W. Richard Webb, MD Chief, Thoracic Imaging Professor Emeritus of Radiology and Biomedical Imaging Emeritus Member, Haile Debas Academy of Medical Educators University of California, San Francisco San Francisco, California William E. Brant, MD, FACR Professor Emeritus Department of Radiology and Medical Imaging University of Virginia Health System Charlottesville, Virginia Nancy M. Major, MD Musculoskeletal Radiologist Professor of Radiology Stony Brook University Hospital Stony Brook, New York Mainline Diagnostic Imaging Bryn Mawr, Pennsylvania FUNDAMENTALS OF BODY CT Fourth Edition 1600 John F. Kennedy Blvd. Ste 1800 Philadelphia, PA 19103-2899 FUNDAMENTALS OF BODY CT, Fourth Edition ISBN: 978-0-323-22146-7 Copyright © 2015 by Saunders, an imprint of Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the Publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Library of Congress Cataloging-in-Publication Data Webb, W. Richard (Wayne Richard), 1945- author. Fundamentals of body CT / W. Richard Webb, William E. Brant, Nancy M. Major. -- Fourth edition. p. ; cm. Includes bibliographical references and index. ISBN 978-0-323-22146-7 (paperback : alk. paper) I. Brant, William E., author. II. Major, Nancy M., author. III. Title. [DNLM: 1. Tomography, X-Ray Computed. WN 206] RC78.7.T6 616.07’572--dc23 2014012926 Content Strategist: Helene Caprari Content Development Specialist: Kelly McGowan Publishing Services Manager: Patricia Tannian Senior Project Manager: Sharon Corell Manager, Art and Design: Steven Stave Printed in China Last digit is the print number: 9 8 7 6 5 4 3 2 1 Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluat- ing and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. With respect to any drug or pharmaceutical products identified, readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administra- tion, and contraindications. It is the responsibility of practitioners, relying on their own experience and knowledge of their patients, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. To Jack and Cole, my grandsons. They understand the value of fundamentals. W.R.W. To my wife and true companion, Barbara. In memory of my daughter Rachel. To our children and their spouses and our ten grandchildren: Evan, Finley, Sophia, Katie, Josie, Danielle, Dylan, Grayson, Amelia, and Noah. W.E.B. To Kenneth…it was meant to be…. N.M.M. vii Preface Despite the fact that we concentrate on the “fundamentals” in this book, the fundamentals keep changing along with advances in CT tech- niques and our improved understanding of vari- ous diseases. This new edition gives us the chance to update important topics and add new material, including a number of new, high-quality images. That is what we have attempted to do, without significantly increasing the size of the book or making it less accessible as a teaching tool. The half-dozen or so years since the third edi- tion was published have seen continued advances in helical CT techniques. In this edition, we review the various spiral/helical CT protocols currently used in clinical practice for the diagno- sis of chest, abdominal, and musculoskeletal abnormalities. We have expanded discussions of high-resolution CT, lung nodule assessment and lung cancer screening, CT pulmonary embolism diagnosis, CT enterography, CT enteroclysis, CT colonography, and optimizing CT tech- niques in musculoskeletal diagnosis. New topics, discussions of new diseases (too numerous to mention here), and new images have been added to all chapters, including updated descriptions and illustrations of normal anatomy and incidental findings. Disease classifications, including those for pulmonary adenocarcinoma, diffuse lung diseases, and pancreatic lesions, have been updated where appropriate. We hope you enjoy and profit from our efforts. W. Richard Webb William E. Brant Nancy M. Major ix Preface to the first edition Instead of writing a text intended to record every- thing that is known about body CT or even everything that we know about body CT, we have attempted to write one that teaches how to per- form and read body CT scans. In doing this, we have tried to limit ourselves to discussing what is key to understanding body CT from a practical clinical standpoint—the key anatomy, the key concepts, the key diseases, and the key controver- sies. We have done this at the risk of leaving a few things out, but it isn’t necessary to read about everything when you are first learning a subject. In other words, Fundamentals of Body CT is not intended to provide more than the best CT texts on the market do, but rather less, with a different emphasis, and in a more manageable package. Each of us has written a different part of this book, obviously depending on our areas of expertise. Since each of us teaches in a slightly different way, each of the three sections of the book—the thorax, the abdomen, and the muscu- loskeletal system—is somewhat different in approach. We hope that by preserving our indi- vidual styles we have made the book more inter- esting to read, and for us, it certainly made this book easier to write. FUNDAMENTALS OF BODY CT Fourth Edition 3 C H A P T E R 1 IntroductIon to ct of the thorax: chest ct technIques W. Richard Webb Spiral or helical computed tomography (CT) allows the entire chest to be easily imaged during a single breath hold, with volumetric acquisition and exact registration or overlapping of slices. Two- and three-dimensional reformations may be performed if desired. Because scanning is rapid, contrast agents can be injected quickly, excellent vascular opacification can be achieved, and reduced volumes of contrast agent can be used. SPIRAL CT IN CHEST DIAGNOSIS Multiple-detector CT (MDCT) scanners have multiple parallel rows (e.g., 64, 16, or 4) of x-ray detectors, although the number varies with the machine. Generally speaking, MDCT scanners with fewer rows (e.g., 16 or 4) are being replaced with scanners having more rows (e.g., 64), but this is an expensive and, therefore, gradual process. With MDCT, each of the detector rows records data independently as the gantry rotates; consequently, a volume of the patient is imaged with each gantry rotation. For example, the detector rows in one 64-detector scanner are 0.625 mm in width, and the multiple scanner rows add up to 40 mm (0.625 mm × 64). The term pitch refers to the table excursion during a complete gantry rotation divided by the width of all the detectors used (e.g., detector width × number of detector rows). With MDCT, pitch usually ranges from 1 to 2. The higher the pitch, the faster the scan is, but images are generally noisier, spatial resolution is reduced somewhat, and the effective slice thickness (the thickness of the patient that is actually imaged) is increased. Gantry rotation time is usually about 0.5 seconds. The formula relating scan parameters for MDCT is shown in Fig. 1-1. Assuming a scan volume (patient length from the lung apices to bases) of 30 cm (300 mm), which is generally sufficient to image the chest, a 64-detector scanner with a detector width of 0.625 mm, pitch of 1.5, and gantry rotation time of 0.5 seconds, the total scan duration would be 2.5 seconds. This easily allows imaging of the entire chest during a single breath hold. For MDCT scanners with fewer detector rows (e.g., 16 or 4), the scan duration is longer, but scanning during a single breath hold is still achievable in cooperative patients. For example, with a four-detector-row scanner and a detector width of 1.25, the scan time for a tho- racic CT would be 20 seconds. If the patient is dyspneic or uncooperative, respiratory motion may occur during the scan, with resulting degra- dation of image quality. SPIRAL CHEST CT: GENERAL PRINCIPLES The specific protocols used for chest CT depend on the scanner used, the scanner manufacturer, and the reason for the study. However, several general principles apply to all chest scans (Table 1-1). Scan Levels Chest CT is usually obtained from a level just above the lung apices (near the suprasternal notch) to the level of the posterior costophrenic angles; these scans also encompass the diaphragm and the upper abdomen. The distance (or volume) needed to cover the thorax is determined by a preliminary projection scan (e.g., “scout view”) and is usually about 25 to 30 cm. Patient Position Routinely, patients are scanned supine. Prone scans may be obtained for high-resolution CT The Thorax 4 (HRCT) or to assess movement of pleural fluid collections. The patient may also be positioned prone for biopsy of posterior lung lesions or drainage of pleural fluid collections. Lung Volume Scans are routinely obtained after a full inspiration (i.e., at total lung capacity) and during suspended respiration. Post-expiratory scans may be performed in some cases (particularly on HRCT) to assess air trapping. Gantry Rotation Time A rapid gantry rotation time is used to reduce the scan time. Times are usually about 0.5 seconds. Slice Thickness and Pitch (Table Excursion) Usually, scan data are acquired using the thinnest detector width, and the slice thickness used for scan interpretation is determined by the indication for the scan. For example, if data are recorded using 0.625-mm detectors, slices can be reconstructed at any thickness from 0.625 to 5 mm for viewing. Thinner slices are required for some specific indica- tions, whereas thicker slices may be appropriate for other studies, are quicker to interpret, and do not occupy as much memory when they are stored. Most chest scans are reconstructed with a thick- ness of 1 to 1.25 mm. When viewing a study recon- structed with 2.5- or 5-mm-thick slices, if the scan data were collected using thinner detectors and if the scan data are still available (data are usually preserved on the scanner disk for a day or two), thinner slices can be reconstructed at a later time. Keep in mind that with the spiral technique, the actual thickness of the slice viewed (i.e., “effective slice thickness”) may be greater than the slice thickness selected (e.g., 1.25 mm), depending on the pitch or table excursion during gantry rotation; the greater the pitch, the greater the effective slice thickness. Thus, there is a trade-off; with a higher pitch, the study is quicker, but the scans are not quite as good. Usually, slices are reconstructed at an interval equal to the slice thickness (e.g., 1.25 mm) to provide a volumetric data set. On occasion, scans are reconstructed at overlapping levels (e.g., 1.25-mm slices reconstructed at 0.625-mm inter- vals), although this is not generally necessary. Scan Duration MDCT of the chest can be easily obtained during a single breath hold, generally avoiding respira- tory motion artifacts, except in very dyspneic or uncooperative patients. However, if thin slices, high resolution, or limited table excursion is desired, the scan will take longer to obtain and artifacts resulting from respiratory motion or cardiac pulsation are more likely to occur. Reconstruction Algorithm Once the scans have been performed, the scan data are reconstructed using an algorithm that determines some characteristics of the resulting FIGURE 1-1 n Parameters for multidetector computed tomography (MDCT). TABLE 1-1 Chest CT: General Principles Scan levels Lung apices to the posterior costophrenic angles Patient position Supine; prone scans sometimes used for diagnosis of lung disease or pleural effusions Lung volume Full inspiration, single breath hold; expiratory scans sometimes used to diagnose air trapping Gantry rotation time Approximately 0.5 s in most instances Scan duration Approximately 2.5 s for the thorax using MDCT and fast scanning Detector width The thinnest detectors (e.g., 0.625 mm) are typically used for image acquisition Pitch (table excursion) Depends on tolerable image noise; increased if noise is OK, decreased if there is a desire for high resolution Reconstruction algorithm High-resolution algorithm used for most studies; standard or soft-tissue algorithm usually used for vascular studies Two- or three- dimensional reconstruc- tions Not routine; occasionally useful for lung, airway, or vascular studies Contrast agents Intravenous contrast injection in some cases; oral contrast agents only for gastrointesti- nal abnormalities 1 IntroductIon to ct of the thorax: chest ct technIques 5 image. For routine chest imaging, a high-resolu- tion algorithm is often used to optimize detail, but this makes the image somewhat grainy in appearance. A standard or soft-tissue algorithm, which produces a smoother image, is better for assessing thoracic vascular structures (e.g., studies per- formed for diagnosis of pulmonary embolism, aneurysm, or aortic dissection) but is not optimal for other chest imaging. This algorithm is often used for abdominal imaging. Two- and Three-Dimensional Reconstruction Because scan data are acquired continuously and volumetrically using spiral CT, scans may be reconstructed in any plane desired, if appropriate workstations are available. A variety of display techniques have been used for imaging the tho- rax. These include multiplanar reconstructions, three-dimensional shaded surface display or vol- ume rendering from an external perspective, or shaded surface or volume rendering from an internal (i.e., endoluminal) perspective, also known as virtual bronchoscopy. Multiplanar, two-dimensional reconstructions offer the advantage of being quickly performed and are sufficient for diagnosis in most cases in which a reformation is considered desirable. Subsequent chapters provide a number of examples of two- dimensional reconstructions. Three-dimensional techniques, such as shaded surface display and vol- ume rendering, can be valuable in selected cases, but they are time-consuming and require consider- able operator experience. These techniques are not commonly used in clinical chest imaging, with the exception of virtual bronchoscopy or airway imag- ing and specialized vascular imaging. Maximum- or minimum-intensity projection images representing a slab of three-dimensional data reconstructed from a volumetric data set may sometimes be useful in imaging pulmonary, airway, or vascular abnormalities. Window Settings For chest CT, scans must be viewed using at least three different window settings. Scans are usually viewed using a workstation having preset windows available. Presets used when reading chest CT are generally termed “lung,” “soft-tissue” (or “mediastinal”), and “bone” windows, names that also describe their primary use. Often, these preset windows are adjusted by the viewer during scan interpretation to optimize the visibility of certain structures or abnormalities of interest. Lung windows typically have a window mean of approximately −600 to −700 HU (Hounsfield units) and a window width of 1000 to 1500 HU. Lung windows best demonstrate lung anatomy and pathology, contrasting soft-tissue structures with surrounding air-filled lung parenchyma. Mediastinal or soft-tissue windows (window mean, 20 to 40 HU; width, 450 to 500 HU) dem- onstrate soft-tissue anatomy in the mediastinum and in other areas of the thorax, allowing the dif- ferentiation of fat, fluid, tissue, calcium, and con- trast-opacified vessels. These windows is also of value in providing information about consoli- dated lung, the hila, pleural disease, and struc- tures of the chest wall. Subsequent chapters discuss more specific uses of these two windows. In the assessment of vascular structures (e.g., for pulmonary embolism or dissection diagnosis), a wider window or higher window mean than that used for a routine mediastinal window is often selected by the radiologist to see better within the dense contrast column. Bone windows typically have a window mean of approximately 300 to 500 HU and a window width of 2000 HU. They best demonstrate skel- etal structures or very dense objects. This win- dow is sometimes valuable in looking at densely opacified vascular structures. SPIRAL CHEST CT: PROTOCOLS In most patients, chest CT is performed using a routine protocol. This technique is designed to provide useful information about the lung, medi- astinum, hila, pleura, and chest wall. It is valuable in the diagnosis of a variety of diseases and types of abnormalities. Modified CT techniques are used in specific clinical settings or to look for spe- cific abnormalities (e.g., pulmonary embolism, aortic dissection, and diffuse lung disease). Subse- quent chapters provide detailed reviews of some specific protocols. For current scanners with a large number of detector rows (e.g., 64), protocols for evaluation of different sorts of thoracic abnormalities have become more alike, since scanning of thin slices and with excellent contrast opacification can eas- ily be performed during a single breath hold, regardless of why the scan is being done. A gen- eral understanding of the principles involved in obtaining CT for specific indications is much more important than knowing detailed specific protocols, because these vary with different scan- ners and manufacturers, and among different institutions. Use of Contrast Agents Chest CT can be performed with or without the administration of an intravenous contrast agent,