clinically and radiographically. Clinicians usually consider the possibility of a fungal infection of bone when a patient has not responded well to antibiotic therapy for a bacterial infection or bacteria have been isolated from the bone culture.29

FIGURE 18-18. Hematogenous osteomyelitis, juvenile pattern. A: Anteroposterior view. B: Lateral view. There is significant rarefaction involving the metaphysis and adjacent diaphysis; however, the epiphysis (e) is unaffected. Arrows identify exuberant periosteal reaction. Syphilitic osteomyelitis can be either congenital or acquired with tertiary syphilis. In the fetus, the newborn, or young infant, bone destruction can develop. A syphilitic osteochondritis can occur, causing changes in endochondral ossification. This leads to broad, horizontal radiolucent bands on a radiograph. Metaphyseal abnormalities can appear, leading to epiphyseal separation. Diaphyseal osteomyelitis can appear in the untreated newborn or infant. Osteolytic lesions with bony eburnation and underlying periostitis are found in the diaphyseal bone. Periostitis can also be seen in congenital syphilis. It is diffuse, symmetric, and widespread.1 Acquired syphilis can cause bone destruction as well. The bone changes usually occur in the tertiary stage of the disease. Irregular cortical thickening may be present. The tibia and femur are the most commonly affected bones in the lower extremity. The irregular cortical thickening of the anterior tibia has been described as a “saber shin” appearance.16 Periosteal reactions may be irregular or lacelike and show spicules radiating perpendicular to the shaft of the bone, mimicking osteogenic sarcoma.1,16 SPECIAL IMAGING MODALITIES This section provides an overview of special imaging studies and their application to skeletal infection in the foot. Adjunctive studies are discussed in Section 6, Special Imaging Procedures, and the reader is encouraged to refer to those chapters for further information. Radiography remains the initial study of choice for imaging infection. If there is obvious evidence of osteomyelitis radiographically in the patient who has not had surgery recently performed at the site and there is no other superimposed pathology (fracture, arthritis, tumor, etc.), there is no need to perform additional adjunctive studies unless considering surgical intervention. However, when the diagnosis is not obvious or is compounded

by superimposed pathology, several adjunctive studies are available (Table 18-2). Special imaging modalities are extremely helpful in differentiating osteomyelitis from soft tissue infection, underlying neuropathic disease, or postoperative bone healing, which all may mimic osteomyelitis radiographically with increased soft tissue volume and density, rarefaction, periosteal reaction, erosion, subchondral resorption, or osteolysis. MRI is probably the most widely ordered imaging study after radiography, but radionucleotide techniques are still used for the early detection of osteomyelitis and in patients where MRI is not available or contraindicated. In addition to CT, ultrasound and positron emission tomography (PET) scan are being used to aid in the diagnosis of osteomyelitis. Often a combination of imaging studies is used to confirm the presence of infection. Nuclear Medicine Radionucleotides that can be used for detecting osteomyelitis include technetium-99 ethylene diphosphonate (99mTc-MDP), gallium-67 citrate, and indium-111 (In-111). Localization of Tc-99 is related to both osteoblastic activity and skeletal vascularity. A technetium bone scan can often be positive 24 hours after the onset of symptoms and 10 to 14 days before any radiographically visible changes have occurred.32 A positive scan shows a well-defined, localized, increased uptake of Tc-99m in an area of inflammation or infection, also referred to as a “hot spot”; there will be increased uptake in all three phases, the angiogram, blood pool phase, and the actual bone scan (Figure 18-19). A positive bone scan does not prove that osteomyelitis is present, because Tc-99m uptake is relative to osteoblastic activity and therefore anything that causes increased osteoblastic activity produces a positive scan. Occasionally, 1 to 2 days after development of symptoms due to infection a technetium scan may be normal or show an area of decreased uptake or a “cold spot” because the medullary microcirculation is compressed by intraosseous pus.11 A cold spot shows up as a lack of radionucleotide accumulation in the affected bone area. Triphasic bone scintigraphy cannot differentiate osteomyelitis from other causes of active bone remodeling such as recent fracture, neoplasm, loose prosthesis, septic arthritis, or diabetic osteoarthropathy.33 For this reason, other nuclear scans

have been used with technetium to help in making a diagnosis in those instances (Figure 18-20A). TABLE 18-2 Primary Use of Imaging Studies for Osteomyelitis Sensitivity/Specificity Imaging Method Use (%)a Radiography Initial study of choice 43–75/75–83 Soft tissue abscess; periosteal elevation; Ultrasound guidance for biopsy, incision and drainage Early detection when radiograph unremarkable; after diagnosis, useful for MRI 82–100/75–96 determining extent of involvement; preoperative planning; diabetic foot Detecting/documenting 67/50 (chronic CT sequestra in chronic osteomyelitis)b osteomyelitis Bone scintigraphy Early detection; (three-phase bone identifying multifocal 85/≈25 scan) involvement Increases specificity of Bone scan + gallium bone scan for early ≈60/≈80 detection For complicated White blood cell scan osteomyelitis when there (indium, Tc-HMPAO) is superimposed pathology Confirm or exclude 98/91 (chronic FDG-PET chronic osteomyelitis osteomyelitis) aFrom Pineda C, Espinosa R, Pena A. Radiographic imaging in

osteomyelitis: the role of plain radiography, computed tomography, ultrasonography, magnetic resonance imaging, and scintigraphy. Semin Plast Surg. 2009;23:80.

bFrom Termaat MF, Raijmakers PG, Scholtein HJ, et al. The accuracy of

diagnostic imaging for the assessment of chronic osteomyelitis: a systematic review and meta-analysis. J Bone Joint Surg Am. 2005;87:2464. Gallium-67 scan is another modality that can be useful in detecting osteomyelitis. Gallium has several possible mechanisms of concentration in a lesion: granulocyte uptake, direct bacterial uptake, lactoferrin binding at the site of injection, or uptake of gallium in reactive bone.32 Gallium was initially proposed as a bone-scanning agent, because it showed increased uptake in areas of increased bone turnover. Because gallium accumulated in leukocytes and bacteria uptake can occur in infected bone or with soft tissue infection, osteomyelitis cannot be differentiated by gallium scan alone. To improve the specificity of gallium imaging at the site of preexisting skeletal disease, investigators compared gallium scans with triphasic bone scans (Figure 18-20B).2,33 If a gallium scan is negative in light of a positive technetium scan, most likely no osteomyelitis is present. In cases of osteomyelitis, a gallium scan should show a local increase uptake that is equal to or greater than technetium uptake. A positive gallium scan in association with a negative technetium scan may suggest cellulitis. Gallium can be positive as early as 30 minutes after its injection in the presence of an active bone infection.34 It is very useful in detecting soft tissue abscesses throughout the body. By showing a decreased uptake as infection improves, gallium can be used as a prognosticator in determining clinical response and termination of antibiosis in treatments of osteomyelitis. Gallium uptake was found to parallel the clinical course, with scans reverting to normal after successful antibiotic therapy.34

FIGURE 18-19. Early calcaneal osteomyelitis. A: Soft tissue density has increased along the plantar aspect of the calcaneus and subtle rarefaction in the calcaneal body, but these findings alone are not specific for osteomyelitis. B: Third-phase technetium bone scan. Focal increased uptake is seen in the heel. In the proper clinical setting, this is highly suggestive of osteomyelitis. Indium-111 leukocyte imaging can be useful for detecting both acute and chronic osteomyelitis. However, it is much more reliable in acute osteomyelitis. The labeled cells aggregate at the site of infection because leukocytes respond in the act of healing. The main drawback to the use of In111 is the more complex preparation, high cost, and relatively high radiation dose to the spleen.32,35 Because leukocytes labeled with In-111 are not usually incorporated into areas of increased bone turnover, they are reported to be specific for infection in cases where osteomyelitis is superimposed on diseases that also cause increased bone turnover.32,35 According to Seabold and associates,33 combined technetium and indium scan imaging improves the specificity over indium alone. Indium-labeled WBCS (white blood cell scan) is the best diagnostic technique to exclude the diagnosis of osteomyelitis. When differentiating a soft tissue infection from a bone infection or postoperative infection, indium combined with technetium is

more specific and permits a more reliable diagnosis (Figure 18-20C). False positives have been reported, caused by early fractured callus, acute bone infarct, heterotrophic bone formation, inflammatory arthritis, and rare neoplasms. Also, indium localization has been reported in uninfected fractures for up to 3 months.33 Leukocytes have also been labeled with Tc-99m using hexamethylpropyleneamine oxine (HMPAO) to identify superficial and deep wound infection. Tc-99m HMPAO is used to detect inflammatory change as well as an infectious process, in a manner similar to that used with indium111.36,37 Tc-99m-labeled leukocytes have some advantages over indium-111 in that Tc-99m has a shorter half-life of 6 hours, compared with a 2.8-day half-life for indium. Technetium images are said to have better resolution, the process is less expensive, and the patient is exposed to less ionizing radiation.37 A newer modality that is being investigated for evaluating osteomyelitis is FDG-PET (fluorine-18 fluorodeoxyglucose-positron emission tomography). In comparison to bone scintigraphy, MRI, and leukocyte scintigraphy, FDGPET has the greatest diagnostic accuracy for confirming or excluding chronic osteomyelitis.38,39 Magnetic Resonance Imaging MRI has become the most useful imaging study for evaluating suspected osteomyelitis (Figure 18-21).40 It can detect early osteomyelitis, before visible on radiographs, as early as 3 days after the start of infection.41 MRI is also valuable when determining the extent of involvement, especially for preoperative planning. MRI is excellent for excluding osteomyelitis; its negative predictive value is 100%; unfortunately, its positive predictive value (distinguishing osteomyelitis from Charcot neuropathic osteoarthropathy, for example) is lower.42 Dinh and colleagues43 performed a meta-analysis to determine the pooled sensitivity and specificity of radiography, MRI, bone scan, and leukocyte scan for the diagnosis of osteomyelitis in patients with underlying foot ulcer

(Table 18-3). Pooled sensitivity for radiography is reduced in the first 2 to 4 weeks due to the fact that bony changes will be evident when 40% to 70% of bone has been resorbed due to the infection. Dinh et al. concluded that, for the diagnosis of osteomyelitis, MRI was the most accurate imaging study.

FIGURE 18-20. Nuclear imaging combined studies of a patient suspected of having septic arthritis of the ankle. A: Third phase of technetium bone scintigram. B: Gallium scan. C: Indium scan. There is intense uptake of technetium in the periarticular ankle and talocalcaneal joint regions. A similar amount of gallium uptake, compared to the technetium, suggested

osteomyelitis. However, the diffuse, less intense indium uptake suggests soft tissue infection and no bone involvement. It turned out that the patient did have soft tissue infection but not osteomyelitis, which meant that the gallium scan was a false positive in this case. Aside from its cost and lack of availability, a disadvantage of MRI is its occasional inability to distinguish infectious from reactive inflammation.42 Multislice Computed Tomography Multislice computed tomography (MSCT) has a definite use in the diagnosis of osteomyelitis and can be used in conjunction with plain radiographs to evaluate patients both for acute and chronic osteomyelitis. CT shows reliable detection or cortical destruction, periosteal proliferation, and soft tissue extension even when radiographs are normal.44 It is a sensitive method to detect sequestra, which appear as isolated bony fragments completely separated from adjacent bone or as free fragments within the medullary cavity. The detection of a sequestrum is very important when planning surgical intervention; in chronic osteomyelitis the presence of sequestrum represents an active infection. CT is also used to detect cortical defects leading to subcutaneous sinus tracts through which pus, granulation tissue, or sequestra are excreted. Soft tissue defects in abscess formation can be seen on CT more readily than on radiographs. In most cases, MRI is preferable over CT because of the latter’s exposure to ionizing radiation and poorer soft tissue contrast.

FIGURE 18-21. Hallux osteomyelitis, MRI. A: Axial T2-weighted image with fat saturation and B: coronal T1-weighted image through the forefoot showing classic signs of osteomyelitis. Marrow fat signal intensity is replaced with material that is lower on the noncontrast T1-weighted image and increased on the T2 fat-saturated image. While this pattern is typical, it is not pathognomonic for osteomyelitis. Correlation with gadolinium enhancement and clinical signs and symptoms is also essential to establish the diagnosis. (Images: Courtesy of David P. Mayer, MD.) Cone-beam computed tomography (CBCT) has, for over a decade, been widely used for dentomaxillofacial imaging45 and recently has become available for extremity use.46 It has been demonstrated to provide similar if not better definition of the fine osseous structure than MSCT. The indications for its use in evaluating pedal osteomyelitis have not yet been investigated; however, they should at least be the same as those for MSCT mentioned above. If readily available, such as in the office setting, and the fact that its ionizing radiation dose is significantly less compared to MSCT,46,47 CBCT may warrant an increased selection over MRI for detecting early osteomyelitis. Diagnosis of Osteomyelitis in Patients with Diabetic Foot TABLE 18-3   Ulcer: Statistics of Imaging Studies Based on Reports from Dinh et al.43 Pooled

 

Pooled Sensitivity %

Radiography MRI Bone scan Leukocyte scan

54 90 81 74

Specificity % 68 79 28 68

aModified from Dinh MT, Abad CL, Safdar N. Diagnostic accuracy of

physical examination and imaging tests for osteomyelitis underlying diabetic foot ulcers: meta analysis. Clin Infect Dis. 2008;47(4):519,Table 6. Ultrasound Musculoskeletal diagnostic ultrasound (US) has been used for assessing infection in soft tissue and bone. Because it can detect increased fluid collections, US can distinguish between abscess and noninfectious masses, such as tumor, as well as delineate the extent of soft tissue infection.48 US can also detect subtle changes along the periosteal surface of bone, such as early periostitis or erosion, before being seen radiographically.40,42 The advantages of US are its safety, portability, noninvasiveness, and real time characteristics. It is also inexpensive and can yield useful diagnostic results. Disdvantages are its limitation with bony pathology, and it is very operator dependant.

FIGURE 18-22. Osteomyelitis imaging algorithm. When infection is suspected clinically, radiographs will be ordered first. If radiographs reveal obvious findings associated with osteomyelitis, such as osteolysis, then the diagnosis is made and no further studies are necessary. If, however, the radiographs are entirely normal, order either bone scintigraphy or MRI. If all three phases of the bone scintigram demonstrate focal uptake in the area of concern, this would be consistent with osteomyelitis. (Gallium could be used as an adjunct to bone scintigraphy to increase its specificity.) MRI can be ordered with gadolinium to enhance the images; some imaging centers include it as normal protocol for suspected infection. MRI will demonstrate high-signal intensity in areas of concern on T2, STIR, and fat-suppressed T1 images (with gadolinium) if infection is present. If the initial radiographic study reveals a superimposed or coexisting pathology or process, such as Charcot neuropathic osteoarthropathy, osteoarthritis, or prior surgery, then differentiation of osteomyelitis becomes much more challenging. In this case, one could follow six different paths: (1) MRI; (2) third-phase bone scintigram + gallium; (3) third-phase bone scintigram + indium; (4) technetium HMPAO scan; (5) FDG-PET study; or (6) SPECT/CT study. The latter two are newer modalities that may not be readily available. In any case, if surgery were being planned, an MRI study would be obtained for preoperative planning, which is one reason why it is preferred initially over all the others after radiography.

SUMMARY Figure 18-22 provides an algorithm for the use of adjunctive imaging studies when assessing osteomyelitis. The American College of Radiology has Appropriateness Criteria for suspected osteomyelitis of the foot in patients with diabetes mellitus.49,50 There are several variant scenarios, but they all include radiography and MRI (with or without contrast) as the appropriate procedures. If MRI is contraindicated or not available, then a white blood cell scan such as In-111 plus Tc-99m triple phase scan may be indicated if suspicious of osteomyelitis or CT if suspicious of Charcot foot. Radiographic follow-up of the treated patient is helpful. If antibiotic therapy is adequate in the acute phase of osteomyelitis, further destruction should be arrested. Clinical improvement usually precedes radiographic changes consistent with healing of the affected site. In a chronic case, after surgical debridement one can use radiographs to follow bone remodeling or to watch for signs of chronic osteomyelitis. REFERENCES   1. Resnick D. Diagnosis of Bone and Joint Disorders. 4th ed. Philadelphia, PA: WB Saunders; 2002.   2. Christman R. The radiographic presentation of osteomyelitis in the foot. Clin Podiatr Med Surg. 1990;7(3):443.   3. Edeiken J, Dalinka M, Karasick D. Edeiken’s Roentgen Diagnosis of Diseases of Bone. 4th ed. Baltimore, MA: Lippincott Williams & Wilkins; 1990.   4. Bravo AA, Bruskoff BL, Perner R. A review of osteomyelitis with case presentation. J Am Podiatr Med Assoc. 1985;75(2):83.   5. Bonakdapour A, Baines V. The radiology of osteomyelitis. Orthop Clin North Am. 1983;14(1):21.   6. Kehr LE, Zulli LP, McCarthy DJ. Radiographic factors in osteomyelitis. J Am Podiatr Assoc. 1977;67(10):716.

  7. Chusid MS, Jacobs WM, Sty JR. Pseudomonas arthritis following puncture wounds of the foot. J Pediatr. 1979;94(3):429.   8. Kothari NA, Pelchovitz DJ, Meyer JS. Imaging of musculoskeletal infections. Radiol Clin North Am. 2001;39(4):653.   9. Pineda C, Vargas A, Rodriguez AV. Imaging of osteomyelitis: current concepts. Infect Dis Clin North Am. 2006;20:789.  10. Patzakis M. Management of acute and chronic osteomyelitis. In: Chapman M, Szabo R, Marder R, et al, eds. Chapman’s Orthopaedic Surgery. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2001:3523–3531.  11. Septimus EJ, Muscher DM. Osteomyelitis: recent clinical and laboratory aspects. Orthop Clin North Am. 1977:10(2):347.  12. Donovan A, Schweitzer ME. Current concepts in imaging diabetic pedal osteomyelitis. Radiol Clin North Am. 2008;46(6):1105.  13. Williams ML, Fleetor M. Radiographic evaluation of subacute osteomyelitis in children. J Curr Podiatr Med. 1990;39(6):81.  14. Manaster BJ, Disler DG, May DA. Musculoskeletal Imaging: The Requisites. 2nd ed. St Louis, MO: Mosby; 2002:554.  15. Jennin F, Bousson V, Parlier C, et al. Bony sequestrum: a radiologic review. Skeletal Radiol. 2011;40:963.  16. Greenfield GB. Radiology of Bone Diseases. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 1986.  17. Lew DP, Waldvogel FA. Osteomyelitis. N Engl J Med. 1997;336:999.  18. Waldvogel FA, Vassey H. Osteomyelitis: the past decade. N Engl J Med. 1980;303(8):360.  19. Waldvogel FA, Medoff G, Swartz NM. Osteomyelitis: a review of clinical features, therapeutic considerations and unusual aspects (first of three

parts). N Engl J Med. 1970;282:198.  20. Frykberg RG. Osteoarthropathy. Clin Podiatr Med Surg. 1987;4(2):351.  21. Kaschok TJ, Laine W. Radiology of the diabetic foot. Clin Podiatr Med Surg. 1988;5(4):849.  22. Trueta J. Three types of acute hematogenous osteomyelitis: a clinical and vascular study. J Bone Joint Surg Br. 1959;41:671.  23. Robbins SL, Cotran RS. Pathologic Basis of Disease. 2nd ed. Philadelphia, PA: WB Saunders; 1979:1477–1486.  24. Cierny G III, Mader JT, Pennick JJ. A clinical staging system for adult osteomyelitis. Clin Orthop. 2003;414:7.  25. Lew DP, Waldvogel FA. Osteomyelitis. Lancet. 2004;364:369.  26. Rao N, Ziran BH, Lipsky BA. Treating osteomyelitis: antibiotics and surgery. Plast Reconstr Surg. 2011;127(suppl 1):177S.  27. Joseph WS. Handbook of Lower Extremity Infections. 3rd ed. Towson, MD: Data Trace Publishing; 2009.  28. Schlossberg D. Orthopedic infection: chronic infective arthritis. New York, NY: Springer; 1988.  29. Waldvogel FA, Medoff G, Swartz MN. Osteomyelitis: clinical features, therapeutic considerations, and unusual aspects. Springfield, IL: Charles C Thomas; 1971.  30. Somerville FW, Wilkinson MC. Girdlestone’s Tuberculosis of Bone and Joint. 3rd ed. New York, NY: Oxford University Press; 1965.  31. Nielsen F, Helmeg O, de Caraulho A. Case report 533. Skeletal Radiol. 1989;18:153.  32. Schauwecker DS, Park HM, Mock BH, et al. Evaluation of complicating osteomyelitis with Tc-99m MDP, In-111 granulocytes, and Ga-

67 citrate. J Nucl Med. 1984;25:849.  33. Seabold JE, Nepola JV, Conrad GR, et al. Detection of osteomyelitis at fracture nonunion sites: comparison of two scintigraphic methods. AJR Am J Roentgenol.1989;152:1021.  34. Visser HJ, Jacobs AM, Oloff L, et al. The use of differential scintigraphy in the clinical diagnosis of osseous and soft tissue changes affecting the diabetic foot. J Foot Surg. 1984;23(11):74.  35. Schauwecker DS, Braunstein EM, Wheat JL. Diagnostic imaging of osteomyelitis. Infect Dis Clin North Am. 1990;4(3):441.  36. Sampson CB, Solanki C, Barber RW. 99Tcm-exametazime-labeled leukocytes: effect of volume and concentration of exametazime or labeling efficiency: multi-dose radiolabeling. Nucl Med Commun. 1991;12:719.  37. Reynolds JH, Graham D, Smith FW. Imaging inflammation with 99Tcm HMPAO labeled leukocytes. Clin Radiol. 1990;42:195.  38. Termaat MF, Rajimakers PGHM, Scholten HJ, et al. The accuracy of diagnostic imaging for the assessment of chronic osteomyelitis: a systematic review and meta analysis. J Bone Joint Surg Am. 2005;87(11):2464.  39. Santiago-Restrepo C, Gimenez CR, McCarthy K. Imaging of osteomyelitis and musculoskeletal soft tissue infections: current concepts. Rheum Dis Clin North Am. 2003;29:89.  40. Pineda C, Espinosa R, Pena A. Radiographic imaging in osteomyelitis: the role of plain radiography, computed tomography, ultrasonography, MRI, and scintigraphy. Semin Plast Surg. 2009;23:80.  41. Kocher MS, Lee B, Dolan M, et al. Pediatric orthopedic infections: early detection and treatment. Pediatr Ann. 2006;35:112.  42. Palestro CJ, Love C, Miller TT. Infection and musculoskeletal conditions: imaging of musculoskeletal infections. Best Pract Res Clin Rheumatol. 2006;20:1197.

 43. Dinh MT, Abad CL, Safdar N. Diagnostic accuracy of physical examination and imaging tests for osteomyelitis underlying diabetic foot ulcers: meta analysis. Clin Infect Dis. 2008;47(4):519.  44. Wang A, Weinstein D, Greenfield L, et al. MRI and diabetic foot infections. Magn Reson Imaging. 1990;8:805.  45. De Cock J, Koen Mermuys K, Goubau J, et al. Cone-beam computed tomography: a new low dose, high resolution imaging technique of the wrist, presentation of three cases with technique. Skeletal Radiol. 2012;41:93.  46. Tuominen EK, Kankare J, Koskinen SK, et al. Weight-bearing CT imaging of the lower extremity. AJR Am J Roentgenol. 2013;200:146.  47. Carrino JA, Al Muhit A, Zbijewski W, et al. Dedicated cone-beam CT system for extremity imaging. Radiology. 2014;270:816.  48. Cardinal E, Bureau N, Aubin B. Role of the ultrasound in musculoskeletal infections. Radiol Clin North Am. 2001;39:191.  49. Schweitzer ME, Daffner RH, Weissman BN, et al. ACR appropriateness criteria® on suspected osteomyelitis in patients with diabetes mellitus. J Am Coll Radiol. 2008;5:881.  50. American College of Radiology ACR Appropriateness Criteria®. Suspected Osteomyelitis of the Foot in Patients with Diabetes Mellitus; 2012. http://www.acr.org/Quality-Safety/AppropriatenessCriteria/Diagnostic/Musculoskeletal-Imaging. Accessed April 21, 2014. SUGGESTED READINGS Aiyenuro O, Shariff S. Acute staphylococcal hind foot osteomyelitis in a healthy adult: a case report. Foot Ankle Surg. 2012;18:e12. Al-Sheikh W, Sfakianakis GN, Mnaymneh W, et al. Subacute and chronic bone infections: diagnosis using In-111, Ga-67, and Tc-99m MDP bone scintigraphy, and radiography. Radiology. 1985;155:501–506. Bamberger DM, Davis GP, Gerding DN. Osteomyelitis in the feet of diabetic

patients. Am J Med. 1987;83:653–660. Beltran J, McGhee RB, Shaffer PB, et al. Experimental infections of the musculoskeletal systems: scintigraphy. Radiology. 1988;167:167–172. Beltran J, Noto AM, McGhee RB, et al. Infections of the musculoskeletal system: high field strength MR imaging. Radiology. 1987;164:449–454. Berendt AR, Peters EJ, Bakker K, et al. Diabetic foot osteomyelitis: a progress report on diagnosis and a systematic review of treatment. Diabetes Metab Res Rev. 2008;24(suppl 1):S145. Buze BH, Hawkins RA, Marcus CS. Technetium-99m white blood cell imaging: false negative result in salmonella osteomyelitis associated with sickle cell disease. Clin Nucl Med. 1989;14:104–106. Carek PJ, Dickerson LM, Sack JL. Diagnosis and management of osteomyelitis. Am Fam Physician. 2001;15;63(12):2413. Chandnani VP, Beltran J, Morris C, et al. Acute experimental osteomyelitis and abscess: detection with MR imaging vs CT. Radiology. 1990;174:233– 236. Donohue TW, Kanat IO. Radionuclides: their use in osteomyelitis. J Am Podiatr Med Assoc. 1997;77(6):284–289. Ewing R, Fainstein V, Musher D, et al. Articular and skeletal infections caused by Pasteurella multocida. South Med J. 1980;73(10):​1349–1352. Gelfand MJ, Silberstein EB. Radionuclide imaging use in the diagnosis of osteomyelitis in children. JAMA. 1977;237:245–247. Guillerman RP. Osteomyelitis and beyond. Pediatr Radiol. 2013;43​(suppl 1):S193. Ingerman J, Abrutyn E. Osteomyelitis: a conceptual approach. J Am Podiatr Med Assoc. 1986;7(69):487–492. Jacobs AM, Oloff LM. Osteomyelitis. In: McGlamry ED, ed. Comprehensive

Textbook of Foot Surgery. Vol 2. Baltimore, MA: Williams and Wilkins; 1987. Jacobson HG. Musculoskeletal applications of magnetic resonance imaging. JAMA. 1989;262(17):2420–2426. Jerome JT. Tuberculosis of the midtarsal joints: a case report. J Am Podiatr Med Assoc. 2008;98:246. Khan AN. Acute pyogenic osteomyelitis imaging. Medscape. http://emedicine.medscape.com/article/393120-overview. Updated August 28, 2013. Accessed April 20, 2014. Khan AN. Chronic osteomyelitis imaging. Medscape. http://emedicine.medscape.com/article/393345-overview. Updated April 16, 2013. Accessed April 20, 2014. Kinberg P. Osteomyelitis of an epiphyseal region. J Foot Surg. 1983;22(3): 251–256. Laitinen R, Tahtinen J, Lantto T, et al. Tc99m labeled leukocytes in imaging of patients with suspected acute abdominal inflammation. Clin Nucl Med. 1990;15:597–602. Ledermann HP, Morrison WB, Schweitzer ME. MR image analysis of pedal osteomyelitis: distribution, patterns of spread, and frequency of associated ulceration and septic arthritis. Radiology. 2002;223:747. Lisbona R, Rosenthall L. Observations on the sequential use of 99MTCphosphate complex and 67Ga imaging in osteomyelitis, cellulitis and septic arthritis. Radiology. 1977;123:123–129. Mader JT, Calhoun JH. Long bone osteomyelitis: an overview. JAMA. 1989;79(10):476–481. Majcen ME, Wilfinger CC, Pilhatsch A. Interpretation of radiologic abnormalities in patients with chronically infected ingrown toenails with regard to a possible exogenic osteomyelitis. J Pediatr Surg. 2009; 44:2179.

Mansor IA. Typhoid osteomyelitis of the calcaneus due to direct inoculation. J Bone Joint Surg Am. 1967;49(4):732–734. Miller ER, Semian DW. Gram negative osteomyelitis following puncture wounds of the foot. J Bone Joint Surg Am. 1975;57(4):535–537. Quinn SF, Murray W, Clark RZ, et al. MR imaging of chronic osteomyelitis. J Comput Assist Tomogr. 1988;12(1):113–117. Saigal G, Azouz EM, Abdenour G. Imaging of osteomyelitis with special reference to children. Semin Musculoskelet Radiol. 2004;8:255. Sia IG, Berbari EF. Infection and musculoskeletal conditions: osteomyelitis. Best Pract Res Clin Rheumatol. 2006;20:1065. Stone RA, Lehlman RA, Zeichner AM. Acute hematogenous osteomyelitis: a case report. J Am Podiatr Assoc. 1982;72(1):31–34. Stumpe KD, Strobel K. Osteomyelitis and arthritis. Semin Nucl Med. 2009;39:27. Varzos P, Galinski A, Gelling WJ, et al. Osteomyelitis associated with monofilament wire fixation. J Foot Surg. 1983;22(3):212–217. Vorne M, Lanatto IS, Paakkinen S. Technetium 99m HM-PAO labeled leukocytes in detection of inflammatory lesion: comparison with gallium-67 citrate. J Nucl Med. 1989;30:1332–1336. Vorne M, Salo S, Anttolainen I, et al. Septic Haemophilus influenza: polyarthritis demonstrated best with Tx-99m HMPAO labeled leukocytes. Clin Nucl Med. 1990;15:883–886. Wheat J. Diagnostic strategies in osteomyelitis. Am J Med. 1985;78: 218– 224. Wing VW, Jeffrey RB, Federle MP, et al. Chronic osteomyelitis examined by CT. Radiology. 1985;154:171–174. Yazdi H, Shirazi MR, Eghbali F. An unusual presentation of subacute

osteomyelitis: a talus brodie abscess with tendon involvement. Am J Orthop (Belle Mead NJ). 2012;41:E36. Yuh WT, Corson JD, Baraniewski HM, et al. Osteomyelitis of the foot in diabetic patients: evaluation with plain film, 99m Tc-MDP bone scintigraphy, and MR imaging. AJR Am J Roentgenol. 1989;152:795–800. Zeiger LS, Fox IM. Use of indium-111 labeled white blood cells in the diagnosis of diabetic foot infections. J Foot Surg. 1990;29(1):46–51.

19 Joint Disease CRISTINA MARCHIS-CRISAN AND ROBERT A. CHRISTMAN Several arthritides have a predilection for the foot (Box 19-1). Without question, osteoarthritis is the most common, primarily because of the mechanical wear and tear that weight-bearing activities place on cartilage. However, it is not unusual for the inflammatory rheumatic diseases (rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, and reactive arthritis) to first appear or be diagnosed in the feet. Furthermore, gouty arthritis and diabetic neuropathic osteoarthropathy target the foot as well. Radiography remains the initial study of choice for most rheumatic disorders.1 BOX 19-1 Joint Disorders with Predilection for the Foot Osteoarthritis Rheumatoid arthritis Psoriatic arthritis Reactive arthritis (formerly called Reiter syndrome) Ankylosing spondylitis Gouty arthritis Charcot neuropathic osteoarthropathy Septic (infectious) arthritis The archetypal radiographic presentations of joint disorders that have been described in the literature are not necessarily what the clinician encounters in everyday practice. The patient presenting with a classic radiographic picture

was probably diagnosed with the joint disease many years or even decades before. In contrast, the patient with acute symptomatology may initially come for help at the onset of disease or soon thereafter. In such a case, the radiographic findings are frequently subtle and nonspecific, the clinical findings are vague, and the diagnosis is often elusive. Furthermore, atypical cases are common. The challenge, therefore, is to identify the subtle, early radiographic findings, because the classic features of any particular joint disorder may not manifest until many years later. To maximize the detection of early arthritis, you “must know how to look, where to look, and what to look for.”2 Then, after considering clinical and laboratory findings, along with advanced imaging studies, if warranted, you must list and consider the probable differential diagnoses. A detailed, systematic approach for radiographically evaluating joint disease in the foot entails three considerations3 (Box 19-2). For obvious reasons, the symptomatic joint or joints are assessed first. The asymptomatic joints of both extremities should also be evaluated, for two reasons: Most joint disorders target both extremities, and joints may be affected that are clinically asymptomatic. (Joint disease is one of the few pathologic conditions that justifies requesting a bilateral radiographic study.) The examination does not stop here, however. Sites distant from involved joints, osseous and soft tissue, are also considered. Abnormal findings at the calcaneal entheses and heel pain, for example, can be associated with joint disease. Finally, the distribution of radiographic findings must be assessed for specific patterns.4 Many articular disorders demonstrate characteristic patterns of joint involvement that help distinguish one disease from another. BOX 19-2 A Systematic Approach to Evaluating Joint Disease ROENTGEN FEATURES AT OR ADJACENT TO INVOLVED JOINTS PRIMARY FINDINGS Osteophyte Osseous erosion

Subchondral resorption Arthritis mutilans SECONDARY FINDINGS Bone production Joint space alteration Soft tissue edema, mass, calcification Detritus Geode Para-articular osteopenia Alignment abnormalities EXTRA-ARTICULAR INVOLVEMENT OF JOINT DISEASES Erosion Enthesopathy Soft tissue mass PATTERNS OF JOINT DISEASE ACCORDING TO DISTRIBUTION OF ROENTGEN FINDINGS Joints involved Targeted joints Bilateral versus unilateral Symmetry versus asymmetry Extra-articular sites involved

CATEGORIZATION OF THE JOINT DISORDERS Articular disorders affecting the foot may involve one or multiple joints. Monoarticular joint disease is divided into two groups: inflammatory disease and mechanical/infiltrative disorders.1 Foot examples are generally attributed to trauma, infection, or acute gouty arthritis (Box 19-3). Less common causes in the foot include rheumatoid monoarthritis and pigmented villonodular synovitis. Polyarticular joint disorders are divided into two similar categories, inflammatory and noninflammatory. Examples affecting the foot are listed in Box 19-4. BOX 19-3 Causes of Monoarticular Joint Disease in the Foot INFLAMMATORY Infectious (septic arthritis) Crystal-induced Gout (acute) Pyrophosphate arthropathy Hydroxyapatite Systemic disease but monoarticular involvement Rheumatoid monoarthritis Psoriatic arthritis Reactive arthritis MECHANICAL/INFILTRATIVE (NONINFLAMMATORY) Trauma Osteonecrosis

Benign tumor Pigmented villonodular synovitis Osteochondroma Differentiation of joint disorders can be simplified by applying a general classification system to the presenting features. One categorization of arthritis has been based on underlying pathologic processes: degenerative, inflammatory, and metabolic5 (Box 19-5). This classification, unfortunately, does not include Charcot neuropathic osteoarthropathy. BOX 19-4 Polyarticular Joint Disease Affecting the Foot INFLAMMATORY Rheumatoid arthritis Seronegative spondyloarthritis Psoriatic arthritis Ankylosing spondylitis Reactive arthritis Chronic tophaceous gout NONINFLAMMATORY Primary osteoarthritis Charcot neuropathic osteoarthropathy Pigmented villonodular synovitis (midfoot) In the first edition of this textbook, Christman6 categorizes joint disorders affecting the foot as either hypertrophic or atrophic arthritis, based on predominant radiographic features (Box 19-6). As far back as 1904 these

terms were used to distinguish between osteoarthritis and rheumatoid arthritis.7 Hypertrophic joint disease features bone overgrowth and enlargement. The characteristic findings are subchondral sclerosis and osteophyte formation at the joint margin. Detritus arthritis, a subcategory of hypertrophic arthritis, includes those disorders that exhibit fragmentation in addition to exaggerated hypertrophic features. BOX 19-5 Categories of Joint Disease (Based on Underlying Pathology) DEGENERATIVE Osteoarthritis INFLAMMATORY Rheumatoid arthritis Seronegative spondyloarthritis Psoriatic arthritis Reactive arthritis Ankylosing spondylitis Septic arthritis METABOLIC Gouty arthritis Pyrophosphate arthropathy The loss of bone substance, primarily through erosion, and joint space narrowing, with or without periarticular osteoporosis, characterize the atrophic joint disorders. A subdivision of this group is commonly associated clinically with an adjacent soft tissue mass and the preservation of joint space; the term “lumpy-bumpy” joint disease has been used to characterize this group.8

BOX 19-6 Categories of Joint Disease (Based on Radiographic Features) HYPERTROPHIC JOINT DISEASE Osteoarthritis (primary) Detritus arthritis Tarsus and midfoot Charcot neuropathic osteoarthropathy Posttraumatic arthritis ATROPHIC JOINT DISEASE Rheumatoid arthritis Seronegative spondyloarthritis Psoriatic arthritis Ankylosing spondylitis Reactive arthritis Septic arthritis Forefoot Charcot neuropathic osteoarthropathy “Lumpy-bumpy” joint disease Gouty arthritis Multiple reticulohistiocytosis Pigmented villonodular synovitis Charcot neuropathic osteoarthropathy is divided into two subtypes: forefoot and the combined midfoot and tarsus. The radiographic features vary depending on location: Forefoot sites exhibit findings characteristic of

atrophic joint disease; the midfoot and tarsal sites display features of detritus (hypertrophic) arthritis. Since the first edition, Christman has developed a new categorization for joint disorders in the foot based upon four primary radiographic findings: osteophyte, erosion, subchondral resorption, and arthritis mutilans (Box 197). The joint disorders are then further distinguished by secondary (associated) findings, including joint space alteration, soft tissue abnormality, bone production, bone loss, and detritus (Box 19-8). This system is valuable for the process of differentiation between the disorders. BOX 19-7 Categories of Joint Disease (Based on the Primary Radiographic Finding) OSTEOPHYTE Osteoarthritis EROSION Rheumatoid arthritis Seronegative spondyloarthritis Psoriatic arthritis Ankylosing spondylitis Reactive arthritis Gouty arthritis SUBCHONDRAL RESORPTION Charcot neuropathic osteoarthropathy Septic arthritis ARTHRITIS MUTILANS

Psoriatic arthritis Charcot neuropathic osteoarthropathy (forefoot only) Rheumatoid arthritis (fifth MPJ only) Each of the roentgen features associated with joint disease is discussed individually in the following section. Remember the radiographic categories of joint disorders; you can recognize associations between certain roentgen findings and arthritis categories, improving your diagnostic acumen. RADIOGRAPHIC FINDINGS IN JOINT DISEASE OF THE FOOT Primary Radiographic Findings Osteophyte An osteophyte is a spur at the margin of a joint (Figure 19-1). Numerous figurative terms have been applied to this lesion, including dorsal flag (along the first metatarsal head), lipping (if at both sides of the joint, especially intertarsal), and beaking (along the talar head). However, the talar beak may not truly represent an osteophyte; its location is between the articular margins of the talonavicular and talocrural joints, where a ridge is normally found.9 The presence of an osteophyte is pathognomonic of osteoarthritis. BOX 19-8 Joint Disease Secondary Findings BONE PRODUCTION Periostitis Whiskering; ivory phalanx (hallux distal phalanx) Subchondral sclerosis (eburnation) Diffuse sclerosis Overhanging margin (Martel sign)

JOINT SPACE ALTERATION Narrow: uneven vs. even Normal Wide SOFT TISSUE ABNORMALITY Increased volume/density: joint effusion versus “sausage toe” Mass: “lumpy-bumpy” vs. fusiform Calcification DETRITUS Loose body (“joint mouse”) Fragmentation GEODE (WITH OR WITHOUT SCLEROTIC MARGIN) PARA-ARTICULAR OSTEOPENIA ALIGNMENT Less than 100% apposition Misalignment Osseous Erosion Erosion is a localized wearing away of bone that begins along its outer surface. It has varying appearances, which are helpful when determining whether or not the associated pathology is acute as well as distinguishing between diseases (Box 19-9). It is the primary feature of all joint disorders affecting the foot except osteoarthritis, Charcot neuropathic osteoarthropathy,

and septic arthritis. Generally speaking, the early appearance of erosion is small, ill-defined, and irregular (Figure 19-2). This characterization contrasts with the larger, well-defined C-shaped erosion classically seen in the disorders associated with an adjacent soft tissue mass (Figure 19-3). The presence of erosion excludes osteoarthritis as a primary diagnosis; however, geodes (or subchondral bone cysts, which are a secondary finding of osteoarthritis) and trauma-induced subchondral bone defects can mimic the appearance of erosion (Figure 19-4). BOX 19-9 Presentations of Erosion PREEROSION (SUBCHONDRAL BONE PLATE) “Dot-dash” appearance “Skip pattern” MARGIN Ill-defined Well-defined With or without sclerotic margin LOCATION Periarticular (extra-articular) Intra-articular Medial vs. lateral vs. both WITH OR WITHOUT JOINT SPACE NARROWING WITH OR WITHOUT ASSOCIATED NEW BONE PRODUCTION

FIGURE 19-1. Osteophyte (osteoarthritis). A: “Flagging” along dorsum of first metaratsal head. B: “Lipping” at intermediate naviculocuneiform joint (arrows) and “beaking” at the talar ridge (arrowhead), probably not a true osteophyte.

Erosion is an early finding in the course of most inflammatory joint diseases. It is usually intra-articular and typically begins along the medial and/or lateral margins of the joint. Between where the cartilage ends and the joint capsule inserts is a bony surface covered only by periosteum or perichondrium; this surface is in contact with the synovium and its fluid and is known as the “bare” area.4 The inflamed synovium, known as pannus, invades the bare area; on a radiograph, the outer margin of subchondral bone quickly disappears. Initially this disappearance may be as subtle as a “dot-dash” appearance or “skipping” along the thin white line that comprises the subchondral bone plate (Figure 19-5). Eventually the localized loss of marginal bone (decreased density) progresses so far that the form of the affected bone appears abnormal. These findings may be recognized days or weeks after the onset of symptoms and contribute to the erosion’s ill-defined and irregular appearance.

FIGURE 19-2. Irregular, ill-defined erosion (arrow) (ankylosing spondylitis).

FIGURE 19-3. Well-defined, C-shaped erosion (gouty arthritis).

FIGURE 19-4. Subchondral geode (arrow) mimicking an erosion (osteoarthritis). The well-defined erosion, in contrast, appears radiographically several months or years after the initial onset of symptoms. It frequently results from chronic pressure atrophy secondary to direct apposition of a soft tissue mass with concomitant infiltration and replacement of bone (Figure 19-3). Or, it may result months or years after acute, ill-defined erosion remodels. Subchondral Resorption The subchondral bone plate is the thin white line at an articular margin; it is where the cartilage ends and bone begins. Normally it is well-defined and continuous (Figure 19-5). Subchondral resorption typically involves the central portion of the joint, in contrast to erosions that are marginal. There are three presentations (Box 19-10): disappearance of the subchondral bone plate centrally; the subchondral bone plate is visible but there is rarefaction of bone immediately adjacent to it, which may appear as early osteolysis; and a combined presentation that demonstrates loss of the bone plate and rarefaction (osteolysis) (Figure 19-6). Subchondral resorption is the primary radiographic feature of septic arthritis and Charcot neuropathic osteoarthropathy. BOX 19-10 Presentations of Subchondral Resorption Disappearance of subchondral bone plate centrally Visibility of subchondral bone plate but rarefaction of bone is immediately adjacent to it Loss of subchondral bone plate and rarefaction/osteolysis

FIGURE 19-5. Early erosion: dot-dash or skip pattern, second metatarsal head (arrowheads). Compare it to the obvious erosion of the third metatarsal

head and the normal subchondral bone plate of the fourth metatarsal head (rheumatoid arthritis). Also note the geode in the center of the third metatarsal head and narrowing of the third metatarsophalangeal joint. Arthritis Mutilans Erosions that involve both margins (medial and lateral) of any metatarsophalangeal or interphalangeal joint (IPJ) proximally accompanied by central resorption of the adjacent phalangeal base can result in a condition known as arthritis mutilans. Also called resorptive arthropathy, arthritis mutilans is characterized by concentric bone resorption and primary joint destruction (osteolysis), and has a very characteristic radiographic picture (Figure 19-7).5 Bone resorption may even expand to include the nearby metadiaphyseal cortex. This has figuratively been described by several terms, most commonly the “pencil-in-cup” deformity. Other terms used to describe arthritis mutilans are listed in Box 19-11. Arthritis mutilans has characteristically been associated with psoriatic arthritis (Figure 19-7). However, forefoot Charcot neuropathic osteoarthropathy (Figure 19-8) and, to a lesser degree, rheumatoid arthritis at the fifth metatarsophalangeal joint (Figure 19-9) present similar pictures.

FIGURE 19-6. Subchondral resorption. A: Central (arrow) (septic arthritis). B: Rarefaction adjacent to subchondral bone plate (arrows) and osteolysis (Charcot neuropathic osteoarthropathy) (B: Courtesy of Philip Bresnahan, DPM.) BOX 19-11 Arthritis Mutilans Figurative Terms Pencil-in-cup deformity Mortar and pestle Sucked candy stick Whittling

Pencil sharpening Secondary Radiographic Findings Bone Production Aside from the osteophyte, other forms of bone production may be associated with the primary radiographic findings listed earlier. The presence or absence of secondary findings can further narrow down the differential diagnoses. For example, new bone production is rarely seen at joints affected by rheumatoid arthritis. However, periostitis is frequently seen with seronegative spondyloarthritis, and an overhanging margin of new bone is associated with the C-shaped erosions encountered with gouty arthritis (Table 19-1).

FIGURE 19-7. Arthritis mutilans: psoriatic arthritis.

FIGURE 19-8. Arthritis mutilans: neuropathic osteoarthropathy.

FIGURE 19-9. Arthritis mutilans: rheumatoid arthritis (fifth metatarsophalangeal joint). TABLE 19-1   Forms of Bone Production and Associated Joint Disorders Form of Bone Production Differential Diagnosis Osteophyte (dorsal flag, lipping, Osteoarthritis beaking) Subchondral sclerosis (eburnation) Osteoarthritis Charcot neuropathic osteoarthropathy Diffuse sclerosis (midfoot and tarsus) Seronegative spondyloarthritis (IPJ or MPJ) Psoriatic (also whiskering type) Periostitis Septic arthritis Charcot neuropathic osteoarthropathy (forefoot) Overhanging margin of bone (Martel Gouty arthritis sign) Psoriatic arthritis (distal phalanx Ivory phalanx hallux) IPJ, interphalangeal joint; MPJ, metatarsophalangeal joint. Subchondral sclerosis, also referred to as eburnation (Figure 19-10), is an ill-defined increased density found in periarticular bone. It is commonly associated with posttraumatic osteoarthritis.

FIGURE 19-10. Osteoarthritis, first metatarsophalangeal joint. Osteophytes (white arrows) and subchondral sclerosis (black arrows). Diffuse sclerosis involving an entire bone is associated with the repair and remodeling phases of Charcot neuropathic osteoarthropathy in the tarsus (Figure 19-8). The presence of periostitis near the metaphysis of a symptomatic metatarsophalangeal or interphalangeal joint is highly suggestive of seronegative arthritis (Figure 19-11). Unfortunately the periostitis is shortlived: Within a few weeks it quickly remodels and becomes continuous with the bony margin. Periostitis may also be seen with septic arthritis and forefoot Charcot neuropathic osteoarthropathy; the latter is difficult to differentiate from infection. A variation of periostitis seen particularly with psoriatic arthritis is referred to as “whiskering”10; its spiculated appearance radiates away from the bone margin. It characteristically involves the hallux and, less frequently, the lesser digit distal phalanges (Figure 19-12). Ill-defined bony sclerosis frequently accompanies this finding. Whiskering appears to represent concomitant new bone formation and erosion at the capsular and ligamentous entheses. Occasionally the distal phalanx of an affected digit becomes quite radiodense or sclerotic relative to normal bone density. This is seen especially in the hallux (Figure 19-13). Known as the “ivory” phalanx, it is another presentation associated, in the proper clinical setting, with psoriatic arthritis.11 The well-defined erosions of chronic gouty arthritis occasionally have an overhanging margin of bone (Figure 19-14). This finding, described by Martel,12 represents new bone production at the margin of erosion. The body seems to be responding to the presence of the tophus and attempting to encapsulate it or wall it off. Its presence strongly suggests gouty arthritis.13 Another finding sometimes associated with the developing erosion of gouty arthritis is ill-defined surrounding sclerosis, which may become well-defined over time.

Joint Space Alteration The joint space seen radiographically corresponds to the cartilage lining each bony surface. Kaye14 has correlated three types of joint space alteration with the articular disorders: narrowing, normal, and widening (Table 19-2).

FIGURE 19-11. Psoriatic arthritis: periostitis (arrows).

FIGURE 19-12. Psoriatic arthritis: whiskering (arrows) of hallux distal phalanx, bilateral. Erosion or lysis of articular cartilage eventually appears as joint space narrowing radiographically, as the two opposing surfaces converge. This is an early radiographic finding of rheumatoid arthritis but is seen eventually in most joint disorders. Joint space narrowing may be either even or uneven. The narrowing seen with inflammatory arthritis usually is even or uniform across the joint (Figure 19-15). This is because inflammatory pannus is distributed throughout the joint and affects all cartilage. In contrast, osteoarthritis secondary to wear and tear or trauma usually only involves one section of the cartilage or subchondral bone, not the entire surface; as a result, joint space narrowing has an uneven or nonuniform presentation radiographically (Figure 19-16). The unaffected joint section has normal spacing. The presence of a normal joint space associated with periarticular erosion is characteristic of joint disorders associated with soft tissue masses (Figure 1917). Chronic tophaceous gout, for example, is not primarily an inflammatory

disorder. Although intense inflammation is clinically seen with acute attacks of gout, these symptoms last only a short period of time. Several years may lapse before radiographic evidence of erosion is evident. Furthermore, since erosion associated with gouty arthritis is commonly periarticular, that is, outside the joint capsule, cartilage may not be directly involved until much later in the course of disease. As a result, the presence of normal joint space in light of obvious erosion is a characteristic finding. This is in strict contrast to inflammatory joint disease. The aggressive nature of this latter group of disorders and associated intense synovitis quite rapidly cause bone and cartilage destruction.

FIGURE 19-13. Psoriatic arthritis: ivory phalanx.

FIGURE 19-14. Gouty arthritis: overhanging margin of bone (arrows)

(Martel sign). Bony or fibrous ankylosis may occur between two bones as an end stage of some joint diseases. This is especially true of the inflammatory joint disorders. Bony ankylosis is more commonly associated with seronegative arthritis. The interphalangeal joints are targeted in psoriatic arthritis (Figure 19-18A). Midfoot ankylosis may be seen in the rheumatoid foot (Figure 1918B). Ankylosis is seldom seen with gouty arthritis and is not associated with pedal osteoarthritis. However, the superimposition of osteophytes and joint space narrowing may simulate bony ankylosis. Grouping of Joint Disease Based on Joint Space Alteration (Modified Kaye Classification9)   Arthritis Inflammatory arthritis Uniform (rheumatoid) Degenerative arthritis (osteoarthritis)

TABLE 19-2   Joint Space Narrowing

 

 

Nonuniform

Ankylosis (end stage)

Posttraumatic arthritis Septic arthritis Seronegative spondyloarthritis (IPJ) Midfoot rheumatoid arthritis

Normal or near normal joint   space Widening

Uniform

 

Nonuniform

IPJ, interphalangeal joint.

Gouty arthritis Early inflammatory arthritis (with acute synovitis) Early psoriatic arthritis (extensive erosion with fibrous tissue deposition)

FIGURE 19-15. Rheumatoid arthritis: even joint space narrowing. Excess fluid accumulates in a joint that is acutely inflamed. To accommodate this fluid, the capsule becomes stretched and the opposing bones are distracted. Radiographically, this may appear as widening of the joint space (Figure 19-19A). Unfortunately, joint space widening secondary to acute synovitis can be a very subtle finding. Furthermore, the finding is short-lived; its radiographic presence is a hit-or-miss incident.

FIGURE 19-16. Osteoarthritis: uneven joint space narrowing. Other findings include osteophytes and geodes.

FIGURE 19-17. Gout: sparing of joint space despite erosive disease. Faint tophus calcification (arrow) is noted medial to the erosion. Extensive resorption of subchondral bone also gives the appearance of joint space widening (Figure 19-19B). Resorption and subsequent fibrous tissue deposition between bones contribute to the widening seen in psoriatic arthritis.3 Soft Tissue Edema, Mass, and Calcification Soft tissue edema may be generalized throughout the foot, regional (forefoot, midfoot, rearfoot, or a toe), or localized to a joint or extra-articular site (Table 19-3). It is viewed radiographically as an increased soft tissue density and/or volume relative to normal expectation. Generalized soft tissue edema can be related to abnormal systemic conditions (cardiac disease, acromegaly), diffuse inflammatory states (cellulitis), or peripheral vascular disease (venous insufficiency, lymphedema).5 It is not, however, a specific finding in joint disease. Many patients with pedal joint disease have concomitant generalized soft tissue edema that is secondary to the conditions just noted. Regional soft tissue edema is confined to a smaller segment of the body. An entire digit, for example, may be edematous from acute inflammatory conditions including infection, seronegative arthritis (the so-called sausage toe), and gout. The edema associated with an acute gouty attack at the first metatarsophalangeal joint may extend to the midfoot. This clinical presentation may certainly mimic an infectious process. Posttraumatic states also can show regional soft tissue edema. Neuropathic osteoarthropathy of the midfoot and tarsus shows either regional or diffuse edema. Localized soft tissue edema may be related to synovial inflammation or to a mass. The edema associated with synovitis surrounds the joint and is quite well-defined radiographically. Increased soft tissue density and volume secondary to synovitis is known as joint effusion. This condition, although nonspecific, is highly associated with inflammatory joint disease. However, synovitis secondary to trauma, either acute or chronic and repetitive, can appear radiographically identical to that caused by inflammatory rheumatic

disease (rheumatoid and seronegative arthritis). Periarticular soft tissue masses are associated with a few joint disorders (Table 19-4). The most common example is the gouty tophus. Tophi are well-defined masses that are found adjacent to joints or at extra-articular sites (Figure 19-20). They occasionally exhibit calcification. Lesions are distributed asymmetrically in the foot.

FIGURE 19-18. Bony ankylosis: A: Psoriatic arthritis. B: Rheumatoid arthritis.

FIGURE 19-19. Joint space widening: A: Secondary to inflammatory exudate (septic arthritis). B: Secondary to central erosion (psoriatic arthritis). Tophi may be seen several years after the initial onset of symptoms. They are a characteristic feature of chronic tophaceous gout; erosions develop adjacent to tophi. It has been reported that the clinical presence of tophi are strongly associated with the characteristic radiographic features of gouty arthritis.15 Another soft tissue erosive joint disorder is multiple reticulohistiocytosis. These masses radiographically appear similar to those seen in gout. However, masses associated with multiple reticulohistiocytosis are widespread, symmetric, and do not calcify. Soft tissue masses are occasionally seen with rheumatoid arthritis. Rheumatoid nodules are seldom found in the foot but, when present, may radiographically appear indistinguishable from a gouty tophus except that the former rarely calcifies.16

Soft tissue tumors and tumor-like lesions may manifest in periarticular locations and cause articular erosions. Although not common, an example of one such lesion occurring in the foot is pigmented villonodular synovitis. As a rule, it is monoarticular. However, a rare, polyarticular (diffuse) manifestation can appear in the midfoot (Figure 19-21). This is probably related to the unique synovial compartmentalization in this anatomic region. Well-defined erosions develop adjacent to soft tissue masses. Grouping of Joint Disease Based on Soft Tissue Edema Classification Soft Tissue Edema Arthritis Generalized/diffuse None Seronegative spondyloarthritis (“sausage toe”)

TABLE 19-3  

Gouty arthritis (acute) Regional Septic arthritis (acute) Charcot neuropathic osteoarthropathy (midfoot/tarsus) Posttraumatic arthritis Localized

Inflammatory (rheumatoid or seronegative spondyloarthritis)

Numerous disorders are associated with soft tissue calcification in the foot. Widespread soft tissue calcification in otherwise normal tissues is associated with disorders that demonstrate elevated calcium or phosphate levels in the serum. Hyperparathyroidism, for example, may cause diffuse periarticular, capsular, and vessel calcification. The majority of soft tissue calcifications seen in the foot, however, are probably dystrophic or idiopathic. Dystrophic calcifications occur in soft tissues that are damaged or altered but have no underlying disturbance in calcium or phosphorus metabolism.17,18 Soft tissue calcifications, when associated with joint disease, may be diagnostic for a group of disorders known as the crystal deposition diseases (Table 19-5). The crystals are monosodium urate (MSU) (gouty arthritis),

calcium pyrophosphate dihydrate (CPPD), and hydroxyapatite (HA). Calcifications can be found in the periarticular tissues, joint capsule, or cartilage. Characteristics of Erosive Joint Disorders with Associated Soft Tissue Mass Joint Disorder Soft Tissue Mass Characteristics Well-defined (lumpy-bumpy); Chronic tophaceous gout articular and/or periarticular; asymmetric; occasionally calcifies Widespread; periarticular; symmetric; Multiple reticulohistiocytosis no calcification Well-defined (fusiform); extraRheumatoid arthritis articular; asymmetric; rarely calcifies Well-defined; articular or periarticular; asymmetric; two Pigmented villonodular synovitis presentations: monoarticular (nodular) and diffuse; no calcification TABLE 19-4  

FIGURE 19-20. Gout: tophus at hallux interphalangeal joint (arrows). The crystals associated with gouty arthritis may be deposited in the joint capsule, synovium, cartilage, subchondral bone, or periarticular tissues.19 In the foot, radiographic visualization of calcified crystals is best appreciated in the periarticular soft tissues. A collection of MSU crystals in the soft tissue is known as a tophus. Tophus calcification is occasionally seen with chronic tophaceous gout. Although not a pathognomonic finding, calcification of a periarticular soft tissue mass, especially if situated adjacent to erosion, is highly suggestive of gouty arthritis in the proper clinical setting. Small, punctate calcifications can be identified in the soft tissue mass (Figure 1922).

FIGURE 19-21. Pigmented villonodular synovitis, tarsus. There are multiple, well-defined C-shaped erosions adjacent to the soft tissue mass. TABLE 19-5   Arthritis

Characteristics of Joint Diseases Associated with Soft Tissue Calcifications Crystal Deposited Location/Characteristics Periarticular soft tissues (tophus); joint capsule;

Gouty arthritis

Pyrophosphate arthropathy

HADD

Monosodium urate

synovium; cartilage; subchondral bone

Small, punctate calcifications Cartilage (chondrocalcinosis); periarticular soft tissue of Calcium pyrophosphate MPJ, tarsal, and ankle dihydrate (CPPD) joint. Curvilinear calcifications parallel subchondral bone plate Periarticular, including tendons, and bursae. Calcifying tendonitis Hydroxyapatite

Small punctate calcifications vs. large, amorphous calcification adjacent to a joint

Calcium pyrophosphate dihydrate (CPPD) deposition disease is associated with several patterns of joint involvement.20 In general, radiographic features include joint space narrowing, subchondral sclerosis, and loose osseous bodies; these collective findings are known as pyrophosphate arthropathy when associated with CPPD deposition.21 Calcifications can occur in articular and periarticular soft tissues. However, cartilage calcification, or chondrocalcinosis, has received the most attention. The primary crystal associated with chondrocalcinosis appears to be CPPD. Little has been reported in the literature regarding pedal CPPD involvement. Perhaps this is because microscopic examination for crystals is not performed routinely for the workup of acutely symptomatic joints. However, the literature refers to metatarsophalangeal, tarsal (talonavicular, in particular), and ankle joint involvement.22 Chondrocalcinosis is not readily recognized at the tarsal joints, because other bones are superimposed.

FIGURE 19-22. Gout: calcified tophus. Faint calcifications are seen in the soft tissues medial to first metatarsal head (arrows). Calcification of periarticular structures, including tendons and bursae, is also seen with hydroxyapatite crystal deposition disease (HADD).23 The clinical course may mimic the single-joint symptomatology seen with gout and pseudogout.24 The radiographic presentation of HADD, also referred to as calcifying tendonitis, consists of round or oval calcifications within the course of a tendon.25 Linear or punctate calcific densities may be seen along the margins of affected joints. Another presentation can be a rather large, amorphous calcification adjacent to a joint. Detritus Ossicles may be seen in the joint and are referred to as loose osseous bodies. Also referred to as “joint mice,” loose bodies vary considerably in size and architecture. They are occasionally seen in osteoarthritic joints (Figure 1923). Trauma can cause osteophytes or subchondral bone with overlying cartilage to break off. These fragments of bone and/or cartilage can float or become wedged within the joint or synovium. Because many of these loose bodies contain cartilage, faint calcifications may be identified. They tend to enlarge over time. Large osseous bodies or fragments in the midfoot and tarsal joints are suggestive of either posttraumatic arthritis or Charcot neuropathic osteoarthropathy (Figure 19-24). Geode Geode, also referred to as subchondral bone cyst or pseudocyst, is a subarticular area of rarefaction, usually presenting as a geographic lytic lesion. It may be seen in osteoarthritis, rheumatoid arthritis, osteonecrosis, and CPPD disease.26 The geode may mimic erosion viewed en face; this is especially true along the medial aspect of the first metatarsal head. (Because the term cyst implies an epithelial-lined cavitary lesion, the term geode is deemed to be more appropriate.26–28)

FIGURE 19-23. Loose bodies (arrow), osteoarthritis. An osteophyte is also seen along the superior aspect of the first metatarsal head.

FIGURE 19-24. Detritus arthritis (arrow): tarsal Charcot neuropathic osteoarthropathy. Additional findings include diffuse sclerosis and subchondral resorption (talonavicular region and posterior ankle). The typical subchondral geode with sclerotic margin is commonly associated with osteoarthritis (Figures 19-4 and 19-16).27 Its pathogenesis is controversial; the two probable mechanisms are bone contusion29 and synovial intrusion.26 Geode has also been associated with rheumatoid arthritis (Figure 19-25). Its radiographic appearance is identical to the degenerative geode but lacks a sclerotic margin.30 The mechanism of formation is thought to be pannus invading the subchondral bone.31 The mechanism and radiographic appearance of the CPPD geode is similar to that associated with osteoarthritis, except the geode may be larger. The osteonecrosis geode forms as a result of osteoclastic resorption of necrotic bone; radiographically there may be surrounding sclerosis.26

MSU crystals, typically deposited in the soft tissues in patients with gout, may also be deposited in bone.32 This deposition has been associated with chronic tophaceous gout. Multiple focal, geographic areas of bone loss (rarefaction) are seen at these sites (see Figure 19-26). Christman (unpublished data) observed, in a retrospective study of 40 patients with gouty arthritis, that localized rarefaction at the first metatarsophalangeal joint with the absence of erosion is frequently an early radiographic finding. The rarefaction is localized in the medial and superior aspects of the first metatarsal head. Although this finding is nonspecific, in the proper clinical setting it suggests gouty arthritis.

FIGURE 19-25. Rheumatoid arthritis: geode (arrow). Other findings include medial third metatarsal head erosion and joint space narrowing.

FIGURE 19-26. Early gout: rarefaction first metatarsal head and proximal

phalanx base, medially. The latter was actually an erosion viewed en face. Osteopenia Para-articular osteopenia has been described as an early, though not specific, finding in rheumatoid arthritis. It is considered to be an indirect sign of synovitis, a collateral effect of osteoclast activation.33 Alignment Abnormalities Positional deformities may be encountered with joint disorders. Abnormalities range from nonspecific misalignment of bones to subluxation and dislocation. Fibular deviation of the digits, especially the hallux, is associated with rheumatoid arthritis, especially late in the course of the disease (Figure 1927). This finding generally does not involve the fifth digit, however; the constraints of shoe gear probably prevent lateral deviation of this toe. Erosion may or may not accompany misalignment. It is important to note that hallux abductovalgus and lesser toe deformities are nonspecific; these abnormalities are frequently seen in the absence of rheumatoid arthritis. Subluxation and dislocation are frequently encountered in the rheumatoid forefoot. These changes especially affect the lesser metatarsophalangeal joints. The digits dislocate superiorly; superimposition of the proximal phalanx base on the metatarsal head may appear as ankylosis in the dorsoplantar view (Figure 19-28). Metatarsophalangeal joint dislocation is best appreciated with the lateral view, although it is difficult to visualize the joint structures because the adjacent bones are superimposed.

FIGURE 19-27. Rheumatoid arthritis: fibular deviation. Midfoot joint subluxation and dislocation are characteristic features of tarsal Charcot neuropathic osteoarthropathy. This is especially noteworthy at the tarsometatarsal joints, although it can also occur at the intertarsal joints (Figure 19-29). The forefoot dislocates superolaterally relative to the rearfoot. Posterosuperior calcaneal displacement is noted when the talocalcaneal joint is involved. Misalignment between two articular surfaces can result in cartilage damage and subsequent osteoarthritis. Examples include hallux abductovalgus and other medial column misalignments associated with pes planus and pes cavus.

FIGURE 19-28. Rheumatoid arthritis: second metatarsophalangeal joint dislocation simulating ankylosis.

FIGURE 19-29. Neuropathic osteoarthropathy: midfoot subluxation and dislocation. Pes planovalgus is a frequent deformity in the rheumatoid arthritis midfoot. It is also seen with Charcot neuropathic osteoarthropathy. Alignment abnormalities are not commonly observed with seronegative and gouty arthritis. Extra-articular Involvement of Joint Disease Erosion With rheumatoid and seronegative arthritis, erosions may also be found at sites distant from involved joints. The calcaneus is a common location. The site most frequently affected is the bursal projection (posterosuperior aspect). The retrocalcaneal bursa lies over this portion of bone. The bursa is lined by synovium, and the bursal projection is covered with cartilage.34 The bursitis accompanying rheumatoid arthritis and the seronegative arthritides frequently causes rarefaction and erosion of the adjacent calcaneus (Figure 19-30). The erosion may be bounded by sclerosis in some instances. Retrocalcaneal bursitis caused by local trauma or irritation should not in turn cause underlying bone pathology in the absence of infection or systemic inflammatory rheumatic disease. Erosion along the inferior surface of the medial tuberosity can also be encountered. Rarely, calcaneal erosion is associated with gout. Psoriatic arthritis can erode the hallux ungual tuberosity (Figure 19-31). This may be an isolated finding. The outline of the tuberosity may appear irregular and sometimes spiculated (Figure 19-13). This finding alone is not pathognomonic for psoriatic arthritis; normal variation appears similar.

FIGURE 19-30. Rheumatoid arthritis: erosion along calcaneal bursal projection (arrow). Enthesopathy Enthesopathy represents an alteration at any ligamentous or tendinous attachment to bone (i.e., enthesis). It may present as spur formation, erosion, or a combination thereof (Figure 19-32A). Enthesopathy has been associated with many joint disorders (Box 19-12).35,36 Common sites of enthesopathy in the foot are the inferior calcaneal tuberosities and the posterior calcaneus. The fifth metatarsal tuberosity is infrequently affected. BOX 19-12 Differential Diagnosis of Enthesopathy in the Foot Trauma Degenerative disease (well-defined, pointed or hook shaped, inferior calcaneal spur)

Osteoarthritis Diffuse idiopathic skeletal hyperostosis Inflammatory joint disease Rheumatoid arthritis (well-defined, inferior calcaneal spur) Seronegative spondyloarthritis (irregular, large calcaneal spur) Crystal deposition disease CPPD (probable) HADD (probable) Gout (possible) (ill-defined, small, inferior calcaneal spurs) Endocrine disorders Diabetes mellitus Inferior calcaneal spur formation associated with degenerative joint disease and rheumatoid arthritis is generally well-defined. Degenerative spurs commonly are pointed and sometimes hook shaped. However, early spur development, regardless of etiology, may be ill-defined. Calcaneal spur formation related to seronegative arthritis tends to be large and irregular. Illdefined erosion and adjacent sclerosis frequently accompany these spurs (Figure 19-32B). Inferior calcaneal spurs may be seen with gout. They are smaller and ill-defined.

FIGURE 19-31. Psoriatic arthritis: ungual tuberosity erosion (acroosteolysis). Soft Tissue Mass Gouty tophi may be found anywhere in the foot, not just intra- or

periarticular. Rheumatoid nodules are rarely encountered in radiographs of the foot but could appear similarly at extra-articular sites. Patterns of Joint Disease and Distribution of Roentgen Findings Joints Involved Each of the joint disorders (with the exception of septic arthritis) consistently targets specific sites in the foot. Furthermore, joints may be involved that are clinically asymptomatic. Radiographs of both feet (dorsoplantar and lateral views, at a minimum) are needed to assess the pattern of joint disease and distribution of roentgen findings. Box 19-13 lists the primary joints targeted by the more common pedal disorders. The patterns of joint involvement and distribution of roentgen findings are discussed in more detail with the following characteristic descriptions of each joint disorder. BOX 19-13 Primary Target Joints Osteoarthritis

First MPJ

Rheumatoid arthritis Seronegative spondyloarthritis Gout Neuropathic osteoarthropathy

MPJs Lesser MPJs, DIPJs First MPJ Tarsometatarsal and intertarsal joints

DIPJ, distal interphalangeal joint; MPJ, metatarsophalangeal joint. Extra-articular Sites Involved Calcaneal abnormality is frequently associated with joint disease. Both spur and erosion may be encountered at inferior and retrocalcaneal locations. For this reason, lateral views should always be included with dorsoplantar views of the feet when evaluating for joint disease. It is unusual to see erosions of the calcaneus unless they are associated with inflammatory rheumatic disease or infection. Occasionally erosion may be encountered with gouty arthritis at an enthesis, adjacent to a tophus. CHARACTERISTIC PRESENTATIONS OF JOINT DISEASE IN THE FOOT (TABLE 19-6)

Osteoarthritis Osteoarthritis is the most frequently encountered pedal joint disorder. It is not just a disease of the elderly population; in the foot and ankle, regardless of age and weight, it occurs secondary to abnormal wear and tear to a particular joint (pronatory biomechanical abnormality is most frequent37) or is posttraumatic (related to either repetitive low-impact injury, or to a single event, such as a high-impact sports injury38). The term osteoarthritis is misleading, because the disease is not primarily inflammatory in nature. It has also been referred to as degenerative joint disease (DJD), degenerative arthritis, and osteoarthrosis. The term degenerative joint disease is more general and pertains to synovial and nonsynovial locations.39 Osteoarthritis specifies joint disease at a synovial articulation. The disorder is one of chronic cartilage and subchondral bone deterioration; if inflammation is present, it is mild in severity and not a primary pathologic feature. Because the term osteoarthritis is widely accepted, it will be used in this discussion.

FIGURE 19-32. Enthesopathy. A: Rheumatoid arthritis: spur and erosion (arrow). B: Psoriatic arthritis: ill-defined inferior calcaneus erosion and spur surrounded by diffuse sclerosis. TABLE 19-6   Presentation of Joint Disorders in the Foot

Another misnomer is use of the term “destructive” to denote a late-stage osteoarthritic joint. Degenerative implies a slow or chronic “wear and tear” process, whereas destructive reflects an aggressive, acute, and lytic process. Because the latter term may have different connotations, its use should be avoided in this context. Osteoarthritis targets joints along the medial column of the foot (Box 19-14). Its distribution is typically asymmetric, but it is not uncommon to see the same joint or joints affected bilaterally; also, if bilateral, the findings may not be symmetric in severity. It is unusual to see osteoarthritis at the remaining pedal joints unless related to previous trauma. Osteoarthritis develops from one or both of the following settings40,41: (1) abnormal concentration of force/mechanical stress on a normal articulation, which includes its supporting connective tissues (for example, direct trauma, accumulation of repetitive microtrauma, misalignment); and (2) normal concentration of force on an abnormal articulation (an underlying cartilage or subchondral bone abnormality, such as osteonecrosis).

BOX 19-14 Target Joints: Osteoarthritis Hallux interphalangeal First metatarsophalangeal Second metatarsocuneiform Intermediate naviculocuneiform Talonavicular Radiographic evidence of osteoarthritis typically takes years to develop. The characteristic features (summarized in Box 19-15 and illustrated in Figure 1933) include (collectively or individually): BOX 19-15 Osteoarthritis: Radiographic Features Primary finding: osteophyte Uneven joint space narrowing Subchondral sclerosis (eburnation) Geode with sclerotic margin (subchondral cyst) Detritus (loose body, “joint mouse”) usually posttraumatic Gross joint deformity seen in severe osteoarthritis Targets: medial column, especially first metatarsophalangeal joint; other joints if posttraumatic Foot distribution: U/L, B/L, any joint possible; monoarticular or polyarticular. If bilateral, typically asymmetric in location and severity. Seldom involves the lesser toe IPJs, unlike the hand. Ankle joint involvement primarily with history of fracture Elsewhere: hands, knees, spine

FIGURE 19-33. Osteoarthritis. A: First metatarsophalangeal joint. B: Secondary to Freiberg disease or previous trauma, third metatarsophalangeal joint. C: Second metatarsal-intermediate cuneiform joint.  1. Osteophytosis. An osteophyte is a bony outgrowth (spur) at the margin of the affected joint (Figures 19-1, 19-10, 19-16, and 19-23). It frequently is an isolated finding but can present in combination with any or all of the following four findings, especially in more severe cases. Evidence suggests that osteophyte formation does not predict disease progression.42 Normal bone density is maintained and trabecular reorganization occurs in the osteophyte. Osteophytes have also been found in the absence of cartilage loss. As such, osteophyte formation may be a reparative process, or possibly a redistribution response to changes in weightbearing, stress distribution, or microfractures. Though experimentally induced osteoarthritis in the rabbit knee has demonstrated the identification of osteophytes in as little as 2 weeks,43,44 it is a chronic, gradual process under normal circumstances.  2. Joint space narrowing. The joint space narrowing typically is uneven (or nonuniform). Narrowing occurs at the focus of the applied abnormal force or at the site of cartilage or subchondral bone abnormality (Figure 19-16); however, uniform narrowing may be seen if the entire cartilaginous surface is affected (Figure 19-10).  3. Subchondral sclerosis (also called eburnation). This finding is represented by periarticular increased bone density. Subchondral sclerosis is

frequently found adjacent to the site of joint space narrowing. In more severe cases, its appearance is diffuse (Figure 19-10). Sharma et al.45 demonstrated that osteophytosis and sclerosis are inversely related phenomena. There also does not appear to be any correlation between subchondral sclerosis and cartilage degeneration, at least not in the first metatarsal head,46 which may suggest that sclerosis is the process of supporting the cartilage so it does not collapse.42 However, subchondral sclerosis does increase as the disease progresses.1  4. Geode (subchondral cyst) formation. The geode or subchondral (aka “degenerative”) cyst is a geographic, radiolucent, eccentric lesion frequently associated with the osteoarthritic joint. It characteristically has a thin, sclerotic margin (Figures 19-4 and 19-16). The location of this lesion at the margin of a bone may be mistaken for erosion; however, erosion is not a characteristic feature of osteoarthritis and its presence suggests an underlying inflammatory condition.  5. Loose osseous body (or “joint mouse”). The loose body appears as a bone fragment or ossicle within the joint or along its margins (Figure 19-23). It more than likely is related to a traumatic event, which the patient may not ever recall; it could be an osteochondral bone fragment that initiates osteoarthritis (a subchondral defect may be identified) or can represent a fractured osteophyte in an already existing osteoarthritic joint. MRI (Magnetic Resonance Imaging) of Cartilage There is poor to moderate correlation between the radiographic characteristics suggesting degenerative cartilage and the actual degree of cartilage damage. Also, no correlation has been found between subchondral plate thickness, porosity of bone, or osteophytosis and the degree of cartilage degeneration.38,46–49 Recently developed MRI techniques and sequences allow for a sensitive analysis of cartilage from focal lesions to generalized disease.38 The SPGR (spoiled gradient-recalled echo) sequence produces a high cartilage signal, while the signal for the adjacent joint fluid is low, thus making the SPGR sequence the current standard for quantitative morphologic imaging of cartilage.50 Having a long-term precision error of only 1.4 to

3.9%, SPGR sequence may be used for diagnosis, patient follow-up, and longitudinal studies.38,51 Newer MRI techniques, such as dGEMRIC, T1ρ MRI, and NaMRI (sodium MRI), provide a validated quantitative measurement of specific articular matrix constituents, making them useful in the future for early articular cartilage damage evaluation and follow-up.38 Interphalangeal Joints Osteoarthritis seldom affects the interphalangeal joints of the lesser toes, unlike the hands. Joint space narrowing and osteophyte formation are the usual findings; however, joint space narrowing is difficult to ascertain in most cases, because of digital contracture. Furthermore, the tubercles along the margins of the phalangeal bases may appear as osteophytes if the toe is rotated into adductovarus. Hallucal interphalangeal osteoarthritis is frequently posttraumatic in nature and affects the medial aspect of this joint; radiographic findings generally are limited to irregular joint space narrowing. Loose osseous bodies and small geodes are occasionally associated with osteoarthritis at this site. First Metatarsophalangeal Joint The first metatarsophalangeal joint is by far the most commonly affected pedal joint and may demonstrate few or all of the classic radiographic features of osteoarthritis (Figure 19-33A; also, Figures 19-1A, 19-10, 19-16, and 19-23). Mild osteoarthritis typically features an osteophyte, small or large, but with only minimal uneven joint space narrowing. Osteophytes are more likely to be seen at the proximal side of the joint, but may also present along the adjacent proximal phalangeal base. A larger superior osteophyte may be associated with medial enlargement of the metatarsal head. Joint space narrowing is best appreciated in the dorsoplantar view. Loose osseous bodies are more commonly associated with severe osteoarthritis but also can be seen in mild disease. They typically present superiorly and are therefore best visualized in the lateral view. Osteochondral defect is infrequently seen centrally in the first metatarsal head.

Mild to moderate osteoarthritis demonstrates progressive joint space narrowing and osteophytosis in varying degrees. Loose osseous bodies are also more frequent. Mild subchondral sclerosis may become evident. (A welldefined, increased density is normally present along the concave hallux proximal phalanx base centrally and may be mistaken for eburnation. The eburnation associated with osteoarthritis is typically ill-defined and diffuse in appearance.) Osteochondral defects and/or geodes are infrequent but may appear exaggerated relative to the other radiographic findings. Significant joint space narrowing, osteophytosis, and eburnation are seen with moderate to severe osteoarthritis. Geode formation, at either side of the joint or within the hypertrophied medial eminence, is also frequent. Gross joint deformity is seen in severe osteoarthritis. The first metatarsal head and proximal phalanx base are hypertrophied both medially and laterally. Subchondral sclerosis is exaggerated, and subchondral geodes and loose osseous bodies are predominant. Metatarsosesamoid osteoarthritis is not uncommon and is best visualized in the sesamoid axial view. Irregular joint space narrowing and small osteophytes may be seen. Similar findings may be seen in the lateral view but are unlikely, because numerous osseous structures are superimposed. Lesser Metatarsophalangeal Joint Lesser metatarsophalangeal joint osteoarthritis is generally not seen unless the patient has had epiphyseal osteonecrosis (Freiberg disease) or a history of injury to the joint. Gross deformity of the metatarsal head and proximal phalangeal base are predominant findings (Figure 19-33B). Metatarsocuneiform Joints Osteophytosis, subchondral sclerosis, and joint space narrowing can be visualized in both the dorsoplantar and lateral views. Osteoarthritis of the first metatarsocuneiform joint is frequently associated with osteoarthritis of the second metatarsocuneiform joint; however, the converse is not. Osteoarthritis of the second metatarsocuneiform joint is best identified in the

lateral view (Figure 19-33C). The primary finding is dorsal osteophytosis; lack of visualization of a joint space and subchondral sclerosis may also be apparent. Joint disease is not easily visualized in the dorsoplantar or medial oblique views unless the arthritis is severe. Osteoarthritis of the third metatarsocuneiform joint is not frequently encountered. When present, joint space irregularity and eburnation are best noted in the medial oblique and lateral views. Naviculocuneiform Joints Osteoarthritis of the medial naviculocuneiform joint primarily consists of joint space irregularity and subchondral sclerosis, with or without mild osteophytosis. These findings are best seen in the dorsoplantar view, although joint space narrowing can also be seen along the inferior aspect of this joint in the lateral view. Intermediate naviculocuneiform osteoarthritis is best visualized in the lateral view and is characterized by dorsal osteophytosis, joint margin irregularity, and eburnation (Figure 19-1B). Similar findings may be seen in the medial oblique view; the dorsoplantar view seldom is useful for assessment at this location. Intermediate naviculocuneiform osteoarthritis is frequently associated with osteoarthritis of the second metatarsocuneiform and medial naviculocuneiform joint. Talonavicular Joint Talonavicular joint osteoarthritis is best appreciated in the lateral view. Radiographic findings vary from large osteophyte formation superiorly with normal joint space to significant joint narrowing with subchondral sclerosis. Occasionally an osteochondral defect will be seen along the navicular posterior articular surface associated with osteoarthritis at the talonavicular joint. Calcaneocuboid Joint Osteoarthritis is seldom seen at the calcaneocuboid joint. The typical finding, when present, includes an enlarged, hook like exostosis laterally, originating

from the calcaneus and seen primarily in the dorsoplantar view; the joint space is relatively spared, and the lateral view appears normal. When more pronounced, eburnation and joint space narrowing can also be seen in both the dorsoplantar, medial oblique, and lateral views. Talocalcaneal Joint The middle and posterior talocalcaneal joints are not easily visualized with radiography, and osteoarthritis is not common. However, posterior talocalcaneal osteoarthritis may be recognized in the medial oblique and lateral ankle views as joint space narrowing and osteophyte production, and less frequently with calcaneal axial views at varying degrees of tube head angulation. Incomplete or fibrocartilaginous coalition is often associated with evidence of middle talocalcaneal joint osteoarthritis in the lateral view, which presents as joint space narrowing and eburnation. Computed tomography (CT) is valuable for further assessment of this joint. Ankle Joint Unlike the knee and hip, primary osteoarthritis of the ankle is rare.52 It has a very high association with a previous history of fracture or significant ankle trauma (78%53 and 70%52). Of the world’s adult population, approximately 15% is affected by osteoarthritis54; approximately 1% of that 15% have ankle involvement.55 Fracture of the malleoli is the most severe risk factor for development of ankle osteoarthritis, accounting for about 39% of the ankle osteoarthritis cases.53 In contrast, the painful ankle in a geriatric patient with no prior history of trauma seldom demonstrates radiographic evidence of osteoarthritis, or mild at best. Posttraumatic ankle arthritis demonstrates varied radiographic features, depending on the type of injury and its extent. Findings may include osteophytosis (best seen in the lateral view), uneven joint space narrowing, subchondral sclerosis, and subchondral lucency (related either to the fracture site itself, an osteochondral defect, or geode formation). Rheumatoid Arthritis Rheumatoid arthritis targets synovial tissue, especially of the foot and hand.

The acute synovitis and associated pannus formation result in lysis of intraarticular structures, including cartilage and subchondral bone. Capsular and ligamentous laxity also occurs. Clinically, the onset is insidious and its course demonstrates periods of remission and exacerbation. The majority of these patients are seropositive for rheumatoid factor. Its etiology, however, is unknown but is regarded as an autoimmune disorder33. A new classification system has redefined the 1987 American College of Rheumatology system to one that features earlier stages of the disease.56 It has been widely tested in many patient populations and appears to be a sensitive instrument.57 In general, there are four classification criteria: joint involvement, serology, acute-phase reactants, and duration of symptoms. The radiographic appearance of early rheumatoid arthritis of the foot differs from its well-established form. The latter classically involves all metatarsophalangeal joints bilaterally and occasionally the hallux interphalangeal joints. In contrast, early rheumatoid arthritis is often monoarticular or polyarticular but asymmetric. The fifth metatarsophalangeal joint is frequently the first joint affected. Findings at the first metatarsophalangeal joint are generally subtle or unremarkable. Although rheumatoid arthritis may involve the tarsal (especially the talocalcaneal) joints, early recognition is difficult with radiography. CT may be necessary for further evaluation of these joints. Unlike the hand, it is rare to see involvement of lesser toe interphalangeal joints. The radiographic grading system proposed by Larsen et al.58 is still widely used for grading rheumatoid arthritis (Table 19-7). Scott et al.59 later modified the Larsen system to provide more sensitivity for grading early and mild radiographic changes. These two sets of descriptors were then combined by Edmonds (Table 19-8).60 Although radiography still plays a great role for imaging rheumatoid arthritis, it is not sensitive enough to detect the earliest signs (hyperemia, synovitis, effusion, and para-articular osteopenia) prior to erosion, which are valuable for early diagnosis and treatment to delay joint destruction. TABLE 19-7  

Larsen System for Radiographic Grading of Rheumatoid

Grade 0

1

2

3

4

5

Arthritisa,58 Definition

Radiographic Findings Abnormalities not related No radiographic changes to rheumatoid arthritis may be present Periarticular swelling, para-articular osteopenia, Slight abnormality slight joint space narrowing Erosion (except in weightDefinite early abnormality bearing joints), joint space narrowing Erosion (in all types of Medium destructive joints), joint space abnormality narrowing Erosion, joint space Severe destructive narrowing, bone abnormality deformation in weightbearing joints Mutilating abnormality

Disappearance of the original articular surfaces, gross bone deformities in weight-bearing joints

aImages (“standards”) were included in the original publication for more

specific correlation to the different grades. MRI and high-resolution ultrasound have been increasingly used to diagnose and monitor synovial and early bony alterations.61 However, ultrasound, like radiography, is unable to show the entire spectrum of the disease. CT, unlike MRI, is not able to image acute inflammation and bone edema, early features of rheumatoid arthritis; for these reasons, and because CT uses ionizing radiation, it may be limited to preoperative settings where a precise assessment of bone destruction and stability is absolutely needed.33 Thus, MRI has been the best imaging modality because it shows soft tissue changes as well as damage to cartilage and bone at an earlier stage.

TABLE 19-8   Grade 0

1

2 3

Edmonds System for Radiographic Grading of Rheumatoid Arthritis60 Definition Radiographic findings Abnormalities that are not No radiographic changes related to rheumatoid arthritis may be present Major periarticular soft tissue swelling, and/or major para-articular osteopenia, and/or slight Slight abnormality joint space narrowing, and/or suggestion of erosion or geode at two sites that are 1 Definite abnormality mm, joint space narrowing Erosion (of significant Medium destructive size), but with some joint abnormality preservation

4

Severe destructive abnormality

5

Mutilating abnormality

Erosion (large), significant loss of articular surface but only partial joint preservation, and/or subluxation Disappearance of the original articular surfaces, gross bone deformities, and/or surgical modification

MPH-SPECT (multipinhole single-photon emission computed tomography), a new technology method initially developed for small-animal imaging and recently employed for human arthritis imaging, not only demonstrates pathologic tracer accumulation in areas matching the bone marrow edema and erosions detected by MRI, but also shows increased metabolism in parts of the bone that appeared normal on MRI.62,63 MPH-SPECT provides

improved spatial resolution compared to bone scintigraphy, with central tracer uptake in early rheumatoid arthritis and eccentric uptake in early osteoarthritis. In early rheumatoid arthritis, although synovial inflammation may extend into bone, MR images may be negative, not capturing early bone alteration, but rather showing normal signal intensity of bone marrow.62,64– 66

Due to the fact that the course of disease progression is not linear and that joint involvement is not uniform, different joints in the same individual may present different radiologic findings at the same time, especially in the early stages of rheumatoid arthritis. As such, each of the involved joints needs to be evaluated individually and independently. Since there is no general consensus as to which joints to be imaged, typically the involved (symptomatic) joints should be evaluated by radiography for diagnostic purposes. The foot joints could be used to evaluate the disease activity when initially affected, but, in general, the wrist and hand joints are used to assess the progression of the disease.33 The radiographic features of rheumatoid arthritis in the foot (summarized in Box 19-16 and illustrated in Figure 19-34) include the following: BOX 19-16 Rheumatoid Arthritis: Radiographic Features Primary finding: erosion along medial side of affected joints only (except at fifth metatarsophalangeal joint) Even joint space narrowing Para-articular osteopenia Loss of joint apposition (subluxation/dislocation) Unlike the hands, rarely involves the interphalangeal joints of lesser toes Fibular deviation of toes 1–4 at metatarsophalangeal joint level Erosion of calcaneal bursal projection associated with retrocalcaneal bursitis

Progressive ankylosis (tarsus) may occur end stage Targets: all metatarsophalangeal and hallux interphalangeal joints; calcaneus; tarsus (infrequent) Foot distribution: B/L and symmetrical (classic); monoarticular initially, then polyarticular Elsewhere: hand/wrist, cervical spine  1. Increased soft tissue density and volume. Synovial hyperemia is the first step in the inflammatory process. However, if there is no profound soft tissue edema or joint effusion, no radiographic changes will be seen.67,68 MRI is best for identifying pathology in this stage.69 The association of synovial hyperemia and synovium enhancement has been documented in the literature, with the degree of enhancement being assessed qualitatively and quantitatively. Although results are not uniformly accepted, Huang et al.70 associated the enhancement rates with the degree of joint inflammation and the probability of developing erosion after 1 year.

FIGURE 19-34. Classic rheumatoid arthritis. A: Left foot. B: Right foot. Findings include medial metatarsal head erosion (medial and lateral at fifth),

no evidence of new bone formation, dislocation (left foot), even joint space narrowing, and no lesser toe IPJ involvement.  2. Joint space widening. As a result of synovitis, the inflammatory tissue thickens at the bare areas and gradually extends into the joint space across the cartilaginous surfaces; this can become more pronounced with joint effusion.33 Effusion is frequently associated with acute inflammation in early stage of the disease, as well as in exacerbations. Radiography demonstrates only indirect signs of effusion, such as joint space widening, increased soft tissue density and volume, and shifting of fat pads. However, MRI can be used to assess synovial volume and swelling as well as differentiating between effusion and synovium; MRI is, therefore, clearly superior to radiography for assessing synovial disease.71,72  3. Para-articular osteopenia. Osteoclast activation near the joint causes radiolucency of bones adjacent to joints. It can be quite subtle and is considered an early radiographic sign of joint involvement.33  4. Geode. Geographic, subchondral radiolucent lesions (sometimes referred to as subarticular cysts or “pseudocysts”) occasionally are seen centrally at affected joints (Figure 19-25). They may contain fluid, synovium, or both and may or may not communicate with the joint.33 The advanced stages of synovitis are almost always accompanied by geodes, which are considered a preerosion stage. The subchondral bone plate of affected bones is frequently ill-defined with their presence.  5. Erosion. Fex et al.73 concluded that radiographic evidence of joint damage occurred early in the disease course; during the first 2 years, its progression is most rapid. Erosion occurs at the insertion of interosseous ligaments and at synovial reflections.74 Erosion typically is seen along the medial side of affected forefoot joints (Figures 19-5, 19-15, 19-25, and 1928); lateral involvement is infrequent except at the fifth metatarsophalangeal joint, which is common (Figure 19-9). Erosion may also be seen along the posterosuperior aspect of the calcaneus in the lateral view adjacent to the retrocalcaneal bursal recess (Figure 19-30). It is not unusual to see erosion along the inferior aspect of the calcaneal medial tuberosity, with or without adjacent spur formation (Figure 19-32A). Later in the course of disease, the

size of the erosion, rather than the number, appears to increase, becoming a more sensitive indicator of erosion progression.75 Unfortunately, radiography has been shown to be insensitive for cancellous bone erosion, especially in the early stages of the disease; superimposition of overlapping structures may also obscure erosion.74 Out of all imaging modalities, MRI shows the greatest sensitivity for detecting and monitoring bone erosion even before manifested on the radiograph.76,77 This may be due to the fact that MRI detects early bone marrow edema and geodes, which may resemble preerosion.33,78,79 Sometimes, contrast enhancement helps distinguish erosion and preerosion from geodes, which are less likely to enhance.78 MRI can also detect and follow preerosive changes in rheumatoid arthritis, such as synovitis, bone marrow edema or osteitis, and tendinous and ligamentous abnormalities. Scintigraphy is very sensitive but not very specific in detecting synovial and bone erosions, as it detects inflammation and bone turnover at sites of active erosions.74 Although scintigraphy lacks spatial resolution, it offers greater anatomical coverage making whole body assessment possible. Even so, scintigraphy offers a great correlation between the presence of synovitis and future joint erosion in early rheumatoid arthritis.80  6. Joint space narrowing. This feature of an affected joint often is evenly distributed (or uniform) across the joint (Figures 19-5, 19-15, 19-25, and 1928). However, mild subluxation can occur simultaneously, giving the appearance of uneven narrowing. In contrast to the eccentric joint space narrowing in osteoarthritis, the narrowing is predominantly concentric in rheumatoid arthritis. This finding, initially considered a sign of early disease (when radiography was the only imaging available), now it is considered a sign of advanced rheumatoid arthritis, being caused by the progression of the destructive process and the formation of scar tissue and fibrosis.33,81  7. Digital misalignment and joint subluxation/dislocation. All toes (except the fifth) generally deviate in a fibular direction relative to the metatarsals (Figure 19-27); however, this finding is not specific for rheumatoid arthritis. As the disease progresses, the proximal phalangeal bases may only partially appose their respective metatarsal heads (subluxation) and can further result in dislocation (0% apposition) (Figures 19-27 and 19-28).

 8. Ankylosis. In the late stage of rheumatoid arthritis, the joint space may disappear entirely and the two bones appear united as one. Ankylosis appears more frequently in the tarsus (Figure 19-18B) but can also affect the first ray. Due to medical therapy improvements and earlier detection of the disease, ankylosis is much less frequent then in the past. Classic Forefoot Rheumatoid Arthritis The classic radiographic picture of pedal rheumatoid arthritis is bilateral and symmetric (Figure 19-34): medial erosions at the first through fourth metatarsophalangeal and the hallux interphalangeal joints, and medial and lateral erosion at the fifth metatarsophalangeal joint; uniform joint space narrowing at affected joints, but subluxation and/or dislocation prevents this from being visualized in many cases; fibular deviation of the toes (with the exception of the fifth toe) at the metatarsophalangeal joints; and para-articular osteopenia. Hallux Interphalangeal Joint Erosions, when found at the hallux interphalangeal joint, are typically along the medial aspect. It is not an isolated finding and usually appears in association with metatarsophalangeal joint involvement. Metatarsophalangeal Joint Initial findings in early rheumatoid arthritis may present as monoarticular or oligoarticular and unilateral. Erosion is frequently identified early in the disease at the fifth metatarsophalangeal joint. Or, only two or three metatarsophalangeal joints may be affected. Asymmetric, bilateral joint involvement is also seen. The erosion of rheumatoid arthritis tends to target the medial margin of metatarsophalangeal joints. Tenosynovitis, ligamentous laxity, and resultant fibular deviation of digits associated with this disease leave the medial margins of the joints unprotected from the destructive pannus; it is believed that this accounts for the medial erosion.39 However, one of the earliest sites for erosion in the rheumatoid foot is the lateral aspect of the fifth metatarsal head. The fifth digit usually does not deviate laterally, because of the

constraints within the shoe. This may allow equal distribution of the pannus along both margins. Marginal erosions eventually progress to involve the entire subchondral surface. Bursitis is a characteristic clinical feature of rheumatoid arthritis. It is interesting to note that several small bursae are found adjacent to all metatarsal heads: Subcutaneous bursae are found along the dorsomedial and dorsolateral aspects of the first and fifth metatarsal heads, respectively; subfascial synovial bursae are located between the medial collateral ligaments of the lesser metatarsal heads and the interosseous tendons; another bursa is found between the abductor tendon and the fifth metatarsal head.82 Coexisting bursitis may contribute to the specific marginal erosions. Tarsometatarsal and Intertarsal Joints Occasionally, the tarsometatarsal and intertarsal joints are afflicted, frequently without associated forefoot involvement. Typical radiographic features are even joint space narrowing and, occasionally, large geographic lucent lesions or geodes. Eventually osseous ankylosis results (see Figure 1918B). To recognize involvement of the talocalcaneal joints, cross-sectional imaging studies may be necessary. In patients with rearfoot pain, posterior tibial tendon synovitis may be observed even in the absence of sinus tarsi synovitis.83 Ankle Joint Radiographic findings are often nonspecific at the ankle joint. Increased soft tissue density and volume (joint effusion) may be the only feature. Eventually, even joint space narrowing and irregular subchondral surfaces are recognized. Secondary osteoarthritis, including subchondral sclerosis and osteophytes, frequently accompany these findings. There may be symptoms associated with sinus tarsi and tarsal tunnel syndromes.33 Calcaneus Bursitis is often associated with rheumatoid arthritis. Radiographically, erosion may be seen inferiorly along the medial tubercle or at the bursal

projection (along the posterosuperior aspect of the calcaneus) adjacent to the retrocalcaneal bursa (Figure 19-30). Spurs have been associated with rheumatoid arthritis. However, when present, they are well-defined. They do not demonstrate the associated new bone production seen with the seronegative spondyloarthritides. Psoriatic Arthritis Psoriatic arthritis, a seronegative spondyloarthritis, presents in anywhere from 1% to 39% of patients with psoriasis.84 However, of those patients with psoriatic arthritis, nearly 70% develop psoriasis before joint involvement; approximately 15% of patients develop psoriasis following the arthritis. The presence of psoriatic arthritis is frequently associated with a long duration of psoriatic skin and/or nail involvement.85,86 Most of these patients are seronegative for rheumatoid factor but show strong association with the HLA-B27 antigen (positive in 25%–60% of patients with “central arthritis”).87,88 Pathologically, synovial involvement is similar to rheumatoid arthritis but is not as intense, with less hyperplasia but more tortuous vascularity89. Psoriatic arthritis has many variants reported in the literature, making it difficult to classify and diagnose. Manifestations range from spinal disease; enthesitis; dactylitis; and peripheral mono-, oligo-, or polyarticular arthritis that may or may not be symmetric.90 The small joints of the hands and feet are most commonly involved. Since there is no confirmatory lab test and because there are various clinical presentations, numerous diagnostic/classification criteria have been proposed over the past three plus decades91; they have been summarized and reviewed in the literature by Cantini et al.84 and Helliwell and Taylor.92 Moll and Wright initially defined psoriatic arthritis in patients that had the following criteria: seronegative rheumatoid factor; current psoriasis of nails or skin; and inflammatory joints of axial disease.93,84 Five clinical subsets were also identified that included distal interphalangeal joint involvement, oligoarthritis (four or less joints affected in an asymmetric distribution), polyarthritis (symmetrical and indistinguishable from rheumatoid arthritis),

spondylitis and sacroiliitis, and arthritis mutilans (Box 19-17).94 This system has been the simplest and most frequently used.92 However, the various clinical patterns may overlap.93 More recently, another rare subset has been reported, psoriatic onychopachydermoperiostitis (POPP), which appears as a drumstick-like deformity of the digits.95 BOX 19-17 Clinical Presentations of Psoriatic Arthritis 1. Predominant distal interphalangeal joint involvement (“classic” psoriatic arthritis) 2. Asymmetric metatarsophalangeal and interphalangeal joint involvement (mono- or oligoarticular) with associated swelling (“sausage toe”/ray distribution); the most common foot presentation 3. Symmetric metatarsophalangeal joint involvement (polyarthritis identical to rheumatoid arthritis) 4. Spondylitis 5. Arthritis mutilans (osteolysis; associated with sacroiliitis) 6. Psoriatic onychopachydermoperiostitis; rare The most recent criteria, known as the Classification of Psoriatic Arthritis (CASPAR),96 has taken the Moll and Wright subgroups and applied more specific criteria. It uses a scoring system to identify psoriatic arthritis to facilitate early diagnosis, even in the absence of psoriasis or family history of psoriasis.91 Spinal involvement with any form of peripheral joint disease is more common than the isolated spondylitis presentation and is seen in 20% to 40% of patients.88 Spondylitis seen in psoriatic arthritis is less severe than in patients with ankylosing spondylitis. According to Gladman,89 only 2% of patients with psoriatic arthritis have isolated spinal involvement, while 10% of patients with ankylosing spondylitis have psoriasis. Ankylosing spondylitis, reactive arthritis, and diffuse idiopathic skeletal hyperostosis

(DISH), which is a variant of osteoarthritis, should be included in the differential diagnosis of psoriatic arthritis. Axial joints are affected in up to 36% of patients with psoriatic arthritis.90,97 Spondylitis is uncommon in the absence of sacroiliitis. Bony bridges are large, coarse, and unilateral or bilateral but asymmetrical. They commonly occur in the upper lumbar, lower thoracic, and lower cervical vertebrae88,98,99. Psoriatic arthritis presents more cervical involvement and less severe changes in the spine compared to ankylosing spondylitis. The paramarginal “chunky” syndesmophytes of psoriatic arthritis do not appear in consecutive vertebrae and are uncommon in ankylosing spondylitis.100–102 Evidence of sacroiliac joint radiographic change may be seen in 10% to 25% of patients with moderate to severe psoriatic skin disease and in 30% to 50% of patients with peripheral joint disease.88 CT may be useful in assessing elements of spinal disease, but has little role in peripheral joints. As discussed for rheumatoid arthritis, MRI has been more sensitive than radiographs for detecting synovitis and early erosion in inflammatory joint disorders.102,103 However, while bone edema may be a strong predictor of bone erosions in rheumatoid arthritis, this has not been specifically demonstrated in psoriatic arthritis.104 Subchondral bone edema may be associated with pronounced periarticular edema of the soft tissues, spreading to the subcutis, and potentially constituting a “psoriatic pattern” on MRI.105 Extensive MRI bone edema was described in the proximal phalanx with inflammatory changes in the overlying soft tissues of a patient with reactive arthritis, which has identical radiographic features with psoriatic arthritis.106– 108

Bone scintigraphy confirms the presence of inflammation at joints and entheses that may not be apparent radiographically.84,102 The lack of periarticular uptake in psoriasis may be valuable in confirming the lack of subclinical inflammatory arthritis.88 Ultrasound has also been used to assess early enthesitis.109 The radiographic features of psoriatic arthritis in the foot (summarized in Box 19-18 and illustrated in Figures 19-35 through 19-38) include the following:

 1. Bone proliferation. Several forms of new bone production may be seen with psoriatic arthritis, which help to distinguish it from rheumatoid arthritis. Examples include “whiskering,” “ivory phalanx,” and periostitis. Bone proliferation may also appear as ill-defined increased bone density. BOX 19-18 Psoriatic Arthritis: Radiographic Features Primary findings: erosion or arthritis mutilans; erosion occurs at medial and/or lateral sides of affected joints and is marginal initially Joint space alteration: narrowing or widening Periostitis adjacent to affected joint (linear or fluffy) Whiskering, ivory phalanx (hallux distal phalanx) Sausage toe Acro-osteolysis Lack of juxta-articular osteopenia Enthesitis (especially calcaneus) Targets: hallux interphalangeal and all metatarsophalangeal joints; distal interphalangeal joint lesser toes; joints of same ray; hallux distal phalanx; calcaneus; tuft Foot distribution: bilateral, asymmetrical (classic) or symmetrical; polyarticular, one ray Elsewhere: hands, sacroiliac joints, spine (lower thoracic and upper lumbar)

FIGURE 19-35. Psoriatic arthritis: ill-defined bone proliferation adjacent to medial and lateral marginal erosions, producing the “mouse-ear” appearance (hallux interphalangeal joint).       Whiskering and the ivory phalanx, sometimes found together, typically affect the hallux distal phalanx. Whiskering appears as small spicules of bone arising perpendicular from the phalanx shaft (Figure 19-12). The ivory phalanx appears as an overall increased density of the phalanx (Figure 19-13); this is almost always associated with soft tissue swelling and nail abnormality of the same digit.11,88 Increased bone density may also involve proximal phalanges adjacent to affected metatarsophalangeal joints.       Periosteal new bone production is seen adjacent to affected metatarsophalangeal joints in early psoriatic arthritis, especially along the phalangeal shaft (Figure 19-11).88,110,111 The only radiologic differentiation between psoriatic arthritis and rheumatoid arthritis found by the most recent classification of psoriatic arthritis in the CASPAR study is the juxta-articular bone formation.89,98       Bony proliferations may also present as irregular, spiculated or fluffy outgrowths along the margins of periarticular erosions; radiographically, at the hallux interphalangeal joint, this may produce a “mouse-ear” appearance (Figure 19-35) (in contrast to the “gull-wing” appearance seen in erosive osteoarthritis).88,112       Significant plantar and posterior calcaneal spurring is another form of bone proliferation.  2. Enthesopathy. Although more common in reactive arthritis, calcaneal changes may be noted in psoriatic arthritis. Findings include spurs, sometimes quite large, that are irregular due to ill-defined erosion and proliferative new bone, which may be found along both the posterior and plantar surfaces of the calcaneus (Figure 19-32B).88,34 Large portions of the calcaneus may even become sclerotic.86       Isolated entheseal pain may be the first clinical presentation of

psoriatic arthritis. With MRI, calcaneal enthesopathy has been described as perientheseal bone marrow edema and peritendinous soft tissue swelling around the Achilles tendon.88,106,113 Bone edema, however, is not always seen adjacent to regions of enthesitis, being noted in less than 50% of cases.107 MRI reveals typical abnormalities at entheses, which correspond to hot spots on radionuclide scans.       Ultrasound has been used to detect early signs of enthesitis, demonstrating a higher sensitivity than MRI in this regard; however, the finding is not specific for psoriatic arthritis.114 Enthesitis at the Achilles tendon insertion has been identified by ultrasonography much more frequently than on clinical examination in patients with psoriatic arthritis.109  3. Erosion. Loss of the marginal subchondral bone plate may appear no different from that seen in rheumatoid arthritis in the early stages. However, as the disease progresses, the erosions become irregular and ill-defined secondary to new bone formation adjacent to them.102 Erosion may be seen at any forefoot joint, including interphalangeal (Figure 19-36) and metatarsophalangeal joints. In contrast to rheumatoid arthritis, erosions frequently affect both the medial and lateral margins joints; severe erosion can lead to arthritis mutilans. Erosion can also occur centrally within the joint and be quite destructive visually, leading to widening of interphalangeal joints.       Erosion may also occur at extra-articular locations. The smooth or irregular erosion/resorption of the distal phalangeal tuft (acro-osteolysis) is associated with soft tissue edema and adjacent interphalangeal joint abnormality (Figure 19-31).88 Ill-defined enthyseal erosion also occurs.  4. Joint space narrowing. Narrowing of joint spaces is more frequently seen early in disease at the metatarsophalangeal joints, similar to that in rheumatoid arthritis (Figure 19-37).  5. Joint space widening. Central erosion of an interphalangeal joint results in a relative widening of the joint space (Figure 19-19B). Aggressive erosion along the medial and lateral aspects of a metatarsal head also results in apparent joint space widening.

 6. Arthritis mutilans. Excessive bone erosion along both the medial and lateral margins of a metatarsophalangeal joint can result in arthritis mutilans (Box 19-11, Figure 19-7). It may also affect the interphalangeal joint.  7. Bone density. Although some patients might be found to have osteoporosis, bone density is usually normal, even in patients with advanced articular changes of psoriasis.88,112 Juxta-articular osteoporosis, seen in rheumatoid arthritis, is not a feature of psoriatic arthritis.  8. Ankylosis. Osseous ankylosis may involve interphalangeal joints of lesser toes (Figure 19-18A).

FIGURE 19-36. Psoriatic arthritis: distal interphalangeal joint.

FIGURE 19-37. Psoriatic arthritis: ray involvement, fourth toe (interphalangeal and metatarsophalangeal joints), including distal interphalangeal joint erosion and increased soft tissue volume. There also is involvement of the hallux (erosion and increased bone density), narrowing of the third metatarsophalangeal joint, and periostitis of the third toe proximal phalanx shaft and proximal metaphysis, medially. Lesser Toe Distal Interphalangeal Joints

Psoriatic arthritis may target the lesser digit distal interphalangeal joints (Figure 19-36). It can affect one or multiple joints. Findings include erosion at the margins of the joint, which may progress to osteolysis of the entire articular surface (Figure 19-7). Adjacent sclerosis may accompany the erosion. Hallux Distal Phalanx and Interphalangeal Joint The hallux distal phalanx and interphalangeal joint are common sites of involvement. Erosions, frequently accompanied with new bone production, present along the medial and lateral aspects of the joint (Figure 19-35). This helps to differentiate psoriatic from rheumatoid arthritis; the latter generally affects only the medial aspect of this joint.115 Furthermore, rheumatoid arthritis rarely has new bone production associated with hallux interphalangeal joint erosion. New bone production (whiskering and the ivory phalanx) frequently accompanies hallucal psoriatic arthritis (Figures 19-12 and 19-13). Ungual tuberosity erosion may be an isolated radiographic finding. Psoriatic onychopachydermoperiostitis,116 also known as psoriatiform onychopachydermoperiostitis of the great toes or OP3GO syndrome,117 is a rare form of psoriatic arthritis, with less than 20 cases reported worldwide. It is characterized by nail changes (onycholysis), painful soft tissue swelling at distal end of digits (giving the drumstick shape),88 “whiskering-type” periosteal changes of the distal phalanx,118 generalized bone marrow edema of the distal phalanx (on MRI), while sparing the interphalangeal joint and often not yet having the hallmark psoriatic skin lesions (Figure 19-38). This type of psoriatic arthritis does not fit into the established Moll and Wright classification,93 but rather in the Helliwell et al. classification,119 which contains three subgroups: peripheral arthritis, spondyloarthritis, and extraarticular osseous disease. Inflammation beginning at entheses and then spreading toward articular structures may cause this rare disease.118,120

FIGURE 19-38. Psoriatic arthritis: psoriatic onychopachydermoperiostitis (POPP). Drumstick-shaped end of toe, ungual tuberosity erosion, whiskering. Ray Involvement Another presentation of psoriatic arthritis targets the joints of an entire ray. The metatarsophalangeal and proximal and distal interphalangeal joints are affected (Figure 19-37). This is commonly associated with the clinical presentation of a “sausage” toe. Periostitis may also be seen near affected joints. Involvement of all the joints of the same digit has been referred to as “ray” distribution in psoriatic arthritis.89 Metatarsophalangeal Joints Psoriatic arthritis frequently involves multiple metatarsophalangeal joints.

One extremity alone may be affected. If bilateral, the distribution usually is asymmetric; for example, the second and third metatarsophalangeal joints may be affected in the right foot and only the fourth metatarsophalangeal joint in the left foot. Erosions could be medial only, but typically affect both sides of the joint; this may lead to arthritis mutilans (Figure 19-7). Metatarsophalangeal joint findings are frequently associated with adjacent periostitis and digital involvement (Figures 19-11 and 19-37). Calcaneus Enthesitis may be seen along the posterior and/or inferior calcaneus as large, ill-defined spurs with adjacent erosion (Figure 19-32B). These findings are identical to those seen in reactive arthritis and ankylosing spondylitis. The Other Seronegative Spondyloarthritides All seronegative spondyloarthritides, psoriatic arthritis, ankylosing spondylitis, reactive arthritis, and enteropathic arthritis, share common clinical and radiologic features: an association with the histocompatibility antigen HLAB27; peripheral arthritis of the foot that is oligoarticular and asymmetric; and the possibility of sacroiliitis, spondylitis, enthesitis, and uveitis.121 The predominant symptoms are chronic low back pain and/or peripheral arthritis. Erosion will often have new bone proliferation adjacent to it, unlike rheumatoid arthritis. The pathogenic hallmark of seronegative spondyloarthritis is enthesitis, inflammation at entheses, which is where soft tissue structures (ligament, tendon) insert on to bone.36 Therefore, in the proper clinical setting, erosion along the inferior calcaneus associated with new bone proliferation is a particularly suggestive finding (Figure 19-32B). Ankylosing spondylitis (Box 19-19) is the prototypical seronegative spondyloarthritis.122 It primarily involves the sacroiliac joint and the axial skeleton, but also affects entheses and peripheral joints. Enthesitis of the calcaneus is a hallmark presentation, including ill-defined bone production and erosion, especially plantarly. The typical pattern of peripheral joint involvement is asymmetrical and mono- or oligoarticular, mimicking that of

psoriatic arthritis in the foot. Ankylosing spondylitis, although seldom affecting the foot, has a predilection for multiple metatarsophalangeal and interphalangeal joints. Periostitis near affected joints and hallux interphalangeal joint involvement are not uncommon. Another hallmark of ankylosing spondylitis has been described as “ankylosis that occurs within a short period of time.”123 BOX 19-19 Ankylosing Spondylitis: Radiographic Features Primary finding: erosion (medial and/or lateral sides of affected joints) Even joint space narrowing Periostitis adjacent to affected joint Targets: metatarsophalangeal joints, hallux interphalangeal joint; calcaneus Foot distribution: bilateral, symmetric or asymmetrical; polyarticular Elsewhere: sacroiliac joints, hips, shoulders, spine (thoracolumbar and lumbosacral junctions) Reactive arthritis (Box 19-20), previously known as Reiter syndrome, is clinically characterized by the triad of arthritis, urethritis, and conjunctivitis, which occurs between a few days and up to 6 weeks after a preceding bacterial infection of the gut with enterobacteriae (Shigella) or the urogential tract (Chlamydia).121 The infection remotely triggers a sterile, reactive inflammatory oligoarthropathy (asymmetrical).88 The characteristic anatomic radiographic pattern (asymmetrical involvement of lower extremities and entheses, and ill-defined bony erosions with adjacent bony proliferation) aid in the diagnosis of reactive arthritis. Although it can involve any forefoot joint, reactive arthritis affects fewer joints than do psoriatic arthritis and ankylosing spondylitis. It also is the only other joint disorder that shares with psoriatic arthritis the features involving the lesser toe distal interphalangeal joints and hallux (Figure 19-36), including dactylitis.124 However, there is inconstant spine involvement, similar to psoriatic arthritis but unlike ankylosing spondylitis. Reactive arthritis, unlike psoriatic, spares the hand,

tends to have juxta-articular osteopenia, and demonstrates less ankylosis.125 BOX 19-20 Reactive Arthritis: Radiographic Features Primary finding: erosion (medial and/or lateral sides of affected joints) Joint space narrowing and widening Sausage toe Periostitis adjacent to affected joint Targets: metatarsophalangeal and interphalangeal joints; calcaneus Foot distribution: bilateral and asymmetrical (classic); polyarticular Elsewhere: knee, ankle, sacroiliac joints The inflammatory arthritis that accompanies Crohn’s disease and ulcerative colitis is referred to as enteropathic arthritis. The most common radiographic findings are increased soft tissue volume and density and paraarticular osteopenia. The peripheral arthritis is described as pauciarticular, asymmetric, and migratory.1 Small erosions may be observed, but permanent destruction is rare.125 Gouty Arthritis The clinical presentation of acute gout is quite dramatic. Findings include severe swelling, redness, and pain at the affected joint. Left untreated, chronic gout may lead to disabling bizarre deformity with complete destruction of the affected joint.126 Patients with gout may exhibit elevated levels of uric acid in the blood (hyperuricemia) due to either increase in production or decrease in excretion of uric acid. MSU crystals can be deposited in joints, bone, and soft tissue near or distant to a joint. Crystal deposition does not always lead to the clinical symptoms or signs of inflammation described above127. Classic gout passes through three stages: (1) asymptomatic hyperuricemia;

(2) acute intermittent gout; and (3) advanced (chronic tophaceous) gout.1 In stage 1, despite hyperuricemia, patients may never develop symptoms, such as gouty arthritis or nephrolithiasis. Asymptomatic hyperuricemia may last for decades before an initial gouty attack occurs. Stage 2 starts with the first attack of gouty arthritis, which in 85% to 90% of cases is monoarticular.128 Hyperuricemia, which is considered to have a strong association with gout, might not be present in up to 42% of patients with acute gouty attacks.129 Gouty arthritis targets the first metatarsophalangeal joint (also referred to as podagra127) in over 50% of patients128 and may affect the sesamoids.130 Eventually, 90% of patients with gout will have involvement of the first metatarsophalangeal joint; however, gout may present at any foot articulation.1 Joint involvement is asymmetrically distributed. Repeat attacks may become polyarticular. During the acute gouty attack soft tissue swelling surrounds the affected joint and joint effusion is noted. These symptoms resolve during intercritical periods, which occur between acute gouty attacks. Characteristic radiographic findings are not seen for several years after the initial onset of symptoms. Advanced gout develops after having acute intermittent gout for at least 10 years. The transition occurs when there are no more pain-free intercritical periods. Chronic tophaceous gout (stage 3) is characterized by the presence of MSU deposits in the soft tissue; these deposits (tophus) may be large enough to be visible on radiographs and may occur at any site. Most commonly tophi are found at the first metatarsophalangeal joint. When urate crystals are deposited within the lining of capsular or synovial tissues, it leads to arthropathy with characteristic radiographic findings.128 Articular tophaceous gout results in destructive arthropathy. The diagnosis of gout is typically made by clinical presentation and laboratory findings, as the imaging findings occur late in the disease process. Radiography is commonly used in the initial evaluation of gouty arthritis to rule out septic arthritis, if applicable, and to determine if tophi are present. Ultrasonography, CT, MRI, and nuclear medicine are rarely used for the diagnosis of gout.128 Characteristic radiographic features of gouty arthritis (summarized in Box

19-21 and illustrated in Figure 19-39) include the following:  1. Soft tissue mass (tophus). The tophus is a characteristic feature of advanced gout. Its radiographic feature is that of a distinct increased soft tissue volume or lump (often large), adjacent to a joint (Figure 19-20). BOX 19-21 Gouty Arthritis: Radiographic Features Primary finding: erosion, which tends to be periarticular (extra-articular), Cshaped, and with an overhanging margin (Martel sign) Normal joint space (unless erosion becomes central in joint) Soft tissue mass (tophus) (“lumpy-bumpy”), which occurs adjacent to erosion in the soft tissue but also may be intraosseous Rarefaction in bone occurs secondary to intraosseous tophus Occasional calcification of tophus Targets: first metatarsophalangeal and hallux interphalangeal joints Foot distribution: U/L, B/L (asymmetric or symmetric); any joint possible; monoarticular or polyarticular Elsewhere: knees, elbows  2. Erosion. Para-articular or extra-articular erosion occurs adjacent to a tophus, resulting from pressure atrophy of bone. Intra-articular erosion starts at the margin of the joint and extends toward the center.128 Unlike rheumatoid arthritis, the classical gouty erosion tends to occur outside the joint (extra-articular) (Figure 19-39A). Initially, erosion is ill-defined but becomes well-defined and C-shaped over an extended period of time (Figure 19-3). The latter has been described as “punched out erosion,” and many develop an overhanging margin of bone.12 Erosion tends to target the medial aspect of the first metatarsophalangeal and hallux interphalangeal joints. In significantly advanced cases, erosions may coalesce to present a honeycomb appearance (Figures 19-14 and 19-17).131

 3. Overhanging margin of bone. As the tophus causes pressure atrophy and erosion of adjacent bone, new bone production may form in an attempt to contain it.128 Erosion will demonstrate thin, elevated bone at its periphery with outward displacement as the tophus gradually expands. This finding, originally described by Martel,12 is often referred to as Martel sign (Figures 19-14 and 19-39A).  4. Normal joint space. In contrast to the inflammatory joint diseases (rheumatoid, seronegative, and septic arthritis), the joint space is relatively spared (Figures 19-17 and 19-39A,B) until late in the disease when either erosion encroaches on the joint center or secondary osteoarthritis occurs with subsequent joint space narrowing (Figure 19-14). No periarticular osteopenia is present but disuse osteoporosis may develop.  5. Rarefaction. Intraosseous tophi have also been described.130 Before calcification, they cause rarefaction within the bone (Figures 19-26 and 1939B). This has also been referred to as “subcortical erosion.”132  6. Calcification. Small, speckled, increased densities may be recognized in a soft tissue tophus (Figures 19-17 and 19-22). This is uncommon in early phases of gout but occurs in advanced gout.127 An intraosseous tophus may also calcify (Figure 19-39B); this occurs in about 6% of patients with advanced gout.32 Advanced imaging may be necessary to differentiate an intraosseous tophus from enchondroma or bone infarct.128 Calcification is more common with alteration of calcium metabolism or with coexisting renal disease.

FIGURE 19-39. Gouty arthritis. A: Extra-articular erosion first metatarsal medially with overhanging margin (arrow). B: Rarefaction (intraosseous tophus with calcification, arrow) within superomedial aspect of first metatrarsal head. C: Distal interphalangeal joint. There is “cupping” of the middle phalanx head secondary to subchondral bone collapse. Cross-sectional imaging (CT) may be used for further evaluation and surgical planning, especially when a tophus has an unusual presentation that may suggest infection or neoplasm.128 Intraosseous tophi are easily identified by CT, either being entirely within the bone, or having a cortical breach and erosion.131,133 Dalbeth et al.133 showed that the presence of erosion is associated with adjacent tophi in 82% to 100% of joints, depending on the erosion size. Dual-energy CT (DECT) can reliably identify, quantitatively, the MSU crystal deposits within joints, tendons, and periarticular soft tissues and may document successful treatment.134,135 DECT can be a useful diagnostic tool in the management of acute gout in clinically challenging cases with normal serum uric acid levels; it also has the potential to differentiate gout from pseudogout (calcium pyrophosphate deposition).134,136 DECT is useful in confirming the MSU crystal deposition at entheses (Achilles tendon, plantar fascia) with previous negative aspiration.127 Physical measurements are more feasible for superficial tophi than advanced imaging methods, but DECT can

be useful in detecting the intra-articular or deep tophi.137 Ultrasonography is helpful in detecting joint effusions and in directing joint aspirations and biopsies. The identification of MSU crystals in the joint aspirate is considered the gold standard in the diagnosis of gout but may not be identifiable in up to 25% of acute gout cases.138 The combination of joint effusion, tophus, erosion, and the double-contour sign on ultrasound are said to be diagnostic of gout in 97% of cases.127,139 Ultrasound is able to clearly identify tophi in bursae, tendons, ligaments, and soft tissue. In regards to tendons, de Avila Fernandes et al.140 has found that in 45% of cases the MSU deposits are enveloping the tendon, and only in a small proportion the deposits are intratendinous. The most affected tendons in the foot are the Achilles and peroneal tendons. MRI may be indicated over ultrasound for imaging deeper structures and other areas not visible by ultrasound, such as bone (intraosseous gouty deposits). However, the true diagnostic accuracy of MRI in gout is yet to be determined.141 MRI is unable to specifically identify the MSU crystal deposition.142 MRI may be helpful in assessing the extent and distribution of gouty arthritis; however, the MRI appearance of tophaceous gout is nonspecific and sometimes variable and nonspecific.128,142,143 Bone erosion adjacent to a tophus can produce cortical destruction and variable bone marrow edema, identifiable by MRI. Because cortical erosion may also be seen with osteomyelitis, the lack of an adjacent soft tissue ulcer is an important finding that suggests the diagnosis of gout.127 First Metatarsophalangeal Joint The earliest finding associated with acute gout is increased soft tissue density and volume medial to the first metatarsal head. There may be an associated rarefaction of the first metatarsal head’s medial aspect (Figure 19-26). Soft tissue calcification is infrequently encountered (Figure 19-22). Bloch and associates describe the appearance of a “lace pattern” of erosion along the dorsomedial aspect of the first metatarsal head in the medial oblique view as an early finding in gout.132 Erosion tends to develop

medially, which is not always clearly isolated in the dorsoplantar view; it may appear as a somewhat geographic rarefaction if viewed en face (Figure 19-26). The sesamoid axial view may also prove valuable in these cases. Erosion associated with gouty arthritis is classically described as being Cshaped (Figure 19-3), especially in chronic disease. An overhanging margin of bone (Martel sign) accentuates this presentation (Figure 19-14). However, developing erosions frequently are ill-defined (Figure 19-17). Notice in Figure 19-3 that, despite the significant presence of erosion, the joint space is relatively spared. Joint space narrowing is frequently a result of secondary osteoarthritis or long-standing intra-articular involvement. The sesamoids inferior to the first metatarsal head can also be affected. Lesser Metatarsophalangeal Joints Lesser metatarsophalangeal joint involvement is rare, but, when present, it often is selective for the fifth metatarsophalangeal joint. Findings may mimic rheumatoid arthritis or cause large geographic defects in the metatarsal head. Interphalangeal Joints Medial marginal erosion, similar to that seen at the first metatarsophalangeal joint, affects the hallux interphalangeal joint. Well-defined soft tissue density and volume presents adjacent to this finding. However, large geographic defects frequently involve the lesser toe interphalangeal joints (Figure 1939C) and occasionally the hallux. Tophaceous gout usually follows repeated episodes of acute gout. In rare instances, the presence of tophi may manifest as an expansile destructive bone lesion with soft tissue involvement suggestive of bony metastasis in the absence of a history of gout. In these cases, a needle biopsy with histopathologic features of tophus is helpful.144 Tarsometatarsal and Intertarsal Joints Tarsal bone involvement is atypical but demonstrates profound findings when it occurs. Large geographic defects are seen, frequently affecting multiple bones. The subtalar joint (STJ), midtarsal joint and cuneonavicular joint, as well as calcaneus, talar dome, cuboid and cuneiforms involvement has been

described. The joint space may be preserved but a tophus may appear as prominently large, well-defined, subchondral radiolucency with possible internal calcification. Charcot Neuropathic Osteoarthropathy Charcot neuropathic osteoarthropathy, or Charcot foot, is characterized by early inflammation and affects the bones, joints, and soft tissues of the foot and ankle.145 It may be active or inactive and is often indolent.146 Charcot neuropathic osteoarthropathy is most commonly associated with diabetes mellitus, although it can also be seen in patients with a history of alcoholism or leprosy. It is now considered an inflammatory syndrome, characterized by varying degrees of bone and joint disorganization, secondary to underlying neuropathy, trauma, and perturbations of bone metabolism.145 Charcot neuropathic osteoarthropathy is associated with varying degrees and patterns of bone destruction, a high frequency of fracture nonunion, joint dislocation, joint subluxation, foot deformity (“rocker-bottom foot”), skin ulceration, and risk of amputation.145,147 The location, magnitude of deformity, and timecourse are patient specific, and may take up to 24 months for the foot to become stable147,148. Cofield and associates149 listed three target areas in the foot for Charcot neuropathic osteoarthropathy: the metatarsophalangeal joints, the tarsometatarsal joints, and the combined talonavicular, naviculocuneiform, and intercuneiform joints. However, any foot joint can be affected. Bone resorption predominates in the forefoot; this is in gross contrast to the hypertrophic arthritis seen in the neuropathic tarsus and midfoot. Schon et al.150 developed a more specific classification system for midtarsus deformity; it is based on four types of anatomic location and three stages of severity or degree of collapse, This system incorporates clinical and radiographic parameters (Table 19-9). The most commonly used Charcot classification is based on the three stage Eichenholtz classification that consists of the following phases: (1) developmental or acute; (2) coalescent or quiescent; (3) consolidation or reconstruction.151,152 Johnson modified the classification terminology, which

is “temporally based,” to one based on the natural history of the disease: (1) dissolution; (2) coalescence; (3) resolution.153 Shibata et al.154 added stage 0 to account for the early disease when there are minimal to no radiographic findings. The modified Eichenholtz classification is summarized in Table 1910. Although these classification systems describe the course of Charcot foot in linear stages, the time of progression through each stage may vary from weeks to years.155 The five patterns of foot and ankle involvement described by Sanders and Frykberg are summarized in Table 19-11.156,157 The fivestage anatomic-based classification system provided by Sella and Barrette is found in Table 19-12.157,158 These classifications are useful in staging or describing the location of the joint involvement.159 Herbst has suggested that the pattern of Charcot neuropathic osteoarthropathy be separated into one of the following three groups: (1) fracture; (2) dislocation; (3) combined fracture-dislocation.160 However, Jones encountered three cases of Charcot neuropathic osteoarthropathy with rapid and aggressive resorption without the presence of fracture or subluxation/dislocation.161 Roger’s Charcot classification describes the location and the severity of the condition on a two-axis system and helps in the prognosis determination, including risk of amputation159,162 (Table 19-13). TABLE 19-9   TYPE I II III IV STAGE A B C

Radiographic Findings in Charcot Neuropathic Osteoarthropathy (Schon Classification)

Lisfranc pattern Naviculocuneiform pattern Perinavicular pattern Transverse tarsal pattern Minimal deformity; loss of arch height; no “rocker” or negative arch Greater derangement than stage A; loss of medial or lateral arch to plantar level; obvious plantar prominence Severe destruction of the midtarsus; collapse of arches, both medially and laterally; midfoot prominence more plantar than

heel or ball of foot From Schon LC, Weinfeld SB, Horton GA, et al. Radiographic and clinical classification of acquired midtarsus deformities. Foot Ankle Int. 1998;19:394. Radiographic Findings in Charcot Neuropathic TABLE 19-10   Osteoarthropathy (Eichenholtz152/Johnson153/Shibata154 Classifications) Stage Name Description Minimal to no radiographic findings (increased soft tissue density and volume); 0 Early (inflammatory) clinically: local warmth, swelling, and instability due to ligamentous laxity Regional osteopenia

1

Development/dissolution (acute)

Periarticular/subchondral bone fragmentation Subluxation and dislocation Bone debris at the articular margins Absorption of osseous debris in soft tissues Organization and early healing of fracture fragments

2

Coalescence (healing)

Periosteal new bone formation Fusion of large fragments to adjacent bones Sclerosis of bone ends Smoothing of edges of large

bony fragments 3

Reconstruction/resolution

Decrease in degree of sclerosis Osseous or fibrous ankylosis

Attempted reformation of bone architecture Radiographic Findings Pertinent to Sanders and TABLE 19-11   Frykberg156 Classification System of Charcot Neuropathic Osteoarthropathy Pattern Joints Involved Radiographic Findings Osteopenia Osteolysis Resorption of metatarsal and phalangeal shafts I

Forefoot

Juxta-articular cortical bone defects Subluxation of metatarsophalangeal joints

II

III

Tarsometatarsal (metatarsal bases; cuneiforms; cuboid)

Chopart (NCJs)

Atrophic bone destruction Subluxation or fracture/dislocation of metatarsal bases Rocker-bottom foot deformity Osteolysis with fragmentation and dorsal/plantar osseous debris Erosion of bone/cartilage Extensive joint destruction

IV

Ankle with/without STJ

Possible complete joint collapse and dislocation

V

Calcaneus

Severe unstable deformity Avulsion of Achilles tendon off the posterior tubercle; no joint involvement

NCJs, naviculocuneiform joints; STJ, subtalar joint. Sella and Barrette158 Classification System of Medial TABLE 19-12   Column Neuropathic Joint Disease with Radiographic Findings Phase Radiographic Findings 0 Normal Early bone involvement • Localized osteopenia/osteoporosis 1

• Subchondral cysts (geodes) • Erosions • Possible diastasis

2

Joint subluxation • Joint dislocation

3 • Joint collapse Signs of healing 4 • Sclerosis and fusion of affected joints/bones Unilateral involvement in Charcot neuropathic osteoarthropathy is most common, but bilateral involvement has been noted in about 30% of patients.159,163 The condition recurs rarely in the same extremity unless weight bearing is discontinued too early. Even though Charcot foot may occur in the contralateral foot in 20% to 25% of cases, late recurrence in the same foot is very rare once the condition goes into remission after several

months.164 The diagnosis of Charcot neuropathic osteoarthropathy is challenging, and imaging plays a pivotal role. Early and accurate diagnosis with immediate intervention is important to prevent progressive and destructive Charcot foot deformity, reducing morbidity and mortality.165 Radiographs are the primary initial imaging method for evaluation of diabetic foot in virtually all settings. They are readily available, inexpensive, and provide information on bone form, alignment, density, and architecture.165 However, radiographic changes are typically delayed and have low sensitivity to early abnormality. And a negative radiograph should not offer any confidence for the lack of disease.145,151 Clinical diagnosis is difficult in the acute phase (stage 0) of Charcot neuropathic osteoarthropathy and radiographs may fail to document any evidence of pathology. The same goes for CT. TABLE 19-13   Rogers’ Charcot 2-Axis Classification

Technetium-99m bone scintigraphy may show abnormalities months before clinical or radiographic abnormality becomes apparent, with uptake noted in all three phases.166,167 It has good sensitivity, but poor specificity for acute osseous pathology including osteoarthropathy. Also, diminished circulation can result in false-negative exams.168 MRI detects subtle change in the early stage of disease, revealing bone edema, occult fracture, and joint effusion in stage 0, when radiographs are normal.168,169 There also is significant correlation between the intensity of bone marrow edema and clinical pain and soft tissue edema.145,151 Until recently, the follow-up of diabetic patients with acute Charcot foot under treatment has been based solely on clinical signs and serial radiographs. The clinical signs (increased skin temperature, pain, swelling, and erythema) are useful in indicating the outcomes of the disease, but may lack sensitivity and specificity.170–172 The radiologic signs (fracture healing,

bone sclerosis, decreased soft tissue swelling, reduction of calcaneal inclination angle) are usually late signs of progression of the disease from acute to chronic phase, and may lack accuracy.145,173 A noninvasive, more accurate method to objectively assess the outcome of the disease and evaluate treatment efficacy is dynamic magnetic resonance imaging (D-MRI) with gadolinium.162 Multidetector computed tomography (MDCT) is indicated for evaluation after immobilization therapy for Charcot neuropathic osteoarthropathy.165 The radiographic features of Charcot neuropathic osteoarthropathy (summarized in Box 19-22 and illustrated in Figures 19-40 and 19-41) include the following:  1. Subchondral resorption. This is the primary radiographic finding (Figure 19-6B). Bone resorption without subluxation and/or dislocation of adjacent joints, and without bone fracture has been described in the midfoot or rearfoot.161 BOX 19-22 Charcot Neuropathic Osteoarthropathy: Radiographic Features FOREFOOT Primary finding: subchondral resorption or arthritis mutilans (osteolysis) Periostitis (occasionally) Targets: MPJs and IPJs Distribution: unilateral; monoarticular or polyarticular MIDFOOT/REARFOOT/ANKLE Primary finding: subchondral resorption Loss of joint apposition (subluxation/dislocation) Detritus

Diffuse sclerosis Targets: metatarsal-tarsal, naviculocuneiform joints; can also involve talocalcaneal, intercuneiform, calcaneocuboid, and ankle joints Distribution: unilateral

FIGURE 19-40. Forefoot neuropathic osteoarthropathy: acute, early involvement.  2. Osteopenia. One of the earliest osseous findings in Charcot neuropathic osteoarthropathy may be osteopenia, which may be the only radiographic finding if Charcot foot is diagnosed before the onset of bony destruction.174

Osteopenia encountered in the later stages of Charcot neuropathic osteoarthropathy may be related to immobilization, which is the mainstay treatment of this condition, which may be slowly and partially reversible.164,175  3. Subluxation and dislocation. Loss of apposition eventually occurs at affected joints and can progress to gross dislocation (Figure 19-29). Subluxation of the first and second tarsometatarsal joints can be evaluated on a dorsoplantar film of the foot, while the third through the fifth tarsometatarsal joints are best assessed with a medial oblique foot position.165  4. Detritus. Bone fragments (also called bony debris) are seen, typically associated with subluxation and dislocation (Figure 19-24). The fragments may be quite large. When joint effusion is present, the effusions may decompress along the fascial planes, carrying bony debris far from the joint.165  5. Osteolysis. Focal osteolysis will be seen at any joint where subchondral resorption leads to detritus formation (Figure 19-40).  6. Periostitis. Periosteal new bone production accompanies acute manifestations (Figure 19-40), sometimes with cloudy callus formation around old fracture sites.146  7. Arthritis mutilans. Osteolysis is typically more pronounced than that seen in psoriatic arthritis. A large portion of bone, affecting phalanges and metatarsals alike, seemingly disappears, individually or en masse (Figure 198).  8. Diffuse sclerosis. Increased density is seen at all affected joints and bones (Figures 19-8 and 19-29).

FIGURE 19-41. Tarsal neuropathic osteoarthropathy. A: Early manifestations include subchondral resorption (second and third metatarsocuneiform joints), mild sclerosis (intermediate cuneiform and second metatarsal base), and subluxation (first and second metatarsocuneiform joints). B: In a different patient, additional findings include diffuse sclerosis and fragmentation of the navicular and cuneiform bones. Forefoot The radiographic presention of forefoot Charcot neuropathic osteoarthropathy may vary consiberably. It targets all metatarsophalangeal and interphalangeal joints; the distribution is unilateral, but can be monoarticular or polyarticular. The primary radiographic feature is subchondral resorption. This initially apears in the form of periarticular rarefaction, while the subchondral bone plate is still visible. This quickly progresses to fragmentation and osteolysis. Periostitis may or may not accompany acute manifestations (Figure 19-40). A less severe presentation, especially at a lesser metatarsophalangeal joint, may mimic osteonecrosis. Following the primary finding of periarticular rarefaction, the metatarsal head may collapse and be accompanied by underlying sclerosis. Arthritis mutilans is the most severe presentation. Osteolysis is typically more pronounced than that seen in psoriatic arthritis. A large portion of bone, affecting phalanges and metatarsals alike, seemingly disappears, individually

or en masse (Figure 19-8). Tarsus The metatarsal-tarsal and naviculocuneiform joints are commonly targeted; however, all tarsal joints can also be affected, including the talocalcaneal, intercuneiform, calcaneocuboid, and ankle joints. The distribution is generally unilateral (Figure 19-41). As in the forefoot, subchondral resorption is the primary radiographic finding, especially in the form of loss of the subchondral bone plate. This is accompanied by loss of apposition (subluxation) at affected joints and progresses to gross dislocation, which is accompanied by detritus (fragmentation), sometimes quite large (Figure 19-24). Diffuse sclerosis will involve all affected bones as the body attempts to repair and later remodel (Figure 19-29). Diabetic patients may present with posttraumatic osteoarthritis at the metatarsal-tarsal joints that could be misdiagnosed as Charcot foot. The lack of subchondral resorption at articular margins with the presence of joint space narrowing, subchondral sclerosis, and loose bodies indicates osteoarthritis. Septic Arthritis Infection at a joint is known as septic arthritis. (Please refer to Chapter 18 for more information.) The primary radiographic finding is subchondral resorption (Figure 19-42). Depending on the articulation affected, joint space widening may be an early finding (Figure 19-19A). If left untreated, this will progress to osteolysis. Though not specific for infection, increased soft tissue density and volume will accompany these findings. Septic arthritis can target any joint, and its distribution is monoarticular (Box 19-23). BOX 19-23 Septic Arthritis: Radiographic Features Subchondral resorption (primary finding) Osteolysis

Increased soft tissue density and volume Targets: any joint Distribution: monoarticular

FIGURE 19-42. Septic arthritis. Subchondral resorption along distal aspect of hallux interphalangeal joint with resultant osteolysis and sequestra laterally (medial ossicle is an eccentric os interphalangeus). REFERENCES   1. Klippel JH. Primer on the Rheumatic Diseases. 13th ed. New York, NY: Springer; 2008.   2. Rubin DA. The radiology of early arthritis. Semin Roentgenol. 1996;31(3):185.   3. Christman RA. A systematic approach for radiographically evaluating joint disease in the foot. J Am Podiatr Med Assoc. 1991;81(4):174.   4. Resnick D. The target area approach to articular disorders: a synopsis. In: Resnick D, Niwayama G, eds. Diagnosis of Bone and Joint Disorders. 2nd ed. Philadelphia, PA: WB Saunders; 1988:1913.   5. Forrester DM, Brown JC. The radiographic assessment of arthritis: the plain film. Clin Rheum Dis. 1983;9(2):291.   6. Christman RA. Joint disease. In Christman RA, ed. Foot and Ankle Radiology. St Louis, MO: Churchill Livingstone; 2003.   7. Goldthwaite JE. The differential diagnosis and treatment of the socalled rheumatoid disease. Boston Med Surg J. 1904;151:529.   8. Chew FS. Musculoskeletal Imaging. Philadelphia, PA: Lippincott Williams & Wilkins; 2003.   9. Resnick D. Talar ridges, osteophytes, and beaks: a radiologic commentary. Radiology. 1984;151:329.  10. Edeiken J, Dalinka M, Karasick D. Edeiken’s Roentgen Diagnosis of Diseases of Bone. 4th ed. Baltimore, MD: Lippincott Williams & Wilkins; 1990:693.

 11. Resnick D, Broderick TW. Bony proliferation of terminal toe phalanges in psoriasis: the “ivory” phalanx. Can Assoc Radiol J. 1977;28:187.  12. Martel W. The overhanging margin of bone: a roentgenologic manifestation of gout. Radiology. 1968;91:755.  13. Resnick D. The radiographic manifestations of gouty arthritis. CRC Crit Rev Diagn Imaging. 1977;9:265.  14. Kaye JJ. Arthritis: roles of radiography and other imaging techniques in evaluation. Radiology. 1990;177(3):601.  15. Barthelemy CR, Nakayama DA, Carrera GF, et al. Gouty arthritis: a prospective radiographic evaluation of sixty patients. Skeletal Radiol. 1984;11:1.  16. Keil H. Rheumatic subcutaneous nodules and simulating lesions. Medicine. 1938;17:261.  17. Edeiken J, Dalinka M, Karasick D. Edeiken’s Roentgen Diagnosis of Diseases of Bone. 4th ed. Baltimore, MD: Lippincott Williams & Wilkins; 1990:1369.  18. Greenfield GB. Radiology of Bone Diseases. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 1986:688.  19. Jaffe HL. Metabolic, Degenerative, and Inflammatory Diseases of Bones and Joints. Philadelphia, PA: Lea & Febiger; 1972.  20. McCarty DJ. Diagnostic mimicry in arthritis: patterns of joint involvement associated with calcium pyrophosphate dihydrate crystal deposits. Bull Rheum Dis. 1974;25:804.  21. Bencardino JT, Hassankhani A. Calcium pyrophosphate dihydrate crystal deposition disease. Semin Musculoskelet Radiol. 2003;7:175.  22. Resnick D, Niwayama G, Goergen TG, et al. Clinical, radiographic, and pathologic abnormalities in calcium pyrophosphate dihydrate deposition disease (CPPD): pseudogout. Radiology. 1977;122:1.

 23. Gruneberg R. Calcifying tendinitis in the forefoot. Br J Radiol. 1963;36:378.  24. Bonavita JA, Dalinka MK, Schumacher HR. Hydroxyapatite deposition disease. Radiology. 1980;134:621.  25. Yochum TR, Rowe LJ. Essentials of Skeletal Radiology. 3rd ed. Baltimore, MD: Lippincott Williams & Wilkins, 2005.  26. Resnick D, Niwayama G, Coutts RD. Subchondral cysts (geodes) in arthritic disorders: pathologic and radiographic appearance of the hip joint. AJR Am J Roentgenol. 1977;128:799.  27. Bullough PG, Bansal M. The differential diagnosis of geodes. Radiol Clin North Am. 1988;26(6):1165.  28. Jacobs AJ, Ginsburg WW, Bryan RS. Large geode in elbow with spontaneous fracture. J Rheumatol. 1982;9:340.  29. Ondrouch AS. Cyst formation in osteoarthritis. J Bone Joint Surg. 1963;45(B):755.  30. Resnick D, Niwayama G. Diagnosis of Bone and Joint Disorders. 2nd ed. Philadelphia, PA: WB Saunders; 1988:1696.  31. Magyar E, Talerman A, Feher M, et al. The pathogenesis of the subchondral pseudocysts in rheumatoid arthritis. Clin Orthop. 1974;100:341.  32. Resnick D, Broderick TW. Intraosseous calcifications in tophaceous gout. AJR Am J Roentgenol. 1981;137:1157.  33. Sommer OJ, Kladosek A, Weiler V, et al. Rheumatoid arthritis: a practical guide to state-of-the-art imaging, image interpretation, and clinical implications. Radiographics. 2005;25:381.  34. Resnick D, Feingold ML, Curd J, et al. Calcaneal abnormalities in articular disorders: rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis and Reiter syndrome. Radiology. 1977;125:355.

 35. Resnick D, Niwayama G. Entheses and enthesopathy: anatomical, pathological, and radiological correlation. Radiology. 1983;146:1.  36. Ash Z, Marzo-Ortega H. Ankylosing Spondilitis—the changing role of imaging. Skeletal Radiol. 2012;41:1031.  37. Pack LG. Get out of your OA Box. Rheumatologist. 2010;111:32.  38. Bekkers JEJ, Creemers LB, Dhert WJA, et al. Diagnostic modalities for diseased articular cartilage from defect to degeneration: a review. Cartilage. 2010;1(3):157.  39. Resnick D. Bone and Joint Imaging. 2nd ed. Philadelphia, PA: WB Saunders; 1996.  40. Mitchell NS, Cruess RL. Classification of degenerative arthritis. Can Med Assoc J. 1977;117:763.  41. Brandt KD, Radin EL, Dieppe PA, et al. Yet more evidence that osteoarthritis is not a cartilage disease. Ann Rheum Dis. 2006;65:1261.  42. Watt I. Osteoarthritis revisited—again! Skeletal Radiol. 2009;38:419.  43. Alexander CJ. Idiopathic osteoarthritis: time to change the paradigms? Skeletal Radiol. 2004;33:321.  44. Langenskiold A, Michelsson JE, Videman T. Osteoarthritis of the knee in the rabbit produced by immobilization: attempts to achieve a reproducible model for studies on pathogenesis and therapy. Acta Orthop Scand. 1979;50:1.  45. Sharma S, Rogers J, Watt I, et al. Bone mineral density and fractal signature analysis in hip osteoarthritis—a study of a post-mortem and postoperative population. Clin Radiol. 1997;52:872. Cited by: Watt I. Osteoarthritis revisited—again! Skeletal Radiol. 2009;38:419.  46. Raudenbush D, Sumner DR, Panchal PM, et al. Subchondral thickness does not vary with cartilage degeneration on the metatarsal. J Am Podiatr Med Assoc. 2003;93(2):104.

 47. Wada M, Baba H, Imura S, et al. Relationship between radiographic classification and arthroscopic findings of articular cartilage lesions in osteoarthritis of the knee. Clin Exp Rheumatol. 1998;16(1):15.  48. Kijowski R, Blankenbaker DG, Stanton PT, et al. Radiographic findings of osteoarthritis versus arthroscopic findings of articular cartilage degeneration in the tibiofemoral joint. Radiology. 2006;239:818.  49. KIjowski R, Blankenbaker D, Stanton P, et al. Arthroscopic validation of radiografic grading scales of osteoarthritis of the tibiofemoral joint. AJR Am J Roentgenol. 2006;187(3):794.  50. Gold GE, Chen CA, Koo S, et al. Recent advances in MRI of articular cartilage. AJR Am J Roentgenol. 2009;193(3):628.  51. Eckstein F, Burstein D, Link TM. Quantitative MRI of cartilage and bone: degenerative changes in osteoarthritis. NMR Biomed. 2006;19(7):822.  52. Saltzman CL, Salamon ML, Blanchard GM, et al. Epidemiology of ankle arthritis. Iowa Orthop J. 2005;25:44.  53. Valderrabano V, Horisberger M, Russell I, et al. Etiology of ankle osteoarthritis. Clin Orthop Relat Res. 2009;467:1800.  54. Felson DT. The epidemiology of osteoarthritis: prevalence and risk factors. In: Kuettner KE, Goldberg VM, eds. Osteoarthritic Disorders. Rosemont, IL: American Academy of Orthopaedic Surgeons; 1995:13–24.  55. Peyron JG. The epidemiology of osteoarthritis. In: Moskowitz RW, Howell DS, Goldberg VM, et al., eds. Osteoarthritis: Diagnosis and Treatment. Philadelphia, PA: WB Saunders; 1984:9–27.  56. Aletaha D, Neogi T, Silman AJ, et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum. 2010;62:2569.  57. Radner H, Neogi T, Smolen JS, et al. Performance of the 2010

ACR/EULAR classification criteria for rheumatoid arthritis: a systematic literature review. Ann Rheum Dis. 2014;73:114.  58. Larsen A, Dale K, Eek M. Radiographic evaluation of rheumatoid arthritis and related conditions by standard reference films. Acta Radiol Diagn (Stockh). 1977;18:481.  59. Scott D, Houssien D, Laasonen L. Proposed modification to Larsen’s scoring method for hand and wrist radiographs. Br J Rheumatol. 1995;34:56.  60. Edmonds J, Saudan A, Lassere M, et al. Introduction to reading radiographs by the Scott modification of the Larsen method. J Rheumatol. 1999;26:740.  61. McQueen FM, Ostergaard M. Established rheumatoid arthritis—new imaging modalities. Best Pract Res Clin Rheumatol. 2007;21:841.  62. Ostendorf B, Mattes-Gyorgy K, Reichelt DC, et al. Early detection of bony alterations in rheumatoid and erosive arthritis of finger joints with highresolution single photon emission computed tomography, and differentiation between them. Skeletal Radiol. 2010;39:55.  63. Ostendorf B, Scherer A, Wirrwar A, et al. High-resolution multipinhole single-photon-emission computed tomography in experimental and human arthritis. Arthritis Rheum. 2006;54:1096.  64. Jimenez-Boj E, Redlich K, Turk B, et al. Interaction between synovial inflammatory tissue and bone marrow in rheumatoid arthritis. J Immunol. 2005;175:2579.  65. Jimenez-Boj E, Nobauer-Huhmann I, Hanslik-Schnabel B, et al. Bone erosions and bone marrow edema as defined by magnetic resonance imaging reflect true bone marrow inflammation in rheumatoid arthritis. Arthritis Rheum. 2007;56:1118.  66. McQueen FM, Ostendorf B. What is MRI bone edema in rheumatoid arthritis and why does it matter? Arthritis Res Ther. 2006;8:222.

 67. Resnick D. Rheumatoid arthritis and the seronegative spondyloarthropathies: radiographic and pathologic concepts. In: Resnick D, ed. Bone and Joint Imaging. 2nd ed. Philadelphia, PA: WB Saunders; 1989;195–209.  68. Resnick D. Rheumatoid arthritis and the seronegative spondyloarthropathies: radiographic and pathologic concepts. In: Resnick D, ed. Bone and Joint Imaging. 2nd ed. Philadelphia, PA: WB Saunders; 1989:210–234.  69. Emery P. Magnetic resonance imaging: opportunities for rheumatoid arthritis disease assessment and monitoring long-term treatment outcomes. Arthritis Res. 2002;4(suppl 2):S6–S10.  70. Huang J, Stewart N, Crabbe J, et al. A 1-year follow-up study of dynamic magnetic resonance imaging in early rheumatoid arthritis reveals synovitis to be increased in shared epitope-positive patients and predictive of erosions at 1 year. Rheumatology (Oxford). 2000;39:407.  71. McQueen FM. Magnetic resonance imaging in early inflammatory arthritis: what is its role? Rheumatology. 2000;39:700.  72. Peterfy CG. New developments in imaging in rheumatoid arthritis. Curr Opin Rheumatol. 2003;15:288.  73. Fex E, Jonsson K, Johnson U, et al. Development of radiographic damage during the first 5–6 years of rheumatoid arthritis: a prospective follow-up study of a Swedish cohort. Br J Rheumatol. 1996;35:1106.  74. Guermazi A, Taouli B, Lynch JA, et al. Imaging of bone erosion in rheumatoid arthritis. Semin Musculoskelet Radiol. 2004;8(4):269.  75. Buckland-Wright JC, Clarke GS, Walker SR. Erosion number and area progression in the wrists and hands of rheumatoid patients: a quantitative microfocal radiographic study. Ann Rheum Dis. 1989;48:25.  76. Backhaus M, Burmester GR, Sandrock D, et al. Prospective two year follow up study comparing novel and conventional imaging procedures in

patients with arthritic finger joints. Ann Rheum Dis. 2002;61:895.  77. McGonagle D, Conaghan PG, O’Connor P, et al. The relationship between synovitis and bone changes in early untreated rheumatoid arthritis: a controlled magnetic resonance imaging study. Arthritis Rheum. 1999;42:1706.  78. McQueen F, Stewart N, Crabbe J, et al. Magnetic resonance imaging of the wrist in early rheumatoid arthritis reveals a high prevalence of erosions at 4 months after symptom onset. Ann Rheum Dis. 1998;57:350.  79. McQueen F, Stewart N, Crabbe J, et al. Magnetic resonance imaging of the wrist in early rheumatoid arthritis reveals progression of erosions despite clinical involvement. Ann Rheum Dis. 1999;58:156.  80. Mottonen TT, Hannonen P, Toivanen J, et al. Value of joint scintigraphy in the prediction of erosiveness in early rheumatoid arthritis. Ann Rheum Dis. 1988;47:183.  81. Lehtinen JT, Lehto MU, Kaarela K, et al. Radiographic joint space in rheumatoid glenohumeral joints: a 15-year prospective follow-up study in 74 patients. Rheumatology (Oxford). 2000;39:288.  82. Kelikian AS. Sarrafian’s Anatomy of the Foot and Ankle. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2011.  83. Bouysset M, Tebib J, Tavernier T, et al. Posterior tibial tendon and subtalar joint complex in rheumatoid arthritis: magnetic resonance imaging study. J Rheumatol. 2003;30:1951.  84. Cantini F, Niccoli L, Nannini C, et al. Psoriatic arthritis: a systematic review. Int J Rheum Dis. 2010;13:300.  85. Bartolomei FJ. Pedal radiographic manifestations of the seronegative spondyloarthritides. II: Psoriatic arthritis. J Am Podiatr Med Assoc. 1986;76(5):266.  86. Kumar R, Madewell JE. Rheumatoid and seronegative arthropathies of

the foot. Radiol Clin North Am. 1987;25(6):1263.  87. Kettering JM, Towers JD, Rubin DA. The seronegative spondyloarthropathies. Semin Roentgenol. 1996;31(3):220.  88. Klecker JR, Weissman BN. Imaging features of psoriatic arthritis and Reiter’s Syndrome. Semin Musculoskeletal Radiol. 2003;7(2):115.  89. Gladman DD. Clinical, radiological, and functional assessment in psoriatic arthritis: is it different from other inflammatory diseases? Ann Rheum Dis. 2006;65:iii22–iii24.  90. Lowell DL, Osher LS, Grady AF. Incidental findings of massive heel spurs in a veteran with a variant of psoriatic arthritis. J Am Podiatr Med Assoc. 2012;102(5):422.  91. Dhir V, Aggarwal A. Psoriatic arthritis: a critical review. Clin Rev Allergy Immunol. 2013;44:141.  92. Helliwell PS, Taylor WJ. Classification and diagnostic criteria for psoriatic arthritis. Ann Rheum Dis. 2005;64(suppl 2):ii3.  93. Moll JMH, Wright V. Psoriatic arthritis. Semin Arthritis Rheum. 1973;3:55.  94. Moll JM. The clinical spectrum of psoriatic arthritis. Clin Orthop Relat Res. 1979;143:66.  95. Bongartz T, Harle P, Friederich S, et al. Successful treatment of psoriatic onycho-pachydermo periostitis (POPP) with adalimumab. Arthritis Rheum. 2005;52:280.  96. Taylor WJ, Gladman DD, Helliwell PS, et al. Classification criteria for psoriatic arthritis. Arthritis Rheum. 2006;54:2665.  97. Weger W. An update on the diagnosis and management of psoriatic arthritis. G Ital Dermatol Venereol. 2011;146:1.  98. Killebrew K, Gold RH, Sholkoff SD. Psoriatic spondylitis. Radiology.

1973;108:9.  99. Lambert JR, Wright V. Psoriatic spondylitis: a clinical and radiological description of the spine in psoriatic arthritis. Q J Med. 1977;46:411. 100. Gladman DD, Brubacher B, Buskila D, et al. Differences in the expression of spondyloarthropathy: a comparison between ankylosing spondylitis and psoriatic arthritis. Genetic and gender effects. Clin Invest Med. 1993;16:1. 101. Helliwell PS, Hickling P, Wright V. Do the radiological changes of classic ankylosing spondylitis differ from the changes found in the spondylitis associated with inflammatory bowel disease, psoriasis, and reactive arthritis? Ann Rheum Dis. 1998;57:135. 102. Ory PA, Gladman DD, Mease PJ. Psoriatic arthritis and imaging. Ann Rheum Dis. 2005;64 (suppl II):ii55. 103. Weishaupt D, Schweitzer ME, Alam F, et al. MR imaging of inflammatory joint diseases of the foot and ankle. Skeletal Radiol. 1999;28:663. 104. McQueen FM, Benton N, Perry D, et al. Bone oedema scored on magnetic resonance scans of the dominant carpus at presentation predicts radiographic joint damage at the hands and feet six years later in patients with rheumatoid arthritis. Arthritis Rheum. 2003;48:1814. 105. Giovagnoni A, Grassi W, Terili F, et al. MRI of the hand in psoriatic and rheumatoid arthritis. Eur Radiol. 1995;5:590. 106. McQueen F, Lassere M, Ostergaard M. Magnetic resonance imaging in psoriatic arthritis: a review of the literature. Arthritis Res Ther. 2006;8(2):207. 107. Jevtic V, Watt I, Rozman B, et al. Distinctive radiological features of small hand joints in rheumatoid arthritis and seronegative spondyloarthritis demonstrated by contrast-enhanced (Gd-DTPA) magnetic resonance imaging. Skeletal Radiol. 1995;24:351.

108. Vasey FB, Seleznck MJ, Fenske NA, et al. New sign-posts on the road to understanding psoriatic arthritis. J Rheumatol. 1989;6:1405. 109. De Simone C, Gueriero C, Giampetruzzi AR, et al. Achilles tendinitis in psoriasis: clinical and sonographic findings. J Am Acad Dermatol. 2003;49:217. 110. Forrester DM, Kirkpatrick J. Periostitis and pseudoperiostitis. Radiology. 1976;118:597. 111. Resnick D, Niwayama G. Rheumatoid arthritis and the seronegative apondyloarthropathies: radiographic and pathologic concepts. In: Resnick D, Niwayama G, eds. Diagnosis of Bone and Joint Disorders. 2nd ed. Philadelphia, PA: WB Saunders; 1988:895–953. 112. Martel W, Stuck KJ, Dworin AM, et al. Erosive osteoarthritis and psoriatic arthritis: a radiologic comparison in the hand, wrist, and foot. AJR Am J Roentgenol. 1980;134:125. 113. McGonagle D, Gibbon W, O’Connor P, et al. Characteristic magnetic resonance imaging entheseal changes of knee synovitis in spondyloarthropathy. Arthritis Rheum. 1998;41:694. 114. Weiner SM, Jurenz S, Uhl M, et al. Ultrasonography in the assessment of peripheral joint involvement in psoriatic arthritis: a comparison with radiography, MRI and scintigraphy. Clin Rheumatol. 2008;27:983. 115. Resnick D. The interphalangeal joint of the great toe in rheumatoid arthritis. Can Assoc Radiol J. 1975;26:255. 116. Dans M, Hivnor C, Van Voorhees AS. Psoriatic onychopacydermoperiostitis: improvement with etanercept. Br J Dermatol. 2005;153:858. 117. Zeimer A, Heider M, Goring HD. Psoriasiform onychopachydermoperiostitis of the large toes: the OP3GO syndrome. Hautarzt. 1998;49:859. Cited by: Mahoney JM, Scott R. Psoriatic onychopachydermoperiostitis (POPP): a perplexing case study. J Am Podiatr

Med Assoc. 2009;99(2):140. 118. Mahoney JM, Scott R. Psoriatic onychopachydermoperiostitis (POPP): a perplexing case study. J Am Podiatr Med Assoc. 2009;99(2):140. 119. Helliwell P, Marchesoni A, Peters M, et al. A reevaluation of the osteoarticular manifestations of psoriasis. Br J Rheumatol. 1991;30:339. 120. McGonacle D, Conaghan PG, Emery P. Psoriatic arthritis: a unified concept twenty years on. Arthritis Rheum. 1999;42:1080. 121. Sieper J, Rudwaleit M, Khan MA, et al. Concepts and epidemiology of spondyloarthritis. Best Pract Res Clin Rheumatol. 2006;20:401. 122. Mansour M, Cheema GS, Naguwa SM, et al. Ankylosing spondylitis: a contemporary perspective on diagnosis and treatment. Semin Arthritis Rheum. 2007;36:210. 123. Resnick D. Patterns of peripheral joint disease in ankylosing spondylitis. Radiology. 1974;110:523. 124. McHugh NJ. Other seronegative spondyloarthropathies. Medicine. 2010;38(4):190. 125. Levesque RY, Azouz EM. Radiologic findings in adults with seronegative spondyloarthropathies: distinguishing the chronic arthropathies in adulthood. Can Fam Physician. 1992;38:2105. 126. Wunschel M, Wulker N, Walter C. Surgical treatment of a young patient with bilaterally destroyed first metatarsophalangeal joints suffering from gout. J Am Podiatr Med Assoc. 2012;102(4):334. 127. Girish G, Glazenbrook KN, Jacobson JA. Advanced imaging of gout. Musculoskeletal Imaging Rev. 2013;201:515. 128. Gentilli A. Advanced imaging of gout. Sem Musculoskeletal Radiol. 2003;7(3):165. 129. Schlesinger N, Baker DG, Schumacher HR Jr. Serum urate during bouts

of acute gouty arthritis. J Rheumatol. 1997;24:2265. 130. Fauci AS, Braunwald E, Isselbacher KJ, et al. Harrison’s Principles of Internal Medicine. 15th ed. USA: International Edition; 2001:2268–2271. 131. McQueen FM, Reeves Q, Dalbeth N. New insights into an old disease: advanced imaging in the diagnosis and management of gout. Postgrad Med J. 2013;89:87. 132. Bloch C, Hermann G, Yu TF. A radiologic reevaluation of gout: a study of 2,000 patients. AJR Am J Roentgenol. 1980;134:781. 133. Dalbeth N, Clark B, Gregory K, et al. Mechanisms of bone erosion in gout: a quantitative analysis using plain radiography and computed tomography. Ann Rheum Dis. 2009;68:1290. 134. Choi HK, Al-Arfaj AM, Eftekhari A, et al. Dual energy computed tomography in tophaceous gout. Ann Rheum Dis. 2009;68:1609. 135. Glazebrook KN, Kahar S, Ida CM, et al. False-negative dual-energy computed tomography in a patient with acute gout. J Clin Rheumatol. 2012;18:138. 136. Nicolaou S, Yong-Hing CJ, Galea-Soler S, et al. Dual-energy CT as a potential new diagnostic tool in the management of gout in the acute setting. AJR Am J Roentgenol. 2010;194:1072. 137. Dalbeth N, Schauer C, Macdonald P, et al. Methods of tophus assessment in clinical trials of chronic gout: a systematic literature review and pictorial reference guide. Ann Rheum Dis. 2011;70:597. 138. Swan A, Amer H, Dieppe P. The value of synovial fluid assays in the diagnosis of joint disease: a literature survey. Ann Rheum Dis. 2002;​61:493. 139. Thiele RG, Schlesinger N. Diagnosis of gout by ultrasound. Rheumatology (Oxford). 2007;46:1116. 140. de Avila Fernandes E, Sandim GB, Mitraud SA, et al. Sonographic description and classification of tendinous involvement in relation to tophi in

chronic tophaceous gout. Insights Imaging. 2010;1:143. 141. Dalbeth N,Doyle AJ. Imaging of gout: an overview. Best Pract Res Clin Rheumatol. 2012;26:823. 142. Thiele RG. Role of ultrasound and other advanced imaging in the diagnosis and management of gout. Curr Rheumatol Rep. 2011;13:146. 143. Yu JS, Chung C, Recht M, et al. MR imaging of tophaceous gout. Am J Roentgenol. 1997;168:523. 144. Dobson M, Alwahab Y, Fazal M. Interphalangeal bone involvement of the big toe in gout: a rare presentation. J Am Podiatr Med Assoc. 2012;102(3):256. 145. Rogers LC, Frykberg RG, Armstrong DG, et al. The Charcot foot in diabetes. J Am Podiatr Med Assoc. 2011;101(5):437. 146. Wunschel M, Wulker N, Gesicki M. Charcot arthropathy of the first metatarsophalangeal joint. J Am Podiatr Med Assoc. 2012;102(2):161. 147. Hastings MK, Johnson JE, Strube MJ, et al. Progression of foot deformity in Charcot neuropathic osteoarthropathy. J Bone Joint Surg Am. 2013;95:1206. 148. de Souza LJ. Charcot arthropathy and immobilization in a weightbearing total contact cast. J Bone Joint Surg Am. 2008;90(4):754. 149. Cofield RH, Morrison MJ, Beabout JW. Diabetic neuroarthropathy in the foot: patient characteristics and patterns of radiographic change. Foot Ankle. 1983;4(1):15. 150. Schon LC, Weinfeld SB, Horton GA, et al. Radiographic and clinical classification of acquired midtarsus deformities. Foot Ankle Int. 1998;19:394. 151. Fernando Grover Páez, Sylvia Elena Totsuka Sutto, Sara Pascoe González, Ernesto G. Cardona Muñóz and Carlos Enrique Medina García (2013). The Diabetic Charcot Foot—New Insights on Treatment, Type 2 Diabetes, Prof. Kazuko Masuo (Ed.), ISBN: 978-953-51-1171-9, InTech,

DOI: 10.5772/56399. Available from: http://www.intechopen.com/books/type-2-​diabetes/the-diabetic-charcot-footnew-insights-on-treatment 152. Eichenholtz SN. Charcot Joints. Springfield, IL: Charles C Thomas; 1966. 153. Johnson JE. Operative treatment of neuropathic arthropathy of the foot and ankle. J Bone Joint Surg. 1998;80:1700. 154. Shibata T, Tada K, Hashizume C. The results of arthrodesis of the ankle for leprotic neuroarthropathy. J Bone Joint Surg Am. 1990;72(5):749. 155. Wukich DK, Sung W. Charcot arthropathy of the foot and ankle: modern concepts and management review. J Diabetes Complications. 2009;23(6):409. 156. Sanders LJ, Frykberg RG. The Charcot foot. In: Frykberg, ed. The High Risk foot in Diabetes Mellitus. New York, NY: Churchill Livingstone; 1991:325–335. 157. Bevilacqua NJ. Current insights on classifying Charcot arthropathy. Podiatry Today. 2009;22(4):22. 158. Sella EJ, Barrette C. Staging of Charcot neuroarthropathy along the medial column of the foot in diabetic patient. J Am Podiatr Med Assoc. 1999;38(1):34. 159. Viswanathan V, Kesavan R, et al. Evaluation of Roger’s Charcot foot classification system in South Indian diabetic subjects with Charcot foot. J Diabet Foot Complications. 2012;4(3, no 2):67. 160. Herbst SA, Jones KB, Saltzman CL. Pattern of diabetic neuropathic arthropathy associated with the peripheral bone mineral density. J Bone Joint Surg Br. 2004;86:378. 161. Jones CW, Agolley D, Burns K, et al. Charcot arthropathy presenting with primary bone resorption. Foot (Edinb). 2012;22:258.

162. Rogers LC, Bevilacqua NJ. The diagnosis of Charcot foot. Clin Podiatr Med Surg. 2008;25:43. 163. Sayner LR, Rosenblum VI. External fixation for Charcot foot reconstruction. Curr Surg. 2005;62(6):618. 164. Rudrappa S, Game F, Jeffcoate W. Recurrence of the acute Charcot foot in diabetes. Diabet Med. 2012;29:819. 165. Schoots IG, Slim FJ, Frederick JS, et al. Neuro-osteoarthropathy of the foot—radiologist: friend or foe? Semin Musculoskeletal Radiol. 2010;14(3):365. 166. Botek G, Hild G. Differential diagnosis of Charcot arthropathy. J Vasc Surg. 2011;53(1):220. 167. Classen JN, Rolley RT, Carneiro R, et al. Management of foot conditions of the diabetic patient. Am Surg. 1976;42(2):81. 168. Varma AK. Charcot neuroarthropathy of the foot and ankle: a review. J Foot Ankle Surg. 2013;52:740. 169. Chantelau E, Poll LW. Evaluation of the diabetic Charcot foot by MR imaging or plain radiography—an observational study. Exp Clin Endocrinol Diabetes. 2006;114:428. 170. Zampa V, Bargellini I, Rizzo L, et al. Role of dynamic MRI in the follow-up of acute Charcot foot in patients with diabetes mellitus. Skeletal Radiol. 2011;40:991. 171. Boyko EJ, Ahroni JH, Stensel VL. Skin temperature in the neuropathic diabetic foot. J Diabetes Complications. 2001;15:260. 172. Armstrong DG, Lavery LA. Monitoring neuropathic ulcer healing with infrared dermal thermometry. J Foot Ankle Surg. 1996;35:335. 173. Tomas MB, Patel M, Marwin SE, et al. The diabetic foot. Br J Radiol. 2000;73:443.

174. Greenhagen RM, Wukich DK, Jung RH, et al. Peripheral and central bone mineral density in Charcot’s neuroarthropathy compared in diabetic and nondiabetic populations. J Am Podiatr Med Assoc. 2012;102(3):213. 175. Rogers LC, Frykberg RG, Armstrong DG, et al. The Charcot foot in diabetes. Diabetes Care. 2011;34:2123.

20 Tumors and Tumorlike Lesions LAWRENCE OSHER, ROCCO PETROZZI, AND ROBERT A. CHRISTMAN GENERAL CONSIDERATONS Primary bone tumor and tumorlike conditions rarely occur in the foot, with most reports ranging from approximately 2% to 3.5%. Studies generally concur that the phalanges and metatarsal bones are the most frequently involved pedal sites. Likewise, neoplasms of the midfoot bones are decidedly rare, leaving the talus and calcaneus as the second most common site behind the small tubular forefoot bones. Looking at the entire skeletal system, a given bone tumor, when diagnosed, is approximately 4 times more likely to be benign versus malignant. When encountered, a given bone tumor is approximately 95 times more likely to be metastatic spread than a primary lesion. In addition, as a general rule, most metastatic lesions tend not to occur distal to the knee. If one adds all of this together, it is easy to see why the majority of pedal bone tumors encountered are benign. Many practitioners are of the opinion that enchondroma is the most commonly encountered tumor of the foot, based on a well-known study by Coley and Higinbotham.1 However, the frequency of pedal enchondroma in the small tubular bones is probably overestimated, as this lesion is far more commonly encountered in the hand. For example, a number of more recent studies indicate that lesions such as osteochondroma and chondroblastoma are more frequent encounters. Of the primary malignant bone tumors, Ewing sarcoma is most common in younger age groups and chondrosarcoma appears the most frequent in adults. As noted earlier, metastases are not common distal to the knee. When acrometastatic lesions occur in the foot, they are most frequently secondary to

lung carcinoma or an aggressive visceral tumor. Lung (and some visceral) tumors may demonstrate symptomatic periostitis as a manifestation of hypertrophic pulmonary osteoarthropathy. When present, periostitis can commonly extend to the ankles and present with pain and swelling. However, foot involvement with hypertrophic osteoarthropathy is rare. Tumors with metastatic extension to the vertebral bodies are also known for the ability to extend distal to the knee (theoretically via the Batson vertebral plexus). In long bones of the extremities, metastatic lesions may be lytic, blastic, or, more frequently, mixed lytic and blastic, with a propensity to demonstrate moth-eaten destruction. Periosteal reactions are variable, and may or may not comprise part of the overall picture. Acrometastasis is more likely to manifest gross lysis without periostitis. When an acrosclerotic lesion occurs, the differential diagnosis should include underlying conditions such as systemic lupus erythematosus, seronegative arthritis, and intraosseous sarcoidosis. As a general rule, radiography is best used for diagnosis and differential diagnosis of a bone tumor, whereas staging is performed with computed tomography (CT) or magnetic resonance imaging (MRI). The cardinal radiographic characteristics with respect to patterns of internal bony lysis, periosteal reaction, cortical erosion, and matrix evaluation are used to assess bone tumors. Beyond this, probabilities in diagnosis frequency rest on epidemiologic data such as patient age and specific tumor location. Clinical examination generally yields the nonspecific finding of pain, which often begins insidiously with an intermittent character. Duration of pain can be a clue as to the aggressiveness of the lesion, and progression to a severe, constant pain can occur with the more aggressive neoplasms or pathologic fracture. Other findings may include swelling, distension, and/or palpable mass; variable inflammation and local venous dilation; effusion into an adjacent joint; and occasional constitutional symptoms. SYSTEMATIC EVALUATION OF SOLITARY BONE LESIONS Solitary bone lesions are occasionally encountered in the foot and distal leg. Fortunately, most of these lesions are benign. Examples include fibrous cortical defect (FCD) and nonossifying fibroma (NOF), solitary bone cyst, enchondroma, and bone island. Some lesions have characteristic radiographic

features; however, many appear similar to one another or only demonstrate subtle differences, and it may be impossible to distinguish between them by radiography alone. Determining the epicenter of the lesion and preferential direction of spread are often important discriminating clues. Unfortunately, much of this information cannot be determined when bone tumors are encountered in the small tubular bones of the foot as most, if not all, of the bone is involved by the time of discovery. On the other hand, early or incidental detection of benign lesions in small tubular bones tends to recapitulate behavior in extremity long bones. An efficient, reliable approach is needed to assess these lesions. The radiographic features must first be recognized. These features can be used not only as diagnostic clues but also for determining the growth rate or aggressiveness of the lesion. A list of potential differential diagnoses can then be formulated based on these radiographic findings. TABLE 20-1  

Diagnostic Indicators for Evaluating the Solitary Bone Lesion

Radiographic Clues  1. Destructive patterna  2. Size and shapea  3. Cortical involvementa  4. Periosteal reactiona  5. Anatomic position (transverse and longitudinal planes)  6. Skeletal location  7. Trabeculation  8. Matrix production Nonradiologic Clues

 1. Clinical course  2. Age of patient aFeatures useful for determining a lesion’s growth rate or aggressiveness.

Ten diagnostic clues are used for assessing a solitary bone lesion (Table 201). Eight of these clues are radiographic features, and, of these, four are valuable for determining the aggressiveness or growth rate of the lesion. Two nonradiologic diagnostic clues complete the initial assessment: the patient’s age and clinical course. These ten criteria (eight radiographic and two clinical) should be considered when evaluating any solitary bone lesion. After these data are collected, differential diagnoses are determined. Destructive Pattern Three types of destructive (lytic) patterns have been described in the literature pertaining to solitary bone lesions: geographic, moth-eaten, and permeative (Figure 20-1).2 The term geographic implies large area, and lesions are typically 1 cm or greater in size. Geographic lytic lesions are often referred to as “cystic.” A sharply outlined lesion with a narrow zone of transition demonstrates a definite form or shape. However, the term cystic should not be used, because it implies that the lesion is not solid; a greater percentage of cases are solid.3 The interface with normal bone is known as the margin; it may be well defined, with or without reactive bone formation (“sclerotic halo”), or ill defined. In general, well-defined geographic lesions demonstrate less aggressive behavior and slower growth activity (Figure 202). Examples of geographic lesions include most benign bone tumors. If, over sequential studies, the margin becomes increasingly less defined, increased biologic activity should be suspected.4

FIGURE 20-1. Schematic diagram of patterns of bone destruction and their margins. Arrows indicate the most frequent transitions or combinations of these margins. Transitions imply increased activity and a greater probability of malignancy. (From Madewell JE, Ragsdale BD, Sweet DE. Radiologic and pathologic analysis of solitary bone lesions: part I, internal margins. Radiol Clin North Am. 1981;19(4):722, figure 6.)

FIGURE 20-2. Calcaneal bone cyst demonstrating geographic grade IB destruction. As noted earlier, well-defined geographic bone lesions bounded by a sclerotic margin tend to have slow growth activity (Figure 20-3). The sclerotic margin represents the surrounding normal bone’s attempt to wall off the lesion. In contrast, normal bone cannot be deposited quickly enough around an aggressive moth-eaten or permeative lesion to be visible radiographically. When a sclerotic margin rims a geographic lytic lesion, the lesion is usually

slowly progressive and therefore probably benign in nature.5 It is nonetheless important to note that slow growth should not automatically be equated with the term “benign,” as low-grade malignant lesions (e.g., myxoid chondrosarcoma) may also manifest with well-defined geographic destruction.

FIGURE 20-3. Calcaneal lesion demonstrating a geographic type IA destructive pattern; arrows indicate the sclerotic margin.

FIGURE 20-4. Permeated (type III) destruction involves the entire distal tibial diametaphysis, characteristic of fast growth rate. When a narrow zone of transition exists between normal bone and lytic tumor, the lesion typically takes form, or has a defined shape. Therefore, in contradistinction, moth-eaten and permeative destructive patterns do not have a definite form or shape (Figure 20-4). Multiple, small, lytic lesions appear to spread across the affected area of bone. A moth-eaten pattern appears as multiple, small holes in cancellous or medullary bone; the permeative pattern typically presents as multiple, small streaks running through cortical bone. Lesions with either of these destructive patterns highly suggest aggressive activity. Examples include Ewing sarcoma, osteosarcoma, and untreated osteomyelitis. Acute osteopenia can also demonstrate a permeative or motheaten appearance, except it involves multiple bones (Figure 20-5).

FIGURE 20-5. Rapidly forming, acute regional osteoporosis can demonstrate permeative (grade III) destruction of bone. In osteopenic disorders (e.g., complex regional pain syndrome), these patterns typically coalesce in the long term, leading to generalized osteopenia. Size and Shape

Generally speaking, aggressive lesions tend to occupy larger areas of bone than slow-growing lesions. Regarding shape, geographic lytic lesions tend to be round to oval. As a general rule, lesions that appear in long bones during periods of skeletal growth and arise outside of the epiphysis tend to be oval, paralleling longitudinal bone growth, whereas intraepiphyseal lesions remain round. Evaluating the contour of the lesion can also be quite useful in foretelling the nature of a bone tumor or tumorlike condition. The contours of many lesions are smooth. However, when contours appear lobulated, cartilaginous and occasionally fibrous lesions are suspect. In some instances, outlines of geographic zones of pathology (not necessarily lytic lesions) are “serpiginous,” as typified by medullary infarcts or occasionally bone infection. Cortical Involvement A solitary bone lesion’s growth rate also can be estimated by the pattern of resorption along the adjacent cortex. Slow growth may only cause scalloping of the endosteal surface (Figure 20-6). With increasing growth rate, the cortex may be nearly fully resorbed yet still intact adjacent to the lesion. As endosteal resorption occurs, subperiosteal apposition may accompany it, giving the cortex an “expanded” appearance (Figure 20-7). A more aggressive lesion penetrates or breaks through the cortex and invades the soft tissues.

FIGURE 20-6. This lesion’s slow growth rate has resulted in scalloping of the endosteal surface of the cortex (arrows). Periosteal Reaction Periosteal reactions can have many different presentations (Figure 20-8). When a periosteal reaction is present, a degree of aggressiveness does exist, and the type of periosteal reaction present can be used to further estimate the growth rate (Table 20-2). For example, lesions that are accompanied by a continuous single layer or solid periosteal reaction are less aggressive than those demonstrating multiple layers, spicules, or fine lace-like patterns. Unless associated with a solid periosteal reaction, interrupted periosteal reactions suggest an aggressive lesion, as this signifies soft tissue extension and is therefore an upstaging sign. This is also the case with complex periosteal reactions.6 As a general rule, the more “porous” the periosteal pattern, the greater the likelihood of an aggressive lesion. Finally, layered appearances can be associated with underlying benign processes, especially when layers appear to be “stuck” or “glued” together (not distinctly separated by lucent layers).

FIGURE 20-7. Chronic endosteal resorption caused by a slow-growing lesion and subsequent periosteal apposition results in the appearance of an “expanded” cortex. Anatomic Position The anatomic position of the lesion in a tubular bone can provide valuable diagnostic information. Many tumors tend to occur in specific anatomic locations. Conceptually, bone tumors are thought to spread like “waves on a pond,” and, in this regard, one should always endeavor to determine the center (or, epicenter) of a lesion. Position can be assessed in two planes of the image: horizontal and vertical (Figure 20-9). In the horizontal plane, the center of the lesion can have a position that is central (solitary bone cyst, fibrous dysplasia, enchondroma), eccentric (NOF, aneurysmal bone cyst [ABC]), cortical (osteoid osteoma), or parosteal (osteoma). A lesion’s vertical position can be epiphyseal (chondroblastoma, giant-cell tumor [GCT]), metaphyseal (enchondroma, chondromyxoid fibroma [CMF]), diametaphyseal (CMF, osteosarcoma), or diaphyseal (fibrous dysplasia, osteoblastoma, Ewing sarcoma). Skeletal Location Some solitary bone lesions have a predilection for certain bones. For example, enchondroma is frequently found in the phalanges of the hand and foot, although far more commonly situated in the hand. Osteoid osteoma may be found in the dorsal talar neck. Intraosseous lipoma and the solitary bone cyst are rare, but when encountered in the foot, they are most frequently situated in the calcaneal body. Trabeculation Trabeculation, when present, is found with only a few lesions (Box 20-1). Also, the type of trabeculation may further narrow the choice of diagnoses. Fine, delicate trabeculation (Figure 20-7) is associated with GCT; horizontally oriented fine, delicate trabeculation is a feature of ABC; coarse, thick trabeculations are seen with CMF; FCD has lobulated trabeculations that impart the appearance of “soap bubbles” (Figure 20-10); and radiating

trabeculations are seen with hemangioma.7 Matrix Production Most tumors do not radiographically demonstrate visible extracellular matrix. However, for those that do, matrix mineralization can narrow the list of differential diagnoses for any particular lesion to a specific category of primary bone tumors (Figure 20-11).8 Ossific mineralized matrix patterns manifest with solid, cloud-like, homogeneous or “ivory-like” increased density. Examples of osteoblastic tumors include osteoblastoma and osteosarcoma. In contradistinction, cartilaginous or chondroid matrix patterns typically present with stippled, punctate, flocculent, popcorn-like, or curvilinear areas of increased density. Although these patterns are mainly encountered in cartilaginous tumors, they may also be noted in areas of tissue necrosis and infarction. Examples include chondroma (Figure 20-12) and chondrosarcoma. Occasionally, a periphery of “pearls” of neoplastic cartilage will calcify yielding a “rings-and-arcs” appearance. Fibroblastic lesions generally do not radiographically demonstrate a matrix and therefore appear lucent, with exception of fibrous dysplasia, which can produce the so-called “ground-glass” matrix, which appears smoky, milky, or smudgy. Clinical Course The clinical presentation of most tumors is nonspecific; complaints may include pain and swelling. Inflammatory signs are not necessarily present. In some instances, pain is secondary to pathologic fracture. There are a few instances where lesions have a fairly characteristic clinical presentation. One well-known example is osteoid osteoma, which classically presents with night pain that is relieved by aspirin.

FIGURE 20-8. Schematic diagram of periosteal reactions. The arrows indicate that the continuous reactions may be interrupted. (From Ragsdale BD, Madewell JE, Sweet DE. Radiologic and pathologic analysis of solitary bone lesions: part II, periosteal reactions. Radiol Clin North Am. 1981;19(4):751, figure 2.) TABLE 20-2   Periosteal Reactions and Lesion Activity

Age of Patient Most bone tumors tend to have peak occurrences in certain age ranges (Table 20-3).9 For example, Ewing tumor, simple bone cyst, and chondroblastoma are found in the first two decades of life. In contrast, multiple myeloma, fibrosarcoma, and metastasis present in the fifth to seventh decade. Adjunctive Imaging Studies

The use of ancillary imaging modalities such as CT and MRI in the evaluation of musculoskeletal lesions is routine nowadays, and it has also become commonplace in the approach to bone tumors and tumorlike lesions. Some of the significant advantages over radiography that both MRI and CT inherently possess include sectional imaging, improved resolution (especially contrast resolution with MRI for soft tissue evaluation), ability to detect fluid–fluid levels, and ability to employ contrast agents like gadolinium (for MRI) to help characterize lesions and accurately determine extent. Nevertheless, CT, MRI, nuclear medicine, positron emission tomography (18FDG-PET), and ultrasound are rightfully labeled “ancillary” imaging modalities, as radiography is still the imaging modality of first choice.

FIGURE 20-9. Composite diagram illustrating frequent sites of bone tumors. The diagram depicts the end of a long bone that has been divided into the epiphysis, metaphysis, and diaphysis. The typical sites of common primary bone tumors are labeled. Bone tumors tend to predominate in those ends of long bones that undergo the greatest growth and remodeling; hence, they have the greatest number of cells and amount of cell activity (shoulder and knee regions). When small tumors, presumably detected early, are analyzed, preferential sites of tumor origin become apparent within each bone (as shown in this illustration), suggesting a relationship between the type of tumor and the anatomic site affected. In general, a tumor of a given cell type arises in the field where the homologous normal cells are most active. These regional variations suggest that the composition of the tumor is affected or may be determined by the metabolic field in which it arises. (From Madewell JE, Ragsdale BD, Sweet DE. Radiologic and pathologic analysis of solitary bone lesions: part I, internal margins. Radiol Clin North Am. 1981;19(4):716, figure 1.)

BOX 20-1 Tumors Demonstrating Trabeculation Giant-cell tumor Aneurysmal bone cyst Chondromyxoid fibroma Hemangioma Fibrous cortical defect SELECTED MALIGNANT TUMORS The “big three” primary mesenchymal malignancies—osteosarcoma, chondrosarcoma, and fibrosarcoma—are rarely encountered in the foot. Osteogenic Sarcoma (Osteosarcoma) Osteogenic sarcoma is an aggressive bone tumor pathologically classified as a malignant mesenchymal neoplasm. The tumor cells directly produce immature bone, which is characterized by profound osteoid elaboration by sarcomatous cells. Pathologically, osteoid may be mixed with cartilage and/or fibrous tissue of varying amounts, and this may have radiologic consequences. Statistically, osteogenic sarcoma is the second most common primary malignant bone tumor bodywide, second only to multiple myeloma, and is approximately three times as common as Ewing sarcoma. That being said, it should be noted that when the foot is involved, several larger series have found Ewing sarcoma to be more prevalent.10,11 There are a number of osteosarcoma subtypes, and even extraskeletal variants are known (Table 204). Nevertheless, the majority (approximately 75%) of all lesions are intramedullary osteosarcomas. In addition to primary lesions, osteosarcomatous degeneration has been described in association with a number of benign conditions such as Pagetic bone, chronic osteomyelitis, osteonecrosis, fibrous dysplasia, and osteochondroma.12

FIGURE 20-10. Examples of fibrocortical defects (NOFs) at varying stages of development. A: The trabeculations appear fine and delicate in early stages. B: In an older patient, the lesion remodels, and the “soap-bubble” appearance is more obvious.

FIGURE 20-11. Schematic diagram of mineralized matrix patterns. Tumor osteoid appears as increased density with a solid (sharp-edged) or cloudy to ivory-like (ill-defined edge) pattern. Tumor cartilage creates stippled, flocculent, and solid density patterns. Rings and arcs represent bony rims around tumor cartilage lobules. Dystrophic mineralization and ischemic osteoid tend to mimic the stippled, flocculent, or patchy solid density pattern. (From Sweet DE, Madewell JE, Ragsdale BD. Radiologic and pathologic analysis of solitary bone lesions: part III, matrix patterns. Radiol Clin North Am. 1981;19(4):788, figure 1.)

FIGURE 20-12. Enchondroma, second toe proximal phalanx, demonstrating a rings-and-arcs mineralized matrix pattern. (Courtesy of Gary F. Bjarnason, Roanoke Rapids, NC.) TABLE 20-3   Age Ranges of Bone Tumor Distributiona Tumor Peak Age Range (Years) Benign Tumors Osteoma 15–45 Osteoid osteoma 10–23 Benign osteoblastoma 10–30 Osteochondroma 10–30 Central chondroma 10–40 Chondroblastoma 10–20 Chondromyxoid fibroma 10–30 Eosinophilic granuloma 5–10 Nonosteogenic fibroma 10–20 Desmoplastic fibroma 10–30 Intraosseous lipoma 30–50 Neurilemoma Hemangioma Giant-cell tumor Simple bone cyst Aneurysmal bone cyst Enchondroma Primary Malignant Tumors Osteogenic sarcoma Parosteal osteosarcoma Chondrosarcoma Fibrosarcoma Malignant giant-cell tumor Adamantinoma Hemangioendothelioma Ewing sarcoma Reticulum cell sarcoma Myeloma Chordoma

10–30 40–50 25–40 5–20 10–30 30s 10–20 20–40 30–60 30–40 30–50 10–30 30–40 10–20 30–60 50–80 30–60

Other Tumors Leukemia: Acute Leukemia: Chronic Metastatic neuroblastoma Metastatic carcinoma

2–6 40–70