Diagnostic Abdominal Imaging [1 ed.] 9780071807258, 007180725X, 9780071623537, 0071623531

A detailed, pattern-based approach to abdominal imaging interpretation Diagnostic Abdominal Imaging provides a comprehen

825 110 98MB

English Pages [961] Year 2012

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Diagnostic Abdominal Imaging [1 ed.]
 9780071807258, 007180725X, 9780071623537, 0071623531

  • Commentary
  • TRUE PDF

Table of contents :
Contents
Contributors
Preface
Acknowledgments
Chapter 1 Fluid, Fat, Blood, Calcium, and Contrast: General Imaging Features
Chapter 2 Imaging of the Bowel
Chapter 3 Imaging of the Liver
Chapter 4 Imaging of the Gallbladder and Biliary System
Chapter 5 Imaging of the Pancreas
Chapter 6 Imaging of the Kidney
Chapter 7 Imaging of the Urinary System
Chapter 8 Imaging of the Adrenal
Chapter 9 Imaging of the Uterus and Cervix
Chapter 10 Imaging of the Ovary and Fallopian Tubes
Chapter 11 Imaging of the Prostate and Seminal Vesicles
Chapter 12 Imaging of the Scrotum and Penis
Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts
Chapter 14 Imaging of the Spleen
Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis
Index
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
R
S
T
U
V
W
X
Y
Z

Citation preview

Diagnostic Abdominal Imaging

Notice Medicine is an ever-changing science. As new research and clinical experience broaden our knowledge, changes in treatment and drug therapy are required. The authors and the publisher of this work have checked with sources believed to be reliable in their efforts to provide information that is complete and generally in accord with the standards accepted at the time of publication. However, in view of the possibility of human error or changes in medical sciences, neither the authors nor the publisher nor any other party who has been involved in the preparation or publication of this work warrants that the information contained herein is in every respect accurate or complete, and they disclaim all responsibility for any errors or omissions or for the results obtained from use of the information contained in this work. Readers are encouraged to confirm the information contained herein with other sources. For example and in particular, readers are advised to check the product information sheet included in the package of each drug they plan to administer to be certain that the information contained in this work is accurate and that changes have not been made in the recommended dose or in the contraindications for administration. This recommendation is of particular importance in connection with new or infrequently used drugs.

Diagnostic Abdominal Imaging Wallace T. Miller Jr., MD Associate Professor of Radiology and Medicine University of Pennsylvania School of Medicine Philadelphia, Pennsylvania

New York Chicago San Francisco Lisbon London Madrid Mexico City Milan New Delhi San Juan Seoul Singapore Sydney Toronto

Copyright © 2013 by The McGraw-Hill Companies, Inc. All rights reserved. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the prior written permission of the publisher. ISBN: 978-0-07-180725-8 MHID: 0-07-180725-X The material in this eBook also appears in the print version of this title: ISBN: 978-0-07-162353-7, MHID: 0-07-162353-1. All trademarks are trademarks of their respective owners. Rather than put a trademark symbol after every occurrence of a trademarked name, we use names in an editorial fashion only, and to the benefit of the trademark owner, with no intention of infringement of the trademark. Where such designations appear in this book, they have been printed with initial caps. McGraw-Hill eBooks are available at special quantity discounts to use as premiums and sales promotions, or for use in corporate training programs. To contact a representative please e-mail us at [email protected]. TERMS OF USE This is a copyrighted work and The McGraw-Hill Companies, Inc. (“McGraw-Hill”) and its licensors reserve all rights in and to the work. Use of this work is subject to these terms. Except as permitted under the Copyright Act of 1976 and the right to store and retrieve one copy of the work, you may not decompile, disassemble, reverse engineer, reproduce, modify, create derivative works based upon, transmit, distribute, disseminate, sell, publish or sublicense the work or any part of it without McGraw-Hill’s prior consent. You may use the work for your own noncommercial and personal use; any other use of the work is strictly prohibited. Your right to use the work may be terminated if you fail to comply with these terms. THE WORK IS PROVIDED “AS IS.” McGRAW-HILL AND ITS LICENSORS MAKE NO GUARANTEES OR WARRANTIES AS TO THE ACCURACY, ADEQUACY OR COMPLETENESS OF OR RESULTS TO BE OBTAINED FROM USING THE WORK, INCLUDING ANY INFORMATION THAT CAN BE ACCESSED THROUGH THE WORK VIA HYPERLINK OR OTHERWISE, AND EXPRESSLY DISCLAIM ANY WARRANTY, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. McGraw-Hill and its licensors do not warrant or guarantee that the functions contained in the work will meet your requirements or that its operation will be uninterrupted or error free. Neither McGraw-Hill nor its licensors shall be liable to you or anyone else for any inaccuracy, error or omission, regardless of cause, in the work or for any damages resulting therefrom. McGraw-Hill has no responsibility for the content of any information accessed through the work. Under no circumstances shall McGraw-Hill and/or its licensors be liable for any indirect, incidental, special, punitive, consequential or similar damages that result from the use of or inability to use the work, even if any of them has been advised of the possibility of such damages. This limitation of liability shall apply to any claim or cause whatsoever whether such claim or cause arises in contract, tort or otherwise.

Dedication To my Mom, my first teacher.

This page intentionally left blank

Contents Contributors.............................................................................................................................................................................ix Preface .....................................................................................................................................................................................xi Acknowledgments...................................................................................................................................................................xiii

Chapter 1 Fluid, Fat, Blood, Calcium, and Contrast: General Imaging Features ........................................................... 1 Chapter 2 Imaging of the Bowel ...................................................................................................................................... 15 Chapter 3 Imaging of the Liver ...................................................................................................................................... 161 Chapter 4 Imaging of the Gallbladder and Biliary System...........................................................................................255 Chapter 5 Imaging of the Pancreas ...............................................................................................................................319 Chapter 6 Imaging of the Kidney.................................................................................................................................. 387 Chapter 7 Imaging of the Urinary System .................................................................................................................... 451 Chapter 8 Imaging of the Adrenal .................................................................................................................................. 511 Chapter 9 Imaging of the Uterus and Cervix ............................................................................................................... 545 Chapter 10 Imaging of the Ovary and Fallopian Tubes ..................................................................................................591 Chapter 11 Imaging of the Prostate and Seminal Vesicles ............................................................................................631 Chapter 12 Imaging of the Scrotum and Penis .............................................................................................................685 Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts.................................................................................. 773 Chapter 14 Imaging of the Spleen .................................................................................................................................. 819 Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis ................................................................867

Index .................................................................................................................................................................................... 927

vii

This page intentionally left blank

Contributors Skip M. Alderson, MD Staff Radiologist Department of Radiology Radiology Affiliates Imaging Hamilton, New Jersey Stanley Chan, MD Associate Radiology Residency Program Director Director of Body Imaging Mercy Catholic Medical Center Darby, Pennsylvania Lauren Ehrlich, MD Assistant Professor of Radiology Department of Diagnostic Radiology Yale-New Haven Hospital New Haven, Connecticut Narainder K. Gupta, MD, DRM, MSc, FRCR Associate Professor of Clinical Radiology Department of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania Susan Hilton, MD Co-Chief and Modality Chief, CT Section Professor of Clinical Radiology Department of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania Jason N. Itri, MD, PhD Assistant Professor Director of Quality and Safety Department of Radiology, University of Pittsburgh Medical Center Pittsburgh, Pennsylvania Saurabh Jha, MBBS, MRCS Assistant Professor Department of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania

Lisa P. Jones, MD, PhD Assistant Professor of Clinical Radiology Department of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania Sharyn I. Katz, MD, MTR Assistant Professor of Radiology Department of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania Stephanie Carruth Kurita, MD Assistant Professor of Radiology Department of Radiology and Radiological Sciences Vanderbilt University Medical Center Nashville, Tennessee Jill Langer, MD Associate Professor of Radiology Department of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania Wallace T. Miller Jr., MD Associate Professor of Radiology and Medicine Department of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania Edward R. Oliver, MD, PhD Assistant Professor of Radiology Department of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania Nicholas Papanicolaou, MD, FACR Co-Chief, Body CT Section Professor of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania ix

x

Contributors

Parvati Ramchandani, MD Section Chief, Genitourinary Radiology Professor of Radiology and Surgery The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania

Drew A. Torigian, MD, MA Associate Professor of Radiology Department of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania

Mark Alan Rosen, MD, PhD Associate Professor of Radiology Department of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania

Hanna M. Zafar, MD, MHS Assistant Professor of Radiology Department of Radiology The Perelman School of Medicine at the University of Pennsylvania Philadelphia, Pennsylvania

Preface It is the intention of this book to provide a comprehensive review of pattern recognition in abdominal imaging. At some level, diagnostic imaging is primarily an exercise in predicting disease states based on gross pathology. The underlying pathology of disease is constant and so the imaging characteristics of diseases are essentially the same regardless of imaging modality. Thus, it is the intent of this text to unify abdominal x-ray, ultrasonography, CT, MRI, and nuclear radiology interpretations of these underlying disease states, emphasizing when individual modalities have characteristics that can also provide unique insight into disease diagnosis. Furthermore, it is my belief that accurate imaging diagnosis of abdominal diseases can only be performed with the addition of appropriate clinical history. Therefore, the characteristic clinical presentations of the various diseases will be discussed in conjunction with the imaging findings. The book will be divided into

chapters based the organ of interest. Chapters will be organized to stress differentiation of diseases based on imaging patterns. In general this pattern based approach will follow a similar format emphasizing differentiation of disease based on “focal,” “multifocal,” and “diffuse,” abnormalities within the organ in interest followed by a discussion of diseases causing unique imaging features. Each section will follow a similar format. The most salient pathologic, histologic, and clinical features of the disorder will be reviewed, followed by a discussion of imaging characteristics. When possible, differentiating features, both clinical and radiographic, between etiologies producing similar imaging patterns will be emphasized. Many diseases have a variety of imaging features. When this occurs, they will be discussed under each pattern but the most comprehensive review of the subject will occur under the pattern that is most characteristic of the disease.

xi

This page intentionally left blank

Acknowledgments I would like to begin by thanking the many residents and fellows I have had the good fortune to work with. Through their many questions and input, they constantly refine my understanding of human illness and the imaging manifestations of disease. I could not have completed a textbook on Abdominal Imaging of this magnitude without the excellent help of the many chapter authors. They are my coworkers in the Department of Radiology at the University of Pennsylvania’s Perelman School of Medicine and I am very proud that this work represents a collective effort of our department. They worked tirelessly on the manuscript and graciously accepted the editing necessary to maintain a uniform format throughout the text.

There would be no book if I didn’t have clinical colleagues. The many cases illustrating this text often have detailed clinical histories as well as final answers thanks to the feedback we have received from the many clinicians who consult with us. Since the completion of my residency, I am sure that I have learned more from my many friends in clinical medicine than from any other source. I would also like to thank my secretary Tisha Grant for her help in preparing the manuscript. Lastly, I would like to thank Cristina for her patience, understanding, and support.

xiii

This page intentionally left blank

CHAPTER

1

Fluid, Fat, Blood, Calcium, and Contrast: General Imaging Features Wallace T. Miller Jr., MD Sharyn I. Katz, MD, MTR

I. IMAGING FEATURES OF FLUID II. IMAGING FEATURES OF FAT AND OTHER LIPIDS III. IMAGING CHARACTERISTICS OF HEMORRHAGE

Certain abnormalities can be seen in different organs but have a common imaging feature regardless of the organ involved. Before discussing abnormalities of specific organs, we will note some of these common features.

IMAGING FEATURES OF FLUID Simple serous fluid has characteristic features on computed tomography (CT), ultrasonography (US), and magnetic resonance imaging (MRI). This fluid can exist freely in various potential anatomical spaces, for example, ascites, pleural effusion, and pericardial effusion, or can be loculated in cysts within solid organs. On CT images, simple physiologic fluid has an attenuation of 0 to 20 Hounsfield units (HUs). This measurement can be made during evaluation of the computer data at the imaging workstation. A visual estimate can also be made by comparing the abnormality to known structures. Simple fluid will usually have an attenuation between that of skeletal muscle and fat (see Figure 1-1). When the accumulated fluid is not a simple transudate of blood but rather becomes “complex,” admixed with proteinaceous material such as that derived from infection or hemorrhage, the attenuation will increase to greater than 20 HU (see Figure 1-2). One potential caveat that must be considered is that volume averaging within a reconstructed slice can confound Hounsfield unit density measurements and so it should be confirmed that the measurement of the tissue in question be made in a slice that has tissue in the slice above and below it so that volume averaging with adjacent distinct tissues does not occur. Simple fluid in US examinations is uniformly anechoic because there are no interfaces to reflect the sound, and therefore it appears uniformly black (see Figure 1-1B). Fluid-containing structures will also demonstrate increased through transmission of sound, which appears as a band of bright echoes behind the fluid structure (see Figure 1-1B). Simple fluid has no structures to

IV. IMAGING CHARACTERISTICS OF CALCIFICATION V. CONTRAST ENHANCEMENT

reflect sound, and therefore more sound passes through to the tissues deep to a fluid-containing structure. Therefore, compared to an ultrasound wave that passes through solid tissue, there is more sound to reflect back when it strikes an interface in the tissues deep to the fluid-containing structure. The US computer records this increased sound as brighter lines. So, the tissues deep to the fluid-containing structure appear brighter than adjacent tissues that are not deep to the fluid-containing structure. This artifact is known as “increased through transmission” of sound and is characteristic of any structure that contains fluid: cysts, gallbladder, urinary bladder, and pleural effusions. Complex fluid, which contains internal debris, can be observed in a number of scenarios such as hemorrhagic collections, abscesses, or central necrosis within a malignancy. US of these processes will often demonstrate multiple internal bright floating specks or “echoes.” These echoes represent the reflection of sound from floating particulate debris within the fluid. If plentiful in debris or clot and relatively low in water content, complex fluid can actually

Imaging Notes 1-1. Imaging Characteristics of Simple Fluid CT

Attenuation between 0 and 20 HU Nonenhancing

US

Anechoic Increased through transmission

MRI

High signal T2W images Intermediate signal T1W images Nonenhancing 1

2

Diagnostic Abdominal Imaging

Imaging Notes 1-2. Imaging Features of Nonsimple Fluid Internal septations Internal debris • Echogenic foci and increased through transmission on US • Attenuation >20 HU and nonenhancing on CT • Decreased T2 signal on MRI Fluid-debris levels

have a similar appearance to solid tissues on US. In these cases, the inherently higher than normal fluid content usually preserves the enhanced through transmission typical of fluid on US, a property characteristic of fluid structures (see Figure 1-3). In some cases, further imaging with MRI or CT will be necessary to confirm a diagnosis of a fluid-containing structure.

A

B

C

D

In some instances, there will be septations within the fluid. These will appear as fine lines of increased echogenicity within the fluid-filled structure. US is often the most sensitive imaging test for the presence of septations. However, MRI can have an advantage in evaluating fluidfilled structures where US is limited because of large body habitus. For example, evaluation of complex renal cysts in an obese patient can be impaired by poor sound transmission through excess adipose tissue. MRI, on the other hand, is not limited in the setting of obesity and will accurately characterize the internal architecture of the cystic structure. CT is the least sensitive cross-sectional modality for detection of internal septations. Occasionally, CT can detect septations when there is enhancement of the septa by intravenous (IV) contrast, thereby increasing tissue contrast. Otherwise, septations will typically remain unrecognized by CT studies (see Figure 1-4). Fluid on an MRI examination appears uniformly high intensity (bright white) on T2-weighted images and appears a uniform intermediate intensity (gray) on T1-weighted images (see Figures 1-1 and 1-4). The bright signal intensity typical of fluid on T2-weighted images is a result of

Figure 1-1 Simple Cyst of the Kidney This 72-year-old woman had undergone a left nephrectomy for renal cell carcinoma. A. CT image through the right kidney demonstrates a small, round, uniform, nonenhancing lesion in the upper pole of the kidney. This region measured 13 HU. B. US of the same lesion shows it to be round and anechoic. The tissue below the lesion appears brighter than the adjacent tissues, representing increased through transmission of sound. C. MRI images through the lesion show it to be uniformly high signal on T2-weighted images. D. Low to intermediate signal on T1-weighted sequences. These findings are all characteristic of a simple cyst.

Chapter 1 Fluid, Fat, Blood, Calcium, and Contrast: General Imaging Features 3

A

B

Figure 1-2 Ascites versus Hemoperitoneum A. Contrast-enhanced coronal CT image from a 58-year-old man with cirrhosis shows a large amount of low attenuation acites (arrowheads) surrounding the liver and small bowel mesentery and filling the pelvis. Note how the attenuation of the fluid is less than the attenuation of muscle and liver but is greater than fat. B. Unenhanced coronal CT image from a 51-year-old woman with abdominal pain status post renal transplantation shows similar low attenuation fluid (arrowheads); however, there is also an amorphous region of higher attenuation in the pelvis and lower abdomen. This represents acute clotted blood within the peritoneum. Notice how it is slightly higher attenuation than the attenuation of unenhanced muscle.

A

B

Figure 1-3 Complicated Cyst This 68-year-old woman was being evaluated for voiding dysfunction. A. US of the lower pole of the right kidney reveals a round hypoechoic structure (large arrow) in the lower pole with faint internal echoes. These internal echoes could indicate either a solid lesion or a cyst with internal debris. However, in the tissues deep to the lesion, there is a band of increased echogenicity (small arrows), an artifact called increased through transmission that indicates

C that this lesion is a cyst. Unenhanced (B) and enhanced (C) CT images through the cyst show the cyst (large arrows) to be faintly higher attenuation than some adjacent ascites (small arrows). The cyst measured 28 HU on the unenhanced image and 25 HU on the enhanced image. This indicates the lesion is nonenhancing and is consistent with a “hyperdense” cyst. Note also the faint rim calcification in the cyst on CT seen as thin foci of high attenuation in the medial rim of the cyst.

4

Diagnostic Abdominal Imaging

A

B

Figure 1-4 Complicated Cysts with Internal Septations in 2 Patients A. This 26-year-old woman had a palpable adnexal lesion. Transvaginal ultrasound shows a large anechoic structure with several echogenic septations projecting from the left ovary. These are features of a cyst with internal septations. Surgical resection demonstrated a mucinous cystadenocarcinoma.

B. This 49-year-old woman was being evaluated for carcinoid tumor. T2-weighted MRI sequence shows a multiloculated hyperintense lesion (large arrow) in the liver. High intensity on T2-weighted sequences will usually indicate the presence of fluid. The internal septations indicate that this represents a complex cyst. Note the high signal fluid in the gallbladder (arrowhead) and cerebrospinal fluid (small arrow).

the slower loss of phase coherence of the proton nuclei in the transverse plane immediately following the imaging radiofrequency pulse. Interactions between adjacent proton nuclei results in loss of spin synchrony and causes measured MR signal in the horizontal plane. This process requires dissipation of energy, which is not efficient within water, thus leading to a prolonged or “brighter” T2 signal. As with other imaging modalities, an estimate of the composition of the structure of interest can be inferred from comparison to normal structures of known composition. For example, when considering the fluid nature of a cyst, the T2-weighted signal may be compared to the spinal fluid within the spinal canal, which is mostly composed of water. If the fluid being studied is “complex,” T2-weighted images may reveal low T2-weighted signal intensity debris or septae within otherwise high T2-weighted signal intensity water (see Figure 1-4). Cysts containing hemorrhage can have a variety of imaging appearances based on the MR signal characteristics of various phases of hemoglobin. However, in general, hemorrhagic cysts will be hyperdense on T1-weighted imaging. This phenomenon is discussed later in the section Imaging Characteristics of Hemorrhage. Simple fluid can collect as cysts in many organs. Besides the various characteristics of fluid in different modalities, there are several common characteristics of simple cysts: (1) smooth, thin, well-defined walls; (2) round or oval shape; and (3) internal characteristics of simple fluid within the cyst: attenuation between 0 and 20 HU on CT,

anechoic, increased through transmission on US, and high T2-weighted signal on MRI (see Figure 1-1). Furthermore, if IV contrast is administered, there is no enhancement of the structure.

IMAGING FEATURES OF FAT AND OTHER LIPIDS Like fluid, fat has a unique appearance on cross-sectional imaging studies. Fat is less dense than other soft tissues, and because density is the major determinant of x-ray attenuation, pure fat will have a lower attenuation than other soft tissues on unenhanced CT exams. Moreover, because fat is also relatively hypovascular compared with other soft tissues, it will remain low attenuation on

Imaging Notes 1-3. Imaging Features of Macroscopic Fat CT

Attenuation between –40 and –120 HU

US

Hyperechoic

MRI

High signal T1W images Intermediate-high signal T2W images Etching artifact on opposed-phase T1W images

Chapter 1 Fluid, Fat, Blood, Calcium, and Contrast: General Imaging Features 5

A

B

C

Figure 1-5 Imaging Characteristics of Fat A. This ultrasonographic image shows a brightly echogenic nodule (cursors) in the lower pole of the kidney. Note how the echogenicity of the nodule is similar to the subcutaneous fat (arrow). This represents a small angiomyolipoma, a fatty tumor of the kidney. A metastasis or second renal cell carcinoma would usually be less echogenic. B. This 63-year-old woman had a lung mass discovered on chest x-ray. This unenhanced CT image at the level of the adrenals shows a normal left adrenal and a 2.5 cm right adrenal mass (arrow). The mass has a thin rim of tissue resembling muscle, but the majority of the mass is low

attenuation, resembling retroperitoneal fat. Mature fat within an adrenal mass is diagnostic of an adrenal myelolipoma, a benign adrenal tumor. C. This 75-year-old woman had back pain and a history of breast cancer. Sagittal T1-weighted MRI image through the thoracic spine shows the normal vertebral bodies to have a relatively high signal because of the intramedullary fat. In the central thoracic spine are 2 spherical areas of decreased signal representing bone metastasis. The breast cancer is very conspicuous on these T1-weighted sequences because of the relative differences in intensity of the fatty marrow and the nonfatty metastasis.

contrast-enhanced CT images. It has been shown that mature fat has an attenuation between –40 and –120 HU. The attenuation can be measured directly at the PACS workstation, but most often it is easiest to compare the attenuation of the structure in question with the attenuation of subcutaneous fat. If the attenuations are similar, then the structure contains macroscopic fat (see Figure 1-5). On US examinations, fat appears as regions of densely increased echogenicity. The numerous small fibrous septae that surround small fat globules act as innumerable specular reflectors, resulting in fine even echoes throughout the fat (see Figure 1-5). The appearance of densely increased echogenicity is not unique to fat and can be seen with structures that contain psammomatous (numerous fine punctate) calcifications or with structures that are composed of a complex network of interfaces, such as hepatic hemangiomas (see Figure 1-6). On MRI, the majority of tissue contrast is derived from the intrinsic T1 time or spin-lattice relaxation time of a tissue. This T1 time refers to the time required for the radiofrequency energy of magnetization from the imaging pulse to dissipate through the infrastructure of the tissue, allowing the proton nuclear spin to return—or “relax”—to its original orientation within the magnetic field. This T1 time is a result of proton–proton nuclear interactions of different molecules within the tissue being imaged and generates relatively unique T1 times

for different tissue compositions. By exploiting the relative tissue properties of T1 imaging, not only can the presence of fat be detected but also macroscopic fat can be differentiated from microscopic fat (ie, microscopic lipid admixed with fluid). These additional fatty characteristics can provide further information as to the nature of the tissue being studied. Macroscopic fat typically appears uniformly bright on T1-weighted sequences and intermediate to high signal intensity on T2-weighted sequences although typically not as bright as fluid. The bright T1-weighted signal of fat is a result of the fast dissipation of radiofrequency energy of magnetization within the tissue structure, resulting in a short T1 time leading to bright T1-weighted signal because the protons have relaxed, are synchronized, and are now generating signal in the plane of imaging signal acquisition (see Figure 1-5). As with CT, the easiest method of identifying fat in a structure is to compare the intensity of the structure in question with the intensity of subcutaneous fat. If the signal pattern on all sequences parallels that of subcutaneous fat, then the structure contains macroscopic fat. Note that it is important to compare all sequences. Some other substances, for example blood, can be very bright on T1-weighted sequences but will not follow subcutaneous fat on all sequences and are generally low in T2-weighted signal, with some exceptions. Macroscopic fat will also produce an artifact on opposedphase images called the etching artifact. With this pulse

6

Diagnostic Abdominal Imaging

Figure 1-6 Hemangioma This 17-year-old woman was being evaluated for abdominal pain. Longitudinal US of the liver shows an echogenic mass (arrow) in the left lobe of the liver. This is a typical appearance of a hepatic hemangioma. The multiple vascular channels of the tumor serve as specular reflectors, resulting in the increased echogenicity of the mass.

sequence, voxels containing both lipid and water molecules will lose signal. At places where macroscopic fat is adjacent to nonfatty soft tissue, voxels will contain both lipid and water molecules, and a thin black line will be

artifactually produced that appears to outline the abdominal organs (see Figure 1-7). Microscopic fat—lipid molecules admixed with water molecules—yields a signal intensity that is generally intermediate on T1- and T2-weighted imaging. However, microscopic lipid can be identified with the use of T1-weighted in-phase and opposed-phase imaging. As mentioned earlier, proton nuclei of lipids dissipate energy more efficiently and hence have a shorter T1 relaxation time than proton nuclei in water. In fact, the T1 time for water is approximately twice that of lipid. Therefore, in a voxel with approximately 50% lipid and 50% water, if the time-to-echo is timed at multiples of the T1 time of lipid, the signals between lipid and water will be opposed, termed opposedphase, and thus cancel out at the odd multiples and be additive, termed in-phase, at even multiples of the lipid T1 time. When compared to in-phase T1-weighted images, the signal generated by a tissue where water and lipid are mixed within a voxel will result in a drop in signal in out-of-phase T1-weighted images, and the drop in signal will be maximal when the mix of molecules within that voxel is close to 50% water and 50% lipid. For this reason, macroscopic fat tissue, which has proportionally much more fat than water in any given voxel, will not exhibit a drop in signal from in- to out-of-phase T1-weighted imaging. This phenomenon is primarily used to identify the presence of cholesterol in adrenal adenomas but occasionally can be used for other purposes such as the detection of microscopic lipid in tumors of hepatic origin and in the differentiation

A

B

Figure 1-7 Etching Artifact on Opposed-Phase Images This 66-year-old woman was being evaluated for an adrenal mass. In-phase (A) and opposed-phase (B) T1-weighted MRI images through the upper abdomen show thin black lines that outline the liver, spleen, stomach, psoas muscles, and erector spinae muscles on the opposed-phase images but are not present on the inphase images. This is an artifact of the way the image is created.

There is fat surrounding the surface of each of these organs and therefore the voxels present at the margins of the organs contain both fat and water protons. In the in-phase image, the signal from these protons is added; however, in the opposed-phase image, the fat signal and the water signal is subtracted, resulting in the thin black lines. Thus, this technique can be used to identify the presence of microscopic or macroscopic lipid within tissues.

Chapter 1 Fluid, Fat, Blood, Calcium, and Contrast: General Imaging Features 7

A

B

Figure 1-8 Subtraction Images in Adrenal Adenomas This 73-year-old man had lung cancer. In-phase (A) T1-weighted and out-of-phase (B) T1-weighed images show a 2.5-cm nodule in the left adrenal gland that loses signal in

the out-of-phase images. This is indicative of the presence of microscopic lipid within the lesion and is diagnostic of an adrenal adenoma.

of thymic hyperplasia, which normally contains diffuse microscopic lipid, from thymomas (see Figure 1-8).

As a clot ages, it begins to degrade and becomes progressively less dense, and so the attenuation falls until it approaches or equals simple fluid attenuation (0-20 HU). In an intermediate phase, hemorrhage can equal the attenuation of unenhanced tissues and as a consequence can be difficult to identify. In many cases, an old liquefied hematoma will contain simple fluid in the nondependent portion of the hematoma and debris will settle by gravity into the dependent portion of the hematoma, generating a “fluiddebris level.” The presence of debris within a fluid collection will often indicate the presence of prior hemorrhage but can also be seen in abscesses and some neoplasms. On US examinations, acute hematoma will appear as an echogenic mass within the tissues. It is believed that the multiple fibrin strands act as specular reflectors and lead to many small echogenic foci. This can be confused with other echogenic masses such as fat-containing lesions or hemangiomas. As the hematoma liquefies, debris from within the clot will cause internal echoes within the fluid collection. This will appear as an echogenic cyst with increased through transmission. Similar to CT, the debris will often settle dependently within the collection, leading to a fluiddebris level. The complex cyst of a liquefying hematoma cannot be distinguished from other debris-filled cysts that can be seen with abscesses and some neoplasms. The imaging properties of blood products on MRI is dependent on location of the hemorrhage within the body, and hence surrounding tissues, and timing of imaging relative to time of hemorrhage. In the brain, the MRI appearance of a hemorrhage has been extensively studied and can yield very specific information on the age of the hemorrhagic products. In the hyperacute phase of intracranial

IMAGING CHARACTERISTICS OF HEMORRHAGE A hematoma undergoes various stages of evolution. Initially, extravasated blood forms a dense coagulated mass. With time, the hematoma undergoes degradation via the action of extracellular proteases and macrophage phagocytosis leading to liquification of the clot. Coincident with this process, hemoglobin undergoes a transition from oxygenated (oxyhemoglobin) to deoxygenated (deoxyhemoglobin) to methylated hemoglobin (methemoglobin). Various characteristics of this transition process can be identified by imaging exams and can suggest the presence of extravasated blood. Computed tomographic examinations will often show acute hemorrhage as a high attenuation region relative to unenhanced tissues (see Figure 1-2). This high attenuation is a direct result of increased density of acute iron-containing clot relative to soft tissues. Although initially characterized for evaluating subarachnoid, subdural, epidural, intraparenchymal, and intraventricular hemorrhage in the brain, this phenomenon is now routinely used to identify acute hemorrhage throughout the body. It is important to note that extravasated blood is only high attenuation relative to surrounding vascularized tissues when those tissues are unenhanced with IV contrast. Use of IV contrast increases the CT attenuation of vascularized soft tissues such that acute hemorrhage will be relatively lower in attenuation. However, acute hemorrhage will always remain higher in CT attenuation than simple fluid.

8 Diagnostic Abdominal Imaging

hemorrhage (less than 12 hours from the hemorrhagic event), the predominant form of hemoglobin within the blood clot is oxyhemoglobin, which is isointense to brain parenchyma on T1-weighted signal and high intensity on T2-weighted signal. Over the initial 12 hours, the predominant blood product is oxyhemoglobin, but during the acute stage (12-72 hours posthemorrhage), the hemoglobin within the hematoma becomes increasingly deoxygenated and results in deoxyhemoglobin, which is a paramagnetic

substance and leads to marked shortening (darkening) of the T2-weighted MR signal within the hematoma. During the early subacute phase of hemorrhage (3-7 d posthemorrhage), failure of extravasated red blood cells to maintain an intracellular functional reductase enzyme system results in oxidation of deoxyhemoglobin to methemoglobin, which is a strongly paramagnetic substance resulting in marked shortening of T1-weighted signal (signal brightening), the most dominant effect, and further shortening of T2-weighted signal (signal darkening) (see Figure 1-9). This is followed by release of methemoglobin into the extracellular environment, as the red blood cells subsequently degenerate and lyse, resulting in further T1-weighted signal shortening (brightening). In the chronic phase, which begins in the weeks following the hemorrhagic event, ongoing macrophage activity along the periphery of the evolving chronic intracranial hemorrhage will result in steady conversion of methemoglobin to hemosiderin, which results in marked T2 shortening (darkening) because of the strongly ferromagnetic properties of hemosiderin. This hemosiderin ring of markedly low T2-weighted signal around the intracranial hematoma increases in thickness with time, whereas the central scar contracts and ultimately collapses, usually resulting in a band-like hypointense ferromagnetic scar that sometimes persists indefinitely. Another location in the body where the MR signal characteristic of hemorrhage is studied, temporally, is the aorta. Similar to the brain, in the setting of an acute intramural

A

B

Figure 1-9 Differentiation Between Fluid and Hemorrhage on MRI Sagittal T1-weighted (A) and T2-weighted (B) MRI images through the pelvis reveal a dilated tubular structure posterior to the uterine fundus in keeping with the Fallopian tube (thick white arrow). There is also a small amount of fluid (thin white arrow) in the endometrial canal. The fluid in these structures

is high signal on T1-weighted sequence (A) and intermediate signal on the T2-weighted sequence (B) typical of early subacute hemorrhage related to menses. By contrast, the predominately water-filled bladder (star) exhibits the typical MRI appearance of simple fluid, that is, low in signal intensity on T1-weighted images and high in signal intensity on T2-weighted images.

Imaging Notes 1-4. MRI Characteristics of Evolving Brain Hemorrhage Phase of Hemorrhage

Hemoglobin State

T1W

T2W

Hyperacute (94%) in the detection of liver metastases.182 There are no published series of the gastric findings of ZE syndrome as detected by CT or MRI. However, ZE syndrome would be expected to demonstrate diffuse gastric wall thickening and hyperemia by CT and MRI examinations (see Figure 2-24).

disease is histologically characterized by thickening and hyperplasia of the gastric mucosa with cystic dilation and elongation of the gastric mucous glands and deepening of the foveolar pits.191 Ménétrier disease also often leads to a protein loosing enteropathy that results in hypoalbuminemia and soft-tissue edema. There are 2 forms of disease, 1 seen in childhood and the second in adulthood. The adult form of the disease is progressive, whereas the childhood form has been associated with cytomegalovirus (CMV) infection and resolves spontaneously.190,192 The mean age of presentation is approximately 55 years and patients typically present with abdominal pain, nausea, vomiting, anemia (due to gastric blood loss), hypochlorhydria, and peripheral edema.190 There is an increased incidence of thrombotic events throughout the body, possibly due to a decrease in intravascular volume. Ménétrier disease is believed to carry an increased risk of gastric malignancy; however, the magnitude is not known. Upper GI examinations will demonstrate massively thickened, lobulated rugal folds, which are most pronounced in the gastric fundus and body along the greater curvature.193 Although the antrum was initially thought to

Ménétrier disease Ménétrier disease, also called hyperplastic, hypersecretory gastropathy, is a rare disorder of the stomach resulting in dramatic thickening of the gastric rugae such that they resemble the gyri of the brain. It is now thought to be due to overexpression of transforming growth factor-α (TGF-α) that results in the selective expansion of surface mucous cells in the body and fundus of the stomach.190 The expansion of these surface mucous cells is at the expense of parietal and chief cells that, when reduced, result in decreased output or absence of gastric acid (achlorhydria). Ménétrier

44

Diagnostic Abdominal Imaging

A

B

C

D

Figure 2-22 Scirrhous Metastasis to the Stomach This 92-year-old woman with breast carcinoma complained of dysphagia. A. Single-contrast UGI examination shows diffuse narrowing of the body of the stomach that could not be distended despite repeated swallows of barium. This appearance is indicative of rigidity of the stomach and suggests the presence of scirrhous metastasis to the stomach.

Coronal oblique reconstruction (B) and axial contrastenhanced CT images (C and D) show diffuse thickening of the gastric wall (arrowheads). On CT examinations, it is difficult to determine whether this is due to inadequate distension or wall thickening. Endoscopic biopsy confirmed a diagnosis of breast cancer metastasis.

be spared, some studies have shown antral fold thickening in Ménétrier disease in nearly 50% of patients.194 Rarely, the disease results in focal fold thickening that can be confused with lymphoma or infiltrating forms of carcinoma. Similar to ZE syndrome, there is an excess of gastric fluid that can cause poor coating of the mucosa by high-density barium. Cross-sectional imaging will demonstrate massive thickening of the gastric wall, which enhances avidly following contrast administration.171,195 In some cases, rugal fold hypertrophy can appear as a mass lesion that can mimic the appearance of a gastric lymphoma or carcinoma.196

Eosinophilic gastroenteritis Eosinophilic gastroenteritis is a rare, poorly defined condition characterized by eosinophilic infiltration of the wall of the GI tract that can variably involve the mucosal, muscularis, or serosal portions of the intestines.197 It most often involves the stomach, followed by the small bowel and colon.198 The pathogenesis of this disease is not certain, but because of the association of eosinophilic gastroenteritis with a variety of allergic conditions, including seasonal allergies, food sensitivity, eczema, and asthma, it is believed that hypersensitivity likely plays a role.199 Some

Chapter 2 Imaging of the Bowel 45

A

B

C Figure 2-23 Severe Acute Gastritis This 19-year-old woman complained of several weeks of severe epigastric pain, nausea, and vomiting. A and B. A contrastenhanced CT was ordered and demonstrated severe, lowattenuation thickening (white arrows) of the antral wall. Note the mucosal enhancement on the nondependent wall. This low attenuation thickening usually indicates wall edema. C. Upper

GI examination confirms the presence of multiple thickened antral folds and also shows punctate collections of barium (black arrows), indicating the presence of small ulcerations on the surface of the thickened folds. Note the thickened duodenal folds suggesting duodenitis. Gastric biopsy demonstrated acute and chronic gastritis and duodenitis.

46

Diagnostic Abdominal Imaging

A

B

C

D

E

F

G

H

Figure 2-24 Zollinger-Ellison Syndrome in 3 Patients A and B. This 37-year-old man had severe abdominal pain. AP (A) and lateral (B) projections from a double-contrast barium enema show increased number and thickness of rugal folds in the body and fundus of the stomach. Gastrin levels were elevated and a diagnosis of Zollinger-Ellison syndrome was made. C-E. This 49-year-old woman complained of abdominal pain. Enhanced CT shows apparent thickening of the wall of the

stomach (arrowheads). Careful observation suggests that this is due to apposition of multiple enlarged rugal folds. When the stomach is collapsed, the wall can appear thickened and so a normal variation cannot be excluded as a cause for this apparent thickening. F-H. Axial T2-weighted MRI sequences in a third patient show similar findings but more easily demonstrate the increased number of rugal folds. Both of these patients were diagnosed with Zollinger-Ellison syndrome.

Chapter 2 Imaging of the Bowel 47 studies suggest that degranulation of eosinophils in the tissues of the GI tract leads to damage and subsequent symptoms. Patients typically present with nonspecific GI symptoms, including abdominal pain, nausea, vomiting, diarrhea, weight loss, and/or abdominal distension. Patients will often have peripheral eosinophilia and elevated IgA. The site of involvement can influence clinical symptoms. Mucosal involvement will characteristically present with features of malabsorption and protein-losing enteropathy. Muscular involvement is more likely to present with mass lesions that can occasionally lead to obstructive symptomatology, whereas serosal involvement is associated with ascites. Upper GI examinations will typically demonstrate thickened mucosal folds, mucosal nodularity, and luminal narrowing, which predominantly involves the distal half of the stomach.200,201 In approximately 50% of patients, there is associated thickening and nodularity of small-bowel folds. Occasionally, eosinophilic gastroenteritis can present with luminal narrowing and obstruction of either the stomach or small bowel.202 There are no studies that have directly evaluated the cross-sectional imaging manifestations of eosinophilic gastroenteritis. However, there are scattered case reports of eosinophilic gastroenteritis that include descriptions of CT findings. In many cases, CT failed to detect the presence of disease.202 However, in 1 case, focal wall thickening of the stomach was noted on enhanced CT scans.203

Crohn disease Crohn disease is a common autoimmune disorder primarily involving the small bowel and colon. Gastric involvement is uncommon. Upper GI examinations will typically demonstrate aphthoid ulcers. In more advanced disease, 1 or more ulcers can be found, associated with diffusely thickened gastric folds.162 The cross-sectional imaging features of gastric involvement by Crohn disease have not been systematically evaluated. Similar to other causes of gastritis, however, it is expected that Crohn involvement of the stomach is likely to cause thickening of the gastric wall. Given the predilection of Crohn disease to involve the gastric body and antrum, this fold thickening will predominantly involve the distal stomach. Crohn disease has been more completely covered later in this chapter under the heading: Long-Segment Wall Thickening of the Small Bowel and Colon.

FOCAL DISORDERS OF THE SMALL BOWEL Focal Polyps and Masses of the Small Bowel The majority of focal mass lesions of the small bowel represent either benign or malignant neoplasms. The most common neoplasms of the small bowel are adenocarcinoma and carcinoid tumor. Other neoplasms include lymphoma, GISTs, lipomas, and hamartomatous polyps (Table 2-3).

Table 2-3. Polyps and Masses of the Small Bowel 1. Epithelial lesions a. Adenoma/adenocarcinoma b. Carcinoid tumor 2. Submucosal lesions a. Lymphoma b. Gastrointestinal stromal tumor (GIST) c. Lipoma d. Hamartomatous polyps i. Peutz-Jeghers ii. Juvenile e. Kaposi sarcoma f. Other mesenchymal tumors i. Leiomyoma/leiomyosarcoma ii. Neurofibroma iii. Schwannoma

Adenoma and adenocarcinoma Adenocarcinoma is the most common tumor of the small bowel but is still only half as common as colonic adenocarcinoma.204-206 Among small-bowel adenocarcinomas, 50% are found in the duodenum, mainly near the ampulla of Vater, or in the proximal jejunum within 30 cm of the ligament of Treitz.204,207 Risk factors for the development of adenocarcinoma of the small bowel include a history of Crohn disease, celiac sprue, Peutz-Jeghers syndrome, Lynch syndrome II, congenital bowel duplication, ileostomy, and duodenal or jejunal bypass surgery.205 Patient symptoms vary with tumor size, location, blood supply, and the presence of ulceration.208 Smaller lesions are typically asymptomatic. Annular lesions eventually produce obstruction209 and, rarely, polypoid lesions can result in bowel obstruction due to intussusception. Duodenal adenocarcinomas can also present with obstructive jaundice, because of biliary obstruction of the tumor, or pancreatitis, because of pancreatic duct obstruction of the tumor.207 Unlike adenocarcinoma of the colon and rectum, the size of small-bowel adenocarcinoma is not correlated with its tendency to spread. Small tumors can manifest with distant metastases.210 Unlike the remainder of the small bowel, the duodenum is retroperitoneal in location, and direct spread into the retroperitoneum is an important additional mode of spread that often leads to unresectability. Approximately half of small-bowel adenocarcinomas are unresectable, and the 5-year survival of these tumors is approximately 20%.208 The imaging appearance of a small-bowel adenocarcinoma varies. Typically, it will involve a short segment of bowel and will appear as one of the following: (1) an annular region of luminal narrowing with abrupt “shelf-like” margins or “overhanging edges,” (2) as a discrete tumor mass, or (3) as an ulcerative lesion (see Figure 2-25).205 Distal small-bowel adenocarcinomas are usually annular,

48

Diagnostic Abdominal Imaging

A

B

C

D

E

F

Figure 2-25 Duodenal Obstruction due to Adenocarcinoma This 75-year-old woman complained of melena and vomiting. A. Examination of the upper abdomen shows 2 air-fluid levels (arrows), 1 underneath the left hemidiaphragm and the other in the right upper quadrant. This is the characteristic appearance of the double bubble sign and indicates a duodenal obstruction. B. Barium swallow confirms the presence of focal narrowing of the descending duodenum with a shelflike shoulder between the narrowed and normal segments of the bowel. This appearance is typical of an annular carcinoma. C-F. CT images of the upper abdomen show a distended stomach and proximal duodenum with a normal-caliber distal duodenum. Images D and E show a concentric mass involving the descending biopsy that was diagnostic of adenocarcinoma of the duodenum.

Chapter 2 Imaging of the Bowel 49 constricting, and partially ulcerated, whereas duodenal adenocarcinomas are typically polypoid with little ulceration or infiltration.208,211 On barium examination, these tumors will be seen as an annular lesion with shelflike overhanging margins or a polypoid mass within the lumen of the bowel. Because barium coats the mucosa of these lesions, ulceration of these tumors is more readily demonstrated on barium compared to CT. On CT, these tumors will present as heterogeneously attenuating soft-tissue masses with moderate levels of enhancement.16 They can appear as small polypoid lesions, larger eccentric masses, or annular masses of the wall of the small bowel. Although a large, aggressive, ulcerated adenocarcinoma can be mistaken for lymphoma, the presence of bulky metastatic lymphadenopathy favors lymphoma rather than adenocarcinoma.212 Metastatic lymphadenopathy typically occurs in the root of the mesentery and in the pyloric, hepatic, peripancreatic, cecal, and ileocolic regions.213 CT can also detect hematogenous metastasis to the liver and other abdominal organs. Both barium examinations and CT can also demonstrate complications of the tumor such as obstruction and intussusception.214-216

Carcinoid tumor Gastrointestinal carcinoid tumors are relatively uncommon, although their incidence has increased over the past 30 years. Approximately two-thirds of carcinoid tumors in the body are located in the GI tract, with most of the remaining one-third located in the tracheobronchial system.217 Within the GI tract, these tumors are located most frequently in the small intestine (41.8%), followed by the rectum (27.4%), appendix (24.1%), and stomach (8.7%).85,217 Carcinoid tumors are a diverse group of typically lowgrade, slow-growing malignancies that arise from specialized endocrine cells populating the GI mucosa and submucosa.85,213 The wide variety of hormones produced by these tumors can produce multiple syndromes, including carcinoid syndrome and ZE syndrome. Alternatively, carcinoid tumors can be a manifestation of the inherited syndromes: multiple endocrine neoplasia (MEN) type 1 and neurofibromatosis (NF) type 1.85 Approximately 90% of carcinoid tumors arise in the ileum.213 Primary carcinoid tumors of the small intestine are typically small, rarely exceeding 3.5 cm.85 At the time of diagnosis, almost two-thirds of patients with small intestinal carcinoid tumors have disease that has spread beyond the intestine to regional lymph nodes or the liver.217 Unlike the primary tumor, metastasis to the lymph nodes, mesentery, and liver often attain large sizes, overshadowing the primary tumor. Involvement of the subserosa and adjacent mesentery stimulates a desmoplastic reaction that results in kinking, retraction, and angulation of the bowel, a finding that is characteristic of carcinoid and can suggest the correct diagnosis.85

Complications of carcinoid tumors include hormonal syndromes (listed above), intestinal ischemia, and intestinal obstruction. The classic carcinoid syndrome consists of cutaneous flushing, sweating, bronchospasm, colicky abdominal pain, diarrhea, and right-sided cardiac valvular fibrosis.85 Approximately 10% of patients who do develop carcinoid syndrome will have a primary tumor of the small bowel. Carcinoid syndrome develops when vasoactive substances produced by the tumor, including serotonin, gastrin, somatostatin, cholecystokinin, and secretin enter systemic circulation without undergoing metabolic degeneration.85 These substances are cleared from the circulation by the liver and, therefore, only 10% of patients with bowel carcinoid tumors develop carcinoid syndrome.218 Unsurprisingly, most patients with a GI carcinoid who develop carcinoid syndrome have extensive hepatic metastasis such that the vasoactive secretions of the metastasis do not undergo degradation by liver enzymes and instead enter the systemic circulation. Although duodenal carcinoids are rare, approximately 62% will secrete gastrin, onethird of which have sufficient hormone production to cause ZE syndrome.82,219 Secretion of serotonin and other vasoactives substances can produce thickening, multifocal stenoses, or occlusion of mesenteric arteries and veins with resultant local, regional, or diffuse intestinal ischemia.220 Both the primary tumor and associated desmoplastic reaction can result in partial or complete SBO. The imaging appearance of jejunal and ileal carcinoids varies by tumor size, extent of mesenteric involvement, and presence or absence of lymph node or liver metastases.85 Enteroclysis is the most sensitive test for the detection of small primary lesions, although these lesions may be seen on routine small-bowel follow-through (SBFT). When visualized, these tumors typically present as small (3 cm in the ileum.531,533 This dilation is typically most pronounced in the mid- and distal jejunum and can be seen with superimposed atonic and featureless

104

Diagnostic Abdominal Imaging

A

B

C

D

E

F

Figure 2-69 Lymphoma Complicating Celiac Disease This 57-year-old woman had chronic diarrhea. Small-bowel follow-through (A) and concurrent abdominal CT (B and C) demonstrates a decreased number of small-bowel folds in the proximal small bowel (arrow) and an increased number of small-bowel folds in the distal small bowel (arrowhead), features virtually diagnostic of celiac disease. Two years later, the patient developed weight loss. Spot film from a small-bowel follow-through (D) and concurrent abdominal CT (E and F) now show thickening of the wall and folds of the distal small bowel (arrows). This is a worrisome feature and can indicate development of a lymphoma. Biopsy was diagnostic of a T-cell lymphoma.

Chapter 2 Imaging of the Bowel 105

A

B

C

D

Figure 2-70 Celiac Sprue This 45-year-old woman had chronic abdominal pain and bloating. Enhanced CT shows thickened proximal small-bowel wall (arrowheads) with an increased number of folds in the distal

small bowel in image D. These are features that strongly suggest the diagnosis of celiac disease. There is also mild para-aortic lymphadenopathy (arrow) in A.

appearance of the small-bowel segments.24,533,539 In addition, CT examinations of patients with celiac disease have demonstrated small-bowel luminal dilation in 24% to 66% of cases.532,540,541 Excess of fluid and/or air in the bowel are subjective findings that have been reported as a manifestation of celiac disease.532,540,541 Excess fluid, characterized by dilution and/or flocculation of the intraluminal barium, has been observed in 56% to 64% of cases on CT examinations.540,541 Mesenteric adenopathy, defined as nodes measuring greater than 10 mm in short axis, has been variably demonstrated in 11% to 43% of patients with celiac disease on abdominal CT examinations (see Figure 2-70).531,532,539-541 Rarely, celiac disease can be associated with an uncommon lymph node pathology called “the cavitating mesenteric lymph node syndrome.” The cause of this rare complication of celiac disease is not known but has been associated with refractory celiac disease. It is characterized by moderately enlarged, centrally cystic mesenteric lymph nodes. (This unusual phenomenon is discussed in greater detail in Chapter 13, Imaging of the Lymph Nodes and Lymphatic Ducts.)

Other reported extraintestinal manifestations of celiac disease include the following: (1) increased mesenteric vessel diameter seen in up to 25% to 83% of patients on CT;532,540,541 (2) splenic atrophy present in 30% to 50% of adults with celiac disease;531,532 and (3) mesenteric panniculitis characterized by increased attenuation of  the small-bowel mesentery fat seen in up to 6% of the general population and 11% (5/44) of patients with celiac disease in 1 study.174 Patients with untreated celiac disease have an increased incidence of small-bowel lymphoma (see Figure 2-69).

Other causes of bowel inflammation There are a variety of unusual causes of bowel wall inflammation that can lead to bowel wall edema. These include some chemotherapeutic medications and graft versus host disease. These 2 disorders result in toxic- or immunologic-mediated inflammation of the bowel mucosa, leading to mucosal hyperemia and submucosal edema. Imaging examinations will demonstrate similar findings of bowel wall inflammation, including bowel wall thickening, mucosal hyperenhancement, submucosal

106

Diagnostic Abdominal Imaging

A

B

C

D

Figure 2-71 Other Causes of Diffuse Bowel Wall Inflammation A and B. This 80-year-old man had 3 days of abdominal pain and diarrhea after 5-fluorouracil therapy for a pancreatic carcinoma. Axial CT images show mild colonic (arrowheads) and small-bowel (arrow) wall thickening. No cause could be identified, and this was assumed to be due to 5FU bowel toxicity. C and D. This 59-year-old woman complained of

nausea, vomiting, and abdominal pain after bone marrow transplant for acute myelogenous leukemia. CT enterography demonstrates intense small-bowel and colonic mucosal enhancement with submucosal edema indicating a diffuse enterocolitis. There is also moderate ascites and engorged mesenteric vessels. This patient responded to therapy for graft versus host disease.

edema, and mural stratification. They can also result in secondary pericolonic signs of inflammation including fat stranding (see Figure 2-71 and Imaging Notes 2-13).

vague symptoms such as nausea, vomiting, and crampy abdominal pain. GI bleeding is identified in only 30% of patients. Variability in symptoms is thought to reflect the speed with which the hemorrhage occurs as well as the volume of blood in the bowel wall.542 The most common sites of spontaneous intramural small-bowel hemorrhage are the duodenum and the proximal jejunum. This distribution is likely due to both the rich vascularity of the duodenum and the absence of a complete circumferential serosal layer. As a result, the duodenal wall demonstrates variable elasticity, which may facilitate expansion of intramural hematomas. In addition, the duodenal mesentery is short and relatively rigid because of fixation of the pylorus and the ligament of Treitz at the origin and termination of this bowel segment. Therefore, sudden

Bowel wall hemorrhage Intramural hemorrhage is an uncommon cause of segmental wall thickening, which can be seen in patients who are undergoing anticoagulation therapy or have an underlying bleeding diathesis. Other conditions, including alcoholism, leukemia, lymphoma, carcinoma, collagen vascular disorders, and pancreatitis can also be associated with intramural hemorrhage but are seen less frequently.542 Almost half of patients are asymptomatic at presentation. Those who are symptomatic typically present with

Chapter 2 Imaging of the Bowel 107 pressure changes can result in a shearing of the bowel wall layers, resulting in a tear of the vascular plexus. This likely explains why patients with bleeding diathesis can develop intramural bowel hemorrhage through benign activities as a cough or a valsalva maneuver.542 Abdominal x-rays can demonstrate thickened folds in the small bowel or colon that are indistinguishable from bowel wall edema. CT will also demonstrate circumferential and symmetric bowel wall thickening involving a long segment of bowel.231,543 In some cases, the bowel wall thickening will be nonspecific; however, in some cases the bowel wall thickening will be high in attenuation, ranging from 50 to 80 HU. This high attenuation is virtually specific for acute hemorrhage into the bowel wall. As this hemorrhage evolves, the attenuation will become isodense to the surrounding tissues at 10 days and hypodense after a few weeks. This hyperattenuation is more easily detected on unenhanced examinations. Therefore, when hemorrhage is suspected, unenhanced CT should be performed initially (see Figure 2-72).542,543 Subsequently, IV contrast can be administered and the hematoma will appear homogenous in attenuation with no appreciable enhancement compared to the unenhanced examination.542,543

of the GI tract and appear as long segment of symmetric bowel wall thickening (see Figures 2-69 and 2-73).231,232,544

Bowel Dilation The 2 most common causes of diffuse bowel dilation are bowel obstruction and a dynamic ileus. Taken together, they are both frequent causes of hospitalization and surgical consultations, representing 20% of all surgical admissions for acute abdominal pain.545,546 Imaging plays an important role in the diagnosis and management of these 2 conditions.

Terminology associated with bowel dilation

Infiltrating neoplasms are the least common cause of longsegment bowel wall thickening.530 These are most often due to intestinal lymphoma but can occasionally be due to carcinomas of the bowel wall, especially in individuals with chronic inflammatory conditions such as ulcerative colitis and Crohn disease. These tumors most often present as a focal concentric or eccentric mass of the bowel wall. However, rarely, they can infiltrate along the submucosa

Mechanical blockage of forward flow of intestinal contents is variably called “bowel obstruction,” “mechanical obstruction,” and “mechanical ileus.” Functional blockage of the bowel as a result of absent or diminished peristalsis is variably called “ileus,” “adynamic ileus,” “functional ileus,” and “paralytic ileus.” The term ileus is most often used to indicate bowel obstruction due to impaired peristalsis of the bowel but can occasionally be synonymous with a mechanical bowel obstruction. For example, the terms gallstone ileus and meconium ileus refer to mechanical obstruction by a gallstone and meconium, respectively. In this text, the terms smallbowel obstruction and colonic obstruction are used to indicate mechanical obstruction of the small bowel and colon, respectively, and the term adynamic ileus is used to indicate bowel blockage as a result of absent or diminished peristalsis. Obstruction can be classified further into simple obstruction, meaning the lumen is variably occluded but that normal blood flow is preserved, and strangulation whereby the blood flow is compromised, leading to edema, ischemia, and eventually necrosis and perforation. Simple obstruction

A

B

Figure 2-72 Bowel Wall Hematoma This 66-year-old man was anticoagulated for a pontine stroke when he developed abdominal pain. A. Unenhanced CT images demonstrate circumferential wall thickening of a segment of small bowel (arrows) associated with mesenteric fat stranding

(arrowheads). Note how the abnormal loop appears faintly higher attenuation than the adjacent abdominal wall. It measured 60 HU, suggesting intramural blood from a spontaneous bleed. B. The same image seen with a narrower window shows the high attenuation (arrow).

Infiltrating neoplasms

108 Diagnostic Abdominal Imaging

A

B

C

D

Figure 2-73 Long-Segment Bowel Wall Thickening due to Lymphoma This 36-year-old man had received a heart transplant in the previous year and now presented with abdominal pain, nausea, and vomiting. A-D. Axial CT images demonstrate a long segment of wall thickening involving the proximal jejunum. Note the

abrupt transition in A from normal to diseased bowel, a feature favoring a neoplastic rather than inflammatory cause for the wall thickening. Endoscopic biopsy was diagnostic of a B-cell lymphoma and is indicative of PTLD as a complication of the immunosuppression related to the heart transplant.

can further be classified into complete or partial depending on whether there is no passage or some passage of fluid or gas beyond the site of obstruction, respectively. An openloop obstruction denotes distal blockage with open loops proximally that are amenable to decompression by vomiting or drainage by placement of a tube in the stomach or proximal small bowel. A closed-loop obstruction denotes that flow into and out of the site of obstruction is blocked leading to progressive accumulation of fluid and gas within this loop. These concepts will be discussed in detail in the following section focusing on how radiology is critical to guiding management of both bowel obstruction and adynamic ileus.

through the use of nasointestinal decompression for patients with low-grade or uncomplicated obstruction.548 Historically, all patients with suspected bowel obstruction were taken to the operating room because of the limited ability to diagnose complications of SBO, specifically strangulation, on clinical examination and imaging, prompting the adage “Never let the sun rise or set on a small-bowel obstruction.”549 The present role of imaging, however, has moved beyond simply determining the presence of obstruction to addressing the severity, location, and etiology of the obstruction as well as the presence of complications of obstruction such as strangulation.547,550 This radiologic information is critical to triaging patients appropriately to conservative versus surgical management. In the past, the imaging evaluation of suspected bowel obstruction began with a combination of a supine and erect plain film of the abdomen, an examination known as an “obstruction series.” This approach is limited in sensitivity as plain films are diagnostic in only 50% to 60% of cases; equivocal in about 20% to 30%; and normal, nonspecific, or misleading in 10% to 20%. However, because these examinations are readily available, quickly obtained and low in cost they can be useful as an initial examination

Role of imaging in guiding management of bowel obstruction Over the past 2 decades, imaging has assumed a primary role in both the diagnosis and treatment of patients with bowel obstruction.547 This approach has evolved due mainly to 2 factors. First, the diagnosis of SBO has improved overall as well as in the early diagnosis of strangulation in particular, using cross-sectional imaging. Second, nonsurgical treatment of bowel obstruction has also increased, typically

Chapter 2 Imaging of the Bowel 109 Imaging Notes 2-16. Terminology of Bowel Obstruction Adynamic ileus:

Blockage of the bowel due to diminished peristalsis

Small-bowel obstruction:

Mechanical blockage of the small bowel

Colonic obstruction:

Mechanical blockage of the colon

Strangulation:

Bowel obstruction associated with compromise of the vascular supply leading to ischemia and infarction

Close-loop obstruction:

Bowel obstruction of both inflow and outflow of a single bowel loop leading to progressive accumulation of fluid and air within the loop

Intussusception:

Telescoping of the bowel

Volvulus:

Twisting of the bowel around a point

from which to triage patients for further imaging and management.551-553 Currently, many centers use CT directly in the evaluation of suspected bowel obstruction because it provides detailed information about both the site and cause of bowel obstruction and can also provide alternative diagnoses when bowel obstruction is excluded. Moreover, CT has a sensitivity of 82% to 100% for high-grade and complete SBO and the early demonstration of strangulation.554-559 As such, CT is useful in determining which patients would benefit from conservative management and close follow-up and which patients would benefit from immediate surgical intervention. In patients with equivocal CT scans or in whom there is persistent clinical concern for obstruction despite normal CT examinations, further evaluation with small-bowel enteroclysis or CT enteroclysis may be indicated.560 These examinations distend the bowel using high volumes of fluid that exaggerate the effects of mild or subclinical obstructions. Enteroclysis is further advantageous in that it utilizes frequent intermittent “real-time” fluoroscopic monitoring during the examination that facilitates the recognition of fixed and nondistensible segments and can point to the location of the obstruction.561-563 In addition, CT enteroclysis is more advantageous in that it is readily available and reproducible across practice settings. However, the largest drawbacks to both routine barium and CT enteroclysis is the placement of a nasoenteric tube, with resultant patient discomfort and slow transit in dilated hypotonic bowel segments.564

Imaging of bowel obstruction Abdominal plain films and abdominal CT are the primary imaging means of evaluating suspected bowel obstruction.

On occasion, barium studies, including enterocleisis, SBFT, and barium enemas are used to evaluate suspected obstruction. Although CT is highly sensitive and specific in the diagnosis of high-grade bowel obstruction, CT enteroclysis is recommended as the primary method of investigation in patients with suspected low-grade or subclinical SBO in certain patients, CT enteroclysis can be therapeutic as well as diagnostic.184

Plain film evaluation of bowel obstruction: With respect to the bowel, the obstruction series is designed to detect 3 major abnormalities as follows: (1) air-fluid levels within the bowel, (2) dilation of air-filled bowel, and (3) pneumoperitoneum. Detection and evaluation of these findings allow for the diagnosis of SBO, colonic obstruction, adynamic ileus, enteritis, and bowel perforation. Differentiation of these clinical conditions is based on 2 principles. First, the characteristic feature of obstruction of any peristaltic tube is proximal dilation and distal collapse. Second, air-fluid levels within the bowel are never normal on abdominal plain films except in the stomach. A horizontal beam film is necessary to detect air-fluid levels. The horizontal orientation of the x-ray beam, perpendicular to gravity, will show the border between dependent fluid and nondependent air, known as an “air-fluid level.” Horizontal beam films are typically taken with the patient standing erect but can also done with the patient in a decubitus position or as a crosstable lateral radiographs, where the patient lies on his or her back with the x-ray beam oriented horizontally. The primary differential diagnoses of air-fluid levels include SBO, colonic obstruction, adynamic ileus, and enteritis/colitis (Table 2-9). The first rule of obstruction of any peristaltic tube is proximal dilation and distal collapse. Tubular structures of the body, including the GI tract and the ureters, all undergo peristalsis regularly and spontaneously. This peristalsis propels the contents of the tube from proximal to distal locations. If there is obstruction of forward flow, the luminal contents will progressively increase proximal to the site of obstruction causing dilation. Continuation of peristalsis beyond the site of obstruction results in evacuation of the distal portion of the tube. Because no new contents are propelled into the portion of the tubular structure distal to the obstruction, that portion of the tube collapses. It is, therefore, possible to determine the site of obstruction

Table 2-9. Causes of Air-Fluid Levels in the Bowel 1. Obstruction a. Small bowel obstruction b. Colonic obstruction 2. Adynamic ileus 3. Enteritis/colitis (diarrhea)

110

Diagnostic Abdominal Imaging

Imaging Notes 2-17. Evaluation of Bowel Obstruction Based on the Tenant: Proximal Dilatation and Distal Collapse Use imaging to determine which loops of bowel are dilated • Abdominal x-rays indirectly determine which loops are dilated based on location and characteristics of bowel folds • CT, MRI, and barium studies directly evaluate the dilation of the bowel

by determining the location where the caliber of the tube changes. In SBO, there will be dilated air- and fluid-filled loops in the small-bowel proximal to the site of obstruction with collapse and a paucity of gas and fluid in the smallbowel and colon distal to the obstruction (see Figure 2-74). In colonic obstruction, there is dilation of the entire small bowel and the portion of the colon proximal to the obstruction with collapse of the distal portion of the colon (see Figure 2-75). Patients with a generalized adynamic ileus will present with dilation of the entire bowel from the duodenum to the rectum (see Figure 2-76). To determine the site of obstruction, the observer must be able to distinguish between air-filled small bowel and airfilled colon. This is occasionally a difficult task even for an experienced observer, but there are a variety of clues that help in distinguishing small from large bowel. The most important separating feature in differentiating small bowel and the colon is the location of the bowel segments. The ascending and descending colons are retroperitoneal structures and are therefore fixed in the lateral aspects of the abdomen. Consequently, they are the most lateral colon segments and travel in a craniocaudal direction. The rectum is also retroperitoneal and fixed in location in the midpelvis. The transverse colon is suspected on a mesentery and will typically form a U-shaped tubular structure between the splenic and hepatic flexures. These colonic structures are usually readily distinguished from small-bowel loops based on their characteristic locations. Both the small bowel and sigmoid colon can occupy the midabdomen and because of their mesenteries, their position can be variable. Consequently, the sigmoid colon may be confused with small bowel (see Figure 3-4). The different appearance of the folds of the small bowel and colon also help to differentiate these structures. Plica circularis, the folds of the small bowel, are thin, regularly spaced, uniform in size, and extend across the entire circumference of the lumen (see Figures 2-76 and 2-2). Haustra, the folds of the colon, are thick, irregularly spaced, and usually do not extend across the entire luminal diameter (see Figures 2-76 and 2-2). The site of obstruction in the bowel can be approximated by determining which portion of the small bowel is dilated. The jejunum is largely present in the left upper and

midabdomen. If air-fluid levels are confined to this location then the obstruction is usually in the jejunum. The ileum is predominantly located in the mid- and lower abdomen and pelvis. When the small bowel is dilated to this level, then the obstruction is usually in the distal ileum. When the site of obstruction is located in the distal small bowel, there is a “stepladder” appearance to the stacked segments of small bowel. Similar to the small bowel, the site of colonic obstruction can be suggested by determining the point at which the colon transitions from dilated to collapsed segments. Although both obstruction and adynamic ileus will produce dilation of the bowel, in patients with enteritis or colitis, or other causes of diarrhea, the bowel is usually not distended but air-fluid levels will often be present on horizontal beam films (see Figure 2-77). The normal small bowel is typically less than 2.5 cm in diameter.555 Maximal diameter of the colon varies based on location. The cecum is the most distensible portion of the colon but is usually less than 9 cm in diameter. The transverse colon is usually less than 6 cm and the descending and sigmoid colons are usually slightly smaller in caliber.278 Bowel loops larger than these thresholds are dilated. Duodenal obstruction causes a unique appearance termed the double bubble sign. In this case, only the stomach and duodenal bulb are dilated. Thus, 2 rounded air collections are seen in the upper abdomen, the larger stomach in the left upper quadrant (see Figure 2-25). This is most commonly seen in the newborn and may be due to duodenal atresia, duodenal stenosis, and rotational anomalies with or without congenital peritoneal bands, also known as Ladd bands.565 The “String of Pearls” sign refers to a row of small bubbles that resembles a string of pearls (see Figure 2-78). This sign is seen when the dilated small-bowel segments proximal to the site of obstruction are filled primarily with fluid and only a small amount of gas. In this setting, small amounts of air become trapped beneath individual plica circularis along the superior or nondependent wall of the fluid-filled dilated small bowel. On horizontal-beam radiographs, the meniscal effect of the gas outlined by the fluid gives the rounded appearance of a pearl. Although this sign can rarely be seen in adynamic ileus and gastroenteritis, it is considered virtually diagnostic of an SBO in the appropriate setting.566-568

CT evaluation of bowel obstruction: Multidetector CT plays a primary role in the evaluation of patients with acute obstruction of the small bowel and colon.547 Further, CT is superior to plain film in the detection of simple and closed loop obstruction and can provide pertinent additional information including the following: (1) the anatomic site(s) of obstruction in the small bowel or colon; (2) the etiology of the obstruction; (3) the severity of obstruction; and (4) the presence or absence of findings of vascular compromise (strangulation). The reader should systematically address each of these questions in their report as these data are essential in guiding treatment to surgical and nonsurgical approaches.556,557,569

A

B

C

D

E

F

Figure 2-74 Small-bowel Obstruction This 55-year-old woman complained of crampy abdominal pain, nausea, and vomiting. A. Erect abdominal radiograph demonstrates multiple air-fluid levels (white arrowheads). B. Although the loops in the upper abdomen (white arrows) in this supine radiograph appear large enough to represent colon, the folds are thin, regular, and pass all the way across the loop, findings characteristic of the plica circularis of small bowel. Therefore, these findings are indicative of a small-bowel

obstruction. Note the single surgical clip in the left of the pelvis (black arrowhead) indicting prior pelvic surgery. This smallbowel obstruction is most likely due to adhesions from the prior surgery. C-F. CT images through the abdomen and pelvis show dilated proximal small-bowel loops (white arrowheads) with collapsed distal small bowel (white arrows) and colon (large white arrows) diagnostic of a small-bowel obstruction, in this case as a result of abdominal adhesions from prior surgery.

112

Diagnostic Abdominal Imaging

A

B

C

D

E

F

Figure 2-75 Colonic Obstruction from Diverticulitis This 76-year-old woman with a history of ovarian cancer complained of abdominal pain and fevers. A. Erect view demonstrates multiple air-fluid levels in the colon (white arrowheads) and small bowel (black arrowheads). B. Supine view confirms the presence of a dilated, redundant transverse colon (white arrowheads) and multiple dilated loops of small bowel (black arrowheads). No gas is seen in the distal colon or rectum. These findings are typical of a colonic obstruction and in the setting of ovarian cancer could indicate obstruction due to peritoneal spread of the cancer. C-F. CT images were obtained in the prone position. Note that the air in the colon (black arrowhead in E) is below the fluid. We have inverted the images to appear in the supine position by convention. The CT examination confirms the presence of dilated loops of colon (white arrowheads). There is a long segment of wall thickening in the sigmoid colon (white arrows). There serosa appears brighter than the muscularis layer, indicating serosal enhancement and edema of the mucosa. Compare the fat adjacent to the sigmoid colon in E with the perinephric fat. The fat adjacent to the sigmoid colon has high attenuation streaks within it, called “fat stranding.” This finding can be associated with malignancies but it more commonly indicates edema due to an inflammatory process. The combination of segmental thickening and fat stranding in the colon will usually indicate diverticulitis. Surgical exploration confirmed a diagnosis of diverticulitis.

Chapter 2 Imaging of the Bowel 113

A

B

C

D

Figure 2-76 Adynamic Ileus This 64-year-old man complained of abdominal distension following a recent laminectomy. The patient was unable to stand and so a left lateral decubitus view of the abdomen (A) was obtained demonstrating multiple air-fluid levels (white arrowheads). Notice the distended rectum with air-fluid level (black arrowhead). B. Supine radiograph demonstrates the characteristic appearance of distended cecum (“C”), transverse colon (“T”), and descending colon (“D”). The other bowel loops represent dilated small bowel. Note that the rectum (“R”) contains air but is not dilated. This is because the rectum is a posterior structure and air rises out of the rectum and into the remainder of the colon. This examination shows how a decubitus radiograph can be useful in confirming free passage of air into the rectum, a finding that indicated bowel dilatation due to an adynamic ileus. This ileus is not due to the spine surgery but is a manifestation of narcotics used to treat the surgical pain. C and D. This 61-year-old woman underwent hysterectomy 2 days prior. She now complains of nausea and vomiting. C. Erect radiograph shows multiple air-fluid levels. D. Supine radiograph shows dilated loops of bowel. The loops in the left upper quadrant have thin continuous folds, closely stacked with each other, typical of small bowel. The loops in the right upper quadrant have fewer folds that appear thicker (black arrows), typical of the colon. Air is seen in the rectum (white arrowhead). This clinical history and radiographic appearance is typical of a postoperative adynamic ileus.

114

Diagnostic Abdominal Imaging Imaging Notes 2-18. Reporting of Bowel Obstruction The CT interpretation of bowel obstruction should include determination of the: 1. Anatomic site(s) 2. Etiology of obstruction 3. Severity of obstruction 4. Presence or absence of vascular compromise (strangulation)

Figure 2-77 Giardia Enteritis This 22-year-old HIV-positive man complained of abdominal pain and diarrhea. Erect radiograph demonstrates multiple air-fluid levels in nondilated small bowel and colon. This combination of findings is usually a manifestation of an enteritis. Stool cultures confirmed the presence of Giardia species.

The CT evaluation of bowel obstruction follows the same general tenants as abdominal plain films: proximal dilation and distal collapse. Computed tomographic criteria for SBO are the presence of dilated small-bowel segments, defined as segments with diameter >2.5 cm from outer wall to outer wall, proximal to the site of obstruction with normal-caliber or collapsed segments distal to the site of obstruction.547,570 At the transition point between dilated and collapsed bowel segments, CT can evaluate for the various causes of obstruction, including an intrinsic or extrinsic mass, bowel herniation, or abnormal thickening of the bowel wall. When no abnormality is visualized at the transition point the presumed etiology of obstruction is adhesions. This is presumed both because adhesions are the most common cause of obstruction and because they are not visualized on CT. Severity of bowel obstruction can be evaluated using the degree of distal collapse, proximal bowel dilation, the passage of oral contrast, and occasionally through the presence of the “small-bowel feces” sign. If positive oral contrast material is given, the passage of a sufficient amount of contrast material through the transition point indicates an incomplete low-grade or partial SBO.547,571 A high-grade partial SBO is diagnosed when there is some stasis and delay in the passage of the contrast medium, so that diluted oral contrast material appears in the distended proximal bowel and minimal contrast material appears in the collapsed distal loops. Finally, a high-grade complete obstruction is defined by the

absence of contrast beyond the point of obstruction.547 Highgrade obstruction, either partial and complete, can also be diagnosed on the basis of an approximately 50% difference in caliber between the proximal dilated bowel and the distal collapsed bowel.547 This discrepancy in bowel caliber is due to accumulation of unabsorbed fluid proximal to the obstruction and complete evacuation of the bowel contents distal to the obstruction point after several days,572 which exaggerates the discrepancy in caliber between the proximal and distal small-bowel loops. The “small-bowel feces” sign can be an additional clue to the diagnosis of SBO, but as explained below, it should be interpreted with caution. The normal contents of the small bowel are liquid and air. When there is delayed smallbowel transit stasis of bowel contents, it leads to the development of particulate, feculent-appearing material with gas bubbles that resembles the appearance of stool in the colon on CT scans. This is called the small-bowel feces sign and is thought to reflect incompletely digested food, bacterial overgrowth, or increased water absorption of the distal small-bowel contents due to stasis. Although the small-bowel feces sign can be seen in SBO, it is neither sensitive nor specific for this diagnosis.573 The incidence of this sign in the setting of SBO is variable ranging from 7% to 55%. When present in the setting

Imaging Notes 2-19. Grading of Bowel Obstruction

Contrast Beyond Transit ion Point

≥50% Difference in Caliber of Bowel Proximal and Distal to Obstruction

Incomplete, low-grade

Moderate

No

Incomplete, high-grade

Minimal

Yes

Complete, high-grade

None

Yes

Grade

Chapter 2 Imaging of the Bowel 115

A

B

C

D

E

Figure 2-78 String of Pearls Sign This 43-year-old man presented with intermittent abdominal pain. A. Supine radiograph shows a general paucity of bowel gas. However, there is a dilated loop with multiple thin folds characteristic of small bowel (black arrows) and a paucity of gas in the colon. This is strongly suggestive of a small-bowel obstruction, where the loops are predominantly fluid filled. There is a cluster of oval gas bubbles (white arrowheads) in a row in the left upper quadrant. This finding is known as the “string of pearls” sign and is associated with a diagnosis of small-bowel obstruction. B-E. CT images through the abdomen confirm the presence of multiple dilated loops of small bowel that are primarily filled with fluid. There are some collapsed loops of small bowel (large arrowheads) and colon (white arrows). This constellation of findings meets the imaging criteria for a small-bowel obstruction. Note the row of bubbles (small arrowheads) in B. This row is caused by air trapped under individual plica circularis. This phenomenon is what causes the “string of pearls” appearance on the abdominal x-ray. There was no identifiable cause for this obstruction and so a diagnosis of adhesions was suggested.

116

Diagnostic Abdominal Imaging

A

B

Figure 2-79 Small-Bowel Feces Sign This 42-year-old woman with a history of a prior hysterectomy complained of intermittent crampy abdominal pain for several months. In image A, there is normal-appearing proximal small bowel (arrow) and colon (large arrowhead). Note the appearance of feces in the ascending colon. In image B, there is a dilated loop

of small bowel containing a mixture of air and solid material (small arrowheads). This is an abnormal finding called the smallbowel feces sign that is most often associated with a low-grade chronic small-bowel obstruction. Surgical exploration confirmed a diagnosis of SBO due to pelvic adhesions.

of SBO, it has been observed more often in patients with moderate or severe obstruction. However, it is not able to discern between low-grade, subacute obstruction and moderate or high-grade obstruction.547,573-575 A helpful guide to determining the significance of this sign is that when it is present in the setting of obstruction, it is typically located just at or proximal to the transition point and is therefore a clue to the location of obstruction.573,574 On the other hand, when this sign is present in the setting of normal or mildly dilated small bowel, it is usually indicative of a diversity of etiologies of delayed intestinal transit from causes other than obstruction including the following: cystic fibrosis, infectious enteritis, rapid jejunostomy tube feedings, bezoars, and reflux of fecal matter from the cecum (see Figure 2-79 and Table 2-10).574,576,577 The site of obstruction or transition point is determined by identifying a caliber change between the dilated proximal and collapsed distal bowel loops. This should be accomplished in a systematic fashion starting from the terminal ileum and moving proximally through the small bowel in the setting of suspected SBO and starting from the rectum and moving proximally in the setting of suspected colon obstruction. Once the site of obstruction is identified, this should be confirmed by starting at the distal duodenum and moving distally or at the cecum and moving distally in the setting of small and large bowel obstruction, respectively. Multiplanar reformats

also can be helpful in confirming the suspected site of obstruction.558

Barium studies in the evaluation of bowel obstruction: Routine CT demonstrates accuracy of greater than 90% in the diagnosis of high-grade SBO. However, the sensitivity and specificity of routine CT in the diagnosis of lowgrade SBO is 50% and 94%, respectively. As above, when low-grade small obstruction is suspected, CT enteroclysis is superior to routine CT with sensitivity and specificity of 89% and 100%, respectively. For the indication of suspected SBO, CT enteroclysis is usually performed with positive oral contrast. This permits performance of fluoroscopy during oral contrast

Table 2-10. Causes of the Small-Bowel Feces Sign 1. Small-bowel obstruction 2. Cystic fibrosis 3. Infectious enteritis 4. Rapid jejunostomy feeds 5. Bezoar 6. Reflux of feces from the colon

Chapter 2 Imaging of the Bowel 117 infusion, which can detect subtle delays in the passage of contrast material. This technique is particularly sensitive in the detection of adhesions, both obstructive and nonobstructive. In addition, CT enteroclysis is advantageous in patients with symptoms of proximal jejunal obstruction, particularly when the stomach may have been decompressed either through the use of a nasogastric tube or patient emesis prior to the scan.

Causes of SBO Once the site of obstruction is discovered, the cause of the obstruction should be determined. Over the past 50 years, the etiology of SBO in Western society has shifted from predominantly hernias to adhesions, Crohn disease, and malignancy as the top 3 causes in decreasing order.547 Hernias continue to represent the predominant cause of SBO in developing countries.578 Etiologies of SBO can be categorized into extrinsic, intrinsic, and intraluminal. Intrinsic bowel lesions are usually seen at the transition point and include causes of focal wall thickening that may lead to intussusception, such as Crohn disease, celiac disease, primary small-bowel neoplasms, hematomas, and ischemia. Extrinsic causes include adhesions, internal and external hernias, secondary neoplasms, endometriosis, and hematomas. Finally, intraluminal lesions are identified by their location and imaging characteristics that differ from other enteric contents and include gallstones, bezoars, and foreign bodies (see Table 2-11).547

Adhesions: Adhesions are the most common cause of SBO, accounting for approximately 50% to 80% of all cases. Adhesions are mostly due to prior intra-abdominal intervention, with only a minority due to peritonitis.570,579-582 Adhesive bands are not seen directly on imaging examinations but are implied when there is an abrupt change in the caliber of the bowel without associated extrinsic mass lesion, intraluminal foreign body, or inflammatory changes at the transition point (see Figures 2-74, 2-78, and 2-79).547 Kinking and tethering of the adjacent nonobstructed smallbowel segments also suggest the presence of adhesion (see Figure 2-80).

Hernia: Hernias of the peritoneum are responsible for approximately 10% of SBOs in developed countries. In developing countries, they are the leading cause of SBO.578 External hernias result from a defect in the abdominal or pelvic wall at sites of congenital weakness or prior surgery. Examples of external hernias include direct and indirect inguinal, incisional, umbilical, femoral, obturator, diaphragmatic, and Spigelian hernias. Diagnosis of these hernias is usually, but not always, obvious on physical examination. When the herniated bowel is filled with air, the hernia can also be visualized on abdominal plain films as an abnormally positioned small-bowel loop, in a location exterior to the peritoneum in the setting of an SBO. On barium, these hernias are best depicted in the lateral projection but

Table 2-11. Causes of Small-Bowel Obstruction 1. Extrinsic Causes a. Adhesions (50%-80%)a i. Prior surgery ii. Prior peritoneal inflammation i. Peritonitis ii. Pelvic inflammatory disease iii. Other b. Hernia (all subtypes) (10%) c. Endometrioma d. Hematoma 2. Intrinsic causes a. Neoplasm (10%) i. Primary tumors ii. Peritoneal metastasis iii. Direct invasion from adjacent neoplasms iv. Hematogenous metastasisb b. Inflammation i. Crohn disease (7%) ii. Tuberculosis iii. Parasites (Ascaris) c. Intussusception i. Post viral infection ii. Meckel diverticulum iii. Polyp iv. Lipoma v. Celiac disease vi. Other lead point. d. Vascular lesions i. Postradiation ii. Ischemia 3. Intraluminal causes a. Gallstone ileus b. Bezoar c. Foreign body 4. Congenital causes a. Midgut volvulus b. Duodenal atresia and other small-bowel atresia aNumbers in parenthesis indicate the approximate percentage of SBO

attributed to this cause. bMetastasis acts as a lead point causing intussusception and SBO.

can also be detected on frontal projections where there is compression and deformity of the bowel segments entering and exiting the hernia sac. When detected, reduction can be attempted during fluoroscopic evaluation.583 On CT examinations, hernias can be easily identified as protrusion of bowel loops through a defect in the abdominal wall (see Figure 2-81). Internal hernias are less common than external hernias and occur when there is protrusion of the viscera through the peritoneum or mesentery and into a compartment within the abdominal cavity. These hernias include,

118

Diagnostic Abdominal Imaging

A

B

Figure 2-80 Adhesions This 75-year-old man had undergone total collectomy years before as therapy for ulcerative colitis and complained of nausea and vomiting. A and B. Two spot films from a small-bowel

follow-through show angulation (white arrow) of some loops of bowel and regions of tethering (black arrows). These are typical features of adhesions.

in order of decreasing frequency, paraduodenal (53%), pericecal (13%), foramen of Winslow (8%), transmesenteric and transmesocolic (8%), intersigmoid (6%), and retroanastomotic (5%).584,585 The overall incidence of internal hernias is 0.2% to 0.9%; however, they account for approximately 0.5% to 5.8% of all cases of intestinal obstruction.585 Internal hernias can be silent but the majority cause symptoms such as epigastric discomfort, periumbilical pain, and recurrent episodes of intestinal obstruction.584 Internal hernias are almost always diagnosed by imaging and are associated with a high mortality rate, exceeding 50% in some series.585 On CT and MRI, internal hernias typically produce a saclike mass or cluster of dilated smallbowel loops in an abnormal anatomic location within the abdomen, in the presence of SBO. The vascular pedicle of these loops is often engorged, stretched, and displaced from its normal position, converging at the hernial orifice.584 Paraduodenal hernias are more common in men, and are located on the left in 75% of cases.585 Left paraduodenal hernias occur when bowel prolapses through the Landzert fossa, which is present in approximately 2% of the population and is located behind the ascending (fourth portion) duodenum. On CT, they are characterized by an abnormal cluster or saclike mass of dilated small-bowel loops lying between the pancreas and stomach to the left of the ligament of Treitz. These dilated segments typically exert a mass effect on adjacent structures, including displacement of the posterior stomach wall anteriorly, of the duodenojejunal junction inferomedially, and of the transverse colon

inferiorly. Abnormalities of the mesenteric vessels include engorgement, crowding, and stretching of the vessels at the entrance of the hernia sac as well as displacement of the IMV and ascending left colic artery along the anterior and medial border of the sac.584,585 Right paraduodenal hernias occur when bowel prolapses through the fossa of Waldeyer located immediately behind the SMA and inferior to the transverse (third portion) duodenum. On CT and MRI, these hernias will present as a cluster of dilated small-bowel loops in the right midabdomen near the root of the small-bowel mesentery. Usually, right paraduodenal hernias occur in the setting of a small-bowel malrotation with a normally or incompletely rotated colon. Accordingly, right paraduodenal hernias are

Imaging Notes 2-20. Neoplastic Mechanisms of Bowel Obstruction Primary neoplasm • Annular narrowing by the primary mass • Intussusception of primary neoplasm Secondary neoplasm • Peritoneal implants with invasion • Direct extension with invasion • Intussusception of hematogenous metastasis

Chapter 2 Imaging of the Bowel 119

A

B

D Figure 2-81 Small-Bowel Obstruction due to Inguinal Hernia This 98-year-old man complained of abdominal pain, nausea, and vomiting for 2 days. A-C. Axial CT images demonstrate dilation of the proximal small bowel (large arrowheads) with collapse of the distal small bowel and colon (small arrowheads), diagnostic of a small-bowel obstruction. Careful observation shows protrusion of a distal small-bowel loop (arrow) through a right inguinal hernia. D. Coronal image shows the inguinal hernia (arrow) to better advantage. C

associated with abnormal location of the SMV to the left of, and ventral to, the SMA and with absence of the normal horizontal duodenum. A hallmark of right paraduodenal hernia is visualization of the SMA and right colic vein along the anterior-medial border of the encapsulated small-bowel loops.584,585

Neoplasm: Neoplasms are among the most common cause of SBO and can cause obstruction through a variety of mechanisms. Serosal implants from peritoneal carcinomatosis are the most common neoplastic mechanism of

SBO. This is most often due to ovarian cancer but can also be a result of appendiceal, colon, and endometrial carcinomas but is rarely due to other neoplasms such as primary peritoneal mesothelioma. These implants invade and then obstruct the small bowel. Their presence is suggested by visualization of extrinsic serosal soft-tissue nodules or masses located near the transition point. Direct growth into the small bowel from a non-small bowel primary abdominal malignancy is a rare cause of SBO. This will appear as a large mass arising from an adjacent organ, which invades the small bowel, causing

120

Diagnostic Abdominal Imaging

obstruction. This phenomenon is most commonly seen from tumors of the pelvic organs such as cervical or ovarian cancers. If cecal or colonic malignancy involves the ileocecal valve, it can also result in SBO. Primary small-bowel carcinomas can grow into an annular mass similar to colon carcinoma leading to obstruction (see Figure 2-25). Primary small-bowel neoplasms are a rare cause of SBO because intrinsic small-bowel neoplasms constitute less than 2% of GI malignancies. When they occur, small-bowel neoplasms are usually advanced and are characterized by pronounced, asymmetric, and irregular mural thickening of the small-bowel wall at the transition point.556,586 Finally, hematogenous metastasis and small primary neoplasms can act as a lead point for small-bowel intussusception—a very rare mechanism of small-bowel and colonic obstruction whereby the bowel telescopes upon itself. This is most frequently a result of melanoma or Kaposi sarcoma metastasis. When intussusception is visualized on CT, a leading mass can be identified but should be differentiated from the soft-tissue pseudotumor of the intussusception itself.570,579,581,582,587

Crohn disease: Crohn disease accounts for approximately 7% of SBOs.578 SBO in Crohn disease can be a manifestation of 3 clinical situations: (1) luminal narrowing secondary to the transmural acute inflammatory process, (2) chronic fibrotic stenosis of affected segments, and (3) adhesions, incisional hernias, or postoperative strictures in patients who have undergone previous intestinal surgery.547,556,571,586,588,589 Distinguishing between these conditions is essential for proper patient treatment. Crohn disease is discussed more completely in the previous section LongSegment Wall Thickening of the Small Bowel and Colon.

Intussusception: Intussusception is a rare cause of SBO in adults and accounts for less than 5% of cases; however, this is a relatively common cause of small-bowel and colonic obstruction in infants and young children.571,590 In intussusception, a soft-tissue projection such as a polyp, inverted Meckel diverticulum or bowel neoplasm is pulled forward with other endoluminal contents by normal peristalsis. As this lesion is pulled forward, it brings along the bowel wall to which it is attached, causing that portion of bowel to invert and telescope within the lumen of the more distal small bowel. Pediatric intussusception is idiopathic in up to 90% of cases. In these cases, lymphoid hyperplasia (hypertrophied Peyer patches) is believed to act as the lead point. The etiology of adult intussusception, on the other hand, remains controversial. Reviews of hospital discharge summaries and surgical reports have suggested that up to 80% of adult intussusception cases are associated with an underlying pathologic lead point such as a neoplasm, adhesion, inverted Meckel diverticulum, and foreign body.16,571,587,591 However, studies based on CT and MRI have shown that almost 50% of adult intussusceptions visualized on CT and MRI are idiopathic. These

imaging-detected intussusceptions are typically transient, resolve spontaneously, are without clinical significance, and are thought to be a result of transient dysmotility (see Figure 2-45). These transient intussusceptions are predominantly located in the small bowel and have a higher incidence in patients with celiac disease and scleroderma in whom there is a higher incidence of small-bowel dysmotility.571,592 Because of limited soft-tissue resolution, abdominal radiographs can only demonstrate small-bowel or colonic obstruction, which is rarely associated with intussusception. However, cross-sectional imaging including CT, US, and MRI will depict the collapsed, intussuscepted proximal bowel (intussusceptum) with the mesenteric fat and vessels lying within the wall of the distal bowel (intussuscipiens). In profile, the intussusceptum appears as a sausage-shaped structure within the lumen of the bowel and, in cross section, it produces a targetlike appearance because of the alternating layers of mucosa and mesenteric fat. In some cases, careful observation will demonstrate the underlying lesion at the lead point of the intussusceptum; however, cross-sectional imaging has a limited ability to distinguish a lead point from thickened or edematous bowel wall (see Figure 2-82).571,592-595 In the absence of clinical symptoms, the majority of shortsegment intussusceptions will represent a transient, idiopathic finding without significance and require no further evaluation (see Figure 2-45).

Midgut volvulus: Midgut volvulus is a rare cause of SBO in adults but is an important cause of SBO in infants and young children. Midgut volvulus is a direct complication of malrotation of the small bowel. In normal fetal development, the developing bowel undergoes 2 rotations that result in the duodenal sweep placing the ligament of Treitz and proximal jejunum in the left upper quadrant and the cecum and ileocecal valve in the right lower quadrant. In this alignment, the small-bowel mesentery is a long band extending from the ligament of Treitz to the ileocecal valve. This broad band of mesentery is resistant to twisting of the small bowel. In small-bowel malrotation, the small-bowel mesentery comes to a single point in the right upper quadrant at the site of the SMA rather than a long band, and the entire small bowel is located in the right side of the abdomen. At the same time the entire colon is located in the left side of the abdomen. Because the entire mesentery comes to a single point, it is possible for the entire small bowel to rotate around this point, a phenomenon known as “midgut volvulus.” Midgut volvulus will typically result in obstruction of the small bowel and compromise of the blood supply of the small bowel leading to ischemia and infarction of nearly the entire small bowel. Most cases of midgut volvulus will present in children during the first month of life. Approximately 60% to 80% of patients will present with bilious vomiting.596,597 Emergency surgery must be performed to repair the malrotation because a delay in diagnosis can result in infarction of the

Chapter 2 Imaging of the Bowel 121

A

B

C

D

Figure 2-82 Intussusception in 2 Patients This 26-year-old woman presented with cramp abdominal pain 2 months following cesarean section. A. Ultrasound of the left lower quadrant shows 2 inner lines (arrowheads) surrounded by 2 outer lines (arrows). This is the ultrasound appearance of intussusception of the bowel. B-D. CT images show a masslike

area in the left lower quadrant. Within the mass are a curvilinear area of fat (arrowheads) and multiple enhancing vessels (arrows) that represent the mesentery of the bowel as the intussusceptum passes into the intussuscipiens. These findings are diagnostic of intussusception. Surgical exploration showed a colon cancer as the lead point of this colo-colic intussusception.

small bowel and death of the child. As the use of CT in emergency departments has increased, however, midgut volvulus has increasingly been recognized in adults. Adults can present with abdominal pain, nausea, and vomiting similar to the presentation in infants. In some adults, midgut volvulus can manifest as chronic intermittent abdominal pain that resolves when the volvulus spontaneously reduces.596,598 Conventional radiography is rarely helpful in the diagnosis of midgut volvulus. Fluoroscopic upper GI and

small-bowel examinations are the gold standard for revealing the characteristic abnormal position of most of the small bowel in the right abdomen and the resultant abnormal location of the ligament of Treitz. Normally, on upper GI images the duodenal sweep terminates at the ligament of Treitz in the left upper quadrant of the abdomen at, or to the left of, the left L1 pedicle. In patients with malrotation, the duodenum fails to curve upward and to the left and instead remains in the right side of the abdomen. The

122

Diagnostic Abdominal Imaging

twisted proximal small bowel has a characteristic corkscrew-like appearance on fluoroscopic images.597 In addition to the abnormal position of the small bowel and colon, malrotation will often result in an abnormal relationship between the SMV and the SMA. In the normal individual, the SMV is located to the right of the SMA; however, in patients with malrotation, the SMV may occupy a position directly anterior or to the left of the SMA.597,599 US can sometimes illustrate the abnormal position of the SMV and SMA. However, a normal SMA-SMV relationship does not exclude malrotation, and the upper GI examination remains the preferred imaging modality to demonstrate malrotation of the bowel. Computed tomographic examinations can demonstrate all of the typical manifestations of malrotation and midgut volvulus, including the abnormal relationship of the SMV to the SMA, the abnormal location of the small bowel in the right side of the abdomen, and the entire colon in the left side of the abdomen and distention of the volvulized loops (see Figure 2-83). CT can also demonstrate swirling of vessels in the mesenteric root, which can be a sign of mesenteric vovulus597,598,600 but can also occur in normal patients. Therefore, application of the mesenteric swirl sign should be used very carefully. In most cases, swirling of the mesenteric vessels should be assumed to represent a normal variant, unless there are other findings of malrotation and midgut volvulus present. Familiarity with the CT findings of midgut volvulus is important because adult patients will typically present with nonspecific symptoms.

2 groups: adynamic (paralytic) ileus and mechanical ileus. The older term mechanical ileus is now called either SBO or colonic obstruction. Thus, the term gallstone ileus should in current terminology be called “gallstone obstruction” but the older term gallstone ileus remains the terminology that is generally used. Gallstone ileus is a rare complication of recurrent acute cholecystitis whereby a large gallstone erodes through the gallbladder directly into the small bowel via a biliary-intestinal fistula. As a result of its large size, the stone becomes impacted at a narrow segment of the bowel, typically the ileocecal valve, resulting in SBO. This pathophysiology results in a classic triad of findings on both plain film and CT: (1) obstructed bowel gas pattern, (2) pneumobilia secondary to the biliary-intestinal fistula, and (3) an ectopic gallstone, usually located in the right lower quadrant at the site of the ileocecal valve (see Figure 4-41).601 Gallstone ileus has become an increasingly rare event in developed countries. This is probably due to improved diagnosis of acute cholecystitis with cross-sectional and hepatobiliary imaging.

Parasitic infection: Globally 1.5 billion people, approxi-

not capable of identifying the cause of SBO. One exception to this rule is gallstone ileus. Although current usage of the term ileus is confined to bowel obstruction due to diminished peristalsis, historically, the ileus was subdivided into

mately 25% of the world population, are infested with ascariasis.602 Ascaris infection is among the most common causes of SBO in the developing world. Although ascariasis infection occurs at all ages, it is most common to affect children 2 to 10 years old with a decreasing trend over the age of 15 years.603 Children are at higher risk for developing Ascaris-related intestinal obstruction because of the smaller diameter of the lumen of the bowel and often a greater worm load than that seen in adults.602 In a 2-year study of a tertiary care hospital in India, 63% (131/207) SBOs in children were a result of ascaris infection.604 Transmission occurs primarily via ingestion of water or food contaminated with Ascaris lumbricoides eggs.602

A

B

Figure 2-83 Malrotation Without Volvulus This 19-year-old man was being evaluated for testicular cancer. A. Contrast-enhanced axial CT images shows an abnormal position of the larger SMV (white arrowhead), anterior rather than to the right of the smaller SMA (black arrowhead).

B. At the level of the umbilicus, the contrast-filled small bowel (small arrows) occupies the right of the abdomen and the feces-filled colon (large arrows) is seen only in the left of the abdomen. These findings are typical of malrotation of the bowel.

Gallstone ileus: In most cases, the abdominal plain film is

Chapter 2 Imaging of the Bowel 123 Children playing in contaminated soil can acquire the parasite from their hands. The number of adult worms per infested person relates to the degree of continued exposure to infectious eggs over time because adult worms do not multiply in the human host. Most patients infected with A lumbricoides are clinically asymptomatic. Symptomatic patients present with the typical symptoms of obstruction such as abdominal pain, nausea, and vomiting. However, occasionally passage of worms in the stool or vomiting of worms can also be seen.602 Obstruction usually requires a relatively large worm burden, estimated to be greater than 60 worms in most cases.445 The greater the worm burden, the higher the likelihood of death due to obstruction. Obstruction due to ascariasis typically occurs at the terminal ileum, although worms can be found throughout the bowel. Delay in the management of the intestinal obstruction can lead to bowel perforation and contamination of the peritoneum with worms and eggs.605 Abdominal x-rays will demonstrate the typical features of SBO. However, occasionally multiple linear or serpentine structures, representing the worms, can be seen as filling defects within the air-filled lumen of the small bowel, also called the “cigar bundle” appearance.602 Abdominal ultrasound can demonstrate the intraluminal worms. In longitudinal section, the Ascaris worm can be seen as a linear intraluminal mass with 3 or 4 internal linear echogenic interfaces. In cross section, the worm will appear as a round or oval structure, with multiple layers resembling a target.606-608 On real-time US, the worms demonstrate a curling movement within the lumen of the bowel.602,608 CT and MRI can also demonstrate the presence of worms within the lumen of the bowel. They appear as approximately 0.5-cm by 10-cm linear or curvilinear filling defects within the lumen of the bowel.609,610 In some cases, the worms will ingest the oral contrast and their gut will appear radiodense.

Duodenal and other small-bowel atresia: Small-bowel atresia is an uncommon cause of intestinal obstruction, which most commonly occurs in the duodenum. A 25-year review of intestinal atresia from a single pediatric center demonstrated 277 cases of intestinal atresia, of which 138 (50%) were in the duodenum and 128 (46%) were at sites in the jejunum and ilium.611 Patients typically present with bilious emesis, abdominal distension, and/or feeding intolerance. Duodenal atresia is associated with Down syndrome; approximately one-quarter of patients with duodenal atresia will have associated Down syndrome.611 The double bubble sign of duodenal obstruction will be present in the majority of patients with duodenal atresia and was evident in 78% (108/138) of cases in one series.611 Jejunoileal atresia can be associated with gastroschisis in a minority of patients.

of an SBO. History of prior surgery, peritonitis, or pelvic inflammatory disease will suggest adhesions as a cause of obstruction. Hernias are often clinically palpable and their presence will denote this as a cause of obstruction. Some individuals will have a history of an abdominal or pelvic malignancy, which may directly invade the small bowel or metastasize via the peritoneum and produce obstruction.

Indications for emergency surgery in patients with SBO In most cases, SBO can initially be treated with nasointestinal decompression. Evacuation of the gastric contents will progressively remove excess fluid proximal to the obstruction and result in decreased caliber of the obstructed loops. Many cases of obstruction due to adhesions can be successfully managed with nasointestinal decompression, without surgical intervention. Some other causes of SBO may require surgical intervention but in most cases can initially be treated conservatively with delayed surgical intervention when necessary. However, patients with closed-loop obstruction have a high morbidity and mortality because of the increased risk for mesenteric ischemia.547,550,569,571 These patients require emergency laparotomy to prevent bowel perforation, sepsis, and other complications of mesenteric ischemia. Therefore, it is of critical importance that patients with imaging evidence of SBO be evaluated for the presence of closed-loop obstruction and for evidence of mesenteric ischemia of the obstructed loops.

Closed-loop obstruction: Closed-loop obstructions are diagnosed when a bowel loop of variable length is occluded at 2 adjacent points along its course. This is most commonly due to adhesive bands that cross the adjacent loops, but can also be secondary to hernias. In this process, the obstructed loops and their mesentery are fixed at a point that allows the closed loop to twist along its long axis, producing a small-bowel volvulus. Flow into and out of the closed loop is blocked, causing progressive accumulation of fluid and gas. If it is not relieved, the progressive distention of the closed loop leads to compression of the entrapped mesentery and the development of ischemia of the closed loop.550 On CT, closed-loop obstruction is diagnosed by characteristic fixed radial distribution of several dilated, usually fluid-filled bowel loops. These radial loops have either a C-shaped or U-shaped configuration, with the open end of the loop pointing at the site of obstruction. In addition, there will be stretched and prominent mesenteric vessels converging toward the point of torsion. This produces a “beak sign” on CT at the site of the torsion due to fusiform tapering or a “whirl sign,” reflecting rotation of the bowel loops and mesenteric vessels around the fixed point of obstruction (see Figure 2-84).550

Other causes of SBO: There are a variety of other causes of SBOs, some of which are listed in Table 2-11. Often clinical history or physical examination will indicate the cause

Strangulation: Strangulation is defined as a mechanical obstruction associated with intestinal ischemia,

124

Diagnostic Abdominal Imaging

B

C

A

D

Figure 2-84 Closed Loop Obstruction This 49-year-old man underwent a bowel resection for Burkett lymphoma 6 months previously. He now complains of intermittent severe abdominal pain. A. Supine view of the abdomen shows multiple dilated loops of bowel in the left upper quadrant. Note the multiple thin folds (plica circularis) in the loops characteristic of small bowel. A few tiny bubbles of gas are seen in the right upper quadrant, probably within collapsed colon. These findings are typical

of a high-grade small-bowel obstruction. B-D. CT images through the abdomen show multiple dilated small-bowel loops with collapsed colon (white arrows) and collapsed distal small bowel (black arrow). Note how there is a cluster of loops in (C) (arrowheads) which appear contained in the right lower quadrant. Centrally, they form a beak (black arrowhead). These are CT features indicating a closed loop obstruction. Surgical exploration confirmed a closed loop obstruction due to adhesions.

which if left untreated will progress to bowel necrosis. This is seen in approximately 10% of patients with SBO and almost always occurs in the setting of a closed-loop obstruction. The distinction between closed-loop obstruction (incarceration) and strangulation (ischemia) should be emphasized as these are related phenomena but separate pathologic entities. Strangulation always develops because of a closed loop; however, a closed loop can be only partially obstructed, may not be associated with strangulation, and can resolve spontaneously.550 This

differentiation is critical to the interpretation of obstruction. When strangulation occurs in the setting of obstruction, mortality rates rise to 20% to 37%, compared with 5% to 8% for a simple obstruction.571,612-614 This high mortality rate in strangulating obstruction is mainly attributed to a delay in establishing the correct diagnosis and initiating appropriate treatment. Because CT examinations are the primary means of diagnosing closed-loop obstruction and strangulation, imaging plays a critical role in the management of these patients.

Chapter 2 Imaging of the Bowel 125 Imaging Notes 2-21. Features of Obstruction with Strangulation (Mesenteric Ischemia) Presence of the typical features of obstruction and any of the following features: 1. Lack of enhancement of the bowel wall following intravenous contrast 2. Thickened bowel wall 3. Target sign in the bowel wall 4. Pneumotosis intestinalis 5. Portal venous gas

Table 2-12. Causes of Colonic Obstruction 1. Neoplasm a. Primary malignancy b. Direct invasion of extra-GI tumor c. Peritoneal metastasis 2. Volvulus a. Cecal b. Sigmoid c. Other 3. Diverticulitis 4. Inflammatory bowel disease a. Crohn disease b. Ulcerative colitis 5. Ischemic stricture

CT findings suggesting the presence of mesenteric ischemia include (1) thickening of the bowel wall, (2) alterations in the attenuation of the affected bowel wall, producing a halo or “target sign,” (3) pneumatosis intestinalis, and (4) portal venous gas. However, these findings are not specific for mesenteric ischemia and can be seen with infections of the small bowel. A specific finding is lack of wall enhancement; asymmetric enhancement or even delayed enhancement may also be found. Localized fluid and hemorrhage in the mesentery can also be seen in patients with mesenteric ischemia.550,569,571 Strangulation is usually associated with adhesions or internal or external hernias.

Causes of colonic obstruction The causes of colonic obstruction are very similar to the causes of SBO with the exception that the 2 most common causes of SBO, adhesions and hernias, virtually never cause colonic obstruction. Table 2-12 lists some of the causes of colonic obstruction. As with SBOs the combination of radiographic findings and clinical history often allows for a presumptive diagnosis. Colonic obstruction can be diagnosed through a variety of radiologic examinations. Plain films have 84% sensitivity and 72% specificity,615 contrast enema has 96% sensitivity and 98% specificity,616 and multidetector CT has a sensitivity and specificity of about 90%617,618 in the diagnosis of colonic obstruction. Colonic obstruction is an abdominal emergency with high morbidity and significant mortality. In colonic obstruction, the mural tension is highest in the cecum where the colonic radius is greatest, according to Laplace’s law. Therefore, in the setting of a competent ileocecal valve, the cecum is most often the site of colonic ischemia or perforation when it is involved in a mechanical obstruction.615 A cecal diameter greater than 9 cm is abnormal, and in the setting of colonic obstruction the cecal wall should be evaluated for early findings of ischemia.

Neoplasms: Neoplasms are responsible for 60% of colonic obstruction.16,615,619 Unlike SBOs where neoplastic

6. Post irradiation stricture 7. Parasites (ascaris) 8. Intussusception 9. Hirschsprung disease 10. Meconium ileus 11. Anal atresia and other colonic atresia

causes of obstruction are most often due to peritoneal implants or direct extension from an extraintestinal primary malignancy, colonic obstruction is most often due to primary colon carcinoma (see Figure 2-38). The mechanism of colonic obstruction by neoplasms is the same as previously described for SBO.

Colonic volvulus: Volvulus is the second leading cause of colonic obstruction, accounting for approximately 10% to 15% of cases.16,615,619 Volvulus occurs when the bowel rotates around a single point and can, therefore, only occur in colon segments that are suspended from a mesentery. In the colon, volvulus is seen most frequently in the sigmoid colon, which accounts for 76% of cases, followed by the cecum, which accounts for 22% of cases.596,619 Rarely, portions of the transverse colon, especially the splenic flexure, can develop volvulus. The various types of colonic volvulus often have characteristic imaging appearances at conventional radiography, which is sufficient for diagnosis in a large percentage of patients.596

Sigmoid volvulus: Sigmoid volvulus is an acquired condition with an increased prevalence among those with sigmoid colonic redundancy. Colonic redundancy can be due to laxative abuse usually for chronic constipation, high-fiber diet, pregnancy, chronic institutionalization, Chagas disease, Parkinson disease, and Hirschsprung disease.596,615,621,622 Patients with colonic volvulus typically

126

Diagnostic Abdominal Imaging

present with nonspecific abdominal pain and symptoms of obstruction and are typically older than patients with cecal volvulus. Sigmoid volvulus will usually appear on abdominal radiograph as a markedly enlarged loop of bowel extending from the pelvis or right lower quadrant of the abdomen to the left upper quadrant of the abdomen. The dilated loop often extends beyond the level of the transverse colon, a finding termed the “northern exposure” sign.620,623 The dilated sigmoid loop typically sharply folds in the left upper quadrant and comes to a point at the site of the twist in the right lower quadrant. Thus, the loop of the sigmoid appears as an oval air collection with a thin central line (the wall between the ascending and descending loop of the sigmoid colon) that has an appearance similar to a coffee bean and has been termed the “coffee bean” sign (see Figure 2-85).620,624 On barium enema, contrast will abruptly terminate at the transition point of the volvulus, forming a beak-shaped cone of contrast without opacification of the dilated air-filled sigmoid colon beyond the site of obstruction (see Figure 2-85). Barium enema is additionally advantageous in that it can potentially reduce the volvulus, thereby offering both diagnosis and therapy. CT examinations will demonstrate both the abnormal position of the sigmoid colon and swirling of the mesentery at the level of the volvulus. Often coronal and sagittal reformats are useful for locating the mesenteric swirl and evaluating the orientation of the rotated bowel segment (see Figure 2-85).596

Cecal volvulus: In contrast to sigmoid volvulus, which is acquired, cecal volvulus is usually the consequence of a congenital abnormality. Specifically, this condition occurs when there is persistence of a cecal mesentery with resultant abnormal mobility that creates the potential for volvulus. This term is a misnomer, however, because most patients with cecal volvulus will in fact have torsion of the ascending colon located above the level of the ileocecal valve. Patients with cecal volvulus are younger than those presenting with sigmoid volvulus (30-60 years old). Clinically they present with symptoms of obstruction, including nausea, vomiting, constipation, and acute cramping pain. Patient history typically reveals prior abdominal surgery, presence of a pelvic mass, violent coughing, atonia of the colon, extreme exertion, unpressurized air travel, recent colonoscopy, or third-trimester pregnancy.596,625 Abdominal radiographs of cecal volvulus will reveal a massive air collection with ovoid morphology in the left upper quadrant or midabdomen that represents the dilated cecum (see Figure 2-86). Occasionally, however, the distended cecum can be displaced to other locations within the abdomen. Depending on the acuity of the obstruction, proximal obstruction may or may not be present.596,621 Of note, the reader should be careful not to overdiagnose the presence of cecal volvulus. When the cecum is loosely attached to its mesentery, air will rise into the nondependent portion and the cecum may fold on itself without torsion. As a consequence, patients with diminished bowel

motility will often have a mildly distended air-filled cecum visualized as a dilated bowel segment in the midabdomen, a phenomenon termed a “cecal bascule.” This finding does not indicate bowel obstruction and does not require surgical intervention. A diagnosis of cecal volvulus should be suggested only when the cecum is massively distended. In questionable cases, evaluation with barium enema or CT should be performed. Contrast enema evaluation will show the distal colon to be decompressed with a beaklike tapering of the ascending colon at the site of the volvulus. In most cases, contrast will fail to pass beyond the site of volvulus. Although the contrast enema findings are characteristic of cecal volvulus, most patients are initially evaluated with CT. CT examinations will demonstrate an abnormally positioned cecum, often in the upper mid- and left abdomen, that can be traced back to the level of the volvulus. The volvulus appears as an area of swirling of the bowel and its mesentery, a finding known as the “whirl” sign.596,626

Diverticulitis: Diverticulitis is the third leading cause of colonic obstruction, accounting for 10% of cases (see Figure 2-75).16,615,619 Although the third most common cause of colonic obstruction, colonic obstruction is a rare presentation of diverticulitis. Instead, patients with diverticulitis usually present with fever and abdominal pain but without evidence of obstruction. In 1 study of diverticulitis over a 20-year period, 7% of cases (29/422) presented with symptoms of obstruction requiring surgical intervention.211 Diverticulitis is discussed earlier in this chapter under the heading Focal Polyps or Masses of the Colon.

Other pericolonic infection: Rarely, other causes of pericolonic abscess will also cause colonic obstruction.627,628

Inflammatory bowel disease: Although SBO is a common complication of IBD, colonic obstruction is relatively rare. In 1 surgical series of 306 patients, 16% required surgical relief of SBO whereas none required surgery for colonic obstruction.629 In a second surgical series of patients with ulcerative colitis, only 3% (2/72) patients required surgery for colonic obstruction.630 The outcome of obstruction is dependent on the underlying cause. Acute obstructions are usually caused by inflammatory narrowing and will respond to medical therapy. However, chronic obstructions are usually due to strictures and will often require surgical intervention.631

Fecal impaction: Fecal impaction is a common problem in the elderly and can also be found in individuals with spinal cord injury, cystic fibrosis, and Hirschsprung disease. However, only rarely is fecal impaction sufficiently severe to cause colonic obstruction. When it occurs, it is most characteristically seen in the very elderly with a mean age of 79  years.632 Patients typically complain of abdominal pain and distention and are frequently bedridden or have chronic mental status changes. Abdominal x-rays will demonstrate dilation of the small bowel and colon. Air-fluid levels are uncommon; however, abundant feces are demonstrated within the rectum and colon.632

Chapter 2 Imaging of the Bowel 127

A

B

C

D

Figure 2-85 Sigmoid Volvulus in 3 Patients A-D. This 91-year-old woman was transferred from a nursing home because of crampy abdominal pain. A. Left lateral decubitus film shows an air-fluid level (black arrow) within a markedly distended loop of colon. This loop resembles a coffee bean (white arrows). This appearance is diagnostic of a cecal volvulus. B. Barium enema from a different patient shows

barium within the rectum (black arrows) and an occlusive twist at the rectosigmoid junction. C and D. This 20-year-old man was chronically presented with abdominal pain, nausea, and vomiting for 2 days. Coronal reconstructions from a noncontrast abdominal CT demonstrate a massively distended sigmoid colon (arrowheads) and a swirl of mesenteric vessels (arrow) in the left lower quadrant, diagnostic of a sigmoid volvulus.

128

Diagnostic Abdominal Imaging

A

B

Figure 2-86 Cecal Volvulus This 66-year-old man was bedridden as a result of several strokes when he complained of abdominal pain. A. Erect radiograph shows an air-fluid level in a large air collection in the

right lower quadrant. B. Supine radiograph showed a 30-cm air collection in the midabdomen. There are also minimally dilated loops of small bowel in the pelvis below the air collection. This appearance is diagnostic of a cecal volvulus.

Meconium ileus and distal ileal obstruction:

migration of colonic ganglion cells during gestation.634 As a consequence, varying lengths of the distal colon remain constricted, causing functional colonic obstruction. The disease usually involves the rectosigmoid colon but can affect the entire colon and, rarely, the small intestine. Approximately 80% of patients will present in infancy with difficult bowel movements, poor feeding, poor weight gain, and progressive abdominal distention. Rarely, patients can present with persistent, severe constipation later in life. Imaging studies will show the typical features of colonic obstruction, although contrast enemas can be normal in the first few months of life.634

Meconium ileus represents obstruction at the terminal ilium or colon due to impaction of abnormally thick, tenacious, sticky meconium. It is one of the more common causes of neonatal intestinal obstruction, and the majority of cases are seen in patients with cystic fibrosis. Patients typically present with failure to pass meconium in the first 24  hours of life associated with bilious emesis and abdominal distention. Abdominal x-rays will show abnormal distention of multiple bowel loops. Barium enema will demonstrate a microcolon with obstruction of the terminal ilium. Distal ileal obstruction, previously known as meconium ileus equivalent, represents the increased incidence of fecal impaction and obstruction in children and adults with cystic fibrosis. This disease has a mechanism similar to that of meconium ileus.

Intussusception: In most cases, intussusception occurs within the small bowel and causes SBO. However, rarely, intussusception can involve the colon and be a cause of colonic obstruction.633 Intussusception is described most completely earlier in this chapter under the heading Causes of SBO.

Hirschsprung

disease: Hirschsprung disease, also called congenital megacolon, is caused by failure of

Colonic atresia: Colonic atresia is an uncommon cause of colonic obstruction in neonates and the least common site of bowel atresia. A 25-year review of intestinal atresia from a single pediatric center demonstrated only 277 cases of intestinal atresia and only 21 cases of colonic atresia.611 Patients presented with bilious emesis, abdominal distension, and failure to pass meconium through the anus. Nearly one-quarter of patients (4/21) will have concomitant gastroschisis.611

Other uncommon causes of colonic obstruction: Other uncommon causes of colonic obstruction include ischemic strictures,635 gallstone ileus,636 bezoars,637 endometriosis,638 pancreatitis,639 radiation-related strictures,640 and diaphragmatic hernias.641

Chapter 2 Imaging of the Bowel 129 Table 2-13. Causes of Adynamic Ileus 1. Recent abdominal or thoracic surgery 2. Peritonitis 3. Drugs a. Narcotic b. Anticholinergics c. Other 4. Metabolic derangement a. Hypokalemia b. Hyponatremia c. Hypomagnesemia d. Hypothyroidism e. Acidosis f. Hypothermia 5. Abdominal and thoracic inflammatory processes a. Appendicitis b. Cholecystitis c. Pancreatitis d. Diverticulitis e. Pelvic inflammatory disease f. Pneumonia g. Other 6. Thoracic and abdominal trauma 7. Head trauma and neurosurgical procedures 8. Intestinal ischemia

Adynamic ileus Adynamic ileus is a common problem among hospitalized patients. The exact physiologic mechanisms of adynamic ileus remain obscure but a variety of factors appear to increase the incidence of adynamic ileus (Table 2-13). Most common of these is abdominal and thoracic surgery.642,643 Inflammation of the peritoneum irritates the entire bowel and causes the cessation or diminution of peristalsis. Other causes of peritonitis including

peritoneal infection and inflammation will also cause a generalized adynamic ileus. Medications, usually narcotics but occasionally anticholinergic medications along with other medications, are also a common cause of adynamic ileus.644-646 Finally, metabolic derangements such as hypokalemia and hyponatremia alter the normal metabolic activity of the bowel will also affect bowel peristalsis and cause adynamic ileus.642 All of the causes of ileus that have been previously mentioned will affect the entire bowel relatively uniformly and result in a “generalized” adynamic ileus. In generalized adynamic ileus, the bowel becomes dilated and fluid filled as a result of diminished peristalsis. Thus, generalized adynamic ileus is distinguished from bowel obstruction by demonstrating the presence of gas throughout the bowel, including the rectum (see Figure 2-76). It is not uncommon for the rectum to be collapsed in the supine and erect radiographs of the abdomen because in both positions it is dependent and therefore gas will rise out of it. Cross-table prone view of the rectum and left lateral decubitus views of the pelvis image the rectum in a nondependent position and often will demonstrate air in the rectum in patients with generalized adynamic ileus. Either of these views are simple methods of distinguishing a distal colonic obstruction from an adynamic ileus without the use of cross-sectional imaging. Occasionally, adynamic ileus will involve only a focal region of the bowel. This is most often seen in adynamic ileus related to inflammatory disease of the abdomen or pelvis. Inflammatory process tends to preferentially affect peristalsis of nearby loops of bowel. Thus, in appendicitis the ileal loops can be dilated without dilation of other bowel segments and in pancreatitis, the transverse colon can be dilated without dilation of other bowel segments (see Figure 2-87). Therefore, focal adynamic ileus can mimic both SBO and colonic obstruction. The only potential clue to distinguishing between a focal adynamic ileus and smallbowel or colonic obstruction is appropriate clinical history.

Acute

colonic pseudo-obstruction: Acute colonic pseudo-obstruction (ACPO) is an uncommon form of ileus characterized by massive colonic dilation. The mechanism

Imaging Notes 2-22. Bowel Gas Patterns Dilation and air-fluid levels Dilation stops in small bowel Dilation stops in colon Dilation throughout bowel

Small-bowel obstruction Focal adynamic ileus (uncommon) Colonic obstruction Focal adynamic ileus (uncommon) Generalized adynamic ileus

Normal caliber and air-fluid levels

Enteritis/colitis

Normal caliber and no air-fluid levels

Normal/Airophagia

130

Diagnostic Abdominal Imaging

A

B

Figure 2-87 Colon Cut-off Sign This 40-year-old man recently experienced a seizure. He now complains of boring abdominal pain. A. Erect radiograph demonstrates multiple air fluid levels. B. Supine radiograph shows dilated small bowel with multiple thin folds (black arrow) and dilated transverse colon and splenic flexure (white arrowheads). The descending colon is collapsed and contains

a small amount of barium and appears as a thin string (white arrow). This pattern will usually indicate a colonic obstruction. However, further evaluation confirmed a diagnosis of acute pancreatitis. This is representative of a focal adynamic ileus due to local inflammation of the transverse mesocolon and jejunal small-bowel mesentery caused by the pancreatitis. This is known as the “colon cut-off” sign.

of pseudo-obstruction is most likely multifactorial because of uncoordinated, nonperistaltic, or attenuated colonic muscle contractions originating from either increased thoracic sympathetic stimulation or decreased sacral parasympathetic activity. Patients usually present with mild, diffuse, abdominal discomfort and distension as well as minimal systemic toxicity that develops slowly over several days.647 Passage of stool or gas is absent in 50% of patients, but patients occasionally have diarrhea.621 Most often, pseudoobstruction presents in hospitalized patients with underlying disorders, including postoperative status, recent child birth, cardiopulmonary disease, nonoperative trauma, drugs, and electrolyte disorders.642,648 ACPO can only be diagnosed after exclusion of a mechanical large-bowel obstruction using barium enema or CT. Abdominal radiographs typically demonstrate massive colonic dilation, which is greater in the right colon than the left colon. This pattern of colonic dilation favors the diagnosis of ACPO rather than colonic obstruction.227 When visualized, a colonic cut-off point favors the diagnosis of colonic obstruction; however, 40% of patients with

ACPO appear to have a cut-off point most often at the splenic flexure, descending, or sigmoid colon.616 Up to 80% of patients with ACPO have simultaneous dilation of small bowel.649 Therefore, contrast enemas, CT, or colonoscopy is usually necessary to make this diagnosis. Contrast enemas, using either barium or water-soluble contrast, or abdominal CT scans are often performed to exclude colonic obstruction. Abdominal CT has a sensitivity of 96% and specificity of 98% for ACPO.616,650 CT has the additional advantages of allowing more accurate measurement of bowel diameter and a better appraisal of the condition of the mucosa to detect coexisting inflammation and ischemia. Most patients with ACPO improve within 72 hours with supportive care. However, approximately 3% to 15% of patients with ACPO will develop a major complication, including ischemia, sepsis, and perforation. Perforation, when it occurs, is typically located in the cecum, which has the greatest level of wall tension. Surgery is reserved for patients with clinical deterioration or with evidence of colonic ischemia or perforation.642

Chapter 2 Imaging of the Bowel 131

UNIQUE FINDINGS RELATED TO THE BOWEL GI bleeding, bowel perforation, pneumoperitoneum, dissection of gas in the wall of the bowel, and portal venous gas are uncommon but important complications of bowel pathology. There are also 2 unusual disorders of the bowel that can cause abdominal pain, called epiploic appendagitis and segmental omental infarction. These disorders will be reviewed.

Gastrointestinal Bleeding Bleeding from the GI tract is a moderately common clinical problem that can be caused by a relatively wide range of disorders, including inflammation, neoplasms, dilated submucosal veins, vascular malformations, traumatic tears, and coagulopathies (see Table 2-14).651 Approximately 75% of GI bleeding will cease spontaneously; however, bleeding can recur in up to 25% of cases.652 Mortality rates related to GI bleeding are approximately 3% to 5%, but can be as high as 23% with massive bleeding or bleeding that recurs after hospitalization.653 GI bleeding is commonly divided into upper GI causes and lower GI causes. Upper GI bleeding indicates a source proximal and lower GI bleeding indicates a source distal to the ligament Treitz. Upper GI bleeding is more common than lower GI bleeding occurring in approximately

Table 2-14. Causes of Gastrointestinal Bleeding 1. Upper a. Inflammation i. Esophagitis ii. Gastritis/duodenitis/peptic ulcers b. Neoplasms c. Esophageal and gastric varices d. Vascular malformations and vascular ectasia e. Superior mesenteric artery syndrome f. Bleeding from pancreatic duct g. Bleeding from the bile duct 2. Lower a. Colitis/enteritis i. Inflammatory bowel disease ii. Ischemic iii. Infectious iv. Postradiation b. Neoplasms c. Hemorrhoids d. Diverticular disease i. Diverticulosis ii. Diverticulitis e. Angiodysplasia f. Anal fissure g. Coagulopathy h. Vasculitis

100 per 100 000 individuals and 20 per 100 000 individuals, respectively.651 The most common causes of upper GI bleeding are peptic ulcers accounting for approximately 35% to 60% of cases, followed by varices (4%-31%) and Mallroy-Weiss tears (4%-13%). However, the precise cause of upper GI bleeding can be unknown in up to 25% of cases.651 The most common cause of lower GI bleeding is diverticulosis, accounting for 33% to 50% of cases, followed by vascular ectasia (8%-20%), neoplasms (19%), and colitis (18%).231 Meckel diverticulum is the most common cause of lower GI bleeding in children.651 Bleeding can present in 1 of 5 ways: (1) hematemesis, (2) melena, (3) hematochezia, (4) positive fecal occult blood tests, or (5) symptoms of blood loss or anemia such as lightheadedness and dyspnea.651 In most cases, hematemesis will indicate an upper GI cause and hematochezia will indicate a lower GI cause. Melena will more often indicate an upper GI cause and positive occult blood tests and symptoms of blood loss or anemia can be seen with both upper and lower causes. Endoscopy and colonoscopy are the primary methods of diagnosis of GI bleeding.654 However, imaging can be an important adjunct test in patients where endoscopy and colonoscopy fail to identify the source of bleeding. This is most common in situations where the bleeding is intermittent and recurrent. In the past, Technetium 99m-labeled red cells [Tc-99m red blood cell (RBC)] and Technetium 99m (Tc-99m) sulfur colloid scintigraphy were the principal imaging studies used in determining the site of occult bleeding. More recently, CT examinations using angiographic technique is used to identify the presence and the site of bleeding. Tc-99m RBC or Tc-99m sulfur colloid scintigraphy can be used to detect and localize GI bleeding. These agents circulate in the bloodstream. Extravasation of blood into the lumen of the bowel will result in increased activity in the distribution of the bowel. Location of the first site of activity indicates the site of bleeding (see Figure 2-88). Tc-99m RBC scintigraphy is approximately 93% sensitive and 95% specific for the detection of active arterial or venous bleeding and can detect bleeding rates as low as 0.2 mL/min.655 Tc-99m RBC scintigraphy is additionally advantageous in that delayed scans can be performed up to 24 hours after administration of the radioisotope. This allows for the detection of intermittent bleeding through repeat evaluation at discrete intervals after initial scans. However, blood can travel in either a retrograde or antegrade direction between the time of bleeding and the time of imaging, which most likely accounts for false localization rates of 22%.656 Furthermore, scintigraphic examinations cannot identify structural lesions causing the bleeding. Computed tomographic angiography is the current accepted method for the detection of GI bleeding.657-661 Examinations are performed following the administration of IV contrast, without the use of oral contrast. Extravasation of contrast medium results in hyperattenuating contrast in the lumen of the bowel and indicates the site of bleeding (see Figure 2-89). A meta-analysis of 9 studies with 198 patients

132

Diagnostic Abdominal Imaging

Figure 2-88 Gastrointestinal Bleeding due to a Radiation Ulcer This 66-year-old man had received radiation therapy for a cholangiocarcinoma and presented with melena, anemia, and hypotension. Images from a 99mTc red blood cell study show a faint blush of activity (arrow) in the right upper quadrant at 22 minutes after injection that progressively increased at 36 and 48 minutes indicating the presence of active bleeding. Colonoscopy revealed a radiation-induced ulcer in the hepatic flexure.

of CT angiography demonstrated a pooled sensitivity of 89% and specificity of 85% for the diagnosis of GI bleeding.654 In addition to identifying the site of bleeding, CT offers the ability to identify a structural cause of the bleeding in some cases. Mesenteric angiography can also be used to detect GI bleeding when the bleeding rates are greater than 0.5  mL/min (see Figure 2-89). Unfortunately, angiography only has a sensitivity of 40% to 86%.652,662 In general, angiography is primarily used as a therapeutic intervention through transcatheter embolization and not a diagnostic study.

Pneumoperitoneum and Bowel Perforation Pneumoperitoneum, often referred to as “free air,” is defined by the presence of air within the peritoneal space that is outside the bowel lumen.

on optimal-quality radiographs. Although erect plain films will usually show resolution of pneumoperitoneum within 1 week, imaging identification of pneumoperitoneum can persist for several weeks following surgical intervention depending on the volume of air and absorption rate of the peritoneum.665,666 This is especially true of CT scans, which can show pneumoperitoneum for up to 3 weeks following abdominal procedures.239 In general, a smaller amount of pneumoperitoneum is expected in patients who undergo laparoscopic procedures, compared to open resection.665 Regardless of the initial volume, the quantity of air should become progressively less with serial examinations.664 If the amount of air is seen to increase over time, this should raise suspicion for the development of a complication of the recent procedure. In the absence of a sterile iatrogenic cause for pneumoperitoneum, the recognition of free air is a cause for concern because the majority of noniatrogenic cases are associated with perforated hollow viscus and will require surgical treatment. In 5% to 15% of spontaneous cases, pneumoperitoneum will be due to causes other than perforation and does not require emergent surgery.663,667,668 Perforation of a hollow viscus can be due to a variety of causes and differs depending on the location of the affected bowel and the clinical setting. Gastric or duodenal ulcers are among the most common causes of bowel perforation. Colonic perforation can be due to obstruction (benign or malignant), ischemia, adynamic ileus, toxic megacolon, and inflammatory conditions, including appendicitis, diverticulitis, tuberculosis, and necrotizing enterocolitis. Although acute diverticulitis and appendicitis are both associated with intestinal perforation, free air is seldom observed with either process. This is thought to be due to containment of the free air associated with diverticulitis by the associated inflammatory reaction as well as the very small amount of gas contained within the appendix.666 Pneumoperitoneum in the setting of trauma should raise the possibility of bowel perforation, which can be seen in both blunt and penetrating trauma. In premature infants, the most common cause of bowel perforation is necrotizing enterocolitis.

Causes of pneumoperitoneum The most common causes of pneumoperitoneum are iatrogenic, including laparotomy and laparoscopy, percutaneous gastrostomy, vigorous respiratory resuscitation, peritoneal dialysis, colonoscopy, and other abdominal interventions (Table 2-15). The majority of iatrogenic causes of pneumoperitoneum leave the peritoneum sterile and without a source of further gas. In these cases, pneumoperitoneum will resolve spontaneously, without intervention and without complication. Free air is recognized on left lateral decubitus abdominal radiographs 3 days after surgery in 44% to 60% of patients who undergo open surgery and 25% of patients who undergo laparoscopic surgery.663,664 By day 6, 8% of patients have persistent evidence of pneumoperitoneum

Imaging Notes 2-23. Evaluation of Pneumoperitoneum • Most cases of pneumoperitoneum will be iatrogenic • Search first for causes of abdominal intervention ▪ Recent surgery ▪ Percutaneous gastrostomy ▪ Peritoneal dialysis ▪ Other • In the absence of an iatrogenic cause, bowel perforation should be excluded • Most common causes of bowel perforation ▪ Peptic ulcer ▪ Neutropenic colitis

Chapter 2 Imaging of the Bowel 133

A

B

C

D

Figure 2-89 Gastrointestinal Bleeding Following Polypectomy This 63-year-old man had a polypectomy performed 1 week prior. He presented with hematochezia. A and B. CT angiogram performed without oral contrast shows a small foci of contrast (white arrows) in the ascending colon that had not been present

on the precontrast images, indicating the presence of active bleeding. C and D. SMA angiogram performed later that day confirms the site of bleeding (black arrow) in the hepatic flexure. The feeding arteries were subsequently embolized.

Pulmonary barotrauma is a rare cause of pneumoperitoneum. This is caused by high pressures in the bronchial tree that results in pulmonary interstitial emphysema, which dissects into the mediastinal tissues causing pneumomediastinum and subcutaneous emphysema in the neck. Rarely, air travels in an inferior rather than superior direction and dissects into the retroperitoneum, which directly communicate with the mediastinum. Air can then perforate the parietal peritoneum and the peritoneal space.665 In many cases, it can be impossible to distinguish barotrauma as a cause of pneumoperitoneum from bowel perforation. In general, when peritoneal signs are absent and clinical suspicion for a ruptured viscus is low, conservative management with clinical and imaging follow-up is recommended. However, if there is an increased clinical suspicion for bowel perforation, fluoroscopic evaluation with water-soluble contrast material administered via oral, rectal, or enterostomy tube may be necessary.663 Other miscellaneous causes of pneumoperitoneum include medications, pneumatosis coli, or pneumatosis

intestinalis (discussed below) and female genital tract-related causes.118,120 Medications leading to gastric ulcers such as corticosteroids and nonsteroidal anti-inflammatory drugs can be a cause of pneumoperitoneum. The fallopian tubes communicate with the peritoneum at one end and with the uterus at the other. Forceful transmission of air and fluid through the vagina related to douching, sexual intercourse, insufflation, water skiing, and other mechanisms can result in the transmission of air from the vagina into the peritoneum.

Imaging manifestations of pneumoperitoneum Pneumoperitoneum can be detected by chest and abdominal radiographs, CT, and MRI.

Radiographic detection of pneumoperitoneum: Air within any space will rise to the nondependent surface of the space. Pneumoperitoneum can be visualized in profile via a horizontal beam film, either an erect chest or abdomen radiograph that includes the diaphragms or a left

134

Diagnostic Abdominal Imaging

Table 2-15. Causes of Pneumoperitoneum 1. Iatrogenica a. Laparotomy and laparoscopy b. Percutaneous gastrostomy c. Peritoneal dialysis d. Paracentesis e. Breakdown of surgical anastomosis f. Other abdominal interventions 2. Bowel perforation a. Peptic ulcera b. Penetrating and blunt traumaa c. Small bowel and colonic infections i. Neutropenic colitis (typhlitis)a ii. Diverticulitis iii. Necrotizing enterocolitis (infants) iv. Appendicitis v. Other bowel infections d. Ruptured diverticulum i. Colonic pulsion diverticula ii. Meckel diverticulum iii. Congenital diverticula e. Ischemic bowel f. Inflammatory bowel disease i. Toxic megacolon g. Bowel obstruction h. Sigmoidoscopy/colonoscopy i. Adynamic ileus 3. Pulmonary barotraumas 4. Pneumatosis intestinalis 5. Passage of air through the female genital tract a. Intercourse b. Douching c. Other aMost common causes.

lateral decubitus radiograph of the abdomen that includes the nondependent surface of the abdomen and the liver surface. It is generally believed that a left lateral decubitus radiograph is the most sensitive radiographic projection for the detection of pneumoperitoneum.664 On erect films, the pneumoperitoneum will be seen as a thin or thick black band outlining the undersurface of the diaphragm (see Figures 2-90 and 2-91). On decubitus films, air will be seen as a thin or thick black band outlining the undersurface of the abdominal wall that is easiest appreciated when outlining the liver surface on a left lateral decubitus examination. Supine and semierect films are much less sensitive for the detection of pneumoperitoneum. On semierect radiographs, air will typically rise to the central and anterior surface of the diaphragm, just beneath the heart. This is seen as a bandlike lucency across the central superior abdomen, inferior to the cardiac silhouette (see Figure 2-90).

Pneumoperitoneum can be seen on either an upright or left lateral decubitus view following the injection of as little as 1 to 2 mL of free air into the peritoneal cavity. Detection of small volumes of free air is facilitated by prolonged positioning of least 10 minutes prior to exposure.669 Failure to maintain the patient in the erect or decubitus position for at least 10 minutes increases the likelihood that air will remain trapped in the dependent peritoneal recesses and will not be visualized. There are several subtle signs of pneumoperitoneum that can be detected on supine radiographs. In most cases, these signs require large volumes of peritoneal gas to be detected. However, 1 study of the ability of supine films to detect pneumoperitoneum suggested that there was evidence of pneumoperitoneum on 59% (26/44) of supine examinations.670 The right upper quadrant sign is thought to be the most sensitive sign of free air on supine radiographs, seen in 41% to 49% of cases.666,670,671 This sign is characterized by the presence of triangular or linear lucencies located along the diagonal surface of the liver edge. The linear collections are thought to represent gas in the right subhepatic space, whereas the triangular collections are thought to represent gas in Morrison pouch.670 The second most commonly seen sign on the supine radiograph is the Rigler sign found in 17% to 32% of supine radiographs of patients with pneumoperitoneum.666,670 When a bowel loop is filled with air, the air outlines the internal border of the bowel wall. Thus, the lumen of the bowel is seen as lucent and the tissues external to the bowel appear opaque. If there is air in both the lumen of the bowel and within the peritoneum, external to the bowel, then the bowel wall will appear as a thin white line bordered by the lucent bowel lumen and the lucent peritoneum; this appearance is called Rigler’s sign. Studies suggest that at least 750 mL to 1L of free air is necessary to identify this finding (see Figure 2-91).666 The falciform ligament is a band of peritoneum that is attached to the surface of the left lobe of the liver that contains the umbilical vein. In the normal individual, the falciform ligament is indistinguishable from the other soft tissues of the abdomen. However, in the supine position, free abdominal air will rise to the anterior surface of the abdomen and can outline the 2 sides of the falciform ligament. In this situation, the falciform ligament appears as a white band overlying the left lobe of the liver known as the falciform ligament sign (see Figure 2-91). In some situations, the peritoneal cavity may appear as an oval gas shadow known as the “Football sign,” whereby the falciform ligament appears as the laces of the football. The reader should be aware that both of these findings require extensive peritoneal gas before they are visualized (see Figure 2-91).666 Although abdominal radiographs are advised as the initial method of evaluating for pneumoperitoneum, they are of limited sensitivity and specificity. Studies suggests that a combination of a supine and left lateral radiographs have a sensitivity of 47% to 59%670,672 and erect films have

Chapter 2 Imaging of the Bowel 135

A

B

Figure 2-90 Pneumoperitoneum on Erect and Semierect Radiographs A. This 65-year-old woman underwent right hemicolectomy for a colon cancer 3 days previously. Posteroanterior chest radiograph demonstrates a thin sliver of air beneath the right hemidiaphragm. This is the typical appearance of pneumoperitoneum on an erect radiograph. This air is related to the recent abdominal procedure and is an expected

postoperative finding, without clinical significance. B. This 74-year-old woman was hospitalized with chronic heart failure following aortic valve repair. This radiograph demonstrates a bandlike lucency in the superior midabdomen. This is a common appearance of pneumoperitoneum on a semierect chest radiograph because in the semierect position the anterior portion of the diaphragm is the most nondependent portion of the abdomen.

a sensitivity of 38% for the detection of pneumoperitoneum when CT is used as the reference standard.673 Radiographs are particularly poor at detecting small amounts of free air and are limited in obese patients.670,673 Even when pneumoperitoneum is detected on radiographs, the etiology of this free air is usually uncertain.

adjacent to the site of suspected perforation. Gastric and duodenal ulcers are not typically associated with an adynamic ileus, whereas colonic perforation causes a bacterial peritonitis that causes almost immediate ileus involving both the small and large bowel.

CT detection of pneumoperitoneum: Computed tomo-

The term pneumatosis intestinalis means air within the bowel wall. It is important to understand that pneumatosis can be attributable to both benign causes and life-threatening causes, even in the presence of some abdominal symptoms (see Table 2-16).674-679 Pneumatosis intestinalis is an imaging finding, not a diagnosis and requires clinical correlation in order to differentiate the etiology and prognosis.680 There are 2 dominant theories as to how gas enters the wall of the bowel. The mechanical theory postulates that gas enters the bowel wall either from the enteric lumen or from the lungs via the mediastinum because of increased pressure. This would explain the etiology of pneumatosis in the setting of bowel obstruction and emphysema. The bacterial theory postulates that gas produced by bacteria enters the submucosa through mucosal rents or increased mucosal permeability that produces gas in the bowel wall.681 Bacterial overgrowth in the GI tract from a variety

graphic scans have been shown to detect as little as 2.5 mL of pneumoperitoneum. As a consequence, CT scans are the gold standard in the evaluation of pneumoperitoneum.672 Gas will predominantly collect in the nondependent portions of the peritoneum anterior to the liver just beneath the anterior abdominal wall, but can also be seen within the mesentery, lesser sac, Morrison pouch, and immediately underlying the rectus abdominis muscles.673 This air is best visualized using lung windows. CT will in some cases identify the cause of pneumoperitoneum, especially in the setting of bowel perforation where often the site of perforation can be discerned (see Figure 3-62). Helpful clues to determining the source of a suspected bowel perforation include (1) the location of the gas and (2) the presence or absence of an associated adynamic ileus. The largest volume of air is often located

Pneumatosis intestinalis and portal venous gas:

136

Diagnostic Abdominal Imaging

A

B

C

D

Figure 2-91 Pneumoperitoneum on the Supine Film This 50-year-old man underwent sigmoidoscopy when he complained of severe abdominal pain. A. Erect film of the abdomen shows a massive amount of air beneath both hemidiaphragms. B. Supine radiograph shows an air-filled colon and 2 subtle signs of pneumoperitoneum on a supine radiograph. C. Magnified view of the right midabdomen shows the hepatic flexure. Note how the

colon wall is seen as a thin white line (arrowheads). This finding is known as “Rigler” sign and is seen because there is air within the colon and also outside of the colon (pneumoperitoneum) that outline both sides of the bowel wall. D. Magnified view of the right upper quadrant shows air in the peritoneum outlining the borders of the liver (black arrowheads) and the falciform ligament (white arrowheads), creating the “football” sign.

Chapter 2 Imaging of the Bowel 137 Table 2-16. Causes of Pneumatosis Intestinalis 1. Life-Threatening Causes: a. Intestinal Ischemia b. Mesenteric vascular disease c. Intestinal obstruction (especially strangulation) d. Infectious enteritis/colitisa e. Toxic megacolon f. Trauma g. Organ transplantation (especially bone marrow transplant)a h. Collagen vascular diseasea 2. Benign Causes: a. Pulmonary disease i. Asthma ii. Chronic obstructive pulmonary disease iii. Cystic fibrosis iv. Positive end expiratory pressure (ventilatorassociated barotraumas) b. Systemic i. Sclerodermaa ii. Systemic lupusa iii. AIDS c. Intestinal causes i. Infectious enteritis/colitisa ii. Intestinal obstructiona iii. Inflammatory bowel disease 1. Crohn disease 2. Ulcerative colitis d. Medication related i. Corticosteroids ii. Chemotherapy e. Organ Transplant i. Graft vs host disease ii. Bone marrowa iii. Solid organ transplant (kidney, liver) f. Primary i. Idiopathic (usually involves colon) ii. Pneumatosis cystoides intestinalis aSome causes are listed under both benign and life threatening.

of causes can lead to excessive hydrogen gas production, bowel distention, and subsequently, dissection of intraluminal hydrogen gas into the bowel wall.124-127,682 Prior literature has demonstrated that the gas collections in the bowel wall have a hydrogen content up to 50%,681 which supports the bacterial theory because hydrogen is a byproduct of bacterial metabolism. However, the bacterial theory is not supported by the finding that long-standing pneumoperitoneum associated with pneumatosis rarely results in peritonitis.683 It is likely that the cause of pneumatosis varies depending on the clinical condition and reflects a combination of both theories. Patient presentation is rarely helpful in the diagnosis of pneumoperitoneum and often reflects the consequence

of the underlying cause or condition.684 However, because of benign causes, such as pulmonary disease, patients are usually asymptomatic or may have mild abdominal discomfort.122,128-130 Conversely, the presence of peritoneal signs can imply intestinal perforation from life-threatening causes.675 Abdominal radiography and CT are both used for the diagnosis of pneumatosis intestinalis. However, CT is more sensitive than radiography and should be used when there are questionable findings on plain radiographs.675,685-689 On CT examinations, pneumatosis intestinalis is best identified using lung windows. Pneumatosis typically adopts 1 of 3 appearances on radiographs and CT images: linear, bubbly, or circular patterns of gas collection. In general, the circular form of pneumotosis intestinalis (PI) is usually benign and is classically associated with pneumatosis cystoides intestinalis—a subset of pneumatosis intestinalis that almost always occurs in the colon and its mesentery (see Figure 2-92).675,686,690 Conversely, linear or bubblelike pneumatosis intestinalis can be due to either benign or life-threatening causes (see Figure 2-93).675 However, these are general guides rather than absolute rules, and correlation with patient symptoms is imperative in distinguishing between benign and life-threatening causes. The presence or absence of other findings on CT scans can be helpful in distinguishing between the likelihood of benign versus life-threatening causes. The presence of a normal bowel wall will usually indicate a benign cause. Conversely, the presence of bowel wall thickening, either absent or intense mucosal enhancement, dilated bowel, arterial or venous occlusion, ascites, and hepatic portal or portomesenteric venous gas raise suspicion for pneumatosis due to a life-threatening cause.675,691,692 Pneumatosis intestinalis confined to a specific vascular territory should raise suspicion of underlying ischemia.252 The extent of pneumatosis is often inversely related to the severity of the disease.674,693 This is probably because life-threatening processes such as ischemia may develop so rapidly that there is less time to form pneumatosis than in less serious conditions. The location of pneumatosis in the colon versus the small bowel or stomach cannot be used

Imaging Notes 2-24. Features Suggesting LifeThreatening Pneumatosis Intestinalis 1. Bowel wall thickening 2. Absent or increased mucosal enhancement 3. Bowel dilation 4. Arterial or venous occlusion 5. Ascites 6. Portal venous gas

138

Diagnostic Abdominal Imaging

A

B

Figure 2-92 Pneumatosis Cystoides Intestinalis This 63-year-old man with Crohn disease was status post ileocolica anastamosis and receiving corticosteroids when he complained of increased flatus. CT images in the axial (A) and coronal (B) planes using lung windows demonstrate both linear (white arrowhead) and bubbly (black arrowhead) pneumatosis

involving essentially the entire colon. There are also scattered foci of pneumoperitoneum (long white arrow). Despite the extensive nature of the pneumatosis, the patient was asymptomatic, with stable vital signs and benign physical examination at the time of the scan. Following the scan, he was admitted for observation and discharged 2 days later without intervention in stable condition.

to distinguish the cause of the pneumatosis and does not indicate its clinical significance.681,690,693 Portal venous gas is classically visualized as branching linear or nodular gas collections in the peripheral aspect of the liver (see Figure 2-93). Further, CT is more sensitive than radiography in the detection of hepatic portal and portomesenteric venous gas (see Figure 2-94).675,687,691,694 When visualized in the setting of pneumatosis, this finding is most often due to bowel ischemia in adults and necrotizing enterocolitis in infants. In this setting, portal venous gas reflects the passage of gas from the bowel wall into the mesenteric veins and then into the portal venous system.692,695-697 Examples of nonischemic causes of hepatic portal or portomesenteric venous gas include mesenteric abscess formation, portomesenteric thrombophlebitis, sepsis, abdominal trauma, severe enteritis, cholangitis, chronic cholecystitis, pancreatitis, IBD, diverticulitis, and following GI surgery or liver transplantation.675,679,688,691,696,698-702 There have also been several reports of gas in the portal vein in conditions with benign pneumatosis,674,687,693 again emphasizing the importance of clinical correlation.

Ultrasound with color Doppler is very sensitive and specific, both in early detection and follow-up of portal venous gas. Gray-scale US typically demonstrates highly echogenic intravascular particles that move in the direction of blood flow. Alternatively, this gas can present as poorly defined patchy hepatic gas accumulation in the nondependent hepatic parenchyma (see Figure 2-94). Color Doppler tracings will often demonstrate bidirectional spikes superimposed on the background of a normal spectral portal vein tracing. These bidirectional spikes are thought to represent artifact caused by acoustic reflections from the mobile intravascular gas bubbles (see Figure 2-94).133-135 The main drawback of US is that when portal venous gas is visualized, further exploration with CT is warranted to evaluate for pneumatosis intestinalis. Recent literature has suggested that patients with isolated pneumatosis intestinalis are more likely to have partial or nontransmural ischemia, whereas patients with both pneumatosis and portomesenteric venous gas are more likely to have transmural infarction.40,136 Overall survival is higher among patients with nontransmural intestinal

Chapter 2 Imaging of the Bowel 139

A

B

C

D

Figure 2-93 Pneumatosis Intestinalis in 2 Patients A. This 2-week-old premature infant had fevers, abdominal distension, and increased gastric residuals. AP view of the abdomen demonstrates linear pneumatosis intestinalis (arrows) in the right lower quadrant. There are also small bubbles of gas in the right upper quadrant characteristic of portal venous gas. These findings are indicative of severe necrotizing enterocolitis. B-D. This 56-year-old man had received a liver transplant several months

prior to presenting with abdominal pain. B. Supine view of the abdomen demonstrates fine curvilinear collections of gas outlining the wall of the transverse colon (arrows), typical of pneumatosis intestinalis. CT images in standard soft tissue windows (C) and lung windows (D) confirm the presence of extensive pneumatosis intestinalis (arrows). Surgical exploration confirmed the presence of intramural air within an otherwise normal colon and so the bowel was left intact without subsequent complication.

ischemia compared with those patients with transmural intestinal infarction.

appendicitis, cholecystitis, and diverticulitis. However, both of these diagnoses demonstrate a unique appearance on CT, which allows confident diagnosis of the disorder.

UNIQUE FOCAL LESIONS OF THE BOWEL

Epiploic Appendagitis

Epiploic appendagitis and idiopathic segmental omental infarction are 2 rare complications of the mesenteric fat that present with acute abdominal pain due to infarction.377 The clinical presentation of these diagnoses can mimic more common inflammatory conditions such as

The epiploic appendages are small pedunculated fatty projections that protrude from the antimesenteric border of the colon into the peritoneal cavity and extend from the cecum to the rectosigmoid junction. These appendages are approximately 1 to 2 cm thick and 2 to

140

Diagnostic Abdominal Imaging

A

B

C

D

Figure 2-94 Portal Venous Gas in 2 Patients A and B. This 32-year-old man had a congenital cardiomyopathy with chronic heart failure when he presented with right lower quadrant pain, fevers, and leucocytosis. A. Axial CT image through the liver demonstrates 6 small black dots (arrows) that represent air in the portal venous system. This finding usually indicates necrosis of some portion of the bowel wall with passage of bowel gas into the portal system. B. CT of the right lower quadrant shows focal thickening of the cecum consistent with bowel edema. Surgical

exploration confirmed a diagnosis of necrotic ischemic cecum and ascending colon. C and D. This 47-year-old male with congestive heart failure, diabetes, and hypertension presented with distended abdomen. C. Transverse image of the right hepatic lobe demonstrates multiple linear-branching echogenic structures extending peripherally to the liver capsule (white arrowhead). D. Spectral Doppler interrogation of the main portal vein demonstrates portal venous waveform with superimposed bidirectional spikes (white arrowhead), typical of portal venous gas.

5 cm long. Usually, they are imperceptible as they blend with the surrounding mesenteric and retroperitoneal fat. Each appendage is supplied by 1 or 2 small end arteries branching from the vasa recta longa of the colon and is drained by a vein that passes through its narrow pedicle.312,377

Rarely the appendices epiploicae will undergo spontaneous torsion or spontaneous venous thrombosis because of their limited blood supply, pedunculated shape, and excessive mobility.377,703,704 When they torse, they result in a focal inflammatory process called epiploic appendagitis. Patients typically present with severe, acute abdominal

Chapter 2 Imaging of the Bowel 141

Figure 2-95 Epiploic Appendagitis This 32-year-old woman complained of acute left lower quadrant pain. The CT image shows a ring of soft-tissue attenuation surrounding a fatty center (arrow) adjacent to the descending colon. There is fat stranding surrounding the ring and there is mild thickening of the adjacent colon wall. This appearance is diagnostic of epiploic appendagitis.

pain that is localized to 1 of the lower quadrants and mimics acute appendicitis or diverticulitis. Computed tomographic findings of epiploic appendagitis are highly diagnostic and include visualization of the following: (1) a 1- to 4-cm paracolonic oval fatty mass representing the infarcted or inflamed appendix epiploica, (2) a well-circumscribed hyperattenuated rim that surrounds the mass and represents the inflamed visceral peritoneal lining, and (3) rarely a high-attenuation central dot representing engorged or thrombosed central vessels (see Figure 2-95).377,705,706 A hallmark of this disease process is that the paracolonic inflammatory changes are disproportionately severe relative to the mild local reactive thickening of the adjacent colonic wall.377 This pattern of inflammation denotes that the pathologic process is centered in the mesentery adjacent to the bowel wall rather than in the bowel wall. When visualized in a paracolonic distribution, this helps to narrow the differential diagnosis to include epiploic appendagitis and diverticulitis.377 Recognition of this unique appearance on noninvasive imaging and absence of diverticula within the region of inflammation are important because epiploic appendagitis is a self-limited process that does not require surgical intervention.707,708

Segmental Omental Infarction The greater omentum is the largest peritoneal fold consisting of a double layer of peritoneum that extends inferiorly from the greater curvature of the stomach as low as the pelvis and drapes over the anterior surfaces

of both the transverse colon and the pancreas. The right and left gastroepiploic arteries provide its main blood supply. On CT, the greater omentum typically appears as a layer of fat located just anterior to the transverse colon invested with small blood vessels.377 The size of the omentum is larger in obese patients and smaller in lower-weight patients. Segmental omental infarction preferentially occurs on the right side of the abdomen. It is believed that an embryologic variant of the blood supply to the right portion of the omentum predisposes it to venous thrombosis.709 About 85% of cases occur in adults. Patients typically present with abdominal pain located just to the right of the umbilicus,710 a clinical presentation that can mimic appendicitis, gallbladder disease, or pyelitis.377 The exact etiology and pathogenesis of omental infarction is unknown; however, anomalous arterial supply to the omentum, kinking of veins secondary to increases in intraabdominal pressure, and vascular congestion after large meals have been proposed as possible mechanisms.710 Risk factors for the development of omental infarction include obesity, recent surgery, and coughing.709,710 On US examinations, segmental omental infarction will appear as a solid, well-circumscribed, hyperechoic, and noncompressible mass immediately underlying the site of focal tenderness. This mass can adhere to the adjacent peritoneum. This phenomenon is best appreciated in the sagittal plane, which shows the mass to move in synchrony

Imaging Notes 2-25. CT-Identifiable Causes of GI-Related Acute Abdominal Pain Diverticulitis Primary finding: Focal wall thickening of the colon Secondary findings: Fat stranding, abscess Appendicitis Primary finding: Distended (>7 cm) enhancing appendix Secondary findings: Fat stranding, cecal wall thickening, abscess Cholecystitis Primary finding: Distended gallbladder with thickened wall (>1 mm) Secondary findings: gallstones, fat stranding Epiploic Appendagitis Primary finding: Small ring of edema around fatty epiploic appendage Secondary findings: none Segmental Omental Infarction Primary finding: Focal inflammatory mass in the right side of the greater omentum Secondary findings: fat stranding in the greater omentum

142

Diagnostic Abdominal Imaging

with the abdominal wall during respiration rather than separately from the abdominal wall like other intraperitoneal contents.28,710,711 In most cases, the diagnosis of segmental omental infarction is made on CT examinations. CT will demonstrate a large ovoid or cakelike mass with heterogenous regions of mixed fat and soft-tissue attenuation within the anterior abdomen. The mass is usually located in a paraumbilical location between the rectus abdominis muscles and the transverse colon in the expected region of the greater omentum.377,710 Ancillary findings occasionally include thickening of either or both the adjacent peritoneum and the bowel wall. Because the inflammatory process is centered in the omentum, the degree of fat stranding is disproportionately more severe than the colonic wall thickening, an important finding distinguishing segmental omental infarction from diverticulitis and other primary bowel pathologies.377 At times, distinguishing between omental infarction and epiploic appendagitis can be challenging because of the overlap in imaging appearance on CT. The clinical relevance of this distinction is limited, however, because management for both conditions is conservative.377,710

REFERENCES 1. Paulsen SR, Huprich JE, Fletcher JG, et al. CT enterography as a diagnostic tool in evaluating small bowel disorders: review of clinical experience with over 700 cases. Radiographics. 2006;26(3):641-657; discussion 657-662. 2. Maglinte DD, Sandrasegaran K, Lappas JC. CT enteroclysis: techniques and applications. Radiol Clin North Am. 2007;45(2):289-301. 3. Mazzeo S, Caramella D, Battolla L, et al. Crohn disease of the small bowel: spiral CT evaluation after oral hyperhydration with isotonic solution. J Comput Assist Tomogr. 2001;25(4):612-616. 4. Doerfler OC, Ruppert-Kohlmayr AJ, Reittner P, et al. Helical CT of the small bowel with an alternative oral contrast material in patients with Crohn disease. Abdom Imaging. 2003;28(3): 313-318. 5. Wold PB, Fletcher JG, Johnson CD, et al. Assessment of small bowel Crohn disease: noninvasive peroral CT enterography compared with other imaging methods and endoscopy— feasibility study. Radiology. 2003;229(1):275-281.

10. Pickhardt PJ, Choi JR, Hwang I, et al. Computed tomographic virtual colonoscopy to screen for colorectal neoplasia in asymptomatic adults. N Engl J Med. 2003;349(23):2191-2200. 11. Pickhardt PJ. Screening CT colonography: how I do it. AJR Am J Roentgenol. 2007;189(2):290-298. 12. Macari M, Bini EJ, Jacobs SL, et al. Significance of missed polyps at CT colonography. AJR Am J Roentgenol. 2004;183(1):127-134. 13. Johnson CD, Dachman AH. CT colonography: the next colon screening examination? Radiology. 2000;216(2):331-341. 14. Dachman AH, Kuniyoshi JK, Boyle CM, et al. CT colonography with three-dimensional problem solving for detection of colonic polyps. AJR Am J Roentgenol. 1998;171(4):989-995. 15. Macari M, Milano A, Lavelle M, et al. Comparison of timeefficient CT colonography with two- and three-dimensional colonic evaluation for detecting colorectal polyps. AJR Am J Roentgenol. 2000;174(6):1543-1549. 16. Gore RM, Levine MS, Laufer I. Textbook of Gastrointestinal Radiology. Vol. 2. Philadelphia: W.B. Saunders Co.; 1994:2716. 17. Pickhardt PJ. Primary 2D versus primary 3D polyp detection at screening CT colonography. AJR Am J Roentgenol. 2007;189(6):1451-1456. 18. Macari M, Megibow AJ. Pitfalls of using three-dimensional CT colonography with two-dimensional imaging correlation. AJR Am J Roentgenol. 2001;176(1):137-143. 19. Taylor SA, Halligan S, Slater A, et al. Polyp detection with CT colonography: primary 3D endoluminal analysis versus primary 2D transverse analysis with computer-assisted reader software. Radiology. 2006;239(3):759-767. 20. Beaulieu CF, Jeffrey RB Jr, Karadi C, et al. Display modes for CT colonography. Part II. Blinded comparison of axial CT and virtual endoscopic and panoramic endoscopic volume-rendered studies. Radiology. 1999;212(1):203-212. 21. Mang T, Maier A, Plank C, et al. Pitfalls in multi-detector row CT colonography: a systematic approach. Radiographics. 2007;27(2):431-454. 22. Yee J, Kumar NN, Hung RK, et al. Comparison of supine and prone scanning separately and in combination at CT colonography. Radiology. 2003;226(3):653-661. 23. Laks S, Macari M, Bini EJ. Positional change in colon polyps at CT colonography. Radiology. 2004;231(3):761-766. 24. Macari M, Berman P, Dicker M, et al. Usefulness of CT colonography in patients with incomplete colonoscopy. AJR Am J Roentgenol. 1999;173(3):561-564. 25. Morrin MM, Kruskal JB, Farrell RJ, et al. Endoluminal CT colonography after an incomplete endoscopic colonoscopy. AJR Am J Roentgenol. 1999;172(4):913-918.

6. Pickhardt PJ. Three-dimensional endoluminal CT colonography (virtual colonoscopy): comparison of three commercially available systems. AJR Am J Roentgenol. 2003;181(6):1599-1606.

26. Fenlon HM, McAneny DB, Nunes DP, et al. Occlusive colon carcinoma: virtual colonoscopy in the preoperative evaluation of the proximal colon. Radiology. 1999;210(2):423-428.

7. Johnson CD, Chen MH, Toledano AY, et al. Accuracy of CT colonography for detection of large adenomas and cancers. N Engl J Med. 2008;359(12):1207-1217.

27. O’Malley ME, Wilson SR. US of gastrointestinal tract abnormalities with CT correlation. Radiographics. 2003;23(1):59-72.

8. Macari M, Bini EJ, Jacobs SL, et al. Filling defects at CT colonography: pseudo- and diminutive lesions (the good), polyps (the bad), flat lesions, masses, and carcinomas (the ugly). Radiographics. 2003;23(5):1073-1091.

28. Puylaert JB. Ultrasound of acute GI tract conditions. Eur Radiol. 2001;11(10):1867-1877.

9. Lefere PA, Gryspeerdt SS, Dewyspelaere J, Baekelandt, et al. Dietary fecal tagging as a cleansing method before CT colonography: initial results polyp detection and patient acceptance. Radiology. 2002;224(2):393-403.

29. Fleischer AC, Muhletaler CA, James AE Jr. Sonographic assessment of the bowel wall. AJR Am J Roentgenol. 1981;136(5):887-891. 30. Ledermann HP, Borner N, Strunk H, et al. Bowel wall thickening on transabdominal sonography. AJR Am J Roentgenol. 2000;174(1):107-117.

Chapter 2 Imaging of the Bowel 143 31. Leyendecker JR, Bloomfeld RS, DiSantis DJ, et al. MR enterography in the management of patients with Crohn disease. Radiographics. 2009;29(6):1827-1846.

49. Ming SC, Goldman H. Gastric polyps; a histogenetic classification and its relation to carcinoma. Cancer. 1965; 18:721-726.

32. Masselli G, Casciani E, Polettini E, et al. Comparison of MR enteroclysis with MR enterography and conventional enteroclysis in patients with Crohn’s disease. Eur Radiol. 2008;18(3):438-447.

50. Ming SC. The adenoma-carcinoma sequence in the stomach and colon: II. Malignant potential of gastric polyps. Gastrointest Radiol. 1976;1:121-125.

33. Rieber A, Wruk D, Potthast S, et al. Diagnostic imaging in Crohn’s disease: comparison of magnetic resonance imaging and conventional imaging methods. Int J Colorectal Dis. 2000;15(3):176-181. 34. Siddiki HA, Fidler JL, Fletcher JG, et al. Prospective comparison of state-of-the-art MR enterography and CT enterography in small-bowel Crohn’s disease. AJR Am J Roentgenol. 2009;193(1):113-121. 35. Lee SS, Kim AY, Yang SK, et al. Crohn disease of the small bowel: comparison of CT enterography, MR enterography, and small-bowel follow-through as diagnostic techniques. Radiology. 2009;251(3):751-761. 36. Low RN, Francis IR, Politoske D, et al. Crohn’s disease evaluation: comparison of contrast-enhanced MR imaging and single-phase helical CT scanning. J Magn Reson Imaging. 2000;11(2):127-135. 37. Schmidt S, Lepori D, Meuwly JY, et al. Prospective comparison of MR enteroclysis with multidetector spiral-CT enteroclysis: interobserver agreement and sensitivity by means of “sign-bysign” correlation. Eur Radiol. 2003;13(6):1303-1311. 38. Negaard A, Paulsen V, Sandvik L, et al. A prospective randomized comparison between two MRI studies of the small bowel in Crohn’s disease, the oral contrast method and MR enteroclysis. Eur Radiol. 2007;17(9):2294-2301. 39. Schreyer AG, Geissler A, Albrich H, et al. Abdominal MRI after enteroclysis or with oral contrast in patients with suspected or proven Crohn’s disease. Clin Gastroenterol Hepatol. 2004;2(6): 491-497. 40. Horton KM, Fishman EK. Current role of CT in imaging of the stomach. Radiographics. 2003;23(1):75-87.

51. Joffe N, Antonioli D. Atypical appearances of benign hyperplastic polyps. AJR. 1978;131:147-152. 52. Smith HJ, Lee EL. Large hyperplastic polyps of the stomach. Gastrointest Radiol. 1983;8:19-23. 53. Burt RW. Gastric fundic gland polyps. Gastroenterology. 2003;125:1462-1469. 54. Tsuchigame T, Saito R, Ogata Y, et al. Clinical evaluation of gastric fundic gland polyps without familial polyposis coli. Abdom Imaging. 1995;20:101-105. 55. Iida M, Yao T, Itoh H, et al. Natural history of fundic gland polyposis in patients with familial adenomatosis coli/Gardner’s syndrome. Gastroenterology. 1985;89:1021-1025. 56. Iida M, Yao T, Watanabe J, et al. Fundi gland polyposis in patients without familial adenomatosis coli: Its incidence and clinical features. Gastroenterology. 1984;86:1437-1442. 57. Watanabe H, Enjoji M, Yao T, et al. Gastric lesions in familial adenomatosis coli. Hum Pathol. 1978;9:269-283. 58. Nishiura M, Hirota T, Itabashi M, et al. A clinical and histopathological study of gastric polyps in familial polyposis coli. Am J Gastroenterol. 1984;79(2):98-103. 59. Buck J, Harned RK, Lichtenstein JE, et al. Peutz-Jeghers syndrome. Radiographics. 1992;12:365-378. 60. Manfredi M. Hereditary hamartomatous polyposis syndromes understanding the disease risks as children reach adulthood. Gastroenterol Hepatol (NY). 2010;6(3):185-196. 61. Moore JR. Gastric carcinoma: 30-year review. Can J Surg. 1986;29(1):25-28. 62. Parkin DM, Bray F, Ferlay J, et al. Global cancer statistics, 2002. CA Cancer J Clin. 2005;55(2):74-108.

41. Rossi M, Broglia L, Maccioni F, et al. Hydro-CT in patients with gastric cancer: preoperative radiologic staging. Eur Radiol. 1997;7(5):659-664.

63. Ba-Ssalamah A, Prokop M, Uffmann M, Pokieser P, Teleky B, Lechner G. Dedicated multidetector CT of the stomach: spectrum of diseases. Radiographics. 2003;23(3):625-644.

42. Hori S, Tsuda K, Murayama S, et al. CT of gastric carcinoma: preliminary results with a new scanning technique. Radiographics. 1992;12(2):257-268.

64. Watanabe H, Bass J, Sobin L. Histological Typing of Esophageal and Gastric Tumors. 2nd Ed. 1990, Berlin: Springer-Verlag.

43. Baert AL, Roex L, Wilms G, et al. Computed tomography of the stomach with water as an oral contrast agent: technique and preliminary results. J Comput Assist Tomogr. 1989;13(4): 633-636.

65. Balthazar EJ, Rosenberg H, Davidian MM. Scirrhous carcinoma of the pyloric channel and distal antrum. AJR Am J Roentgenol. 1980;134(4):669-673.

44. Hundt W, Braunschweig R, Reiser M. Assessment of gastric cancer: value of breathhold technique and two-phase spiral CT. Eur Radiol. 1999;9(1):68-72. 45. Ming SC. The classification and significance of gastric polyps. Monogr Pathol. 1977(18):149-175. 46. Abraham SC, Singh VK, Yardley JH, et al. Hyperplastic polyps of the stomach: associations with histologic patterns of gastritis and gastric atrophy. Am J Surg Pathol. 2001;25(4):500-507. 47. Abraham SC, Singh VK, Yardley JH, et al. Hyperplastic polyps of the stomach associations with histologic patterns of gastritis and gastric atrophy. Am J Clin Pathol. 2001;115(2):224-234. 48. Tomosulo J. Gastric polyps: Histologic types and their relationship to gastric carcinoma. Cancer. 1971;27:1346-1355.

66. Levine MS, Kong V, Rubesin SE, et al. Scirrhous carcinoma of the stomach: radiologic and endoscopic diagnosis. Radiology. 1990;175(1):151-154. 67. Hisamichi S. Screening for gastric cancer. World J Surg. 1989;13:31-37. 68. Dicken BJ, Bigam DL, Cass C, et al. Gastric adenocarcinoma: review and considerations for future directions. Ann Surg. 2005;241(1):27-39. 69. Oiso T. Incidence of stomach cancer and its relation to dietary habits and nutrition in Japan between 1900 and 1975. Cancer Res. 1975;35(11 Pt. 2):3254-3258. 70. Fukuya T, Honda H, Hayashi T, et al. Lymph-node metastases: efficacy for detection with helical CT in patients with gastric cancer. Radiology. 1995;197(3):705-711.

144

Diagnostic Abdominal Imaging

71. Montesi A, Graziani L, Pesaresi A, et al. Radiologic diagnosis of early gastric cancer by routine double-contrast examination. Gastrointest Radiol. 1982;7(3):205-215. 72. Gold RP, Green PH, O’Toole KM, et al. Early gastric cancer: radiographic experience. Radiology. 1984;152(2):283-290. 73. Laufer I, Lavine MS. Double Contrast Gastrointestinal Radiology. 2nd ed. Philadelphia, PA: Saunders; 1992:xii, 701. 74. Levine MS, Igor L. Stomach. In: Double Contrast Gastrointestinal Radiology. Philadelphia: W.B. Saunders; 2000:213-225. 75. Maruyama M, Baba Y. Diagnosis of the invasive depth of gastric cancer. Abdom Imaging. 1994;19(6):532-536. 76. Hustinx R. PET imaging in assessing gastrointestinal tumors. Radiol Clin North Am. 2004;42(6):1123-1139, ix. 77. Stahl A, Ott K, Weber WA, et al. FDG PET imaging of locally advanced gastric carcinomas: correlation with endoscopic and histopathological findings. Eur J Nucl Med Mol Imaging. 2003;30(2):288-295. 78. Ott K, Fink U, Becker K, et al. Prediction of response to preoperative chemotherapy in gastric carcinoma by metabolic imaging: results of a prospective trial. J Clin Oncol. 2003;21(24):4604-4610. 79. Mochiki E, Kuwano H, Katoh H, et al. Evaluation of 18F-2deoxy-2-fluoro-D-glucose positron emission tomography for gastric cancer. World J Surg. 2004;28(3):247-253. 80. Yoshioka T, Yamaguchi K, Kubota K, et al. Evaluation of 18F-FDG PET in patients with advanced, metastatic, or recurrent gastric cancer. J Nucl Med. 2003;44(5):690-699. 81. Turlakow A, Yeung HW, Salmon AS, et al. Peritoneal carcinomatosis: role of (18)F-FDG PET. J Nucl Med. 2003;44(9):1407-1412. 82. Aaltonen LA, Hamilton Stanley R, World Health Organization, et al. Pathology and genetics of tumours of the digestive system. World Health Organization classification of tumours., Lyon, Oxford: IARC Press; Oxford University Press; 2000 (distributor. 314 p.) 83. Capella C, Solcia E, Sobin LH, et al. Endocrine tumours of the stomach. In: Hamilton SR, Aaltonen LA, eds. World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of the Digestive System. IARC: Lyon, France; 2000:53-57. 84. Modlin I, Sandor A, Tang LH, et al. A 40-year analysis of 265 gastric carcinoids. Am J Gastroenterol. 1997;92:633-638. 85. Levy AD, Sobin LH. From the archives of the AFIP: Gastrointestinal carcinoids: imaging features with clinicopathologic comparison. Radiographics. 2007; 27(1):237-257. 86. Binstock AJ, Johnson CD, Stephens DH, et al. Carcinoid tumors of the stomach: a clinical and radiographic study. AJR. 2001;176:947-951. 87. Rindi G, Luinetti O, Cornaggia M, et al. Three subtypes of gastric argyrophil carcinoid and the gastric neuroendocrine carcinoma: a clinicopathologic study. Gastroenterology. 1993; 104:994-1006. 88. Rindi G, Bordi C, Rappel S, et al. Gastric carcinoids and neuroendocrinecarcinomas: pathogenesis, pathology, and behavior. World J Surg. 1996;20:168-172.

91. Gossios K, Katsimbri P, Tsianos E. CT features of gastric lymphoma. Eur Radiol. 2000;10(3):425-430. 92. Yoo CC, Levine MS, Furth EE, et al. Gastric mucosa-associated lymphoid tissue lymphoma: radiographic findings in six patients. Radiology. 1998;208(1):239-243. 93. Ghai S, Pattison J, O’Malley ME, et al. Primary gastrointestinal lymphoma: spectrum of imaging findings with pathologic correlation. Radiographics. 2007;27(5):1371-1388. 94. Smith C, Kubicka RA, Thomas CR Jr. Non-Hodgkin lymphoma of the gastrointestinal tract. Radiographics. 1992;12(5):887-899. 95. Levine AM, Meyer PR, Begandy MK, et al. Development of B-cell lymphoma in homosexual men. Clinical and immunologic findings. Ann Intern Med. 1984;100(1):7-13. 96. Mengoli M, Marchi M, Rota E, et al. Primary non-Hodgkin’s lymphoma of the esophagus. Am J Gastroenterol. 1990;85(6): 737-741. 97. Araki K, Ogata T, Kobayashi M, et al. A morphological study on the histogenesis of human colorectal hyperplastic polyps. Gastroenterology. 1995;109:1468-1474. 98. Craig O, Gregson R. Primary lymphoma of the gastrointestinal tract. Clin Radiol. 1981;32(1):63-72. 99. Menuck LS. Gastric lymphoma, a radiologic diagnosis. Gastrointest Radiol. 1976;1(2):157-161. 100. ReMine SG, Braasch JW. Gastric and small bowel lymphoma. Surg Clin North Am. 1986;66(4):713-722. 101. Levine MS, Pantongrag-Brown L, Aguilera NS, et al. NonHodgkin lymphoma of the stomach: a cause of linitis plastica. Radiology. 1996;201(2):375-378. 102. Buy JN, Moss AA. Computed tomography of gastric lymphoma. AJR Am J Roentgenol. 1982;138(5):859-865. 103. Miller FH,Kochman ML, Talamonti MS, et al. Gastric cancer. Radiologic staging. Radiol Clin North Am. 1997;35(2):331-349. 104. Ciftci AO, Tanyel FC, Kotiloğlu E, Hiçsönmez A. Gastric lymphoma causing gastric outlet obstruction. J Pediatr Surg. 1996;31(10):1424-1426. 105. Miettinen M, Lasota J. Gastrointestinal stromal tumors— definition, clinical, histological, immunohistochemical, and molecular genetic features and differential diagnosis. Virchows Arch. 2001;438(1):1-12. 106. Miettinen M, Sarlomo-Rikala M, Sobin LH, et al. Gastrointestinal stromal tumors and leiomyosarcomas in the colon: a clinicopathologic, immunohistochemical, and molecular genetic study of 44 cases. Am J Surg Pathol. 2000;24(10):1339-1352. 107. Suster S. Gastrointestinal stromal tumors. Semin Diagn Pathol. 1996;13(4):297-313. 108. Nauert TC, Zornoza J, Ordonez N. Gastric leiomyosarcoma. AJR Am J Roentgenol. 1982;139(2):291-297. 109. Pannu HK, Hruban RH, Fishman EK. CT of gastric leiomyosarcoma: patterns of involvement. AJR Am J Roentgenol. 1999;173(2):369-373. 110. Libshitz HI, Lindell MM, Dodd GD. Metastases to the hollow viscera. Radiol Clin North Am. 1982;20(3):487-499.

89. Balthazar EJ, Megibow A, Bryk D. Gastric carcinoid tumors: radiographic features in eight cases. AJR. 1997;139:1123-1127.

111. Dunnick NR, Harell GS, Parker BR. Multiple “bull’seye” lesions in gastric lymphoma. AJR Am J Roentgenol. 1976;126(5):965-969.

90. Berger MW, Stephens DH. Gastric carcinoid tumors associated with chronic hypergastrinemia in a patient with ZollingerEllison syndrome. Radiology. 1996;201:371-373.

112. Goldstein HM, Beydoun MT, Dodd GD. Radiologic spectrum of melanoma metastatic to the gastrointestinal tract. AJR Am J Roentgenol. 1977;129(4):605-612.

Chapter 2 Imaging of the Bowel 145 113. Rose HS, Balthazar EJ, Megibow AJ, et al. Alimentary tract involvement in Kaposi sarcoma: radiographic and endoscopic findings in 25 homosexual men. AJR Am J Roentgenol. 1982;139(4):661-666.

135. Lanza F, Royer G, Nelson R. An endoscopic evaluation of the effects of non-steroidal anti-inflammatory drugs on the gastric mucosa. Gastrointest Endosc. 1975;21(3):103-105.

114. McLeod AJ, Zornoza J, Shirkhoda A. Leiomyosarcoma: computed tomographic findings. Radiology. 1984;152(1):133-136.

136. Laufer I, Trueman T. Multiple superficial gastric erosions due to Crohn’s disease of the stomach. Radiologic and endoscopic diagnosis. Br J Radiol. 1976;49(584):726-728.

115. Sharp RM, Ansel HJ, Keel SB. Best cases from the AFIP: gastrointestinal stromal tumor. Armed Forces Institute of Pathology. Radiographics. 2001;21(6):1557-1560.

137. Ariyama J, Wehlin L, Lindstrom CG, et al. Gastroduodenal erosions in Crohn’s disease. Gastrointest Radiol. 1980;5(2): 121-125.

116. Scatarige JC, Fishman EK, Jones B, et al. Gastric leiomyosarcoma: CT observations. J Comput Assist Tomogr. 1985;9(2):320-327.

138. Cronan J, Burrell M, Trepeta R. Aphthoid ulcerations in gastric candidiasis. Radiology. 1980;134(3):607-611.

117. Hasegawa S, Semelka RC, Noone TC, et al. Gastric stromal sarcomas: correlation of MR imaging and histopathologic findings in nine patients. Radiology. 1998;208(3):591-595.

139. McLean AM, Paul RE Jr, Philipps E, et al. Chronic erosive gastritis—clinical and radiological features. J Can Assoc Radiol. 1982;33(3):158-162.

118. Rose H, Balthazar EJ, Megibow AJ, et al. Alimentary tract involvement in Kaposi sarcoma: Radiographic and endoscopic findings in 25 homosexual men. AJR. 1982;139:661-666.

140. Dooley CP, Cohen H, Fitzgibbons PL, et al. Prevalence of Helicobacter pylori infection and histologic gastritis in asymptomatic persons. N Engl J Med. 1989;321(23):1562-1566.

119. Friedman SL. Gastrointestinal and hepatobiliary neoplasms in AIDS. Gastroenterol Clin North Am. 1988;17:465-486.

141. Levine MS, Rubesin SE. The Helicobacter pylori revolution: radiologic perspective. Radiology. 1995;195(3):593-596.

120. Saltz R, Kurtz RC, Lightdale CJ, et al. Kaposi’s sarcoma: Gastrointestinal involvement correlation with skin findings and immunologic function. Dig Dis Sci. 1984;29:817-823.

142. Graham DY, Lew GM, Klein PD, et al. Effect of treatment of Helicobacter pylori infection on the long-term recurrence of gastric or duodenal ulcer. A randomized, controlled study. Ann Intern Med. 1992;116(9):705-708.

121. Wall SD, Friendman SL, Margulis AR. Gastrointestinal Kaposi’s sarcoma in AIDS: Radiographic manifestations. J Clin Gastroenterol. 1984;6:165-171. 122. Hadjiyane C, Lee YH, Stein L, et al. Kaposi’s sarcoma presenting as linitis plastica. Am J Gastroenterol. 1991;86: 1823-1825. 123. Thompson WM. Imaging and findings of lipomas of the gastrointestinal tract. AJR Am J Roentgenol. 2005;184(4): 1163-1171. 124. Taylor AJ, Stewart ET, Dodds WJ. Gastrointestinal lipomas: a radiologic and pathologic review. AJR Am J Roentgenol. 1990;155(6):1205-1210. 125. Agha FP, Dent TL, Fiddian-Green RG, et al. Bleeding lipomas of the upper gastrointestinal tract. A diagnostic challenge. Am Surg. 1985;51(5):279-285. 126. Menuck L, Amberg J. Metastatic disease involving the stomach. Am J Dig Dis. 1975;20:903-913. 127. Klein M, Sherlock P. Gastric and colonic metastases from breast carcinoma. Am J Dig Dis. 1972;17:881-886. 128. Das Gupta TK, Brasfield RD. Metastatic melanoma of the gastrointestinal tract. Arch Surg. 1964;88:969-973. 129. Meyers M, McSweeney J. Secondary neoplasms of the bowel. Radiology. 1972;105:1-11. 130. Goldstein HM, Beydonn MT, Dodd GD. Radiologic spectrum of metastatic melanoma to the gastrointestinal tract. AJR. 1977;129:605-612. 131. Lipshutz H, Lindell M, Dodd G. Metastases to the hollow viscera. Radiol Clin North Am. 1982;20:487-499. 132. Caskey CI, Scatarige JC, Fishman EK. Distribution of metastases in breast carcinoma: CT evaluation of the abdomen. Clin Imaging. 1991;15:166-171.

143. NIH Consensus Conference. Helicobacter pylori in peptic ulcer disease. NIH Consensus Development Panel on Helicobacter pylori in Peptic Ulcer Disease. JAMA. 1994;272(1):65-69. 144. Poplack, Paul RE, Goldsmith M, Matsue H, Moore JP, Norton R. Demonstration of erosive gastritis by the doublecontrast technique. Radiology. 1975;117(3 Pt 1):519-521. 145. Laufer I. Assessment of the accuracy of double contrast gastroduodenal radiology. Gastroenterology. 1976;71(5): 874-878. 146. Den Orth JO, Dekker W. Gastric erosions: radiological and endoscopic aspects. Radiol Clin (Basel). 1976;45(2-4):88-99. 147. Tragardh B, Wehlin L, Ohashi K. Radiologic appearance of complete gastric erosions. Acta Radiol Diagn (Stockh). 1978;19(4):634-642. 148. Catalano D, Pagliari U. Gastroduodenal erosions: radiological findings. Gastrointest Radiol. 1982;7(3):235-240. 149. Laufer I, Hamilton J, Mullens JE. Demonstration of superficial gastric erosions by double contrast radiography. Gastroenterology. 1975;68(2):387-391. 150. Levine MS, Verstandig A, Laufer I. Serpiginous gastric erosions caused by aspirin and other nonsteroidal antiinflammatory drugs. AJR Am J Roentgenol. 1986;146(1):31-34. 151. Laveran-Stieber RL, Laufer I, Levine MS. Greater curvature antral flattening: a radiologic sign of NSAID-related gastropathy. Abdom Imaging. 1994;19(4):295-297. 152. Levine MS, Creteur V, Kressel HY, Laufer I, Herlinger H. Benign gastric ulcers: diagnosis and follow-up with doublecontrast radiography. Radiology. 1987;164(1):9-13.

133. Taal BG, Peterse H, Boot H. Clinical presentation, endoscopic features, and treatment of gastric metastases from breast carcinoma. Cancer. 2000;89:2214-2221.

153. Sun DC, Stempien SJ. The Veterans Administration Cooperative Study on Gastric Ulcer. 3. Site and size of the ulcer as determinants of outcome. Gastroenterology. 1971;61(4):(suppl 2):576-584.

134. Furth EE, Rubesin SE, Levine MS. Pathologic primer on gastritis: an illustrated sum and substance. Radiology. 1995;197(3):693-698.

154. Thompson G, Somers S, Stevenson GW. Benign gastric ulcer: a reliable radiologic diagnosis? AJR Am J Roentgenol. 1983;141(2):331-333.

146

Diagnostic Abdominal Imaging

155. Gelfand DW, Dale WJ, Ott DJ. The location and size of gastric ulcers: radiologic and endoscopic evaluation. AJR Am J Roentgenol. 1984;143(4):755-758.

177. DeVita VT, Hellman S, Rosenberg SA. Cancer, principles & practice of oncology. Philadelphia, PA: Lippincott Williams & Wilkins; 2005:1788-1813.

156. Amberg JR, Zboralske FF. Gastric ulcers after 70. Am J Roentgenol Radium Ther Nucl Med. 1966;96(2):393-399.

178. Wolfe MM, Jensen RT. Zollinger-Ellison syndrome: Current concepts in diagnosis and management. N Engl J Med. 1987;317:1200-1209.

157. Sheppard MC, Holmes GK, Cockel R. Clinical picture of peptic ulceration diagnosed endoscopically. Gut. 1977;18(7): 524-530. 158. Kikuchi Y, Levine MS, Laufer I, et al. Value of flow technique for double-contrast examination of the stomach. AJR Am J Roentgenol. 1986;147(6):1183-1184.

179. Amberg J, Ellison EH, Wilson SD, et al. Roentgenographic observations in the Zollinger-Ellison syndrome. JAMA. 1964;190:185-187. 180. Missakian MM, Carlson HC, Huzenga KA. Roentgenographic findings in Zollinger-Ellison syndrome. AJR. 1965;94:429-437.

159. Urban BA, Fishman EK, Hruban RH. Helicobacter pylori gastritis mimicking gastric carcinoma at CT evaluation. Radiology. 1991;179(3):689-691.

181. Nelson S, Christoforidis A. Roentgenologic features of the Zollinger-Ellison syndrome: Ulcerogenic tumor of the pancreas. Semin Roentgenol. 1968;3:254-266.

160. Jacobs JM, Hill MC, Steinberg WM. Peptic ulcer disease: CT evaluation. Radiology. 1991;178(3):745-748.

182. Akerstrom G, Hellman P. Surgery on neuroendocrine tumours. Best Pract Res Clin Endocrinol Metab. 2007;21(1): 87-109.

161. Wagtmans M, Verspaget HW, Lamers CB, et al. Clinical aspects of Crohn’s disease of the upper gastrointestinal tract: A comparison with distal Crohn’s disease. Am J Gastroenterol. 1997;92:1467-1470. 162. Marshak RH, Maklansky D, Kurzban JD, et al. Crohn’s disease of the stomach and duodenum. Am J Gastroenterol. 1982;77:340-343. 163. Levine MS. Crohn’s disease of the upper gastrointestinal tract. Radiol Clin North Am. 1987;25(1):79-91. 164. Nelson SW. The discovery of gastric ulcers and the differential diagnosis between benignancy and malignancy. Radiol Clin North Am. 1969;7(1):5-25. 165. Levine MS. Erosive gastritis and gastric ulcers. Radiol Clin North Am. 1994;32(6):1203-1214. 166. Meeroff M, Gollan JRM, Meeroff JC. Gastric diverticulum. Am J Gastroenterol. 1967;47:89-203. 167. Martin L. Diverticula of the stomach. Ann Intern Med. 1936;10:447-465. 168. Schwartz AN, Goiney RC, Graney DO. Gastric diverticulum simulating an adrenal mass: CT appearance and embryogenesis. AJR Am J Roentgenol. 1986;146:553-554. 169. Harford W, Jeyarajah R. Diverticula of the pharynx, esophagus, stomach, and small intestine. In: Feldman M, Friedman L, Brandt LJ, eds. Sleisenger & Fordtran’s gastrointestinal and liver disease. Philadelphia, PA: Saunders. 2006;465-477. 170. Akerlund A. Diverticula of the stomach from a roentgenological point of view. Acta Radiol (Stockh). 1923;2:476-485. 171. Chen C, Jaw YS, Wu DC, et al. MDCT of giant gastric folds: differential diagnosis. AJR Am J Roentgenol. 2010;195(5): 1124-1130. 172. Lauren P. The two histological main types of gastric carcinoma: Diffuse and so-called intestinal-typecarcinoma. Acta Pathol Microbiol Scand. 1965;64:31-49. 173. Raskin M. Some specific radiological findings and consideration of linitis plastica of the gastrointestinal tract. CRC Crit Rev Clin Radiol Nucl Med. 1976;8(1):87-106. 174. Sohn J, Levine MS, Furth EE, et al. Helicobacter pylori gastritis: radiographic findings. Radiology. 1995;195(3):763-767. 175. Fishman EK, Urban BA, Hruban RH. CT of the stomach: spectrum of disease. Radiographics. 1996;16(5):1035-1054. 176. Jensen RT. Gastrinomas: Advances in Diagnosis and Management. Neuroendocrinology. 2004;80(suppl 1):23-27.

183. Rockall AG, Reznek RH. Imaging of neuroendocrine tumours (CT/MR/US). Best Pract Res Clin Endocrinol Metab. 2007;21(1):43-68. 184. Wank SA, Doppman JL, Miller D, et al. Prospective study of the ability of computed axial tomography to localize gastrinomas in patients with Zollinger-Ellison syndrome. Gastroenterology. 1987;92(4):905-912. 185. Balci NC, Semelka RC. Radiologic features of cystic, endocrine and other pancreatic neoplasms. Eur J Radiol. 2001;38(2): 113-119. 186. Doppman JL. The localization and treatment of parathyroid adenomas by angiographic techniques. Ann Radiol (Paris). 1980;23(4):253-258. 187. Ichikawa T, Peterson MS, Federle MP, et al. Islet cell tumor of the pancreas: biphasic CT versus MR imaging in tumor detection. Radiology. 2000;216(1):163-171. 188. Owen NJ, Sohaib SA, Peppercorn PD, et al. MRI of pancreatic neuroendocrine tumours. Br J Radiol. 2001;74(886):968-973. 189. Semelka RC, Custodio CM, Cem Balci N, Woosley JT. Neuroendocrine tumors of the pancreas: spectrum of appearances on MRI. J Magn Reson Imaging. 2000;11(2): 141-148. 190. Coffey RJ, Washington MK, Corless CL, Heinrich MC. Ménétrier disease and gastrointestinal stromal tumors: hyperproliferative disorders of the stomach. J Clin Invest. 2007;117(1):70-80. 191. Wolfsen H, Carpenter H, Talley N. Ménétrier’s disease: A form of hypertrophic gastropathy or gastritis? Gastroenterology. 19093;104:1310-1319. 192. Occena R, Taylor SF, Robinson CC, et al. J Pediatr Gastroenterol Nutr. 1993;17(2):217-224. 193. Reese D, Hodgson J, Dockerty M. Giant hypertrophy of the gastric mucosa (Ménétrier’s disease): A correlation of the roentgenographic, pathologic, and clinical findings. AJR. 1962;88:619-626. 194. Olmsted W, Cooper P, Madewell J. Involvement of the gastric antrum in Ménétrier’s disease. AJR. 1976;126:524-529. 195. Duprey K, Ahmed S, Mishriki Y. Ménétrier disease in an acquired immunodeficiency syndrome patient. South Med J. 2010;103(1):93-95. 196. Wilkerson M, Meschter S, Brown R. Menetrier’s disease presenting with iron deficiency anemia. Ann Clin Lab Sci. 1998;28(1):14-18.

Chapter 2 Imaging of the Bowel 147 197. Klein N, Hargrove R, Sleisenger M, et al. “Eosinophilic gastroenteritis”. Medicine (Baltimore). 1970;49(4):299-319. 198. Naylor A. “Eosinophilic gastroenteritis”. Eosinophilic gastroenteritis. Scott Med J. 1990;35(6):163-165. 199. Baig M, Qadir A, Rasheed J. A review of eosinophilic gastroenteritis. J Natl Med Assoc. 2006;98(10):1616-1619. 200. Goldberg H, O’Kieffe D, Jenis EH, et al. Diffuse eosinophilic gastroenteritis. AJR. 1973;119:342-351. 201. Wehunt WD, Lewis BS, Frankel A, et al. Eosinophilic gastritis. Radiology. 1976;120:85-89. 202. Sheikh RA, Baba AA, Ahmad SM, et al. Unusual presentations of eosinophilic gastroenteritis: Case series and review of literature. World J Gastroenterol. 2009;15(17):2156-2161. 203. Tursi A, Rella G, Inchingolo CD, et al. Gastric outlet obstruction due to gastroduodenal eosinophilic gastroenteritis. Endoscopy. 2007;39(suppl 1):E184. 204. Rosai J. Ackerman’s Surgical Pathology. Vol. 2. 8th ed. St. Louis: Mosby; 1996: (xiv, 2732. 50 p.). 205. DiSario JA, Burt RW, Vargas H, et al. Small bowel cancer: epidemiological and clinical characteristics from a populationbased registry. Am J Gastroenterol. 1994;89(5):699-701. 206. Rosai J. Gastrointestinal tract. In: Rosai J. ed. Ackerman’s Surgical Pathology. St. Louis, MO: Mosby-Year Book: 1995.

220. Pantongrag-Brown L, Buetow PC, Carr NJ, et al. Calcification and fibrosis in mesenteric carcinoid tumor: CT findings and pathologic correlation. AJR Am J Roentgenol. 1995;164(2): 387-391. 221. Herlinger H, Maglinte DDT, Birnbaum BA. Clinical Imaging of the Small Intestine. 2nd ed. New York: Springer; 1999:xvi, 576. 222. Horton KM, Kamel I, Hofmann L, et al. Carcinoid tumors of the small bowel: a multitechnique imaging approach. AJR Am J Roentgenol. 2004;182(3):559-567. 223. Bader TR, Semelka RC, Chiu VC, et al. MRI of carcinoid tumors: spectrum of appearances in the gastrointestinal tract and liver. J Magn Reson Imaging. 2001;14(3):261-269. 224. Scarsbrook AF, Ganeshan A, Statham J, et al. Anatomic and functional imaging of metastatic carcinoid tumors. Radiographics. 2007;27(2):455-477. 225. Quan GM, Pitman A, Slavin J, et al. Soft tissue metastasis of carcinoid tumour: a rare manifestation. ANZ J Surg. 2004;74(3):164-166. 226. Isidori AM, Kaltsas G, Frajese V, et al. Ocular metastases secondary to carcinoid tumors: the utility of imaging with [(123)I]meta-iodobenzylguanidine and [(111)In]DTPA pentetreotide. J Clin Endocrinol Metab. 2002;87(4):1627-1633.

207. Herbsman H, Wetstein L, Rosen Y, et al. Tumors of the small intestine. Curr Probl Surg. 1980;17(3):121-182.

227. Robboy SJ, Scully RE, Norris HJ. Carcinoid metastatic to the ovary. A clinocopathologic analysis of 35 cases. Cancer. 1974;33(3):798-811.

208. Gore RM. Small bowel cancer. Clinical and pathologic features. Radiol Clin North Am. 1997;35(2):351-360.

228. Pelage JP, Soyer P, Boudiaf M, et al. Carcinoid tumors of the abdomen: CT features. Abdom Imaging. 1999;24(3):240-245.

209. Maglinte DD, O’Connor K, Bessette J, et al. The role of the physician in the late diagnosis of primary malignant tumors of the small intestine. Am J Gastroenterol. 1991;86(3):304-308.

229. Levine MS, Rubesin SE, Pantongrag-Brown L, et al. NonHodgkin’s lymphoma of the gastrointestinal tract: radiographic findings. AJR Am J Roentgenol. 1997;168(1):165-172.

210. Dudiak KM, Johnson CD, Stephens DH. Primary tumors of the small intestine: CT evaluation. AJR Am J Roentgenol. 1989;152(5):995-998. 211. Lai EC, Doty JE, Irving C, Tompkins RK. Primary adenocarcinoma of the duodenum: analysis of survival. World J Surg. 1988;12(5):695-699. 212. Laurent F, Raynaud M, Biset JM, et al. Diagnosis and categorization of small bowel neoplasms: role of computed tomography. Gastrointest Radiol. 1991;16(2):115-119. 213. Buckley JA, Fishman EK. CT evaluation of small bowel neoplasms: spectrum of disease. Radiographics. 1998;18(2): 379-392. 214. Maglinte DD, Gage SN, Harmon BH, et al. Obstruction of the small intestine: accuracy and role of CT in diagnosis. Radiology. 1993;188(1):61-64.

230. Koh PK, Horsman JM, Radstone CR, et al. Localised extranodal non-Hodgkin’s lymphoma of the gastrointestinal tract: Sheffield Lymphoma Group experience (1989-1998). Int J Oncol. 2001;18(4):743-748. 231. Macari M, Balthazar EJ. CT of bowel wall thickening: significance and pitfalls of interpretation. AJR Am J Roentgenol. 2001;176(5):1105-1116. 232. Balthazar EJ, Noordhoorn M, Megibow AJ, et al. CT of smallbowel lymphoma in immunocompetent patients and patients with AIDS: comparison of findings. AJR Am J Roentgenol. 1997;168(3):675-680. 233. Byun JH, Ha HK, Kim AY, et al. CT findings in peripheral T-cell lymphoma involving the gastrointestinal tract. Radiology. 2003;227(1):59-67.

215. Kazerooni EA, Quint LE, Francis IR. Duodenal neoplasms: predictive value of CT for determining malignancy and tumor resectability. AJR Am J Roentgenol. 1992;159(2):303-309.

234. Shojaku H, Futatsuya R, Seto H, et al. Malignant gastrointestinal stromal tumor of the small intestine: radiologic-pathologic correlation. Radiat Med. 1997;15(3): 189-192.

216. Merine D, Fishman EK, Jones B. CT of the small bowel and mesentery. Radiol Clin North Am. 1989;27(4):707-715.

235. Zua MS. Familial Adenomatous Polyposis Syndrome. Hospital Physician. 1999;35(5):61-68.

217. Modlin IM, Lye KD, Kidd M. A 5-decade analysis of 13,715 carcinoid tumors. Cancer. 2003;97(4):934-959.

236. Burt RW. Inherited colorectal cancer syndrome. American Society for Gastrointestinal Endoscopy Clinical Update. 1998; 5:1-4.

218. Sweeney JF, Rosemurgy AS. Carcinoid Tumors of the Gut. Cancer Control. 1997;4(1):18-24. 219. Capella C, Solcia E, Sobin LH, et al. Endocrine tumours of the small intestine, in World Health Organization classification of tumours: pathology and genetics of tumours of the digestive system, Hamilton SR, Aaltonen LA, Editors. IARC: Lyon, France. 2000;77-82.

237. Giardiello F, Welsh SB, Hamilton SR, et al. Increased risk of cancer in the Peutz. Jeghers syndrome. N Engl J Med. 1987;316:1151-1154. 238. Sener R, Kumcuoglu Z, Elmas N, et al. Peutz-Jeghers syndrome: CT and US demonstration ofsmall bowel polyps. Gastrointest Radiol. 1991;16:21-23.

148 Diagnostic Abdominal Imaging 239. Margulis AR, Burhenne HJ. Practical Alimentary Tract Radiology. St. Louis: Mosby-Year Book; 1993:xii, 512. 240. Koehler RE. Neoplasms. In: Freeny PC, Stevenson GW, eds. Margulis and Burhenne’s Alimentary Tract Radiology. St Louis, MO: Mosby-Year Book; 1994;627-648.

260. Estrada RG, Spjut HJ. Hyperplastic polyps of the large bowel. Am J Surg Pathol. 1980;4:127-133. 261. Goldman H, Ming S, Hickok DF. Nature and significance of hyperplastic polyps of the human colon. Arch Pathol. 1970;89:349-354.

241. Levine MS, Rubesin SE, Laufer I. Pattern approach for diseases of mesenteric small bowel on barium studies. Radiology. 2008;249(2):445-460.

262. Ott DJ, Gelfand DW, Wu WC, Ablin DS. Colon polyp morphology on double-contrast barium enema: its pathologic predictive value. AJR Am J Roentgenol. 1983;141(5):965-970.

242. Moss S, Calam J. Helicobacter pylori and peptic ulcers: the present position. Gut. 1992;33(3):289-292.

263. Gollub MJ, Schwartz LH, Akhurst T. Update on colorectal cancer imaging. Radiol Clin North Am. 2007;45(1):85-118.

243. Rodriguez HP, Aston JK, Richardson CT. Ulcers in the descending duodenum. Postbulbar ulcers. Am J Roentgenol Radium Ther Nucl Med. 1973;119(2):316-322.

264. Laufer IaK H. Principles of double contrast diagnosis. In: Double Contrast Gastrointestinal Radiology. W.B. Saunders; 2000;8-46.

244. Jayaraman MV, Mayo-Smith WW, Movson JS, Dupuy DE, Wallach MT. CT of the duodenum: an overlooked segment gets its due. Radiographics. 2001;21 Spec No:S147-S160.

265. Youker JE, Welin S. Differentiation of true polypoid tumors of the colon from extraneous material: a new Roentgen sign. Radiology. 1965;84:610-615.

245. Cooke WT, Cox EV, Fone DJ, et al. The clinical and metabolic significance of jejunal diverticula. Gut. 1963;4:115-131.

266. Simms SM. Differential diagnosis of the bowler hat sign. AJR Am J Roentgenol. 1985;144(3):585-587.

246. Rossi P, Gourtsoyiannis N, Bezzi M, et al. Meckel’s diverticulum: imaging diagnosis. AJR Am J Roentgenol. 1996;166(3):567-573.

267. Tobin KD, Young JW. The bowler hat: a valid sign of colonic polyps? Gastrointest Radiol. 1987;12(3):250-252.

247. Park JJ, Wolff BG, Tollefson MK, et al. Meckel diverticulum: the Mayo Clinic experience with 1476 patients (1950-2002). Ann Surg. 2005;241(3):529-533. 248. Matsagas MI, Fatouros M, Koulouras B, et al. Incidence, complications, and management of Meckel’s diverticulum. Arch Surg. 1995;130(2):143-146. 249. Levy AD, Hobbs CM. From the archives of the AFIP. Meckel diverticulum: radiologic features with pathologic Correlation. Radiographics. 2004;24(2):565-587. 250. Bani-Hani KE, Shatnawi NJ. Meckel’s diverticulum: comparison of incidental and symptomatic cases. World J Surg. 2004;28(9):917-920.

268. Miller WT Jr, Levine MS, Rubesin SE, et al. Bowler-hat sign: a simple principle for differentiating polyps from diverticula. Radiology. 1989;173(3):615-617. 269. Bussey JHR, Veale AMO, Morson BC. Genetics of gastrointestinal polyposis. Gastroenterology. 1978;74:1325-1330. 270. Nandakumar G, Morgan JA, Silverberg D, Steinhagen RM. Familial polyposis coli: Clinical manifestations, evaluation, management and treatment. Mt Sinai J Med. 2004;71:384-391. 271. Welling DR, Beart RW. Surgical alternatives in the treatment of polyposis coli. Semin Surg Oncol. 1987;3:99-104. 272. Taylor SA, Halligan S, Moore L, et al. Multidetector-row CT duodenography in familial adenomatous polyposis: a pilot study. Clin Radiol. 2004;59(10):939-945.

251. Jain TP, Sharma R, Chava SP, et al. Pre-operative diagnosis of Meckel’s diverticulum: report of a case and review of literature. Trop Gastroenterol. 2005;26(2):99-101.

273. Healy JC, Reznek RH, Clark SK, et al. MR appearances of desmoid tumors in familial adenomatous polyposis. AJR. 1997;169:465-472.

252. Craig O, Murfitt J. Radiological demonstration of Meckel’s diverticulum. Br J Surg. 1980;67(12):881-883.

274. Brooks AP, Reznek RH, Nugent K, et al. CT appearances of desmoid tumours in familial adenomatous polyposis: further observations. Clin Radiol. 1994 49(9):601-617.

253. Hughes JA, Hatrick A, Rankin S. Computed tomography findings in an inflamed meckel diverticulum. Br J Radiol. 1998;71(848):882-883.

275. Howe J, Mitros F, Summers R. The risk of gastrointestinal carcinoma in familial juvenile polyposis. 1988;5(8):751-756.

254. Mortele KJ, Govaere F, Vogelaerts D, et al. Giant Meckel’s diverticulum containing enteroliths: typical CT imaging findings. Eur Radiol. 2002;12(1):82-84.

276. Schreibman IR, Baker M, Amos C, et al. The hamartomatous polyposis syndromes. A clinical and molecular review. 2005;100: 476-490.

255. Hoff G, Foerster A, Vatn MH, et al. Epidemiology of polyps in the rectum and colon. Recovery and evaluation of unresected polyps 2 years after detection. Scand J Gastroenterol. 1986;21(7):853-862.

277. Kao KT, Patel JK, Pampati V. Cronkhite-Canada Syndrome: A Case Report and Review of Literature. Gastroenterology Research and Practice. 2009:1-4.

256. Arthur JF. The significance of small mucosal polyps of the rectum. Proc R Soc Med. 1962;55:703-705. 257. Vatn MH, Stalsberg H. The prevalence of polyps of the large intestine in Oslo: an autopsy study. Cancer. 1982;49: 819-825.

278. Horton KM, Corl FM, Fishman EK. CT evaluation of the colon: inflammatory disease. Radiographics. 2000;20(2):399-418. 279. Caroline DF, Friedman AC. The radiology of inflammatory bowel disease. Med Clin North Am. 1994;78(6):1353-1385. 280. Carucci LR, Levine MS. Radiographic imaging of inflammatory bowel disease. Gastroenterol Clin North Am. 2002;31(1):93-117, ix.

258. Levine MS, Rubesin SE, Laufer I, et al. Diagnosis of colorectal neoplasms at double-contrast barium enema examination. Radiology. 2000;216(1):11-18.

281. Scotiniotis I, Rubesin SE, Ginsberg GG. Imaging modalities in inflammatory bowel disease. Gastroenterol Clin North Am. 1999;28(2):391-421, ix.

259. Williams AR, Balasooriya BA, Day DW. Polyps and cancer of the large bowel: a necropsy study in Liverpool. Gut. 1982;23:835-842.

282. Lichtenstein JE. Radiologic-pathologic correlation of inflammatory bowel disease. Radiol Clin North Am. 1987;25(1):3-24.

Chapter 2 Imaging of the Bowel 149 283. Use of colorectal cancer tests—United States, 2002, 2004, and 2006. MMWR Morb Mortal Wkly Rep. 2008;57(10):253-258. 284. Winawer SJ, Fletcher RH, Miller L, et al. Colorectal cancer screening: clinical guidelines and rationale. Gastroenterology. 1997;112(2):594-642. 285. Muto T, Bussey HJ, Morson BC. The evolution of cancer of the colon and rectum. Cancer. 1975;36(6):2251-2270. 286. Morson BC. Evolution of cancer of the colon and rectum. Cancer. 1974;34(3):suppl:845-849. 287. Levin B, Lieberman DA, McFarland B, et al. Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. CA Cancer J Clin. 2008;58(3):130-160. 288. Kelvin FM, Maglinte DD. Colorectal carcinoma: a radiologic and clinical review. Radiology. 1987;164(1):1-8. 289. Fischel RE, Dermer R. Multifocal carcinoma of the large intestine. Clin Radiol. 1975;26(4):495-498. 290. Levine MS, Glick SN, Rubesin SE, et al. Double-contrast barium enema examination and colorectal cancer: a plea for radiologic screening. Radiology. 2002;222(2):313-315. 291. McCarthy PA, Rubesin SE, Levine MS, et al. Colon cancer: morphology detected with barium enema examination versus histopathologic stage. Radiology. 1995;197(3):683-687. 292. National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released April 2008, based on the November 2007 submission. 293. Lieberman DA. Clinical practice. Screening for colorectal cancer. N Engl J Med. 2009;361(12):1179-1187. 294. Johns LE, Houlston RS. A systematic review and metaanalysis of familial colorectal cancer risk. Am J Gastroenterol. 2001;96(10):2992-3003. 295. Butterworth AS, Higgins JP, Pharoah P. Relative and absolute risk of colorectal cancer for individuals with a family history: a meta-analysis. Eur J Cancer. 2006;42(2):216-227. 296. Bernstein CN, Blanchard JF, Kliewer E, et al. Cancer risk in patients with inflammatory bowel disease: a population-based study. Cancer. 2001;91(4):854-862. 297. Pickhardt PJ, Choi JR, Hwang I, et al. Nonadenomatous polyps at CT colonography: prevalence, size distribution, and detection rates. Radiology. 2004;232(3):784-790. 298. Winawer SJ. Natural history of colorectal cancer. Am J Med. 1999;106(1A):3S-6S; discussion 50S-51S. 299. Horton KM, Abrams RA, Fishman EK. Spiral CT of colon cancer: imaging features and role in management. Radiographics. 2000;20(2):419-430. 300. Scott NA, Wieand HS, Moertel CG, et al. Colorectal cancer. Dukes’ stage, tumor site, preoperative plasma CEA level, and patient prognosis related to tumor DNA ploidy pattern. Arch Surg. 1987;122(12):1375-1379.

304. Willett CG, Tepper JE, Cohen AM, et al. Failure patterns following curative resection of colonic carcinoma. Ann Surg. 1984;200(6):685-690. 305. Easson AM, Barron PT, Cripps C, et al. Calcification in colorectal hepatic metastases correlates with longer survival. J Surg Oncol. 1996;63(4):221-225. 306. Kinkel K, Lu Y, Both M, et al. Detection of hepatic metastases from cancers of the gastrointestinal tract by using noninvasive imaging methods (US, CT, MR imaging, PET): a meta-analysis. Radiology. 2002;224(3):748-756. 307. Selzner M, Hany TF, Wildbrett P, et al. Does the novel PET/ CT imaging modality impact on the treatment of patients with metastatic colorectal cancer of the liver? Ann Surg. 2004;240(6):1027-1034; discussion 1035-1036. 308. Rydzewski B, Dehdashti F, Gordon BA, et al. Usefulness of intraoperative sonography for revealing hepatic metastases from colorectal cancer in patients selected for surgery after undergoing FDG PET. AJR Am J Roentgenol. 2002;178(2): 353-358. 309. de Geus-Oei LF, Vriens D, van Laarhoven HW, et al. Monitoring and predicting response to therapy with 18F-FDG PET in colorectal cancer: a systematic review. J Nucl Med. 2009;50(suppl 1):43S-54S. 310. Gollub MJ, Akhurst T, Markowitz AJ, et al. Combined CT colonography and 18F-FDG PET of colon polyps: potential technique for selective detection of cancer and precancerous lesions. AJR Am J Roentgenol. 2007;188(1):130-138. 311. Meyerhardt JA, Mayer RJ. Follow-up strategies after curative resection of colorectal cancer. Semin Oncol. 2003;30(3): 349-360. 312. Ohlsson B, Palsson B. Follow-up after colorectal cancer surgery. Acta Oncol. 2003;42(8):816-826. 313. Eddy DM. Screening for colorectal cancer. Ann Intern Med. 1990;113(5):373-384. 314. Williams CB, Macrae FA, Bartram CI. A prospective study of diagnostic methods in adenoma follow-up. Endoscopy. 1982;14(3):74-78. 315. Steine S, Stordahl A, Lunde OC, Løken K, Laerum E. Doublecontrast barium enema versus colonoscopy in the diagnosis of neoplastic disorders: aspects of decision-making in general practice. Fam Pract. 1993;10(3):288-291. 316. Pickhardt PJ. By-patient performance characteristics of CT colonography: importance of polyp size threshold data. Radiology. 2003;229(1):291-393; author reply 293; discussion 293. 317. Balthazar EJ, Megibow AJ, Hulnick D, Naidich DP. Carcinoma of the colon: detection and preoperative staging by CT. AJR Am J Roentgenol. 1988;150(2):301-306. 318. Earls JP, Colon-Negron E, Dachman AH. Colorectal carcinoma in young patients: CT detection of an atypical pattern of recurrence. Abdom Imaging. 1994;19(5):441-445.

301. Cohen AM, Minsky BD, Schilsky RL. Cancer of the colon. In: Devita VT, Hellman S, eds. Principles and Practice of Oncology. Philadelphia, PA: Lippincott-Raven; 1997:1177-1205.

319. Freeny PC, Marks WM, Ryan JA, et al. Colorectal carcinoma evaluation with CT: preoperative staging and detection of postoperative recurrence. Radiology. 1986;158(2):347-353.

302. August DA, Ottow RT, Sugarbaker PH. Clinical perspective of human colorectal cancer metastasis. Cancer Metastasis Rev. 1984;3(4):303-324.

320. Gazelle GS, Gaa J, Saini S, et al. Staging of colon carcinoma using water enema CT. J Comput Assist Tomogr. 1995;19(1): 87-91.

303. Tong D, Russell AH, Dawson LE, et al. Second laparotomy for proximal colon cancer. Sites of recurrence and implications for adjuvant therapy. Am J Surg. 1983;145(3):382-386.

321. Thompson WM, Halvorsen RA, Foster WL Jr, et al. Preoperative and postoperative CT staging of rectosigmoid carcinoma. AJR Am J Roentgenol. 1986;146(4):703-710.

150 Diagnostic Abdominal Imaging 322. Pickhardt PJ, Choi JR, Hwang I, et al. Flat colorectal lesions in asymptomatic adults: implications for screening with CT virtual colonoscopy. AJR Am J Roentgenol. 2004;183(5): 1343-1347. 323. Sato T, Konishi F, Togashi K, et al. Prospective observation of small “flat” tumors in the colon through colonoscopy. Dis Colon Rectum. 1999;42(11):1457-1463. 324. Muto T, Kamiya J, Sawada T, et al. Small “flat adenoma” of the large bowel with special reference to its clinicopathologic features. Dis Colon Rectum. 1985;28(11):847-851. 325. Fidler J, Johnson C. Flat polyps of the colon: accuracy of detection by CT colonography and histologic significance. Abdom Imaging. 2009;34(2):157-171. 326. Soetikno RM, Kaltenbach T, Rouse RV, et al. Prevalence of nonpolypoid (flat and depressed) colorectal neoplasms in asymptomatic and symptomatic adults. JAMA. 2008;299(9):1027-1035. 327. Adachi M, Muto T, Morioka Y, Ikenaga T, Hara M. Flat adenoma and flat mucosal carcinoma (IIb type)—a new precursor of colorectal carcinoma? Report of two cases. Dis Colon Rectum. 1988;31(3):236-243. 328. Fidler JL, Johnson CD, MacCarty RL, et al. Detection of flat lesions in the colon with CT colonography. Abdom Imaging. 2002;27(3):292-300. 329. Rembacken BJ, Fujii T, Cairns A, et al. Flat and depressed colonic neoplasms: a prospective study of 1000 colonoscopies in the UK. Lancet. 2000;355(9211):1211-1214. 330. Adachi M, Muto T, Okinaga K, et al. Clinicopathologic features of the flat adenoma. Dis Colon Rectum. 1991;34(11):981-986. 331. Rubesin SE, Saul S, Lauffer I, Levine M. Carpet Lesions of the Colon. Radiographics. 1985;5(4):537-552. 332. Galdino GM, Yee J. Carpet lesion on CT colonography: a potential pitfall. AJR Am J Roentgenol. 2003;180(5): 1332-1334. 333. Glick SN, Teplick SK, Balfe DM, et al. Large colonic neoplasms missed by endoscopy. AJR Am J Roentgenol. 1989;152(3):513-517. 334. Chintapalli KN, Esola CC, Chopra S, et al. Pericolic mesenteric lymph nodes: an aid in distinguishing diverticulitis from cancer of the colon. AJR Am J Roentgenol. 1997;169(5):1253-1255.

341. Abdel-Nabi H, Doerr RJ, Lamonica DM, et al. Staging of primary colorectal carcinomas with fluorine-18 fluorodeoxyglucose whole-body PET: correlation with histopathologic and CT findings. Radiology. 1998;206(3): 755-760. 342. de Bree E, Koops W, Kroger R, et al. Peritoneal carcinomatosis from colorectal or appendiceal origin: correlation of preoperative CT with intraoperative findings and evaluation of interobserver agreement. J Surg Oncol. 2004;86(2):64-73. 343. You YT, Chang Chien CR, Wang JY, et al. Evaluation of contrast-enhanced computed tomographic colonography in detection of local recurrent colorectal cancer. World J Gastroenterol. 2006;12(1):123-126. 344. Thoeni RF, Rogalla P. CT for the evaluation of carcinomas in the colon and rectum. Semin Ultrasound CT MR. 1995;16(2): 112-126. 345. Wald C, Scheirey CD, Tran TM, et al. An update on imaging of colorectal cancer. Surg Clin North Am. 2006;86(4):819-847. 346. Huebner RH, Park KC, Shepherd JE, et al. A meta-analysis of the literature for whole-body FDG PET detection of recurrent colorectal cancer. J Nucl Med. 2000;41(7):1177-1189. 347. Hine KR, Dykes PW. Serum CEA testing in the postoperative surveillance of colorectal carcinoma. Br J Cancer. 1984;49(6):689-693. 348. Flanagan FL, Dehdashti F, Ogunbiyi OA, et al. Utility of FDGPET for investigating unexplained plasma CEA elevation in patients with colorectal cancer. Ann Surg. 1998;227(3):319-323. 349. Libutti SK, Alexander HR Jr, Choyke P, et al. A prospective study of 2-[18F] fluoro-2-deoxy-D-glucose/positron emission tomography scan, 99mTc-labeled arcitumomab (CEA-scan), and blind second-look laparotomy for detecting colon cancer recurrence in patients with increasing carcinoembryonic antigen levels. Ann Surg Oncol. 2001;8(10):779-786. 350. Flamen P, Hoekstra OS, Homans F, et al. Unexplained rising carcinoembryonic antigen (CEA) in the postoperative surveillance of colorectal cancer: the utility of positron emission tomography (PET). Eur J Cancer. 2001;37(7):862-869.

335. Phatak MG, Frank SJ, Ellis JJ. Computed tomography of bowel perforation. Gastrointest Radiol. 1984;9(2):133-135.

351. Crane CH, Skibber JM, Feig BW, et al. Response to preoperative chemoradiation increases the use of sphincterpreserving surgery in patients with locally advanced low rectal carcinoma. Cancer. 2003;97(2):517-524.

336. Zerhouni EA, Rutter C, Hamilton SR, et al. CT and MR imaging in the staging of colorectal carcinoma: report of the Radiology Diagnostic Oncology Group II. Radiology. 1996;200(2):443-451.

352. Minsky BD. Sphincter preservation in rectal cancer. Preoperative radiation therapy followed by low anterior resection with coloanal anastomosis. Semin Radiat Oncol. 1998;8(1):30-35.

337. Acunas B, Rozanes I, Acunas G, et al. Preoperative CT staging of colon carcinoma (excluding the recto-sigmoid region). Eur J Radiol. 1990;11(2):150-153.

353. Kotanagi H, Fukuoka T, Shibata Y, et al. The size of regional lymph nodes does not correlate with the presence or absence of metastasis in lymph nodes in rectal cancer. J Surg Oncol. 1993;54(4):252-254.

338. McDaniel KP, Charnsangavej C, DuBrow RA, et al. Pathways of nodal metastasis in carcinomas of the cecum, ascending colon, and transverse colon: CT demonstration. AJR Am J Roentgenol. 1993;161(1):61-64. 339. Granfield CA, Charnsangavej C, Dubrow RA, et al. Regional lymph node metastases in carcinoma of the left side of the colon and rectum: CT demonstration. AJR Am J Roentgenol. 1992;159(4):757-761. 340. Kantorova I, Lipska L, Belohlavek O, et al. Routine (18)F-FDG PET preoperative staging of colorectal cancer: comparison with conventional staging and its impact on treatment decision making. J Nucl Med. 2003;44(11):1784-1788.

354. Bipat S, Glas AS, Slors FJ, et al. Rectal cancer: local staging and assessment of lymph node involvement with endoluminal US, CT, and MR imaging—a meta-analysis. Radiology. 2004;232(3):773-783. 355. Kulinna C, Eibel R, Matzek W, et al. Staging of rectal cancer: diagnostic potential of multiplanar reconstructions with MDCT. AJR Am J Roentgenol. 2004;183(2):421-427. 356. Beets-Tan RG, Beets GL. Rectal cancer: review with emphasis on MR imaging. Radiology. 2004;232(2):335-346. 357. Heriot AG, Grundy A, Kumar D. Preoperative staging of rectal carcinoma. Br J Surg. 1999;86(1):17-28.

Chapter 2 Imaging of the Bowel 151 358. Pihl E, Hughes ES, McDermott FT, et al. Disease-free survival and recurrence after resection of colorectal carcinoma. J Surg Oncol. 1981;16(4):333-341.

377. Pereira JM, Sirlin CB, Pinto PS, et al. Disproportionate fat stranding: a helpful CT sign in patients with acute abdominal pain. Radiographics. 2004;24(3):703-715.

359. Beets-Tan RG, Beets GL, Borstlap AC, et al. Preoperative assessment of local tumor extent in advanced rectal cancer: CT or high-resolution MRI? Abdom Imaging. 2000;25(5): 533-541.

378. Doria AS, Moineddin R, Kellenberger CJ, et al. US or CT for Diagnosis of Appendicitis in Children and Adults? A MetaAnalysis. Radiology. 2006;241(1):83-94.

360. Tessier DJ, McConnell EJ, Young-Fadok T, et al. Melanoma metastatic to the colon: case series and review of the literature with outcome analysis. Dis Colon Rectum. 2003;46(4):441-447. 361. Elsayed AM, Albahra M, Nzeako UC, et al. Malignant melanomas in the small intestine: a study of 103 patients. Am J Gastroenterol. 1996;91(5):1001-1006. 362. Lee HJ, Han JK, Kim TK, et al. Primary colorectal lymphoma: spectrum of imaging findings with pathologic correlation. Eur Radiol. 2002;12(9):2242-2249. 363. Yatabe Y, Nakamura S, Nakamura T, et al. Multiple polypoid lesions of primary mucosa-associated lymphoid-tissue lymphoma of colon. Histopathology. 1998;32(2):116-125. 364. Hoeffel C, Crema MD, Belkacem A, et al. Multi-detector row CT: spectrum of diseases involving the ileocecal area. Radiographics. 2006;26(5):1373-1390.

379. Barger RLJ, Nandalur KR. Diagnostic performance of magnetic resonance imaging in the detection of appendicitis in adults: a meta-analysis. Acad Radiol. 2010;17(10):1211-1216. 380. Blumenfeld YJ, Wong AE, Jafari A, Barth RA, El-Sayed YY. MR imaging in cases of antenatal suspected appendicitis—a metaanalysis. J Matern Fetal Neonatal Med. 2011;24(3):485-488. 381. Singh A, Danrad R, Hahn PF, Blake MA, Mueller PR, Novelline RA. MR imaging of the acute abdomen and pelvis: acute appendicitis and beyond. Radiographics. 2007;27(5): 1419-1431. 382. Heaston D, McClellan J, Heaston D. Community hospital experience in 600+ consecutive patients who underwent unenhanced helical CT for suspected appendicitis. AJR. 2000;174 [American Roentgen Ray Society 98th Annual Meeting Abstract Book suppl]:53.

365. Wyatt SH, Fishman EK, Hruban RH, Siegelman SS. CT of primary colonic lymphoma. Clin Imaging. 1994;18(2):131-141.

383. Hill BC, Johnson SC, Owens EK, et al. CT scan for suspected acute abdominal process: impact of combinations of IV, oral, and rectal contrast. World J Surg. 2020;34(4):699-703.

366. Lee HJ, Han JK, Kim TK, et al. Peripheral T-cell lymphoma of the colon: double-contrast barium enema examination findings in six patients. Radiology. 2001;218(3):751-756.

384. Hlibczuk V, Dattaro JA, Jin Z, et al. Diagnostic accuracy of noncontrast computed tomography for appendicitis in adults: a systematic review. Ann Emerg Med. 2010;55(1):51-59.

367. Hama Y, Okizuka H, Odajima K, et al. Gastrointestinal stromal tumor of the rectum. Eur Radiol. 2001;11(2):216-219.

385. Wise SW, Labuski MR, Kasales CJ, et al. Comparative assessment of CT and sonographic techniques for appendiceal imaging. AJR Am J Roentgenol. 2001;176(4):933-941.

368. van den Berg JC, van Heesewijk JP, van Es HW. Malignant stromal tumour of the rectum: findings at endorectal ultrasound and MRI. Br J Radiol. 2000;73(873):1010-1012. 369. Swischuk LE, Hayden CK, Boulden T. Intussusception: indications for ultrasonography and an explanation of the doughnut and pseudokidney signs. Pediatr Radiol. 1985;15(6):388-391. 370. Wong WD, Wexner SD, Lowry A, et al. Practice parameters for the treatment of sigmoid diverticulitis—supporting documentation. The Standards Task Force.The American Society of Colon and Rectal Surgeons. Dis Colon Rectum. 2000;43:290-297. 371. Perkins JD, Shield CF 3rd, Chang FC, et al. Acute diverticulitis. Comparison of treatment in immunocompromised and nonimmunocompromised patients. Am J Surg. 1984;148: 745-748. 372. Detry R, James J, Kartheuser A, et al. Acute localized diverticulitis: Optimum management requires accurate staging. Int J Colorectal Dis. 1992;7:38-42.

386. Anderson SW, Soto JA, Lucey BC, et al. Abdominal 64-MDCT for suspected appendicitis: the use of oral and IV contrast material versus IV contrast material only. AJR Am J Roentgenol. 2009;193(5):1282-1288. 387. Jacobs JE, Birnbaum BA, Macari M, et al. Acute appendicitis: comparison of helical CT diagnosis focused technique with oral contrast material versus nonfocused technique with oral and intravenous contrast material. Radiology. 2001;220(3): 683-690. 388. Curtin KR, Fitzgerald SW, Nemcek AA Jr, et al. CT diagnosis of acute appendicitis: imaging finding. Am. J. Roentgenol. 1995;164:905-990. 389. Callahan MJ, Rodriguez DP, Taylor GA. CT of Appendicitis in Children. Radiology. 2002;224(2):325-332. 390. Cobben L, Groot I, Kingma L, et al. A simple MRI protocol in patients with clinically suspected appendicitis: results in 138 patients and effect on outcome of appendectomy. Eur Radiol. 2009;19(5):1175-1183.

373. Janes S, Meagher A, Frizelle A. Elective surgery after diverticulitis. Br J Surg. 2005;92:133-142.

391. Rao PM, Wittenberg J, McDowell RK, et al. Appendicitis: use of arrowhead sign for diagnosis at CT. 1997. 1997;202:363-366.

374. Bahadursingh AM, Virgo KS, Kaminski DL, Longo WE, et al. Spectrum of disease and outcome of complicated diverticular disease. Am J Surg. 2003;2003:696-701.

392. Jeffrey RBJ, Laing FC, Townsend RR. Acute appendicitis: sonographic criteria based on 250 cases. Radiology. 1988;167:327-329.

375. Ambrosetti P, Chautems P, Soravia C, et al. Long-term outcome of mesocolic and pelvic diverticular abscesses of the left colon: a prospective study of 73 cases. Dis Colon Rectum. 2005;48:787-791.

393. Jeffrey RB, Jain KA, Nghiem HV. Sonographic diagnosis of acute appendicitis: interpretive pitfalls. AJR. 1994;162:55-59.

376. Salem L, Veenstra DL, Sullivan SD, et al. The timing of elective colectomy in diverticulitis: A decision analysis. J Am Coll Surg. 2004;199:904-912.

394. Kessler N, Cyteval C, Gallix B, et al. Appendicitis: Evaluation of Sensitivity, Specificity, and Predictive Values of US, Doppler US, and Laboratory Findings. Radiology. 2004;230:472-478. 395. Levine CD, Aizenstein O, Wachsberg RH. Pitfalls in the CT diagnosis of appendicitis. Br J Radiol. 2004;77(921):721-729.

152

Diagnostic Abdominal Imaging

396. Commane DM, Arasaradnam RP, Mills S, et al. Diet, ageing and genetic factors in the pathogenesis of diverticular disease. World J Gastroenterol. 2009;15(20):2479-2488.

418. Bradley SJ, Weaver DW, Maxwell NP, et al. Surgical management of pseudomembranous colitis. Am Surg. 1988;54(6):329-332.

397. Eide TJ, Stalsberg H. Diverticular disease of the large intestine in Northern Norway. Gut. 1979;20:609-615.

419. Fishman EK, Kavuru M, Jones B, et al. Pseudomembranous colitis: CT evaluation of 26 cases. Radiology. 1991;180(1):57-60.

398. Parks TG. The clinical significance of diverticular disease of the colon. Practitioner. 1982;226:643-648, 650-654.

420. Kawamoto S, Horton KM, Fishman EK. Pseudomembranous colitis: spectrum of imaging findings with clinical and pathologic correlation. Radiographics. 1999;19(4):887-897.

399. West BA. The pathology of diverticulosis: classical concepts and mucosal changes in diverticula. J Clin Gastroenterol. 2006;40:S126–S131. 400. Manousos O, Day NE, Tzonou A, et al. Diet and other factors in the aetiology of diverticulosis: an epidemiological study in Greece. Gut. 1985;26:544-549.

421. Ros PR, Buetow PC, Pantograg-Brown L, et al. Pseudomembranous colitis. Radiology. 1996;198(1):1-9. 422. Merine D, Fishman EK, Jones B. Pseudomembranous colitis: CT evaluation. J Comput Assist Tomogr. 1987;11(6): 1017-1020.

401. Brodribb AJ, Humphreys DM. Diverticular disease: three studies. Part I—Relation to other disorders and fibre intake. Br Med J. 1976;1:424-425.

423. Wagner ML, Rosenberg HS, Fernbach DJ, et al. Typhlitis: a complication of leukemia in childhood. Am J Roentgenol Radium Ther Nucl Med. 1970;109(2):341-350.

402. Rosemar A, Angerås U, Rosengren A. Body mass index and diverticular disease: a 28-year follow-up study in men. Dis Colon Rectum. 2008;51:450-455.

424. Wall SD, Jones B. Gastrointestinal tract in the immunocompromised host: opportunistic infections and other complications. Radiology. 1992;185(2):327-335.

403. Dobbins C, Defontgalland D, Duthie G, et al. The relationship of obesity to the complications of diverticular disease. Colorectal Dis. 2006;8:37-40.

425. Urbach DR, Rotstein OD. Typhlitis. Can J Surg. 1999;42: 415-419.

404. Painter NS, Truelove SC. The Intraluminal pressure patterns in diverticulosis of the Colon. I. Resting patterns of pressure. II. the effect of morphine. Gut. 1964;5:201-213. 405. Painter NS, Burkitt DP. Diverticular disease of the colon: a deficiency disease of Western civilization. Br Med J. 1971;2: 450-454. 406. Jung HK, Choung RS, Locke GR 3rd, et al. Diarrheapredominant irritable bowel syndrome is associated with diverticular disease: a population-based study. Am J Gastroenterol. 2010;105(3):652-661. 407. Fearnhead NS, Mortensen NJ. Clinical features and differential diagnosis of diverticular disease. Best Pract Res Clin Gastroenterol. 2002;16:577-593.

426. Gorschlüter M, Mey U, Strehl J, et al. Invasive fungal infections in neutropenic enterocolitis: a systematic analysis of pathogens, incidence, treatment and mortality in adult patients. BMC Infect Dis. 2006;6:35-45. 427. Moir CR, Scudamore CH, Benny WB. Typhlitis: selective surgical management. Am J Surg. 1986;151(5):563-566. 428. Shamberger RC, Weinstein HJ, Delorey MJ, et al. The medical and surgical management of typhlitis in children with acute nonlymphocytic (myelogenous) leukemia. Cancer. 1986;57(3):603-609. 429. Frick MP, Maile CW, Crass JR, et al. Computed tomography of neutropenic colitis. AJR Am J Roentgenol. 1984;143(4):763-765. 430. Maher D, Raviglione M. Global epidemiology of tuberculosis. Clin Chest Med. 2005;26(2):167-182, v.

408. Sugihara S, Fujii S, Kinoshita T, et al. Giant sigmoid colonic diverticulitis: case report. Abdom Imaging. 2003;28(5):640-642.

431. Burrill J, Williams CJ, Bain G, et al. Tuberculosis: a radiologic review. Radiographics. 2007;27(5):1255-1273.

409. Naing T, Ray S, Loughran CF. Giant sigmoid diverticulum: a report of three cases. Clin Radiol. 1999;54(3):179-181.

432. Lin PY, Wang JY, Hsueh PR, et al. Lower gastrointestinal tract tuberculosis: an important but neglected disease. Int J Colorectal Dis. 2009;24(10):1175-1180.

410. Versaci A, Macri A, Terranova M, et al. Volvulus due to a giant sigmoid diverticulum: a rare cause of intestinal occlusion. Chir Ital. 2008;60(3):487-491. 411. Sutorius DJ, Bossert JE. Giant sigmoid diverticulum with perforation. Am J Surg. 1974;127(6):745-748. 412. Roger T, Rommens J, Bailly J, et al. Giant colonic diverticulum: presentation of one case and review of the literature. Abdom Imaging. 1996;21(6):530-533. 413. Ona FV, Salamone RP, Mehnert PJ. Giant sigmoid diverticulitis. A cause of partial small bowel obstruction. Am J Gastroenterol. 1980;73(4):350-352. 414. Mehta DC, Baum JA, Dave PB, et al. Giant sigmoid diverticulum: report of two cases and endoscopic recognition. Am J Gastroenterol. 1996;91(6):1269-1271. 415. Grover H, Nair S, Hertan H. Giant true diverticulum of sigmoid colon. Am J Gastroenterol. 1998;93(11):2267-2268. 416. Kelly CP, Pothoulakis C, LaMont JT. Clostridium difficile colitis. N Engl J Med. 1994;330(4):257-262. 417. Lipsett PA, Samantaray DK, Tam ML, et al. Pseudomembranous colitis: a surgical disease? Surgery. 1994;116(3):491-496.

433. Lam KY, Lo CY. A critical examination of adrenal tuberculosis and a 28-year autopsy experience of active tuberculosis. Clin Endocrinol (Oxf). 2001;54(5):633-639. 434. Leder RA, Low VH. Tuberculosis of the abdomen. Radiol Clin North Am. 1995;33(4):691-705. 435. Harisinghani MG, McLoud TC, Shepard JA, et al. Tuberculosis from head to toe. Radiographics. 2000;20(2):449-470; quiz 528-529. 532. 436. Denton T, Hossain J. A radiological study of abdominal tuberculosis in a Saudi population, with special reference to ultrasound and computed tomography. Clin Radiol. 1993;47(6):409-414. 437. Healy E, Rogers S. Tuberculosis verrucosa cutis in association with bovine tuberculosis. J R Soc Med. 1992;85(11):704-705. 438. Bargallo N, Nicolau C, Luburich P, et al. Intestinal tuberculosis in AIDS. Gastrointest Radiol. 1992;17(2):115-118. 439. Zissin R, Gayer G, Chowers M, et al. Computerized tomography findings of abdominal tuberculosis: report of 19 cases. Isr Med Assoc J. 2001;3(6):414-418.

Chapter 2 Imaging of the Bowel 153 440. Suri S, Gupta S, Suri R. Computed tomography in abdominal tuberculosis. Br J Radiol. 1999;72(853):92-98. 441. Fenollar F, Puéchal X, Raoult D. Whipple’s Disease. N Engl J Med. 2007;356:55-66. 442. Horton KM, Corl FM, Fishman EK. CT of Nonneoplastic Diseases of the Small Bowel: Spectrum of Disease. J Comput Assist Tomogr. 1999;23(3):417-428. 443. Fleming JL, Wiesner RH, Shorter RG. Whipple’s disease: clinical, biochemical, and histopathologic features and assessment of treatment in 29 patients. Mayo Clin Proc. 1988:1988;63:539-551. 444. Rijke AM, Falke TH, de Vries RR. Computed tomography in Whipple disease. J Comput Assist Tomogr. 1983;7(6):1101-1102.

463. Toner M, Condell D, O’Briain D. Obstructive colitis: ulceroinflammatory lesions occurring proximal to colonic obstruction. Am J Surg Pathol. 1990;14:719-728. 464. Ko GY, Ha HK, Lee HJ, et al. Usefulness of CT in patients with ischemic colitis proximal to colonic cancer. AJR Am J Roentgenol. 1997;168(4):951-956. 465. Storesund B, Gran JT, Koldingsnes W. Severe intestinal involvement in Wegener’s granulomatosis: report of two cases and review of the literature. Br J Rheumatol. 1998;37(4):387-390. 466. Trimble MA, Weisz MA. Infarction of the sigmoid colon secondary to giant cell arteritis. Rheumatology (Oxford). 2002;41(1):108-110.

445. Herlinger H. Radiology in malabsorption. Clin Radiol. 1992; 45(2):73-78.

467. Senadhi V. A rare cause of chronic mesenteric ischemia from fibromuscular dysplasia: a case report. Med Case Reports. 2010;4:373.

446. Philpotts LE, Heiken JP, Westcott MA, et al. Colitis: use of CT findings in differential diagnosis. Radiology. 1994;190(2):445-449.

468. Collin DA, Duke O. Systemic vasculitis presenting with massive bowel infarction. J R Soc Med. 1995;88:692-693.

447. Davis TW, Goldstone SE. Sexually transmitted infections as a cause of proctitis in men who have sex with men. Dis Colon Rectum. 2009;52(3):507-512.

469. Hauer-Jensen M, Wang J, Denham JW. Bowel Injury: Current and Evolving Management Strategies. Seminars in Radiation Oncology. 2003;13(3):357-371.

448. Wiesner W, Hauser A, Steinbrich W. Accuracy of multidetector row computed tomography for the diagnosis of acute bowel ischemia in a non-selected study population. Eur Radiol. 2004;14(12):2347-2356.

470. Gervaz P, Morel P, Vozenin-Brotons MC. Molecular aspects of intestinal radiation-induced fibrosis. Curr Mol Med. 2009;9(3): 273-280.

449. Herbert GS, Steele SR. Acute and chronic mesenteric ischemia. Surg Clin North Am. 2007;87(5):1115-1134, ix. 450. Rha SE, Ha HK, Lee SH, et al. CT and MR imaging findings of bowel ischemia from various primary causes. Radiographics. 2000;20(1):29-42. 451. Chen SC, Wang HP, Chen WJ, et al. Selective use of ultrasonography for the detection of pneumoperitoneum. Acad Emerg Med. 2002;9(6):643-645. 452. Levine JS, Jacobson ED. Intestinal ischemic disorders. Dig Dis. 1995;13(1):3-24. 453. Wiesner W, Khurana B, Ji H, et al. CT of acute bowel ischemia. Radiology. 2003;226(3):635-650. 454. Horton KM, Fishman EK. Multidetector CT angiography in the diagnosis of mesenteric ischemia. Radiol Clin North Am. 2007;45(2):275-288.

471. Green N, Iba G, Smith WR. Measures to minimize small intestine injury in the irradiated pelvis. Cancer. 1975;35(6): 1633-1640. 472. Meagher T, Nolan DJ, Galland RB, et al. The radiology of irradiated gut. London: Edward Arnold. 1990;88-102. 473. Haboubi NY, El-Zammar O, O’Dwyer ST, et al. Radiation bowel disease: pathogenesis and management. Colorectal Disease. 2000;2(6):322-329. 474. Mendelson RM, Nolan DJ. The radiological features of radiation enteritis. Clin Radiol. 1985;36:141-148. 475. Fishman EK, Zinreich ES, Jones B, et al. Computed tomographic diagnosis of radiation ileitis. Gastrointest Radiol. 1984;9:149-152. 476. Baumgart DC, Carding SR. Inflammatory bowel disease: cause and immunobiology. Lancet. 2007;369(9573):1627-1640.

455. Lee R, Tung HK, Tung PH, et al. CT in acute mesenteric ischaemia. Clin Radiol. 2003;58(4):279-287.

477. Xavier RJ, Podolsky DK. Unravelling the pathogenesis of inflammatory bowel disease. Nature. 2007;448(7152): 427-434.

456. Stoker J, van Randen A, Lameris W, et al. Imaging patients with acute abdominal pain. Radiology. 2009;253(1):31-46.

478. Guindi M, Riddell RH. Indeterminate colitis. J Clin Pathol. 2004;57(12):1233-1244.

457. Thoeni RF, Cello JP. CT imaging of colitis. Radiology. 2006;240(3):623-638.

479. Horsthuis K, Stokkers PC, Stoker J. Detection of inflammatory bowel disease: diagnostic performance of cross-sectional imaging modalities. Abdom Imaging. 2008;33(4):407-416.

458. Schoots IG, Levi MM, Reekers JA, et al. Thrombolytic therapy for acute superior mesenteric artery occlusion. J Vasc Interv Radiol. 2005;16(3):317-329.

480. Wills JS, Lobis IF, Denstman FJ. Crohn disease: state of the art. Radiology. 1997;202(3):597-610.

459. Kumar S, Sarr MG, Kamath PS. Mesenteric venous thrombosis. N Engl J Med. 2001;345(23):1683-1688.

481. Roggeveen MJ, Tismenetsky M, Shapiro R. Best cases from the AFIP: Ulcerative colitis. Radiographics. 2006;26(3):947-951.

460. Ludwig KA, Quebbeman EJ, Bergstein JM, et al. Shockassociated right colon ischemia and necrosis. J Trauma. 1995;39(6):1171-1174.

482. Sheth SG, LaMont JT. Toxic megacolon. Lancet. 1998;351(9101):509-513. 483. Gore RM. CT of inflammatory bowel disease. Radiol Clin North Am. 1989;27(4):717-729.

461. Landreneau RJ, Fry WJ. The right colon as a target organ of nonocclusive mesenteric ischemia. Case report and review of the literature. Arch Surg. 1990;125(5):591-594.

484. Furukawa A, Saotome T, Yamasaki M, et al. Cross-sectional imaging in Crohn disease. Radiographics. 2004;24(3):689-702.

462. Ventemiglia R, Khalil KG, Frazier OH, et al. The role of preoperative mesenteric arteriography in colon interposition. J Thorac Cardiovasc Surg. 1977;74(1):98-104.

485. Gore RM, Balthazar EJ, Ghahremani GG, et al. CT features of ulcerative colitis and Crohn’s disease. AJR Am J Roentgenol. 1996;167(1):3-15.

154

Diagnostic Abdominal Imaging

486. Zisman TL, Rubin DT. Novel diagnostic and prognostic modalities in inflammatory bowel disease. Gastroenterol Clin North Am. 2009;38(4):729-752. 487. Nanakawa S, Takahashi M, Takagi K, et al. The role of computed tomography in management of patients with Crohn disease. Clin Imaging. 1993;17(3):193-198.

Enterography—correlation with endoscopic and histologic findings of inflammation. Radiology. 2006;238(2):505-516. 506. Fishman EK, Wolf EJ, Jones B, et al. CT evaluation of Crohn’s disease: effect on patient management. AJR Am J Roentgenol. 1987;148(3):537-540.

488. Jacobs JE, Birnbaum BA. CT of inflammatory disease of the colon. Semin Ultrasound CT MR. 1995;16(2):91-101.

507. Mako EK, Mester AR, Tarjan Z, et al. Enteroclysis and spiral CT examination in diagnosis and evaluation of small bowel Crohn’s disease. Eur J Radiol. 2000;35(3):168-175.

489. Raptopoulos V, Schwartz RK, McNicholas MM, et al. Multiplanar helical CT enterography in patients with Crohn’s disease. AJR Am J Roentgenol. 1997;169(6):1545-1550.

508. Koh DM, Miao Y, Chinn RJ, et al. MR imaging evaluation of the activity of Crohn’s disease. AJR Am J Roentgenol. 2001;177(6):1325-1332.

490. Broome U, Bergquist A. Primary sclerosing cholangitis, inflammatory bowel disease, and colon cancer. Semin Liver Dis. 2006;26(1):31-41.

509. Maccioni F, Colaiacomo MC, Parlanti S. Ulcerative colitis: value of MR imaging. Abdom Imaging. 2005;30(5):584-592.

491. Bansal P, Sonnenberg A. Risk factors of colorectal cancer in inflammatory bowel disease. Am J Gastroenterol. 1996;91(1): 44-48.

510. Madsen SM, Thomsen HS, Schlichting P, et al. Evaluation of treatment response in active Crohn’s disease by low-field magnetic resonance imaging. Abdom Imaging. 1999;24(3): 232-239.

492. Matsumoto T, Iida M, Kuroki F, et al. Dysplasia in ulcerative colitis: is radiography adequate for diagnosis? Radiology. 1996;199(1):85-90.

511. Maccioni F, Viscido A, Broglia L, et al. Evaluation of Crohn disease activity with magnetic resonance imaging. Abdom Imaging. 2000;25(3):219-228.

493. Kiran R, Khoury W, Church JM, et al. Colorectal Cancer Complicating Inflammatory Bowel Disease: Similarities and Differences Between Crohn’s and Ulcerative Colitis Based on Three Decades of Experience. Ann Surg. 2010;252:330-335.

512. Madsen SM, Thomsen HS, Munkholm P, Schlichting P, Davidsen B. Magnetic resonance imaging of Crohn disease: early recognition of treatment response and relapse. Abdom Imaging. 1997;22(2):164-166.

494. Zisman TL, Rubin DT. Colorectal cancer and dysplasia in inflammatory bowel disease. World J Gastroenterol. 2008;14(17):2662-2669.

513. Shoenut JP, Semelka RC, Magro CM, Silverman R, Yaffe CS, Micflikier AB. Comparison of magnetic resonance imaging and endoscopy in distinguishing the type and severity of inflammatory bowel disease. J Clin Gastroenterol. 1994; 19(1):31-35.

495. Rutter MD, Saunders BP, Wilkinson KH, et al. Thirty-year analysis of a colonoscopic surveillance program for neoplasia in ulcerative colitis. Gastroenterology. 2006;130:1030-1038. 496. Kandiel A, Fraser AG, Korelitz BI, et al. Increased risk of lymphoma among inflammatory bowel disease patients treated with azathioprine and 6-mercaptopurine. Gut. 2005;5(8):1121-1125.

514. Maccioni F, Bruni A, Viscido A, et al. MR imaging in patients with Crohn disease: value of T2- versus T1-weighted gadolinium-enhanced MR sequences with use of an oral superparamagnetic contrast agent. Radiology. 2006;238(2): 517-530.

497. Lewis J, Bilker WB, Brensinger C, et al. Inflammatory bowel disease is not associated with an increased risk of lymphoma. Gastroenterology. 2001;121:1080-1087.

515. Rollandi GA, Curone PF, Biscaldi E, et al. Spiral CT of the abdomen after distention of small bowel loops with transparent enema in patients with Crohn’s disease. Abdom Imaging. 1999;24(6):544-549.

498. Loftus E, Tremaine W, Habermann T, et al. Risk of lymphoma in inflammatory bowel disease. Am J Gastroenterol. 1998;95: 2308-2312.

516. Booya F, Fletcher JG, Huprich JE, et al. Active Crohn disease: CT findings and interobserver agreement for enteric phase CT enterography. Radiology. 2006;241(3):787-795.

499. Dijkstra J, Reeders JW, Tytgat GN. Idiopathic inflammatory bowel disease: endoscopic-radiologic correlation. Radiology. 1995;197(2):369-375.

517. Sempere GA, Martinez Sanjuan V, Medina Chulia E, et al. MRI evaluation of inflammatory activity in Crohn’s disease. AJR Am J Roentgenol. 2005;184(6):1829-1835.

500. Hizawa K, Iida M, Kohrogi N, et al. Crohn disease: early recognition and progress of aphthous lesions. Radiology. 1994;190(2):451-454.

518. Jones B, Fishman EK, Hamilton SR, et al. Submucosal accumulation of fat in inflammatory bowel disease: CT/ pathologic correlation. J Comput Assist Tomogr. 1986;10(5): 759-763.

501. Lee SD, Cohen RD. Endoscopy in inflammatory bowel disease. Gastroenterol Clin North Am. 2002;31(1):119-132. 502. Blackstone MO, Riddell RH, Rogers BH, et al. Dysplasiaassociated lesion or mass (DALM) detected by colonoscopy in long-standing ulcerative colitis: an indication for colectomy. Gastroenterology. 1981;80(2):366-374. 503. Loftus EV Jr. Clinical epidemiology of inflammatory bowel disease: Incidence, prevalence, and environmental influences. Gastroenterology. 2004;126(6):1504-1517. 504. Gossios KJ, Tsianos EV. Crohn disease: CT findings after treatment. Abdom Imaging. 1997;22(2):160-163. 505. Bodily KD, Fletcher JG, Solem CA, et al. Crohn Disease: mural attenuation and thickness at contrast-enhanced CT

519. Gore RM. Characteristic morphologic changes in chronic ulcerative colitis. Abdom Imaging. 1995;20(3):275-277. 520. Scott EM, Freeman AH. Prominent omental and mesenteric vasculature in inflammatory bowel disease shown by computed tomography. Eur J Radiol. 1996;22(2):104-106. 521. Ahualli J. The Fat Halo Sign. Radiology 2007. 2007;242:945-946. 522. Harisinghani MG, Wittenberg J, Lee W, et al. Bowel Wall Fat Halo Sign in Patients Without Intestinal Disease. AJR. 2003;181:781-784. 523. Herlinger H, Furth EE, Rubesin SE. Fibrofatty proliferation of the mesentery in Crohn disease. Abdom Imaging. 1998;23(4):446-448.

Chapter 2 Imaging of the Bowel 155 524. Meyers MA, McGuire PV. Spiral CT demonstration of hypervascularity in Crohn disease: “vascular jejunization of the ileum” or the “comb sign”. Abdom Imaging. 1995; 20(4):327-332. 525. Colombel JF, Solem CA, Sandborn WJ, et al. Quantitative measurement and visual assessment of ileal Crohn’s disease activity by computed tomography enterography: correlation with endoscopic severity and C reactive protein. Gut. 2006;55(11):1561-1567. 526. Yousem DM, Fishman EK, Jones B. Crohn disease: perirectal and perianal findings at CT. Radiology. 1988;167(2):331-334. 527. Ribeiro MB, Greenstein AJ, Yamazaki Y, et al. Intra-abdominal abscess in regional enteritis. Ann Surg. 1991;213(1):32-36. 528. Karahan OI, Dodd GD 3rd, Chintapalli KN, et al. Gastrointestinal wall thickening in patients with cirrhosis: frequency and patterns at contrast-enhanced CT. Radiology. 2000;215(1):103-107. 529. Kagnoff M. Celiac disease: pathogenesis of a model immunogenetic disease. J Clin Invest. 2007;117(1):41-49. 530. Lomoschitz F, Schima W, Schober E, et al. Enteroclysis in adult celiac disease: diagnostic value of specific radiographic features. Eur Radiol. 2003;13(4):890-896. 531. Rubesin SE, Herlinger H, Saul SH, et al. Adult celiac disease and its complications. Radiographics. 1989;9(6):1045-1066. 532. Soyer P, Boudiaf M, Dray X, et al. CT enteroclysis features of uncomplicated celiac disease: retrospective analysis of 44 patients. Radiology. 2009;253(2):416-424. 533. Farthing MJ, McLean AM, Bartram CI, et al. Radiologic features of the jejunum in hypoalbuminemia. AJR Am J Roentgenol. 1981;136(5):883-886. 534. Lars-Egil F, Bergseng E, Hotta K, et al. Differences in the risk of celiac disease associated with HLA-DQ2.5 or HLA-DQ2.2 are related to sustained gluten antigen presentation. Nature Immunol. 2009;10:1096-1101. 535. Fasano A, Berti I, Gerarduzzi T, et al. Prevalence of Celiac Disease in At-Risk and Not-At-Risk Groups in the United States: A Large Multicenter Study. Arch Intern Med. 2003;163:286-292. 536. Rubio-Tapia A, Murray JA. Celiac disease. Curr Opin Gastroenterol. 2010;26(2):116-122. 537. Herlinger H, Maglinte DD. Jejunal fold separation in adult celiac disease: relevance of enteroclysis. Radiology. 1986;158(3):605-611. 538. Holmes GK, Prior P, Lane MR, et al. Malignancy in coeliac disease—effect of a gluten free diet. Gut. 1989;30(3):333-338. 539. Horton KM, Fishman EK. Uncommon inflammatory diseases of the small bowel: CT findings. AJR Am J Roentgenol. 1998;170(2):385-388. 540. Mallant M, Hadithi M, Al-Toma A, et al. Abdominal computed tomography in refractory coeliac disease and enteropathy associated T-cell lymphoma. World J Gastroenterol. 2007;13(11):1696-1700. 541. Tomei E, Diacinti D, Marini M, et al. Abdominal CT findings may suggest coeliac disease. Dig Liver Dis. 2005;37(6):402-406. 542. Lane MJ, Katz DS, Mindelzun RE, et al. Spontaneous intramural small bowel haemorrhage: importance of noncontrast CT. Clin Radiol. 1997;52(5):378-380. 543. Balthazar EJ, Hulnick D, Megibow AJ, et al. Computed tomography of intramural intestinal hemorrhage and bowel ischemia. J Comput Assist Tomogr. 1987;11(1):67-72.

544. Balthazar EJ. CT of the gastrointestinal tract: principles and interpretation. AJR Am J Roentgenol. 1991;156(1):23-32. 545. Foster NM, McGory ML, Zingmond DS, et al. Small bowel obstruction: a population-based appraisal. J Am Coll Surg. 2006;203(2):170-176. 546. Welch JP. Bowel Obstruction: Differential Diagnosis and Clinical Management. Philadelphia: Saunders; 1990:xvi, 711. 547. Silva AC, Pimenta M, Guimaraes LS. Small bowel obstruction: what to look for. Radiographics. 2009;29(2):423-439. 548. Peetz DJ Jr, Gamelli RL, Pilcher DB. Intestinal intubation in acute, mechanical small-bowel obstruction. Arch Surg. 1982;117(3):334-336. 549. Silen W, Hein MF, Goldman L. Strangulation obstruction of the small intestine. Arch Surg. 1962;85:121-129. 550. Balthazar EJ. For suspected small-bowel obstruction and an equivocal plain film, should we perform CT or a small-bowel series? AJR Am J Roentgenol. 1994;163(5):1260-1261. 551. Lappas JC, Reyes BL, Maglinte DD. Abdominal radiography findings in small-bowel obstruction: relevance to triage for additional diagnostic imaging. AJR Am J Roentgenol. 2001;176(1):167-174. 552. Nicolaou S, Kai B, Ho S, et al. Imaging of acute small-bowel obstruction. AJR Am J Roentgenol. 2005;185(4):1036-1044. 553. Thompson WM, Kilani RK, Smith BB, et al. Accuracy of abdominal radiography in acute small-bowel obstruction: does reviewer experience matter? AJR Am J Roentgenol. 2007;188(3):W233-W238. 554. Maglinte DD, Heitkamp DE, Howard TJ, et al. Current concepts in imaging of small bowel obstruction. Radiol Clin North Am. 2003;41(2):vi, 263-283. 555. Fukuya T, Hawes DR, Lu CC, et al. CT diagnosis of small-bowel obstruction: efficacy in 60 patients. AJR Am J Roentgenol. 1992;158(4):765-769; discussion 771-772. 556. Qalbani A, Paushter D, Dachman AH. Multidetector row CT of small bowel obstruction. Radiol Clin North Am. 2007;45(3):viii, 499-512. 557. Ros PR, Huprich JE. ACR Appropriateness Criteria on suspected small-bowel obstruction. J Am Coll Radiol. 2006;3(11):838-841. 558. Jaffe TA, Martin LC, Thomas J, et al. Small-bowel obstruction: coronal reformations from isotropic voxels at 16-section multidetector row CT. Radiology. 2006;238(1):135-142. 559. Frager D, Medwid SW, Baer JW, et al. CT of small-bowel obstruction: value in establishing the diagnosis and determining the degree and cause. AJR Am J Roentgenol. 1994;162(1):37-41. 560. Maglinte DD, Kelvin FM, Rowe MG, et al. Small-bowel obstruction: optimizing radiologic investigation and nonsurgical management. Radiology. 2001;218(1):39-46. 561. Maglinte DD, Stevens LH, Hall RC, et al. Dual-purpose tube for enteroclysis and nasogastric-nasoenteric decompression. Radiology. 1992;185(1):281-282. 562. Shrake PD, Rex DK, Lappas JC, et al. Radiographic evaluation of suspected small bowel obstruction. Am J Gastroenterol. 1991;86(2):175-178. 563. Makanjuola D. Computed tomography compared with small bowel enema in clinically equivocal intestinal obstruction. Clin Radiol. 1998;53(3):203-208. 564. Maglinte DD, Sandrasegaran K, Lappas JC, et al. CT Enteroclysis. Radiology. 2007;245(3):661-671.

156

Diagnostic Abdominal Imaging

565. Schmidt H, Abolmaali N, Vogl TJ. Double bubble sign. Eur Radiol. 2002;12(7):1849-1853.

587. Agha FP. Intussusception in adults. AJR Am J Roentgenol. 1986;146(3):527-531.

566. Maglinte DD, Reyes BL, Harmon BH, et al. Reliability and role of plain film radiography and CT in the diagnosis of smallbowel obstruction. AJR Am J Roentgenol. 1996;167(6):1451-1455.

588. Chiorean MV, Sandrasegaran K, Saxena R, et al. Correlation of CT enteroclysis with surgical pathology in Crohn’s disease. Am J Gastroenterol. 2007;102(11):2541-2550.

567. Levin B. Mechanical small bowel obstruction. Semin Roentgenol. 1973;8(3):281-297.

589. Zissin R, Hertz M, Paran H, et al. Small bowel obstruction secondary to Crohn disease: CT findings. Abdom Imaging. 2004;29(3):320-325.

568. Nevitt PC. The string of pearls sign. Radiology. 2000;214(1): 157-158. 569. Herlinger H, Maglinte DDT. Clinical Radiology of the Small Intestine. Philadelphia, PA: Saunders; 1989;497-507. 570. Furukawa A, Yamasaki M, Takahashi M, et al. CT diagnosis of small bowel obstruction: scanning technique, interpretation and role in the diagnosis. Semin Ultrasound CT MR. 2003;24(5):336-352. 571. Furukawa A, Yamasaki M, Furuichi K, et al. Helical CT in the diagnosis of small bowel obstruction. Radiographics. 2001;21(2):341-355. 572. Lee JKT. Computed Body Tomography with MRI Correlation. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2006. 573. Lazarus DE, Slywotsky C, Bennett GL, et al. Frequency and relevance of the “small-bowel feces” sign on CT in patients with small-bowel obstruction. AJR Am J Roentgenol. 2004;183(5):1361-1366. 574. Jacobs SL, Rozenblit A, Ricci Z, et al. Small bowel faeces sign in patients without small bowel obstruction. Clin Radiol. 2007;62(4):353-357. 575. Fuchsjager MH. The small-bowel feces sign. Radiology. 2002;225(2):378-379. 576. Mayo-Smith WW, Wittenberg J, Bennett GL, et al. The CT small bowel faeces sign: description and clinical significance. Clin Radiol. 1995;50(11):765-767. 577. Quiroga S, Alvarez-Castells A, Sebastia MC, et al. Small bowel obstruction secondary to bezoar: CT diagnosis. Abdom Imaging. 1997;22(3):315-317. 578. Miller G, Boman J, Shrier I, et al. Etiology of small bowel obstruction. Am J Surg. 2000;180(1):33-36. 579. Delabrousse E, Destrumelle N, Brunelle S, et al. CT of small bowel obstruction in adults. Abdom Imaging. 2003;28(2): 257-266. 580. Attard JA, MacLean AR. Adhesive small bowel obstruction: epidemiology, biology and prevention. Can J Surg. 2007; 50(4):291-300. 581. Boudiaf M, Soyer P, Terem C, et al. Ct evaluation of small bowel obstruction. Radiographics. 2001;21(3):613-624.

590. Reijnen HA, Joosten HJ, de Boer HH. Diagnosis and treatment of adult intussusception. Am J Surg. 1989;158(1):25-28. 591. Herlinger H, Maglinte DDT. Clinical Radiology of the Small Intestine. Philadelphia, PA: Saunders; 1989:xviii, 605. 592. Warshauer DM, Lee JK. Adult intussusception detected at CT or MR imaging: clinical-imaging correlation. Radiology. 1999;212(3):853-860. 593. Lorigan JG, DuBrow RA. The computed tomographic appearances and clinical significance of intussusception in adults with malignant neoplasms. Br J Radiol. 1990;63(748):257-262. 594. Iko BO, Teal JS, Siram SM, et al. Computed tomography of adult colonic intussusception: clinical and experimental studies. AJR Am J Roentgenol. 1984;143(4):769-772. 595. Yoshimitsu K, Fukuya T, Onitsuka H, et al. Computed tomography of ileoileocolic intussusception caused by a lipoma. J Comput Assist Tomogr. 1989;13(4):704-706. 596. Peterson CM, Anderson JS, Hara AK, et al. Volvulus of the gastrointestinal tract: appearances at multimodality imaging. Radiographics. 2009;29(5):1281-1293. 597. McAlister WH, Kronemer KA. Emergency gastrointestinal radiology of the newborn. Radiol Clin North Am. 1996;34(4): 819-844. 598. Bernstein SM, Russ PD. Midgut volvulus: a rare cause of acute abdomen in an adult patient. AJR Am J Roentgenol. 1998;171(3):639-641. 599. Shimanuki Y, Aihara T, Takano H, et al. Clockwise whirlpool sign at color Doppler US: an objective and definite sign of midgut volvulus. Radiology. 1996;199(1):261-264. 600. Fisher JK. Computed tomographic diagnosis of volvulus in intestinal malrotation. Radiology. 1981;140(1):145-146. 601. Loren I, Lasson A, Nilsson A, et al. Gallstone ileus demonstrated by CT. J Comput Assist Tomogr. 1994;18(2):262-265. 602. Mishra PK, Agrawal A, Joshi M, et al. Intestinal obstruction in children due to Ascariasis: A tertiary health centre experience. Afr J Paediatr Surg. 2008;5:65-70.

582. Sinha R, Verma R. Multidetector row computed tomography in bowel obstruction. Part 2. Large bowel obstruction. Clin Radiol. 2005;60(10):1068-1075.

603. Haswell-Elkins M, Elkins D, Anderson RM. The influence of individual, social group and household factors on the distribution of Ascaris lumbricoides within a community and implications for control strategies. Parasitology. 1989;98:125-134.

583. Zafar HM, Levine MS, Rubesin SE, et al. Anterior abdominal wall hernias: findings in barium studies. Radiographics. 2006;26(3):691-699.

604. Shiekh KA, Baba AA, Ahmad SM, et al. Mechanical small bowel obstruction in children at a tertiary care centre in Kashmir. Afr J Paediatr Surg. 2010;7(2):81-85.

584. Takeyama N, Gokan T, Ohgiya Y, et al. CT of internal hernias. Radiographics. 2005;25(4):997-1015.

605. Dayalan, N, Ramakrishnan MS. The pattern of intestinal obstruction with special preference toascariasis. Indian Pediatr. 1976;13:47-49.

585. Martin LC, Merkle EM, Thompson WM. Review of internal hernias: radiographic and clinical findings. AJR Am J Roentgenol. 2006;186(3):703-717. 586. Mak SY, Roach SC, Sukumar SA. Small bowel obstruction: computed tomography features and pitfalls. Curr Probl Diagn Radiol. 2006;35(2):65-74.

606. Hoffmann H, Kawooya M, Esterre P, et al. In vivo and in vitro studies of the sonographic detection of Ascaris lumbricoides. Pediatr Radiol. 1997;27:226-229. 607. Peck RJ. Ultrasonography of intestinal Ascaris. J Clin Ultrasound. 1990;18:741-743.

Chapter 2 Imaging of the Bowel 157 608. Mahmood T, Mansoor N, Quraishy S, et al. Ultrasonographic appearance of Ascaris lumbricoides in the small bowel. J Ultrasound Med. 2001;20:269-274. 609. Hommeyer SC, Hamill GS, Johnson JA. CT diagnosis of intestinal ascariasis. Abdom Imaging. 1995;20(4):315-316. 610. Beita AO, Haller JO, Kantor A. CT Findings in pediatric gastrointestinal ascariasis. Comput Med Imaging Graph. 1997;21:47-49. 611. Dalla VLK, Grosfeld JL, West KW, et al. Intestinal atresia and stenosis: a 25-year experience with 277 cases. Arch Surg. 1998;133(5):490-496. 612. Megibow AJ. Bowel obstruction. Evaluation with CT. Radiol Clin North Am. 1994;32(5):861-870. 613. Chakrabarty PB, Tripathy BC, Panda K. Acute intestinal obstruction (a review of 1020 operated cases). J Indian Med Assoc. 1976;67(3):64-69. 614. Laws HL, Aldrete JS. Small-bowel obstruction: a review of 465 cases. South Med J. 1976;69(6):733-734. 615. Cappell MS, Batke M. Mechanical obstruction of the small bowel and colon. Med Clin North Am. 2008;92(3):viii, 575-597. 616. Chapman AH, McNamara M, Porter G. The acute contrast enema in suspected large bowel obstruction: value and technique. Clin Radiol. 1992;46(4):273-278. 617. Frager D, Rovno HD, Baer JW, et al. Prospective evaluation of colonic obstruction with computed tomography. Abdom Imaging. 1998;23(2):141-146. 618. Beattie GC, Peters RT, Guy S, et al. Computed tomography in the assessment of suspected large bowel obstruction. ANZ J Surg. 2007;77(3):160-165. 619. Taourel P, Kessler N, Lesnik A, et al. Helical CT of large bowel obstruction. Abdom Imaging. 2003;28(2):267-275. 620. Ott DJ, Chen MY. Specific acute colonic disorders. Radiol Clin North Am. 1994;32(5):871-884.

630. Pal S, Sahni P, Pande GK, et al. Outcome following emergency surgery for refractory severe ulcerative colitis in a tertiary care centre in India. BMC Gastroenterol. 2005;5:39. 631. Eun SK, Won HK. Inflammatory bowel disease in Korea: epidemiological, genomic, clinical, and therapeutic characteristics. Gut Liver. 2010;4(1):1-14. 632. Byun YH, Park YS, Myung SJ, et al. Transient intestinal obstruction due to stool impaction in the elderly. Korean J Gastroenterol. 2005;46(3):211-217. 633. Khan MN, Agrawal A, Strauss P. Ileocolic Intussusception - A rare cause of acute intestinal obstruction in adults; Case report and literature review. World J Emerg Surg. 2008;3:26. 634. Kessman J. Hirschsprung’s disease: diagnosis and management. Am Fam Physician. 2006;74(8):1319-1322. 635. Simi M, Pietroletti R, Navarra L, et al. Bowel stricture due to ischemic colitis: report of three cases requiring surgery. Hepatogastroenterology. 1995;42(3):279-281. 636. Osman N, Subar D, Loh MY, et al. Gallstone ileus of the sigmoid colon: an unusual cause of large-bowel obstruction. HPB Surg. 2010;2010:153740. 637. Chen JH, Chen KY, Chang WK. Intestinal obstruction induced by phytobezoars. CMAJ. 2010;182(17):E797. 638. Indraccolo U, Trevisan P, Gasparin P, et al. Cecal endometriosis as a cause of ileocolic intussusception. JSLS. 2010;14(1):140-142. 639. Pyun DK, Kim KJ, Ye BD, et al. Two cases of colonic obstruction after acute pancreatitis. Korean J Gastroenterol. 2009;54(3):180-185. 640. Dietz DW, Remzi FH, Fazio VW. Strictureplasty for obstructing small-bowel lesions in diffuse radiation enteritis—successful outcome in five patients. Dis Colon Rectum. 2001;44(12): 1772-1777. 641. Aronchick JM, Epstein DM, Gefter WB, et al. Chronic traumatic diaphragmatic hernia: the significance of pleural effusion. Radiology. 1988;168(3):675-678.

621. Feldman M, Friedman LS, Sleisenger MH. Sleisenger & Fordtran’s Gastrointestinal and Liver Disease: Pathophysiology, Diagnosis, Management. 7th ed. Philadelphia, PA: Saunders; 2002. 2 v. (xli, 2385. 98 p.).

642. Batke M, Cappell MS. Adynamic ileus and acute colonic pseudo-obstruction. Med Clin North Am. 2008;92(3): 649-670, ix.

622. Jones IT, Fazio VW. Colonic volvulus. Etiology and management. Dig Dis. 1989;7(4):203-209.

643. Holte K, Kehlet H. Postoperative ileus: a preventable event. Br J Surg. 2000;87(11):1480-1493.

623. Javors BR, Baker SR, Miller JA. The northern exposure sign: a newly described finding in sigmoid volvulus. AJR Am J Roentgenol. 1999;173(3):571-574.

644. Townsend CM, Sabiston DC. Sabiston Textbook of Surgery : the Biological Basis of Modern Surgical Practice. Philadelphia, PA: Saunders; 2004:2388, xxv.

624. Burrell HC, Baker DM, Wardrop P, et al. Significant plain film findings in sigmoid volvulus. Clin Radiol. 1994;49(5):317-319.

645. Kurz A, Sessler DI. Opioid-induced bowel dysfunction: pathophysiology and potential new therapies. Drugs. 2003;63(7):649-671.

625. Moore CJ, Corl FM, Fishman EK. CT of cecal volvulus: unraveling the image. AJR Am J Roentgenol. 2001;177(1): 95-98. 626. Frank AJ, Goffner LB, Fruauff AA, et al. Cecal volvulus: the CT whirl sign. Abdom Imaging. 1993;18(3):288-289. 627. Ng DC, Kwok SY, Cheng Y, et al. Colonic amoebic abscess mimicking carcinoma of the colon. Hong Kong Med J. 2006; 12(1):71-73. 628. Filippou D, Psimitis I, Zizi D, et al. A rare case of ascending colon actinomycosis mimicking cancer. BMC Gastroenterol. 2005;5:1. 629. Platell C, Mackay J, Collopy B, et al. Crohn’s disease: a colon and rectal department experience. Aust N Z J Surg. 1995;65(8):570-575.

646. Fallon MT, Hanks GW. Morphine, constipation and performance status in advanced cancer patients. Palliat Med. 1999;13(2):159-160. 647. Kahi CJ, Rex DK. Bowel obstruction and pseudo-obstruction. Gastroenterol Clin North Am. 2003;32(4):1229-1247. 648. Saunders MD, Kimmey MB. Systematic review: acute colonic pseudo-obstruction. Aliment Pharmacol Ther. 2005;22(10): 917-925. 649. Vanek VW, Al-Salti M. Acute pseudo-obstruction of the colon (Ogilvie’s syndrome). An analysis of 400 cases. Dis Colon Rectum. 1986;29(3):203-210. 650. De Giorgio R, Knowles CH. Acute colonic pseudo-obstruction. Br J Surg. 2009;96(3):229-239.

158 Diagnostic Abdominal Imaging 651. Lavine L. Gastrointestinal Bleeding. In: Fauci AS, et al. eds. Harrison’s Principles of Internal Medicine, 17th ed. New York, NY: McGraw Hill; 2008: Chap 42. 652. Imdahl A. Genesis and pathophysiology of lower gastrointestinal bleeding. Langenbecks Arch Surg. 2001;386(1):1-7. 653. Longstreth GF. Epidemiology and outcome of patients hospitalized with acute lower gastrointestinal hemorrhage: a population-based study. Am J Gastroenterol. 1997;92(3):419-424. 654. Barnert J, Messmann H. Diagnosis and management of lower gastrointestinal bleeding. Nat Rev Gastroenterol Hepatol. 2009;6(11):637-646. 655. Zuckier LS. Acute gastrointestinal bleeding. Semin Nucl Med. 2003;33(4):297-311. 656. Fallah MA, Prakash C, Edmundowicz S. Acute gastrointestinal bleeding. Med Clin North Am. 2000;84(5):1183-1208. 657. Ettorre GC, Francioso G, Garribba AP, et al. Helical CT angiography in gastrointestinal bleeding of obscure origin. AJR Am J Roentgenol. 1997;163(3):727-731. 658. Tew K, Davies RP, Jadun CK, et al. MDCT of acute lower gastrointestinal bleeding. AJR Am J Roentgenol. 2004;182(2):427-430. 659. Ernst O, Bulois P, Saint-Drenant S, Leroy C, Paris JC, Sergent G. Helical CT in acute lower gastrointestinal bleeding. Eur Radiol. 2003;13(1):114-117. 660. Zink SI, Ohki SK, Stein B, et al. Noninvasive evaluation of active lower gastrointestinal bleeding: comparison between contrast-enhanced MDCT and 99mTc-labeled RBC scintigraphy. AJR Am J Roentgenol. 2008;191(4):1107-1114. 661. Jaeckle T, Stuber G, Hoffmann MH, et al. Detection and localization of acute upper and lower gastrointestinal (GI) bleeding with arterial phase multi-detector row helical CT. Eur Radiol. 2008:2008;18(7):1406-1413. 662. Cohn SM, Moller BA, Zieg PM, et al. Angiography for preoperative evaluation in patients with lower gastrointestinal bleeding: are the benefits worth the risks. Arch Surg. 1998;133(1):50-55. 663. Mularski RA, Sippel JM, Osborne ML. Pneumoperitoneum: a review of nonsurgical causes. Crit Care Med. 2000;28(7):2638-2644.

670. Levine MS, Scheiner JD, Rubesin SE, et al. Diagnosis of pneumoperitoneum on supine abdominal radiographs. AJR Am J Roentgenol. 1991;156(4):731-735. 671. Menuck L, Siemers PT. Pneumoperitoneum: importance of right upper quadrant features. AJR Am J Roentgenol. 1976;127(5):753-756. 672. Earls JP, Dachman AH, Colon E, et al. Prevalence and duration of postoperative pneumoperitoneum: sensitivity of CT vs left lateral decubitus radiography. AJR Am J Roentgenol. 1993;161(4):781-785. 673. Stapakis JC, Thickman D. Diagnosis of pneumoperitoneum: abdominal CT vs. upright chest film. J Comput Assist Tomogr. 1992;16(5):713-716. 674. Feczko PJ, Mezwa DG, Farah MC, et al. Clinical significance of pneumatosis of the bowel wall. Radiographics. 1992;12(6): 1069-1078. 675. Ho LM, Paulson EK, Thompson WM. Pneumatosis intestinalis in the adult: benign to life-threatening causes. AJR Am J Roentgenol. 2007;188(6):1604-1613. 676. Lund EC, Han SY, Holley HC, et al. Intestinal ischemia: comparison of plain radiographic and computed tomographic findings. Radiographics. 1988;8(6):1083-1108. 677. Ho LM, Mosca PJ, Thompson WM. Pneumatosis intestinalis after lung transplant. Abdom Imaging. 2005;30(5):598-600. 678. Hwang J, Reddy VS, Sharp KW. Pneumatosis cystoides intestinalis with free intraperitoneal air: a case report. Am Surg. 2003;69(4):346-349. 679. Wood BJ, Kumar PN, Cooper C, et al. Pneumatosis intestinalis in adults with AIDS: clinical significance and imaging findings. AJR Am J Roentgenol. 1995;165(6):1387-1390. 680. Pear BL. Pneumatosis intestinalis: a review. Radiology. 1998;207(1):13-19. 681. Galandiuk S, Fazio VW. Pneumatosis cystoides intestinalis. A review of the literature. Dis Colon Rectum. 1986;29(5): 358-363. 682. Berk JE, Bockus HL. Bockus gastroenterology. 4th ed. Philadelphia, PA: Saunders; 1985:1-7, v. 683. Koss LG. Abdominal gas cysts (pneumatosis cystoides intestinorum hominis); an analysis with a report of a case and a critical review of the literature. AMA Arch Pathol. 1952;53(6):523-549.

664. Gayer G, Hertz M, Zissin R, et al. Postoperative pneumoperitoneum as detected by CT: prevalence, duration, and relevant factors affecting its possible significance. Abdom Imaging. 2000;25(3):301-305.

684. St Peter SD, Abbas MA, Kelly KA. The spectrum of pneumatosis intestinalis. Arch Surg. 2003;138(1):68-75.

665. Gayer G, Hertz M, Zissin R. Postoperative pneumoperitoneum: prevalence, duration, and possible significance. Semin Ultrasound CT MR. 2004;25(3):286-289.

685. Caudill JL, Rose BS. The role of computed tomography in the evaluation of pneumatosis intestinalis. J Clin Gastroenterol. 1987;9(2):223-226.

666. Cho KC, Baker SR. Extraluminal air. Diagnosis and significance. Radiol Clin North Am. 1994;32(5):829-844.

686. Connor R, Jones B, Fishman EK, et al. Pneumatosis intestinalis: role of computed tomography in diagnosis and management. J Comput Assist Tomogr. 1984;8(2):269-275.

667. Roh JJ, Thompson JS, Harned RK, et al. Value of pneumoperitoneum in the diagnosis of visceral perforation. Am J Surg. 1983;146(6):830-833. 668. Winek TG, Mosely HS, Grout G, et al. Pneumoperitoneum and its association with ruptured abdominal viscus. Arch Surg. 1988;123(6):709-712. 669. Miller RE, Nelson SW. The roentgenologic demonstration of tiny amounts of free intraperitoneal gas: experimental and clinical studies. Am J Roentgenol Radium Ther Nucl Med. 1971;112(3):574-585.

687. Federle MP, Chun G, Jeffrey RB, et al. Computed tomographic findings in bowel infarction. AJR Am J Roentgenol. 1984;142(1): 91-95. 688. Hutchins WW, Gore RM, Foley MJ. CT demonstration of pneumatosis intestinalis from bowel infarction. Comput Radiol. 1983;7(5):283-285. 689. Kelvin FM, Korobkin M, Rauch RF, et al. Computed tomography of pneumatosis intestinalis. J Comput Assist Tomogr. 1984;8(2):276-280.

Chapter 2 Imaging of the Bowel 159 690. Meyers MA, Ghahremani GG, Clements JL Jr, et al. Pneumatosis intestinalis. Gastrointest Radiol. 1977;2(2):91-105.

701. Huurman VA, Visser LG, Steens SC, et al. Persistent portal venous gas. J Gastrointest Surg. 2006;10(5):783-785.

691. Schindera ST, Triller J, Vock P, et al. Detection of hepatic portal venous gas: its clinical impact and outcome. Emerg Radiol. 2006;12(4):164-170.

702. Griffith J, Apostolakos M, and Salloum RM. Pneumatosis intestinalis and gas in the portal venous system. J Gastrointest Surg. 2006;10(5):781-782.

692. Smerud MJ, Johnson CD, Stephens DH. Diagnosis of bowel infarction: a comparison of plain films and CT scans in 23 cases. AJR Am J Roentgenol. 1990;154(1):99-103.

703. Ghahremani GG, White EM, Hoff FL, et al. Appendices epiploicae of the colon: radiologic and pathologic features. Radiographics. 1992;12(1):59-77.

693. Knechtle SJ, Davidoff AM, Rice RP. Pneumatosis intestinalis. Surgical management and clinical outcome. Ann Surg. 1990;212(2):160-165.

704. Legome EL, Sims C, Rao PM. Epiploic appendagitis: adding to the differential of acute abdominal pain. J Emerg Med. 1999;17(5):823-826.

694. Fisher JK. Computed tomography of colonic pneumatosis intestinalis with mesenteric and portal venous air. J Comput Assist Tomogr. 1984;8(3):573-574.

705. Rao PM, Wittenberg J, Lawrason JN. Primary epiploic appendagitis: evolutionary changes in CT appearance. Radiology. 1997;204(3):713-717.

695. Epelman M, Daneman A, Navarro OM, et al. Necrotizing enterocolitis: review of state-of-the-art imaging findings with pathologic correlation. Radiographics. 2007;27(2):285-305.

706. Torres GM, Abbitt PL, Weeks M. CT manifestations of infarcted epiploic appendages of the colon. Abdom Imaging. 1994;19(5):449-450.

696. Wiesner W, Mortele KJ, Glickman JN, et al. Pneumatosis intestinalis and portomesenteric venous gas in intestinal ischemia: correlation of CT findings with severity of ischemia and clinical outcome. AJR Am J Roentgenol. 2001;177(6): 1319-1323.

707. Legome EL, Belton AL, Murray RE, et al. Epiploic appendagitis: the emergency department presentation. J Emerg Med. 2002;22(1):9-13.

697. Kernagis LY, Levine MS, Jacobs JE. Pneumatosis intestinalis in patients with ischemia: correlation of CT findings with viability of the bowel. AJR Am J Roentgenol. 2003;180(3): 733-736. 698. Liebman PR, Patten MT, Manny J, et al. Hepatic—portal venous gas in adults: etiology, pathophysiology and clinical significance. Ann Surg. 1978;187(3):281-287. 699. Wiesner W, Mortele KJ, Glickman JN, et al. Portal-venous gas unrelated to mesenteric ischemia. Eur Radiol. 2002;12(6): 1432-1437. 700. Hou SK, Chern CH, How CK, et al. Hepatic portal venous gas: clinical significance of computed tomography findings. Am J Emerg Med. 2004;22(3):214-218.

708. van Breda Vriesman AC, de Mol van Otterloo AJ, Puylaert JB. Epiploic appendagitis and omental infarction. Eur J Surg. 2001;167(10):723-727. 709. Epstein LI, Lempke RE. Primary idiopathic segmental infarction of the greater omentum: case report and collective review of the literature. Ann Surg. 1968;167(3): 437-443. 710. Puylaert JB. Right-sided segmental infarction of the omentum: clinical, US, and CT findings. Radiology. 1992;185(1):169-172. 711. Puylaert JB. Ultrasonography of the acute abdomen: gastrointestinal conditions. Radiol Clin North Am. 2003;41(6):1227-1242, vii.

This page intentionally left blank

CHAPTER

3

Imaging of the Liver Mark Alan Rosen, MD, PhD Stanley Chan, MD Wallace T. Miller Jr., MD

I. NORMAL HEPATIC ANATOMY II. IMAGING OF THE LIVER a. US Evaluation of the Liver b. CT Evaluation of the Liver c. MRI of the Liver d. Multiphase Vascular Imaging of the Liver with CT and MRI III. UNIFOCAL, SPHERICAL LIVER LESIONS a. Solitary Cysts and Cystic-Appearing Lesions i. Idiopathic cysts ii. Solitary cystic or cystic appearing neoplasms iii. Abscesses b. Solitary Solid Hepatic Masses or Nodules i. Solitary solid appearing hepatic neoplasms ii. Peliosis hepatis iii. Inflammatory pseudotumor iv. Pseudolymphoma v. Hepatic pseudolesions IV. MULTIPLE MASSES OR NODULES OF THE LIVER a. Multiple Cystic or Cystic-Appearing Liver Masses i. Idiopathic cysts ii. Polycystic liver disease iii. Multiple cystic or cystic-appearing neoplasms iv. Multifocal abscesses b. Multiple Solid Hepatic Masses or Nodules i. Neoplasms appearing as multiple solid masses or nodules ii. Granulomatous diseases of the liver iii. Peliosis hepatis V. NONSPHERICAL, FOCAL LESIONS OF THE LIVER a. Liver Trauma b. Spontaneous (Nontraumatic) Hemorrhage c. Hepatic Infarction d. Arterial-Portal Shunting (Transient Hepatic Attenuation [or Intensity] Difference) e. Focal Steatosis and Focal Fatty Sparing VI. DIFFUSE LIVER DISEASES a. Hepatomegaly i. Acute hepatitis

ii. Hematologic malignancies iii. Sarcoidosis iv. Gaucher disease v. Glycogen storage disease vi. Amyloidosis b. Liver Atrophy c. Hepatic Steatosis d. Iron Deposition e. Amiodarone Deposition VII. HEPATIC DISORDERS WITH A SPECIFIC APPEARANCE a. The Cirrhotic Liver and Hepatocellular Carcinoma i. Etiology and evolution of cirrhosis ii. Secondary findings in chronic liver inflammation and early cirrhosis iii. The end-stage cirrhotic liver iv. Portal hypertension, thrombosis, and ascites v. Regenerative nodules and HCC vi. Hepatocellular carcinoma b. Pseudocirrhosis c. Radiation Hepatitis d. Hepatic Congestion, Budd-Chiari Syndrome, and the Nutmeg Liver e. Vascular Malformations f. Posttraumatic Fistula VIII. IMAGING THE POSTOPERATIVE LIVER a. Nonsurgical Therapies i. Coil embolization ii. Transjugular intrahepatic portosystemic shunt placement iii. Transarterial chemoembolization iv. Percutaneous ablation b. Surgical Treatment of the Liver i. Wedge resection ii. Hepatic segmentectomy and lobectomy iii. Evaluation of the potential partial liver donor c. Liver Transplantation i. Normal transplant appearance ii. Complications of liver transplantation 161

162

Diagnostic Abdominal Imaging

NORMAL HEPATIC ANATOMY The liver primarily lies in the right upper quadrant, underneath the right hemidiaphragm, although in most individuals the left portion of the liver will cross the midline of the upper abdomen into the left upper quadrant. With the exception of a small portion of the posterior liver bordering the diaphragm, the liver is invested by the peritoneum, and is thus a peritoneal organ. The surface anatomy of the liver is categorized by various peritoneal reflections, including the falciform ligament (residual ligaments from the embryonic coelomic cavity reflection), the ligamentum teres (peritoneal investment about the residual umbilical vein), and the ligamentum venosum (peritoneal investment about the residual ductus venosum between the left portal vein and the inferior vena cava [IVC]). Posteriorly and inferiorly, the liver is bordered by the gallbladder (within the gallbladder fossa), the duodenum, the hepatic flexure of the colon, and the IVC. The falciform ligament connects the liver to the diaphragm, and divides the anatomic right and left lobes. The ligamentum teres extends inferiorly from the falciform ligament toward the anterior abdominal wall. The liver also has attachments through the gastrohepatic ligament (which connect the liver to the stomach, and through which the left gastric artery and vein run), and the hepaticoduodenal ligament (connecting the liver to the duodenum, and through which run the gastroduodenal artery and vein). The liver is functionally served by 2 distinct vascular supplies, the hepatic artery and the portal vein, with blood exiting via the 3 hepatic veins (right, middle, and left) and into the IVC. Functionally, the liver is divided into right and left lobes (distinct from the anatomic lobes). The functional left lobe is supplied by the left hepatic and left portal vein branches, whereas the right lobe is supplied by the right hepatic and right portal vein branches. The boundary between the right and left lobes on cross-sectional imaging is defined by the middle hepatic vein superiorly, and the gallbladder fossa inferiorly. The left lobe is divided by the falciform ligament into the left lateral segment and the left medial segment (also termed the quadrate lobe). The right lobe is divided by the right hepatic vein into the anterior and posterior segments. The caudate lobe is a small, more isolated area of the liver between the portal vein and the ligamentum venosum and has a more variable arterial and portal vein supply, as well as variant venous drainage. Surgically, the liver is often divided into smaller segments, defined by the Couinaud system. The Couinaud system uses the existing functional lobar anatomy and further divides these segments superiorly/inferiorly by the plane of the portal vein. Couinaud segment I represents the caudate lobe; segments II and III represent the superior and inferior portions of the left lateral segment, respectively; segments IVa and IVb represent the superior and inferior portions of the left lobe medial segment (quadrate lobe), respectively. In the right lobe, the Couinaud segments begin in the inferior aspect of the anterior right lobe and run clockwise

(when the liver is viewed from the anterior aspect). Thus, segment V is the inferior right anterior segment, segment VI the inferior right posterior segment, segment VII the superior posterior right segment, and segment VIII the superior right anterior segment (see Figure 3-1). The porta hepatis region is the root of the liver, where the proper hepatic artery, the main portal vein, and the common hepatic duct lie. The primary lymphatic drainage  of the liver also lies in this region, although secondary lymphatic drainage may exist about the hepatic capsule areas, particularly across the diaphragm to the cardiophrenic lymph nodes.

IMAGING OF THE LIVER In current radiology practice, 3 imaging techniques: ultrasonography (US), computed tomography (CT), and magnetic resonance imaging (MRI), have become the mainstay of liver imaging. The general principles of these imaging techniques are discussed in Chapter 1. However, for liver evaluation, specific technical details for optimization of these techniques are discussed below. Also, for certain disorders, these cross-sectional imaging methods may be supplemented by other imaging modalities, including angiography, cholangiography, and nuclear medicine techniques. In the field of cancer imaging (within the liver and elsewhere in the body), fluorodeoxyglucose positron emission tomography (FDG-PET) has become ubiquitous and will often be used as the primary means of cancer staging, including depiction of hepatic metastatic disease. Imaging of the biliary system is often an important component of liver assessment, either directly (through endoscopic retrograde cholangiopancreatography [ERCP] or transhepatic cholangiography [THC]), or indirectly (via magnetic resonance cholangiopancreatography [MRCP]). The appearance of biliary abnormalities with these techniques is discussed in detail in Chapter 4.

US Evaluation of the Liver US imaging of the liver is often performed in conjunction with dedicated imaging of the gallbladder and biliary system (discussed in Chapter 4), with the patient in the fasting state to ensure adequate gallbladder distention with bile. Parenchymal evaluation with US is a routine part of any hepatobiliary US examination, and can include Doppler interrogation of the hepatic vasculature when required to exclude vascular abnormalities. Though not used routinely in clinical examinations, microbubble intravenous (IV) contrast agents can greatly enhance the US examination of the liver,1 and has been used to more efficaciously identify and characterize focal liver lesions.2-4 Normal hepatic parenchyma is uniformly hypoechoic in US imaging bounded by the more hyperechoic ligaments, capsule, and diaphragmatic surface (see Figure 3-2). The vascular structures in the liver are generally anechoic on non-Doppler imaging. Intrahepatic biliary ducts are

Chapter 3 Imaging of the Liver 163

7

2 8

4a 1 3 4b

6

5

A

B

C

D

Figure 3-1 Liver Segments and the Couinaud Division of the Liver In the Couinaud system, the liver is divided vertically by radiating planes created by the hepatic veins and horizontally by the right and left portal veins, with the caudate lobe acting as a separate unit. These divisions are illustrated in the 3-dimensional figure in (A) and in the axial images (B-D) and are numbered in black lettering. The conventional division of the liver is performed only by the vertical division along the hepatic veins and are annotated with white lettering (A, anterior; P, posterior; M, medial; L, lateral). Thus, the liver is divided into right and left lobes by the middle hepatic vein. The right lobe is divided into anterior and posterior segments by the right hepatic vein, and the left hepatic lobe is divided into medial

and lateral segments by the plane from the left hepatic vein to the falciform ligament. Thus, Couinaud regions 2 and 3 make up the lateral segment, Counaud regions 4a and 4b make up the medial segment, Counaud regions 5 and 8 make up the anterior segment, and Couinaud regions 6 and 7 make up the posterior segment. The small portion of the liver immediately adjacent to the inferior vena cava (C) is called the caudate lobe or segment 1 and is separated from the remainder of the liver because it drains directly into the inferior vena cava rather than through the hepatic veins like most of the remainder of the liver. (Part A: Reproduced, with permission, from Smithuis R. Liver: Segmental Anatomy. 7-5-2006 http://www.radiologyassistant.nl/ en/4375bb8dc241d.)

usually not well-seen on US imaging, except when dilated, when they are noted as anechoic tubular structures. The bile ducts have relatively hyperechoic walls, whereas the portal veins have imperceptible walls on US examinations, a feature that can help identify biliary ductal dilation.

contraindicated because of patient allergy or impaired renal function. In certain instances, for example, suspected hepatocellular carcinoma (HCC) or hypervascular liver metastases, the examination can include a dedicated “arterial-phase” image, generally acquired 20 to 30 seconds after rapid IV bolus contrast administration, in order to improve conspicuity of hypervascular liver lesions.5,6 Precontrast imaging may also be required, for example, to exclude fatty infiltration of the liver7,8 or to aid in determining if focally treated liver lesions (such as following

CT Evaluation of the Liver CT imaging of the liver is almost always performed with iodinated contrast agents administered IV, except when

164

Diagnostic Abdominal Imaging

A

B

C

D

E

F

Figure 3-2 Normal US of Liver A-C. Transaxial US images of the liver from superior to inferior locations and (D-F) sagittal US images of the liver from right to

left. D, diaphragm; HV, hepatic veins; IVC, inferior vena cava; PV, portal veins; RK, right kidney; Ao, aorta; Liv, liver; Duo, duodenum; MPV, main portal vein; Panc, pancreas.

chemoembolization with radiopaque embolic agents) contain areas of enhancing viable tissue.9,10 When cholangiocarcinoma is suspected or known to be present, imaging in the delayed phase of contrast (roughly 15 minutes after administration) may also be useful to improve conspicuity of these tumors.11,12 On CT imaging, the normal liver is relatively hyperintense, even in the absence of contrast, with more hypointense vascular structures. On contrast-enhanced imaging, the parenchyma enhances uniformly, with more hyperintense vessels, depending on the phase of dynamic contrast administration (see Figure 3-3).

and/or background hepatic tissue.13-15 Specialized, heavily T2-weighted (fluid-weighted) imaging to depict ductal structures, also known as MRCP is also a part of the liver MRI study, and is discussed in more detail in Chapter 4. Gadolinium-enhanced T1-weighted imaging is routinely used in liver evaluation, except when contraindicated in the rare patient with gadolinium allergy, or in patients who are susceptible to nephrogenic sclerosing fibrosis (NSF) because of impaired renal function.16 As MRI does not produce ionizing radiation, repeated imaging does not pose additional risk to the patient, and multiphase T1 imaging before and during IV bolus administration of gadolinium is the norm, including arterial-, portal-, and delayed-phase imaging. When imaging is performed with newer gadolinium contrast agents that undergo hepatobiliary excretion, the contrast-enhanced examination can be supplemented with delayed biliaryphase imaging. This phase of imaging has been shown to be useful for improving conspicuity and characterization of focal liver lesions.17,18 Iron oxide–based contrast agents are also available for liver MRI. Although these agents will also produce T1 shortening (brightening on imaging), they are often not approved for high-rate bolus injection, thus limiting their ability to act effectively as gadolinium T1 shortening agents for liver MRI. However, the iron oxide particles are selectively taken

MRI of the Liver MRI of the liver will always include imaging with both T1- and T2-weighted contrast, in order to provide means of characterizing focal and diffuse liver abnormalities. T1-weighted imaging is in particular useful because of the relatively high T1 signal intensity of normal liver, and comparable high T1 signal intensity of focal lesions derived from hepatocellular lineage (ie, focal nodular hyperplasia [FNH], hepatic adenoma, and HCC). T1-weighted imaging will also in general utilize chemical shift–sensitive techniques (eg, “Dixon,” or “in-phase/opposed-phase” imaging) in order to demonstrate features of fatty infiltration in focal lesions

Chapter 3 Imaging of the Liver 165

A

B

C

D

E

F

G

H

I

Figure 3-3 Normal CT of the Liver A-C. Unenhanced, (D-F) arterial phase, and (G-I) portal venous phase CT images of the liver. Note how the arteries are much brighter and the veins remain unopacified in the arterial phase.

During the venous phase, both the arteries and the veins are brighter than the normal liver but less so than the arteries on the arterial phase. HA, hepatic artery; Ao, aorta; Ce, celiac axis; IVC, inferior vena cava; HV, hepatic veins; PV, (main) portal vein.

166

Diagnostic Abdominal Imaging

up by the reticuloendothelial system, including the Kupffer cells in the liver. This leads to modest T2 and marked T2* shortening of the liver parenchyma, whereas liver lesions are spared this shortening. As such, most focal liver lesions will appear as brighter objects on T2*-weighted sequences performed after administration of these iron oxide contrast agents.19,20 However, distribution times of these agents into the liver reticuloendothelial system may vary, and timing of optimal T2* liver imaging with these agents may vary.21-23 On MRI, the liver is uniformly hyperintense on T1-weighted imaging, and is hypointense on T2-weighted imaging. Intrahepatic vessels generally appear hypointense on T1 imaging, and hyperintense on T2 imaging, although rapidly flowing blood, such as in the IVC or portal vein, may appear hypointense on spin-echo imaging because of the phenomenon of “flow voids” (see Figure 3-4). Like all fluid structures, biliary ducts are low signal on T1 imaging, and very high signal on T2 imaging (the latter being the basis of MRCP, discussed in detail in Chapter 4). Post gadolinium, the T1-weighted appearance of the liver is similar to that of CT, with a uniformly enhancing liver parenchyma, and more avidly enhanced vascular structures (see Figure 3-4). More recently, diffusion-weighted imaging (DWI) by MRI has been studied as a method for the detection and classification of liver lesions. In DWI, echo-planar T2-weighted images are acquired with and without the addition of oppositely paired diffusion gradients, applied briefly prior to signal acquisition. The diffusion gradients serve to diphase MR signal intensity from water in various tissues. In cases where the mobility of tissue water is highly restricted, the effects of the opposed diffusion gradients will cancel out each other, leaving signal intensity intact. In cases where water mobility is higher, random water motion will occur between the paired gradients, leading to incomplete compensation of the phase dispersion, and resulting signal loss. Application of DWI in the liver has demonstrated that malignant tumors display a higher degree of restricted diffusion than background liver parenchyma.24 This results in tumors displaying relatively brighter signal on images with stronger diffusion gradients. Resulting DWI and apparent diffusion coefficient (ADC) maps can be used to document areas of restricted diffusion, leading in some cases to improved detection and staging of malignant liver lesions.25,26

Multiphase Vascular Imaging of the Liver with CT and MRI With modern CT and MRI techniques, the liver is often imaged with a dynamic or multiphase contrast protocol. In this regimen, the dynamic appearance of the liver enhancement can be shown, reflecting the unique nature of hepatic blood supply. Normally, the liver will obtain 20% to 25% of its blood supply from the hepatic artery, whereas the remaining 75% to 80% of the supply derives from the portal vein. Although this ratio can be altered in certain pathologic conditions, such as cirrhosis and portal hypertension,

the basic premise of dual vascular supply, with the majority supply from the portal circulation, remains true in almost all individuals. During bolus IV contrast enhancement, the hepatic artery receives the initial pass of high-concentration contrast from the aorta sooner than does the portal vein, because the latter receives contrast only after it has passed through the splenic and mesenteric circulation. Relative to the rather rapid transit of contrast from the aorta to the hepatic artery and then the liver, the circulation time from aorta to portal vein in the normal patient is generally 30 to 40 seconds longer. As such, the use of multiphase serial breathhold imaging during enhanced CT or MRI allows for evaluation of hepatic vascular physiology. In multiphase imaging, the liver parenchyma will enhance modestly in the early (arterial) phase of imaging but will demonstrate peak enhancement during the portal phase, when filling of the portal sinusoids with contrast yields maximal signal enhancement (see Figure 3-4). More-delayed postcontrast imaging can also be employed, usually to define pathologies with delayed hyperenhancement. After peak portal phase enhancement, the liver and associated vascular structures will gradually lose contrast, with the blood vessels maintaining higher contrast levels and hence brighter appearance on CT or MRI. After several minutes, the degree of signal difference between liver and blood vessels will be minimal, a phenomenon that will also be seen on MRI, though the liver–blood vessel contrast generally persists for a longer time in MRI. Very delayed biliary-phase MRI may be performed with specialized gadolinium contrast agents with partial biliary excretion, as discussed later in this chapter.

UNIFOCAL, SPHERICAL LIVER LESIONS Unifocal liver lesions can be subdivided into those that are roughly spherical in shape, typically called masses when large and nodules when small. Spherical lesions can be further subdivided into those that are cystic or cystic appearing and those that are solid appearing.

Solitary Cysts and Cystic-Appearing Lesions Cystic or cystic-appearing nodules or masses of the liver are derived from 3 classes of lesions: idiopathic cysts, cystic or cystic-appearing neoplasms, and abscesses. The most common of these lesions are simple cysts and hemangiomas, but a variety of other neoplastic and infections lesions are represented by this category (Table 3-1).

Idiopathic cysts Simple hepatic cysts are among the most commonly discovered abnormalities of the abdomen on cross-sectional imaging examinations. These are idiopathic abnormalities of the liver with no clinical significance, other than that they can be confused with metastasis and other hepatic neoplasms. They are more commonly multiple rather than

Chapter 3 Imaging of the Liver 167

A

B

C

D

E

F

Figure 3-4 MRI of the Normal Liver MRI of the liver seen in (A) T1-weighted in-phase, (B) T1weighted opposed-phase, and (C) fat-saturated T2-weighted images. Hepatic parenchyma is bright on T1-weighted image and dark on T2-weighted images. Hepatic ducts are bright on T2-weighted images (C). Hepatic vessels are dark on T1-weighted images, whereas appearance on T2-weighted imaging is variable based on velocity and direction of blood flow. Dynamic enhanced imaging of the liver includes a (D) pregadolinium T1-weighted

fat-saturated image, demonstrating intermediately bright hepatic parenchyma and dark vessels. Only bowel contents and pancreas (not shown) are brighter than liver on fat-saturated T1-weighted imaging. (E) During the late arterial phase of gadoliniumenhanced imaging, there is mild enhancement of the hepatic parenchyma, with enhancement of both hepatic arteries and variably early filling of the portal vein. (F) On portal-phase imaging, the portal sinusoids have filled, leading to peak liver parenchymal enhancement.

solitary27 and are therefore discussed in detail in the subsequent heading: Multifocal Lesions of the Liver.

reasons. Although hemangiomas are solid vascular tumors, because they are largely composed of blood-filled channels, they can have imaging features of a cystic lesion on CT and MRI examinations. They are typically asymptomatic, and although rarely, when very large they can be painful, necessitating surgical resection.29 In infants, large hemangiomas can result in significant shunting of blood and cause high-output congestive heart failure.30 Their major clinical significance is the confusion with malignant hepatic masses on imaging. On US, hepatic hemangiomas appear densely echogenic, because of the multitude of vascular channels, which act as acoustic interfaces reflecting the sound beam (see Figure 3-5).31,32 Although the vast majority of echogenic

Solitary cystic or cystic appearing neoplasms There are a variety of benign of malignant liver masses that can appear cystic on imaging examinations. The most common of these is the hepatic hemangioma, but also included are biliary hamartomas, mucinous cystadenoma/cystadenocarcinoma, epithelioid hemangioendotheliomas (EHEs), and liver metastasis.

Hemangiomas: Hemangiomas are the most common benign tumor of the liver28 and are usually incidentally discovered on cross-sectional imaging performed for unrelated

168

Diagnostic Abdominal Imaging

Table 3-1. Solitary Cystic-Appearing Nodules or Masses of the Liver 1. Idiopathic cyst a. Simplea b. Complex 2. Neoplasms a. Hemangioma b. Biliary hamartoma c. Biliary cystadenoma/cystadenocarcinoma d. Cystic and mucinous metastasis e. Epithelioid hemangioendothelioma 3. Abscesses a. Pyogenic b. Amoebic c. Echinococcal 4. Hematoma aLesions in bold are most common.

masses in the liver will represent hemangiomas, some malignancies can also appear hyperechoic.33 Therefore, depending on the clinical scenario, hyperechoic masses on hepatic US may require additional imaging evaluation with CT or MRI to ensure benignity. With the advent of dynamic helical CT scanning, multiphase contrast-enhanced CT has been shown to demonstrate with high accuracy the characteristic imaging pattern of hemangiomas. On unenhanced CT, hemangiomas will typically appear as uniform, low-attenuation nodules or masses that resemble hepatic cysts. On early (arterial-phase) imaging, hemangiomas will demonstrate incomplete globular peripheral enhancement, termed discontinuous peripheral enhancement. On more delayedphase contrast CT imaging, increasing portions of the

Imaging Notes 3-1. Imaging Features of Hepatic Hemangiomas US

Hyperechoic nodule

Noncontrast CT

Hypoattenuating nodule

Noncontrast MRI

Uniformly hyperintense nodule on T2-weighted sequences

Contrast CT/MRI

Discontinuous globular peripheral enhancement Centripetal enhancement over time Uniform dense enhancement (parallels aortic enhancement)

hemangioma will enhance from the periphery of the lesion to the center of the lesion, a phenomenon termed centripetal enhancement (see Figure 3-5).34-38 On later phase images, the hemangioma will, typically, enhance uniformly and intensely. More recently, MRI has become the modality of choice in many institutions for characterizing hepatic lesions as a hemangioma, avoiding the radiation exposure of multiphase CT studies. Like cysts, hemangiomas are classically markedly hyperintense to liver on T2-weighted imaging.39,40 Furthermore, when imaged with greater degrees of T2 weighting (ie, longer echo times), hemangiomas and cysts will become more hyperintense relative to background liver, whereas metastases or other solid hepatic lesions will become less intense (see Figure 3-5). MRI can also demonstrate dense centripetal enhancement following gadolinium administration similar to that seen with CT scans.41,42 On occasion, the imaging appearance of hemangiomas on enhanced CT or MRI will differ from the typical features seen in most cases.43 Very small hemangiomas will often demonstrate diffuse, uniform, early enhancement termed flash-filling, resembling a hypervascular solid mass.44,45 Unlike hypervascular masses, flash-filling hemangiomas will usually remain hyperintense on more delayed imaging, paralleling the degree of aortic enhancement. When a hepatic lesion is shown to be uniformly hyperintense on early-phase contrast–enhanced imaging, portal- or delayedphase imaging can be used to distinguish between a flashfilling hemangioma and other hypervascular solid masses. Hemangiomas will remain hyperintense on more delayed postcontrast imaging, whereas other hypervascular neoplasms will typically become more isointense with background parenchyma on portal- and delayed-phase imaging (see Figure 3-6). Larger hemangiomas can also demonstrate an atypical imaging appearance, especially those that have undergone partial internal sclerosis. These lesions may enhance slowly, and portions may never enhance. Usually, careful inspection of the lesion on dynamic enhanced imaging will demonstrate the characteristic early nodular discontinuous peripheral enhancement.46 When the character of a large hepatic lesion is uncertain by US or CT, MRI will often be definitive in diagnosing a hemangioma, because of both the high sensitivity of MRI to gadolinium enhancement and the added value of the T2-weighted imaging appearance. In rare circumstances, giant or sclerosed hemangiomas may require additional imaging with technetium-labeled red blood cells (RBCs), or even biopsy, to confirm the nature of these lesions.47,48 Traditionally, hepatic hemangiomas could be diagnosed through a nuclear medicine scan with technetium99m–labeled RBCs (tagged RBC) study.49 In this study, hemangiomas appeared as focally high activity nodules or masses within the liver parenchyma. However, tagged RBCs are insensitive for smaller hemangiomata.50,51 With the improvements in liver US, the advent of dynamic

Chapter 3 Imaging of the Liver 169

A

B

C

D

Figure 3-5 Typical Hepatic Hemangioma in 2 Patients A. This 36-year-old woman had right upper quadrant pain. Axial US image through the liver demonstrates an echogenic mass in the right lobe of the liver. This finding will usually represent a hepatic hemangioma, which was confirmed by T2-weighted MRI (not shown). B-D. This 58-year-old man had rectal carcinoma. Contrast-enhanced CT scan (B and C) through

the liver shows a low-attenuation mass with small puddles of contrast in the peripheral aspects of the mass (arrows). Although this man is at risk for hepatic metastasis, the appearance of this mass is diagnostic of a hepatic hemangioma. T2-weighted MRI examination of the liver (D) shows a lobulated, very intense (bright white) mass also diagnostic of a hepatic hemangioma.

helical CT, and the evolution of liver MRI, the use of labeled red cell scans to confirm the identity of suspected hemangiomas is now uncommon.

Biliary cystadenoma and cystadenocarcinoma: Biliary

Biliary hamartoma: Biliary hamartomas are small benign tumors of the bile ducts with interspersed fibrous or hyaline stroma. These are relatively common hepatic lesions, occurring in slightly less than 1% of autopsies and characteristically appear as small simple cysts.52-57 Occasionally, they can occur as a solitary lesion. Biliary hamartomas are discussed most completely under the heading Multiple Cystic or Cystic-Appearing Liver Masses.

cystadenoma and cystadenocarcinoma are rare mucinproducing cystic neoplasms of the biliary system that resemble mucinous cystic neoplasms of the pancreas and ovary. Biliary cystadenomas are typically seen in women, particularly in the fifth to sixth decade of life,58 whereas malignant biliary cystadenocarcinomas can be seen in both men and women.59 Typically, cystadenomas and cystadenocarcinomas will histologically contain both hepatobiliary epithelium and ovarian stromal tissues.60 Cross-sectional imaging typically demonstrates a multilocular cystic mass.61 The fluid contents of the cyst will have the usual imaging features of fluid on US, CT,

170

Diagnostic Abdominal Imaging

Imaging Notes 3-2. Imaging Features of Biliary Cystadenoma/Cystadenocarcinoma • Cystic mass with internal septations • Cyst contents can be simple or complex fluid • Connection to intrahepatic bile ducts can be seen in some cases • The thicker the septations and the larger the solid components, the greater the likelihood of malignancy

and MRI. However, those cases with mucinous secretions can demonstrate more hyperintense T1 signal intensity on MRI examinations.62 Connection to the intrahepatic ducts can be demonstrated in some cases,63 particularly on T2-weighted MRI and MRCP sequences.64 The walls of the biliary cystadenomas will usually be very thin, although the tumor will often demonstrate fine internal septations that can be detected on cross-sectional imaging, particularly on T2-weighted MRI and US. Contrast-enhanced imaging by CT and MRI can also demonstrate fine internal septations in some cases (see Figure 3-7). In many cases, the distinction between benign biliary cystadenomas and malignant cystadenocarcinomas is impossible. In general, the greater the proportion of solid elements and the thicker the walls, the  greater the likelihood that the neoplasm is malignant. The presence of enhancing papillary nodules also increases the likelihood of malignancy.61,63,65

A

B

Figure 3-6 Flash-filling Hemangioma This 55-year-old man had malignant mesothelioma. A. T2weighted MRI sequence shows a small bright nodule typical of either a small hemangioma or simple cyst. (B) Pre-contrast, (C) hepatic arterial phase, and (D) portal venous phase T1-weighted postgadolinium enhanced images show immediate bright

Epithelioid hemangioendothelioma: Most commonly found in women in their mid 40’s, EHEs are rare malignant tumors that are typically found in the periphery of the liver.66 They are more common than the more highly aggressive angiosarcomas, and EHEs represent 1% of all vascular hepatic neoplasms.67 Epithelioid hemangioendotheliomas have an abundance of myxoid and fibrous stroma resulting in retraction of Glisson capsule (see Figure 3-8).68 Epithelioid hemangioendotheliomas also have a propensity to grow into tributaries of the hepatic and portal veins, which causes occlusion of the blood supply of the neoplasm and results in self-induced ischemia. Imaging characteristics include solitary or multifocal variants.69 Large lesions can present as dominant masses with small satellite regions of increased vascularity. Although the vascular characteristics of EHE evoke the appearance of hemangiomas on contrast-enhanced CT and MRI, several distinct features of EHE are notable. On enhanced CT or MRI, EHEs usually demonstrate coalescing nodules of heterogenous vascularity, contrasting with the more uniform appearance of the enhancing portions of hemangiomas. T2 signal characteristics on MRI are often also distinctive, with heterogeneous areas of milder hyperintensity in EHEs rather than the more common uniformly hyperintense appearance of hemangiomas (see Figure 3-8). In addition, although focal hemangioendotheliomas can be seen, more commonly the masses show more infiltrative indistinct margins than hemangiomas.70 The “halo” or target sign, demonstrating areas of more intermediate enhancement surrounding less-enhancing regions (an appearance not seen in hemangiomas), is often described.71 A target appearance has also been described because of

C

D

enhancement of the nodule following gadolinium enhancement. This pattern of enhancement can be seen with focal nodular hyperplasia, hepatic adenomas, hypervascular metastases, and flash-filling hemangiomas. However, the T2-weighted characteristics and persistent hyperenhancement on portal-phase imaging is typical for hemangiomas.

Chapter 3 Imaging of the Liver 171

A

B

C

Figure 3-7 Biliary Cystadenoma A. T2-weighted, (B) T1-weighted, and (C) T1-weighted postgadolinium MR evaluation in this 84-year-old woman shows a complex multicystic mass with enhancing mural nodules and

thickened septa in (C) The T2-weighted sequence also shows shading of the fluid in multiple locules, indicating the presence of a gradient of dependent debris. This is believed to represent a biliary cystadenoma, but the lesion was never biopsied.

in-growth and thrombosis of vascular channels.72 When diagnosis is uncertain, biopsy may be required, noting that core biopsy may provide a more reliable diagnosis than fine-needle aspiration.73,74

misinterpreted as benign idiopathic cysts. Malignant cystic pancreatic neoplasms, particularly mucinous tumors, will often demonstrate hematogenous cystic metastasis to the liver. On imaging, mucinous metastases will appear as solitary or multiple cystic lesions that casually resemble idiopathic cyst or hemangioma imaging. However, US will typically demonstrate a complex imaging appearance with septations and nodular solid areas, the latter easily confirmed by Doppler imaging. Careful evaluation of the enhancement pattern of these lesions on CT or MRI examinations will reveal peripheral enhancement without the globular discontinuous character that is typical of hemangiomas. Diffusion-weighted MRI can also be useful for distinguishing cystic metastases from other cystic lesion mimickers, revealing areas of restricted diffusion not seen in cysts and hemangiomas.77

Cystic and mucinous metastases: Metastatic disease to the liver commonly presents as solitary or multiple solid masses, and is described later in this chapter. However, cystic or mucinous metastases deserve special mention, as they can be casually misidentified as a benign lesion because of their partially cystic appearance. Mucinous tumors are complex multiseptated masses filled with mucinous, gelatinous material. Solid tissue, when visible, will usually present as small papillary nodules arising from the periphery of the mass or along thickened septations. These tumors can arise from a variety of organs including the ovary, gastrointestinal (GI) tract, and pancreas and can be confused with benign cystic lesions.75 Malignant mucinous tumors most commonly arise from the ovary. However, these tumors generally spread by peritoneal implantation and when involving the liver will appear as a surface mass along the capsule of the liver. Hematogenous metastasis to any organ, including the liver, is unusual with mucinous adenocarcinoma of the ovary but occasionally occur. Appendiceal carcinomas are also often mucinous in nature. However, these tumors also demonstrate a propensity for peritoneal, rather than hematogenous, metastatic spread.76 Malignant mucinous GI primary tumors, particularly stomach and colon, though less common than ovarian and appendiceal mucinous malignancies, are more likely to demonstrate hematogenous spread to the liver. When this occurs, these lesions can appear as multiple cystic lesions that can be

Abscesses Abscesses to the liver include pyogenic, amebic, echinococcal, and fungal causes. Fungal abscesses characteristically cause multiple tiny abscesses and will be discussed in the subsequent section titled Multifocal Lesions of the Liver. However, pyogenic, amebic, and echinococcal abscesses can all present as a solitary cystic mass, and will be discussed here.

Pyogenic abscesses: Bacterial (pyogenic) hepatic abscesses represent the most common infectious complication of the liver in the developed world. Although the source of many hepatic abscesses are unknown,64 bacteria can spread hematogenously via the hepatic artery to the liver from a site of

172

Diagnostic Abdominal Imaging

A

B

C

D

E

F

Figure 3-8 Epithelioid Hemangioendothelioma A. Precontrast and (B) postcontrast axial CT images in this 53-year-old woman demonstrate a small hypoenhancing mass in the posterior segment of the liver. C. T2-weighted MRI sequence shows slight high attenuation of the mass. D. Precontrast, in-phase; (E) precontrast, out-of-phase; and

(F) postcontrast T1-weighted MRI sequences show a small low signal, nonenhancing nodule. Note the subtle retraction of the hepatic capsule, seen best in B. Biopsy of the mass revealed an epithelioid hemangioendothelioma, an aggressive malignant vascular tumor. Multiple lung metastases were found on chest CT (not shown). 

infection elsewhere in the body.78,79 Hematogenous spread can also occur via the portal vein from sites of enteric infections such as appendicitis, diverticulitis, or bowel perforation.80-84 Spread from an infected biliary tree is a third source of liver infection.82,85 Inoculation of the liver via direct penetrating trauma or during accidental contamination during surgery is a less common source of hepatic infection.

Approximately half of pyogenic hepatic abscesses are polymicrobial.86 Escherichia coli is the most common organism, though many other bacteria, both aerobic and anaerobic, can cause hepatic abscesses.87 Abscesses that arise from seeding via the arterial or portal system will generally form in the periphery of the liver in a subcapsular location. Rarely, diffuse military involvement can also be seen.88 Abscesses

Chapter 3 Imaging of the Liver 173 Imaging Notes 3-3. Sources of Hepatic Abscesses 1. Hematogenous dissemination via the hepatic artery from a systemic source a. Endocarditis b. Infected catheter or in vascular hardware c. Other deep-tissue infection 2. Hematogenous dissemination via the portal vein from a gut infection a. Diverticulitis b. Appendicitis c. Abdominal abscess 3. Spread from biliary infection 4. Direct inoculation a. Hepatic surgery b. Penetrating trauma 5. Idiopathic

images.98 Often, mild perilesional T2 bright edema is noted around the lesion. The T2 signal intensity of the central fluid is often more moderate and heterogeneously bright than that seen with simple cysts or hemangiomas. Following the administration of IV contrast, the wall of the abscess will usually enhance. In general, MRI is more sensitive than CT for the detection of enhancement.99-101 Diffusion-weighted imaging can be useful in discriminating abscesses from cysts. Abscesses will typically demonstrate increased signal intensity on DWIs but will generally demonstrate restricted diffusion throughout, whereas cysts and cystic/necrotic regions of tumors will more frequently demonstrate elevated ADC values.77 It is important to recognize that on follow-up imaging, US imaging can demonstrate residual lesions even after successful completion antibiotic therapy.102 On CT and MRI, treated abscesses tend to decrease in size and lose their hyperenhancing rims but may persist as hypoattenuating areas on CT or hypointense areas on MRI, representing areas of granulation tissue. Partially treated, organizing abscesses can appear solid and mimic enhancing tumors on CT or MRI.103

Amebic abscess: Amebic abscesses are a result of mesenthat arise from biliary seeding, for example, as a complication of ascending cholangitis, are typically more centrally located, and when multifocal, will be clustered to a single segment of the liver.89,90 The segmental distribution of liver abscesses on imaging can suggest a ductal source of infection. Patients with hepatic abscesses typically present with fever, elevated white blood cell (WBC) count, and right upper quadrant pain. Pleuritic and/or right shoulder pain can also be noted in cases of subdiaphragmatic spread. Symptoms are often insidious in onset, and can develop over several weeks. Laboratory evaluation will reveal elevation in liver enzymes, particularly alkaline phosphatase and bilirubin, with more variable transaminitis.91 On US examinations, pyogenic hepatic abscesses typically appear as one or several heterogeneously hypoechoic masses with ill-defined margins.92 In one study, US examinations had an 85% sensitivity for the detection of hepatic abscesses, with most false-negative examinations due to abscesses located in the liver dome.93 On CT scans, pyogenic abscesses will appear as one or several hypoattenuating masses within the liver parenchyma. However, careful observation will show the center to have an attenuation greater than that of simple fluid, with measurements greater than 20 Hounsfield units (HU) (see Figure 3-9).94 On contrast-enhanced CT, faint illdefined enhancing rims are often present.64 Gas bubbles or air-fluid levels are an uncommon finding on CT examinations but when present can be an important imaging clue to the diagnosis of a hepatic abscess.95 Calcifications are rare.96 Larger abscesses can cause portal vein thrombosis in some patients.97 On MRI, hepatic abscesses are low-intermediate signal intensity on T1-weighted and bright on T2-weighted

teric and portal venous dissemination from colonic infection with Entamoeba histolytica. Approximately 10% of the world’s population is asymptomatically colonized with this organism that is acquired by drinking contaminated water.104,105 In one study, nearly three-quarters of patients with liver abscess were found to have asymptomatic intestinal colonization.106 The liver is, by far, the most common nonintestinal site of disease, and extrahepatic spread elsewhere in the abdomen or chest is often the result of rupture of a hepatic abscess.107 Patients with hepatic amebic abscess typically present with right upper quadrant pain, fevers, and rigors. Symptoms are often acute, although more indolent presentations can be seen manifesting as weakness and weight loss.108 Because intestinal colonization, which precedes the development of hepatic abscesses, can be asymptomatic, gastrointestinal symptoms are more variable. Laboratory evaluation is nonspecific, and eosinophilia is rare. Stool cultures and serologic testing are now used routinely to diagnose amebic infection. Aspiration may be for diagnostic confirmation or for therapeutic drainage when risk of rupture is high.109 On US examinations, amebic abscess are usually seen as well-defined cystic lesions, although solid appearances early on in infection can be seen.110 As with pyogenic abscesses, sonographic resolution can lag behind clinical response to therapy.111,112 On CT examinations, amebic abscesses are often round or oval, and are mostly seen in the right lobe. On contrast-enhanced imaging, a thick enhancing wall is often seen.113 On MR, lesions are uniformly dark on T1-WI with bright T2 cystic appearance.114 There is often peri-lesion edema of the adjacent liver.115 Contrast-enhanced appearance is similar to that of CT, with a rind of enhancement about the abscess wall.

174

Diagnostic Abdominal Imaging

A

B

C

D

Figure 3-9 Pyogenic Liver Abscess in 3 Patients A. This 56-year-old man complained of fever, abdominal pain, nausea, and lethargy. Contrast-enhanced CT shows a multiloculated cyst with a thick enhancing wall (arrowheads), features typical for a liver abscess. B. This 76-year-old man had persistent fevers following cecal resection for a perforated toothpick. Contrast-enhanced CT images through the liver demonstrate a large unilocular low attenuation mass (arrows) in

the left lobe of the liver. Note the fluid debris level (arrowhead) within the mass, indicating a complicated cyst. Cultures of aspirated fluid from the mass demonstrated enterococcal species. C and D. This 19-year-old man with Hodgkin disease had been septic several days earlier and then complained of fever and chest pain. Contrast-enhanced CT shows multiple indistinctly marginated low attenuation nodules in the liver that were proven to represent multiple hematogenous liver abscesses.

Echinococcal cyst: Echinococcal infection is a result of

imaging examinations. Some echinococcal abscesses will appear as a unilocular cyst that can be virtually indistinguishable from idiopathic cysts. However, most echinococcal cyst will demonstrate early rim enhancement indicating the inflammatory nature of the cystic lesion.116 In some cases, the cyst will contain small calcifications, which layer dependently within the cyst and are termed hydatid sand. When present, hydatid sand is an important clue to the diagnosis of echinococcal cysts and differentiates echinococcal cysts from other causes of cystic masses.117 Many echinococcal cysts will appear as a multilocular cystic mass called the “alveolar form”

the accidental ingestion of raw food contaminated with organisms contained within canine feces. Echinococcus granulosus and multilocularis are the most common species responsible for infection in humans, resulting in the cystic and alveolar forms of the disease, respectively. The organism invades the human via the gut, passing through the mesenteric and portal venous systems to infect most commonly the liver, and rarely other organs of the body such as lungs, heart, and brain. Echinococcal abscesses of the liver will most often appear as a round or oval mass on all cross-sectional

Chapter 3 Imaging of the Liver 175

A

B

Figure 3-10 Echinococcal Cyst This 51-year-old immigrant from India complained of shooting abdominal pain. A. Contrast-enhanced axial CT image shows a large cystic mass involving the posterior segment and caudate

lobe of the liver. Note the rim of smaller cysts within the dominant cyst that represent “daughter” cysts and is diagnostic of an echinococcal cyst. B. There is a similar echinococcal cyst in the pelvis, indicating peritoneal spread of the infection.

of the disease.118 The alveolar form of echinococcus has a characteristic appearance as a large cyst with many small cysts lining the internal rim of the dominant cyst. These smaller cysts are called “daughter” cysts and this appearance is pathogmnemonic of echinococcal cysts in the liver, spleen, and other organs where they arise (see Figure 3-10). MRI can demonstrate several additional imaging features of echinococcal cysts.119 These include peripheral liver edema100 and a low signal intensity rim.120,121

Focal nodular hyperplasia: Focal nodular hyperplasia,

Solitary Solid Hepatic Masses or Nodules The majority of solitary solid hepatic masses will represent primary or secondary neoplasms of the liver but can rarely represent other unusual conditions such as peliosis hepatis, inflammatory pseudotumor and hepatic pseudo-lesions such as focal fatty infiltration or focal fatty sparing.

Solitary solid appearing hepatic neoplasms

as indicated by its name, represents hyperplastic proliferation of hepatic parenchyma, hepatocytes, bile canaliculi and reticuloendothelial cells, in a focal nodule. The lesion is thought to be a hyperplastic response to a central arteriovenous malformation. Focal nodular hyperplasia is the second most common benign tumor of the liver, following hemangiomas, and is a common incidental finding on imaging. It is associated with oral contraceptive use,122

Table 3-2. Solitary Solid-Appearing Nodules or Masses of the Noncirrhotic Liver 1. Benign primary neoplasms a. Hemangiomaa b. Focal nodular hyperplasia c. Hepatic adenoma

Solitary, solid hepatic neoplasms include hemangioma, FNH, hepatic adenoma, HCC, cholangiocarcinoma, angiosarcoma and solitary hepatic metastasis (Table 3-2).

2. Malignant primary neoplasms a. Cholangiocarcinoma b. Lymphoma c. Fibrolamellar HCC

Hemangioma: Hemangiomas have been extensively dis-

3. Metastasis

cussed in the previous section entitled: Solitary cystic or cystic-appearing neoplasms. On CT and MRI examinations these tumors will typically have imaging features indicating the presence of fluid within the lesion and will appear cystic. However, some small hemangiomas will fill with contrast so rapidly that they can appear as a uniformly enhancing solid appearing nodule on contrast-enhanced CT. Furthermore, on US examinations, the majority of hemangiomas will appear as a uniformly hyperechoic solid appearing mass.

4. Peliosis hepatis 5. Inflammatory pseudotumor 6. Pseudolymphoma 7. Pseudolesions a. Focal fatty infiltration b. Focal fatty sparing aLesions in bold are most common.

176

Diagnostic Abdominal Imaging

although not as strongly as with hepatic adenomas. Its lesions need no additional imaging follow-up or treatment because they represent benign proliferations of hepatocytes with no potential for malignant transformation or spontaneous hemorrhage. When FNH lesions are followed, the majority demonstrate stability over time, although regression or rarely growth has been shown.123-126 Definitive diagnosis of a FNH lesion is important so as to exclude other more clinically significant liver lesions. On US, FNH lesions typically appear as isoechoic nodules that faintly distort the hepatic echotexture.127,128 Because their echotexture is nearly identical to normal parenchyma, small FNH nodules are often difficult to detect on US. Definitive diagnosis of FNH is typically made using CT and/or MRI examinations. On both unenhanced CT imaging and pregadolinium T1 MRI, FNH lesions will typically appear isointense to the background liver parenchyma (see Figure 3-11).129 The isointense character of FNH on noncontrast T1-weighted imaging is particularly important, because metastatic lesions will typically appear hypointense relative to normal liver parenchyma. On T2-weighted imaging, FNH lesions are also generally isoechoic to liver, although faintly hyperintense pattern are seen. When visible, the central scar of an FNH lesion will be hyperintense on T2-weighted imaging, a distinguishing feature from the MR appearance of hepatic adenomas.130 The appearance of FNH on dynamic contrast-enhanced CT and MRI are relatively invariant.130-132 FNH will characteristically appear as a uniformly hyperenhancing nodule on arterial-phase imaging because of the enhanced arterial supply of the nodule relative to that of normal hepatic parenchyma (see Figure 3-11). On CT arteriography, anomalous feeding artery and draining veins can be shown in some cases.133 On portal venous-phase CT images, the nodule is usually isointense or faintly hyperintense to normal enhanced hepatic parenchyma. In many cases, even moderately large FNH nodules will be undetectable on venous phase imaging (see Figure 3-11). A consequence of this phenomenon is that previously present FNH nodules go undetected by prior venous-phase-only imaging and can be discovered as “new” hypervascular nodules on arterialphase imaging and can be mistaken for hypervascular metastasis. Large FNH lesions will characteristically contain a central fibrous scar, although central scars are uncommon in small FNH lesions.131,134 Central scars often go undetected on unenhanced CT images because they are isoattenuating to liver parenchyma. However, the central scar is usually seen as a stellate central defect within the mass that is hypointense on T1-weighted imaging, and hyperintense on T2-weighted MRI. With IV contrast enhancement, the scar will typically not enhance and appear as hypoattenuating/hypointense stellate lesion within the brightly enhancing mass. With images performed after the portal venous phase of enhancement (delayed phase imaging), the mass will typically appear as a brightly enhancing stellate lesion within the less enhancing FNH nodule (see Figure 3-11).

Imaging Notes 3-4. Imaging Features of Focal Nodular Hyperplasia Ultrasonography

Isoechoic nodule or mass

Noncontrast CT

Isoattenuating to liver parenchyma

Noncontrast MRI

Isointense T1-weighted images, T2-weighted images

Contrast CT/MRI

Hyperenhancing on arterial-phase images Isointense on venous-phase images

In most cases, FNH lesions can be definitively diagnosed according to their imaging characteristics on CT and MRI examinations; however, there are 2 notable exceptions. If only contrast-enhanced CT images are obtained, both small FNH lesions and small “flash filling” hemangiomas can appear as a uniformly hyperenhancing nodule within the liver parenchyma. Fortunately, as both of these lesions are benign and without clinical significance, differentiation between the lesions is unnecessary. FNH lesions can also be confused with the rare fibrolamellar variant of HCC. This variant of HCC can also contain a central scar. Although the central scar is hypointense on T1-weighted images and resembles that seen in FNH, unlike FNH, the central scar of fibrolamellar HCC is not typically hyperintense on T2-weighted imaging. Also, fibrolamellar HCCs will typically demonstrate heterogenous enhancement, rather than demonstrate a more uniform early-enhanced appearance typical of FNH lesions.135 Before the widespread use of cross-sectional imaging, sulfur colloid nuclear medicine examinations were commonly used to definitively diagnose a liver mass as an FNH. Focal nodular hyperplasia has the unique characteristic among hepatic masses of containing reticuloendothelial cell, in common with the normal liver. As sulfur colloid is phagocytized by the reticuloendothelial cells of the liver, spleen, and lymph nodes, these organs demonstrate high levels of activity on sulfur colloid scans. All mass lesions other than FNH will appear as a photopenic region with the liver on sulfur colloid, whereas FNH lesions will typically take up sulfur colloid as well, therefore presenting as an area of equal or more intense tracer uptake on planar or single-photon emission computed tomographic (SPECT) imaging.136 Recently, the use of a newer-generation gadoliniumbased MRI agents with partial hepatobiliary excretion has been shown to be useful for characterizing FNH lesions. Two such agents are currently approved for use in the United States. The first agent, gadobenate dimeglumine (MultiHance, Bracco), demonstrates approximately 2% biliary excretion in humans with normal renal and hepatic function.137 A more recently approved agent, gadoxetate

Chapter 3 Imaging of the Liver 177

A

B

C

D

E

F

G

H

Figure 3-11 Focal Nodular Hyperplasia in 2 Patients A-C. This 34-year-old man had a history of malignant melanoma. Noncontrast (A), arterial-phase (B), and venous-phase (C) CT images through the same location in the liver demonstrate a 2.5-cm mass in the medial segment of the liver. This is hypoattenuating on noncontrast images, hyperenhancing on arterial-phase images, and isoenhancing on venous-phase images. This appearance is common among the hepatocellularderived tumors of the liver but is most typical of focal nodular hyperplasia. Note also the small hypoattenuating, nonenhancing central region within the mass. This is typical of a central scar and is characteristic of focal nodular hyperplasia. D-H. This

37-year-old woman had abdominal pain. T1-weighted (D); T1weighted, fat saturation (E); and T2-weighted (F) MRI images through the same location in the liver demonstrate a large mass that is isointense with normal liver parenchyma. Note that there is no loss of signal in the mass following the fat saturation pulse, making a hepatic adenoma less likely. Note also the hypointense central scar (arrow) in (D) and (E) is hyperintense on the T2weighted images. With intravenous contrast, the mass became hyperintense to liver on the arterial-phase (G) and isointense on venous-phase (H) images. The central scar demonstrates delayed enhancement on the venous-phase images (arrow). These findings are typical of focal nodular hyperplasia.

disodium (Eovist, Bayer) demonstrates approximately 50% biliary excretion.138,139 In both cases, the delayed excretion into the biliary canaliculi provides prolonged and increased T1-weighted image enhancement after the first pass and blood pool phases of contrast accumulation have subsided. This effect has been shown to increase conspicuity of liver lesions without functioning bile canaliculi. Because FNH lesions contain biliary canaliculi, biliary-phase imaging

with either agent demonstrates delayed hyperenhancement of the mass. Most other rapidly enhancing lesions in the liver will demonstrate contrast washout on the biliaryphase images.140,141 This effect is seen earlier and is more pronounced with gadoxetate, given that this agent undergoes a greater relative degree of biliary uptake and excretion than does gadobenate. However, the use of these agents can alter the delayed-phase appearance of the central scar

178 Diagnostic Abdominal Imaging in FNH, rendering them hypointense to the remainder of the lesion on biliary-phase imaging.142

Hepatic adenoma: Hepatic adenoma is a benign tumor of hepatocytes that is most often seen in young and middle-aged women and is typically associated with the use of oral contraceptives.143 The incidence of hepatic adenoma is directly related to the dose of estrogens within the medication and the duration of use.144 Withdrawal of oral contraceptives will frequently lead to tumor regression.145 Use of anabolic androgenic steroids is also associated with an increased incidence of hepatic adenoma.146 Adenomas are also associated with glycogen storage disease types I and III.147 Although often solitary, multiple hepatic adenomas are not uncommonly reported in patients’ oral contraceptive use.148 Multiplicity is also associated with hepatic steatosis.149 Hepatic adenomatosis is a distinct entity resulting in the appearance of multiple adenomas unrelated to steroid use or glycogen storage disease. The appearance of the individual lesions in hepatic adenomatosis is similar to that of “acquired” isolated tumors. However, management is more challenging as there are no options to remove the precipitating agent. Discovery of multiple hepatic adenomatous lesions unassociated with glycogen storage disease and oral contraceptive or anabolic steroid use can suggest the presence of hepatic adenomatosis.150-152 Differentiation from other presentations such as multiple FNH lesion is critical for clinical management. Hepatic adenomas are composed of benign hepatocytes without bile ducts and usually without Kupffer cells. Adenoma cells can contain increased intracellular lipid, a feature that can be used to identify the tumor on chemical shift–sensitive MRI (see Figure 3-12).153 Hepatic adenomas are often detected incidentally. However, spontaneous rupture and hemorrhage of these lesions is not uncommon, especially for larger lesions (see Figure 3-13).154-156 When spontaneous rupture occurs, patients will typically present with acute right upper quadrant pain. In these cases, hemorrhage in some cases will completely or partially obscure the lesion on cross-sectional imaging. One consequence

Imaging Notes 3-5. Hepatic Masses with Intracellular Lipid Intracellular lipid as measured by chemical shift imaging (in-phase and opposed-phase T1-weighted images) within a hepatic mass is highly associated with primary neoplasms of the hepatocyte: 1. Hepatic adenoma 2. Hepatocellular carcinoma 3. Focal nodular hyperplasia

of this is that when a young woman presents with spontaneous intrahepatic hemorrhage, an underlying adenoma should be considered as the potential etiology, even if the lesion itself is not initially apparent. Like other tumors of hepatocellular origin, hepatic adenomas demonstrate an imaging appearance that is similar to normal hepatic parenchyma. Nonhemorrhagic hepatic adenomas will often appear as a isoechoic or faintly hypoechoic spherical lesion in the liver on US (see Figure 3-12).157 On noncontrast CT, the lesion will be isoattenuating or slightly hypoattenuating relative to background liver.158 The MRI appearance of hepatic adenomas can be variable but is most often nearly isointense to liver on both T1- and T2-weighted sequences (see Figure 3-12).54,159-161 Intravenous contrast on both CT and MRI will show moderate to strong enhancement of the nodule relative to liver on arterial-phase images, and often slightly decreased enhancement relative to liver on venous-phase images, although the degree of enhancement can be variable (see Figure 3-12). As noted previously, tumors of hepatic origin, including hepatic adenomas, will often demonstrate the presence of microscopic lipid on chemical shift–sensitive MRI. Thus, the lesion will lose signal intensity on opposedphase images when compared with in-phase images (see Figure 3-12). The presence of spontaneous hemorrhage on either CT or MRI can also be an important clue to the diagnosis of hepatic adenoma (see Figure 3-13). Larger hepatic adenomas will often develop a central stellate scar. Scar tissue will typically demonstrate delayed enhancement on contrast-enhanced CT and MRI images. Focal nodular hyperplasia is the most common hepatic lesion to resemble hepatic adenomas. As noted in the discussion of FNH, both sulfur colloid scanning and the hepatobiliary excreted gadolinium agents (gadobenate and gadoxetate) can be useful to distinguish between hepatic adenoma and FNH.140,158,162-164 Both sulfur colloid and the hepatobiliary excreted gadolinium agents are taken up by FNH but not hepatic adenomas. Therefore, hepatic adenomas will appear as photopenic regions on sulfur colloid scans and will appear hypointense to enhanced liver on these delayed postgadolinium images, whereas FNH will have the same or increased activity as normal liver on sulfur colloid scans and will have increased signal on delayedphase MRI images (see Figure 3-14). Hepatocellular carcinoma: HCC in almost all cases is a complication of cirrhosis. As a consequence, the imaging appearance of HCC is intrinsically connected with the imaging features of cirrhosis and will be discussed with cirrhosis under the heading: Hepatic Disorders with a Specific Appearance later in this chapter. Fibrolamellar HCC: Fibrolamellar HCC is a rare tumor that has several clinical and radiologic features that distinguish it from the much more common variant of HCC associated with cirrhosis. Fibrolamellar HCC tends to occur in younger patients and is not associated with underlying liver disease.165 As such, fibrolamellar HCC can often be treated

Chapter 3 Imaging of the Liver 179

A

B

C

D

E

F

Figure 3-12 Hepatic Adenoma in 2 Patients A. This 40-year-old woman had been using oral contraceptives when she complained of right upper quadrant pain. US of the left lobe of the liver demonstrates a slightly hypoechoic mass (arrow) subsequently shown to represent a hepatic adenoma. B-F. This 47-year-old woman used oral contraceptives and complained of abdominal pain. T1-weighted in-phase (B), T1-weighted out-of-phase (C), T1-weighted postcontrast arterialphase (D), T1-weighted postcontrast venous-phase (E), and

T2-weighted (F) sequences through the same region of the liver. A low signal mass (arrow) is seen in the medial segment of the liver in the out-of-phase image (C) but is isointense to liver on T1- and T2-weighted sequences and is virtually invisible on these sequences. Careful evaluation of the arterial-phase images (D) shows a slight increased enhancement of the mass (arrow) and venous-phase images (E) show slightly decreased enhancement of the mass (arrow) relative to normal liver parenchyma. This was subsequently demonstrated to represent a hepatic adenoma.

more aggressively, and patients with HCC, even when large, often have a better prognosis when compared to that of patients with advanced-stage HCC associated with liver disease.166 Fibrolamellar HCCs tend to be large at presentation, as small lesions are asymptomatic. Calcification is present in many cases, and is easily demonstrated by CT. Lesions are unifocal and lobular, with low signal intensity on unenhanced CT and variable low-isointense appearance on pregadolinium T1-weighted MR images. Lesions are usually

marked heterogeneous on T2-weighted imaging, with relatively high signal intensity. Fibrolamellar HCCs are notable for containing a central scar, a characteristic associated with the more common benign FNH lesion, and occasionally with giant sclerosed hemangiomas. As such, the presence of a lesion with a central scar on cross-sectional imaging can pose a diagnostic dilemma initially. However, the central scar of fibrolamellar HCC will not enhance or will enhance only mildly, in contrast with the central scar of FNH, which is generally

180 Diagnostic Abdominal Imaging

A

B

Figure 3-13 Ruptured Hepatic Adenoma This 27-year-old woman developed fever and abdominal pain while on oral contraceptives. A and B. Contrast-enhanced CT shows a heterogeneous mass (arrows) in the right lobe of the liver. This is associated with a mixed-attenuation subcapsular fluid collection (arrowheads) compressing the surface of the liver.

The mixed attenuation is typical of a subcapsular hematoma with differing phases of blood. Although the mass is nonspecific, the combination of the clinical history of oral contraceptive use and the presence of a subcapsular hematoma is highly suggestive of spontaneous rupture of a hepatic adenoma.

hyperenhancing on delayed-phase imaging. Furthermore, fibrolamellar HCCs typically demonstrate markedly heterogenous enhancement, a feature common to many malignant lesions. As such, fibrolamellar HCCs can often be distinguished from benign entities that also present with a central scar.135

In some cases, cholangiocarcinomas grow as a focal well-defined, intrahepatic mass that can be confused with other causes of hepatic masses.176-178 However, mass-forming intrahepatic cholangiocarcinomas will often obstruct the biliary system, a feature that is more commonly seen in cholangiocarcinomas than other primary or secondary liver masses (see Figure 3-15). If CT or MRI examinations are obtained several minutes after the injection of IV contrast, there can be delayed enhancement of cholangiocarcinomas.179 Combination of dynamic and delayed-phase imaging can therefore improve depiction of cholangiocarcinoma.180 This delayed enhancement is characteristic of the scirrhous nature of these lesions, with abundant fibrous tissue interspaced with malignant tumor cells.11 Cholangiocarcinoma is discussed most completely in Chapter 4: Imaging of the Gallbladder and Biliary System.

Cholangiocarcinoma: Cholangiocarcinoma (CCA) is the second most common hepatic primary malignancy and is typically found in patients in their fifth and sixth decade of life.167 This tumor can arise sporadically or be associated with a variety of chronic inflammatory disorders of the bile ducts, including sclerosing cholangitis168 and parasitic infections of the biliary tree.169,170 As biliary parasites are more commonly found in East Asia, cholangiocarcinoma is more frequently seen in Japan, Korea, and China than in Western countries.171 Patients with Caroli disease (congenital dilation of the intrahepatic bile ducts; see Chapter 4) and patients with biliary stones are also at an increased risk for developing cholangiocarcinoma.172,173 Cholangiocarcinomas can arise from any portion of the bile ducts, both intrahepatic and extrahepatic, but most typically occur at the junction of the left and right bile ducts in the porta hepatis. Tumors at this location are termed Klatskin tumors. Cholangiocarcinomas are divided into types based on their growth pattern: (1) mass forming, (2) periductal infiltrating, and (3) intraductal-growing types.174,175 The periductal and intraductal types of intrahepatic cholangiocarcinoma, as well as extrahepatic tumors, will be discussed in Chapter 4: Imaging of the Gallbladder and Biliary System.

Angiosarcoma: Angiosarcoma is among the rarest primary hepatic neoplasms. It is estimated that only 25 cases will occur annually in the United States.181 Angiosarcoma is a highly malignant tumor with a very poor prognosis. Most patients will die within a year of diagnosis.182 They are of historical interest because they were discovered to be induced by the intravascular contrast agent: thorium dioxide (Thorotrast). Thorium dioxide colloidal solution, or Thorotrast, was introduced as a radiographic contrast agent in 1928 and used until the early 1950s. Thorotrast accumulates in the cells of the reticuloendothelial system, where it is retained throughout life. Thorium, an alpha-particle emitter with a biologic halflife of 200 to 400 years, resulted in a cumulative radiation

Chapter 3 Imaging of the Liver 181

A

B

C

D

Figure 3-14 Hepatic Adenomas Demonstrated by Absence of Hepatobiliary Contrast A 54-year-old woman with multiple known hepatic adenomas. A and B. Arterial-phase MRI demonstrates 2 small hypervascular lesions (arrows). C and D. Hepatobiliary phase image at

20 minutes demonstrates diffuse uptake (brightening in the liver) with washout of vessels. The adenomas are hypointense relative to background liver, reflecting the lack of biliary uptake in the tumors.

exposure leading to an increased incidence of cholangiocarcinoma, HCC, and angiosarcoma of the liver.183 Other risk factors for angiosarcoma include arsenic and polyvinyl chloride exposure, use of anabolic steroids, and a history of hemochromatosis or cirrhosis.184 However, most angiosarcomas develop sporadically without a known risk factor.185

Angiosarcoma will typically present in the seventh decade of life and is seen more commonly in men, with a male-female ratio of 4:1.186 Patients will most often present with right upper-quadrant abdominal pain, abdominal discomfort, anorexia, or weight loss.182 Patients will often have associated thrombocytopenia, disseminated intravascular

182

Diagnostic Abdominal Imaging

A

B

Figure 3-15 Cholangiocarcinoma This 82-year-old man presented with jaundice. A and B. Contrast-enhanced CT shows mild intrahepatic ductal dilation

(arrowheads) and a faintly enhancing mass (arrows) in the porta hepatis. Biopsy was diagnostic of a cholangiocarcinoma.

coagulation (DIC), and/or hemolytic anemia. Most patients will have metastasis discovered at presentation. Common sites of metastasis include the spleen and lung.187 Angiosarcoma is typically a multifocal neoplasm of the liver.188 In approximately one-half of patients, angiosarcoma will appear as a large dominant mass within the liver with or without scattered smaller satellite nodules.189 In most of the remainder, it will appear as multifocal nodular lesions scattered throughout the liver (see Figure 3-16). Rarely, angiosarcoma will appear as diffuse nodular infiltration of the liver. On CT, the masses will usually appear hypoattenuating on precontrast190 images, occasionally with focal regions of higher attenuation.191 On MRI examinations, angiosarcomas will typically appear hypointense on T1-weighted images and hyperintense on T2-weighted images. Small focal regions of T1-weighted hyperintensity and areas with fluid-fluid levels on T2-weighted images are commonly found and suggest the presence of intratumoral hemorrhage.192 Following contrast administration, typically heterogeneous regions of the mass will enhance, and the extent of enhancement will increase with delayed images, although rarely tumors appeared hyperenhancing to liver parenchyma.193

appearance on US.197 On CT, lymphoma is hypodense to liver181 and demonstrates little enhancement following contrast administration (see Figure 4-59). On MRI, lymphoma appears hypointense on T1, with variable signal intensity

Lymphoma: Secondary involvement of the liver by lymphomas is not uncommon, with an 8% incidence of Hodgkin disease and 25% incidence of non-Hodgkin lymphoma.194,195 However, primary hepatic lymphoma is extremely rare. Hepatic lymphoma most often appears as diffuse hepatomegaly but can also appear as one or several focal masses.196 When presenting as a focal mass in the liver, lymphoma will demonstrate a uniform hypoechoic

Figure 3-16 Angiosarcoma This 55-year-old woman presented with dyspnea. Contrastenhanced CT shows a large heterogenous mass filling and expanding the left hepatic lobe. There is also a small amount of perihepatic ascites. An infiltrative mass of this appearance will usually indicate a primary hepatic neoplasm, most often HCC. Liver biopsy was diagnostic of an epithelioid angiosarcoma.

Chapter 3 Imaging of the Liver 183 on T2 imaging, and modest enhancement post gadolinium administration.198,199

Peliosis hepatis Peliosis hepatis is a rare benign vascular lesion characterized by solitary or multifocal randomly distributed regions of sinusoidal dilation and blood-filled hepatic spaces. A variety of agents have been linked with peliosis hepatis, including chemicals such as arsenic and polyvinyl chloride, drugs such as oral contraceptives, anabolic steroids, and corticosteroids;200 and infectious agents such as Bartonella in AIDS.201 Peliosis hepatitis is also seen in patients with renal or cardiac transplants,202 and malignancies such as HCC.203 The pathophysiology of peliosis hepatis is not known. Hypotheses regarding the etiology include the following: (1) it is a result of sinusoidal epithelial damage; (2) it is a result of pressure from obstruction of hepatic venous outflow; or (3) it is a result of hepatic necrosis.204,205 In most cases, patients are asymptomatic and the disorder is discovered as a result of elevated liver function tests or imaging examinations performed for other reasons.206 However, occasionally, when severe, peliosis hepatis can be symptomatic, presenting with hepatomegaly, jaundice, liver failure, and/ or hemoperitoneum. It is important to consider the diagnosis of peliosis hepatis because a correct diagnosis can lead to withdrawal of the offending drug or toxin, leading to resolution of the disease and prevention of serious complications such as hepatic failure or intra-abdominal hemorrhage.206 Cross-sectional imaging examinations will typically demonstrate 1 or multiple focal masses or nodules within the liver parenchyma that can be confused with hepatic neoplasms, especially hemangiomas, focal nodular hyperplasia, and other hypervascular liver masses.206-208 Hemorrhagic findings with multiple cystic dilated spaces can be a clue to the diagnosis.209 On US examinations, peliosis hepatis will typically appear as a uniform hyperechoic mass.206 If the mass is complicated by hemorrhage, it can appear heterogeneously hyperechoic and, in patients with hepatic steatosis, it can appear hypoechoic because of increased echogenicity of the background liver parenchyma. On unenhanced CT examinations, peliosis hepatis will typically appear as uniform hypoattenuating liver nodules or masses.206 Contrast-enhanced CT will characteristically demonstrate one or multiple hepatic masses with early globular enhancement with centripetal increase in enhancement during later phases of imaging. Occasionally brisk, uniform, diffuse early enhancement can be seen.210 Also, areas of central enhancement, termed the target sign, can be identified. On delayed-phase images, peliosis hepatis can appear homogeneously hyperintense to surrounding liver parenchyma.211 On MR examinations, the appearance of peliosis hepatis will depend on the age and status of the hemorrhagic component.206 On T2-weighted sequences, peliotic lesions are typically hyperintense to liver parenchyma. Multiple

Imaging Notes 3-6. Hypervascular Lesions of the Otherwise Normal Liver Uniformly hyperenhancing lesions in the arterial phase of enhancement in an otherwise normal-appearing liver are most often due to either flash-filling hemangiomas or regions of focal nodular hyperplasia. Less commonly, they can represent small hepatic adenomas. Rarely, they can be due to hypervascular metastasis (especially neuroendocrine neoplasms) or vascular disorders such as hemangioendotheliomas, peliosis hepatis, and arteriovenous malformations. In the cirrhotic liver, hypervascular nodules typically represent either regenerative nodules or small hepatocellular carcinomas.

small foci of even higher signal intensity are seen within the lesion, which are thought to be due to hemorrhagic necrosis. On T1-weighted sequences, the lesions are usually hypointense because of the presence of subacute blood. Enhancement patterns are similar to CT, with globular centripetal or centrifugal enhancement.

Inflammatory pseudotumor Inflammatory pseudotumor is a pathologically defined disease and specifically refers to a benign proliferation of inflammatory and myofibroblastic spindle cells. Inflammatory hepatic pseudotumors are masses consisting of proliferation of inflammatory cells and fibrous stroma that occur in children and young adults.212 The cause of these lesions is unknown, although they have been hypothesized to represent sequelae of previous hepatic infection. Although approximately one-third of patients with hepatic inflammatory pseudotumor are asymptomatic, the remainder will have symptoms and serum chemistries suggesting inflammation of the liver. In one study, symptoms included fever, fatigue, weight loss, abdominal pain, nausea, and vomiting of 2 weeks to 6 months in duration.213 Laboratory studies showed elevations in C-reactive protein, sedimentation rate, and elevated WBC count in half of patients, and abnormalities in liver function were elevated in a minority. Knowledge of these signs and symptoms of inflammation can be an important clue to the diagnosis of inflammatory pseudotumor, because most solitary hepatic masses will present asymptomatically, without laboratory abnormalities. At US, inflammatory pseudotumors have been described as both hypoechoic and hyperechoic masses with increased through transmission and visualization of multiple septa.214,215 Nonspecific imaging characteristics are seen at CT. Commonly, the tumors are hypoattenuating to liver parenchyma on unenhanced images, and variable patterns of enhancement are seen after contrast material administration, including peripheral enhancement or enhancement of multiple internal septa.213,215 Calcification may be seen by

184 Diagnostic Abdominal Imaging Imaging Notes 3-7. Hepatic Inflammatory Pseudotumor Inflammatory pseudotumor typically appears as a nonspecific liver mass on cross-sectional imaging exams. However, with the exception of liver abscesses, this is the only liver mass that characteristically presents with signs and symptoms of systemic inflammation such as fever and elevations in serum white blood cell count, sedimentation rate, and C-reactive protein.

CT in some cases.215 Variable appearances of inflammatory pseudotumors of the liver have been described at MRI as well. The lesions are typically hypointense on T1-weighted images and hyperintense on T2-weighted images, with variable enhancement patterns.216,217 A delayed hyperenhancing fibrous pseudocapsule may be seen.215 However, as is suggested by its name, pseudotumors can mimic focal malignant lesions, and biopsy may be the only method of definitively excluding malignancy.218

Pseudolymphoma Pseudolymphoma is a term representing neoplastic-like proliferation of inflammatory cells, without evidence of clonal expansion. Most commonly seen in the skin, pseudolymphomas have been reported in multiple visceral compartments. Hepatic pseudolymphomas are relatively rare and are thought to be associated with inflammatory or autoimmune insults to the liver.219 Pseudolymphomas usually present as a focal mass, with nonspecific imaging characteristics similar to that of lymphoma,220 although hypervascularity has been reported.221 The diagnosis requires pathologic evaluation with immunohistochemical and/or flow cytometry evaluation to distinguish this from true hepatic lymphoma.

Hepatic pseudolesions Unlike inflammatory pseudotumor and pseudolymphoma that have defined pathologic characteristics, the term pseudotumor is ubiquitous in the radiology literature, and is nonspecifically used to refer to any masslike lesion without a neoplastic cause. In the liver, diffuse infiltrative processes sparring certain regions or cirrhosis with nodular regeneration can result in a “pseudotumor” on imaging.222,223

Focal and nodular steatosis: Hepatic steatosis is usually diffuse or regional in distribution. However, nodular variants are sometimes encountered.224 Focal hepatic steatosis is a common phenomenon and is usually seen in characteristic locations, most commonly in the medial segment left lobe abutting the ligamentum teres.225 Less typically, steatosis can present as a multinodular variant that may be confused with metastatic disease (see Figure 3-17).226,227 The demonstration of microscopic fat on

chemical shift–sensitive MRI can help avoid this diagnostic dilemma. On contrast-enhanced CT and MR, enhancing vessels can sometimes be seen coursing through steatosis regions, a finding that distinguishes these lesions from primary or secondary hepatic masses (see Figure 3-17).228

Focal fatty sparing: The obverse of focal hepatic steatosis, focal fatty sparing, can also occur. In this situation, there is fatty infiltration of the majority of the liver with small focal regions of sparing that on imaging can resemble a focal liver mass. Focal fatty sparing occurs most commonly around the gall bladder fossa, in the left lobe of the liver, or abutting the falciform ligament (see Figure 3-18).229,230

MULTIPLE MASSES OR NODULES OF THE LIVER Like solitary masses and nodules, multifocal hepatic masses or nodules can be solid or cystic in character, features that can help distinguish the underlying cause of the lesion.

Multiple Cystic or Cystic-Appearing Liver Masses Simple idiopathic cysts are the most common multifocal cystic-appearing lesion of the liver. Other causes include multiple hemangiomas, cystic metastasis, autosomal dominant polycystic liver disease, and multifocal liver abscesses (Table 3-3).

Idiopathic cysts The incidence of idiopathic hepatic cysts in the general population is not known; however, 1 study of 1541 consecutive US examinations of the liver showed an 11.3% incidence of hepatic cysts.231 There were no cysts found in 413  individuals younger than 40 years, suggesting that the incidence of idiopathic hepatic cysts increases with increasing age. The cause for this common hepatic abnormality is not known; however, they have no clinical significance, other than that they can be confused with metastasis and other hepatic neoplasms. Rarely, hepatic cysts can be symptomatic, particularly in patients with polycystic liver or kidney disease, and they can be treated by surgical unroofing procedures or catheter drainage with or without sclerosis.232 They are composed of cuboidal epithelium surrounded by fibrous stroma and contain serous fluid.233 Their incidence increases with age and they are more commonly multiple rather than solitary.27 Except when very small, they unequivocally can be diagnosed by any of the cross-sectional imaging techniques through demonstration of the characteristic imaging features of a cyst. Idiopathic cysts can be subdivided into those that are simple and those that are complex.

Simple cysts: Hepatic cysts are round or oval in cross section with imperceptible walls and contain simple fluid.

Chapter 3 Imaging of the Liver 185

A

B

C

D

E

F

Figure 3-17 Nodular Steatosis in 2 Patients A and B. This 66-year-old man was being evaluated for a renal mass. Unenhanced (A) and enhanced (B) CT images show a focal zone of decreased attenuation within the liver. Note the geographic borders and the presence of undisturbed vessels passing through the lesion. These are typical features of focal fatty infiltration. C-F. This 46-year-old woman had an islet cell

transplant. Opposed-phase T1-weighted MRI sequence (C and D) shows innumerable small, low-signal lesions throughout the liver. In-phase T1-weighted sequence (E and F) shows no evidence of the small lesions. This combination of findings indicates the presence of fat within the lesions and represents multinodular hepatic steatosis.

On US examinations, they will appear anechoic (uniformly black) and have increased through transmission;234 on CT examinations, they will have an attenuation between 0 and 20 HU and will not enhance with IV contrast.235 On MRI

examinations, they will be uniformly low-intermediate intensity (medium gray) on T1-weighted sequences and uniformly high intensity (bright white) on T2-weighted sequences (see Figure 3-19).236,237

A

B

C

D

E

F

Figure 3-18 Focal Fatty Sparing Appearing as a Liver Mass in 3 Patients A and B. This 38-year-old woman had right upper quadrant pain. US of the liver (A) shows diffuse hyperechogenicity, a finding associated with steatosis. Within the right hepatic lobe (B), this hypoechogenic mass (arrow) was discovered. Subsequent MRI (not shown) was pathognomonic for focal fatty sparing. C and D. This 62-year-old women had colon cancer with local spread to the pelvic nodes and peritoneum. Opposed-phase (C) and in-phase (D) imaging demonstrates diffuse hepatic steatosis. There is a focal nodule (arrows) that is both bright on opposed-phase imaging and iso-intense on in-phase imaging. No abnormality was seen on diffusion imaging, and the lesion did not hyperenhance following gadolinium. FDG-PET imaging was also negative for hepatic metastatic disease. On serial imaging, the lesion varied in size based on degree of steatosis. These findings are compatible with nodular sparing of steatosis. E and F. This 62-year-old woman had lung cancer. Unenhanced CT of the abdomen viewed with narrow windows shows a focal masslike region (arrowheads) adjacent to the porta hepatis that could be confused with a liver metastasis. However, the liver is diffusely low attenuation relative to the spleen, indicating hepatic steatosis. This pseudolesion represents focal fatty sparing.

Chapter 3 Imaging of the Liver 187 Table 3-3. Multiple Cystic or Cystic-Appearing Liver Masses 1. Idiopathic cysts a. Simple cystsa b. Complex cysts 2. Polycystic liver disease

Imaging Notes 3-8. Multiple Nodular Lesions of the Liver • Idiopathic cysts and hemangiomas are the most common cause of multiple cystic-appearing nodules of the liver. • Metastases are the most common cause of multiple solid-appearing nodules of the liver.

3. Multiple cystic or cystic-appearing neoplasms a. Multiple hemangiomas b. Biliary hamartoma c. Cystic and mucinous metastases 4. Multifocal abscesses a. Fungal microabscesses b. Acute disseminated tuberculosis c. Bacterial abscesses d. Amebic abscesses aMost common appear in bold.

Complex cysts: Complex cysts are cysts that have septations, a thickened wall, and/or have contents other than simple fluid within them as a result of prior infections or hemorrhage (see Figure 3-19). On US, internal septations and complex mildly echogenic fluid and or debris can be shown.238 Higher CT attenuation values can be seen internally because of increased protein content, such as from old hemorrhage.235 On MRI, proteinaceous or hemorrhagic cysts will have T1 shortening and therefore higher T1 signal.239,240

Polycystic liver disease Polycystic liver disease encompasses a spectrum of imaging presentations. A polycystic liver is often an accompanying feature of autosomal dominant polycystic kidney disease (ADPCK), presenting in upwards of 70% to 80% of such patients.241 However, polycystic liver disease can be present as a distinct entity without renal involvement.242 Patients with polycystic liver disease demonstrate DNA alterations that map to a different genetic locus than that of ADPCK.243-245 Although usually asymptomatic, polycystic liver disease can cause symptoms, including abdominal pain,246 portal hypertension,247,248 and lower extremity swelling (because of IVC obstruction).247,248 Even in the setting of extensive hepatic involvement, liver dysfunction is rare.249 Imaging of polycystic liver disease by either US, CT, or MRI will reveal the extent of liver involvement. Appearance is variable depending on the degree of hepatic involvement. The liver is generally enlarged—sometimes massively. Cysts of varying size and number are scattered throughout the liver, typically with areas of interspersed normal liver (see Figure 3-20). Most cysts are simple in appearance, although septations, hemorrhage, or thickened cyst walls can be seen. Imaging is generally done to confirm the presence of cysts as the cause of hepatomegaly, and to document the size and location of cysts when interventional or surgical therapies

are planned in the setting of intractable symptoms. Often, vascular compression of the portal vein or IVC can be seen, and imaging can be used to plan directed therapy toward the lesion(s), resulting in vascular compromise.

Multiple cystic or cystic-appearing neoplasms There are 2 neoplasms that commonly appear as multiple cystic-appearing lesions in the liver: hemangiomas and biliary hamartomas. There are also some cystic or mucinous malignancies that when they metastasize to the liver cause multiple cystic liver metastasis.

Multiple hemangiomas: Multiple hemangiomas are the most common cause, after idiopathic cysts, of multiple cystic-appearing masses of the liver. Hemangiomas are solid vascular tumors, but because they are largely composed of blood-filled channels, they can have imaging features of a cystic lesion on unenhanced CT and MRI examinations. On contrast-enhanced CT and MRI, hemangiomas have a characteristic enhancement pattern, discontinuous peripheral globular enhancement, a finding pathognomonic for hemangioma. On US examinations, hepatic hemangiomas appear as a densely echogenic solid masses of the liver. Hemangiomas are discussed most completely under the prior heading: Solitary Cysts and Cystic-Appearing Lesions.

Biliary hamartoma: Biliary hamartomas are small benign tumors composed of dilated bile ducts with interspersed fibrous or hyaline stroma. These are relatively common hepatic lesions and can be incidentally found in slightly less than 1%of autopsies, and are not clinically significant.52,53 Biliary hamartomas can be seen by US, contrastenhanced CT, and MRI52,54-57 and characteristically appear as small, usually 200 ms) are used in conjunction with high-resolution 3D or slab 2D techniques to identify smaller intrahepatic ducts via their relationship to the larger intrahepatic and extrahepatic ductal system on these projection and volumetric techniques.

Chapter 4 Imaging of the Gallbladder and Biliary System 285

A

B Figure 4-44 Choledocholithiasis with Biliary Dilation This 18-year-old woman complained of several months of right upper quadrant abdominal pain. A. Transverse US demonstrated intrahepatic biliary ductal dilation. B. The common bile duct (large arrow) measures 13 mm. C. Multiple gallstones are seen within the gallbladder (arrowhead). Note the shadowing behind the gallstones (small arrows). Although no stone is demonstrated in the common bile duct, the presence of a stone can be inferred from the bile duct dilation. A stone was endoscopically removed from the distal common bile duct.

C

Bile duct obstruction Bile duct obstruction can result from a variety of causes, including gallstones, malignancies, infections, autoimmune disorders, and congenital abnormalities (Table 4-3).

Choledocholithiasis: Choledocholithiasis is the most common cause of biliary obstruction. In most cases, gallstones will also be present within the gallbladder, and the patient will present with signs and symptoms of acute biliary ductal obstruction such as elevations in bilirubin, alkaline phosphatase, or γ-glutamyl transpeptidase associated with right upper quadrant pain. Bile duct stones are very small structures and are very difficult to detect with cross-sectional imaging. When common duct stones are suspected, the distal common bile duct should be carefully evaluated for the presence of a stone. In most cases, US

fails to adequately visualize the distal common bile duct, and the presence of a stone can only be inferred from the presence of dilation of the common bile duct and stones in the gallbladder (see Figure 4-44). If the distal common bile duct can be visualized, a common duct stone will appear as a hyperechoic, shadowing structure within the distal duct, with proximal ductal dilation. On CT scans, calcified common duct stones will be seen as a few-millimeter-size hyperattenuating structures within the low attenuating common bile duct but most noncalcified stones will be missed (see Figures 4-47 and 4-48). Heavily T2-weighted MRI sequences will show the common duct stone to appear as a few-millimeter, low-signal structure within high-attenuation distal common bile duct lumen (see Figure 4-47). In general, MRCP is now the diagnostic study of choice for the evaluation of suspected choledocholithiasis. On the

286

Diagnostic Abdominal Imaging

Malignant biliary obstruction: Malignancy is another common cause of bile duct obstruction, especially pancreatic carcinoma involving the head of the pancreas but also other pancreatic malignancies, intrahepatic and extrahepatic cholangiocarcinomas, hepatocellular carcinoma, other hepatic malignancies, ampullary carcinomas, duodenal carcinomas, metastasis, and lymphoma. Except for ampullary carcinoma and some cholangiocarcinomas, most of these malignancies will be easily detected by cross-sectional imaging, centered in the organ in which they arose.

Figure 4-45 Ampullary Carcinoma This 69-year-old man presented with jaundice. ERCP demonstrates a sharply marginated filling defect (arrow) in the distal common bile duct. Note the shouldered margins of the lesion relative to the duct. Differential diagnosis would include a radiolucent stone and small ampullary tumor. Biopsy was diagnostic of a small ampullary carcinoma.

other hand, ERCP and PTC are primarily used for interventions requiring the removal of biliary stones and are usually not necessary for the diagnosis of choledocholithiasis; however, occasionally all noninterventional imaging studies fail to diagnose the presence of biliary stones, which are subsequently diagnosed by ERCP or PTC (see Figure 4-49). For details about types and etiologies of gallstones, please see the heading Unique Disorders of the Biliary System previously discussed in this chapter. Mirizzi syndrome is a rare cause of biliary obstruction that results from one or several gallstones lodged in the cystic duct or gallbladder neck causing common bile duct obstruction and obstructive jaundice. In order for this to occur, the cystic duct must run parallel to the extrahepatic bile duct. It is important to identify this anatomic variation preoperatively to avoid inadvertent ligation of the common bile duct during cholecystectomy (see Figure 4-50).77-79 The obstruction may be caused directly by the gallstone or by the inflammatory reaction surrounding the affected duct. If the impaction is not resolved, erosion of the walls of the cystic duct and common bile duct can rarely result in the formation of a fistulous connection between the Hartman pouch of the gallbladder (the outpouching of the gallbladder wall at the junction of the neck and cystic duct) and adjacent common bile duct (cholecystocholedochal fistula) or stricture.

Cholangiocarcinoma: Cholangiocarcinoma is an adenocarcinoma arising from bile duct epithelium. These are typically infiltrating tumors that spread along the biliary tracts and are often only minimally conspicuous. These neoplasms induce a substantial amount of fibrosis in the surrounding tissues, a feature that causes prolonged delayed enhancement of the tumor with intravenous contrast and can be an important clue to identifying these tumors.80 Cholangiocarcinoma is associated with chronic inflammation of the biliary epithelium81 and is, therefore, seen with increased frequency in patients with biliary stones, primary sclerosing cholangitis, infections with clonorchiasis species, choledochal cysts, and Caroli disease.80 Cholangiocarcinoma can be classified into intrahepatic and extrahepatic cholangiocarcinoma. Intrahepatic cholangiocarcinoma can be divided further into hilar and peripheral cholangiocarcinoma and is classified as mass-forming (the most common), periductal infiltrating, and intraductal.81 The most common site of tumor involvement is the confluence of the right and left hepatic ducts, commonly called a Klatskin tumor.82 In many cases, no discrete mass is identified within the porta hepatis, and so these tumors typically present as dilated intrahepatic ducts without imaging evidence of communication between the right and left hepatic ducts (see Figure 4-51).37 Occasionally, careful observation will demonstrate focal thickening of the walls of the central bile ducts, greater than 5 mm, or faint enhancing tumor in the central ducts associated with the presence of distal duct obstruction on contrast-enhanced CT or MRI examinations (see Figures 4-52 and 4-53).81 Appearance as an exophytic mass can be occasionally identified, most commonly in cholangiocarcinomas of peripheral bile ducts. These tumors appear as a discrete mass within the hepatic parenchyma and will usually demonstrate obstruction of the ducts distal to the mass (see Figures 4-54 and 4-55). This is an important clue to peripheral cholangiocarcinomas. Although other hepatic neoplasms can cause bile duct obstruction, none do it with the frequency of cholangiocarcinoma. Among the least common manifestations of cholangiocarcinoma is the appearance of a small polypoid lesion projecting into the lumen of the bile duct. Most cholangiocarcinomas will have typical imaging characteristics depending on the morphologic type (massforming, periductal infiltrating, or intraductal). Peripheral cholangiocarcinomas are the type of cholangiocarcinoma most likely to be mass-forming.81 On US, the mass-forming

Chapter 4 Imaging of the Gallbladder and Biliary System 287

A

B

C

D

Figure 4-46 Bile Duct Dilation on Unenhanced and Enhanced CT This 68-year-old man presented with jaundice and back pain. A and B. Unenhanced CT shows 2 distinct attenuations in the portal triads, low attenuation branching structures (black arrowheads) that represent moderately dilated bile ducts and slightly higher attenuation branching structures

(white arrowheads) that represent the portal veins. C and D. The dilated bile ducts (black arrowheads) are seen to greater advantage on post contrast-enhanced images. Images of the pancreas demonstrated a small mass in the head of the pancreas (not shown), which was later proven to represent an adenocarcinoma.

288 Diagnostic Abdominal Imaging

A

B

C

D

E

F

G

H

Figure 4-47 Choledocholithiasis This 58-year-old man presented with 2 weeks or right upper quadrant pain. US examination (not shown) demonstrated gallstones and dilation of the common bile duct. A-D. Contrastenhanced CT examination shows gallstones within the gallbladder (white arrowhead), normal intrahepatic bile ducts (black arrow) but mild dilation of the common bile duct (white arrow). Within the pancreatic head is a small linear calcification (black arrowhead) indicating the presence of a distal common bile duct stone. E-G. Highly T2-weighted axial MR images confirm the presence of gallstones (small white arrowhead), mild dilation of the common bile duct (large white arrowhead), and a partially obstructing stone in the distal common bile duct (small white arrow). H. Note how much easier it is to see the stone in the distal common bile duct on this MRCP.

Chapter 4 Imaging of the Gallbladder and Biliary System 289 Table 4-3. Causes of Bile Duct Dilation 1. Gallstonesa 2. Prior surgery 3. Malignancy a. Pancreatic carcinoma b. Other pancreatic malignancies c. Cholangiocarcinoma d. Ampullary carcinoma e. Duodenal carcinoma f. Hepatocellular carcinoma g. Metastasis h. Lymphoma 4. Infections a. Clonorchis sinensis b. Ascaris lumbricoides 5. Inflammatory and postinflammatory strictures a. Prior gallstone passage b. Sclerosing cholangitis c. Ascending cholangitis d. Pancreatitis 6. Congenital a. Choledochal cysts b. Biliary atresia 7. Other a. Sphincter of Oddi spasm aMost common causes are listed in bold.

A

B

Figure 4-48 Choledocholithiasis This 60-year-old woman presented with weight loss and an increased serum alkaline phosphatase and gamma-glutamyl transpeptidase (GGT). A. Coronal MRCP shows intrahepatic and extrahepatic (arrow) biliary ductal dilation to a level near the sphincter of Oddi. Notice the sharp occlusion of the distal common bile duct, suggesting an intraluminal filling defect.

type will usually appear as a homogeneous mass with an irregular well-defined margin. A peripheral hypoechoic rim is seen in approximately 35% of tumors.81,83 The echogenicity of the tumor will depend on tumor size, with greater than 3 cm appearing hyperechoic and tumors less than 3 cm being hypo- or isoechoic.81,84 The CT appearance of the mass-forming type is a mass with homogeneous attenuation and irregular peripheral enhancement with gradual centripetal enhancement, capsular retraction, possibly satellite nodules, and vascular encasement (see Figures 4-54 and 4-55).80,81,85-87 Delayed enhancement is seen in areas with marked fibrosis. Hepatolithiasis and ductal dilation with obliteration of the portal vein and atrophy of the involved segment can also be seen.81,88 On MR, the mass demonstrates an irregular margin, with high signal intensity on T2-weighted images and low signal intensity on T1-weighted images with peripheral centripetal enhancement and delayed enhancement in areas with fibrous stroma.81,89,90 The appearance of the periductal infiltrating type has also been described. Most hilar cholangiocarcinomas are of this type.81,93-95 On US, the infiltrating type of cholangiocarcinoma can appear as a small masslike lesion or diffuse bile duct thickening. The bile duct lumen can be occluded depending on the tumor involvement.81,91,92 On CT, there is diffuse periductal thickening with increased enhancement due to tumor infiltration along with dilated or irregular narrowed ducts and peripheral ductal dilation. During the arterial phase of hepatic enhancement, cholangiocarcinomas can appear faintly hyperattenuating to liver but on portal-phase imaging, the tumor will often be isoattenuating relative to normal liver parenchyma, rendering the tumor mass essentially

C B and C. Enhanced CT images demonstrate low attenuation, branching, tubular structures (arrowheads) in the central liver, typical of intrahepatic biliary ductal dilation. There is also a small high-attenuation filling defect (arrow) within the distal common bile duct, within the head of the pancreas, typical of a biliary ductal stone.

290

Diagnostic Abdominal Imaging

A

B

Figure 4-49 Choledocholithiasis by ERCP and PTC A. Images obtained during ERCP show multiple filling defects (arrows) within the common hepatic duct indicating retained stones after cholecystectomy. These images were taken in preparation for stone removal. B. This patient had persistent

pain following cholecystectomy. CT and US examinations showed persistent bile duct dilation but failed to identify a cause. PTC through a T-tube reveals retained stones in the distal common bile duct as filling defects (arrows) within the contrast column.

invisible (see Figures 3-15 and 4-53). Delayed-phase imaging at 3 to 5 minutes following contrast enhancement will typically show the tumor to be hyperenhancing relative to the normal liver parenchyma. This is the phase of imaging at which the periductal infiltrating forms of cholangiocarcinoma is most evident; unfortunately, this phase is not part of most typical imaging protocols and requires the observer to suspect the presence of the tumor and perform delayedphase imaging. This delayed enhancement is characteristic of scarring of any type and is a result of the scirrhous character of cholangiocarcinomas. Even though CT has greater spatial resolution, MRI examinations are the study of choice for evaluating suspected cholangiocarcinomas because the tumor is more conspicuous both before and after contrast enhancement.21 In addition, MRI examinations allow for production of MR cholangiograms, an important adjunctive imaging process. Cholangiocarcinomas are typically isointense with liver parenchyma on T1-weighted sequences and will appear hypointense on T2-weighted sequences because of the surrounding scirrhous reaction. Following intravenous contrast administration, cholangiocarcinomas will have similar imaging characteristics with CT contrast enhancement:

faintly hyperintense during arterial phases, isointense during portal phases, and hyperintense on delayed-phase imaging (see Figure 4-52). The intraductal type can present with localized or diffuse ductectasia occasionally with an echogenic intraductal component on ultrasound. On CT and MR, there are several appearances. There can be diffuse marked ductal dilation with an intraductal mass that can be either hypoattenuating or isoattenuating on precontrast imaging, with enhancement on postcontrast imaging, an intraductal polypoid mass within a focal dilated duct, an intraductal castlike lesion, or a focal stricture with mild proximal ductal dilation.81 Ampullary neoplasms: Periampullary tumors are defined as arising within 1 cm of the ampulla of Vater and include tumors of the ampulla, distal bile duct, pancreas, and duodenum. In a series of 647 consecutive periampullary tumors over a 20-year period by Michelassi et al, 85% were adenocarcinomas of the head of the pancreas, 6 % were cholangiocarcinomas of the distal common bile duct, and 4% were tumors of the ampulla itself.96,97 Ampullary adenomas and adenocarcinomas are neoplasms that arise from the glandular epithelium at the

Chapter 4 Imaging of the Gallbladder and Biliary System 291

A

B

C

D

Figure 4-50 Mirizzi Syndrome This middle-aged man presented with fever, jaundice, and right upper quadrant pain. A. US of the biliary system shows a distended gallbladder and cystic duct with diffuse thickening of the gallbladder wall. B. The cystic duct (large arrowhead) has a long course, and a large echogenic stone (small arrowhead) with shadowing is discovered at the distal end of the duct. There is also a dilated common bile duct (small arrow). C. Sagittal images

of the liver demonstrate multiple dilated intrahepatic ducts (arrowheads) with their echogenic walls distinguishing them from hepatic and portal veins. This constellation of findings suggests Mirizzi syndrome. D. Percutaneous transhepatic cholangiogram shows dilation of the intrahepatic and extrahepatic bile ducts. The cystic duct is massively distended and there is a stone at the junction of the cystic duct and common bile duct. This appearance is diagnostic of Mirizzi syndrome.

292

A

Diagnostic Abdominal Imaging

B

C Figure 4-51 Cholangiocarcinoma This 86-year-old man presented with jaundice and abdominal pain. A-C. CT images in the late portal phase of enhancement demonstrate multiple low-attenuation tubular structures in the liver, characteristic of dilated small and large intrahepatic biliary ducts (arrowheads). The common bile duct (black arrow) is normal in caliber, indicating obstruction at the porta hepatis. No obvious mass is identified although there is minimal soft tissue (white arrow) in the porta hepatis. This is the typical appearance of a central cholangiocarcinoma. The tumor infiltrates along the bile ducts and is difficult to visualize. D. Percutaneous transhepatic cholangiogram confirms dilated intrahepatic ducts (white arrowheads) with a normal diameter common bile duct (white arrow). The central portions of the ducts are occluded (black arrowheads). A percutaneous biliary drain has been placed across the obstructed bile duct (black arrows).

D

ampulla of Vater within the medial wall of the second portion of the duodenum into which the common bile duct and main pancreatic duct drain.96 These tumors occur in both men and women with a male-to-female ratio of 2:196-98 and an average age of 60 years.97-100 Symptoms are due to biliary ductal obstruction and include intermittent jaundice and acholic stools.101 Partial obstruction can lead to ascending cholangitis with symptoms of fever, chills, and right upper quadrant pain.102 Associated pancreatic ductal

obstruction causing pancreatitis has also been described.103 Patients with adenomas and adenocarcinomas in other sites of the GI tract, including those with familial adenomatous polyposis syndromes, are at increased risk for developing ampullary tumors.24,96,104-109 Ampullary neoplasms are usually small at the time of diagnosis, probably because of the relatively early onset of symptoms, with the average size 3 cm or less.96,98,102 If the tumor is confined to the ampulla, it may be difficult to

Chapter 4 Imaging of the Gallbladder and Biliary System 293 Figure 4-52 Cholangiocarcinoma With Bile Duct Enhancement This 85-year-old man presented with jaundice. A. MR cholangiogram shows dilated peripheral bile ducts as bright enhancing tubes and a dilated gallbladder (arrowhead). Note the narrowing of the central bile ducts in the porta hepatis (white arrow) a finding associated with the Klatskin-type cholangiocarcinoma. B. T2-weighted MRI through the porta hepatis shows mildly dilated peripheral bile ducts, with increased soft tissue surrounding the central bile ducts. T1-weighted MRI (C) before and (D) after gadolinium administration shows enhancement of the central bile ducts. These findings are highly associated with cholangiocarcinoma, which was subsequently proven by endoscopic brush biopsy.

Mil001_Fig_04-52A

A

B

C

visualize on endoscopy.110,111 When visible, endoscopic findings are usually a prominent papilla or submucosal mass with nonvisualization of the tumor itself. Blind biopsy with cytologic and histologic evaluation of material obtained is usually required for definitive diagnosis.96 The biliary or pancreatic duct dilation caused by these tumors is usually the only finding on cross-sectional imaging, including US, CT, and MR. The double-duct sign (dilation of the common bile duct and pancreatic duct) can be readily apparent but the tumor is not visualized in most cases.96,112,113 In some cases, a small soft-tissue mass will be detected at the ampulla of Vater on CT, but this is usually due to the tumor protruding into the duodenum that is distended by contrast material (see Figure 4-56).96 On MR, visible ampullary carcinomas are of low signal intensity relative to the pancreas on both T1- and T2-weighted images and enhance to a lesser extent than pancreas on immediate enhanced images. On delayed postcontrast sequences, these lesions demonstrate heterogeneous enhancement with

D

variable rim enhancement.24,114 Cholangiopancreatography by either MRCP, ERCP, or PTC will demonstrate sharp shouldering rather than gradual tapering of the distal common bile duct, suggesting the presence of a neoplasm, but this diagnosis is difficult because the short transmural segment of the pancreaticobiliary tree contains little to no fluid (see Figure 4-45).24 Unfortunately, this shouldering can be mimicked by gallstone impacted in the distal common bile duct. Fortunately, both impacted stones and periampullary tumors will require evaluation/therapy by ERCP. Pancreatic malignancy: As stated above, pancreatic cancer is the most common malignant cause of bile duct obstruction.96-97 In most cases, CT and MRI will readily identify a mass in the pancreatic head as the cause of biliary obstruction (see Figure 4-57). Rarely, the pancreatic tumor can locally invade the distal common bile duct and infiltrate along the duct, making it very difficult to differentiate between a primary pancreatic neoplasm and a

294

Diagnostic Abdominal Imaging

A

B

Figure 4-53 Cholangiocarcinoma With Subtle Wall Enhancement This 50-year-old man presented with jaundice. A and B. CT images in the early portal phase of enhancement demonstrate diffuse bile duct dilation. Note how the central ducts (arrowheads)

appear slightly higher attenuation than the peripheral ducts (arrows). This represents enhancing intraluminal tumor. Biopsy was diagnostic of cholangiocarcinoma.

primary biliary tumor that has invaded the pancreatic head. Enlarged peripancreatic lymph nodes can also make delineating tissue planes extremely challenging. Other extrabiliary neoplasms: Occasionally, other malignancies can result in biliary obstruction. As previously stated,

primary hepatocellular carcinoma or hepatic metastases are sometimes impossible to differentiate from cholangiocarcinoma (see Figure 4-58).81 Rarely, lymphoma has also been described as a cause for biliary obstruction (see Figure 4-59).115 Duodenal carcinoma and other retroperitoneal cancers can also cause obstruction of the common bile duct.

A

B

Figure 4-54 Cholangiocarcinoma Appearing as a Discrete Mass This 51-year-old woman presented with abnormal liver function tests. A and B. Images from an enhanced CT scan demonstrate an indistinctly marginated mass in the central liver causing

mild bile duct dilation (arrows). In this early phase of contrast enhancement, the peripheral portion of the mass enhances preferentially. Biopsy was diagnostic of cholangiocarcinoma.

Chapter 4 Imaging of the Gallbladder and Biliary System 295

A

B

C

Figure 4-55 Cholangiocarcinoma with Bile Duct Dilation This 72-year-old man presented with new-onset jaundice. A-C. CT images demonstrate multiple low attenuation branching

tubes typical of dilated bile ducts. In the central liver is an irregularly marginated mass with peripheral enhancement, typical of a cholangiocarcinoma.

Disorders of the sphincter of oddi: Disorders of the

gallstones. Papillary dysfunction can be related to spasm of the sphincter or abnormal sphincteric peristalsis.24,116

sphincter of Oddi that can result in intermittent biliary obstruction include papillary dysfunction and stenosis. As with stenosis seen in other parts of the biliary system, ampullary stenosis results from bouts of inflammation at the ampulla of Vater and usually is due to the passage of

Inflammatory biliary strictures: There are a variety of inflammatory causes of biliary strictures, including prior surgery, prior choledocholithiasis, autoimmune diseases,

A

B

Figure 4-56 Ampullary Carcinoma Detected by CT This 86-year-old man presented with intermittent abdominal pain and jaundice. A and B. Images from an unenhanced CT demonstrate mild dilation of the common bile duct (arrowhead)

and a small polypoid tumor (white arrow) in the duodenum at the location of the sphincter of Oddi, characteristic of an ampullary carcinoma.

296

Diagnostic Abdominal Imaging

A

B

C

D

Figure 4-57 Biliary Obstruction Due to Pancreatic Carcinoma This 70-year-old woman presented with jaundice and dizziness. A-D. Contrast-enhanced CT images demonstrate dilation of the intrahepatic (small arrows) and extrahepatic (white arrowhead)

bile ducts, indicating an extrahepatic biliary obstruction. In the head of the pancreas is an inhomogeneous mass (large arrow) typical of a pancreatic adenocarcinoma. There is also metastatic adenopathy near the celiac axis (black arrowhead).

A

B

Figure 4-58 Biliary Obstruction by Metastasis This 60-year-old man had pancreatic carcinoma. A and B. T2-weighted sequences demonstrate normal right bile ducts but dilated left ducts. There is a faint hyperintense mass

C

D

(arrowhead) in the left side of the porta hepatis representing a metastasis. C and D. Post gadolinium T1-weighted images confirm obstruction of the left biliary radicals by a central metastasis (arrowhead).

Chapter 4 Imaging of the Gallbladder and Biliary System 297

A

B

C

D

Figure 4-59 Lymphoma Causing Bile Duct Obstruction This 38-year-old man presented with painless jaundice. A and B. US of the liver demonstrates mild intrahepatic ductal dilation (arrowheads) and a slightly hypoechoic mass (arrows). C and D. CT scan confirms the presence of ductal dilation (arrowheads)

and a hypoattenuating mass (arrows). This might have represented any one of a number of primary hepatic neoplasms but surgical biopsy was diagnostic of primary hepatic lymphoma. No adenopathy or other imaging abnormality was identified.

biliary infections, radiation therapy, and blunt abdominal trauma (Table 4-4).

the duct on both sides of the narrowing.117 The most common cause of postoperative strictures is cholecystectomy, with strictures occurring in approximately 1 of every 400 to 500 cholecystectomies.120 Approximately 2.6% of patients develop biliary strictures after pancreaticoduodenectomy, whether it is performed for benign or malignant disease.121 Patients generally present with intermittent cholangitis with or without jaundice. Biliary strictures are among the most common surgical complications of liver transplantation and can develop within several months to several years after surgery. Strictures characteristically occur at 2 sites. Most common are strictures at the surgical anastomosis as a result

Biliary surgery: Stricture of the bile ducts is relatively uncommon after biliary surgery but is still among the most common causes of bile duct obstruction. These surgeries include open and laparoscopic cholecystectomy, liver transplantation, Whipple procedures, and surgical extraction of bile duct stones. Postoperative biliary strictures manifest by intrahepatic and extrahepatic biliary dilation defined as a diameter of more than 3 or 8 mm, respectively, with abrupt narrowing,117-119 nonvisualization of part of the duct, or focal narrowing of the duct with clear identification of

298

Diagnostic Abdominal Imaging

Table 4-4. Causes of Inflammatory Biliary Strictures 1. Prior biliary surgery 2. Prior choledocholithiasis 3. Sclerosing cholangitis 4. Biliary infections a. Ascending pyogenic cholangitis b. Clonorchis sinensis c. Ascaris lumbricoides d. AIDS-related cholangitis 5. Eosinophilic cholangitis 6. Abdominal trauma 7. Radiation therapy 8. Chronic pancreatitis

of excessive scarring (see Figure 4-60).117 Nonanastomotic strictures are caused by vascular compromise and resultant ischemia. Up to 50% of nonanastomotic strictures are caused by thrombosis of the hepatic artery. Other causes include prolonged preservation time, infectious cholangitis, and rejection. Nonanastomotic strictures usually

Figure 4-60 Biliary Stricture Following Liver Transplantation This 51-year-old who had undergone liver transplantation had rising biliary function tests. ERCP shows a smooth tapered stricture (white arrow) at the site of the biliary anastomosis. This is a known complication of liver transplantation. Note the small blind-ending cystic duct remnant (arrowhead) just distal to the site of the stricture.

involve a significant ductal segment and are located near the hilum.117 Traditionally, percutaneous transhepatic cholangiography or ERCP has been performed in patients in whom a postoperative bile duct injury is suspected (see Figure 4-60). Because both of these modalities are invasive and have limited ability to visualize portions of the biliary tree, MRCP has become the imaging modality of choice at many institutions in evaluation for postsurgical bile duct abnormalities. Choledocholithiasis: In addition to obstruction due to the stone itself, prior choledocholithiasis can also cause obstruction due to chronic inflammation and fibrosis, leading to biliary strictures.123 Patients present with right upper quadrant pain. Cross-sectional imaging demonstrates dilated bile ducts without evidence of a mass. MRCP, ERCP, and PTC will show a smooth tapered stricture of the affected bile duct without other lesion (see Figure 4-61). Sclerosing cholangitis: Sclerosing cholangitis is a rare, idiopathic, autoimmune disorder resulting in progressive inflammation of the biliary tree, causing fibrosis and multiple strictures of the bile ducts. Biliary damage leads to cirrhosis, portal hypertension, and liver failure in the majority of patients, and 10% to 15% of patients will develop cholangiocarcinoma.124 Approximately 70% of the time, sclerosing cholangitis will be associated with inflammatory bowel disease, usually ulcerative colitis but occasionally Crohn disease.124 In the remainder of cases, sclerosing cholangitis will be an isolated finding. Patients typically present in the fourth and fifth decades. Because of the association with inflammatory bowel disease, many patients are initially diagnosed prior to the development of symptoms. Those with symptoms can present with pruritis, jaundice, or fatigue.124,125 Liver function enzymes are usually elevated.124-126 Sclerosing cholangitis appears as alternating areas of narrowing and dilation of the bile ducts that resemble a string of beads, best appreciated by MRCP, ERCP, and PTC (see Figure 4-62). The beaded appearance of the bile ducts is more difficult to appreciate on cross-sectional imaging because the ducts traverse over multiple cross sections. On cross-sectional imaging, the ducts will have a variable diameter, rather than the normal smooth tapering seen with other ductal dilation. In some cases, the ducts appear as multiple cysts along the expected course of the larger bile ducts. CT and MRI can also demonstrate inflammation of the bile ducts as linear areas of enhancement along the portal triads (see Figure 4-63). Infectious cholangitis: Infectious cholangitis is most often due to ascending infection from the duodenum. Bacterial cholangitis is typically a complication of partial or complete biliary obstruction and is often a result of post surgical or post inflammatory strictures or choledocholithiasis.127,128 Patients will usually present with abdominal pain, jaundice and, in some cases, sepsis.127 Hepatic abscesses

Chapter 4 Imaging of the Gallbladder and Biliary System 299

A

B

C

D

E

F

Figure 4-61 Biliary Stricture Due to Prior Choledocholithiasis This 57-year-old woman with a history of prior gallstones and chronic renal failure was being evaluated for the presence of adenopathy. A-D. Enhanced CT scan demonstrates dilated intrahepatic (black arrowheads) and extrahepatic (black arrows) bile ducts and a dilated pancreatic duct (white arrowheads), suggesting an obstruction near the sphincter of Oddi. No

mass is seen in the pancreatic head on image (D). E-H. US examination confirms the presence of intrahepatic (white arrowheads) and extrahepatic (small white arrow) bile duct dilation and pancreatic duct dilation (large white arrowheads) and no mass in the pancreatic head. The examination also demonstrates multiple gallstones that were not seen on the CT scan.

300

Diagnostic Abdominal Imaging

G

H Figure 4-61 Biliary Stricture Due to Prior Choledocholithiasis (Continued) I. ERCP shows a smooth tapered stricture (black arrow) at the distal common bile duct probably due to prior stone passage.

I

are an important complication of ascending cholangitis and are one of the reasons for cross-sectional imaging of patients with suspected cholangitis.129 Uses of imaging include determining the level and etiology of obstruction as well as evaluating for possible complicating abscesses (see Figure 4-64).24 Imaging findings on CT, US, and MR include extrahepatic and/or intrahepatic biliary ductal dilation with or without stones. Note that biliary ductal dilation is usually in the central liver as opposed to dilation in the peripheral liver that is seen with primary sclerosing cholangitis.128 In Asia and Africa, the parasitic infections: Ascaris lumbricoides and Clonorchis sinensis are important causes of bile duct obstruction and chronic liver disease.130,131

Clonorchis infection is often known as “oriental cholangiohepatitis.” These organisms ascend into the biliary tree from the small bowel where they cause inflammation and subsequent fibrosis of bile ducts.24,132,133 They also cause multiple pigmented biliary calculi in the majority of patients. Both fibrotic strictures and biliary calculi can lead to chronic bile duct obstruction and recurrent secondary pyogenic cholangitis. Patients usually present with recurrent abdominal pain, jaundice, and fever requiring removal of stones and debris. Cross-sectional imaging will demonstrate multifocal regions of intrahepatic bile duct dilation (see Figure 4-65). MRCP, ERCP, and PTC will  demonstrate multiple biliary strictures with intraluminal debris and stones.

Chapter 4 Imaging of the Gallbladder and Biliary System 301

A

B

C

D

Figure 4-62 ERCP of Sclerosing Cholangitis A-C. Contrast-enhanced CT in this patient with sclerosing cholangitis demonstrates multiple regions of bile duct dilation (white arrowheads). Note how the ducts are variable diameter and

in some slices appear to be small cysts adjacent to the portal triads. This is the cross-sectional equivalent of beaded ducts. D. ERCP demonstrates the alternating dilated and stenotic appearance typical of “beading” of the bile ducts in sclerosing cholangitis.

302

Diagnostic Abdominal Imaging

A

B

C

D

Figure 4-63 Sclerosing Cholangitis This 40-year-old man had ulcerative colitis. A-C. Contrastenhanced CT images through the liver demonstrate mild intrahepatic biliary ductal dilation (black arrowheads). There is also diffuse enhancement of the bile ducts seen as areas of increased attenuation surrounding the bile ducts

(white arrowheads) that indicates diffuse inflammation of the bile ducts. In the setting of a patient with ulcerative colitis, these findings are indicative of sclerosing cholangitis. D. Image through the pelvis also shows enhancement of the sigmoid wall (white arrow) consistent with colonic inflammation in ulcerative colitis.

AIDS-Related cholangitis: Acalculous inflammation of the biliary system is a rare but well-documented complication in patients with AIDS. AIDS-related cholangitis is thought to be a complication of Cryptosporidium and/or cytomegalovirus infection and acalculous inflammation of the biliary tree.134 Patients typically present with right upper quadrant pain and/or epigastric pain, jaundice, or abnormal liver enzymes.134 Ultrasonographic evaluation generally shows dilation of the common bile duct and, occasionally, the intrahepatic bile ducts with thickening of the duct walls.134,135 A terminal ductal stricture is occasionally identified. Gallbladder wall thickening can be seen in some individuals. Computed tomographic findings are similar to US, and CT appears to be better at demonstrating the

intrahepatic ductal dilation but less sensitive in demonstrating the bile duct wall thickening.134 Pruning, an appearance resembling a shrub after it has lost the outermost leaves, of the distal common bile ducts can also been seen on CT. Patients can present with isolated strictures involving the extrahepatic or intrahepatic biliary ducts or ductal irregularities throughout the biliary tree. Both CT and US often demonstrate extrabiliary disease, including splenomegaly and enlarged portohepatic nodes.134 The overall appearance can closely resemble findings of papillary stenosis and sclerosing cholangitis. Eosinophilic cholangitis: Eosinophilic cholangiopathy is a rarely reported cause of benign biliary obstruction with an unknown cause. Histologically, there is transmural

Chapter 4 Imaging of the Gallbladder and Biliary System 303

A

B

C

Figure 4-64 Hepatic Abscesses Due to Ascending Cholangitis This 86-year-old woman developed fever and jaundice in the first few days following cholecystectomy. A and B. Unenhanced CT images demonstrate a cluster of fluid collections (white arrows) in the liver that had not been on previous examinations and

consistent with hepatic abscesses due to ascending cholangitis. C. Sagittal US also reveals an anechoic, irregularly marginated fluid collection (white arrow) with internal debris, typical of an abscess.

eosinophilic infiltration of the biliary tract, which can affect the gallbladder and/or the bile ducts. There can be peripheral eosinophilia.136 There are no specific radiologic signs for eosinophilic cholangiopathy. It should be considered in patients with eosinophilia and biliary obstruction.

When eosinophilic cholangiopathy is limited to the gallbladder, treatment with steroids may help patients avoid cholecystectomy. On US, CT, and MR, this appears as acalculous cholecystitis with gallbladder distention, gallbladder wall thickening, and pericholecystic fluid.136 When it involves the

A

B

Figure 4-65 Oriental Cholangiohepatitis This 66-year-old man, immigrant from Southeast Asia, presented with fever and right upper quadrant pain. A and B. Unenhanced CT images demonstrate intrahepatic biliary dilation

involving primarily the posterior segment. The extrahepatic ducts were normal in caliber. Further evaluation was diagnostic of oriental cholangiohepatitis.

304

Diagnostic Abdominal Imaging

biliary tree, there is nonspecific segmental or diffuse bile wall thickening, sometimes with biliary obstruction. On ERCP, there is irregularity of the bile duct wall, which may be in a beaded pattern mimicking primary sclerosing cholangitis.136

Other causes of bile duct stricture: Blunt abdominal trauma and radiation therapy have rarely been reported as causes of biliary strictures. Strictures resulting from external beam radiation therapy characteristically appear in the radiation field.137 Both blunt abdominal trauma and strictures from radiation therapy are nonspecific in appearance and present as abrupt termination of the common bile duct with proximal ductal dilation. Correlation with patient history is often necessary. These findings usually have a delayed presentation and can appear many years after the initial insult.138 Chronic pancreatitis can occasionally cause periampullary stenosis.139

Choledochal cysts Congenital cystic dilation of the bile ducts is a rare phenomenon collectively known as choledochal cysts. This dilation

type I

type III

type II

A

B

type IV A D

has been classified into 5 categories (see Figure 4-66). Type I cysts represent fusiform cystic dilation of the common bile duct (see Figures 4-67 and 4-68). Type II cysts are diverticular outpouching of the common bile duct (see Figure 4-69). Type III choledochal cysts are cystic protrusions of the common bile duct into the duodenum, also known as a choledochoceles (see Figure 4-70). Type IV cysts have been further subclassified. Type IVA cysts are dilation of both the intrahepatic and extrahepatic bile ducts. Type IVB cysts involve dilation of multiple segments of the extrahepatic bile duct with a normal intrahepatic biliary tree. Lastly, Type V choledochal cysts are multiple cystic dilations of the intrahepatic bile ducts, with normal extrahepatic bile ducts, a phenomenon called Caroli disease (see Figure 4-71). Caroli disease is an autosomal recessive disorder resulting in multifocal saccular dilation of the intrahepatic bile ducts.140 Ductal stones, hepatic fibrosis, and portal hypertension resulting in variceal bleeding are some associated complications.140,141 Patients also have an increased risk of developing cholangiocarcinoma. Caroli disease is also associated with autosomal dominant polycystic kidney disease.142-145

C

type V

type IV B E

Figure 4.66 Choledochal Cysts The biliary system is shaded green and the duodenum is shaded pink. A. Type I choledochal cyst: fusiform dilation of the common bile duct. B. Type II choledochal cyst: saccular outpouching from the common bile duct. C. Type III choledochal cyst: choledochocele or cystic protrusion of common bile duct into the

F duodenum. D. Type IVA choledochal cyst: dilation of both the intrahepatic and extrahepatic duct. E. Type IVB choledochal cyst: multiple cystic dilations of the common bile duct with normal intrahepatic duct. F. Type V choledochal cyst: Caroli disease or multiple cystic dilations of the intrahepatic ducts.

Chapter 4 Imaging of the Gallbladder and Biliary System 305

Figure 4-67 US of Choledochal Cyst US of the common bile duct (arrow) in this 62-year-old man demonstrates fusiform dilation of the proximal bile duct with a tapered, narrowed distal common bile duct within the head of the pancreas (arrowheads). The intrahepatic ducts appeared normal. This is the typical appearance of a type I choledochal cyst.

A

Choledochal cysts are thought to be the result of a congenital abnormal connection between the pancreatic duct and the common bile duct that results in the chronic reflux of pancreatic secretions into the bile duct, causing irritation and subsequent dilation.141 Although the majority will be diagnosed in infancy and childhood, occasionally they are first discovered during adulthood. The prevalence of choledochal cysts is higher in female patients. The classic triad of abdominal pain, right upper quadrant mass, and jaundice has been described but is present in less than onethird of patients.141 The classic US appearance of a choledochal cyst is a cystic or fusiform structure in the porta hepatis with apparent communication with the common hepatic duct and a normal-appearing gallbladder (see Figure 4-67).141 In this case, CT and MRI evaluation is more accurate in the evaluation of the intrahepatic biliary tree and the distal common bile, duct, which is occasionally obscured by bowel gas during US evaluation (see Figures 4-68 and 4-69).141,146 Further, MRCP, ERCP, and PTC can be more accurate in identifying the intrahepatic forms of choledochal cysts (Type IVA and Caroli disease) because the cystic intrahepatic ducts can be confused with intrahepatic cysts. Confirmation with hepatobiliary scintigraphy is occasionally performed

B

Figure 4-68 Choledochal Cyst Type I by ERCP and MRCP ERCP (A) and MRCP (B) of this 59-year-old woman demonstrates fusiform dilation of the common bile duct (arrows) with normal-appearing intrahepatic ducts (arrowheads), typical of a type I choledochal cyst. The shape of the choledochal cyst is different on ERCP from that on MRCP because the cyst is distended under the pressure of injection with ERCP, whereas it is in the normal physiologic state in MRCP.

306

Diagnostic Abdominal Imaging

A

B

Figure 4-69 Type II Choledochal Cyst A and B. Sequential coronal T2-weighted MRI images demonstrate normal proximal and distal common bile ducts

(small arrowheads) and normal gallbladder (large arrowhead) with a saccular outpouching (arrow) from the mid common bile duct, features diagnostic of a type II choledochal cyst.

and demonstrates accumulation and stasis of tracer in the dilated ducts and choledochal cyst.141,147 The US features of Caroli disease include bile duct dilation with intraluminal protrusions. On CT, there is a classic “central dot sign” which is a high attenuation dot visualized in the dilated intrahepatic bile ducts on unenhanced CT which enhance avidly after contrast administration.142 This enhancing focus is also seen on gadolinium-enhanced MR. The central dots correspond to intraluminal portal veins on ultrasound. Cholangiography demonstrates saccular or fusiform dilations of portions of the intrahepatic ducts that sometimes contain intraluminal filling defects corresponding to intrahepatic calculi.140

systemic hypervolemia, abdominal trauma, or post liver transplantation (Table 4-5).148 A distinguishing feature between periportal edema and biliary dilation is that periportal edema will appear as periportal lucency that surrounds the portal triad on all sides where dilated bile ducts lie only on one side of the portal triad (see Figures 4-72, 4-73, and 4-74).148

Periportal Edema Edema of the portal triads can be due to a variety of conditions, many of which are not related to the biliary system. However, periportal edema can superficially resemble and be confused with biliary ductal dilation, and so we will discuss the disorder here. Periportal edema occurs when there is increased production of hepatic extracellular fluid or decrease in the ability for lymphatics to remove excess fluid. Some common causes include congestive heart failure, hepatitis,

UNIQUE DISORDERS OF THE BILE DUCTS Pneumobilia Pneumobilia is nearly always a result of prior instrumentation of the biliary system that renders the sphincter of Oddi incompetent. This includes sphincterotomy following ERCP and surgeries that result in resection of the sphincter of Oddi, most often a Whipple procedure. Rarely, pneumobilia can be a result of ascending infection, biliary-enteric fistula, emphysematous cholecystitis, and noniatrogenic incompetence of the sphincter of Oddi. Abdominal plain films and chest radiographs will demonstrate pneumobilia as branching tubular hypodensities within the central liver. This is distinguished from portal venous gas that will collect in the periphery of the

Chapter 4 Imaging of the Gallbladder and Biliary System 307

A

B

C

D

Figure 4-70 Choledochocele This 56-year-old man had a history of alcoholic pancreatitis and complained of abdominal pain. A. Image of the duodenum from an upper GI examination shows smoothly marginated intraluminal mass (arrows) in the second portion of the duodenum. B. CT through the central liver shows no evidence

of intrahepatic biliary dilation. C. Image through the gallbladder fossa shows the common bile duct (arrowhead) to be borderline enlarged. D. At the junction of the second with the third portion of the duodenum there is a cystic mass (arrow) projecting into the lumen of the duodenum. This is near the sight of the ampulla of Vater and represented a choledochocele.

308 Diagnostic Abdominal Imaging

A

B

C

Figure 4-71 CT of Caroli Disease This 25-year-old man presented with abdominal pain. Liver function tests were normal. A and B. Enhanced CT images demonstrate moderate multifocal dilation of the central biliary ducts with normal-appearing peripheral ducts. C. The

mid common bile duct (arrow) is markedly dilated but with a normal caliber duct (arrowhead) as it enters the pancreas. The combination of normal liver function tests and normal peripheral and dilated central ducts is consistent with the diagnosis of Caroli disease.

liver (see Figures 4-75 and 4-76). Computed tomographic examinations will readily demonstrate hypoattenuating air within the lumen of the bile ducts, and MRI will show the air as signal voids on all pulse sequences (see Figures 4-75 and 4-77). This will most often collect in the anterior portions of the liver because air will rise into nondependent structures. US will demonstrate linear and branching echogenic structures in the distribution of bile ducts (see Figure 4-75).

Operative and Postoperative Hardware

POSTOPERATIVE FINDINGS RELATED TO BILIARY SURGERY Cholecystectomy, Whipple procedure, and liver transplantation are the most common surgeries to involve the biliary system.

Table 4-5. Causes of Periportal Edema 1. Congestive heart failure 2. Hepatitis 3. Hypervolemia 4. Abdominal trauma 5. Post liver transplantation

The most common surgery performed on the biliary system is cholecystectomy. This procedure can be performed as an open surgery or laparoscopically, depending on the indication. The postsurgical findings are usually similar regardless of how the procedure was performed. Normal postoperative findings after cholecystectomy include a cystic duct remnant, which is usually 1 to 2 cm long; however, remnants of up to 6 cm in length have been reported.77,117 The diameter of the common bile duct can increase after cholecystectomy. A maximum diameter of 13 mm with gradual tapering is within normal range.117,122 Cholecystectomy clips are apparent on plain film, and CT examinations as linear metallic densities in the gallbladder fossa (see Figure 4-78). On US, the clips appear as linear shadowing structures. On MRI, cholecystectomy clips appear as metallic artifacts in the gallbladder fossa. Usually, surgical clips and the absence of the gallbladder are the only findings in patients status post cholecystectomy. Occasionally, there can be mild ductal dilation following the surgery that persists with the common hepatic duct measuring up to 10 to 13 mm in internal diameter (see Figure 4-78).149 The other surgery affecting the biliary system that is commonly encountered is the pancreaticoduodenectomy, also known as the Whipple procedure. The operation includes removal of the pancreatic head, duodenum, gallbladder, and gastric antrum with drainage of the biliary system through a choledochojejunostomy. The Whipple procedure is traditionally thought of as surgery

Chapter 4 Imaging of the Gallbladder and Biliary System 309

A

B

Figure 4-72 Periportal Edema in a Trauma Patient This 18-year-old man was in a motor vehicle accident. A and B. Enhanced CT demonstrates low attenuation regions surrounding the high-attenuation enhancing portal veins. When only on one side of the portal vein, this low attenuation region will usually

indicate bile duct dilation. However, when found surrounding portal veins, this finding is indicative of periportal edema. In the setting of trauma, periportal edema will usually be a manifestation of volume overload common in the acute trauma setting but can be a direct manifestation of liver trauma.

for tumors involving the pancreatic head, but it can also be performed for cholangiocarcinoma or duodenal masses requiring resection. Postsurgical changes from the choledochojejunostomy include a loop of small bowel in the gallbladder fossa usually with some degree of pneumobilia, most commonly affecting the left hepatic lobe. The loop of

small bowel forming the choledochojejunostomy does not fill with oral contrast and can appear as a low attenuation masslike opacity in the porta hepatis (see Figures 4-75 and 4-79). This should not be confused with recurrent tumor or a postoperative collection. Mild bile duct dilation can imply a stricture at the choledochojejunostomy but is often

A

B

Figure 4-73 Periportal Edema Due to Chemotoxicity This 30-year-old man with acute lymphoid leukemia (ALL) had elevated liver function tests. A and B. T2-weighted MRI sequence images show the low signal in the portal veins

surrounded by high-signal edema in the portal triads, findings typical of periportal edema. This was an imaging manifestation of hepatotoxicity because of his chemotherapy for ALL.

310

Diagnostic Abdominal Imaging Figure 4-74 Periportal Edema Following Liver Transplantation This 54-year-old man had received a liver transplant. A and B. Unenhanced CT shows the portal veins to be surrounded by lower attenuation periportal edema. This is probably due to disruption of the lymphatic drainage during liver transplantation and is usually a clinically irrelevant finding.

A

B

Figure 4-75 Pneumobilia Following Whipple Procedure This 64-year-old man had recently undergone a Whipple procedure. A. Abdominal plain film also shows the pneumobilia (arrows) as branching hypodensity in the right upper quadrant. There are also small-bowel air-fluid levels due to an adynamic ileus (arrowheads). B and C. Transverse US images in the same patient demonstrate indistinct branching and linear regions of echogenicity (black arrowheads)—the characteristic appearance of pneumobilia on US examinations. Note the large echogenic region in the porta hepatis (black arrow). This is air within the choledochojejunostomy.

A

B

C

Chapter 4 Imaging of the Gallbladder and Biliary System 311

D

E

F

G

Figure 4-75 Pneumobilia Following Whipple Procedure (Continued) D and E. Contrast-enhanced CT images through the liver shows lines and small dots of gas in the central liver (large arrows) typical of pneumobilia. F and G. There is an amorphous mass within the porta hepatis (white arrowheads). There are small

hyperattenuating foci representing surgical clips and enteric staple lines (small arrows). This is the typical appearance of a choledochojejunostomy. These loops of small bowel lack oral contrast and can be confused with recurrent tumor.

312

Diagnostic Abdominal Imaging present without elevation of liver function tests, in which case it is of no significance (see Figure 4-78).

Complications of Biliary Surgery

Figure 4-76 Portal Venous Gas Associated with Necrotizing Enterocolitis This 8-day-old premature infant had increased residuals and abdominal distention. Anteroposterior view of the abdomen demonstrates multiple linear lucencies (arrowheads) paralleling the bowel loops, indicating pneumatosis intestinalis and typical of necrotizing enterocolitis in premature infants. There are also multiple branching lucencies overlying the liver shadow (arrow) typical of portal venous gas. Note how the more central portal veins are not gas filled. Biliary air typically involves the central ducts, whereas portal venous gas involves the more distal venules.

There are many complications that can occur as a result of both laparoscopic and open cholecystectomy. Complications of laparoscopy in general include abdominal wall bleeding, abdominal vessel injury, GI perforation, solid visceral injury, and wound infection. General complications of cholecystectomy include bile duct injury, bile leakage, and infection. It is important for radiologists to be aware of the radiographic manifestations of these potential complications.150 The complications from laparoscopic procedures are usually readily apparent on cross-sectional imaging. Abdominal wall injury and bleeding will present as hematoma at the trochar site. GI perforation will present with increasing abdominal pain and with pneumoperitoneum that progressively increases over time. Occasionally the site of bowel injury will be visualized directly. A solid visceral injury will appear as a low attenuation region within the affected organ. Wound infections will most often appear as excessive infiltration of the subcutaneous fat, sometimes with rim-enhancing fluid collections in the anterior abdominal wall. Complications from cholecystectomy can be more difficult to evaluate. An injury to the bile duct can be due to accidental clipping of the common hepatic duct and usually presents with jaundice, nausea, and vomiting. Usually, CT, MRI, MRCP, and ERCP examinations will demonstrate bile duct dilation and site of obstruction from an aberrant surgical clip.

A

B

Figure 4-77 Pneumobilia on MRI This 64-year-old man had recently undergone a Whipple procedure. A. T1-weighted and (B) T2-weighted MRI sequences

both show areas of signal void (arrows) within the regions of the portal triads. These are within the central bile ducts and are typical of pneumobilia on MRI examinations.

Chapter 4 Imaging of the Gallbladder and Biliary System 313

A

B

C Figure 4-78 Cholecystectomy A and B. Contrast-enhanced CT shows hyperattenuating surgical clips (arrowhead) in the gallbladder fossa and mild common bile duct dilation (arrow) of 10 mm. These are typical

findings following cholecystectomy. There is also a dissection of the abdominal aorta. C. Abdominal plain film in a different patient shows the typical location of surgical clips after cholecystectomy.

314

A

Diagnostic Abdominal Imaging

B

C

Figure 4-79 Whipple Procedure This 64-year-old man had undergone a Whipple procedure for carcinoma of the pancreas. A. There is pneumobilia (black arrowhead) and mild intrahepatic bile duct dilation (white arrowheads). Liver function tests were normal. B and

C. Within the porta hepatis is an amorphous mixed fluid and soft-tissue attenuation mass (arrows). This represents the choledochojejunostomy and should not be confused with recurrent tumor.

Bile leaks from biliary surgery are also occasionally encountered. Cross-sectional imaging will demonstrate with fluid collections near the site of the bile leak. Hepatobiliary studies will reveal leakage of radiotracer into a nonanatomic location. This will often be the gallbladder fossa and can be confused with the normal gallbladder if the reader is not aware of the previous surgery (see Figure 4-80). A loculated fluid collection from a bile

leak, a biloma, can be drained percutaneously. Lastly, postoperative abscesses can be seen following cholecystectomy, Whipple procedures, and other biliary surgery. The radiologic findings are identical to abscesses seen from other etiologies and include a rim-enhancing fluid collection with surrounding inflammatory changes usually containing foci of gas. In most cases, the abscess will be present within the surgical field.

Figure 4-80 Biloma This patient had recently undergone cholecystectomy. A fluid collection was noted in the operative bed on a postoperative

CT (not shown). Tc-99m HIDA examination demonstrates an amorphous collection of tracer in the gallbladder fossa, indicating the presence of a bile leak. (Courtesy of Jacob Dubroff, MD, PhD.)

Chapter 4 Imaging of the Gallbladder and Biliary System 315 REFERENCES 1. Parulekar SG. Ultrasound evaluation of common bile duct size. Radiology. 1979;133:703-707. 2. Catalano OA, Sahani DV, Kalva SP, et al. MR imaging of the gallbladder: a pictorial essay. Radiographics. 2008;28:135-155. 3. Grand D, Horton KM, Fishman EK. CT of the gallbladder: spectrum of disease. AJR Am J Roentgenol. 2004;183:163-170. 4. Rooholamini SA, Tehrani NS, Razavi MK, et al. Imaging of gallbladder carcinoma. Radiographics. 1994;14:291-306. 5. Klein JB, Finck FM. Primary carcinoma of the gallbladder: review of 28 cases. Arch Surg. 1972;104:769-777. 6. Hamrick RE Jr, Liner FJ, Hastings PR, et al. Primary carcinoma of the gallbladder. Ann Surg. 1982;195:270-273. 7. Roberts JW, Daugherty SF. Primary carcinoma of the gallbladder. Surg Clin North Am. 1986;66:743-749. 8. Yum HY, Fink AL. Sonographic findings in primary carcinoma of the gallbladder. Radiology. 1980;134:693-696. 9. Kane RA, Jacobs R, Katz J, Costello P. Porcelain gallbladder: ultrasound and CT appearance. Radiology. 1984;152:137-141. 10. Robbins SL. Pathology. 3rd ed. Philadelphia, PA: Saunders; 1987:957-959. 11. Vaittinen E. Carcinoma of the gallbladder: a study of 390 cases diagnosed in Finland 1953-1967. Ann Chir Gynaecol Fenn. 1970; 59(suppl 168):1-81. 12. Adson MA. Carcinoma of the gallbladder. Surg Clin North Am. 1973;53:1203-1216. 13. Bergdahl L. Gallbladder carcinoma first diagnosed at microscopic examination of gallbladders removed for presumed benign disease. Ann Surg. 1980;191:19-22. 14. Fahim RB, McDonald JR, Richards JC, et al. Carcinoma of the gallbladder: a study of its modes of spread. Ann Surg. 1962; 156:114-124. 15. Levy AD, Murakata LA, Rohrmann CA Jr. Gallbladder carcinoma: radiologic-pathologic correlation. Radiographics. 2001;21:295-314. 16. Sako M, Ohtsuki S, Hitora S, et al. Diagnostic imaging of thickening of the gallbladder wall: angiographic approach to differentiation between cancer and chronic cholecystitis. Rinsho Hoshasen (Japan J Clin Radiol). 1985;30:697-704. 17. Lane J, Buck JL, Zeman RK. Primary carcinoma of the gallbladder: a pictorial essay. Radiographics. 1989;9:209-227. 18. Dalla Palma L, Rizzatto G, Pozzi-Mucelli RS, et al. Grayscale ultrasonography in the evaluation of carcinoma of the gallbladder. Br J Radiol. 1980;53:662-667.

24. Siegelman ES. Body MR. 1st ed. Philadelphia, PA: ElsevierSaunders ;2001:63-119. 25. Ohtani T, Shirai Y, Tsukada K, et al. Carcinoma of the gallbladder: CT evaluation of lymphatic spread. Radiology. 1993;189:875-880. 26. Sons HU, Borchard F, Joel BS. Carcinoma of the gallbladder: autopsy findings in 287 cases and review of the literature. J Surg Oncol. 1985;28:199-206. 27. Bickel A, Eitan A, Tsilman B, et al. Low-grade B cell lymphoma of mucosa-associated lymphoid tissue (MALT) arising in the gallbladder. Hepatogastroenterology. 1999;46:1643-1646. 28. Chim CS, Liang R, Loong F, et al. Primary mucosa-associated lymphoid tissue lymphoma of the gallbladder. Am J Med. 2002;112:505-507. 29. Pantongrag-Brown L, Nelson AM, Brown AE, et al. Gastrointestinal manifestations of acquired immunodeficiency syndrome: radiologic-pathologic correlation. Radiographics. 1995;15:1155-1178. 30. Donnelly LF, Bisset GS III, Frush DP. Embryonal rhabdomyosarcoma of the biliary tree. Radiology. 1998;208: 621-623. 31. Zielinski MD, Atwell TD, Davis PW, et al. Comparison of surgically resected polypoid lesions of the gallbladder to their pre-operative ultrasound characteristics. J Gastrointest Surg. 2009;13:19-25. 32. Okamoto M, Okamoto H, Kitahara F, et al. Ultrasonographic evidence of association of polyps and stones with gallbladder cancer. Am J Gastroenterol. 1999;94(2):446-450. 33. Meyers RP, Shaffer EA, Bech PL. Gallbladder polyps: epidemiology, natural history, and management. Can J Gastroenterol. 2002;16:187-194. 34. Choi JH, Yun JW, Yong-Sung K, et al. Pre-operative predictive factors for gallbladder cholesterol polyps using conventional diagnostic imaging. World J Gastroenterol. 2008;14(44):6831-6834. 35. Sermon A, Himpens J, Leman G. Symptomatic adenomyomatosis of the gallbladder—report of a case. Acta Chir Belg. 2003;103:225-229. 36. Martel JA, McLean CA, Rankin RN. Best cases from the AFIP: melanoma of the gallbladder. Radiographics. 2009;29:291-296. 37. Middleton WD, Kurtz AB, Hertzberg BS. Ultrasound: the requisites. 2nd ed. St Louis, MO: Mosby;2004:28-46, 87-101. 38. Terzi C, Sokmen S, Seckin S, et al. Polypoid lesions of the gallbladder: report of 100 cases with special reference to operative indications. Surgery. 2000;127:622-627.

19. Jeffrey RB, Laing FC, Wong W, et al. Gangrenous cholecystitis: diagnosis by ultrasound. Radiology. 1983;148:219-221.

39. Mainprize K, Gould S, Gilbert J. Surgical management of polypoid lesions of the gallbladder. Br J Surg. 2000;87: 414-417.

20. Ruiz R, Teyssou H, Fernandez N, et al. Ultrasonic diagnosis of primary carcinoma of the gallbladder: a review of 16 cases. J Clin Ultrasound. 1980;8:489-495.

40. Sugiyama M, Atomi Y, Kuroda A, et al. Large cholesterol polyps of the gallbladder: diagnosis by means of US and endoscopic US. Radiology. 1995;169:493-497.

21. Franquet T, Montes M, Ruiz de Azua Y, et al. Primary gallbladder carcinoma: imaging findings in 50 patients with pathologic correlation. Gastrointest Radiol. 1991;16:143-148.

41. Furukawa H, Kosuge T, Shimada K, et al. Small polypoid lesions of the gallbladder: differential diagnosis and surgical indications by helical computed tomography. Arch Surg. 1998;133:735-739.

22. Weiner SN, Koenigsberg M, Morehouse H, et al. Sonography and computed tomography in the diagnosis of carcinoma of the gallbladder. AJR Am J Roentgenol. 1984;142:735-739.

42. Kozuka S, Tsubone M, Yasui A, et al. Relation of adenoma to carcinoma in the gallbladder. Cancer. 1982;50:2226-2234.

23. Smathers RL, Lee JK, Heiken JP. Differentiation of complicated cholecystitis from gallbladder carcinoma by computed tomography. AJR Am J Roentgenol. 1984;143:255-259.

43. Van Breda Vriesman AC, Engelbrecht MR, Smithuis RHM, et al. Diffuse gallbladder wall thickening: differential diagnosis. AJR Am J Roentgenol. 2007;188:495-501.

316

Diagnostic Abdominal Imaging

44. Rumack CM, Wilson SR, Charboneau JW. Diagnostic ultrasound. 2nd ed. St Louis, MO: Mosby; 1998:175-200. 45. Zissin R, Osadchy A, Shapiro M, et al. CT of a thickened-wall gallbladder. Br J Radiol. 2003;76:137-143. 46. Altun E, Semelka RC, Elias J Jr, et al. Acute cholecystitis: MR findings and differentiation from chronic cholecystitis. Radiology. 2007;244:174-183.

65. Berk RN, van der Vegt JH, Lichtenstein LE. The hyperplastic cholecystoses: cholesterolosis and adenomyomatosis. Radiology. 1983;146:593-601. 66. McCarty WC. Pathology of the gallbladder and some associated lesions: a study of specimens from 365 cholecystectomies. Ann Surg. 1910;51:651-669.

47. Ko CW, Lee SP. Biliary sludge and cholecystitis. Best Pract Res Clin Gastroenterol. 2003;17:383-396.

67. Feldman M, Feldman M Jr. Cholesterolosis of the gallbladder: an autopsy study of 165 cases. Gastroenterology. 1954;27: 641-648.

48. Ralls PW, Colletti PM, Halls JM, et al. Prospective evaluation of 99m-Tc-IDA cholescintigraphy and gray-scale ultrasound in the diagnosis of acute cholecystitis. Radiology. 1982;144: 369-371.

68. Hoefsloot FAM. Histologic investigations of cholecystosis In: Hulst SGT, Ruijs JHJ, eds. Symposium on functional and acalculous anomalies of the gallbladder: a multidisciplinary approach. Amsterdam: Excerpta Medica; 1979:59-66.

49. Ziessman HA, O’Malley JP, Thrall JH. Nuclear medicine: the requisites. 3rd ed. Philadelphia, PA: Elsevier-Mosby ; 2006: 159-214.

69. Holzbach RT, Marsh M, Tang P. Cholesterolosis: physicalchemical characteristics of human and diet-induced canine lesions. Exp Mol Pathol. 1977;27:324-328.

50. Jeffrey RB, Laing FC, Wong W, et al. Gangrenous cholecystitis: diagnosis by ultrasound. Radiology. 1983;148:219-221.

70. Tilvis RS, Aro J, Strandberg TE, et al. Lipid composition of bile and gallbladder mucosa in patients with acalculous cholesterolosis. Gastroenterology. 1982;82:607-615.

51. Bortoff GA, Chen MY, Ou DJ, et al. Gallbladder stones: imaging and intervention. Radiographics. 2000;20:751-766. 52. Kaftori JK, Pery M, Green J, et al. Thickness of the gallbladder wall in patients with hypoalbuminemia: a sonographic study of patients on peritoneal dialysis. AJR Am J Roentgenol. 1987;148:117-118. 53. Yamada K, Yamada H. Gallbladder wall thickening in mononucleosis syndromes. J Clin Ultrasound. 2001;29: 322-325. 54. Cerny E, Husek K, Jelinkova I, et al. Validity of diagnostic criteria of chronic cholecystitis. Scripta Medica. 2000;73: 283-288. 55. Kim PN, Lee SH, Gong GY, et al. Xanthogranulomatous cholecystitis: radiologic findings with histologic correlation that focuses on intramural nodules. AJR Am J Roentgenol. 1999;172:949-953. 56. Itai Y, Araki T, Yoshikawa, et al. Computed tomography of gallbladder carcinoma. Radiology. 1980;137:713-718. 57. Reyes CV, Jablokow VR, Reid R. Xanthogranulomatous cholecystitis: report of seven cases. Am Surg. 1981;47:322-325. 58. Dao AH, Wong SW, Adkins RB Jr. Xanthogranulomatous cholecystitis: a clinical and pathologic study of twelve cases. Am Surg Pathol. 1981;5:653-659. 59. Parra JA, Acinas O, Bueno J, et al. Xanthogranulomatous cholecystitis: clinical, sonographic, and CT findings in 26 patients. AJR Am J Roentgenol. 2000;174:979-983.

71. Levy AD, Murakat LA, Abbott RM, et al. Benign tumors and tumorlike lesions of the gallbladder and extrahepatic bile ducts: radiologic-pathologic correlation. Radiographics. 2002;22: 387-413. 72. Bickel A, Eitan A, Tsilman B, et al. Low-grade B cell lymphoma of mucosa-associated lymphoid tissue (MALT) arising in the gallbladder. Hepatogastroenterology. 1999;46:1643-1646. 73. Chin CS, Liang R, Loong F, et al. Primary mucosa-associated lymphoid tissue lymphoma of the gallbladder. Am J Med. 2002;112:505-507. 74. Bolondi L, Gaiani S, Testa S, Labò G. Gall bladder sludge formation during prolonged fasting after gastrointestinal tract surgery. Gut. 1985;26(7):734-738. 75. Stewart L, Smith AL, Pelligirini CA, et al. Pigment gallstones form as a composite of bacterial microcolonies and pigment solids. Ann Surg. 1987;206(3):242-249. 76. Vitellas KM, Keogan MT, Spritzer CE, et al. MR cholangiopancreatography of bile and pancreatic duct abnormalities with emphasis on the single-shot fast spin-echo technique. Radiographics. 2000;20:939-957. 77. Turner MA, Fulcher AS. The cystic duct: normal anatomy and disease processes. Radiographics. 2001;21:3-22. 78. Htoo MM. Surgical implications of stone impaction in the gallbladder neck with compression of the common hepatic duct (Mirizzi syndrome). Clin Radiol. 1983;34:651-655.

60. Lichtman JB, Varma VA. Ultrasound demonstration of xanthogranulomatous cholecystitis. J Clin Ultrasound. 1987;15:342-345.

79. Cruz FO, Barriga P, Tocornali J, et al. Radiology of the Mirizzi syndrome: diagnostic importance of the percutaneous transhepatic cholangiogram. Gastrointest Radiol. 1983; 8:249-253.

61. Casas D, Perez-Andres R, Jimenez JA, et al. Xanthogranulomatous cholecystitis: a radiological study of 12 cases and review of the literature. Abdom Imaging. 1996;21:456-460.

80. Han JK, Choi BI, Kim AY, et al. Cholangiocarcinoma: pictorial essay of CT and cholangiographic findings. Radiographics. 2002;22:173-187.

62. Kim PN, Ha HK, Kim YH, et al. US findings of xanthogranulomatous cholecystitis. Clin Radiol. 1998;53: 290-292. 63. Chun KA, Ha HK, Yu ES, et al. Xanthogranulomatous cholecystitis: CT features with emphasis on differentiation from gallbladder carcinoma. Radiology. 1997;203:93-97. 64. Furuta A, Ishibashi T, Takahashi S, et al. MR imaging of xanthogranulomatous cholecystitis. Radiat Med. 1996;14:315-319.

81. Chung YE, Kim MJ, Park YN, et al. Varying appearances of cholangiocarcinoma: radiologic-pathologic correlation. Radiographics. 2009;29:683-700. 82. Thorsen MK, Quiroz F, Lawson TL, et al. Primary biliary carcinoma: CT evaluation. Radiology. 1984;152:479-483. 83. Wernecke K, Henke L, Vassallo P, et al. Pathologic explanation for hypoechoic halo seen on sonograms of malignant liver tumors: an in vitro correlative study. AJR Am J Roentgenol. 1992;159(5):1011-1016.

Chapter 4 Imaging of the Gallbladder and Biliary System 317 84. Wibulpolprasert B, Dhiensiri T. Peripheral cholangiocarcinoma: sonographic evaluation. J Clin Ultrasound. 1992;20(5): 303-314.

104. Guyton DP, Schreiber H. Intestinal polyposis and periampullary carcinoma: changing concepts. J Surg Oncol. 1985;29:158-159.

85. Lim JH. Cholangiocarcinoma: morphologic classification according to growth pattern and imaging findings. AJR Am J Roentgenol. 2003;181(3):819-827.

105. Berk T, Friedman LS, Goldstein SD, et al. Relapsing acute pancreatitis as the presenting manifestation of an ampullary neoplasm in a patient with familial polyposis coli. Am J Gastroenterol. 1985;80:627-629.

86. Ros PR, Buck JL, Goodman ZD, et al. Intrahepatic cholangiocarcinoma: radiologic-pathologic correlation. Radiology. 1988;167(3):689-693. 87. Choi BI, Lee JH, Han MC, et al. Hilar cholangiocarcinoma: comparative study with sonography and CT. Radiology. 1989; 172(3):689-692. 88. Vilgrain V, Van Beers BE, Flejou JF, et al. Intrahepatic cholangiocarcinoma: MRI and pathologic correlation in 14 patients. J Comput Assist Tomogr. 1997;21(1):59-65. 89. Valls C, Guma A, Puig I, et al. Intrahepatic peripheral cholangiocarcinoma: CT evaluation. Abdom Imaging. 2000; 25(5):490-496. 90. Asayama Y, Yoshimitsu K, Irie H, et al. Delayed phase dynamic CT enhancement as a prognostic factor for mass-forming intrahepatic cholangiocarcinoma. Radiology. 2006;238(1):150-155. 91. Mittelstaed CA. Ultrasound of the bile ducts. Semin Roentgenol. 1997;32(3):161-171.

106. Jagelman DG, DeCosse JJ, Bussey HJ. Upper gastrointestinal cancer in familial adenomatous polyposis. Lancet. 1988;1(8595):1149-1151. 107. Nannery WM, Barone JG, Abouchedid C. Familial polyposis coli and Gardner’s syndrome. N J Med. 1990; 87:731-733. 108. Cohen SB. Familial polyposis coli and its extracolonic manifestations. J Med Genet. 1982;19:193-203. 109. Ono C, Iwama T, Mishima Y. A case of familial adenomatous polyposis complicated by thyroid carcinoma, carcinoma of the ampulla of Vater, and adrenocortical adenoma. Jpn J Surg. 1991;21:234-240. 110. Ponchon T, Berger F, Chavaillon A, et al. Contribution of endoscopy to diagnosis and treatment of tumors of the ampulla of Vater. Cancer. 1989;64:161-167.

92. Robledo R, Muro A, Prieto ML. Extrahepatic bile duct carcinoma: US characteristics and accuracy in demonstration of tumors. Radiology. 1996;198(3):869-873.

111. Nakao NL, Siegel JH, Stenger RJ, et al. Tumors of the ampulla of Vater: early diagnosis by intraampullary biopsy during endoscopic cannulation—two case presentations and a review of literature. Gastroenterology. 1982;83:459-464.

93. Park HS, Lee JM, Kim SH, et al. CT differentiation of cholangiocarcinoma from periductal fibrosis in patients with hepatolithiasis. AJR Am J Roentgenol. 2006;187(2):445-453.

112. Zenman RK, Burrell MI. Gallbladder and bile duct imaging: a clinical radiologic approach. New York: Churchill Livingstone;1987:575.

94. Lim JH, Park CK. Pathology of cholangiocarcinoma. Abdom Imaging. 2004;29(5):540-547.

113. Pandolfo I, Scribano E, Blandino A, et al. Tumors of the ampulla diagnosed by CT hypotonic duodenography. J Comput Assist Tomogr. 1990;14:199-200.

95. Han JK, Lee JM. Intrahepatic intraductal cholangiocarcinoma. Abdom Imaging. 2004;29(5):558-564. 96. Buck JL, Elsayed AM. From the archives of the AFIP: ampullary tumors: radiologic-pathologic correlation. Radiographics. 1993;13:193-212. 97. Michelassi F, Erroi F, Dawson PJ, et al. Experience with 647 consecutive tumors of the duodenum, ampulla, head of the pancreas, and distal common bile duct. Ann Surg. 1989; 210:544-556. 98. Yamaguchi K, Enjoji M, Ysuneyoshi M. Pancreatoduodenal carcinoma: a clinicopathologic study of 304 patients and immunohistochemical observation for CEA and CA 19-9. J Surg Oncol. 1991;47:148-154. 99. Perzin KH, Bridge MF. Adenomas of the small intestine: a clinicopathologic review of 51 cases and a study of their relationship to carcinoma. Cancer. 1981;48:799-819. 100. Celik C, Venditti JA Jr, Satchidan S, et al. Villous tumors of the duodenum and ampulla of Vater. J Surg Oncol. 1986;33: 268-272. 101. Fenoglio-Preiser CM, Pascal RR, Perzin KH. Tumors of the intestines. In: Atlas of tumor pathology, series 2, fasc 27. Washington, DC: Armed Forces Institute of Pathology;1990:173.

114. Smelka RC, Kelekis NL, Gesine J, et al. Ampullary carcinoma: demonstration by current MR techniques. J Magn Reson Imaging. 1997;7:153-156. 115. Baron RL, Stanley RJ, Lee JKT, et al. A prospective comparison of the evaluation of biliary obstruction using computed tomography and ultrasonography. Radiology. 1982;145:91-98. 116. Takehara Y. Fast MR imaging for evaluating the pancreaticobiliary system. Eur J Radiol. 1999;29:211-232. 117. Hoeffel C, Azizi L, Lewin M, et al. Normal and pathologic features of the postoperative biliary tract at 3D MR cholangiopancreatography and MR imaging. Radiographics. 2006;26:1603-1620. 118. Pavone P, Laghi A, Catalano C, et al. MR cholangiography in the examination of patients with biliary-enteric anastomoses. AJR Am J Roentgenol. 1997;169:807-811. 119. Bowie JD. What is the upper limit of normal for the common bile duct on ultrasound: how much do you want it to be? Am J Gastroenterol. 2000;95:897-900. 120. Williams HJ Jr, Bender CE, May GR. Benign postoperative biliary strictures: dilation with fluoroscopic guidance. Radiology. 1987;163:629-634.

102. Yamaguchi K, Enjoji M, Kitamura K. Non-icteric ampullary carcinoma with a favorable prognosis. Am J Gastroenterol. 1990;85:994-999.

121. House MG, Cameron JL, Schulick RD, et al. Incidence and outcome of biliary strictures after pancreaticoduodenectomy. Ann Surg. 2006;243:571-578.

103. Wise RH Jr, Stanley RJ. Case report: carcinoma of the ampulla of Vater presenting as acute pancreatitis. J Comput Assist Tomogr. 1984;8:158-161.

122. Feng B, Song Q. Does the common bile duct dilate after cholecystectomy? Sonographic evaluation in 234 patients. AJR Am J Roentgen. 1995;165:859-861.

318

Diagnostic Abdominal Imaging

123. Brugge WR, Saleemuddin A, Pande H, Nikoomanesh P. Bile duct strictures. http://emedicine.medscape.com/ article/186850-overview.

138. Yoon KH, Ha HK, Kim MH, et al. Biliary stricture caused by blunt abdominal trauma: clinical and radiologic features in five patients. Radiology. 1998;207:737-741.

124. Vitellas KM, Keogan MT, Freed KS, et al. Radiologic manifestations of sclerosing cholangitis with emphasis on MR cholangiopancreatography. Radiographics. 2008;20:959-975.

139. Kim MJ, Mitchell DG, Ito K, et al. Biliary dilatation: differentiation of benign from malignant causes— value of adding conventional MR imaging to MR cholangiopancreatography. Radiology. 2000;214:173-181.

125. Ueno Y, LaRusso NF. Primary sclerosing cholangitis. J Gastroenterol. 1994;29:531-543. 126. Lindor KD, Wiesner RH, MacCarty RL, et al. Advances in primary sclerosing cholangitis. Am J Med. 1990;89:73-80.

140. Brancatelli G, Federle MP, Vilgrain V, et al. Fibropolycystic liver disease: CT and MR imaging findings. Radiographics. 2005;25:659-670.

127. Balthazar EJ, Birnbaum BA, Naidich M. Acute cholangitis: CT evaluation. J Comput Assist Tomogr. 1993;17:283-289.

141. Kim OH, Chung HJ, Choi BG. Imaging of the choledochal cyst. Radiographics. 1995;15:69-88.

128. Bader TR, Braga L, Beavers KL, et al. MR imaging findings of infectious cholangitis. Magn Reson Imaging. 2001;19:781-788.

142. Choi BI, Yeon KM, Kim SH. Caroli disease: central dot sign in CT. Radiology. 1990;174:161-163.

129. Urban BA, Fishman EK. Tailored helical CT evaluation of acute abdomen. Radiographics. 2000;20:725-749.

143. Caroli J, Sonpault R, Kossakowski J, et al. La dilatation polydystique congenital des voies biliaires intra-hepatiques: essai de classification. Semin Hop Paris. 1958;34:128-135.

130. Seel D, Park Y. Oriental infestational cholangitis. Am J Surg. 1983;146:366-370. 131. Lim J. Oriental cholangiohepatitis: pathologic, clinical, and radiologic features. AJR Am J Roentgenol. 1991;157:1-8. 132. Kim MJ, Cha SW, Mitchell DG, et al. MR imaging findings in recurrent pyogenic cholangitis. AJR Am J Roentgenol. 1999;173:1545-1549. 133. Park MS, Yu JS, Kim KW, et al. Recurrent pyogenic cholangitis: comparison between MR cholangiography and direct MR cholangiography. Radiology. 2001;220:677-682. 134. Dolmatch BL, Laing FC, Federle MP, et al. AIDS-related cholangitis: radiographic findings in nine patients. Radiology. 1987;163:313-316. 135. Kavin H, Jonas RB, Chowdhury L, et al. Acalculous cholecystitis and cytomegalovirus infection in the acquired immunodeficiency syndrome. Ann Intern Med. 1986;104:53-54. 136. Vauthey JN, Loyer E, Chokshi P, et al. Case 57: eosinophilic cholangiopathy. Radiology. 2003;227:107-112. 137. Campbell W, Kirk G, Clements WB. Radiation-induced stricture of the common bile duct. Internet J Surg. 2008;17(2).

144. Mall JC, Ghahremani GG, Boyer JL. Caroli’s disease associated with congenital hepatic fibrosis and renal tubular ectasia. Gastroenterology. 1974;66:1029-1035. 145. Mujahed Z, Glenn F, Evans JA. Communicating cavernous ectasia of the intrahepatic ducts (Caroli’s disease). AJR Am J Roentgenol. 1971;113:21-26. 146. Katyal D, Lees GM. Choledochal cysts: a retrospective review of 28 patients and a review of the literature. Can J Surg. 1992; 35:584-588. 147. Sty JR, Sullivan P, Wagner R, et al. Hepatic scintigraphy in Caroli’s disease. Radiology. 1978;124:732. 148. Ros PR, Mortele KJ. CT and MRI of the abdomen and pelvis: a teaching file. 2nd ed. Philadelphia, PA: Lippincott ;2006:35. 149. Graham MF, Cooperberg PL, Cohen MM, et al. The size of the normal common hepatic duct following cholecystectomy: an ultrasonographic study. Radiology. 1980;135:137-139. 150. Wright TB, Bertino RB, Bishop AF, et al. Complications of laparoscopic cholecystectomy and their interventional radiologic management. Radiographics. 1993;13:119-128.

CHAPTER

5

Imaging of the Pancreas Edward R. Oliver, MD, PhD Wallace T. Miller Jr., MD

I. ANATOMY OF THE PANCREAS II. IMAGING OF THE PANCREAS III. DISEASES OF THE PANCREAS a. Solid Masses of the Pancreas i. Neoplasms of the pancreas ii. Focal pancreatitis iii. Pancreatic trauma iv. Focal fatty sparing and focal fatty infiltration b. Cystic Masses of the Pancreas i. Cystic pancreatic neoplasms ii. Simple unilocular cysts iii. Epithelial cysts iv. Complications of pancreatitis c. Disruption of the Gland: Pancreatic Trauma

ANATOMY OF THE PANCREAS The pancreas is an accessory digestive gland with both exocrine and endocrine functions. Approximately 80% of the gland is exocrine in function and composed of ductal and acinar cells whereas 2% of the pancreas is composed of the endocrine islets cells of Langerhans.1 The remainder of the gland is composed of stromal tissue. During development, the pancreas derives from 2 separate anlage: the dorsal and ventral pancreatic buds. The dorsal anlage gives rise to the pancreatic neck, body, and tail, whereas the ventral anlage forms the pancreatic head and uncinate process.2 Each pancreatic bud develops its own draining duct and at approximately 7 weeks of gestation, the 2 anlage rotate and fuse thereby giving the characteristic comma-shaped configuration. The adult pancreas is a long thin organ that measures 15 to 25 cm in length.3 The mean anteroposterior dimensions of the head, body, and tail are greatest during the third decade, when they measure approximately 2.9 cm, 1.9 cm, and 1.8 cm, respectively.4 There is progressive atrophy of the gland throughout adulthood, with the head, body, and tail measuring approximately 2.1 cm, 1.4 cm, and 1.3 cm during the eighth decade of life.4 The pancreas lies within

d. Diffuse Processes of the Pancreas i. Diffuse pancreatic enlargement ii. Pancreatic atrophy iii. Fatty infiltration of the pancreas iv. Iron deposition in the pancreas: hemochromatosis v. Diffuse calcifications: chronic pancreatitis vi. Diffuse pancreatic ductal dilation e. Congenital Disorders of the Pancreas i. Pancreas divisum ii. Annular pancreas IV. POSTOPERATIVE FINDINGS RELATED TO PANCREATIC SURGERY a. Whipple Procedure b. Distal Pancreatectomy c. Pancreas Transplant

the retroperitoneum at the level of L1-2 and is divided into 5 portions: the uncinate process, head, neck, body, and tail.3 The pancreatic head is bordered laterally on the right and inferiorly by the second and third portions of the duodenum. The uncinate process extends from the head and lies posterior to the superior mesenteric vein. The neck of the pancreas lies anterior to the confluence of the splenic vein and superior mesenteric vein and the body and tail extend into the left upper quadrant anterior to the splenic vein, with the pancreatic tail located in the splenic hilum. The main pancreatic duct, the duct of Wirsung, originates within the dorsal pancreatic anlage and is commonly referred to as the dorsal duct. This duct traverses the length of the pancreas and receives the distal common bile duct within the pancreatic head in proximity to the greater duodenal papilla. Approximately 20 to 35 secondary pancreatic duct branches drain into the main duct along its length and the majority of the pancreatic secretions empty into the duodenum through the main duct and greater duodenal papilla. The accessory pancreatic duct, the duct of Santorini, develops within the anterosuperior pancreatic head in what was originally the ventral pancreatic bud. Accessory pancreatic duct anatomy is variable. In some cases, the accessory duct empties into the minor duodenal 319

320

Diagnostic Abdominal Imaging

papilla whereas in others it partially regresses or empties into the main duct. One congenital variant of ductal anatomy, pancreas divisum, is discussed in greater detail in a later section.

IMAGING OF THE PANCREAS Computed tomography (CT), magnetic resonance imaging (MRI), and ultrasonography (US) are the 3 common imaging modalities by which the pancreas is evaluated. Imaging of the normal gland by these modalities is reviewed below. CT is the most common modality by which the pancreas is evaluated, owing largely to its noninvasive nature and speed. The use of thin multidetector CT scanners has further improved evaluation, allowing for greater resolution and for multiplanar imaging. Most enhanced CT protocols of the abdomen and pelvis administer a positive oral contrast agent (barium or water-soluble contrast) and an iodinated intravenous contrast and view images at a nominal slice thickness of 5 mm during the portal venous phase of enhancement, approximately 70 seconds after the bolus of contrast is given. However, the protocol employed at our institution in the case of suspected pancreatic neoplasm calls for the use of a negative oral contrast agent (eg, water) and multiple scans. Positive oral contrast is not used as it commonly obscures the pancreatic head and duodenal papilla. A noncontrast scan is performed of the abdomen at a nominal slice thickness of 5 mm first to localize the level of the pancreas. Intravenous contrast is administered and the pancreas is scanned at a nominal slice thickness of 3 mm during the late arterial (pancreatic) phase of enhancement (40 seconds following contrast bolus). Finally, the entire abdomen is scanned at a slice thickness of 5 mm during the portal venous phase of enhancement (70 seconds after initiation of the contrast bolus). On unenhanced CT, pancreatic parenchyma is homogenous and has an attenuation close to that of spleen and muscle but less than that of liver.3 Following the administration of intravenous contrast, the pancreas demonstrates avid uniform enhancement owing to its high vascularity (see Figure 5-1). In younger individuals, the normal pancreas should demonstrate a smooth contour with the body and tail narrower than the head. With aging, fat deposition leads to an lobulated contour; however, the body and tail should remain narrower than the head as in younger patients (see Figure 5-2).4 The main pancreatic duct may be identified as a low-attenuation linear structure running the length of the gland (see Figure 5-1). The normal duct diameter is 2 to 3 mm; however, failure to identify the main duct on CT is not uncommon and of no clinical significance. MRI evaluation of the pancreas is best performed on a high field strength magnet (≥1.0 Tesla), which provides a high signal-to-noise ratio and allows for fast breath-hold imaging.5 Increased fat-water frequency shift is also desired as it affords adequate chemical fat suppression.5 Axial T1-weighted gradient echo breath-hold sequences with and

without fat suppression are effective sequences with which to evaluate the pancreas and peripancreatic tissues. Fat suppression provides the greatest contrast between normal and abnormal pancreatic tissue, whereas nonsuppressed sequences allow for the evaluation of extension of disease into the extrapancreatic tissues. Enhanced imaging is performed using 2- or 3-dimensional fat-suppressed gradient echo imaging with scans performed during the capillary and interstitial phases of enhancement, which occur 15 seconds and 45 seconds after contrast arrives in the abdominal aorta.6 Fast spin echo T2-weighted sequences are effective in the evaluation of islet cell tumors, peripancreatic fluid collections, cystic lesions, and liver metastases.5 Although T2-weighted sequences are also useful in evaluating the ducts, heavily T2-weighted sequences (MR cholangiopancreatography [MRCP]) are particularly powerful in demonstrating the ductal anatomy and pathology.6 The normal pancreas is hyperintense to liver and muscle on noncontrast T1-weighted images and slightly hyperintense to muscle on T2-weighted sequences (see Figure 5-3).8,9 In older individual, pancreatic parenchyma may demonstrate decreased T1 signal relative to liver, a finding that may be represent age-related fibrosis.5 Fat suppression increases the difference between the signal intensity of the pancreas and surrounding fat on T1-weighted images but has little effect in T2-weighted sequences.7 Nevertheless, fat-suppressed T2-weighted sequences are particularly helpful in demonstrating liver metastases and islet cell tumors.5 Immediately after the administration of gadolinium contrast, the pancreas demonstrates avid homogenous enhancement, with the pancreatic signal intensity greater than that of liver and adjacent fat (see Figure 5-3).5,8 The pancreatic signal intensity is similar to fat approximately 1 minute after contrast administration and less than fat 2 minutes after contrast administration.5 T2-weighted images and MRCP readily demonstrate the main pancreatic duct as a linear high-signal structure traversing the pancreatic gland. The normal duct measures 2 to 3 mm and side branch ducts are not identified unless enlarged (see Figure 5-4).6 US is an inexpensive and fast way to evaluate the pancreas. Disadvantages include operator dependence and frequent nonvisualization of the entire pancreas either secondary to large body habitus or due to obscuration of the gland by overlying bowel gas. Sonography is best performed in thin patients, during the fasting state and with a high frequency (5-8 MHz) transducer. At US, the pancreas usually exhibits a homogenous appearance with an echotexture that is isoechoic or hyperechoic to liver (see Figure 5-5).1 Fatty replacement of the pancreas is commonplace with aging and obesity and results in increased echogenicity of the pancreatic parenchyma. In cases of fatty replacement, the echogenicity of the gland may approach that of the adjacent retroperitoneal fat.1,11 The normal pancreatic duct appears as a linear hypoechoic structure extending through the pancreas (see Figure 5-6). The normal pancreatic duct diameter has

Chapter 5 Imaging of the Pancreas 321

A

B

C

D

E

F

Figure 5-1 Normal Pancreas This 41-year-old woman presented with elevated biliary enzymes. A triple phase protocol was performed. In all 3 series, the pancreas has a gently lobulated contour with the gland tapering along its length. Unenhanced images (A and B) demonstrate the pancreas (H, head; B, body; T, tail) to have an attenuation similar to spleen and muscle but an attenuation slightly less than liver parenchyma. During the arterial phase of enhancement (C and D), the pancreas demonstrates avid uniform enhancement, with an attenuation slightly greater than liver and muscle. The relationship of the pancreas to the superior mesenteric

artery (white arrowhead) and superior mesenteric vein(s) is well appreciated. The gastroduodenal artery (white arrow in C) is noted passing anterior to the pancreatic head. E and F. Delayed imaging during the portal venous enhancement shows uniform enhancement with attenuation slightly less than that of liver. The main pancreatic duct is partially visualized in cross section (black arrowhead). The position of the pancreas anterior to the splenic vein (black arrows) is partially seen on the provided arterial and venous phase images. Additional annotations: sb, loop of small bowel; d, duodenum; s, superior mesenteric vein; ∗, common hepatic artery.

322

Diagnostic Abdominal Imaging

A

B

C

Figure 5-2 Age-Related Pancreatic Atrophy A-C. CT angiogram of the abdomen and pelvis was performed in this 85-year-old woman for evaluation of an infrarenal aortic

aneurysm. Delayed imaging demonstrates marked atrophy of the pancreas (white arrowheads).

been reported to be between 2 and 2.5 mm;12,13 however, measurements slightly greater than these (∼3 mm) may be considered normal if the duct walls are smooth and parallel and the duct tapers peripherally.1 As with MRI, side branch ducts are not visible unless dilated. Sonographic evaluation of the pancreas is ideally performed during the fasting state as overlying bowel gas often obscures the pancreas. In cases where bowel gas limits evaluation of the pancreas, placing the patient in a right lateral decubitus position may improve visualization as overlying gas moves to the nondependent position.

section. Metastases and lymphoma are other neoplasms that can rarely appear as a solid pancreatic mass (Table 5-1).

DISEASES OF THE PANCREAS Pancreatitis and its sequelae, chronic pancreatitis, pancreatic neoplasms, and pancreatic trauma are the primary disorders of the pancreas. Imaging findings of pancreatic disorders include solid masses, cystic masses, disruption of the gland, ductal dilation, diffuse enlargement, atrophy, calcifications, and fatty infiltration.

Solid Masses of the Pancreas Focal masses of the pancreas will usually represent pancreatic neoplasms. However, rarely pancreatic contusions and acute pancreatitis will appear as a solid pancreatic mass.

Neoplasms of the pancreas There are 4 groups of tumors that account for the majority of primary malignancies of the pancreas: adenocarcinoma, islet cell tumors, solid-pseudopapillary tumor, and a group of tumors called the cystic pancreatic malignancies. The cystic pancreatic malignancies will usually appear as a cystic mass and will be discussed in detail in a subsequent

Adenocarcinoma: Pancreatic ductal adenocarcinoma is the most common malignancy of the exocrine pancreas, accounting for approximately 95% of pancreatic malignancies,8 and is also the most common mass of the pancreas. The American Cancer Society estimates that approximately 37,680 Americans will be diagnosed with pancreatic cancer in 2008. An estimated 34,290 Americans will die from the disease during the same period. Certain risk factors are associated with pancreatic cancer and include age >45 years, cigarette use, obesity, diabetes mellitus, and chronic pancreatitis. The prognosis for pancreatic adenocarcinoma is poor. Approximately 80% of cases are inoperable at the time of diagnosis9 and most of these patients will die within 6 months. The traditional 5-year survival rate following surgical resection has been reported as approximately 5%;16,17 however, survival rates vary in multiple series, with one series reporting a 5-year survival rate of 0.2%10 and one series reporting a postresection survival rate of 10%.11 Although differences in institutional experience may account for the variable survival rate, it has been suggested that other pancreatic neoplasms (ie, nonpancreatic ductal adenocarcinoma) and even benign conditions may be incorrectly diagnosed as pancreatic adenocarcinoma and thereby artificially inflate the postsurgical pancreatic adenocarcinoma survival rates in some series.10 Pancreatic adenocarcinomas are rarely symptomatic at early stages and therefore are usually large in size and at an advanced stage when they present. Jaundice, usually secondary to obstruction of the common bile duct, is the most common symptom at presentation and occurs in more than 90% of patients (see Figure 4-57).12 Other common symptoms include abdominal pain, weight loss, nausea, and anorexia.12 Rarely, patients will present with waxing

Chapter 5 Imaging of the Pancreas 323

A

B

C

D

E

F

Figure 5-3 MRI of Normal pancreas A. Abdominal pain was the presenting complaint in this 25-yearold woman. On T2-weighted images (HASTE sequence, TR 700 ms, TE 104 ms), the pancreas is slightly hyperintense to muscle. In this image, the common bile duct is identified as a hyperintense focus in the pancreatic head and a portion of the main pancreatic duct is seen as a thin hyperintense line in the pancreatic tail. B. On fat-suppressed T2-weighted images (FSE, TR 6680 ms, TE 98 ms), the pancreas is only slightly more hyperintense to surrounding fat-suppressed peripancreatic fat.

C. On nonenhanced T1-weighted images (TR 142 ms, TE 4.8 ms), the pancreas is slightly hyperintense to liver and hyperintense to muscle. D. On nonenhanced fat-suppressed T1-weighted gradient echo images (TR 3.6 ms, TE 1.8 ms), the pancreas is hyperintense to liver and muscle and well distinguished from the peripancreatic fat. E. Following the administration of gadolinium contrast, the pancreas demonstrates avid enhancement in the arterial phase and is hyperintense to liver (TR 3.6 ms, TE 1.8 ms). F. During the venous phase of enhancement, the pancreas is slightly hyperintense to liver (TR 3.6 ms, TE 1.8 ms).

324

Diagnostic Abdominal Imaging

Figure 5-4 Normal MRCP An MR/MRCP was ordered for suspected biliary obstruction. Coronal maximum-intensity projection (MIP) of highresolution heavily T2-weighted sequence demonstrates a normal pancreatic duct (arrow) and the common hepatic duct (arrowhead). Normal pancreatic and biliary ductal anatomy is present. A small simple hepatic cyst in the caudate lobe is identified as a small focus of intense signal superiorly. The high-resolution source images should always be reviewed as small filling defects (eg, stones or neoplasms) may not be well demonstrated on MIP or thick slab sections.

and waning thromboses.9 Tumors of the pancreatic head may come to clinical attention earlier than those within the body and tail secondary to symptomatic involvement of the bile duct and ampulla.12 Jaundice is not usually seen in tumors of the pancreatic tail unless they have metastasized. Cross-sectional imaging demonstrates an indistinctly marginated mass centered within a portion of the pancreatic parenchyma that can range from a few centimeters to as much as 10 cm in diameter. Approximately 60% of adenocarcinomas arise in the pancreatic head, whereas 15% and 5% develop in the body and tail, respectively.21,22 Diffuse pancreatic involvement occurs in roughly 20% of cases. Pancreatic adenocarcinomas are hypovascular and desmoplastic tumors. On unenhanced CT, the tumor is usually isoattenuating to normal pancreatic parenchyma and, therefore, inconspicuous unless it significantly alters the normal pancreatic contour. Contrast administration greatly improves tumor detection, with the tumor’s hypovascular and desmoplastic characteristics usually resulting in a hypoattenuating appearance relative to normal pancreas (see Figures 4-56 and 5-7).23,24 Tumor conspicuity is greater during the pancreatic phase of enhancement, with an average attenuation difference between tumor and pancreas of 67 Hounsfield units (HUs) compared to 39 HU during the portal venous phase of enhancement.13 Multidetector CT with 3-dimensional reconstructions has improved the overall diagnostic accuracy of tumor resectability, with reported accuracies of 73% to 95%.25-28 The

A

B

Figure 5-5 US of Normal Pancreas This 33-year-old woman was being evaluated for possible gallbladder polyp. A and B. The pancreas head, body, and tail are well visualized (asterisks). In this patient, the pancreas is hyperechoic relative to the liver (LL, left hepatic lobe) and spleen

(Spl). The pancreas lies anterior to the splenic vein (SV) and mesenteric vessels (arrowhead, superior mesenteric artery). IVC, inferior vena cava; Ao, Aorta; PS, portal splenic confluence. Arrows in (B) denote the course of the left renal vein.

Chapter 5 Imaging of the Pancreas 325

Figure 5-6 Normal Main Pancreatic Duct This 61-year-old man was being evaluated for possible biliary obstruction. The main pancreatic duct is identified as a thin anechoic line passing through the pancreatic body in this image (small arrowheads). The duct diameter was normal at 2 mm. LL, left hepatic lobe; Ao, aorta; SV, splenic vein. Large arrowhead denotes the superior mesenteric artery.

use of these techniques also provides important information regarding the vascular anatomy and their relationship to pancreatic lesions. Occasionally, pancreatic adenocarcinoma may be isoattenuating relative to normal pancreatic parenchyma and inconspicuous. In one small study, as many as 11% of pancreatic adenocarcinomas could not be differentiated from normal pancreas on the basis of attenuation.14 In each of these cases, ancillary findings of underlying malignancy were detected, allowing for correct diagnosis. These include

Table 5-1. Solid Pancreatic Masses A. Neoplasms 1. Adenocarcinoma 2. Islet cell tumors a. Insulinoma b. Glucagonoma c. Gastrinoma d. Somatostatinoma e. VIPoma f. Carcinoid tumor 3. Metastasis 4. Lymphoma B. Focal Pancreatitis C. Pancreatic Contusion

dilation of the distal main pancreatic duct with an abrupt caliber change at the tumor, the so-called interrupted duct sign; atrophy of the pancreatic parenchyma distal to the tumor; and contour abnormality of the pancreas (see Figure 5-8). Although none of these secondary signs is specific to malignancy, adenocarcinoma should be considered if any of these findings are encountered. MR represents another modality commonly employed for the evaluation of adenocarcinoma of the pancreas. Tumors are characteristically hypointense on fat-suppressed T1-weighted, non–fat suppressed T1-weighted, and postgadolinium T1-weighted MR sequences.30,31 The tumor may be uniform in character or variably necrotic. Regions of cystic necrosis may appear as areas of high signal on T2-weighted MR images (see Figures 5-9 and 5-10). MRCP is a powerful MR application that is at least as specific and possibly slightly more sensitive than endoscopic retrograde pancreatography in diagnosing pancreatic adenocarcinoma.32 In addition, MRCP is particularly helpful in the evaluation of cases where tumors obstruct the pancreatic duct and/or biliary duct. Furthermore, small inconspicuous tumors can manifest as subtle contour abnormalities of the pancreatic duct on MRCP, thereby aiding in the detection of early tumors. Transabdominal US is often the initial imaging modality utilized in cases of suspected biliary tract pathology and can detect some pancreatic adenocarcinomas. On US, pancreatic cancer is typically hypoechoic relative to normal parenchyma (see Figure 5-9). Detection of these lesions, however, is often limited to large lesions and those arising in the pancreatic head. As US is less sensitive in detecting pancreatic cancer, CT or MRI is recommended for evaluation of suspected malignancy and staging. Surgical resection is the only cure for pancreatic adenocarcinoma and tumor resectability depends on the absence of locally advanced and metastatic disease. Therefore, it is

Imaging Notes 5-1. Imaging Features of Pancreatic Adenocarcinoma Direct Signs Indistinctly marginated mass CT: Hypoattenuating 89%—Isoattenuating 11% MRI: Hypointense on T1 (fat suppressed and nonsuppressed) Hypointense on post gad T1 Variable on T2 US: Hypoechoic Focal widening or enlargement of the pancreas Indirect Signs Abrupt termination of dilated pancreatic duct (interrupted duct sign) Atrophy of the distal pancreas

326

Diagnostic Abdominal Imaging

important for the reader to evaluate each imaging examination for features that would contraindicate surgical intervention. Direct extension of tumor that involves the local vasculature such as the celiac axis, hepatic artery, superior mesenteric artery and vein, and portal vein renders the tumor nonresectable. Imaging findings that indicate vascular involvement include tumoral obliteration of the perivascular fat plane, contact of greater than 180 degrees between the tumor and vessel, and narrowing or distortion of the vessel (see Figures 5-10, 5-11, and 5-12).27,33 In the absence of these findings, the presence of isolated dilated peripancreatic veins is highly suggestive of occult vascular

involvement.34-36 Metastatic disease, including peritoneal implants and liver metastases, also precludes curative resection. Pancreatic adenocarcinoma is staged according to the TNM system, and the American Joint Committee on Cancer recently released the seventh edition of its cancer staging manual.15 No changes were made to the TNM staging system for exocrine neoplasms of the pancreas, which already had been validated with respect to predicting survival (see Table 5-2).16 Helical CT has been shown to be the most accurate modality in the TNM system of staging pancreatic adenocarcinoma.39 T classification is determined by tumor size and extent. T0 refers to the absence of evidence of primary tumor. Tis represents carcinoma in situ; however, very few pancreatic adenocarcinomas are discovered at this very early stage. T1 and T2 tumors correspond to those confined to the pancreas, with size being the discriminating factor: tumors less than or equal to 2 cm in greatest dimension are considered T1 and tumors greater than 2 cm are considered T2. In T3 disease, the tumor extends beyond the pancreas into the adjacent peripancreatic tissues, including the duodenum, stomach, biliary system, adrenal glands, spleen and perirenal fat; however, there is no involvement of the celiac axis or superior mesenteric artery (see Figure 5-13). T4 disease is unresectable and corresponds to involvement of adjacent vascular structures, specifically the celiac axis and superior mesenteric artery. TX is reserved for those instances when the main tumor cannot be assessed. N classification is determined by the extent of nodal disease. The absence of regional (peripancreatic) lymph node involvement is classified as N0, whereas involvement

A

B

Figure 5-7 Pancreatic Head Adenocarcinoma with Obstructive Pancreatic Atrophy A and B. Contrast-enhanced CT images show a heterogeneously hypoenhancing mass in the pancreatic head (black arrowhead). There is dilation of the peripheral main pancreatic duct and obstructive atrophy of the pancreas (white arrowheads). The

pancreatic head adenocarcinoma contacts but does not encase the superior mesenteric vein (asterisk). Low attenuation lesions in the liver (black arrows) represent liver metastases. The high attenuation focus in the pancreatic head represents an internal biliary stent.

Imaging Notes 5-2. Findings Indicating Surgical Unresectability of Pancreatic Carcinoma 1. Vascular invasion (celiac axis, hepatic artery, SMA, SMV, portal vein) a. Interruption of perivascular fat plane b. >180 degrees of contact between vessel and cancer c. Narrowing or distortion of vessel d. Dilation of peripancreatic veins (suggestive) 2. Distant metastasis a. Liver b. Peritoneal c. Other

Chapter 5 Imaging of the Pancreas 327

A

B

Figure 5-8 Pancreatic Adenocarcinoma With Hepatic Metastases A CT angiogram of the chest was performed to evaluate a suspected aortic stent graft endoleak in this 77-year-old woman. A and B. Extended cuts through the upper abdomen revealed a very subtle mass (white arrowheads) in the pancreatic tail, which is isoattenuating to minimally hypoattenuating to normal pancreas. Parenchymal atrophy with main pancreatic duct dilation is present distal to the mass (arrows in A and B).

Also present are multiple hypoattenuating hepatic lesions. An enlarged celiac axis lymph node is present (black arrowhead in A). The imaging findings were highly suspicious for pancreatic adenocarcinoma with hepatic metastases and regional nodal metastasis. Hepatic metastasis was confirmed through percutaneous biopsy of one of the hepatic lesions. A celiac axis vascular stent is noted in A.

of regional lymph nodes is classified as N1. In cases where regional lymph node involvement cannot be assessed, the N category is classified as NX. Nodal involvement directly relates to overall prognosis; however, lymph node diameter is a poor indicator for nodal metastasis and has a low accuracy in establishing the N category. As such, pathologic sampling is always performed in cases of resectable pancreatic cancer.40-43 Despite this limitation, any abnormalities of either size or number of regional lymph nodes should be reported because imaging identification of lymph nodes suspicious for metastasis provides surgical guidance to nodes that should be sampled and maximizes lymph node yield (see Figure 5-8).17 Finally, the presence or absence of distant metastases determines the M category and corresponds to the spread of disease to distant lymph nodes (ie, nonregional lymph nodes) and other organs, such as liver and lung. M0 disease corresponds to the absence of metastatic disease whereas M1 disease corresponds to the presence of distant metastatic disease. MX is reserved for cases in which distant metastatic disease cannot be assessed. Although helical CT was shown to be the most accurate imaging modality for the detection of metastatic disease,39 peritoneal implants and small hepatic metastases are frequently too small to detect. Nevertheless, careful evaluation for hepatic metastases and peritoneal deposits should be performed because recognition of their presence results in upstaging of disease and can result in the avoidance of unnecessary surgery and its associated morbidity and mortality.

Islet cell tumors: Neuroendocrine neoplasms constitute a small subset of pancreatic neoplasms and include insulinomas, gastrinomas, glucagonomas, somatostatinomas, VIPomas, and serotonin-secreting (ie, carcinoid) tumors. Although these tumors are commonly referred to as “islet cell tumors,” these neoplasms arise from neuroendocrine cells that are of a ductal and non–islet cell origin.18 These may be sporadic in etiology but are also associated with a number of hereditary conditions, including multiple endocrine neoplasia type I (MEN1), neurofibromatosis type I,  tuberous sclerosis, and von Hippel-Lindau syndrome.46,47 Neuroendocrine tumors are clinically classified into 2 broad categories, syndromic and nonsyndromic, based on whether or not a clinical syndrome results from

Imaging Notes 5-3. Neuroendocrine tumors of the pancreas most often occur sporadically. However, occasionally their occurrence is as part of one of several hereditary syndromes including: 1. Multiple endocrine neoplasia type I (MEN1) 2. Neurofibromatosis type I 3. Tuberous sclerosis 4. von Hippel-Lindau syndrome

328

Diagnostic Abdominal Imaging

A

B

C

D

E

F

Figure 5-9 Pancreatic Adenocarcinoma causing Biliary and Pancreatic Duct Obstruction A 54-year-old woman with adenocarcinoma of the pancreatic head. A and B. Contrast-enhanced fat-suppressed T1-weighted image reveals a mass (white arrowhead in A) within the medial pancreatic head that is hypointense relative to normal pancreatic parenchyma in the lateral pancreatic head (A). Main pancreatic duct dilation in (B) is manifested by a beaded low-signal structure running through the pancreatic body. C. Fat-suppressed

T2-weighted image demonstrates decreased signal intensity of the mass (arrowhead) relative to the remainder of the pancreatic head. D. Fat suppressed T2-weighted image through the pancreatic body reveals an enlarged main pancreatic duct. Marked extrahepatic and intrahepatic biliary dilation is also partially visualized. E and F. Axial and coronal T2-weighted image reveals marked extrahepatic and intrahepatic biliary dilation. Pancreatic duct dilation is partially visualized (arrows).

Chapter 5 Imaging of the Pancreas 329

G

H

Figure 5-9 Pancreatic Adenocarcinoma causing Biliary and Pancreatic Duct Obstruction (Continued) G. Sagittal US of the abdomen at the level of the common bile duct shows dilation of the duct to 10 mm (green arrow),

which terminates at the obstructing hypoechoic pancreatic adenocarcinoma (arrows). H. Transverse US of the epigastrium shows the pancreatic body (arrowheads) with a dilated pancreatic duct (arrow).

the production of hormones. These terms are preferable to the more commonly used designations hyperfunctioning and nonfunctioning tumors as all neuroendocrine tumors produce some amount of hormones.19 Approximately 85% of neuroendocrine tumors are syndromic and named based on the dominant hormone produced.20 These patients typically present with symptoms of

endocrine disturbance such as hypoglycemia with an insulinoma, recurrent and multiple gastric ulcers with a gastrinoma (Zollinger-Ellison syndrome), or new-onset diabetes mellitus with a glucagonoma. VIPomas are named for the secretion of vasoactive intestinal polypeptide (VIP), which results in the secretion of electrolytes and fluid and a characteristic profuse watery diarrhea. Somatostatinomas are

A

B

Figure 5-10 Pancreatic Adenocarcinoma With Vascular Encasement This 60-year-old man was diagnosed with pancreatic adenocarcinoma of the pancreatic head/neck. A and B. Postcontrast fat-suppressed T1-weighted MR images demonstrate an ill-defined hypointense mass in the pancreatic head/neck (asterisk). A small amount of normal-enhancing pancreatic

head parenchyma is seen in B. There is marked narrowing of the superior mesenteric vein (A) with only a slitlike appearance of the vessel (white arrowhead). The superior mesenteric vein (B) peripheral to the mass is slightly dilated secondary to the narrowing (white arrowhead). This patient was not a surgical candidate secondary to the vascular encasement.

330

A

Diagnostic Abdominal Imaging

B

C

Figure 5-11 Pancreatic Adenocarcinoma With Vascular Encasement This 54-year-old man was diagnosed with pancreatic adenocarcinoma of the uncinate process. A-C. CT images obtained during the arterial phase of enhancement reveal a mass within the uncinate process (asterisk). There is circumferential encasement of the junction of the splenic and portal veins (B) with marked narrowing of the involved vessel (white arrowhead).

There is soft-tissue density (C) surrounding the proximal common hepatic artery (black arrowhead), consistent with additional vascular encasement (compare the attenuation of the perivascular tissue to that of uninvolved mesenteric fat). Partially visualized is dilation of the peripheral main pancreatic duct (white arrow). The patient is not a surgical candidate given the degree of vascular encasement.

A

B

Figure 5-12 Pancreatic Adenocarcinoma with Vascular Encasement This 49-year-old woman was diagnosed with pancreatic adenocarcinoma of the uncinate process. A. Arterial-phase CT image demonstrates a hypoenhancing uncinate process mass involving the superior mesenteric artery (white arrowhead) from the 5-11 o’clock position. The superior mesenteric vein(s) is dilated and the mass abuts the vein from the 4-8 o’clock position (90 degree encasement). B. A more cephalad section from

the same CT study reveals marked narrowing of the superior mesenteric vein (black arrowhead) by the hypoattenuating mass, which explains the dilation of vein more peripherally. The superior mesenteric artery is denoted by the white arrowhead in (B). The main pancreatic duct is partially visualized in the pancreatic neck and dilated secondary to obstruction by the mass. A biliary stent is noted in the pancreatic head. The patient is not a surgical candidate given the degree of vascular encasement.

Chapter 5 Imaging of the Pancreas 331 Table 5-2. Staging Classification of Pancreatic Adenocarcinoma and Pancreatic Neuroendocrine Neoplasms TX

Main tumor cannot be assessed

T0

No findings of primary tumor

Tis

Carcinoma in situ

T1

Confined to the pancreas, ≤2cm

T2

Confined to the pancreas, >2 cm

T3

Involvement of peripancreatic tissuesa

T4

Direct involvement of local vesselsb

NX

Regional lymph node involvement cannot be assessed

N0

No regional (peripancreatic) nodal involvement

N1

Regional (peripancreatic) nodal involvement

MX

Distant lymph node and organ metastasis cannot be assessed

M0

No distant lymph node spread or organ metastasis

M1

Distant lymph node spread or organ metastasis

Stage 0

Tis, N0, M0

Stage IA

T1, N0, M0

Stage IB

T2, N0, M0

Stage IIA

T3, N0, M0

Stage IIB

T1-3, N1, M0

Stage III

T4, N0-1, M0

Stage IV

T1-4, N0-1, M1

a

Duodenum, stomach, CBD, adrenal glands, spleen, perirenal fat Hepatic artery, portal vein, SMA, SMV, splenic artery, splenic vein Data from the AJCC Cancer Staging Handbook, 7th edition.

b

rare tumors that result in a variety of signs and symptoms including diarrhea, diabetes mellitus, cholelithiasis, and achlorhydria or hypochlorhydria. Carcinoid tumors of the pancreas are rare and are composed of enterochromaffin cells. These account for less than 1% of gastrointestinal (GI) carcinoid tumors21 and only rarely produce symptoms relating to hypersecretion of serotonin, such as bronchospasm, diarrhea, and flushing (see Table 5-3).8 Syndromic islet cell neoplasms are characteristically small in size (2 mm

2. Complications of Pancreatitis a. Pseudocyst b. Abscess 3. Trauma a. Hematoma 4. Epithelial cysts associated with congenital disorders a. Von Hippel-Lindau disease b. Autosomal dominant polycystic kidney disease c. Cystic fibrosis

lesions are asymptomatic although when symptomatic they typically present with the same constellation of nonspecific symptoms as seen with serous cystadenomas.51 Unlike serous microcystic adenomas, mucinous neoplasms are considered surgical lesions because they all have a malignant potential.53 Histologically, mucinous cystic neoplasms are composed of mucin-producing columnar cells and resemble ovarian mucinous cystic neoplasms. These lesions are found predominantly in the body and tail of the pancreas110,114 and will usually appear as a unilocular mass or a multiseptated cystic mass with at least 1 cyst greater than 2 cm in diameter.45,110 Because the lesion does not originate from pancreatic ductal epithelium, no communication with the pancreatic duct is present. Unenhanced CT typically reveals a smoothly lobulated cystic mass with low attenuation contents (see Figures 5-29 and 5-30). Unlike serous microcystic adenomas, these lesions are composed of fewer (2 cm) cysts.54 Occasionally, intralesional debris and hemorrhage will be present and the cyst will have an attenuation greater than 20 HU. These lesions typically demonstrate enhancing septa and septal nodules following the administration of intravenous contrast (see Figure 5-30). Associated findings include dilation of the main pancreatic duct and pancreatic atrophy distal to the lesion. Peripheral eggshell calcifications are specific for mucinous cystic neoplasms and highly predictive of malignancy; however, these are not a common finding at CT imaging.114,115 Features suggestive of a malignant mucinous neoplasm (ie, mucinous cystadenocarcinoma) include septations, wall, and/or septal calcifications and a wall thickness greater than 2 mm. (see Figure 5-31).55 The presence of more than 1 of these features increases the likelihood of malignancy and when all 3 features are present, there is a 95% probability that the lesion is malignant.55 Adjacent tissues should be assessed for signs of malignant invasion.

MR is superior to CT in characterizing the complex internal architecture of mucinous cystic neoplasms.51 A lobulated cystic mass is identified with increased T2 signal intensity (see Figure 5-29).52 Variable T1 signal intensity is present depending on the relative amounts of intralesional simple fluid and protein.45,113 Any calcifications will manifest as decreased T1 and T2 signal intensity. Gadolinium administration reveals irregular enhancing septa and nodules.18 Pancreatic atrophy and duct dilation may also be identified distal to the lesion.50 US typically reveals a cystic mass within the parenchyma.18 The cystic contents may be simple and anechoic or may demonstrate echogenic debris and/or hemorrhage. Septa, mural nodules, and calcifications can be identified in some cases;18 however, further evaluation by CT or MRI is usually required to optimally characterize these lesions.

Intraductal papillary mucinous neoplasms: Intraductal papillary mucinous neoplasms are rare neoplasms of the pancreas and account for 1% to 2% of exocrine pancreas tumors and approximately 28% of cystic neoplasms.48,56 They arise from the pancreatic duct epithelium and range from benign IPMN adenomas to malignant IPMN carcinomas.57 Histologically, IPMNs appear as papillary projections of mucin-secreting columnar epithelial cells. The papillomatous overgrowth of mucin-secreting cells is responsible for the hallmark features of this lesion: excess mucin secretion causing dilation of the involved pancreatic ducts. Two distinct forms of IPMNs exist: (1) those that involve the main pancreatic duct and (2) those that involve only side branch pancreatic ducts. The former variant is described under the heading Pancreatic Ductal Dilation whereas the latter is described here. Side branch IPMNs are more commonly observed in men aged 60 to 70 years and are frequently located in the pancreatic head and uncinate process.58 Mucinous distention of a side branch duct results in the formation of a cystic lesion with associated atrophy of the surrounding parenchyma.59 Like other cystic pancreatic neoplasms, many cases will be accidentally discovered by cross-sectional imaging. However, because this tumor involves the pancreatic duct, it can be a cause for recurrent pancreatitis secondary to pancreatic duct obstruction from thick mucinous secretions.56 Side branch IPMNs have a 15% 5-year risk of developing high-grade dysplasia or invasive carcinoma.60

Chapter 5 Imaging of the Pancreas 347

A

B

C

D

E

F

Figure 5-28 Microcystic Serous Cystadenoma A CT study of the chest was performed in this 84-year-old woman with a history of breast carcinoma. A and B. Partially visualized was a large cystic lesion arising from the junction of the pancreatic neck and body (arrowheads) for which MRI was recommended for further evaluation. T2-weighted (C), fat-suppressed T2-weighted (D), T1-weighted (E), and contrast-enhanced fat-suppressed T1-weighted (F) sequences demonstrate a multiloculated lesion consisting of many small (less than 2 cm) cystic spaces. The contents of the lesion follow simple fluid on all sequences and there is no enhancement of the lesion following contrast administration (F). No pancreatic duct dilation is present and thin-section imaging through the lesion did not reveal any communication with the pancreatic duct (not shown). The combined imaging findings are characteristic of a microcystic serous cystadenoma. A 2-mm nonaggressive pancreatic cystic lesion is noted in the pancreatic tail (arrow in D). GB, gallbladder; C, hepatic cyst.

348

Diagnostic Abdominal Imaging

A

B

C

D

Figure 5-29 Mucinous Cystadenoma This 33-year-old woman presented with low back pain. A. An unenhanced CT examination was performed to evaluate for suspected renal calculus. A 3-cm low-attenuation cystic lesion arising from the pancreatic tail was detected. Coronal T2weighted (B), axial T1-weighted (C), and enhanced T1-weighted (D) MR sequences confirm a unilocular cystic lesion arising from the pancreatic tail. No nodular or enhancing components

are identified. Pancreatic pseudocyst would be the most common cause of a cystic pancreatic lesion; however, there was no history of pancreatitis. Differential considerations would also include a mucinous cystic neoplasm, such as mucinous cystadenoma or mucinous cystadenocarcinoma. Distal pancreatectomy was performed and revealed mucinous cystadenoma with ovarian stroma.

Both microcystic and macrocystic patterns of side branch IPMNs exist.119,121 The microcystic pattern of side branch IPMNs typically appear as clusters of small 1- to 2-mm cysts with thin septa, thus mimicking the appearance of a serous microcystic adenoma. Macrocystic side branch IPMNs appear as unilocular or multilocular lesions and are similar in appearance to mucinous cystic neoplasms (see Figure 5-32). Distinguishing these side branch lesions from serous or mucinous neoplasms is often possible with the use of MRCP, which is the modality of choice for the

evaluation of suspected side branch IPMNs as it is most sensitive in defining its morphologic features, such as nodules, septa, and communication with the main pancreatic duct.86,122 Thin-section multidetector CT (with curved reformations), though less sensitive, is another useful modality in evaluating side branch IPMNs.86,122 Management of side branch IPMNs is based on lesion size and the presence of additional associated findings.121,123,124 Lesions that are 3 cm or less in size and without mural nodules or involvement of the main pancreatic duct

Chapter 5 Imaging of the Pancreas 349 Imaging Notes 5-8. Intrapapillary Mucinous Tumors Intraductal papillary mucinous tumors can usually be distinguished from other cystic pancreatic lesions by the demonstration of communication with the pancreatic duct.

may be followed with imaging. Lesions greater than 3 cm in size and/or lesions with suspicious features such as mural nodules, main duct dilation, or interval growth should be considered suspicious for malignancy and resected.

Solid pseudopapillary tumors: Solid pseudopapillary tumors (SPTs), formerly termed solid and pseudopapillary epithelial neoplasms, are rare low-grade malignant neoplasms of the pancreas that usually present in young women, most commonly in the second and third decades of life.14,125 As these lesions occur primarily in young women, some have termed this entity the “daughter” lesion.50 These lesions have often been reported to be most common in the pancreatic tail and in women of African and Asian descent;126-128 however, the veracity of these characteristics have been challenged.61 Solid pseudopapillary tumors may become symptomatic when large, and patients present most commonly with an abdominal mass and/or abdominal discomfort.62 Invasion of adjacent structures and hepatic metastases have been reported but are uncommon,126,129,130 and the presence of metastatic lesions does not seem to preclude cure.63 Even rarer are instances of high-grade malignant lesions.132 Macroscopically, SPTs are well-encapsulated masses that may be solid, mixed cystic and solid, or cystic with a thick wall.126,128 Histologically, these lesions are composed of cells forming solid, pseudopapillary, and/or hemorrhagic pseudocystic structures.131,133 These are not “true cysts” as no true epithelial lining is present.63 The cellular origin of this entity remains unclear.

Imaging Notes 5-9. Age Associations of Cystic Pancreatic Neoplasms Three cystic neoplasms are typically seen in women of different ages.

Figure 5-30 Mucinous cystadenoma This 63-year-old woman had a cystic pancreatic lesion that was progressively enlarging on follow up studies. Enhanced CT obtained during the arterial phase of enhancement demonstrates a 2.5-cm low-attenuation cystic lesion at the junction of the pancreatic neck and body (arrowhead). There are subtle enhancing nodular foci within the lesion that were called prospectively. No dilation of the pancreatic duct is present. The lesion was surgically resected, and pathology revealed mucinous cystadenoma. A Bosniak category IIF cystic lesion in the right kidney is noted.

Moreover, SPTs typically appear as large mixed cystic and solid masses within the pancreatic parenchyma. Evaluation with CT typically shows a well-circumscribed complex lesion of central fluid attenuation and peripheral soft-tissue attenuation.126,134-136 Areas of high attenuation, representing foci of hemorrhage, are also usually seen.126,136 Calcifications may be identified peripherally or centrally.135,136 On MRI, a heterogeneous lesion is revealed that is T1 hyperintense, corresponding to hemorrhage, and most frequently T2 hyperintense.126,128,129 Fluid-fluid levels or fluid-debris levels may be present. In addition, MRI may demonstrate a peripheral rim of low T1 and T2 signal, corresponding to the fibrous capsule and characteristic of an SPT.126,128,129 Following the administration of gadolinium contrast, mild peripheral enhancement is usually identified followed by enhancement of the solid elements on portal venous phase imaging.61 US is not particularly helpful in distinguishing the lesion from other cystic lesions of the pancreas, as these lesions appear as well-circumscribed complex cystic and solid masses (see Figure 5-33).126,134

Neoplasm

Age Range Moniker

Serous cystadenoma

>60

Grandmother tumor

Mucinous cystic neoplasm

∼50

Mother tumor

Simple unilocular cysts

Daughter tumor

With modern thin-section CT and MRI, small, 2 cm

Enhancing nodules occ panc duct dil occ panc atrophy

Side Branch IPMT

Multilocular occ. unilocular

1 mm to >2 cm

Communication with pancreatic duct

Solid pseudopapillary tumor

Unilocular Occ multilocular

Variable

Thick enhancing wall Blood products on MRI Fluid-debris levels Low T1/low T2 fibrous capsule

Necrotic adenocarcinoma

Unilocular

>2 cm

Thick enhancing wall Irregular internal wall

Pseudocyst

Unilocular

Variable

Thin sharply def. wall

Abscess

Unilocular

Variable Usually >2 cm

Thick irregular wall Signs of pancreatitis

of these cysts is often uncertain but in one study where incidentally noted pancreatic cysts underwent resection, 25% were found to represent main duct IPMNs and 23% were found to be side branch IPMNs.137 Of the remaining resected cystic lesions, 18% were mucinous cystic neoplasms, 13% were serous cystadenomas, 5% were SPTs, 4% were neuroendocrine neoplasms, and 2% were ductal adenocarcinomas. The majority of these lesions are clinically irrelevant and will not require intervention. In one study, only 8% of their cases developed changes in the cyst that lead to surgical resection.64 In another study of 90 patients with incidental cysts and a mean follow-up of 48 months, malignancy was demonstrated in only 1 patient, 7 years from diagnosis.65,66 As a consequence of these and other

studies, experts advocate a conservative approach of periodic surveillance with CT or MRCP in cases of incidentally discovered small (40%

shielded from the external magnetic field than water protons and as a result resonate at a slower frequency. T1-weighted chemical shift imaging allows 2 sets of sequences to be acquired, an in-phase sequence and an opposed-phase sequence. In the in-phase sequence, the signal from water and lipid protons is added in each voxel and in the opposed-phase sequence, the signal from lipid protons is subtracted from water protons in each voxel. Tissues with voxels containing both lipid and water have signal loss (ie, appear darker) on opposed-phase images. Using the chemical shift technique, the sensitivity and specificity for differentiating adenomas from metastases ranges from 81% to 100% and 94% to 100%, respectively (see Figures 1-8 and 8-4).29,30,32-36 Approximately 30% of adenomas are relatively lipid-poor by CT examinations, and their attenuation on unenhanced CT is >10 HU and up to 20% of adenomas are relatively lipid poor by MRI examinations using chemical shift techniques. In

A

B

Figure 8-5 Lipid-Poor Adenoma This 62-year-old woman was diagnosed with lung cancer. A. Unenhanced CT image shows a 1.5-cm nodule (arrow) in the left adrenal that measured 23 HU. This is an indeterminate adrenal lesion that could represent either an adenoma or

this situation, additional imaging, specifically a dynamic contrast enhancement CT examination, is necessary for adequate characterization of adrenal adenomas. Post IV contrast administration, adrenal adenomas invariably enhance, sometimes to 80 HU or more. Although the degree of enhancement may not differ from that of other adrenal tumors, adenomas are unique, in that, they display rapid washout of the contrast medium compared to adrenal metastases on delayed, 10- to 15-minute post IV contrast injection CT imaging.37-40 This phenomenon has been used to distinguish between lipid-poor adenomas and adrenal malignancies, especially adrenal metastases. Two measurements can be obtained, the absolute and relative enhancement washout (AEW and REW), depending on whether or not an unenhanced CT series is available. The AEW is calculated based on the formula: AEW = (E – D)/(E – U) × 100, where E stands for attenuation value after enhancement, D stands for the delayed attenuation enhancement of the lesion (10 or 15 minutes), and U for the unenhanced attenuation value. A 60% washout at 15 min results in 88% sensitivity and 96% specificity for the diagnosis of an adenoma.16 The REW, used when unenhanced images are not available, is calculated using the formula: REW = (E – D)/E × 100. A 40% REW results in 96% sensitivity and 100% specificity for the diagnosis of an adenoma.16 Of note, lipid-rich and lipid-poor adenomas display near identical washout measurements.41 These attenuation measurements pre and post IV contrast enhancement, including the washout estimates, are powerful diagnostic imaging tools in making the distinction between 2 common adrenal lesions, the benign adenomas and malignant metastases, very often obviating the need for biopsy (see Figure 8-5).

C metastasis. Images obtained 60 seconds after contrast administration (B) and 15 minutes after contrast administration (C) show the nodule (arrows) to measure 80 HU and 40 HU. This gives a washout of 70%, identifying the nodule as a lipid-poor adenoma.

518 Diagnostic Abdominal Imaging Imaging Notes 8-2. Washout Calculations for Adrenal Adenomas Absolute enhancement washout: AEW = (E – D)/ (E – U) × 100 Relative enhancement washout: REW = (E – D)/ E × 100 E – attenuation value after enhancement D – delayed attenuation enhancement after 10 or 15 min U – unenhanced attenuation value Reproduced, with permission, from Caoili EM, Korobkin M, Francis IR, et al. Adrenal masses: characterization with combined unenhanced and delayed enhanced CT. Radiology. 2002;222:629-633

A very small percentage of pathologically proven adrenal adenomas deviates from the typical homogeneous, oval or round nodule. On occasion, adenomas can be large, irregular in contour, heterogeneous, and can contain hemorrhagic material or calcifications.42 Washout calculations cannot be applied for such lesions. Depending on their size, appearance, and clinical context, percutaneous biopsy or excision is often necessary.

Adrenocortical carcinoma Adrenocortical carcinoma is a rare malignant neoplasm derived from the cortex of the adrenal gland that accounts for approximately 0.2% of all cancer deaths. These tumors are usually large infiltrating masses at presentation and have a very poor prognosis. Only 30% of cases will be confined to the adrenal gland at initial diagnosis.43 The majority of adrenal cortical carcinomas are hormonally active and secrete a variety of substances, including cortisol, androgens, estrogens, or aldosterone, and in approximately 60% of cases, the presentation will be due to endocrine-related syndromes, most commonly Cushing syndrome but also including virilization, adrenogenital syndrome, precocious puberty, feminization, hyperaldosteronism, and Conn (primary aldosteronism) syndrome.44,45 Nonfunctioning tumors are discovered at a later stage, typically with symptoms of abdominal pain, a palpable upper abdominal mass, or with symptoms related to metastases. Adrenocortical carcinoma affects  women more often, though nonfunctioning tumors are more prevalent in men. Although very rare among children, adrenocortical carcinoma can be seen in some cases of the Beckwith-Wiedemann syndrome, which is usually associated with Wilms tumor or hepatoblastoma, or in association with hemihypertrophy (also referred to as hemihyperplasia).46 The TNM staging system for adrenal cortical carcinoma is outlined (Table 8-2). Size is an important imaging feature among adrenal lesions. Most primary masses 6 cm are malignant.47 Nonfunctioning

Table 8-2. TNM Staging System for Adrenal Cortical Carcinoma • Tumor ▪



T1 - Tumor confined to adrenal gland and less than 5 cm T2 - Tumor confined to adrenal gland and greater than 5 cm



T3 - Tumor invasion into periadrenal fat



T4 - Tumor invasion of adjacent organs

• Node ▪

N0 - Negative lymph nodes



N1 - Positive lymph nodes

• Metastases ▪

M0 - No metastases



M1 - Distant metastases

Stage I

T1, N0, M0

Stage II

T2, N0, M0

Stage III

T3, N0, M0 or T1-2, N1, M0

Stage IV

Any T, N, M1, or T3-4, N1, M0

Reproduced, with permission, from Norton JA. Adrenal tumors. In: DeVita VT Jr, Hellman S, Rosenberg SA, eds. Cancer: Principles and Practice of Oncology. 7th ed. Philadelphia, PA: Lippincott Williams & Wilkins;2005:1528-1539.

adrenal cortical carcinoma typically presents as a large suprarenal mass with an average cross-sectional diameter of 12 cm. CT typically shows a large, heterogeneous, suprarenal mass, with areas of central hypoattenuation indicating necrosis, scattered calcifications, and high-attenuation regions indicating hemorrhage (see Figure 8-6).16,48 On MRI, adrenal cortical carcinoma is typically T1-weighted hypointense and T2-weighted hyperintense compared to the liver.16,48-51 Like adrenal adenomas, adrenal cortical carcinoma has increased intracellular lipid content. Chemical shift imaging can be applied to detect the high intracellular lipid content of the tumor by showing signal drop on opposed-phase images compared to in-phase images. Adrenal cortical carcinoma will often invade adjacent tissues, obliterating the tissue planes between the mass and adjacent fat and other organs. As a result of this large size and the propensity to invade adjacent tissues, it can be difficult to distinguish adrenal cortical carcinoma from exophytic renal cell carcinomas, other malignant adrenal mass, such as metastasis or lymphoma, and retroperitoneal sarcomas (see Figure 8-6). Multiplanar, isotropic reconstruction of axial CT images and the direct multiplanar capabilities of US and MRI makes determination of the origin of the tumor easier.49 The tumor is often hypovascular, a feature that may help distinguish adrenal

Chapter 8 Imaging of the Adrenal 519

A

B

C

D

Figure 8-6 Adrenal Cortical Carcinoma This 31-year-old man presented with left testicular swelling from a varicocele. A. US of the left retroperitoneum demonstrates a large heterogenous mass (arrows) superior to the left kidney. B-D. Contrast-enhanced CT confirms the presence of a heterogeneously enhancing left suprarenal mass (arrow in B),

which has invaded the left renal vein (small arrowheads in C) and causes left para-aortic adenopathy (large arrowhead in D). This appearance is most likely to represent an adrenal cortical carcinoma or upper-pole renal cell carcinoma. Subsequent needle biopsy was diagnostic of an adrenal cortical carcinoma.

cortical carcinoma from the normally hypervascular renal cell carcinoma. Adrenal cortical carcinoma also has a propensity for intravascular invasion into the renal vein or inferior vena cava.

Most pheochromocytomas/paragangliomas occur sporadically. However, they have also been associated with a wide variety of syndromes, including (1) von Hippel-Lindau disease (VHL), (2) neurofibromatosis type 1, (3) multiple endocrine neoplasia (MEN 2) syndrome, (4) Carney syndrome or triad (also pulmonary chondromas and gastric stromal tumors), (5) tuberous sclerosis, and (6) SturgeWeber disease. A rare, familial type of pheochromocytoma has also been described as an autosomal dominant trait, often leading to presentation in younger patients. The incidence of pheochromocytoma is approximately 5% to 15% of VHL patients, approximately 1% of neurofibromatosis patients, and approximately 25% of Carney syndrome patients.53 The rule of 10% is an often sited maxim concerning pheochromocytomas such that among sporadic pheochromocytomas, approximately 10% will be bilateral, 10% will occur at ectopic sites, and 10% will be malignant. Patients

Pheochromocytoma Pheochromocytoma is the most common neoplasm of the adrenal medulla. The tumor typically presents in the third and fourth decades of life and affects male and female patients equally. The neoplasm is derived from the chromaffin cells of the adrenal medulla. Histologically identical neoplasms occur at sites of other chromaffin cells, including the sympathetic chain and para-aortic bodies, such as the organ of Zuckerkandl and the carotid body.52 When ectopic, these neoplasms are called paragangliomas. Approximately 98% of these tumors arise in the abdomen and 2% in the neck/thorax.

520 Diagnostic Abdominal Imaging with MEN 2 and the familial variety of pheochromocytoma have an even higher incidence of multiplicity and bilaterality. Some studies have identified ectopia in as many as 25% of sporadic pheochromocytomas,54 most of them in the organ of Zuckerkandl or near the adrenals, however, sites of disease can be wide ranging. In patients with Carney triad, most lesions are functioning, ectopic paragangliomas.53 Paragangliomas are more often malignant and, therefore, more likely to metastasize than pheochromocytomas. Catecholamine secretion is the hallmark of pheochromocytomas/paragangliomas and are estimated to be the cause of high blood pressure in 0.1% to 0.5% of patients with newly diagnosed hypertension.55 Rarely, potentially lethal hypertensive crises can occur as a result of catecholamine secretion by these neoplasms. Hypertensive crisis can be elicited unexpectedly during induction of anesthesia, during pregnancy or surgery, and rarely have been described after intravascular administration of ionic contrast media,56 and can be clinically problematic, because the presence of a pheochromocytoma usually is not suspected or known, and its diagnosis not established until after 1 or several hypertensive crises take place. Hypertensive crisis typically presents with headache, hypertension, and palpitations. Other symptoms include excessive sweating, visual disturbance, tremor, nausea, vomiting, and nonspecific thoracic or abdominal pain. In many instances, the symptoms can be paroxysmal or infrequent. Rarely, patients can present with cerebral hemorrhage or myocardial infarction. A small percentage of pheochromocytomas will be detected accidentally on abdominal CT or MRI performed for unrelated reasons. In the presence of symptoms compatible with a pheochromocytoma, patients should undergo appropriate screening for the disease. The most commonly used tests are assays that detect and estimate levels of free catecholamines (epinephrine and norepinephrine) or their metabolites (metanephrine and vanillylmandelic acid) in plasma or 24-hour urine samples.57,58 Measurements of total urine metanephrine are up to 95% accurate, especially in patients with sustained hypertension. In some cases, a timed urine collection within short periods of time after an attack may be the best test for measuring metanephrine levels. The treatment of choice in a patient with clinical and biochemical evidence of pheochromocytoma and an adrenal mass is adrenalectomy. Patients at risk of developing multiple and bilateral pheochromocytomas (MEN 2 or familial types), who present with a unilateral adrenal mass, should undergo excision of the involved adrenal, and placed on surveillance for other tumors in the adrenals or elsewhere. If bilateral adrenal masses are found, one could consider, if technically feasible, cortex-sparing adrenalectomy to avoid the relatively high risk of acute adrenal insufficiency and chronic corticosteroid replacement therapy.59 Patients with pheochromocytomas need careful followup because of the risk of residual or recurrent tumor and metastatic disease. Patients with nonresectable malignant or metastatic tumors can be conservatively treated for long

periods of time. Bone lesions often respond to radiation and soft-tissue masses can be partially controlled by chemotherapy, including administration of I-131 metaiodobenzylguanidine (MIBG).60,61

Localization and imaging of pheochromocytomas: The most appropriate imaging modalities include MIBG scintigraphy, CT, and MRI. Metaiodobenzylguanidine (MIBG), a derivative of guanethidine, is a very suitable radionuclide agent for the localization of adrenergic tissues. The radioactive element is either I-131 or I-123. Radioactive MIBG is given IV, after the thyroid gland has been blocked with iodides to inhibit glandular uptake of free radioactive iodine and whole body scans are generated. MIBG is an excellent imaging modality, with an accuracy of 90% or better in detecting pheochromocytomas (see Figure 8-7).62 If the primary is

Figure 8-7 Pheochromocytoma Metaiodobenzylguanidine scan performed from the back of this 75-year-old man shows normal activity in the liver and spleen (small arrowheads) and the bladder (large arrowhead). There is also a focal area of increased activity medial to the liver (arrow), typical of a right adrenal pheochromocytoma.

Chapter 8 Imaging of the Adrenal 521 malignant, MIBG can detect sites of metastases. The radionuclide can also be used to deliver therapeutic radiation doses to metastatic lesions throughout the body.60 In general, CT and MRI have been shown to be accurate in the detection of pheochromocytomas and are more widely available and easier to perform than MIBG scans.63-65 On cross-sectional imaging, pheochromocytoma typically appears as a large lesion, 4 to 5 cm in largest dimension mass. In addition, CT will often demonstrate regions of necrosis and/or hemorrhage as low-attenuation or high-attenuation foci, respectively. Calcifications occur uncommonly. Pheochromocytomas typically appear T1 hypointense and T2 hyperintense on MR spin echo imaging. The hyperintensity on T2-weighted images is so intense that it is often compared with a “light bulb” (see Figure 8-8). Like CT, larger tumors will often contain central necrosis and hemorrhage giving the mass an inhomogeneous appearance with the typical features of fluid and blood (see Figure 8-8). Unfortunately, the MRI appearance of a pheochromocytoma is not specific, and there is a considerable overlap, in up to one-third of cases, between a pheochromocytoma and other adrenal neoplasms, such as adrenocortical carcinoma.66 The viable portions of the neoplasm will usually briskly enhance following contrast administration on both CT and MRI examinations (see Figure 8-8). In the past, there had been concern for potential hypertensive crisis following use of intravascular ionic contrast media, and patients have received α- and β-adrenergic receptor blockade prior to IV contrast administration. This practice does not appear to be necessary with the use of nonionic agents, and is now abandoned.56,67

Neuroblastoma, ganglioneuroblastoma, and ganglioneuroma The other common type of tumors arising in the adrenal medulla is the neuroblastoma and its counterparts, the more differentiated and less malignant ganglioneuroblastoma, as well as the benign ganglioneuroma. Neuroblastoma is the most common solid abdominal neoplasm in childhood, ranking behind leukemia, lymphoma, and CNS tumors, but ahead of Wilms tumor or rhabdomyosarcoma. Most tumors arise in patients 2 to 3 years old but have been observed in infants and adults also. Its propensity to infiltrate the bone marrow can make it easy to confuse with leukemia. The majority of tumors secrete catecholamines and their metabolites, as a result of their derivation from neuroendocrine cells similar to those found in a pheochromocytoma.68 Signs and symptoms are dependent on the location of the mass. Most patients with masses that arise in the adrenal present with a palpebral abdominal mass. Other symptoms include abdominal pain or back pain, constipation, fever, weight loss, and other constitutional symptoms.

About one-third of neuroblastomas are found in the adrenal glands. Enlargement of the liver, bone pain, or anemia secondary to bone marrow replacement indicate tumor spread. A widely used, though not universally accepted, staging classification has been proposed by Evans, D’Angio, and Randolph (see Table 8-3).69 An alternative to the above clinical staging system is one proposed by the International Neuroblastoma Risk Group (INRG), a system that stratifies risk prior to treatment, based on the experience of a very large international cohort of patients with the disease. The INRG system analyzed the statistical and clinical significance of 13 potential prognostic factors (eg, stage, age, histology, grade of tumor differentiation, status of the MYCN oncogene, chromosomal components) and created survival curves using event-free survival periods as the primary end point.70 Plain radiographs of the abdomen can show a flank mass or posterior mediastinal mass. Stippled calcifications are present on up to 30% of radiographs. Neuroblastomas are typically detected in utero or during infancy or young childhood.71 Imaging studies have shown that approximately two-thirds occur in an intra-abdominal location, most in the region of the adrenal glands, with 15% to 35% occurring as an intrathoracic paraspinal mass.72,73 Rarely, they can appear at a wide variety of other sites. Fetal sonograms have increasingly identified neuroblastomas. US of infants is also frequently used to evaluate for the presence of a suspected abdominal mass. On US, neuroblastomas typically appear as a cystic, hyperechoic, or mixed cystic and solid mass in a suprarenal location.74,75 However, masses have been identified in multiple locations throughout the fetus, including the thorax and brain. Small tumors are homogeneous and hyperechoic, whereas larger ones are usually heterogeneous with intraparenchymal calcification that appear as small regions of shadowing within the mass.74,75 On CT, neuroblastomas typically appear as large softtissue masses with punctate areas of calcification.76 The mass typically crosses the midline and encases or displaces the major surrounding vascular structures (see Figure 8-9). MRI is also commonly used, because of its lack of radiation, and ability to perform multiplanar whole body imaging for detection of metastases. Masses typically are large, measuring as much as 16 cm in the largest dimension, although lesions as small as 2  cm can be detected and are of decreased signal intensity compared to muscle on T1-weighted images and increased signal intensity on T2-weighted images.72,77 In the majority of cases, signal intensity can be heterogeneous on both T1- and T2-weighted sequences, and in a few cases, homogeneous. Calcifications are seen as punctate, low-signal foci within the mass. Invasion of the neural foramina, vascular encasement, and bone marrow invasion is seen in a significant number of patients.72,77 Further, MIBG scans are an important complementary imaging modality in patients with neuroblastoma,

A

B

C

D

E

F

Figure 8-8 Pheochromocytoma in 2 Patients A-C. This 45-year-old man presented with seizures and systemic hypertension. (A) Coronal T2-weighted MRI image shows a heterogeneous mass (arrow) superior to the right kidney in the expected location of the right adrenal. (B) T2-weighted axial image and (C) T1 postcontrast axial image confirm the heterogeneity of the mass. The anterior portion of the mass (arrowheads) is hyperintense on the T2-weighted image and nonenhancing, indicating a cystic region in the mass, probably because of necrosis. The solid portions of the mass enhance with contrast. These are a feature of an aggressive adrenal neoplasm. Biopsy was diagnostic of a high-grade malignant

pheochromocytoma. D-F. This 48-year-old man was being evaluated for testicular carcinoma. (D) Contrast-enhanced CT shows a uniformly enhancing, 1.5-cm, solid nodule (arrow) in the right adrenal gland. Most likely differential was either an adenoma or metastasis. (E) Opposed-phase T1-weighted MRI sequence shows no signal loss in the nodule (arrow). (F) T2-weighted sequence shows uniform high signal in the nodule (arrow). These features are atypical for both an adrenal adenoma and for adrenal metastasis but are often features of small pheochromocytomas. Adrenal biopsy was diagnostic of a benign pheochromocytoma.

Chapter 8 Imaging of the Adrenal 523 Table 8-3. Evans, D’Angio, and Randolph Staging System for Adrenal Cortical Carcinoma

The adrenal glands are among the most common location for metastatic disease, after the lungs, liver, and bones.79 The primary malignancies most likely to metastasize to the adrenals are lung, breast, melanoma, and kidney, although

many more neoplasms can also metastasize there, such as gastrointestinal primaries (esophagus, stomach, pancreas, and colon), hepatocellular carcinoma, uterine tumors, and sarcomas.80 It has long been known that adrenal metastases are relatively common at autopsy series of patients with various malignancies, most of them microscopic.81,82 In such instances, the adrenals will have a normal appearance on cross-sectional imaging, such as CT, MRI, or US. In many cases, the adrenal lesions are frequently seen simultaneously with metastases to other organs. In such cases, correct identification of the adrenal metastases does not affect staging and treatment. If the adrenal lesion(s) is the only potential and critical metastasis found at a staging CT, additional imaging can be successful in excluding a benign lesion, such as an adenoma, or other adrenal pathology. Percutaneous or laparoscopic biopsy remains an option in selected cases, when necessary.83 In other cases, a gradual increase in the size of a small or subtle adrenal lesion on follow-up CT or MRI over time strongly suggests a metastatic lesion. Likewise, a decrease in size of an adrenal mass following successful chemotherapy of the primary tumor strongly suggests metastasis. With few exceptions, appearances of metastasis are nonspecific, with considerable overlap between primary and metastatic adrenal malignancies. The common benign adrenal masses, adenoma and myelolipoma, can be specifically identified based on their imaging characteristics. Also, metastases are much more common than primary adrenal malignancies. Metastases to the adrenals can be unilateral or bilateral, and can measure from a few millimeters to several centimeters in diameter. With bilateral adrenal involvement, a patient can develop adrenal insufficiency, the symptoms of which may be difficult to distinguish from those of the underlying malignancy. Small metastases typically appear

A

B

Figure 8-9 Neuroblastoma This 2-year-old child presented with an abdominal mass. A and B. Contrast-enhanced CT demonstrates a large inhomogeneously

enhancing midline mass with small punctate areas of calcification, typical of a neuroblastoma. Biopsy confirmed a diagnosis of neuroblastoma.

Stage I

Tumor confined to organ of origin (rare, cured by surgery)

Stage II

Tumor extends beyond organ of origin, does not cross the midline, but may involve ipsilateral lymph nodes (65% 5-year survival)

Stage III

Tumor crosses midline, and may involve contralateral or ipsilateral lymph nodes (30% 5-year survival)

Stage IV

Distant metastases (common, liver, bones, bone marrow, etc 3 cm

occasional

hyperintense T2 MRI 131I-MIBG+

Neuroblastoma

usually >3 cm

rare

punctate calcifications 131I-MIBG+

Myelolipoma

any size

never

macroscopic fat

Hemangioma

usually >3 cm

never

phleboliths punctate calcifications globular enhancement

Lymphoma

usually > 3cm

rare

retroperitoneal mass engulfing adrenal

Chapter 8 Imaging of the Adrenal 527

A

B

C

D

E

F

Figure 8-12 Myelolipoma in 4 Patients These 4 patients were being evaluated for reasons other than adrenal pathology. A-C. Unenhanced CT in each patient demonstrates a mixed solid and fat attenuation mass (arrows) in the expected location of the adrenal glands. Note the variable amount of fat in the lesions. D. US of the patient in (C) shows a heterogeneously echogenic mass (arrows) posterior to the liver

and above the kidney, features typical of an adrenal myelolipoma. E. In phase T1-weighted image and (F) opposed phase T1weighted image in a fourth patient shows a large predominantly fatty mass (arrows) in the left retroperitoneum. This could possibly represent a well differentiated liposarcoma but biopsy showed a myelolipoma.

528 Diagnostic Abdominal Imaging

Adrenal hemangioma Hemangioma of the adrenal is extremely uncommon, with only a few reported cases.103 Similar to hemangiomas elsewhere in the body, the uncommon adrenal hemangioma is a benign, nonfunctioning lesion consisting of dilated, tubular spaces lined by endothelium and filled with blood. The presence of phleboliths in a soft-tissue attenuation adrenal lesion is common and helps in making the diagnosis. Some believe that the incidence of adrenal hemangiomas may be higher than generally reported. It is hypothesized that repeat hemorrhage, thrombosis, and calcifications within hemangiomas result in an appearance similar to that of hemorrhagic adrenal cysts. Adrenal hemangiomas are seen more commonly in women and typically present between the ages of 50 and 70 years. The majority are larger than 10 cm at presentation.104 Most lesions are incidentally discovered. However, rarely an adrenal hemangioma can attain such a large size, such that mass effect or spontaneous hemorrhage render the lesion symptomatic and result in its detection.105 On US examinations, adrenal hemangiomas appear as a heterogeneously echogenic suprarenal mass. Diffused irregular anechoic areas with hyperechoic septa are seen throughout the lesion typical of cavernous hemangiomas in any location.106 On CT, adrenal hemangioma appears as a heterogeneous, smoothly marginated, relatively low attenuation soft-tissue mass containing speckled and coarse calcifications from prior hemorrhage and ring calcifications representing phleboliths.104,107 Occasionally, hemangiomas can contain intralesional macroscopic fat that can be detected by CT and MRI.104 MRI shows low signal intensity on T1-weighted images and high signal on T2-weighted images.104 In some cases, multiple internal septations can be seen on the T2-weighted images, representing the walls of the vascular channels. Post IV contrast administration, heterogeneous enhancement is observed on both CT and MRI. The classic nodular, interrupted, peripheral enhancement described in hepatic hemangiomas has also been described in case reports of adrenal hemangiomas.104,108,109 Slow, venous filling has been described on angiography. Large lesions resembling an adrenal hemangioma can rarely represent hemangiosarcomas that will require surgical resection.

Nonneoplastic Causes of Adrenal Nodules and Masses Adrenal hemorrhage is a common nonneoplastic cause of an adrenal nodule or mass.

Adrenal hemorrhage Unilateral and bilateral adrenal hemorrhage is a rare event that can be observed at any age. Causes include blunt and

Table 8-4. Causes of Adrenal Hemorrhage 1. Trauma a. Blunt b. Penetrating 2. Coagulopathy 3. Physiologic stress a. Neonatal asphyxia b. Sepsis c. Burns d. Hypotension

penetrating trauma, coagulopathy, and physiologic stress such as sepsis or neonatal asphyxia (Table 8-4). Neonatal adrenal hemorrhage is the most common cause of an adrenal mass in the newborn. The condition is overwhelmingly unilateral, >90% of cases, and more common in male infants. The etiology is often uncertain, but there is an association with traumatic delivery, especially in the breech presentation, and maternal diabetes.110,111 The clinical picture varies from an asymptomatic abdominal mass, prolonged neonatal jaundice second to breakdown of hemoglobin, to dramatic hypotension and shock, in case of massive bleeding. Imaging is mostly obtained by sonography, though noncontrast CT or MRI are helpful to distinguish simple hemorrhage from one that coincides with or is due to a neuroblastoma. Serial imaging will demonstrate eventual involution of the mass; in some cases, circular or angular calcifications may develop several months later. In the adult, adrenal hemorrhage is uncommon and can be associated with sepsis, burns, stress, blunt trauma, or hypotension. The hematologic disorders, thrombocytopenia, disseminated intravascular coagulation, therapeutic anticoagulation, antiphospholipid antibody syndrome, and systemic lupus erythematosus, have been associated with adrenal hemorrhage. Trauma is also a known cause of adrenal hemorrhage, where it can be bilateral in up to 20% of cases.112 Bilateral hemorrhage can lead to clinical adrenal insufficiency (Addison disease), often several weeks following the incident.113 Adrenal hemorrhage is unilateral in approximately 80% of cases, the majority of which involve the right adrenal.114 One theory suggests that acute elevations in intraabdominal pressure are transmitted into the inferior vena cava and cause elevations in adrenal venous pressures. Because the right adrenal vein is shorter than the left adrenal vein, greater pressures are transmitted to the right than the left adrenal gland. Adrenal hematoma typically appears as a suprarenal, round, or oval mass on cross-sectional imaging examinations (see Figure 8-13). In many cases, it can be indistinctly marginated with stranding of the suprarenal, retroperitoneal fat. On CT, an acute hematoma will appear

Chapter 8 Imaging of the Adrenal 529

A

B

Figure 8-13 Spontaneous Adrenal Hemorrhage This 43-year-old woman underwent lobectomy for lung cancer. In the postoperative period, she was found to have an acute drop in serum hemoglobin. A and B. Enhanced CT demonstrates an inhomogeneous mass in the retroperitoneum just superior to

the right kidney with stranding of the adjacent retroperitoneal fat. Notice how portions of the mass, especially in B, appear hyperattenuating relative to skeletal muscle. This mass was not present on the preoperative CT and is consistent with spontaneous adrenal hemorrhage.

as hyperattenuating (value >50 HU), whereas subacute hematoma often is equal in attenuation to that of soft tissue (35-40 HU) (see Figure 8-14). Chronically, the hematoma will often liquefy and appear as a complex cystic mass. (See the subsequent section entitled Pseudocysts.) Calcifications can eventually form months later. Further, MRI is best suited to detect the presence of blood byproducts. Subacute hemorrhage appears as a high-signal-intensity process on both T1- and T2-weighted scans, because of the paramagnetic effect of methemoglobin.115 Later on, a dark ring can be seen surrounding the hematoma because of formation of hemosiderin or ferritin.115,116 The hematoma usually liquefies and slowly resolves over time (see Figure 1-10).

They are more common among women and can be bilateral in about 10% of cases.119 Histologically, adrenal cysts can be divided into 4  groups: pseudocysts, endothelial cysts, epithelial cysts, and parasitic cysts. There is some debate in the literature whether pseudocysts or endothelial cysts are most common. The incidence of pseudocysts among series ranges from 39% to 78% of cases and the incidence of endothelial cysts ranges from 20% to 45% of cases.118,119-121 Epithelial

Cystic-Appearing Adrenal Nodules and Masses Adrenal cysts are an uncommon finding with an incidence on autopsy studies of 0.064% to 0.18%.117 In the past, they have been associated with abdominal or flank pain, gastrointestinal symptoms as a result of mass effect on adjacent abdominal structures, or as a palpable abdominal mass. However, currently, many adrenal cystic masses are discovered incidentally on cross-sectional imaging studies performed for reasons unrelated to the mass.117 A study from the Mayo Clinic reviewed 41 excised adrenal cystic lesions: 66% were symptomatic, typically with pain or gastrointestinal symptoms, and 33% were incidental.118 However, because most asymptomatic, incidentally discovered adrenal cysts are not surgically resected, the incidence of symptomatic cysts is probably overrepresented in this series.

Figure 8-14 Posttraumatic Adrenal Hemorrhage This 26-year-old man was struck in the right upper quadrant during a soccer game. The right adrenal gland (arrow) is diffusely thickened and hyperattenuating but the left adrenal gland (arrowhead) is normal appearing. These findings are indicative of hemorrhage into the right adrenal as a result of the trauma.

530 Diagnostic Abdominal Imaging

Table 8-5. Causes of Adrenal Cysts 1. Prior hemorrhage (pseudocysts) 2. Endothelial cysts a. Blood vessel ectasia (angiomatous cyst) b. Lymphatic ectasia (lymphangiomatous cyst) 3. Epithelial cyst (true cyst) 4. Paracytic cyst a. Echinococcal infection b. Leishmaniasis

Parasitic cysts Most parasitic adrenal cysts are a result of echinococcal infection but rarely can be due to leishmaniasis.117 However, less than 0.5% of all patients with echinococcal infection will have adrenal involvement. Eosinophilia is noted in 20% of parasitic infections. Many patients may be clinically asymptomatic; however, complications include hemorrhage and shock (anaphylaxis) from a ruptured adrenal cyst or compression on the liver and infrahepatic vessels.117

Imaging features of adrenal cysts

and paracystic cysts account for less than 15% of adrenal cysts (Table 8-5).

Pseudocysts and cystic neoplasms Pseudocysts usually are the result of hemorrhage into normal adrenal tissue and occasionally may be due to hemorrhage into an adrenal tumor. Histologically, they are typically large unilocular cysts, with walls composed of dense, fibrous connective tissue, without an endothelial layer. The walls range from 1 to 5 mm in thickness, and can contain islands of adrenal cortical tissue in up to 19% of cases.117 According to 2 studies, approximately 7% of pseudocysts develop from necrosis within adrenal neoplasms, including adenomas, hemangiomas, pheochromocytomas, adrenal cortical carcinomas, and malignant hemangioendotheliomas.117,121 Rarely, adrenal hematomas can become secondarily infected to create adrenal abscesses. This has most often been reported in neonates.122

Endothelial cysts Endothelial cysts, often called “simple cysts,” are cysts characterized by smooth, flattened endothelial lining, and filled with clear or milky fluid.117 They can be subcategorized into lymphangiomatous and angiomatous cysts. Lymphangiomatous cysts arise from ectasia of lymphatic vessels in adrenal glands or cystic degeneration of a hamartoma. Angiomatous cysts are thought to arise from ectasia of blood vessels or may form after repeated episodes of hemorrhage in adrenal hemangiomas.117

Epithelial cysts Epithelial cysts, or “true cysts,” contain a smooth, flattened epithelial lining. Epithelial-lined cysts are subdivided into 3 groups based on pathogenesis: (1) glandular or retention cysts, which develop because of inclusion of displaced tissue from nonadrenal tissue during fetal development; (2) cystic adenomas; and (3) embryonal cysts.

Most cysts are small, although large ones up to 10 cm in diameter have been reported. Occasionally, adrenal cysts can be identified on chest or abdominal radiographs as thin, peripherally (rim) calcified lesions to the right or left of the spine in the upper abdomen (see Figure 8-15).123,124 Rim calcifications are relatively common among adrenal cysts and are located either in the wall or septa.124,125 The majority of calcified cysts likely represent pseudocysts from prior adrenal hematomas. Noncalcified cysts have imaging features consistent with other cysts anywhere in the body. Endothelial and epithelial cysts typically are unilateral, thin-walled, smoothbordered, round or oval masses with pure cystic internal structure.126 Lymphangiomatous variants of endothelial cysts can contain internal septa.127 On CT, they will have an attenuation usually less than 20 HU. On MRI, they appear as intermediate signal intensity on T1-weighted sequences and high signal on T2-weighted sequences, without softtissue components. The contents of adrenal cysts do not enhance after IV contrast injection.126,127 This is the important feature of the cyst that differentiates it from a lipid-rich adenoma, which has a fluidlike low attenuation on unenhanced CT, but clearly enhances, if IV contrast material is given. Imaging of pseudocysts will typically demonstrate a cystic mass with walls ranging from 1 to 5 mm thick. The contents of the cysts can appear as simple fluid in approximately one-half of patients (anechoic on US, 0-20 HU on CT, and hyperintense on T2-weighed MRI) or can contain internal debris or septations in the remaining (see Figure 8-15).128 Features of blood products can be seen with MRI examinations. Calcification of the wall or septum is moderately common.128 The presence of wall thickening or nodularity, especially if the nodule(s) enhance, in a cystic adrenal lesion on cross-sectional imaging is of concern for a necrotic tumor, such as a pheochromocytoma or adrenal cortical carcinoma. One report of 7 pathologically proven benign pseudocysts showed that the solid-appearing elements in pseudocysts did not enhance on CT images following IV contrast administration.128 Echinococcal cysts can have a variety of appearances. Typically, they demonstrate one of several imaging features that are characteristic of echinococcal cysts.103 Most

Chapter 8 Imaging of the Adrenal 531

A

B

C

D

E

F

Figure 8-15 Calcified Pseudocyst in 2 Patients A-D. This 50-year-old woman had breast cancer. Magnified images from (A) posteroanterior and (B) lateral chest radiograph shows a rim calcified mass (arrows) in the left retroperitoneum. In most cases, this will represent a calcified cyst in the adrenal, pancreas, or kidney. (C) Coronal T2-weighted and (D) coronal T1-weighted postgadolinium MRI images show an adrenal mass (arrows) that is high signal on T2-weighted sequences and does not enhance following gadolinium administration. There is debris seen in the dependent portion of the cyst in C. These findings are consistent with a proteinaceous adrenal cyst,

most likely a pseudocyst from prior hemorrhage. E and F. This 62-year-old woman was being evaluated for pulmonary nodules. (E) Contrast-enhanced CT shows a fluid attenuation mass with small focal regions of peripheral calcification (arrowheads) in the expected location of the adrenal. (F) Ultrasonographic examination show the mass to contain multiple internal echoes, with linear peripheral echogenic foci (arrowheads) representing the rim calcification. Note the shadowing (arrow) due to one of the calcifications. These features are characteristic of a benign, complicated adrenal cyst, probably a pseudocyst.

532

Diagnostic Abdominal Imaging

typical of these is a dominant cystic mass containing multiple smaller daughter cysts that line the internal wall of the dominant cyst. Occasionally, fine calcifications, called “hydatid sand” are found in the dependent portion of the cyst. With death of the organism, the internal membranes can become detached from the wall of the cyst and float within the central contents of the cyst, a finding called the “water lily sign.” Some echinococcal cysts appear as unilocular cysts that are indistinguishable from other simple adrenal cysts. Percutaneous aspiration of adrenal cyst fluid may occasionally be used in case of inconclusive imaging, such as large, suprarenal cystic lesions of unclear origin. Fluid analysis may reveal the presence of adrenal cortical hormones or precursors, indicating adrenal origin.125

BILATERAL ENLARGEMENT OF THE ADRENAL GLAND Bilateral enlargement of the adrenal gland can be due to adrenal hyperplasia and granulomatous infections of the adrenal gland. In some cases, bilateral adrenal metastases or hemorrhage will cause bilateral adrenal masses.

Adrenal Hyperplasia Adrenal hyperplasia represents a nonmalignant proliferation of normal adrenal cortical cells and nearly always involves both adrenal glands. Histologically, this can result from proliferation of any 1 of the 3 layers of the adrenal cortex, the zona glomerulosa, which produces aldosterone; the zona fasciculata, which produces cortisol; and the zona reticularis, which produces androstenedione. Adrenal hyperplasia is most clearly associated with syndromes of increased adrenocortical hormone production: Cushing syndrome (increased cortisol production), primary aldosteronism (increased aldosterone production), and adrenogenital syndrome (increased androstenedione production also called congenital adrenal hyperplasia).129,130 In some cases, adrenal hyperplasia can occur without apparent alterations in serum hormone levels and, despite the enlarged gland, the patient is asymptomatic.131-133 Acute stress will result in diminished size of the adrenal glands, as they use up their stores of cholesterol and cholesterol-derived hormones. However, chronic stress can result in hyperplasia of the gland (Table 8-6). Morphologically, adrenal hyperplasia can take 1 of 2 macroscopic pathologic and 3 imaging appearances. (1) In some cases, there is uniform proliferation of cortical tissue, resulting in uniform enlargement of the glands bilaterally. On cross-sectional imaging, this will often appear as uniform thickening of the limbs of both adrenals (see Figure 8-16). It will appear hyperenhancing following IV administration of iodinated CT contrast or gadolinium-based MRI contrast agents. This appearance

Table 8-6. Causes of Adrenal Hyperplasia A. Associated with hyperfunctioning of the gland 1. Cushing syndrome a. ACTH-secreting pituitary adenoma b. ACTH-secreting nonpituitary tumor i. Bronchogenic carcinoma ii. Thymoma iii. Bronchial carcinoid iv. Pheochromocytoma v. Neuroendocrine tumor of the pancreas 2. Primary aldosteronism 3. Congenital adrenal hyperplasia (adrenogenital syndrome) 4. Chronic physiologic stress B. Without hyperfunctioning of the gland (usually nodular hyperplasia)

is strongly associated with the syndromes of hormone imbalance. (2) In other situations, the proliferation is asymmetric, resulting in multinodular thickening of the glands, called nodular adrenal hyperplasia. In nodular adrenal hyperplasia, the nodules are small and multiple, distributed throughout both glands. Nodular hyperplasia has been associated with systemic hypertension, and can otherwise be clinically asymptomatic, without associated hormonal imbalances.132,133 On cross-sectional imaging, this will appear as multiple, small, typically less-than5-mm, nodules distributed through both adrenal glands (see Figure 8-17). (3) In some cases, the adrenal gland may be of normal thickness on cross-sectional imaging, despite pathologic evidence of hyperplasia. In this latter case, NP-59 scintigraphy may correctly identify cases of adrenal hyperplasia by detecting increased bilateral adrenal uptake.

Granulomatous Infections of the Adrenal Gland Both tuberculosis and histoplasmosis can develop disseminated infection throughout the adrenal gland resulting in a multinodular enlargement of the gland.

Tuberculosis A 28-year autopsy review of tuberculosis in Hong Kong demonstrated 6% incidence of adrenal involvement, the sixth most common site after the lungs, liver, spleen, kidneys, and bones.134 In the same series, 13% of patients with adrenal involvement developed Addison disease, and bilateral involvement at autopsy was seen in 69% of cases. The adrenals were approximately 40% enlarged over the normal gland at autopsy evaluation. In most

Chapter 8 Imaging of the Adrenal 533

A

B

C

D

Figure 8-16 Adrenal Hyperplasia in 2 Patients A and B. This 55-year-old man had a brain mass. Computed tomographic images through the adrenal glands demonstrate diffuse uniform enlargement of the cortex of the adrenals bilaterally, characteristic of adrenal hyperplasia. Notice how the cortex of the adrenals enhances more intensely than the adrenal medulla. Further biochemical and clinical analysis confirmed

a diagnosis of Cushing syndrome. C and D. This 70-year-old woman had a primary lung cancer. Contrast-enhanced CT shows diffuse uniform enlargement of the adrenals bilaterally. The cause was never evaluated. It is likely that she had hyperplasia due to ectopic production of ACTH by the lung cancer but it is possible that the hyperplasia was a response to chronic stress of a metastatic malignancy.

cases, adrenal disease is clinically undetected at the time of infection and discovery of adrenal involvement is identified by imaging examinations months and years after the initial infection. If detected during the acute phase of infection, usually during the evaluation for Addison disease, adrenal tuberculosis will demonstrate bilaterally enlarged adrenals in approximately 90% of cases (see Figure 8-18).135 In approximately half of cases, multiple small hypoattenuating lesions enlarging the adrenal gland or bilateral adrenal masses were identified.134,135 The chronic manifestation of adrenal tuberculosis is atrophy and calcification of the gland.136 Calcifications are commonly bilateral, and can appear as small punctate or larger, chunky calcifications within an atrophied gland.137

Histoplasmosis Histoplasma capsulatum is a ubiquitous dimorphic saprophyte that is endemic in the southern parts of North America but can be found worldwide.138 It naturally occurs in soil rich in bird and bat droppings.138 In the soil, it exists as a mycelium, but in tissue, it forms small budding cells. In most individuals, it causes a self-limited flulike illness. However, in some individuals, especially those who are immunocompromised, it can cause a severe disseminated illness. Adrenal involvement is often asymptomatic but can occasionally cause adrenal insufficiency.138-142 The cross-sectional imaging features of adrenal histoplasmosis vary depending on the stage of the disease.

534

Diagnostic Abdominal Imaging

A

B

C

D

E

F

Figure 8-17 Nodular Adrenal Hyperplasia in 2 Patients A-C. This 68-year-old man had prostate cancer and poorly controlled systemic hypertension. (A and B) Unenhanced CT shows some areas of nodular enlargement and some areas of relatively normal adrenal gland thickness. The nodular areas are slightly decreased attenuation. (C) T2-weighted MRI shows the same nodular thickening and show that the nodules are

slightly lower signal compared with other thickened portions of the gland. D-F. This 59-year-old man had colon cancer and benign essential hypertension. Contrast-enhanced images show multinodular thickening of the adrenal gland. These cases are pathologically unproven but likely represent nodular adrenal hyperplasia.

At the acute phase of infection, imaging will typically demonstrate nodular or masslike enlargement of both adrenal glands. The enlargement can be solid appearing or contain multiple cystic regions that are hypoechoic on US and hypoattenuating on CT.138-142 The lesions can be homogeneously or heterogeneously enhancing, and when heterogenous, they will often rim enhance around

hypoechoic regions. In the healing phase, 1 or multiple calcifications are demonstrated.

Bilateral Adrenal Metastases Adrenal metastases are frequently bilateral in distribution. In most cases, it is obvious that they represent bilateral

Chapter 8 Imaging of the Adrenal 535

Figure 8-18 Adrenal Disease in Tuberculosis This 59-year-old man had known active tuberculosis and had persistent fevers. There is mild diffuse enlargement of the adrenal glands bilaterally, the most common adrenal manifestation of patients with tuberculosis. This can represent either adrenal involvement by tuberculosis or represent hyperplasia due to the stress of chronic infection.

adrenal masses rather than hyperplasia of the gland (see Figure 8-19).

OTHER IMAGING AND CLINICAL FEATURES OF ADRENAL DISEASES Adrenal imaging is occasionally performed to evaluate for the cause of one of a few hormonal syndromes associated with adrenal pathology: Cushing syndrome, primary hyperaldosteronism, and feminization or virilization syndromes. This is an important topic that will be discussed here. The only imaging characteristic of adrenal lesions that are not adrenal masses or diffuse enlargement, is the presence of adrenal calcification, which will also be discussed here.

Hormonal Syndromes Associated with Adrenal Pathology Cushing syndrome Cushing syndrome is a syndrome of hirsutism, amenorrhea, truncal obesity, weakness, hypertension, and hyperglycemia and is a result of sustained high level of serum cortisol. In approximately 25% of cases, Cushing syndrome is caused by an autonomous (non–ACTH dependent), cortisol-secreting adrenal adenoma or adenocarcinoma. Approximately 75% of cases of Cushing syndrome are due to excessive ACTH stimulation, which causes hyperplasia

Figure 8-19 Bilateral Adrenal Metastasis This 49-year-old man had lung cancer. Contrast-enhanced CT shows nodular enlargement of the adrenals bilaterally. This appearance has been seen in patients with tuberculosis and adrenal metastasis but would be unusually large for adrenal hyperplasia. In this clinical setting, these findings are most consistent with bilateral metastasis.

of the adrenal cortex. Causes of excess ACTH include pituitary adenomas (Cushing disease) and nonpituitary neoplasms that secrete ACTH, including bronchogenic carcinoma, thymoma, bronchial carcinoid, pheochromocytoma, and islet cell tumor of the pancreas.129,143,144 Among patients with ACTH dependent Cushing syndrome, females outpace males by a 3:1 margin. Patients with Cushing syndrome will frequently undergo CT or MRI of the brain to search for a pituitary adenoma, and abdominal CT or MRI to evaluate for adrenal characteristics of the syndrome. Occasionally, CT of the thorax and abdomen will be performed to search for nonpituitary ACTH-secreting neoplasms. Evaluation of the adrenal glands will frequently point to the cause of Cushing syndrome. In approximately onequarter of patients, a unilateral nodule or mass will be identified, indicating the presence of a hyperfunctioning adrenal adenoma or adrenal cortical carcinoma.24,49 The remaining parts of the involved adrenal gland, as well as the uninvolved, contralateral adrenal are atrophic, because the levels of the circulating ACTH are very low. Most patients with ACTH-dependent Cushing syndrome will demonstrate CT or MRI evidence of bilateral adrenal hyperplasia (see Figure 8-16).129 In some patients with ACTH dependent Cushing syndrome, the adrenal glands may appear normal. However, in patients with Cushing syndrome, the absence of a unilateral adrenal mass points to bilateral adrenal hyperplasia, even if the glands appear normal on crosssectional imaging. Occasionally, multiple small nodules will

536

Diagnostic Abdominal Imaging

Imaging Notes 8-4. Imaging Features of Cushing’s Syndrome Adrenal

Pituitary

Other Site

Cause

3-cm mass

normal

normal

cortisol-secreting cortical carcinoma or cortisol-secreting large adrenal adenoma

Hyperplasia or normal

nodule/mass

normal

ACTH-secreting pituitary adenoma

Hyperplasia or normal

normal

lung nodule

ACTH-secreting bronchogenic carcinoma or ACTH-secreting carcinoid tumor

Hyperplasia or normal

normal

anterior mediastinal mass

ACTH-secreting thymoma

Hyperplasia or normal

normal

lung nodule

ACTH-secreting bronchogenic carcinoma

Hyperplasia or normal

normal

pancreas mass

ACTH-secreting neuroendocrine carcinoma

be found in the adrenal glands in a patient with Cushing disease. In this situation, it can be difficult to determine whether the elevated cortisol is due to adrenal hyperplasia or whether one of the nodules represents an autonomous hyperfunctioning adrenal adenoma. Furthermore, occasionally patients with bilateral adrenal hyperplasia will also have a CT- or MRI-detected non hyperfunctioning adenoma, which can be mistaken as a functional, unilateral adenoma. In confusing cases, a combination of CT or MR with radionuclide NP-59 scintigraphy may lead to the correct diagnosis in many patients.8 Cushing syndrome due to unregulated ACTH production results in bilateral adrenal cortical uptake and visualization, whereas in cases of an autonomous adenoma, only the lesion is visualized, while there is no uptake at all by the remaining adrenal cortex. In rare instances, adrenal venous blood sampling will be needed to identify the unilateral or bilateral source of excess cortisol.8 Rare types of bilateral hyperplasia in Cushing syndrome, such as massive macronodular and primary pigmented nodular hyperplasia, need to be differentiated from the common nodular hyperplasia. They are associated with low ACTH levels, lack of an adenoma, and adrenal enlargement on CT; the autonomous cortisol production in these cases can only be treated by bilateral adrenalectomies.

Primary aldosteronism (Conn syndrome) Aldosterone is a mineralocorticoid responsible for the regulation of blood pressure. This process occurs through rennin-angiotensin-aldosterone feedback loop. When blood volume is low, juxtaglomerular cells in the kidneys secrete

renin, which stimulates the production of angiotensin I which is then converted to angiotensin II. Angiotensin II has 2 major effects, constriction of blood vessels, resulting in increased blood pressure, and stimulation of the adrenal cortex to produce the hormone aldosterone. Aldosterone causes the renal tubules to increase the reabsorption of sodium and water into the blood, increasing blood volume, which also increases blood pressure. Elevated blood volume results in a decrease in renin and subsequent decrease in aldosterone. Hyperaldosteronism can be caused by direct elevation in aldosterone (primary aldosteronism) or by elevations in renin and angiotensin, leading to secondary increases in aldosterone (secondary aldosteronism). Decreased renin levels indicate primary rather than secondary aldosteronism.129 Approximately 90% of cases of primary aldosteronism are a result of either bilateral adrenal hyperplasia or an aldosteronesecreting adrenal adenoma. Less common causes include unilateral adrenal hyperplasia, aldosterone-secreting adrenal cortical carcinoma, and, rarely, disorders of the renin-angiotensin system.131,145-147 Adenomas are more often detected in women, whereas adrenal hyperplasia is more often detected in men. Recognition of Conn syndrome (primary aldosteronism due to an adrenal adenoma) as a cause of systemic hypertension is an important diagnosis because resection of the adenoma will lead to curing of systemic hypertension.129 Patients typically present with systemic hypertension, which is caused by elevations in serum sodium. Primary aldosteronism is now recognized to be the most common form of secondary hypertension although it still accounts for 2 mm), or of the vas deferens, or visualization of causes of obstruction such as stones, midline cysts, or chronic prostatitis. Seminal vesicle stones and hemorrhagic material in the seminal vesicle can also be seen in the setting of dilation. In equivocal cases, more definitive evaluation for ejaculatory duct obstruction can be obtained by invasive procedures, including TRUS-guided seminal vesicle puncture, vesiculography, and chromotubation.58

Seminal Vesicle Cysts Seminal vesicle cysts can be congenital or acquired. Congenital cysts are present at birth and are typically asymptomatic but can become symptomatic in young adulthood, probably as a result of sexual activity. Congenital seminal vesicle cysts are associated with urinary tract anomalies including renal agenesis, renal maldevelopment, and ectopic insertion of the ureter. Other associations with seminal vesicle cysts include autosomal dominant kidney disease (44%-60%), (in which case the cysts are typically bilateral), hemivertebra, absence of the testes and absence of the vas deferens.77,78 The most common urinary tract abnormality associated with seminal vesical cysts is ipsilateral renal agenesis, occurring in approximately two-thirds of patients.79 This association can be explained by their embryologic origin. The ureteral bud arises from the distal mesonephric duct and is responsible for inducing differentiation of the metanephric blastema into the kidney. The mesonephric duct differentiates into the appendix of the epididymis, paradidymis, epididymis, vas deferens, ejaculatory duct, seminal vesicle, and hemitrigone.80 Maldevelopment of the distal mesonephric duct results in both faulty ureteral budding and atresia of the ejaculatory duct, causing seminal vesicle cysts and ipsilateral renal agenesis. An abnormally cephalic origin of the ureteral bud from the mesonephric duct results in ectopic ureteral insertion into derivatives of the mesonephric duct, most commonly the prostatic urethra and the seminal vesicle.80 Ectopic insertion of the ureter into the seminal vesicle usually results in both cystic dilation of the seminal vesicle and maldevelopment of the ipsilateral kidney. Acquired seminal vesicle cysts are due to acquired obstruction of the ejaculatory duct, most commonly due to prostatitis or stones, and less likely from neoplasm, BPH or prior surgery. Acquired cysts tend to occur in older adults and are more commonly bilateral when compared to congenital cysts. Seminal vesicle cysts are often asymptomatic but can also present with a wide variety of symptoms, including pain, dysuria, hematuria, urinary tract infections,

Müllerian, Ejaculatory Duct and Utricle Cysts When large, ejaculatory duct and müllerian duct cysts (discussed in more depth previously) extend out of the prostate and may be off midline, but may be distinguished from seminal vesicle cysts by the presence of separate normal seminal vesicle (see Figure 11-42).

Cowper Gland Cyst Cowper gland cysts may be congenital or acquired. When acquired, they are usually due to trauma or infection. They are differentiated from other cysts by their characteristic location adjacent to the bulbomembranous portion of the posterior urethra either posterolaterally or posteriorly (see Figure 11-42).81 Small cysts are usually asymptomatic.

Periprostatic Abscess Prostatic abscesses can extend into the periprostatic tissues to involve the seminal vesicles, but should be readily

668

Diagnostic Abdominal Imaging

A

B

C

D

Figure 11-40 Seminal Vesicle Cyst A and B. Axial unenhanced CT image (A) through the pelvis and TRUS image (B) showing a lobulated fluid density structure (arrows) in the expected region of the seminal vesicle in keeping with a seminal vesicle cyst. C. Axial unenhanced image at the level of the kidneys demonstrating “flattened” appearance of the adrenal gland (arrow) in the setting of ipsilateral renal agenesis.

D. Axial T2-weighted image in a different patient with right renal ageneisis (not shown) demonstrating a cyst (straight arrow) communicating with the right seminal vesicle (curved arrow). Note central gland hypertrophy partially visualized (arrowhead). Although ejaculatory duct cysts and müllerian cysts may extend out of the prostate, they do not communicate with the seminal vesicle.

differentiated from other periprostatic cysts by the clinical scenario and by imaging findings associated with abscesses such as rim enhancement, gas within the cyst, and perilesional inflammatory fat stranding (see Figure 11-42).

Postbiopsy Hematoma Postbiopsy hematomas can also occur in the periprostatic tissues. In most cases, there will be a clinical history of prostate biopsy. MRI examinations will demonstrate signal

Chapter 11 Imaging of the Prostate and Seminal Vesicles

669

A

B

Figure 11-41 Cystic Dilation of the Seminal Vesicle Axial T1-weighted (A) and axial T2-weighted images (B) in a patient with dysplastic left kidney (not shown) demonstrating cystic enlargement of the left seminal vesicle (straight arrow),

which contains T1 hyperintense fluid indicative of proteinaceous or hemorrhagic content. Note normal right seminal vesicle (curved arrow).

characteristics of subacute blood, including hyperintense signal on T1-weighted MRI.

are lamellated bodies composed of the mucoproteins in the prostatic secretions and sloughed epithelial cells. Corpora amylacea are usually found in the acini of the peripheral zone or central zone, most commonly along the surgical capsule (see Figure 11-43).84 Calcification of these bodies occurs slowly over time and they usually are incidentally found in the aging prostate with no clinical significance. Endogenous calcification can be associated with pathologic conditions. For example, calcifications occurring in the apical peripheral zone inferior to the verumontanum, or diffuse “blocklike” regional nonshadowing areas

Other Periprostatic Fluid-Filled Lesions Ureteroceles, hydronephrotic kidneys, and bladder diverticula can result in a fluid-filled structure in the retrovesical space but can be differentiated on the basis of other features.79 An ectopic ureterocele is associated with ipsilateral collecting system duplication. A hydronephrotic pelvic kidney will be associated with absence of an ipsilateral orthotopic kidney, reniform shape, and the branching appearance of a dilated collecting system. Bladder diverticula will be adjacent to the bladder and will usually demonstrate communication with the urine in the bladder. Other rare periprostatic cystic lesions include peritoneal inclusion cysts, teratomas, and lymphangiomas.82,83

UNIQUE PROSTATE, SEMINAL VESICLE, AND PERIPROSTATIC LESIONS Prostatic Calcifications Calcifications within the substance of the prostate are a common phenomenon. They may be endogenous (develop within prostatic acini) or exogenous (derived from urine)84 (Table 11-5). The most common prostatic calcification is due to endogenous calcification of corpora amylacea. Corpora amylacea

Table 11-5. Causes of Prostatic or Periprostatic Calcification A. Normal aging 1. BPH 2. Corpora amylacea B. Inflammation 1. Chronic prostatitis (bacterial or abacterial) 2. Granulomatous disease (e.g. TB) 3. Prior radiation C. Other 1. Stasis of secretions due to obstruction 2. Systemic conditions (hypercalcemia) 3. Exogenous calcifications (calculus in urethra or in urethral diverticulum)

670

Diagnostic Abdominal Imaging

A

C

B

D

Figure 11-42 Other Periprostatic Cystic Lesions A and B. Axial T2-weighted image (A) demonstrates a cystic structure (asterisk) in the vicinity of the seminal vesicles. Based on the axial images, the major differential considerations would be seminal vesicle cyst, ejaculatory duct cyst, and müllerian cyst. Sagittal T2-weighted image (B) shows the cyst (asterisk) to be separate from the seminal vesicles with a pear shape and neck approaching the urethra, and appearance most in keeping

with a müllerian cyst. C. Probable Cowper gland cyst. Sagittal T2-weighted image showing a cystic periurethral structure (arrow) in the region of the urogenital diaphragm (in the expected location of Cowper glands), most likely a retention cyst. D. Postbiopsy abscess. Axial T2-weighted image demonstrates at thick-walled collection (asterisk) posterior to the seminal vesicles (arrows) post biopsy, in keeping with an abscess that resolved over time.

of calcification, are suggestive of chronic prostatitis (see Figure 11-43).57 Relatively extensive calcification throughout the prostate has been described due to tuberculosis prostatitis.85 Other causes of endogenous calcifications include dystrophic calcifications related to BPH, calcification related to metabolic disorders such as hypercalcemia, vitamin D intoxication and ochronosis, and calcifications related to

stasis of secretions as may occur by obstruction of prostatic ducts by strictures, hyperplasia, or less likely, cancer.84 Ejaculatory duct calcifications simulate intraprostatic calcifications but may be identified by their characteristic paramedian location (see Figure 11-43). These are discussed in more detail in the subsequent section: Calcifications of the ejaculatory duct.

Chapter 11 Imaging of the Prostate and Seminal Vesicles

671

A

B

C

D

Figure 11-43 Prostate Calcifications A. Corpora amylacea. Echogenic foci (arrows) along the surgical capsule are typical in location for corpora amylacea, an incidental finding in the aging prostate. B. Peripheral zone calcifications. Axial TRUS images showing fine nonshadowing diffuse peripheral zone calcifications (arrows), a pattern typical of chronic prostatitis. C and D. Ejaculatory duct calcifications. Axial (C) and sagittal (D) TRUS images showing

punctate calcification near the midline of the prostate on the axial image (C, arrow), which are in the typical distribution of the ejaculatory duct on the sagittal image (D, arrows). Calcifications in and about the ejaculatory ducts may be sequela of or cause of ejaculatory duct stenosis, which may result in upstream dilation of the seminal vesicles or ejaculatory ducts. Arrowheads point to the urethra.

Exogenous calcification can also be due to migrated urinary bladder calculi lodged in the prostatic urethra or periurethral calculi in a urethra diverticula. Calcifications appear as echogenic foci with or without posterior acoustic shadowing on US, as hyperdense foci on CT, and as signal voids on both T1- and T2-weighted MRI sequences (see Figures 11-34 and 11-43). Brachytherapy seeds can mimic calcification but may be differentiated based on regular pattern and history, as well as the presence of ring down on US, susceptibility artifact on MRI, and density greater than calcium on CT (see Figures 11-15, 11-38, and 11-44).

Calcifications of Other Male Reproductive Structures Calcifications can also be found in the vas deferens and ejaculatory duct.

Calcifications of the Vas Deferens Calcification of the vas deferens can be seen on abdominal plain films and abdominal CT. They appear as curvilinear tubular calcifications extending upward from the central pelvis toward the lateral pelvis (see Figure 11-45). The most

672

Diagnostic Abdominal Imaging

Calcifications of the Ejaculatory Duct Calcifications around the ejaculatory duct as well as calcification within the ejaculatory duct (ED lithiasis) are findings that are seen more commonly in hypofertile men when compared with men with normal fertility.58 Ejaculatory duct calcifications are identified by the characteristic paramedian location (see Figure 11-43). When these calcifications are detected, other imaging findings suggestive of ejaculatory duct obstruction (dilated ejaculatory ducts, dilated seminal vesicles, seminal vesicle, and/or vas deferens lithiasis) should be sought.

Abnormal Low T2 Signal in the Seminal Vesicles on MRI Exams Figure 11-44 Radiation Seeds Brachytherapy seeds. Axial TRUS image showing multiple echogenic foci (arrows) with ringdown artifact typical of metal, in keeping with brachytherapy seeds. Note diffusely decreased signal in the peripheral zone (asterisk) reflecting treatment effect.

common causes are aging and diabetes mellitus but vas deferens calcifications can also be a result of tuberculous or other infections (gonorrhea, syphilis, schistosomiasis, and chronic urinary tract infections). When calcification occurs as a result of aging or diabetes, there is bilaterally symmetric calcification of the muscular elements of the vas deferens without luminal narrowing. It is believed that the calcifications occur at a greater frequency and younger age in the setting of diabetes mellitus. Inflammatory and infectious calcifications develop intraluminally and are more frequently unilateral, segmental, and/or irregular.

On T2-weighted MRI sequences, the seminal vesicles normally appear as bow-tie-shaped convolutions of tubules that typically measure 40 is diagnosed with hematospermia.81,89 Blood products within the seminal vesicle are usually in the subacute phase of evolution at the time of imaging and typically will appear as high signal on T1-weighted images and variable signal on T2-weighted images (see Figure 11-47). The high T1 signal is a clue differentiating seminal vesicle hemorrhage from the other causes of low T2 signal in the seminal vesicles, which are typically low in T1 signal.

Other seminal vesicle tumors Primary seminal vesicle carcinoma is very rare, with fewer than 100 cases reported in the literature, in adults of all ages.87 Symptoms may include hematospermia, hematuria, and outlet obstruction but usually are absent in early disease. On imaging, in the rare situation where the tumor is small or organ confined, soft-tissue enlargement of the seminal vesicle or a rectovesical mass may suggest the diagnosis (see Figure  11-46). However, most seminal vesicle carcinomas present at an advanced stage with invasion of adjacent organs, such as the prostate or urinary bladder, and therefore may be confused with adenocarcinomas arising from these structures. Normal PSA, normal carcinoembryonic antigen, and elevated CA125 as may occur with seminal vesicle carcinoma are clinical clues that may point to the diagnosis.87 Other tumors of the seminal vesicles are exceedingly rare but include secondary involvement by lymphoma or other rare neoplasms such as leiomyosarcoma, angiosarcoma, müllerian adenosarcoma–like tumor, carcinoid, seminoma, and cystosarcoma phyllodes (see Figure 11-46).77

Seminal vesicle hemorrhage Seminal vesicle hemorrhage is the imaging correlate of hematospermia and a frequent cause of low T2 signal in

Seminal vesiculitis Seminal vesiculitis typically is due to extension from bacterial prostatitis; although in endemic countries, tuberculosis and schistosomiasis are other causes. In its subacute to chronic phase, it can also result in proteinaceous or hemorrhagic debris within the seminal vesicle, leading to low T2 signal. The inflammation and debris can result in dilation and cystic change or conversely, atrophy of the seminal vesicle. The seminal vesicle wall can be thickened, and there can also be findings of chronic prostatitis. In the acute phase, other findings such as hyperemia, inflammatory stranding, and abscesses may also be present (see Figure 11-47).77

A

B

C

D

E

F

Figure 11-46 Neoplastic Involvement of the Seminal Vesicles in 3 Patients Neoplastic involvement can be suspected if there is irregular soft-tissue enlargement or infiltration of the seminal vesicles, in combination with history of malignancy. A and B. Seminal vesicle invasion from prostate cancer. (A) Axial and (B) coronal T2-weighted images demonstrating low signal soft tissue (straight arrows) infiltrating the medial seminal vesicles with loss of normal lobular architecture (curved arrows indicating preserved seminal vesicles), in keeping with seminal vesicle invasion from prostate cancer. C and D. Seminal vesicle lymphoma in a patient with aggressive enteropathyassociated T-cell lymphoma. (C) Axial T2-weighted image shows soft-tissue replacement (asterisk) of most of the seminal vesicles (arrow) that was FDG avid, which would be atypical for prostate carcinoma. Note soft-tissue thickening of urinary bladder (arrowhead), also proven to be lymphoma. (D) On US, the seminal vesicles demonstrate enlargement, decreased echogenicity, and loss of normal architecture (arrows). Rectal wall was also diffusely thickened (arrowhead). E and F. Seminal vesicle carcinoma. Axial (E) and coronal (F) images from enhanced CT in a young male shows large mass (asterisk) centered in the right seminal vesicle with extension into of the prostate and left seminal vesicle; note the left seminal vesicle is partially spared (arrow). Biopsy was most consistent with primary seminal vesicle carcinoma.

Chapter 11 Imaging of the Prostate and Seminal Vesicles

675

A

B

C

D

Figure 11-47 Non-neoplastic Low T2 Signal in the Seminal Vesicles A and B. Postbiopsy hemorrhage. Axial T2-weighted image (A) shows low signal (arrow) diffusely throughout the left seminal vesicle with preservation of the lobular architecture, which on an axial T1-weighted image (B) corresponds to high signal (arrow), in keeping with hemorrhage. C. Acute seminal vesiculitis. Axial T2-weighted image demonstrates low signal throughout both seminal vesicles (arrows) compared to the

signal intensity of urine in the bladder (B). D. Presumed amyloid. Axial T2-weighted image in an older gentleman shows diffusely low-signal retracted seminal vesicles with preservation of lobular architecture; this appearance is most consistent with amyloid or end-stage fibrosis of the seminal vesicles from prior inflammation. Treatment effect (due to radiation or hormonal therapy for prostate cancer) may result in a similar appearance.

Amyloidosis

can have  an imaging appearance similar to seminal vesicle  amyloidosis. However, the diagnosis of hormonal/ radiation therapy as the cause of the signal abnormality is based on clinical history of ongoing or prior therapy.

Senile seminal vesicle amyloidosis, which occurs in up to 17% of men older than age 50, is another benign cause of low signal in the seminal vesicles. This phenomenon can be mistaken for invasion of the seminal vesicle by prostate carcinoma.83 Atrophy of the seminal vesicle and preservation of normal lobular pattern can be a clue to the nonneoplastic etiology: seminal vesicle amyloidosis.

Iatrogenic Hormonal and/or radiation therapy commonly leads to low T2 signal and atrophy of the seminal vesicles. This

Seminal vesicle calculi Calcifications can occur in the seminal vesicles, usually as sequel of prior infection or inflammation. Uremia, hyperparathyroidism, and diabetes are other uncommon causes. The calcifications appear as intraluminal low-signal-intensity foci on T2-weighted images (see Figure 11-48).

676

Diagnostic Abdominal Imaging

A

B

Figure 11-48 Seminal Vesicle Calculi A. Axial image from an enhanced CT scan showing hyperdense foci within the seminal vesicles (arrows) in keeping with

calcifications. B. Axial T2-weighted image demonstrates low T2 signal rounded structures within the seminal vesicles (arrows) representing seminal vesicle calculi.

Congenital Anomalies of the Prostate, Seminal Vesicles, and Related Structures Agenesis of vas deferens is present in 1% to 7% of otherwise normal males.60 Bilateral vas deferens agenesis is present in 99% of those with cystic fibrosis, and two-thirds of those with bilateral agenesis have mutations in the cystic fibrosis transporter gene. In cystic fibrosis, agenesis is postulated to be a result of luminal blockage of the vas deferens and seminal vesicles by thick secretions. There is usually

associated seminal vesicle agenesis; however, the kidneys are usually normal in cystic fibrosis since the blockage is believed to occur after 7 weeks in gestation.79 Other associated findings include prominence of the rete testis and the epididymal head as well as decreased T2 signal in the peripheral zone of the prostate on MRI.90 When bilateral, agenesis results in infertility despite normal testicular spermatogenesis. Agenesis can also be due to an early insult to the mesonephric duct during embryogenesis. If an embryologic

A

B

Figure 11-49 Agenesis of the Seminal Vesicle A. Axial T2-weighted image shows a normal left seminal vesicle (straight arrow), with no right seminal vesicle present in the

expected location (curved arrow). B. Axial T1-weighted image more superiorly in the pelvis shows pelvic kidney (arrow) ipsilateral to the side of the seminal vesicle agenesis.

Chapter 11 Imaging of the Prostate and Seminal Vesicles insult occurs before the seventh week of gestation, when the ureteric bud arises from the mesonephric duct, seminal vesicle agenesis will result and is reported in 45% of cases of bilateral vas deferens agenesis and up to 86% of cases of unilateral agenesis.77 Seminal vesicle agenesis is commonly associated with renal agenesis, which occurs in 79% of cases. However, in 9% of cases there is an ipsilateral normal kidney (see Figure 11-49).77

FINDINGS RELATED TO SURGICAL AND MEDICAL INTERVENTION IN THE PROSTATE The mainstays of therapy for prostate cancer include radical prostatectomy, brachytherapy seed placement, external beam radiation, and hormonal therapy.

Normal Postoperative Findings Radical prostatectomy involves removal of the prostate and seminal vesicles with creation of an anastomosis between the bladder neck and membranous urethra and preservation of the neurovascular bundles if possible to preserve potency. A lymphadenectomy is commonly also performed. MRI can be obtained following prostatectomy if there is an abnormal digital rectal examination or if there is a rise in the PSA. This is not an uncommon situation as biochemical failure occurs following prostatectomy in 10% to 53%, most occurring in the first 5 years following surgery.91 The normal appearance of the prostatectomy bed is illustrated (see Figure 11-50). Findings that may be visualized in the surgical bed include fibrosis, surgical clips, seminal vesicle remnants, residual prostatic tissue, postoperative chronic collections, periurethral collagen, and tumor recurrence.92 Fibrosis is manifested as low T1/T2 signal in the region of the anastomosis as well as in the rectovesical region, which demonstrates minimal if any enhancement. Surgical clips can be identified based on the susceptibility artifact they produce. In 1 study, seminal vesicle remnants were reported in 20%, where they were bilateral in 80% and complete seminal vesicles were found in 30%.93 In 53%, these appeared as convoluted fluid-filled remnants, and in the remainder, as low-signal masses. In the same study, an additional 38% of patients demonstrated low-signal masses in the lateral postprostatectomy fossa suggestive of fibrotic tips. Although such low-signal masses can mimic recurrence, the irregular convoluted appearance should suggest the diagnosis of residual seminal vesicles. Periurethral collagen, which is injected to treat incontinence, appears as intermediate- to low-signal masses near the anastomosis (see Figures 11-39 and 11-51). Collagen typically may be differentiated from tumor based on history, normal PSA, and lack of enhancement following gadolinium, as discussed in previous sections.

677

Residual prostatic tissue can rarely be visualized following prostatectomy (see Figure 11-39). The imaging appearance of residual tissue cannot be reliably distinguished from tumor recurrence. Postoperative collections should demonstrate features of either simple or complex fluid, with rim enhancement only, and typically should resolve over time. Local recurrence most commonly appears as a softtissue nodule that is similar to hyperintense relative to muscle on T2-weighted images (see Figure 11-52). This overlaps in appearance with benign processes such as fibrosis or inflammatory tissue. Recently, 2 studies91,94 have demonstrated improved accuracy for identifying local recurrence using a dynamic contrast-enhanced protocol with tumor differentiated from benign processes by the presence of early enhancement. A sensitivity and specificity up to 88% (69%-98%), and 100% (84%-100%) was reported in 1 of the studies.94 In spite of these promising results, in equivocal cases, biopsy may still be necessary. Magnetic resonance spectroscopy and diffusion weighted sequences are evolving techniques technique, under investigation, which may be helpful in distinguishing residual tissue from local recurrence but is beyond the scope of this chapter.95 Following radiation therapy and hormonal therapy, the prostate decreases in size and there is fibrosis and loss of hydration of the prostate and seminal vesicles, resulting in decreased signal on T2-weighted images (see Figure 11-38). If radiation therapy was performed using brachytherapy, then multiple low-signal seeds will be visualized that result in susceptibility artifact on MRI, echogenic foci on US, and high-attenuation foci on CT (see Figures 11-15, 11-38, and 11-44). Imaging can be used to evaluate for satisfactory placement of radiation seeds.

Complications of Treatment of Prostate Cancer Complications are discussed in excellent review by Yablon et al, the contents of which are summarized here.96 Complications of radical prostatectomy can be divided into intraoperative, perioperative, and long-term complications. Intraoperative complications include rectal injury (1%) and ureteral injury (0.05%-1.6% of cases). Rectal injury not recognized at the time of surgery may present in the postoperative period as a fistula to the urethra, skin, and urinary bladder. Ureteral injuries can result in leak with urinoma or urinary obstruction. Perioperative complications include anastomotic leak, collections, and catheter-related stricture and fistula. Anastomotic leak typically occurs posterolaterally, where it is difficult to achieve tension-free sutures because of poor surgical access. On voiding cystourethrogram, anastomotic leaks have indistinct, amorphous contours in contrast to plication defects, which appear as well-defined outpouchings (see Figure 11-53). Anastomotic leak is treated by catheter drainage until resolution. Potential postoperative

678

Diagnostic Abdominal Imaging

A

B

C

D Figure 11-50 Normal Postprostatectomy Appearance A. Axial T2-weighted image showing region of the vesicourethral anastomosis (arrow) with central fluid signal. Note absence of soft-tissue nodularity. B. Axial T1-weighted image at the same level as (A) showing foci of low signal (arrows) reflecting susceptibility artifact from microscopic amounts of metal. C. Sagittal T2-weighted image demonstrating normal appearance of the region of the anastomosis with funneled appearance to the bladder neck. D. Axial T2-weighted images at the level of the seminal vesicles demonstrating low signal areas representing a combination of signal voids due to surgical clips (arrows) and postsurgical fibrosis. E. Axial T2-weighted image in a separate patient demonstrating residual seminal vesicle tips (arrows) and signal voids due to surgical clips (arrowhead).

E

Chapter 11 Imaging of the Prostate and Seminal Vesicles

679

A

B

Figure 11-51 Periurethral Collagen Injections A and B. Periurethral collagen. Axial T2-weighted sequence (A) demonstrating soft-tissue nodules (arrows) around the anastomosis, which demonstrate no internal enhancement

on (B) axial T1-weighted fat saturated postgadolinium images, in keeping with periurethral collagen in this patient with longstanding history of incontinence that eventually required placement of an artificial urethral sphincter.

collections include abscess, hematoma, and lymphocele. Of these, lymphocele is common, occurring in up to 60% who undergo open lymphadenectomy. Most lymphoceles are asymptomatic and small, but treatment, typically with drainage, with or without sclerosis, may be necessary if the lymphocele is > 5  cm, compresses important structures, becomes infected, or is associated with pain. Fistula to adjacent structures such as the rectum is an uncommon complication. Long-term complications include anastomotic strictures, incontinence, erectile dysfunction, and tumor recurrence. A recent study has reported an incidence of anastomotic strictures of 4.8%, which may be due to surgical technique or postoperative fibrosis. These can be treated with transurethral balloon dilation, which is effective in 60% of cases initially but has unknown long-term duration. Incontinence is the most distressing complication, and is common. Total incontinence is reported in up to 17% of men, and stress incontinence is reported in up to 35% of men. Treatment options include injection of a periurethral bulking agent such as collagen, and placement of a total artificial urethral sphincter (AUS). An AUS consists of a cuff placed around the proximal bulbar urethra, which is connected via tubing traveling in the subcutaneous tissues to a sphincter control pump placed in the scrotum (see Figure 11-54). Complications related to the AUS include urethral atrophy beneath the cuff, cuff erosion kinked tubing, pump malfunction, pump migration, and infection. Erectile dysfunction occurs in 10% to 90% and may be treated medically, or by placement of a penile prosthesis. Tumor recurrence has already been discussed.

Common complications of radiation therapy include urinary complications such as cystitis, urethritis, voiding dysfunction, stress urinary incontinence (up to 12% in patients with prior TURP), as well as other complications such as proctitis, urinary fistula, sinus tract formation, radiation-induced osteitis, and erectile dysfunction.

Figure 11-52 Local Recurrence of Prostate Cancer Axial T2-weighted image in a patient with rising PSA showing soft-tissue nodularity (arrows) that is higher in signal than fibrosis, in keeping with locally recurrent tumor. Compare with Figure 11-51. Unlike periurethral collagen, recurrent tumor would be expected to enhance and the patient would not have a history of collagen injections.

680 Diagnostic Abdominal Imaging

A

B

Figure 11-53 Anastomotic Leak versus Plication Defect A. Image from a voiding cysoturethrogram in a patient approximately 10 days following radical prostatectomy shows multiple surgical clips in the pelvis and a Foley balloon in the bladder. There is amorphous contrast material (arrows) surrounding the region of the vesicourethral anastomosis that progressively increased in size during the course of the examination, in keeping with moderate to large anastomotic

leaks. The Foley catheter was left in place to allow for further healing. B. Imaging from a voiding cysoturethrogram from a separate patient 2 weeks following radical prostatectomy shows a Foley balloon in place and multiple pelvic surgical clips. There is a tiny well-defined outpouching of contrast from the right aspect of the anastomosis that did not change over the course of the examination, reflecting a plication defect (arrow).

A

B

Figure 11-54 Artificial Urethral Sphincter A. Image from a retrograde urethrogram show a device (arrows) in the region of the midbulbar urethra that is attached to a reservoir (asterisk). This is the characteristic appearance of an

artificial urethral sphincter. B. Axial CT image from the same patient shows the appearance of a urethral sphincter as a hyperdense structure in the region of the bulbar urethra (arrow).

Chapter 11 Imaging of the Prostate and Seminal Vesicles REFERENCES 1. Kumar V, Abbas AK, Fausto N, Robbins SL, Cotran RS. Robbins and Cotran Pathologic Basis of Disease. Philadelphia, PA: Elsevier Saunders; 2005:1050-1056. 2. Epstein JI. Chapter 91—Pathology of prostatic neoplasia. In: Wein AJ, Kavoussi LR, Novick AC, Partin AW, Peters CA, eds. Campbell-Walsh Urology. 9th ed. Philadelphia, PA: WB Saunders; 2007. 3. Kundra V, Silverman PM, Matin SF, Choi H. Imaging in oncology from the University of Texas M.D. Anderson Cancer Center: diagnosis, staging, and surveillance of prostate cancer. AJR Am J Roentgenol. 2007;189:830-844. 4. Halpern EJ. Prostate and seminal vesicle measurements. In: Goldberg BB, McGahan JP, eds. Atlas of Ultrasound Measurements. Philadelphia, PA: Mosby Elsevier; 2006: 370-374. 5. Curran S, Akin O, Agildere AM, Zhang J, Hricak H, Rademaker J. Endorectal MRI of prostatic and periprostatic cystic lesions and their mimics. AJR Am J Roentgenol. 2007;188:1373-1379.

681

18. Dahnert WF, Hamper UM, Eggleston JC, Walsh PC, Sanders RC. Prostatic evaluation by transrectal sonography with histopathologic correlation: the echopenic appearance of early carcinoma. Radiology. 1986;158:97-102. 19. Yu KK, Hricak H. Imaging prostate cancer. Radiol Clin North Am. 2000;38:59-85, viii. 20. Graser A, Heuck A, Sommer B, et al. Per-sextant localization and staging of prostate cancer: correlation of imaging findings with whole-mount step section histopathology. AJR Am J Roentgenol. 2007;188:84-90. 21. Beyersdorff D, Taupitz M, Winkelmann B, et al. Patients with a history of elevated prostate-specific antigen levels and negative transrectal US-guided quadrant or sextant biopsy results: value of MR imaging. Radiology. 2002;224:701-706. 22. Ikonen S, Karkkainen P, Kivisaari L, et al. Magnetic resonance imaging of clinically localized prostatic cancer. J Urol. 1998;159: 915-919. 23. Futterer JJ, Heijmink SWTPJ, Scheenen TWJ, et al. Prostate cancer localization with dynamic contrast-enhanced MR imaging and proton MR spectroscopic imaging. Radiology. 2006;241:449-458.

6. Hricak H, Choyke PL, Eberhardt SC, Leibel SA, Scardino PT. Imaging prostate cancer: a multidisciplinary perspective. Radiology. 2007;243:28-53.

24. Nakashima J, Tanimoto A, Imai Y, et al. Endorectal MRI for prediction of tumor site, tumor size, and local extension of prostate cancer. Urology. 2004;64:101-105.

7. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin. 2008;58:71-96.

25. Choi YJ, Kim JK, Kim N, Kim KW, Choi EK, Cho K-S. Functional MR imaging of prostate cancer. Radiographics. 2007;27:63-75.

8. Scardino P. Update: NCCN prostate cancer clinical practice guidelines. J Natl Compr Canc Netw. 2005;3(suppl 1):S29-S33.

26. Akin O, Sala E, Moskowitz CS, et al. Transition zone prostate cancers: features, detection, localization, and staging at endorectal MR imaging. Radiology. 2006;239:784-792.

9. Kawachi MH, Bahnson RR, Barry M, Carroll PR, Carter HB, Catalona WJ, Epstein JI, Etzioni RB, Hemstreet GP 3rd, Howe RJ, Kopin JD, Lange PH, Lilja H, Mohler J, Moul J, Nadler RB, Patterson S, Pollack A, Presti JC, Stroup AM, Urban DA, Wake R, Wei JT; National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology: prostate cancer early detection. J Natl Compr Canc Netw. 2007 Aug;5(7):714-736. 10. Carter HB, Allaf ME, Partin AW. Chapter 94—Diagnosis and staging of prostate cancer. In: Wein AJ, Kavoussi LR, Novick AC, Partin AW, Peters CA, eds. Campbell-Walsh Urology. 9th ed. Philadelphia, PA: WB Saunders; 2007. 11. Claus FG, Hricak H, Hattery RR. Pretreatment Evaluation of Prostate Cancer: Role of MR Imaging and 1H MR Spectroscopy. Radiographics. 2004;24:S167-S180. 12. Ramey JR, Halpern EJ, Gomella LG. Chapter 92— Ultrasonography and biopsy of the prostate. In: Wein AJ, Kavoussi LR, Novick AC, Partin AW, Peters CA, eds. CampbellWalsh Urology. 9th ed. Philadelphia, PA: WB Saunders; 2007. 13. McNeal JE, Villers AA, Redwine EA, Freiha FS, Stamey TA. Capsular penetration in prostate cancer. Significance for natural history and treatment. Am J Surg Pathol. 1990;14:240-247. 14. Epstein JI, Carmichael M, Walsh PC. Adenocarcinoma of the prostate invading the seminal vesicle: definition and relation of tumor volume, grade and margins of resection to prognosis. J Urol. 1993;149:1040-1045. 15. Green FL, Page DL, Fleming ID, et al. AJCC Cancer Staging Handbook. 6th ed. New York: Springer-Verlag; 2002. 16. Potter SR, Partin AW. Prostate cancer: detection, staging, and treatment of localized disease. Semin Roentgenol. 1999;34:269-283. 17. Lee F, Torp-Pedersen S, Littrup PJ, et al. Hypoechoic lesions of the prostate: clinical relevance of tumor size, digital rectal examination, and prostate-specific antigen. Radiology. 1989;170:29-32.

27. Li H, Sugimura K, Kaji Y, et al. Conventional MRI capabilities in the diagnosis of prostate cancer in the transition zone. AJR Am J Roentgenol. 2006;186:729-742. 28. Noguchi M, Stamey TA, Neal JE, Yemoto CE. An analysis of 148 consecutive transition zone cancers: clinical and histological characteristics. J Urol. 2000;163:1751-1755. 29. Linzer DG, Stock RG, Stone NN, Ratnow R, Ianuzzi C, Unger P. Seminal vesicle biopsy: accuracy and implications for staging of prostate cancer. Urology. 1996;48:757-761. 30. Partin AW, Kattan MW, Subong EN, et al. Combination of prostate-specific antigen, clinical stage, and Gleason score to predict pathological stage of localized prostate cancer. A multiinstitutional update. J Am Med Assoc. 1997;277:1445-1451. 31. D’Amico AV, Whittington R, Malkowicz SB, et al. Predicting prostate specific antigen outcome preoperatively in the prostate specific antigen era. J Urol. 2001;166:2185-2188. 32. Wang L, Hricak H, Kattan MW, et al. Prediction of seminal vesicle invasion in prostate cancer: incremental value of adding endorectal MR imaging to the Kattan nomogram. Radiology. 2006;242:182-188. 33. Wang L, Hricak H, Kattan MW, Chen H-N, Scardino PT, Kuroiwa K. Prediction of organ-confined prostate cancer: incremental value of MR Imaging and MR spectroscopic imaging to staging nomograms. Radiology. 2005;238:597-603. 34. Yu KK, Hricak H, Alagappan R, Chernoff DM, Bacchetti P, Zaloudek CJ. Detection of extracapsular extension of prostate carcinoma with endorectal and phased-array coil MR imaging: multivariate feature analysis. Radiology. 1997;202:697-702. 35. Cornud F, Flam T, Chauveinc L, et al. Extraprostatic spread of clinically localized prostate cancer: factors predictive of pT3 tumor and of positive endorectal MR imaging examination results. Radiology. 2002;224:203-210.

682

Diagnostic Abdominal Imaging

36. Sala E, Akin O, Moskowitz CS, et al. Endorectal MR Imaging in the evaluation of seminal vesicle invasion: diagnostic accuracy and multivariate feature analysis. Radiology. 2006;238:929-937.

58. Langer JE, Cornud F. Inflammatory disorders of the prostate and the distal genital tract. Radiol Clin North Am. 2006;44: 665-677, vii.

37. Grignon DJ. Unusual subtypes of prostate cancer. Mod Pathol. 2004;17:316-327.

59. Shukla-Dave A, Hricak H, Eberhardt SC, et al. Chronic prostatitis: MR imaging and 1H MR spectroscopic imaging findings—initial observations. Radiology. 2004;231:717-724.

38. Varghese SL, Grossfeld GD. The prostatic gland: malignancies other than adenocarcinomas. Radiol Clin North Am. 2000;38: 179-202.

60. Parsons RB, Fisher AM, Bar-Chama N, Mitty HA. MR imaging in male infertility. Radiographics. 1997;17:627-637.

39. Chang JM, Lee HJ, Lee SE, et al. Unusual tumours involving the prostate: radiological-pathological findings. Br J Radiol. 2008;81:907-915.

61. Schiebler ML, Schnall MD, Pollack HM, et al. Current role of MR imaging in the staging of adenocarcinoma of the prostate. Radiology. 1993;189:339-352.

40. Mazzucchelli R, Lopez-Beltran A, Cheng L, Scarpelli M, Kirkali Z, Montironi R. Rare and unusual histological variants of prostatic carcinoma: clinical significance. BJU Int. 2008;102:1369-1374.

62. Wang L, Zhang J, Schwartz LH, et al. Incremental value of multiplanar cross-referencing for prostate cancer staging with endorectal MRI. AJR Am J Roentgenol. 2007;188:99-104.

41. Khan A, Ramchandani P. Unusual and uncommon prostatic lesions. Semin Roentgenol. 1999;34:350-363. 42. Kajiwara M, Mutaguchi K, Usui T. Ductal carcinoma of the prostate with multilocular cystic formation [in Japanese]. Hinyokika Kiyo. 2002;48:557-560. 43. Zini L, Villers A, Leroy X, Ballereau C, Lemaitre L, Biserte J. Cystic prostate cancer: a clinical entity of ductal carcinoma [in French]. Prog Urol. 2004;14:411-413. 44. Saito S, Iwaki H. Mucin-producing carcinoma of the prostate: review of 88 cases. Urology. 1999;54:141-144. 45. Schwartz LH, LaTrenta LR, Bonaccio E, Kelly WK, Scher HI, Panicek DM. Small cell and anaplastic prostate cancer: correlation between CT findings and prostate-specific antigen level. Radiology. 1998;208:735-738. 46. Abbas F, Civantos F, Benedetto P, Soloway MS. Small cell carcinoma of the bladder and prostate. Urology. 1995;46:617-630.

63. Ishikawa M, Okabe H, Oya T, et al. Midline prostatic cysts in healthy men: incidence and transabdominal sonographic findings. AJR Am J Roentgenol. 2003;181:1669-1672. 64. Nghiem HT, Kellman GM, Sandberg SA, Craig BM. Cystic lesions of the prostate. Radiographics. 1990;10: 635-650. [published erratum appears in Radiographics. 1990;10(5):963] 65. Galosi AB, Montironi R, Fabiani A, Lacetera V, Galle G, Muzzonigro G. Cystic lesions of the prostate gland: an ultrasound classification with pathological correlation. J Urol. 2009;181:647-657. 66. Tsujimoto Y, Satoh M, Takada T, et al. Papillary cystadenocarcinoma of the prostate: a case report [in Japanese]. Hinyokika Kiyo. 2007;53:67-70. 67. Rusch D, Moinzadeh A, Hamawy K, Larsen C. Giant multilocular cystadenoma of the prostate. AJR Am J Roentgenol. 2002;179:1477-1479.

47. Lazar EB, Whitman GJ, Chew FS. Embryonal rhabdomyosarcoma of the prostate. AJR Am J Roentgenol. 1996;166:72.

68. Allen EA, Brinker DA, Coppola D, Diaz JI, Epstein JI. Multilocular prostatic cystadenoma with high-grade prostatic intraepithelial neoplasia. Urology. 2003;61:644.

48. Cheville JC, Dundore PA, Nascimento AG, et al. Leiomyosarcoma of the prostate. Report of 23 cases. Cancer. 1995;76:1422-1427.

69. Ishida K, Kubota Y, Takada T, et al. A case of prostate cancer with cyst formation [in Japanese]. Hinyokika Kiyo. 2003;49:235-237.

49. Sexton WJ, Lance RE, Reyes AO, Pisters PW, Tu SM, Pisters LL. Adult prostate sarcoma: the M. D. Anderson Cancer Center Experience. J Urol. 2001;166:521-525.

70. Tuziak T, Spiess PE, Abrahams NA, Wrona A, Tu SM, Czerniak B. Multilocular cystadenoma and cystadenocarcinoma of the prostate. Urol Oncol. 2007;25:19-25.

50. Hansel DE, Herawi M, Montgomery E, Epstein JI. Spindle cell lesions of the adult prostate. Mod Pathol. 2007;20:148-158. 51. Kaufman JJ, Berneike RR. Leiomyoma of the prostate. J Urol. 1951;65:297-310. 52. Kitajima K, Kaji Y, Imanaka K, Hayashi M, Kuwata Y, Sugimura K. MR imaging findings of pure prostatic leiomyoma: a report of two cases. J Comput Assist Tomogr. 2006;30:910-912. 53. Bostwick DG, Iczkowski KA, Amin MB, Discigil G, Osborne B. Malignant lymphoma involving the prostate: report of 62 cases. Cancer. 1998;83:732-738. 54. Bostwick DG, Mann RB. Malignant lymphomas involving the prostate. A study of 13 cases. Cancer. 1985;56:2932-2938. 55. White S, Hricak H, Forstner R, et al. Prostate cancer: effect of postbiopsy hemorrhage on interpretation of MR images. Radiology. 1995;195:385-390. 56. Tamada T, Sone T, Jo Y, et al. Prostate cancer: relationships between postbiopsy hemorrhage and tumor detectability at MR diagnosis. Radiology. 2008;248:531-539. 57. Wasserman NF. Prostatitis: clinical presentations and transrectal ultrasound findings. Semin Roentgenol. 1999;34:325-337.

71. Outwater E, Schiebler ML, Tomaszewski JE, Schnall MD, Kressel HY. Mucinous carcinomas involving the prostate: atypical findings at MR imaging. J Magn Reson Imaging. 1992;2:597-600. 72. Roehrborn CG, Mcconnell JD. Chapter 86—benign prostatic hyperplasia: etiology, pathophysiology, epidemiology, and natural history. In: Wein AJ, Kavoussi LR, Novick AC, Partin AW, Peters CA, eds. Campbell-Walsh Urology. 9th ed. Philadelphia, PA: WB Saunders; 2007. 73. Berry SJ, Coffey DS, Walsh PC, Ewing LL. The development of human benign prostatic hyperplasia with age. J Urol. 1984;132: 474-479. 74. Lovett K, Rifkin MD, McCue PA, Choi H. MR imaging characteristics of noncancerous lesions of the prostate. J Magn Reson Imaging. 1992;2:35-39. 75. Potter SR, Partin AW. Prostatitis syndromes and benign prostatic hyperplasia. Semin Roentgenol. 1999;34:256-268. 76. Maki DD, Banner MP, Ramchandani P, Stolpen A, Rovner ES, Wein AJ. Injected periurethral collagen for postprostatectomy urinary incontinence: MR and CT appearance. Abdom Imaging. 2000;25:658-662.

Chapter 11 Imaging of the Prostate and Seminal Vesicles 77. Kim B, Kawashima A, Ryu J-A, Takahashi N, Hartman RP, King BF Jr. Imaging of the seminal vesicle and vas deferens. Radiographics. 2009;29:1105-1121.

88. Jinza S, Noguchi K, Hosaka M. Retrospective study of 107 patients with hematospermia [in Japanese]. Hinyokika Kiyo. 1997;43:103-107.

78. Patel B, Gujral S, Jefferson K, Evans S, Persad R. Seminal vesicle cysts and associated anomalies. BJU Int. 2002;90:265-271.

89. Han M, Brannigan RE, Antenor JA, Roehl KA, Catalona WJ. Association of hemospermia with prostate cancer. J Urol. 2004;172:2189-2192.

79. Arora SS, Breiman RS, Webb EM, Westphalen AC, Yeh BM, Coakley FV. CT and MRI of congenital anomalies of the seminal vesicles. AJR Am J Roentgenol. 2007;189:130-135. 80. Livingston L, Larsen CR. Seminal vesicle cyst with ipsilateral renal agenesis. AJR Am J Roentgenol. 2000;175:177-180. 81. Torigian DA, Ramchandani P. Hematospermia: imaging findings. Abdom Imaging 2007;32:29-49. 82. Hatano K, Tsujimoto Y, Ichimaru N, Miyagawa Y, Nonomura N, Okuyama A. Rare case of aggressive angiomyxoma presenting as a retrovesical tumor. Int J Urol. 2006;13:1012-1014. 83. Ramchandani P, Schnall MD, LiVolsi VA, Tomaszewski JE, Pollack HM. Senile amyloidosis of the seminal vesicles mimicking metastatic spread of prostatic carcinoma on MR images. AJR Am J Roentgenol. 1993;161:99-100. 84. Dahnert WF. Prostatic Calcifications. In: Resnick M, Watanabe H, Karr JP, eds. Diagnostic Ultrasound of the Prostate. New York: Elsevier Science; 1989:178-182. 85. Jung YY, Kim JK, Cho K-S. Genitourinary tuberculosis: comprehensive cross-sectional imaging. AJR Am J Roentgenol. 2005;184:143-150. 86. Wymenga LF, Duisterwinkel FJ, Groenier K, Mensink HJ. Ultrasound-guided seminal vesicle biopsies in prostate cancer. Prostate Cancer Prostatic Dis. 2000;3:100-106. 87. Thiel R, Effert P. Primary adenocarcinoma of the seminal vesicles. J Urol. 2002;168:1891-1896.

683

90. Simpson WL Jr, Rausch DR. Imaging of male infertility: pictorial review. AJR Am J Roentgenol. 2009;192:S98-S107. 91. Cirillo S, Petracchini M, Scotti L, et al. Endorectal magnetic resonance imaging at 1.5 Tesla to assess local recurrence following radical prostatectomy using T2-weighted and contrastenhanced imaging. Eur Radiol. 2009;19:761-769. 92. Allen SD, Thompson A, Sohaib SA. The normal post-surgical anatomy of the male pelvis following radical prostatectomy as assessed by magnetic resonance imaging. Eur Radiol. 2008;18: 1281-1291. 93. Sella T, Schwartz LH, Hricak H. Retained seminal vesicles after radical prostatectomy: frequency, MRI characteristics, and clinical relevance. AJR Am J Roentgenol. 2006;186:539-546. 94. Casciani E, Polettini E, Carmenini E, et al. Endorectal and dynamic contrast-enhanced MRI for detection of local recurrence after radical prostatectomy. AJR Am J Roentgenol. 2008;190:1187-1192. 95. Pucar D, Shukla-Dave A, Hricak H, et al. Prostate cancer: correlation of MR imaging and MR spectroscopy with pathologic findings after radiation therapy—initial experience. Radiology. 2005;236:545-553. 96. Yablon CM, Banner MP, Ramchandani P, Rovner ES. Complications of prostate cancer treatment: spectrum of imaging findings. Radiographics. 2004;24:S181-S194.

This page intentionally left blank

CHAPTER

12

Imaging of the Scrotum and Penis Lisa P. Jones, MD, PhD Jason N. Itri, MD, PhD Jill Langer, MD Parvati Ramchandani, MD

I. THE NORMAL SCROTUM a. Anatomy b. Ultrasonographic Appearance c. MRI Appearance II. FOCAL TESTICULAR DISORDERS a. Solid-Appearing Intratesticular Lesion or Mass i. Germ cell neoplasms ii. Other testicular tumors iii. Orchitis and epididymo-orchitis iv. Abscess v. Hematoma vi. Splenogonadal fusion vii. Testicular sarcoidosis viii. Other causes of solid appearing testicular lesion b. Geographic Testicular Lesions i. Neoplasm ii. Orchitis and epididymo-orchitis iii. Testicular fibrosis/atrophy iv. Segmental infarction v. Artifact: the two-tone testicle c. Bilateral Testicular Lesions i. Bilateral neoplasms ii. Bilateral inflammatory or postinflammatory lesions iii. Other unusual causes of bilateral testicular lesions d. Cystic Testicular Lesions i. Simple testicular cysts ii. Complex testicular cysts iii. Mimics of testicular cystic lesions III. DIFFUSE TESTICULAR DISORDERS a. Diffusely Enlarged Testicle with Abnormal Echogenicity or Signal Intensity i. Testicular infarction ii. Orchitis and epididymo-orchitis iii. Trauma and testicular rupture iv. Neoplasms causing diffuse testicular abnormality

v. Testicular fibrosis/atrophy vi. Testicular prosthesis b. Diffusely Hyperemic Testicle i. Orchitis ii. Torsion-detorsion (Intermittent torsion) iii. Neoplasms iv. Other IV. EXTRATESTICULAR DISORDERS a. Focal Epididymal Disorders i. Epididymal neoplasms ii. Inflammatory masses of the epididymis iii. Traumatic mass—hematoma iv. Complex epididymal cyst b. Other Paratesticular Masses i. Neoplasms of the spermatic cord and extratesticular tissues ii. Nonneoplastic masses c. Diffusely Enlarged Epididymis i. Epididymitis and epididymo-orchitis ii. Sarcoidosis iii. Tubular ectasia of the epididymis iv. Torsion of epididymis v. Spermatic cord torsion vi. Diffuse epididymal neoplasm: leukemia and lymphoma V. UNIQUE SCROTAL LESIONS a. Cystic Lesions in the Scrotal Sac i. Simple and complicated hydroceles ii. Epididymal cyst and spermatocele iii. Inguinal hernia iv. Mesothelioma of the tunica vaginalis b. Scrotal Echogenic Foci i. Calcifications and other echogenic foci associated with testicular neoplasms ii. Calcifications associated with benign conditions iii. Intratesticular gas iv. Iatrogenic material

685

686

Diagnostic Abdominal Imaging

c. Extratesticular Echogenic Foci i. Extratesticular calcification ii. Gas iii. Foreign bodies iv. Other d. Nonpalpable Testis VI. ANATOMY OF THE PENIS VII. IMAGING OF THE PENIS VIII. NORMAL IMAGING APPEARANCE OF THE MALE URETHRA AND PENIS a. Ultrasonography b. Magnetic Resonance Imaging IX. DISEASES OF THE PENIS a. Solid Lesions of the Penis i. Primary malignancies of the penis ii. Squamous cell carcinoma iii. Nonsquamous penile malignancies

THE NORMAL SCROTUM Anatomy The scrotum is a multilayered sac divided by a midline septum, with each half containing a testis, epididymis, and intrascrotal portion of the spermatic cord. The tunica vaginalis is a pouch of serous membrane derived from the processus vaginalis that is reflected on the internal surface of the scrotum, the parietal layer, and covers the surface of the testis and the epididymis, the visceral layer. Only the posterior aspect of the testis, which is the site of attachment, is not covered by the tunica vaginalis. The parietal and the visceral layers are normally separated by a few milliliters of fluid. The testes are symmetric ovoid organs covered by a fibrous layer called the tunica albuginea. The tunica dives into the testicle forming the mediastinum testis. Septa arising from the mediastinum divide the testis into 250 to 400 lobules, each containing 1 to 3 seminiferous tubules, spermatocytes that give rise to sperm, Leydig cells that produce testosterone, and supporting Sertoli cells.1 The seminiferous tubules become the tubuli recti, which drain into dilated spaces in the mediastinum called the rete testis. The rete testis drain through 10 to 15 efferent ductules into the ductus epididymis, a 6- to 8-cm tubule that courses through the epididymis, and curves acutely in the region of the tail to become the ductus deferens.2 The epididymis lies along the posterolateral aspect of the testis and is composed of a head, body, and tail. The head (globus major) is located at the superior pole, whereas the tail (globus minor) is located along the inferior pole. The testes contain four appendages: the appendix testis, appendix epididymis, the paradidymis, and vas aberrans. The appendix testis is a paramesonephric duct remnant, whereas the other

b. Penile Metastasis i. Benign neoplasms of the penis ii. Urethral neoplasms iii. Peyronie’s disease and other causes of penile fibrosis iv. Partial cavernosal thrombosis c. Cystic-Appearing Periurethral Lesions X. UNIQUE PENILE DISORDERS a. Penile Trauma b. Dorsal Vein Thrombosis c. Priapism d. Peyronie’s Disease e. Imaging Evaluation of Erectile Dysfunction

appendages are mesonephric duct remnants. The appendix testis and appendix epididymis can occasionally be detected by ultrasonography (US) and magnetic resonance imaging (MRI) examinations and will usually appear as small, 5 cm in greatest dimension

Pathologic involvement pN0 pN1 pN2

pN3

No regional lymph node metastases One or multiple lymph node metastasis with diameter ≤2 cm and 5 or fewer positive nodes One or multiple lymph nodes with the largest having a maximal diameter between 2 and 5 cm, more than 5 nodes positive, with none >5 cm, or evidence of extranodal extension of tumor Metastasis with a lymph node mass >5 cm in greatest dimension

Distant metastases (M) MX M0 M1a M1b

Distant metastasis cannot be assessed No distant metastasis Nonregional lymph node or pulmonary metastasis Distant metastasis other than to nonregional lymph nodes and lungs

Serum tumor markers (S) SX S0 S1 S2 S3

Tumor marker studies not available or not performed Tumor marker levels within normal limits LDH 10 000 ng/mL

Abbreviation: LDH, lactate dehydrogenase. Adapted from AJCC:Cancer staging book.158

seminomas also occur but in most cases the nodules are in continuity with one another, as part of the same tumor, rather than representing separate synchronous tumors in a truly multifocal neoplasm.9 Bilateral seminoma is rare, and is usually due to asynchronous tumors.7 Seminomas are typically confined by the tunica albuginea and rarely extend to paratesticular structures. Therefore, gross invasion of the spermatic cord or tunica albuginea should

prompt consideration of another tumor type, particularly lymphoma, which also tends to be relatively homogenous and hypoechoic. On MRI, seminomas characteristically appear isointense on T1-weighted images and predominantly lowsignal-intensity on T2-weighted images relative to testicular parenchyma. Following contrast material administration, septa within the seminoma typically enhance to a greater

Chapter 12 Imaging of the Scrotum and Penis 693

A

B Figure 12-5 Seminoma With Retroperitoneal Metastases A and B. Grayscale (A) and (B) color Doppler images of the testis demonstrate the presence of a relatively uniform hypoechoic intratesticular mass (asterisk) with through transmission (arrows in B). Although through transmission typically is a feature of a cyst, it can be seen in tumors characterized by a uniform population of cells, such as seminoma and lymphoma. The presence of color flow within the lesion excludes the possibility that this represents a complex cyst. Note dystrophic testicular calcifications (small arrows in A) and a second hypoechoic focus (arrowhead), also likely representing tumor. C. Axial image from an enhanced CT scan demonstrate aortocaval and left para-aortic bulky lymphadenopathy originating in the region of the left renal hilum and extending inferiorly (asterisk). IVC is difficult to visualize in C due to invasion.

C

degree than the tumor itself (see Figure 12-6).11 Similar to the US examination, the tumor is usually confined to the testicle, without extension into the surrounding tissues.

Spermatocytic seminoma: Spermatocytic seminoma should be differentiated from seminoma, which is a histologically and clinically distinct neoplasm. Unlike seminoma, spermatocytic seminoma does not arise from intratubular germ cell neoplasia. Affected individuals with spermatocystic seminoma are generally over the age of 65 years, whereas seminomas typically affect middle-aged men. Spermatocystic seminoma is a slow-growing tumor that rarely, if ever, produces metastases. In general, the prognosis for spermatocystic seminoma is excellent. The tumor appears similar to, but tends to be larger than, classic seminoma, but otherwise is indistinguishable from seminoma on the basis of imaging appearance.6

Nonseminomatous germ cell tumors (NSGCT): NSGCTs arise from a totipotent germ cell, which can remain undifferentiated, resembling embryonic stem cells, as in the case of embryonal carcinoma, or can differentiate into various lineages generating yolk sac tumors,

choriocarcinomas, and teratomas. Further, NSGCTs can either be composed of 1 histologic pattern or can be composed of multiple germ cell components, in which case they are referred to as a mixed germ cell neoplasm. Despite the histologic diversity of NSGCTs, the general therapeutic approach is similar among the different pathologic subtypes, and differs from that for seminoma. In general, NSGCTs are more biologically aggressive than seminoma, with nodal or other metastatic involvement at presentation in 60% to 75%. Additionally, NSGCTs are less radiosensitive than seminomas and are treated with various combinations of chemotherapy and lymphadenectomy, rather than radiotherapy.8 Consequently, the prognosis for NSGCT is not quite as good as for seminoma, but cure is still achievable in up to 80% even in the setting of metastatic disease and lymphadenopathy.10 Interestingly, unlike most other types of tumors, nonseminomatous germ cell tumor metastases can have histologic characteristics that are different from those of the primary testicular tumor, reflecting the totipotential nature of the germ cells. This can have implications for treatment.7 For example, an enlarging retroperitoneal mass following treatment can indicate development of mature teratoma in

694

Diagnostic Abdominal Imaging

A

B Figure 12-6 MRI of Seminoma Sagittal T2-weighted (A) and postgadolinium fat-saturated T1-weighted sequences (B) demonstrate the presence of a homogenous low-T2-signal intratesticular mass (asterisk) with enhancing septa (arrow), which is suggestive but not specific for seminoma. Arrowhead points to a signal void due to susceptibility artifact from metal.

the metastatic lymphadenopathy even if mature teratoma was not present in the primary tumor. This is referred to as “growing teratoma syndrome” and is important to differentiate from relapse,12 as the preferred treatment approach is surgical, since mature teratoma is generally poorly responsive to chemoradiation. Treatment for mature teratoma is aimed at reducing potential morbidity associated with local

invasion as well as reducing the risk of dedifferentiation into a malignant histology. The diagnosis of “growing teratoma syndrome” should be suspected if tumor markers are negative in the setting of enlarging cystic components in a retroperitoneal mass, and/or if the mass is not hypermetabolic on positron emission tomography (PET) scan.13 Persistence of disease at one or several foci despite regression at other foci can also be an indicator of “growing teratoma syndrome.” Mixed germ cell neoplasm is the most common type of NSGCT. Embryonal histology is the most common element occurring in up to 87%, with teratoma occurring in 50%, yolk sac elements in up to 44%, and choriocarcinoma in only 8% to 16%.1,7,9 Compared with seminoma, mixed germ cell neoplasms occur in younger men (average age of 30, compared with 40 for seminoma), tend to present at a more advanced stage, and are associated with elevated tumor markers in 80%, either β-hCG (due  to the choriocarcinoma component) or α-fetoprotein (due to yolk sac and rarely due to teratoma component) or both. Lactate dehydrogenase can also be elevated, and though not specific for testicular carcinoma does correlate with the bulk of the disease and is used in staging. The sonographic and MRI appearance of mixed germ cell tumors reflects the heterogeneous composition of these neoplasms. On US, they often have an inhomogeneous echotexture (71%), irregular or ill-defined margins (45%), echogenic foci (35%), and cystic components (61%) (see Figures 12-7 and 12-8). The cystic components can reflect true cysts due to teratoma, dilated rete testis, or areas of necrosis. Echogenic foci represent areas of hemorrhage, calcification, or fibrosis. Similarly, MRI usually reveals poorly marginated tumors that are inhomogeneous. In addition, MRI can identify the presence of fat indicative of a teratomatous component as a T1 hyperintense area that loses signal with fat suppression or as an area that loses signal on out-of-phase T1-weighted sequences because of the presence of lipid and water in the same voxel.11 Embryonal carcinoma accounts for 3% of tumors and tends to occur in men in their 30s.7 The tumor cells often stain with both β-hCG and α-fetoprotein.6 These tumor markers are elevated in 60% to 70% of cases of embryonal carcinoma. Further, US and MRI examinations typically demonstrate a small irregular mass with inhomogeneous echotexture and heterogeneous T1 and T2 signal, which contain cystic spaces in 20%.7 Embryonal carcinoma is, on average, smaller but more aggressive than seminoma. It is reported to invade the tunica albuginea in up to 20% to 25% of cases, which can be seen as distortion of the testicular contour on imaging examinations.9 Teratoma refers to a tumor composed of various cellular elements reminiscent of normal derivatives of more than 1 germ layer and can be categorized as mature or immature. Pure teratoma is relatively common in children and infants, second only to yolk sac tumors in frequency, but in adults constitutes only 2% to 3% of testicular neoplasms.

Chapter 12 Imaging of the Scrotum and Penis 695

A

B

C

Figure 12-7 Nonseminomatous Germ Cell Tumor A. This transverse US image demonstrates a large heterogeneous intratesticular mass with areas of necrosis (arrows), enlarging the left testis. B. Coronal T2-weighted and (C) postgadolinium T1-weighted images demonstrate a heterogenous mass (small arrows) in the left testicle with areas

of necrosis (asterisk). Note the rind of normal testicle superiorly (arrowhead). Note also the epididymis (curved arrows) and the normal right testicle (larger white arrow in B). Surgical excision demonstrated a nonseminomatous germ cell tumor composed of 80% seminoma and 20% embryonal tumor.

However, teratomatous elements are commonly found in mixed germ cell tumors. α-Fetoprotein is elevated in 38% and β-hCG in 25% of cases of pure teratoma.9 In children, both mature and immature teratomas typically have a benign behavior. However, in the postpubertal male teratomas are usually aggressive tumors and are treated as malignant regardless of whether the elements appear histologically mature or immature. Both mature and immature teratomas can metastasize, and the metastases can contain nonteratomatous elements.7 On US examinations, these tumors tend to be very large and markedly inhomogeneous, frequently containing echogenic foci that represent calcification, cartilage, immature bone, and fibrosis. Cystic

components are more commonly seen in teratomas than in other NSGCTs.1 On MRI, the presence of fat is suggestive of teratomatous elements. Yolk sac tumors, also known as endodermal sinus tumors, are the most common germ cell tumor in children where the prognosis is excellent. In adults, the pure form is rare; however, the presence of yolk sac elements in mixed tumors is common and is associated with a worse prognosis.1 Yolk sac tumors produce α-fetoprotein, which is elevated in 90% of tumors.6,9 Imaging features are nonspecific, especially in children where the only findings may be testicular enlargement without a well-defined mass. In adults, treatment does not differ from other NSGCTs;

A

B

Figure 12-8 Mixed Germ Cell Tumor A. Transverse US image shows replacement of the right testis by a heterogenous mass with necrotic areas (arrow). B. Sagittal T2-weighted image shows similar findings to A with arrows

C indicating fluid signal areas in keeping with necrosis and solid areas of tumor invasion (asterisk). C. Axial enhanced CT image through the retroperitoneum shows a single abnormal precaval lymph (arrows) node likely to represent nodal metastasis.

696

Diagnostic Abdominal Imaging

however, in children, if the tumor is confined and the α-fetoprotein is not elevated, treatment may consist only of orchiectomy followed by observation, with use of chemotherapy only if relapse occurs.7 Choriocarcinoma is a highly malignant neoplasm that, in its pure form, accounts for only 0.3% of testicular germ cell tumors. It typically develops in patients in the second and third decades of life; human chorionic gonadotropin is usually elevated and it causes gynecomastia in 10% of cases.7 Hematogenous metastases occur early and are frequently hemorrhagic, with common sites including lung, liver, gastrointestinal tract, and brain. Pure choriocarcinoma has the worst prognosis of all germ cell tumors, with death usually occurring within 1 year of diagnosis. Patients with mixed germ cell tumors with choriocarcinoma fare better than those with pure choriocarcinoma tumors, but a very high level of human chorionic gonadotropin (>50,000 IU/L) portends a poor prognosis, with a 5-year survival rate of 48%.7 On US and MRI, the primary tumor is usually hemorrhagic resulting in a mixed cystic and solid appearance.

Sertoli cell tumors: Sertoli cell tumors are the next most

Other testicular tumors

Epidermoid cyst: Epidermoid cysts are benign intrat-

Other testicular tumors include Leydig cell tumor, Sertoli tumors, epidermoid cysts, lymphoma, leukemia, and other rare neoplasms.

esticular masses with no malignant potential, constituting approximately 1% of testicular tumors. Like germ cell tumors, they usually present as painless palpable testicular masses. However, unlike germ cell tumors, epidermoid cysts can be treated with enucleation rather than orchiectomy because they are benign. Pathologically, they are cystic cavities with a fibrous wall at least partially lined with stratified squamous epithelium, and containing desquamated keratinized epithelium and keratin debris.7 Despite their cystic character, on US, epidermoid cysts are typically well-circumscribed masses measuring from 1 to 3 cm in diameter that appear solid, without through transmission (see Figure 12-11). This is a result of the complex content. The exact US appearance depends on the maturation, compactness, and quantity of keratin within the cyst, and consequently the internal echogenicity of the lesion varies from hypoechogenic with a few low-level internal echoes to hyperechogenic. Despite this variability, several US patterns have been described that are suggestive of an epidermoid. Of these, the most characteristic is the “onion-ring” appearance, in which there are alternating hyperechoic and hypoechoic layers representing layers of compact keratin and loosely dispersed desquamated squamous epithelium (see Figure 12-12).1 Another typical pattern is the “target” appearance, in which there is a central focus of increased echogenicity, representing an aggregate of keratin, surrounded by a hypoechoic region. Other epidermoids appear as hypoechoic masses with a hyperechoic or calcified rim. As expected with a cyst, there is no flow within epidermoid on Doppler evaluation. On MRI, epidermoid cysts also appear as sharply marginated nonenhancing masses, typically with a lowT2-signal wall and T2 hyperintense contents (representing the keratin). In some cases, epidermoids may demonstrate

Leydig cell tumors: Non–germ cell sex-cord stromal tumors account for 4% of all testicular tumors. Leydig cell tumors are the most common tumor in this group, accounting for 3% of all testicular tumors, mostly occurring in young children 3 to 6 years of age and in adults in the third to fifth decades of life. There is a reported association with Klinefelter syndrome.1 Approximately 30% of patients will have an endocrinopathy secondary to secretion of androgens or estrogens by the tumor.7 The endocrinopathy can manifest as isosexual pseudoprecocity in children, and as gynecomastia or impotence in adults.1 Approximately 10% of Leydig cell tumors are malignant. Malignant tumors typically occur in the elderly and do not have preceding symptoms or endocrinopathy. The appearance of Leydig cell tumors is varied. In 1 older surgical study of 40 cases, the tumors ranged in size from 0.5 to 10 cm with a mean of 3 cm, and were mostly sharply circumscribed except for 7 that had infiltrative margins. However, a relatively recent report of the US appearance of 10 cases described the tumors as ranging in size from 0.4 to 3 cm with most less than 1 cm. This same report found that none of the lesions were calcified, 90% were hypoechogenic, and that 88% showed peripheral hypervascularity.15 However, other imaging appearances have also been reported, including hyperechogenic tumors and tumors with diffusely increased vascularity. Few reports exist about the MRI appearance. In a review of 3 cases of Leydig cell tumor, it was found that marked enhancement of the tumor may favor a diagnosis of Leydig cell histology over conventional germ cell neoplasm (see Figure 12-9).16

common sex cord stromal tumor, accounting 1% of testicular tumors, and like Leydig cell tumors, are benign in approximately 90% of cases. Unlike Leydig cell tumors, sufficient hormone production to result in clinical endocrinopathy is rare. Most often the tumors appear as small, unilateral, and well-circumscribed, but otherwise nonspecific, masses. However, a subtype, the large cell calcifying Sertoli tumor, presents as multiple bilateral calcified masses.

Other stromal tumors of the testis: Other rare sex-cord stromal tumors include granulosa cell tumors, fibromathecomas, and mixed sex-cord stromal tumors (see Figure 12-10). Gonadoblastoma is a special tumor that also contains sex-cord-stromal elements but is distinguished by the presence of germ cells. Gonadoblastoma typically occurs in the setting of gonadal dysgenesis and intersex syndromes.7 These tumors are very rare, and the imaging appearance has not been well characterized in the radiology literature.

Chapter 12 Imaging of the Scrotum and Penis 697

A

B

C

D

Figure 12-9 Leydig Cell Tumor in 3 Patients A. Color Doppler US image shows a relatively homogenous, hypoechoic mass (arrow) with peripheral flow. B. Color Doppler image in a different patient demonstrates diffuse vascularity

throughout a small hypoechoic mass (arrow). C. T2-weighted and (D) T1-weighted postgadolinium axial images show a hyperenhancing mass (arrows). All 3 masses represented Leydig cell tumors at surgical excision.

a lamellated “onion ring”–type pattern on T2-weighted sequences or a “target” appearance (low-signal rim, hyperintense middle, and low-signal central focus), analogous to the US (see Figure 12-12). However, in other cases, the MRI signal characteristics are those of a nonspecific complex cyst (see Figure 12-11). Theoretically, the presence of lipid within an epidermoid might be detected on MRI as signal loss on out-of-phase images; however, this has not yet been reported. Note that although the onion ring appearance is characteristic of epidermoid, it is not pathognomonic as the appearance has also been reported with 2 cases of teratoma.15 Nonetheless, an onion ring appearance in association with negative tumor markers is still highly suggestive of epidermoid. In such cases, enucleation rather than orchiectomy may be considered, with the final surgical procedure determined by the pathologic diagnosis.

primary and only site of lymphoma. However, more commonly, lymphomatous involvement of the testicle is a manifestation of more generalized disease, either as a presenting site (which, in 10% of patients may be the only area of disease initially) or as a secondary site, and/or site of recurrence in patients with established lymphoma.19 Testicular lymphoma carries a poor prognosis with a 5-year survival rate of about 12% and a median survival time of less than 12 months.18,20 Clinically, testicular lymphoma is distinct from germ cell tumors in 2 major ways. First, it occurs in an older population, accounting for 50% of testicular neoplasms in men older than age 60  years.1 Second, although most patients present with painless testicular enlargement, in up to 25%, systemic symptoms, including weight loss, fever, and anorexia, have been reported as the initial complaint.7 On imaging, lymphoma often resembles seminoma, appearing as focal or diffuse homogeneous lesions that are hypoechoic on US and T2 hypointense on MRI, reflecting a uniform cell population and infiltrative nondestructive growth.18,20 Imaging features favoring lymphoma over

Lymphoma: Non-Hodgkin’s lymphoma accounts for 1% to 7% of all testicular tumors.18 The testicle may be the

698

Diagnostic Abdominal Imaging

A

B

Figure 12-10 Low-grade Spindle Cell Neoplasms A. Color and spectral Doppler image demonstrates a hypoechoic lesion with an echogenic border (arrow) and slight through transmission. The gray-scale appearance could indicate either a uniform solid mass or a complex cyst. However, arterial flow is documented on spectral Doppler indicating that this is a solid

mass. B. Coronal postcontrast image confirms this is a solid mass showing avid enhancement (arrow), a feature that raises the possibility of a stromal neoplasm. The mass was isointense to testicle on precontrast images and low signal relative to testicle on T2-weighted images (not shown). Surgical biopsy was diagnostic of a low-grade spindle cell neoplasm.

seminoma include (1) direct invasion of the epididymis and the spermatic cord; (2) indistinct as opposed to lobulated, well-defined margins; (3) preservation of normal testicular contour; and (4) bilateral testicular involvement.1,18,20 The imaging appearance in combination with the patient’s age at presentation, symptoms, and medical history may allow the interpreter to make the appropriate diagnosis of lymphoma.

common, occurring in up to 65% of men with acute leukemia and up to 35% of men with chronic leukemia in autopsy series.1 Clinically detectable disease is less common, reported in 1 study to be present in 8% of children with acute leukemia.21 The testis is a common location for recurrence because the blood–testis barrier allows leukemic cells to be “hidden” during chemotherapy such that the testis serves as a “sanctuary site.” Consequently, testicular involvement by leukemia is often found during bone marrow remission. The US appearance is similar to that of lymphoma. The diagnosis may be suspected in a child in bone marrow remission for leukemia who presents with a painless enlarging testis.22

Leukemia: Primary leukemia of the testis is very rare. However, secondary leukemic involvement of the testis is

Testicular metastasis: Testicular metastases are uncomImaging Notes 12-3. Distinguishing Features Between Testicular Lymphoma and Seminoma Lymphoma

Seminoma

Extension into the epididymis/spermatic cord

Confined to testis

Indistinct margins

Lobulated, well-defined margins

Normal testicular contour

Distortion testicular contour

Can be bilateral

Unilateral

Systemic symptoms

Local symptoms/ no symptoms

mon, occurring in right, remember to consider secondary varicocele

Torsion of appendix testis

Clinically, prepubertal males with “blue dot sign,” avascular mass with peripheral flow along superior pole testis

Splenogonadal fusion

Homogeneous mass with organized branching flow, associated congenital abnormalities, confirm with Tc99 sulfur colloid scan

Polyorchidism

Left > right, homogeneous circumscribed mass similar in appearance to testis, ± associated epididymis; visualization of mediastinum or tunica makes the diagnosis; sometimes bridging flow to ipsilateral testis

Fibrous pseudotumor

May be masslike or present as tunical thickening; on US, can mimic tumor. MRI showing avascular low-T2-signal mass may be helpful

Hematoma

History, other findings of trauma, resolution with time

Spermatic cord lipoma

Echogenic extratesticular mass; main differential is herniated omental fat, and rarely liposarcoma if soft-tissue is present

Malignant spermatic cord tumor

Usually sarcomas, often large and complex, findings of metastatic disease or rapid interval growth

Scrotal metastases

Rare, history of primary malignancy; can either be hematogenous or lymphatic or due to extension of carcinomatosis through patent processes vaginalis

Benign tumors and nonneoplastic masses

Adenomatoid, leiomyoma, hemangioma, lymphangioma, noncalcified scrotal pearls, sperm granuloma (when in the region of the vas deferens)

738

Diagnostic Abdominal Imaging

Table 12-11. Causes of a Diffusely Thickened Epididymis 1. Epididymitis a. Acute bacterial b. Chronic epididymitis c. Granulomatous inflammation i. Tuberculous epididymitis ii. Sarcoidosis 2. Tubular ectasia of the epididymis 3. Epididymal torsion 4. Spermatic cord torsion 5. Tumor a. Leukemia and lymphoma

causes of acute scrotal pain. The scrotal pain is characteristically relieved when the testis is elevated over the symphysis pubis, called the Prehn sign, a finding that may help to clinically differentiate epididymitis from torsion. Epididymitis usually is due to an ascending infection, with inflammation initially involving the tail and progressing to involve the rest of the epididymis. In 20% to 40% of cases, the infection will progress to involve the testis.1 In adolescents, the most commonly responsible organisms are Chlamydia trachomatis and Neisseria gonorrhoeae, with Treponema pallidum and other sexually transmitted diseases less likely, whereas in men older than age 45 and prepubertal males, the most common organisms are urinary coliform bacteria such as E coli and P mirabilis.2 Other uncommon infectious causes of epididymitis include granulomatous organisms such as tuberculosis (discussed in the next heading), as well as opportunistic infections in patients who are immunocompromised, including cryptococcus, candida, brucellosis, other fungi, and cytomegalovirus. Viral causes in the immunocompetent patient include mumps, coxsackie virus A, and varicella and echovirus. Viral infections can be suspected based on the absence of pyuria.2 In Asia, Africa, and the Western Pacific, Wucheria bancrofti, a filarial parasite, can be a cause of granulomatous epididymo-orchitis. Noninfectious causes of epididymitis include vasculitis (Behçet and Henoch-Schönlein purpura), trauma, amiodarone, and chemical epididymitis due to sterile reflux of urine through the vas deferens or due to a congenital anomaly. Epididymitis occurring before the age of 2 should prompt evaluation for predisposing factors, such as ectopic insertion of the ureter into the seminal vesicle, posterior urethral valves, voiding dysfunction, or bladder extrophy.75 Acute complications of epididymitis include epididymal or testicular abscess, pyocele, and testicular infarction or ischemia. Chronic complications include infertility, chronic pain, and testicular atrophy. Sonographically, in acute epididymitis, the epididymis is usually diffusely enlarged and demonstrates variable echogenicity from normal to hypoechoic as a result of edema, to hyperechoic because of the presence of hemorrhage (see

Figure 12-60). Occasionally, epididymitis can appear focal, as was discussed previously. Since reflux of urine via the vas deferens reaches the tail of the epididymis first, focal epididymitis involves the tail more commonly than the head of the epididymis. On color and spectral Doppler, the presence of increased blood flow has sensitivity for acute epididymitis approaching 100%. This is one of the most important imaging manifestation of acute epididymitis because 20% of cases of epididymitis and 40% of cases of orchitis demonstrate normal grayscale findings.1 Associated findings include skin thickening, hydrocele, and echogenic paratesticular fat. MRI is not typically obtained during epididymitis. However, on MRI, acute epididymitis usually appears as epididymal enlargement and hyperenhancement; in some case, epididymal signal may be heterogeneous because of hemorrhage or edema.76

Chronic epididymitis: Chronic epididymitis is characterized by scrotal pain lasting greater than 3 months. It most commonly is a result of granulomatous infection (discussed further below) but can be due to persistent inflammation after an episode of acute bacterial epididymitis. Other causes include drug-induced epididymitis, proximal obstruction following vasectomy, other obstructive phenomena, Behçet vasculitis, and idiopathic.2 The grayscale appearance of the epididymis in chronic epididymitis is similar to that seen in acute ependymitis, with diffuse thickening that may be associated with heterogeneity (see Figure 12-61). However, skin thickening is usually absent and hyperemia is often mild, reflecting the chronic nature of the disorder. The presence of calcifications also is suggestive of prior or chronic inflammation (see Figure 12-49).

Tuberculous epididymitis: Genitourinary tuberculosis is the most common location of extrapulmonary tuberculosis. When the genital organs are involved, the epididymis is the most common site of involvement and has been reported in 7% of those with tuberculosis in autopsy series. Epididymal tuberculosis can be due to spread of infection from the prostate or the kidney, or can be due to

Imaging Notes 12-18. Sonographic Findings of Epididymitis 1. Increased Doppler blood flow. Nearly 100% sensitivity 2. Usually diffuse involvement of epididymis. When focal, tail is the most common location. 3. Grayscale appearance of epididymis is variable, and may be NORMAL in 20% 4. Skin thickening 5. Hydrocele 6. Echogenic paratesticular fat

Chapter 12 Imaging of the Scrotum and Penis 739

A

B

C

D

Figure 12-60 Epididymitis This patient presented with acute scrotal pain. A. Sagittal grayscale and (B) color Doppler images demonstrate marked thickening, heterogeneity, and increased blood flow of the epididymal body and tail (arrows in A and B). Note small hydrocele (asterisk). C. Sagittal color Doppler and (D) transverse

grayscale images of the region of the epididymal head demonstrate a thickened hypervascular epididymis (arrow), complex hydrocele with septations (asterisk), and increased echogenicity of the paratesticular tissue (arrowheads in D). These findings are in keeping with acute epididymitis with an probable associated pyocele.

hematogenous spread. Rarely, epididymal tuberculosis can result from sexual transmission or acquired from intravesical BCG therapy for bladder cancer. In patients with genital tuberculosis, pulmonary disease can be documented in 50% of patients, and renal tuberculosis can be documented in 80% to 85% of patients.77 Tuberculous epididymitis can be either painless or cause scrotal pain. Tuberculous involvement usually results in diffuse enlargement of the epididymis. This enlargement can be uniformly hypoechogenic, heterogeneously hypoechogenic, or with multiple hypoechogenic lesions.26 The heterogeneously hypoechoic patterns of epididymal enlargement are believed to reflect various stages of fibrosis, necrosis, and granulomas and calcifications (see

Figure 12-62).85 The heterogeneously hypoechoic pattern of epididymal enlargement favors a diagnosis of tuberculous over bacterial epididymitis. Diffusely increased blood flow throughout the epididymis is seen in subjects with bacterial epididymitis, whereas focal linear or spotty blood flow within the periphery of the affected epididymis is typical of subjects with tuberculous epididymitis.24 Additionally, tuberculous epididymal abscesses are reported to demonstrate a milder degree of peripheral hypervascularity compared with bacterial abscesses.24 Other features that favor a diagnosis of tuberculous over bacterial epididymitis include sinus tracts, a miliary pattern of testicular orchitis, and a history/diagnosis of tuberculosis at other sites in the body.

740

Diagnostic Abdominal Imaging

A

B

Figure 12-61 Chronic Epididymitis This patient presented with a palpable abnormality of the testis. A. Coronal T2-weighted and (B) postgadolinium fat-saturated

T1-weighted images show a diffusely heterogenous epididymis, in keeping with chronic inflammation.

Sarcoidosis

testis, where involvement is reportedly more commonly bilateral than in the epididymis. Combined testicular and epididymal nodules may be produced by other processes, including granulomatous infection and certain neoplasms, such as lymphoma (see Tuberculous Orchitis and Other Granulomatous Orchitis under Focal Testicular Disorders). Potential clues include African American race, history of sarcoid and elevated serum ACE levels, and in some cases response to steroids.2,14

Sarcoidosis is a systemic disease characterized by formation of noncaseating granulomas in multiple origins. The prevalence of genital involvement on postmortem series in 5%; however, clinically diagnosed disease is present in less than 0.5%.1,79 Of the reported cases of genitourinary sarcoid, 75% involve the epididymis and 50% involve the testis.80 The epididymal involvement can be bilateral in up to one-third of cases.81 The patients can present with systemic symptoms related to sarcoidosis, or with findings referable to the scrotum, including pain and palpable abnormality. On imaging, the epididymis usually is enlarged and heterogeneous with hypoechogenic nodules. Hypoechogenic nodules can also be present in the

A

B

Figure 12-62 BCG Epididymitis This patient had received BCG therapy for bladder cancer and presented with a palpable abnormality in the scrotum. A. Sagittal grayscale and (B) color Doppler images of the epididymal body and tail and (C) grayscale image of the testis demonstrate irregular thickening of the epididymal body and tail with focal

Tubular ectasia of the epididymis Tubular ectasia of the epididymis is believed to be due to upstream obstruction. It most commonly occurs as a sequela of vasectomy, where it is reported to occur in

C areas of decreased echogenicity (arrows in A) and increased vascularity (arrow in B). There are tiny hypoechoic lesions (arrows in C) in the testis. This appearance, in combination with the lack of acute symptoms and history of BCG treatment, is most consistent with chronic granulomatous epididymal and testicular inflammation due to BCG therapy.

Chapter 12 Imaging of the Scrotum and Penis 741

A

B

Figure 12-63 Tubular Ectasia of the Epididymis This patient had undergone previous vasectomy. (A) Grayscale and (B) color Doppler images demonstrate a thickened epididymal body and tail which have a somewhat spongiform appearance. This appearance is typical of tubular ectasia, a

finding that can be seen postvasectomy. This phenomenon can be differentiated from other causes of diffuse epididymal thickening by the lack of acute symptoms, spongiform appearance, and lack of increased blood flow.

43%.82 Tubular ectasia results in an enlarged epididymis that can be differentiated from other causes of epididymal enlargement by (1) spongiform appearance due to visualization of the dilated epididymal tubules, (2) lack of increased vascularity, and (3) history of vasectomy (see Figure 12-63). Other findings commonly found postvasectomy include sperm granuloma (discussed previously under the heading Focal Epididymal Disorders), spermatocele, dilation of the vas deferens, and tubular ectasia of the rete testis, in addition to tubular ectasia of the epididymis.83 Sometimes, the vasectomy clip may be identified by US.

Diffuse epididymal neoplasm: leukemia and lymphoma

Torsion of epididymis

Unique scrotal lesions include cystic lesions of the scrotum, causes of scrotal calcifications, and scrotal gas.

Isolated epididymal torsion, without associated testicular torsion, is rare. Causes include an anomalous attachment of the epididymis to the testis or a long epididymis with a long meso-orchium.2 Such anomalies are more common in patients who have undescended testes. Isolated torsion of the epididymis can be suspected if (1) the body and tail of the epididymis are markedly enlarged and heterogeneous with little or no blood flow detected by Doppler imaging, (2) the head of the epididymis is hypervascular, with a whorled appearance of the vessels indicating the site of torsion, (3) the testicle is normal in vascularity, and (4) the patient has acute scrotal pain.

Spermatic cord torsion Spermatic cord torsion can result in an enlarged epididymis, but by contrast to epididymitis the vascularity is decreased (see Figure 12-64). See also section Torsion of the Epididymis.

Leukemia and lymphoma usually produce focal lesions, but can rarely result in a diffusely enlarged, hypoechogenic, hypervascular epididymis. However, isolated involvement of the epididymis by lymphoma and leukemia is rare, and in most cases the ipsilateral testicle will also be extensively involved by lymphoma.

UNIQUE SCROTAL LESIONS

Cystic Lesions in the Scrotal Sac Fluid within the scrotum will most often be due to a hydrocele or an epdidymal cyst but can occasionally be a result of herniation of fluid or fluid containing structures, and rarely can be caused by mesothelioma of the tunica vaginalis (Table 12-12).

Simple and complicated hydroceles Serous fluid, blood, pus, lymphatic fluid, or urine can accumulate between the layers of the tunica vaginalis resulting in a hydrocele, hematocele, pyocele, and rarely lymphocele or urinoma, respectively. Simple hydrocele is the most common cause of asymptomatic scrotal swelling and can be a congenital or acquired abnormality. Congenital hydroceles are usually seen in infants and are due to a patent processus vaginalis

742

Diagnostic Abdominal Imaging

Imaging Notes 12-19. Differentiating Features of Diffuse Epididymal Enlargement Diffusely thickened epididymis Entity

Helpful Differentiating Features

Acute epididymo-orchitis

Pain, hyperemia, hydrocele, scrotal wall thickening, associated increased testicular vascularity, resolution with treatment

Chronic epididymo-orchitis

Chronic pain, thickening without hyperemia, diagnosis of exclusion

Tuberculous epididymo-orchitis

Systemic findings of TB, heterogeneous hypoechoic enlargement with only mildly increased vascularity, calcification, sinus tracts, miliary involvement of testicle

Tubular ectasia

History of vasectomy, spongiform appearance without internal vascularity; other postvasectomy findings

Torsion of the epididymis (rare)

Acute scrotum with enlarged but hypovascular epididymis, normal testis, whorled vessels near enlarged epididymal head

Secondary involvement with tumor

For lymphoma, epididymal involvement is usually due to infiltration from testis

Sarcoid

Systemic findings of sarcoid, heterogeneous hypoechogenic enlargement, elevated ACE, ± testicular involvement

resulting in collection of peritoneal fluid in the scrotal sac. Most cases of congenital hydrocele will resolve by 18 months of age, but in about 15% a patent processes vaginalis may persist into adulthood.14 The presence of calcifications complicating a congenital hydrocele should suggest the diagnosis of meconium periorchitis. This is a sequela of meconium peritonitis, which is a chemical peritonitis due to intrauterine bowel perforation. In adults, most simple hydroceles are acquired, occurring as a reaction to intrascrotal pathology such as tumors

or infections. However, in some cases no cause for the acquired hydroceles can be identified. For both acquired and congenital hydroceles, when the hydrocele is small, the fluid tends to be located anterolateral to the testis. When large, intravaginal fluid collections can compress the testis, resulting in vascular compromise, which is manifested as diminished flow on color Doppler US or abnormal spectral Doppler waveforms. On US, a hydrocele is usually anechoic but may contain a few low-level echoes due to a small amount of protein

A

B

Figure 12-64 Epididymal Thickening Due to Spermatic Cord Torsion A. Grayscale and (B) color Doppler images in a patient with acute scrotal pain demonstrate an enlarged epididymis and a

hydrocele. However, the testicle and the epididymis demonstrate decreased blood flow. These findings should prompt consideration of spermatic cord torsion as the etiology, rather than epididymo-orchitis, which would have increased blood flow.

Chapter 12 Imaging of the Scrotum and Penis 743 Table 12-12. Large Cystic Lesions in the Scrotal Sac 1. Fluid in the sac of the tunica vaginalis a. Hydrocele b. Hematocele c. Pyocele d. Lymphocele e. Scrotal urinoma 2. Epididymal cyst and spermatocele 3. Inguinal hernia a. Ascitic fluid b. Herniated fluid-filled bowel c. Herniated bladder 4. Mesothelioma of the tunica vaginalis

or cholesterol content (see Figure 12-65).1 On MRI examinations, a simple hydrocele appears as simple fluid surrounding the testis. On both imaging studies, congenital hydrocele can be differentiated from acquired hydrocele by demonstrating a patent processus vaginalis, seen as fluid tracking from the peritoneal cavity through the inguinal region into the sac of the tunica vaginalis. A hydrocele can appear complex because of a variety of reasons, including (1) chronic or prior inflammation, (2) the presence of blood (hematocele), (3) the presence of pus and infectious debris (pyocele), (4) rarely due to complex peritoneal fluid extending through a patent processes vaginalis into the scrotal sac, or primary malignancy of the tunica vaginalis (see Mesothelioma of the tunica vaginalis below). Complex hydroceles from all of these of causes have

A

B

Figure 12-65 Hydrocele and Pyocele, in Different Patients A. Grayscale US image demonstrating simple fluid (asterisk) surrounding and compressing the epididymis (arrowhead) and testis (arrow) characteristic of a large hydrocele. B. Enhanced post contrast fat-saturated T1-weighted MRI image in a different

a similar sonographic appearance, demonstrating variable amounts of internal echoes, fluid-debris levels, septations, and loculations (see Figure 12-65). However, different etiologies can sometimes be distinguished based on other factors, as follows. Hematoceles are usually associated with a history of recent trauma and other imaging findings of scrotal injury. Pyoceles typically present with fever and leucocytosis, and are associated with other imaging findings of infection, such as epididymo-orchitis. Pyoceles occasionally can contain gas, suggesting infectious nature of the collection. On the other hand, complex hydrocele in an asymptomatic patient most likely represents sequela of prior inflammation. Rarely, a rapidly enlarging complex hydrocele can reflect extension of carcinomatosis into the scrotal sac through a patent processus vaginalis or a mesothelioma of the tunica vaginalis, in which case peripheral nodularity may be present. MRI can also occasionally add specificity by demonstrating T1 hyperintense fluid representing blood products in the case of a hematocele or by revealing the presence of enhancing nodules lining the scrotal sac in the case of malignancy (see Figures 12-66 and 12-67). Intravaginal collections of lymphatic fluid can be due to congenital lymphedema or can be due to prior surgery. History is key to the diagnosis in this situation as the imaging appearance is not specific. Likewise, urinoma is indistinguishable based on imaging from a simple hydrocele, but might be suspected if there is new scrotal swelling in the setting of urinary tract injury.

Epididymal cyst and spermatocele Epididymal cysts and spermatoceles are both believed to result from dilation of the epididymal tubules. Epididymal

C patient demonstrates fluid (asterisk) surrounding the testis (arrow) and epididymis (arrowhead) typical of a simple hydrocele. C. Grayscale image in a third patient with epididymo-orchitis demonstrates a scrotal fluid collection (asterisk) containing multiple lacy septations, suggestive of a pyocele.

744

Diagnostic Abdominal Imaging

A

B

Figure 12-66 Hematocele This patient presented with a history of testicular trauma. A. Axial T1 and (B) T2-weighted images demonstrate T1 hyperintense fluid

(asterisk) in keeping with hemorrhage with associated blood clot (small black arrow) surrounding the testis (T), in keeping with hematocele. White arrows point to scrotal edema.

cysts and spermatoceles are differentiated on the basis of the presence of sperm in a spermatocele. Together, these lesions are reported to occur in 29% of asymptomatic men.84 Epididymal cysts and spermatoceles have been reported more frequently in men who have undergone vasectomy or have had intrauterine exposure to diethylstilbestrol, or who demonstrate dilation of the rete testis. Spermatoceles are only found in the epididymal head whereas epididymal cysts can be found throughout the epididymis. Epididymal cysts and spermatoceles

are otherwise indistinguishable by US, and in general, the distinction is not clinically important. Both lesions appear on US as cystic masses within the epididymis (see Figure 12-68). These can appear simple or can contain loculations, septations, and/or internal echoes. When large, they can be difficult to distinguish from a large hydrocele. However, epididymal cysts and spermatoceles tend to displace the testis, by contrast to hydroceles, which envelop the testis. This feature may suggest the correct diagnosis.14

A

B

Figure 12-67 Scrotal Metastases This patient had known peritoneal carcinomatosis from gallbladder carcinoma. A. Coronal T2-weighted and (B) coronal fat-saturated T1-weighted postgadolinium sequences

reveal multiple small enhancing nodules along the tunica vaginalis (arrows) as well as a small hydrocele reflecting scrotal involvement by peritoneal spread through a patent process vaginalis.

Chapter 12 Imaging of the Scrotum and Penis 745

A

D

B

E

C

F

Figure 12-68 Epididymal Cysts in 4 Patients A. Sagittal grayscale image through the epididymal head demonstrates an anechoic mass with a well-defined wall and through transmission characteristic of a simple epididymal cyst. A spermatocele can have an identical appearance. B. Transverse grayscale image demonstrates tubular ectasia of the rete testis (arrowhead), which is commonly associated with epididymal cysts (arrow). C. Grayscale image shows a large fluid collection in the scrotum. This appears to have a wall (arrows) and the testis is located in a nondependent location, features that are suggestive of a large epididymal cyst

as opposed to hydrocele. D. Coronal T2-weighted image of the same patient as in C more clearly shows the cyst (arrows) as well as a small hydrocele (asterisk). The testis is displaced inferiorly to the left of the image. E. Axial T2-weighted and (F) axial T1-weighted images in a patient referred for evaluation of a solid-appearing epididymal mass demonstrates a mass with T1 hyperintensity (arrows) in the region of the epididymal head. This did not enhance postcontrast (not shown), indicative of an epididymal cyst complicated by protein or hemorrhage. This illustrates how MRI may be useful in cases where US is unable to determine if a mass is cystic or solid.

Inguinal hernia

frequently with the indirect type. A diagnosis of strangulation should be suspected on US if (1) the herniated bowel is akinetic and hyperemic or demonstrates wall thickening, (2) the scrotal wall is hyperemic, (3) the loop of bowel is not reducible, or (4) there is acute scrotal pain.1

Inguinal hernias may be classified as direct or indirect inguinal hernias. Both types of inguinal hernias can contain fluid-filled bowel or urinary bladder, each of which can be seen as an intrascrotal fluid collection (see Figure 12-57). The diagnosis of a hernia is made by examination of the inguinal region, which demonstrates communication of the herniated contents through the inguinal canal into the pelvic cavity. Valsalva maneuver can be helpful in eliciting an otherwise occult hernia. Most inguinal hernias are asymptomatic and discovered as a groin bulge on physical examination. However, inguinal hernias can present with acute scrotal pain if bowel is incarcerated and strangulated, which occurs more

Mesothelioma of the tunica vaginalis Mesothelioma of the tunica vaginalis is a rare etiology of recurrent or enlarging hydrocele, with fewer than 100 cases reported in the literature (see Figure 12-69).85 Scrotal mesothelioma arises from the tunica vaginalis and accounts for 0.2% to 5% of cases of mesothelioma. Asbestos exposure is a risk factor, but reported in less than 50% of cases.14 Prospective diagnosis of scrotal

746

Diagnostic Abdominal Imaging A patent process vaginalis containing ascites and tumor nodules related to peritoneal carcinomatosis can produce a similar appearance (see Figure 12-67).

Scrotal Echogenic Foci In most cases, echogenic foci within the testis will be a result of dystrophic calcifications but can occasionally be a result of intratesticular gas or foreign material like sutures or surgical clips deposited in the scrotum during surgery. Rarely, echogenic foci within testicular neoplasms can be due to focal areas of hemorrhage or fibrosis.1 Calcifications can be associated with both malignant and benign conditions (Table 12-13).

Calcifications and other echogenic foci associated with testicular neoplasms Viable neoplasms: A variety of testicular neoplasms can

Figure 12-69 Scrotal Mesothelioma Grayscale image reveals a large complex hydrocele with fine internal echoes and internal strands (arrows), without solid components or internal vascularity. This was thought to represent a hematocele or pyocele but found to represent scrotal mesothelioma. Epididymal head indicated by arrowhead.

mesothelioma is difficult. However, the diagnosis can be suspected in the setting of a recurrent or enlarging hydrocele associated with soft-tissue nodules along the tunica vaginalis. Mesothelioma is an aggressive tumor with metastatic disease found in 15% of patients at presentation, most commonly in retroperitoneal lymph nodes.85

contain echogenic foci, the most common of which are NSGCT, which contain echogenic foci in 35% of cases. Echogenic foci in NSGCT can be due to hemorrhage, fibrosis, or dystrophic calcification and in tumors containing teratomatous elements can also be due to the presence of bone or cartilage (see Figure 12-70).1 Other rare neoplasms with intralesional calcifications include large-cell calcifying Sertoli tumor, carcinoid tumor, osteosarcoma, epidermoid, and hemangioma of the testis.

Burned-out germ cell tumor: Rarely, a small fibrotic mass, without evidence of malignant cells, will be discovered within the testicle of a patient with evidence of metastatic germ cell tumor. It is believed that this fibrotic mass represents the primary testicular malignancy that has autoinfarcted as a result of such rapid growth that the malignancy outstripped its blood supply. This phenomenon has been termed burned-out germ cell tumor and occurs most commonly with teratocarcinoma or choriocarcinoma. The US appearance of burned-out germ cell tumor is variable, ranging from a single area of macroscopic calcification to a small hypoechogenic mass.86

Imaging Notes 12-20. Differentiating Features of Fluid in the Scrotal Sac Large amount of fluid in the scrotal sac Fluid in Scrotal Sac

Differentiating Features

Hydrocele, hematocele, pyocele

Surrounds and envelops the testis and epididymis, often anterolateral in location, can compress the testis

Epididymal cyst/spermatocele

Displaces, as opposed to envelops the testis (compared to hydrocele)

Hernia sac with ascites or bowel

Communication with pelvic cavity via inguinal canal; peristalsis

Mesothelioma

Recurrent or enlarging complex hydrocele with nodules along the tunica vaginalis

Scrotal metastases

Mixed cystic and solid masses associated with the hydrocele

Chapter 12 Imaging of the Scrotum and Penis 747 Table 12-13. Causes of Scrotal Echogenic Foci A. Calcifications 1. Associated with neoplasms a. More common i. Nonseminomatous germ cell tumor (most common) ii. Burnt-out germ cell tumor iii. Epidermoid b. Very rare i. Large-cell calcifying Sertoli tumor ii. Hemangioma iii. Carcinoid tumor iv. Osteosarcoma 2. Calcifications associated with benign conditions a. Dystrophic calcifications i. Prior infarction ii. Prior trauma iii. Prior abscess iv. Granulomatous diseases (tuberculosis, sarcoidosis) b. Testicular microlithiasis (benign, but associated with ITGCN)

Dystrophic macrocalcifications: Benign causes of coarse testicular calcifications include prior infarction, trauma, or abscess and previous granulomatous disease, including sarcoidosis and granulomatous infections. (see Figure 12-71). One unusual posttraumatic cause of dystrophic calcifications is chronic repetitive trauma in equestrians and mountain bikers.87,88 This usually results in bilateral testicular calcifications measuring a few millimeters in size. Although relatively small, they are larger in size than the calcifications of microlithiasis. Rimlike testicular calcification can be seen in the setting of prenatal torsion. This rim calcification is associated with testicular atrophy and thickening of the tunica albuginea.89 Vascular calcifications may also occasionally be identified as a cluster of tiny calcifications.90

Testicular microlithiasis: Testicular microliths are

Calcifications associated with benign conditions can be macroscopic, typically due to prior insult, or tiny punctate calcifications representing microlithiasis.

laminated calcium deposits within the lumen of the seminiferous tubules, which on US appear as nonshadowing echogenic foci measuring less than 2 to 3 mm.91 These are differentiated from macrocalcifications by their small size. The presence of 5 or more microliths per transducer field has come to be accepted as the definition for classic testicular microlithiasis (CTM), with fewer microliths representing limited testicular microlithiasis (LTM) (see Figure 12-72).92 Management of microlithiasis remains controversial because of the uncertain causal relationship between CTM and germ cell neoplasia. Early retrospective studies reported a low prevalence of CTM (0.6% of the general population), associated with a fairly high prevalence of GCT (40% in patients with CTM),94 with a relative risk of testicular GCT of up to 21.6 comparing those with CTM with the general population,95 suggesting a strong relationship between CTM and GCN. However,

A

B

Figure 12-70 Calcifications in Nonseminomatous Germ Cell Tumors A. Grayscale sonographic image demonstrates a coarse shadowing calcification (arrow) associated with a hypoechoic mass (arrowheads). Orchiectomy revealed a mixed germ cell

tumor composed of seminoma and yolk sac components. B. Grayscale sonographic image in a different patient shows a hypoechoic mass with a coarse calcification (arrow). Orchiectomy revealed mixed germ cell tumor composed of mostly seminoma with a small component of embryonal cell carcinoma.

B. Intratesticular gas (postoperative) C. Iatrogenic material (sutures, surgical clips)

Calcifications associated with benign conditions

748

Diagnostic Abdominal Imaging development.92 Long-term large prospective studies have not yet been performed. At this time, it seems reasonable to consider patients with testicular microlithiasis as having an increased risk of developing a primary testicular tumor and to offer some kind of surveillance.96,102 A combination of annual US and periodic self-examination has been suggested for patients at increased risk, such as men with testicular microlithiasis who are subfertile or have a history of prior GCT or who have cryptorchidism.96 There is no consensus regarding management of patients with no risk factors other than testicular microlithiasis, and the decision regarding the type of surveillance, either self-examination or annual US, should be decided on an individual basis.96

Figure 12-71 Dystrophic Calcifications This patient was asymptomatic. Sonogram shows a coarse echogenic focus with shadowing indicating dystrophic calcification without known cause. Note the absence of associated mass or adjacent testicular abnormality.

later studies reported a higher prevalence of CTM of 1.7% in an asymptomatic population and 3.7% to 9% in a referred population and also reported a lower prevalence of testicular neoplasia of between 5.8% and 18% for CTM and between 0% and 5.8 % for LTM when compared with 0.3% to 0.7% in the control population.92,98-101 The data from these newer studies indicates that the association of testicular microlithiasis with malignancy is likely not as strong as originally thought. Furthermore, while there have been case reports of testicular carcinoma developing in the setting of microlithiasis,91,102 other studies with short-term follow-up have shown no tumor

Figure 12-72 Microlithiasis Ultrasonographic examination demonstrates innumerable tiny bilateral nonshadowing echogenic intratesticular foci characteristic of microlithiasis. These are not visible by MRI.

Intratesticular gas Intratesticular gas can be due to iatrogenic intervention, penetrating trauma, or due to a gas-forming infection. Features suggestive of gas on US examinations include ring-down artifact, dirty shadowing, nondependent location, and mobility (see Figure 12-73).

Iatrogenic material Regularly spaced nonshadowing echogenic foci can represent suture material, usually in the setting sperm harvesting procedure (see Figure 12-74).

Figure 12-73 Intratesticular Gas This grayscale US image shows multiple echogenic foci (arrows) in a diffusely heterogeneous testicle (outlined by thin arrows). On grayscale imaging, this could mimic a mass with calcifications; however, the patient had a history of gunshot to the scrotum and the testicle was avascular with tunica disruption (not shown), and therefore the echogenic foci represent gas due to the penetrating injury.

Chapter 12 Imaging of the Scrotum and Penis 749 Imaging Notes 12-21. Differentiating Features of Intratesticular Echogenic Foci Intratesticular Echogenic Foci Entity

Differentiating Features

Burnt-out germ cell neoplasm

History of germ cell neoplasm/retroperitoneal LAN, +tumor markers

NSGCT

Soft-tissue component, +tumor markers, retroperitoneal LAN

Calcifying Sertoli cell tumor

May present as bilateral calcified masses, pediatric age group, ass with Peutz-Jaeger and Carney syndrome

Other rare calcifying tumors (carcinoid, osteosarcoma)

Rare, would be difficult to diagnose prospectively

Macrocalcifications (chronic trauma, hematoma, abscess, granulomatous disease, infarct)

Negative tumor markers, no soft-tissue component, history of repetitive trauma

Intrauterine spermatic cord torsion

Rim calcified atrophic testicle in a neonate

Microlithiasis

Tiny punctate calcifications, nonshadowing; association with testicular cancer

Intratesticular gas

History of surgery, instrumentation, infection, dirty shadowing, ring down, mobility of echogenic foci

Iatrogenic

Regularly spaced, history of testicular biopsy or sperm harvesting

Extratesticular Echogenic Foci Extratesticular echogenic foci can be due to calcifications, gas, or foreign material (Table 12-14).

Extratesticular calcification Most extratesticular calcifications are benign findings due to chronic or prior inflammation, and include scrotal

pearls, and calcification of the tunica vaginalis, appendix testis, and epididymis. These are differentiated by location of the calcification. The most common type of extratesticular calcification is a “scrotolith” or “scrotal pearl,” which is a calcified

Table 12-14. Extratesticular Echogenic Foci 1. Extratesticular calcification a. Scrotolith (“scrotal pearl”) b. Prior infarction of the appendix testis c. Calcification of the tunica vaginalis or tunica albuginea d. Epididymal calcifications i. Chronic epididymitis ii. Sperm granuloma

Figure 12-74 Suture Material Grayscale image shows regularly spaced echogenic foci in a linear distribution (arrows) typical of sutures related to a recent sperm harvest. Note the large complex hydrocele (asterisk) with dependent debris representing a postoperative hematocele.

2. Gas a. Forniers gangrene b. Bowel containing hernia; gas may be extraluminal if perforation has occurred c. Prior recent surgery d. Scrotal abscess e. Fistula from the genitourinary or gastrointestinal tracts f. Dissection of subcutaneous gas from the thorax 3. Foreign bodies a. Prior penetrating trauma b. Prior surgery

750

A

Diagnostic Abdominal Imaging

B

C

Figure 12-75 Extratesticular Calcifications A. Grayscale US image demonstrating a coarse shadowing calcification in the dependent scrotal sac, which was mobile during real-time examination typical of a scrotolith (a.k.a scrotal pearl) B. Grayscale US image in a different patient demonstrating a nonmobile, linear calcification with posterior

acoustic shadowing (arrow) in a nondependent location intimately associated with the tunica (small arrows) characteristic of a tunica albuginea calcification. C. Grayscale US image from a patient with history of prior vasectomy, demonstrating echogenic foci in the epididymal head (arrow) with shadowing, indicating dystrophic calcification related to the prior surgery.

loose  body lying between the membranes of the tunica vaginalis. It is believed to originate from either a fibrinous deposit in the tunica vaginalis or as a remnant of a detached torsed appendix testis or appendix epididymis. Scrotoliths can occasionally be detected by physical examination but in most cases are not palpable because of an associated hydrocele. On US examinations, a scrotolith appears as a dependently located mobile small round or oval mass external to the testis, which can be completely calcified or may contain central calcifications (see Figure 12-75). In most cases, there will be an associated hydrocele. Calcification of the appendix testis can be recognized by its characteristic position near the superior pole of the testis, and is often secondary to prior appendiceal torsion. Dystrophic calcification of the tunica albuginea or tunica vaginalis can be recognized on the basis of a linear or plaquelike configuration along the surface of the testis or along the scrotal wall, respectively and is usually from prior inflammation (see Figure 12-75). Calcification within the epididymis can be on the basis of prior inflammation in an asymptomatic patient, but can in some cases reflect chronic/ granulomatous epididymitis and sperm granuloma (see Figure 12-75). (See Chronic epididymitis and Tuberculous epididymitis under Diffusely Thickened Epididymis and Sperm Granuloma under Focal Epididymal Disorder.) Phleboliths can occur within intrascrotal, extratesticular hemangiomas and are one additional very rare cause of paratesticular echogenic foci.

with position, “ring-down” artifact, and “dirty-shadowing.” Gas can be confirmed by CT or plain-film radiography. The patient’s clinical history and the location of the gas within the scrotum can aid in differentiating among different etiologies of scrotal gas. Necrotizing gangrene of the perineum and genitalia, called Fournier’s gangrene, is a severe infection of the perineum that is associated with underlying illnesses, including diabetes, renal failure, alcohol abuse, and perirectal disease. It is usually a result of polymicrobial infection and is most commonly caused by anaerobic cocci. The mortality rate is high. Patients will typically have fever, leucocytosis, and perineal pain and will frequently have crepitus on physical examination. Imaging examinations will demonstrate diffuse gas within the scrotal wall and surrounding perineum (see Figure 12-76).97 Inguinal hernia can also present with acute scrotum and extratesticular gas; however, unlike in Fournier gangrene, the gas is localized within bowel loops in the hernia sac, unless perforation has occured. Rarely, gas within the skin of the scrotum can be due to extensive subcutaneous gas dissecting from an air leak in the thorax or due to dissection of gas due to a perforated abdominal viscus. Other etiologies of scrotal gas include organized abscess, fistula from the bowel in the abdomen or pelvis, fistula from the urethra, and gas introduced by surgery or trauma.

Gas Causes of extratesticular scrotal gas include Fournier gangrene, inguinal hernia, gas dissection from an air leak in the thorax, abscess, fistula, and recent prior surgery. Features suggesting gas include mobility as indicated by a change

Foreign bodies Shrapnel from penetrating injury or surgical clips appear as highly echogenic foci with shadowing and ring-down artifact. In most cases, this appearance plus a clinical history of penetrating trauma or prior surgery will identify the cause of the echogenic foci. If necessary, radiography can be performed to confirm the metallic composition of the echogenic focus.

Chapter 12 Imaging of the Scrotum and Penis 751

Figure 12-76 Fournier Gangrene This 76-year-old man complained of increasing scrotal pain for 3 days. Images show thickening of the perineum with associated stranding and foci of gas (arrow), suspect for Fournier gangrene.

Other Echogenic foci are commonly present in the epididymis postvasectomy, representing either fibrosis, calcification, or clumps of sperm. Occasionally, the clumps of sperm may be seen to move in real time in the tubules, which has been termed “dancing megasperm” (see Figure 12-77).29

Figure 12-77 Dancing Megasperm This grayscale image demonstrates marked thickening and tubular ectasia of the epididymal tail (between large arrows) with several echogenic foci (small arrows) that were seen to move in the dilated tubules in real time; this has been referred to as “dancing megasperm.”

be obtained. On US, the cryptorchid testis is usually small and hypoechogenic and can be difficult to differentiate from a lymph node or from the distal bulbous portion of the gubernaculum testis.23 Identification of the echogenic mediastinum testis is helpful in confirming that a structure represents the testis.1,23

Nonpalpable Testis A testis can be nonpalpable due to resection, congenital absence, atrophy, retractile location or cryptorchidism, or dislocation due to trauma. The testis descends from the abdomen into the scrotum at 36 weeks’ gestation. Cryptorchidism is defined as complete or partial failure of the testis to descend into the scrotal sac, which is found in 3.5% of newborns, decreasing to 1% at 1 year of age.19 Complications associated with cryptorchidism include infertility and increased risk of testicular cancer, usually seminoma. The estimated rate of testicular cancer was initially reported to be as much as 50 times greater than the general population; however, more recent studies indicate a risk of 2.5- to 8-fold. Orchiopexy is usually performed between ages 1 and 10 and does not change the risk of malignant degeneration but does facilitate surveillance. US is useful for identifying ectopic testicles located in the inguinal canal (72%) or in a prescrotal location, just distal to the external inguinal ring, but US is usually unable to detect intraabdominal testicles (8%) due to overlying bowel gas (see Figure 12-78).1 In cases where the testicle is not identified sonographically, MRI should

Imaging Notes 12-22. Differentiating Features of Extratesticular Echogenic Foci Extratesticular Echogenic Foci

Differentiating Features

Scrotal pearls

Mobile masses in the scrotal sac

Other extratesticular calcifications

Location (most commonly along the tunica albuginea or in the epididymis), usually due to chronic or prior inflammation

Gas in the setting of fistula, surgery/ trauma, trauma or infection/abscess

Dirty shadowing, mobility, ring-down artifact, nondependent location

Gas within a bowel containing hernia

peristalsis

Iatrogenic

Shrapnel, staples—ring-down artifact, confirm with radiographs and history

752

Diagnostic Abdominal Imaging

A

B

C

D

Figure 12-78 Cryptorchidism A. Sagittal and (B) transverse US images through the left inguinal region reveal a small hypoechoic cryptorchid testis (arrows). Note the adjacent epididymis (arrowheads). C. Coronal T1-weighted and (D) fat saturated T2-weighted images in a

different patient show a cryptorchid testis (long arrows) in the left inguinal region. Note the absence of the left spermatic cord in the scrotum when compared with the normal right spermatic cord (arrowheads) and the gubernaculum (short arrow, A).

ANATOMY OF THE PENIS

The root is the proximal portion of the penis containing the crus of the paired corpora cavernosa and the bulb of the corpus spongiosum. The body of the penis extends from the root to the distal ends of the corpora cavernosa and is surrounded by the deep fascial layers. The distal portion of the penis consists of the expanded anterior end of the corpus spongiosum known as the glans penis.103,105,106 Numerous trabeculae extend across the corpora cavernosa, forming sinusoids or cavernous spaces that become engorged during erection. The cavernosal arteries are located slightly medially within the corpora cavernosa and feed capillary networks via helicine branches that open directly into the vascular sinusoids. Drainage of the corporal bodies is through emissary veins within the wall of the tunica albuginea into the deep dorsal vein of the penis,

The penis is composed of three cylindrical bodies of endothelium-lined vascular spaces: the paired dorsolateral corpora cavernosa and a single ventral corpus spongiosum. The corpora cavernosa are in close apposition in the penile shaft and diverge at the base of the penis to become the crura, which are attached to the ischial tuberosities. The bulbar and pendulous portions of the male urethra are contained within the corpus spongiosum, which forms the penile bulb posteriorly and glans penis anteriorly. A fibrous sheath known as the tunica albuginea surrounds the corpus spongiosum and both corpora cavernosa.103,104 External to this is another fascial layer (Buck’s fascia).

Chapter 12 Imaging of the Scrotum and Penis 753 which courses within the shallow groove between the corpora cavernosa, deep to Buck’s fascia. The superficial dorsal vein, which is superficial to Buck’s fascia, drains blood from the pendulous penile skin and glans and communicates with the deep dorsal vein. During erection, engorgement of the corpora cavernosa causes compression of the draining veins, trapping blood within the corporal bodies. Branches from the paired dorsal arteries of the penis, which run adjacent to the deep dorsal vein, supply the glans penis, distal corpus spongiosum and penile skin, whereas the bulbar artery supplies the urethra, proximal corpus spongiosum, and bulbospongiosus muscle.103,105,106

IMAGING OF THE PENIS US is the primary imaging modality to evaluate patients with penile disease. Clinical indications for penile US are palpable abnormality, penile fracture, erectile dysfunction, and priapism. MRI is generally reserved for cases where the ultrasound findings are not diagnostic, or for staging of known penile malignancies.107,108 For most indications, ultrasound of the penis is performed in the flaccid state, and consists primarily of grayscale axial and longitudinal images obtained with a high-frequency linear transducer (>12 megaHz). However, for evaluation of erectile dysfunction, a penile vascular flow study is also usually performed (described in more detail later), in which the penile vessels (cavernosal and dorsal arteries, deep dorsal vein) are evaluated by spectral and color Doppler, in both the flaccid state and the erect state. For these studies, erection is induced by intracavernosal injection of a vasoactive agent (most commonly Prostoglandin E1), if injection is not contraindicated by a condition predisposing to priapism, such as sickle cell disease, multiple myeloma, leukemia, tumors that invade the cavernosa, and partial cavernosal thromosis.156,157 In general, MRI is the preferred imaging modality for the evaluation and staging of penile malignancy due to superior soft-tissue contrast and multiplanar capabilities. The scrotum and penis are elevated by placing a folded towel between the patient’s legs. The penis is subsequently dorsiflexed against the lower abdomen and taped in position to reduce motion during examination. A surface coil is employed to maximize signal-to-noise ratio at small fields of view. A pelvic coil is used when imaging the entire pelvis in staging penile malignancy. In general, both T1- and T2-weighted sequences are obtained in axial, sagittal, and coronal planes. Fat-saturated T1-weighted images are obtained before and after administration of gadolinium contrast agent. As with US, imaging of the erect penis can be performed after the intracavernosal injection of vasoactive drugs.104,107

NORMAL IMAGING APPEARANCE OF THE MALE URETHRA AND PENIS Both sonography and MRI examinations can provide important information regarding diseases of the penis.

Ultrasonography In the flaccid penis, the corpus spongiosum and corpora cavernosa appear as homogeneous cylindrical structures. The urethra can sometimes be seen as a slitlike structure in the corpora spongiosum, but is typically difficult to appreciate when not distended by retrograde injection of fluid. The combination of the tunica albuginea and Buck’s fascia usually appear as a single thin echogenic line surrounding the corpora (see Figure 12-79). The cavernosal arteries appear as narrow tubular structures within the corpora cavernosa on longitudinal scans, measuring 0.3 to 1 mm (mean 0.30.5 mm) in the flaccid state. The dorsal vessels appear as anechoic structures along the dorsal aspect of the penile shaft, with the deep dorsal vein and dorsal arteries located deep to Buck’s fascia and the superficial dorsal vein outside Buck’s fascia (see Figure 12-80). In the flaccid state, peak systolic velocities in the cavernosal and dorsal arteries are similiar, in the range from 11 to 20 cm/s. At rest, flow within the cavernosal arteries demonstrates a high resistance pattern with no diastolic flow. During erection, flow within the cavernosal arteries undergoes a series of characteristic changes. Initially there is increased systolic flow with increased diastolic flow, followed by progressive decrease in diastolic flow due to venous occlusion, eventually resulting in reversal of diastolic flow. By contrast, the dorsal arteries normally maintain diastolic flow during erection (see Figure 12-80), although the peak systolic velocity increases.156,157

Magnetic Resonance Imaging The paired corpora cavernosa and single corpora spongiosum are usually intermediate signal intensity on T1-weighted sequences and high signal intensity on T2-weighted sequences. The corpora cavernosa are typically isointense to each other as fenestrations in the membranous intercavernosal septum allows communication of blood within the cavernosal sinuses. The corpora spongiosum may have a slightly different signal intensity from the corpora cavernosa because of different rates of blood flow within the vascular sinusoids. Fluid levels can be present in the cavernosal bodies as a normal finding because of layering of blood in the cavernosal spaces. Both the tunica albuginea and Buck’s fascia appear as low T1-weighted and low T2-weighted signalintensity bands that surround the corporal bodies, the individual layers of which may not be distinguishable. The cavernosal arteries appear as hypointense foci within the corpora cavernosa on axial T2-weighted sequences (see Figure 12-81).104 The urethra is a low-signal-intensity tubular structure coursing within the midline corpus spongiosum in the root of the penis. Midline sagittal T2-weighted sequences in some cases will show the course of the male urethra extending from the bladder neck to the distal penile urethra. However, the proximal portion of the prostatic urethra and distal penile urethra are rarely seen on MRI in the absence of a Foley catheter.111

754

Diagnostic Abdominal Imaging

A

B

C

D

Figure 12-79 Normal US Appearance of the Penis A. Grayscale transverse image at the mid-shaft penis demonstrating the paired corpora cavernosa (asterisks) and corpora spongiosum (arrowhead). B. Grayscale longitudinal image of the flaccid penis demonstrating the tunica albuginea (arrowhead) surrounding the corpora cavernosa (asterisk). The cavernosal artery (arrow) appears as a narrow tubular structure located centrally within the cavernosal body. C.

Grayscale sagittal image of the flaccid penis demonstrating homogeneous echogenicity of the corpora spongiosum (asterisk). D. Grayscale sagittal image of the erect penis (following injection of prostaglandin E-1) demonstrating the engorged corpora cavernosum (asterisk) and cavernosal artery (arrow). The tunica albuginea is again visualized as a thin echogenic line (arrowhead) surrounding the corpora.

DISEASES OF THE PENIS

Primary malignancies of the penis

Penile pathology can be subdivided into solid lesions of the penis, cystic lesions of the penis, and some unique disorders of the penis.

Primary malignant neoplasms of the penis are rare in Western countries, affecting an estimated 1000 to 1500 men per year, with approximately 300 deaths per year. By contrast, in some parts of Africa and South America, penile cancer may account for 10% to 20% of all male cancers.

Solid Lesions of the Penis Solid lesions of the penis include benign and malignant neoplasms, fibrous plaques, and partial thrombosis of the corpora cavernosa (Table 12-15).

Squamous cell carcinoma Squamous cell carcinoma (SCC) accounts for 95% of penile cancers and usually occurs in the sixth to seventh

Chapter 12 Imaging of the Scrotum and Penis 755

A

B

C

D

Figure 12-80 Normal US Appearance of the Penile Vessels A. Coronal color Doppler image of the paired cavernosal arteries (arrows) demonstrating normal flow. B. Spectral Doppler tracing obtained from the right cavernosal artery 20 minutes following stimulation (injection of 3.5 μg of prostaglandin E-1) shows expected reversal of diastolic flow and a peak systolic velocity of >25 cm/s. C. Spectral Doppler tracing during erection obtained

from a dorsal artery shows measurable diastolic flow, which is normal in the dorsal arteries, by contrast to the cavernosal arteries (B). D. Spectral Doppler tracing obtained following stimulation from the deep dorsal vein demonstrating a venous waveform, which may be present early post-stimulation, but should disappear as erection progresses.

decades of life, most commonly arising from the skin of the glans penis in 48% of patients. SCC is far more common in uncircumcised men, presumably as a result of chronic irritation from smegma. Therefore, circumcision in the neonatal period is considered a well-established prophylactic measure for penile carcinoma. Infection with human papilloma viruses 16 and 18 is an additional risk factor for the development of penile cancer.104,107,111-119 Sonographically, on grayscale imaging, SCC usually appears as a hypoechoic heterogeneous mass that may invade into the adjacent cavernosal bodies. In the region of the glans, invasion of subepithelial tissue is difficult to differentiate from invasion of the corpus spongiosum.

However, elsewhere, deep invasion of the corpora cavernosa appears as interruption of the tunica albuginea, and early invasion may demonstrate focal thickening and decreased

Imaging Notes 12-23. Differentiating Features of Scrotal Wall Thickening Squamous cell carcinoma is the most common malignancy of the penis typically arising from the skin of the glans penis and seen on the sixth and seventh decades of life.

756

Diagnostic Abdominal Imaging

A

B

C

D

Figure 12-81 Normal MRI appearance of the Penis A. Axial T1-weighted MRI of the penis with fat suppression demonstrating the intermediate-signal-intensity corpora cavernosa (asterisks) and corpus spongiosum (arrowhead). The thin, hypointense line surrounding the corpora cavernosa represents the tunica albuginea. B. Axial T1-weighted MRI of the penis with fat suppression obtained after the administration of gadolinium-containing contrast material demonstrating enhancement of the paired corpora cavernosa and corpora spongiosum. The tunica albuginea (white arrow) is thickened

around the corpora cavernosa where it joins with Buck’s fascia. The flattened urethra (arrowhead) is located centrally within the corpora spongiosum. C. Transverse T2-weighted MRI of the penis more clearly demonstrates the hypointense tunica albuginea surrounding the corpora cavernosa (asterisks) and corpus spongiosum (arrowhead). D. Sagittal T2-weighted MRI of the penis demonstrates the low signal tunica albuginea (arrows). The urethra is visible in the proximal corpora spongiosum (arrowhead).

Chapter 12 Imaging of the Scrotum and Penis 757 Table 12-15. Causes of Solid Lesions of the Penis 1. Malignant neoplasms a. Penile shaft i. Squamous cell carcinoma ii. Melanoma iii. Epithelioid sarcoma iv. Leiomyosarcoma v. Rhabdosarcoma vi. Kaposi sarcoma b. Urethra i. Squamous cell carcinoma ii. Transitional cell carcinoma iii. Adenocarcinoma c. Metastasis 2. Benign neoplasms a. Penile shaft i. Hemangioma ii. Neurofibroma b. Urethra i. Fibrovascular polyp ii. Hemangioma 3. Fibrous plaques a. Peyronie disease b. Trauma (accidental and iatrogenic) c. Injection of intracavernosal drugs

Table 12-16. Staging System for Penile Cancer TNM system of staging penile cancer TX

Primary tumor cannot be assessed

T0

No evidence of primary tumor

Ta

Noninvasive verrucous carcinoma

Tis

Carcinoma in situ

T1

Tumor invades subepithelial connective tissue

T2

Tumor invades corpora cavernosa or spongiosum

T3

Tumor invades the urethra or prostate

T4

Tumor invades other adjacent structures

NX

Regional lymph nodes cannot be assessed

N0

No regional lymph node metastases

N1

Metastasis in a single superficial inguinal lymph node

N2

Metastasis in multiple or bilateral superficial inguinal lymph nodes

MX

Presence of distant metastases cannot be assessed

N3

Extranodal extension of lymph node metastasis or pelvic lymph node(s) unilateral or bilateral

M0

No distant metastases

M1

Distant metastases

4. Partial cavernosal thrombosis.

echogenicity of an intact tunica at the point of contact with the adjacent lesion. Color Doppler evaluation generally does not provide much additional information, other than to confirm that a penile lesion is solid. Nonetheless, it has been reported that SCC typically demonstrates poor vascularization, although associated inflammation may result in increased vascularity.120-123 On MRI, SCC of the penis is typically hypointense on both T1- and T2-weighted sequences with poor enhancement after the administration of gadolinium contrast material compared to the corpora. However, associated inflammation can result in areas of increased signal intensity on T2-weighted sequences with marked enhancement. Both T2-weighted and gadolinium-enhanced images can be used to determine the presence and extent of invasion of the corpora. Lymphatic drainage of SCC is via pelvic lymph node chains. The location of the primary lesion determines whether spread is via superficial inguinal nodes (shaft), deep inguinal nodes (glans and shaft), external iliac nodes (glans), or internal iliac nodes (urethra). Because there is significant communication between the lymphatic channels, bilateral lymphadenopathy can be seen with a unilateral tumor.124 There is no universally accepted staging system for primary penile carcinoma, and both the TNM and Jackson systems are used (Table 12-16). Tumors involving the glans and distal shaft are typically managed with partial

Jackson classification for staging penile cancer Stage I

Tumor is confined to the glans, prepuce, or both

Stage II

Tumor extends onto the shaft of the penis

Stage III

Tumor has inguinal metastasis that is operable

Stage IV

Tumor involves adjacent structures and is associated with inoperable inguinal metastasis or distant metastasis

Adapted from references 104 and 158.

penectomy to preserve a cosmetically acceptable and functional penis. Proximal tumors involving the base of the penis can require total penectomy. Surgical margins of 1 to 2 cm have been advocated to reduce the rate of local recurrence. Radiation therapy can be used as an alternative to surgery in young men with small (25 cm/s and is usually due to the presence of a fistula between the cavernosal artery and the vascular spaces of the corpus cavernosum, known as an arterial-lacunar fistula. Patients usually present days or weeks after an injury with a painless nonrigid erection due to unregulated arterial inflow into the sinusoidal spaces. Grayscale US may reveal an irregular hypoechogenic region in the corpus cavernosum due to enlarged vascular spaces or tissue injury. Color Doppler shows a blush of turbulent flow extending from a feeding artery into the cavernosal tissues, with

Imaging Notes 12-26. Clinical and Imaging Features of Priapism High Flow:

1. Fistula between the cavernosal artery and the corpus cavernosum 2. Painless, nonrigid erection 3. Cavernosal artery flow >25 cm/s

Low Flow:

1. Sinusoidal thrombosis and venous occlusion 2. Painful, rigid erection 3. Cavernosal artery flow: high resistance, low velocity 4. Venous flow: Minimal or absent

766

Diagnostic Abdominal Imaging

A

B

Figure 12-91 Arterial-lacunar Fistula A. Color Doppler transverse image in a patient with partial priapism a few days following straddle injury demonstrates a hyperechogenic hematoma in the left corpora cavernosa (arrows)

adjacent to an area of disorganized flow (arrowhead) which shows low resistance on spectral Doppler (B), in keeping with a cavernosal-lacunar fistula in the setting of high-flow priapism.

Peyronie’s Disease

Imaging Evaluation of Erectile Dysfunction

Peyronie’s disease is a common disorder of the penis that typically results in a characteristic curved deformity of the penile shaft when erect. This disorder is caused by fibrous plaques within the tunica albuginea that appear as a focal lesion. This is discussed previously under the heading: Solid Lesions of the Penis.

Erectile dysfunction is defined as the persistent or repeated inability to attain or maintain an erection sufficient for satisfactory sexual performance, with a duration of at least 6 months. The pathophysiology of erectile dysfunction is complex with many potential contributing organic and psychogenic factors, a discussion of which is beyond the scope of this review. Sonographic penile flow evaluation is just one component of the evaluation of erectile dysfunction, which must encompass the complete sequence of male sexual function. US is used to evaluate the arterial and venous flow patterns within the erect penis to distinguish between arterial and venous causes of erectile dysfunction. In a penile flow study, grayscale, color Doppler and spectral Doppler evaluation of penis are performed in the flaccid state. A vasoactive agent is then injected into the penis to stimulate penile erection and the peak systolic velocities in the cavernosal arteries (as well as the arterial diameter) are recorded every 5 minutes. The dorsal penile arteries and deep dorsal vein are also assessed on spectral Doppler. The average peak systolic velocity (PSV) in the cavernosal arteries after cavernosal injection of vasoactive substances is approximately 30 to 40 cm/s. A PSV of less than 25 cm/s with a dampened waveform in the cavernosal artery is the standard criterion for a diagnosis of arterial insufficiency. Other findings of arterial insufficiency include (1) dilation of the cavernosal artery by less than 75%, (2) asymmetry of cavernosal peak systolic velocities by >10 cm/s, (3) focal stenosis or retrograde flow in a cavernosal artery (CA), (4) acceleration time of >100

Figure 12-92 Low Flow Priapism Sagittal color Doppler image obtained of one of the cavernosal bodies in a patient with priapism lasting >30 hours, showing no detectable flow in the cavernosal artery, in keeping with low flow priapism.

Chapter 12 Imaging of the Scrotum and Penis 767 Imaging Notes 12-27. Sonographic Evaluation of Erectile Dysfunction Arterial Insufficiency Primary criteria:

1. Peak systolic velocity in the cavernosal artery 10 cm/s difference in cavernosal peak systolic velocities 3. Focal stenosis/retrograde flow in cavernosal artery 4. Acceleration time >100 ms in the cavernosal artery 5. Peak systolic velocity 5 cm/s throughout all phases of erection (cavernosal arteries)

Secondary criteria:

1. Persistent flow in the deep dorsal vein 2. Resistive index in the cavernosal arteries less than 0.75 measured 20 min after cavernosal injection of vasoactive substance

Note: Venous insufficiency cannot be accurately diagnosed by US in the setting of arterial insufficiency.

ms, and (5) a PSV of less than 10 cm/s in the cavernosal artery in the flaccid state. The peak systolic velocities in the dorsal arteries are also measured for comparison and usually are similar to those in the cavernosal arteries. Normal PSV in the dorsal arteries with low cavernosal artery velocities suggests intrapenile arteriogenic disease whereas low velocities in the both the dorsal and cavernosal arteries suggests more proximal arterial disease (see Figure 12-93). Normally, there is no diastolic flow in the cavernosal arteries during the tumescent phase of erection. An arterial

A

B

Figure 12-93 Penile Arterial Insufficiency in 2 Patients A. Spectral Doppler waveform obtained from sampling the right cavernosal artery following stimulation demonstrates a tardus parvus appearance with abnormally low peak systolic velocity of 12.1 cm/s (normal > 25 cm/s). Similar waveforms were obtained from the left cavernosal artery and from the dorsal arteries, indicative of arterial insufficiency proximal

diastolic velocity of >5 cm/s throughout all phases of erection is abnormal and indicates persistent arterial diastolic flow, a finding that suggests venous leak if the patient has normal arterial function. Venous leak/insufficiency can also be suggested by persistent flow in the deep dorsal vein. Note that in the setting of arterial insufficiency, the cavernosal sinusoids may never fill to the point of occluding the veins, resulting in persistent venous flow even in the setting of normal veins, and therefore venous insufficiency cannot be accurately diagnosed by US in the setting of arterial insufficiency (see Figure 12-94).157

C to the penis. B. Spectral Doppler waveform obtained from the right cavernosal artery in a different patient following stimulation shows a parvus tardus appearance with very low peak systolic velocity, in keeping with arterial insufficiency. However, the spectral Doppler wave from the right dorsal artery is normal (C), suggesting intrapenile arterial disease.

768

Diagnostic Abdominal Imaging 12. Lorigan JG, Eftekhari F, David CL, Shirkhoda A. The growing teratoma syndrome: an unusual manifestation of treated, nonseminomatous germ cell tumors of the testis. Am J Roentgenol. 1988;151:325-329. 13. Aide N, Comoz F, Sevin E. Enlarging residual mass after treatment of a nonseminomatous germ cell tumor: growing teratoma syndrome or cancer recurrence? J Clin Oncol. 2007;25:4494-4496. 14. Woodward PJ, Schwab CM, Sesterhenn IA. From the archives of the AFIP: extratesticular scrotal masses: radiologic-pathologic correlation. Radiographics. 2003; 23:215-240. 15. Maizlin ZV, Belenky A, Kunichezky M, Sandbank J, Strauss S. Leydig cell tumors of the testis: grayscale and color Doppler sonographic appearance. J Ultrasound Med. 2004;23:959-964. 16. Fernandez GC, Tardaguila F, Rivas C, et al. MRI in the diagnosis of testicular Leydig cell tumour. Br J Radiol. 2004;77:521-524.

Figure 12-94 Penile Venous Insufficiency Spectral Doppler waveform obtained from the cavernosal artery following stimulation demonstrates persistent elevated diastolic flow, a finding of venous leak/insufficiency, assuming normal arterial inflow.

REFERENCES 1. Dogra VS, Gottlieb RH, Oka M, Rubens DJ. Sonography of the scrotum. Radiology. 2003;227:18-36. 2. Lee JC, Bhatt S, Dogra VS. Imaging of the epididymis. Ultrasound Q. 2008;24:3-16.

17. Maizlin ZV, Belenky A, Baniel J, Gottlieb P, Sandbank J, Strauss S. Epidermoid cyst and teratoma of the testis: sonographic and histologic similarities. J Ultrasound Med. 2005;24:1403-1409. 18. Mazzu D, Jeffrey RB Jr, Ralls PW. Lymphoma and leukemia involving the testicles: findings on gray-scale and color Doppler sonography. Am J Roentgenol. 1995;164:645-647. 19. Akin EA, Khati NJ, Hill MC. Ultrasound of the scrotum. Ultrasound Q. 2004;20:181-199. 20. Zicherman JM, Weissman D, Gribbin C, Epstein R. Best cases from the AFIP: primary diffuse large B-cell lymphoma of the epididymis and testis. Radiographics. 2005;25:243-248. 21. Stoffel TJ, Nesbit ME, Levitt SH. Extramedullary involvement of the testes in childhood leukemia. Cancer. 1975;35:1203-1211.

3. Gerscovich EO. Scrotal Ultrasound Measurements. In: Goldberg BB, McGahan JP, eds. Atlas of Ultrasound Measurements. 2nd ed. Philadelphia, PA: Mosby Elsevier; 2006:375-381.

22. Lupetin AR, King W 3rd, Rich P, Lederman RB. Ultrasound diagnosis of testicular leukemia. Radiology. 1983;146:171-172.

4. Kim W, Rosen MA, Langer JE, Banner MP, Siegelman ES, Ramchandani P. US MR imaging correlation in pathologic conditions of the scrotum. Radiographics. 2007;27:1239-1253.

23. Dambro TJ, Stewart RR, Carroll BA. Chapter 24: the scrotum. In: Rumack CM, Wilson SR, Charboneau JW, eds. Diagnostic Ultrasound. St Louis, MO: Mosby; 1998:791-821.

5. Langer JE. Ultrasound of the scrotum. Semin Roentgenol. 1993;28:5-18.

24. Muttarak M, Peh WCG. Case 91: tuberculous epididymoorchitis. Radiology. 2006;238:748-751.

6. Epstein JI. Chapter 21—the lower urinary tract and male genital system. In: Kumar V, Abbas AK, Fausto N, eds. Kumar: Robbins and Cotran: Pathologic Basis of Disease. 7th ed. Philadelphia, PA: Elsevier Saunders; 2005.

25. Drudi FM, Laghi A, Iannicelli E, et al. Tubercular epididymitis and orchitis: US patterns. Eur Radiol. 1997;7:1076-1078.

7. Woodward PJ, Sohaey R, O’Donoghue MJ, Green DE. From the archives of the AFIP: tumors and tumorlike lesions of the testis: radiologic-pathologic correlation. Radiographics. 2002;22:189-216. 8. Sohaib SA, Koh D-M, Husband JE. The role of imaging in the diagnosis, staging, and management of testicular cancer. Am J Roentgenol. 2008;191:387-395. 9. Kocakoc E, Bhatt S, Dogra VS. Ultrasound evaluation of testicular neoplasms. Ultrasound Clin. 2007;2:27-44. 10. Ryan CJ, Small EJ, Torti FM. Chapter 90—testicular cancer. In: Abeloff MB, Armitage MD, Niederhuber JO, Kastan JE, eds. Abeloff’s Clinical Oncology. 4th ed. Philadelphia, PA: Churchill Livingstone Elsevier; 2008. 11. Tsili AC, Tsampoulas C, Giannakopoulos X, et al. MRI in the histologic characterization of testicular neoplasms. Am J Roentgenol. 2007;189:W331-W337.

26. Muttarak M, Peh WCG, Lojanapiwat B, Chaiwun B. Tuberculous epididymitis and epididymo-orchitis: sonographic appearances. Am J Roentgenol. 2001;176:1459-1466. 27. Dogra VS, Gottlieb RH, Rubens DJ, Liao L. Benign intratesticular cystic lesions: US features. Radiographics. 2001;21:S273-S281. 28. Turgut AT, Bhatt S, Dogra VS. Acute painful scrotum. Ultrasound Clin. 2008;3:93-107. 29. Stewart VR, Sidhu PS. The testis: the unusual, the rare and the bizarre. Clin Radiol. 2007;62:289-302. 30. Gerscovich EO, Bateni CP, Kazemaini MR, Gillen MA, Visis T. Reversal of diastolic blood flow in the testis of a patient with impending infarction due to epididymitis. J Ultrasound Med. 2008;27:1643-1646. 31. Holloway BJ, Belcher HE, Letourneau JG, Kunberger LE. Scrotal sonography: a valuable tool in the evaluation of complications following inguinal hernia repair. J Clin Ultrasound. 1998;26:341-344.

Chapter 12 Imaging of the Scrotum and Penis 769 32. Fernandez-Perez GC, Tardaguila FM, Velasco M, et al. Radiologic findings of segmental testicular infarction. Am J Roentgenol. 2005;184:1587-1593. 33. Ulbright TM, Young RH. Metastatic carcinoma to the testis: a clinicopathologic analysis of 26 nonincidental cases with emphasis on deceptive features. Am J Surg Pathol. 2008;32:1683-1693.

52. Middleton WD, Middleton MA, Dierks M, Keetch D, Dierks S. Sonographic prediction of viability in testicular torsion: preliminary observations. J Ultrasound Med. 1997;16:23-27; quiz 29-30. 53. Vijayaraghavan SB. Sonographic differential diagnosis of acute scrotum: real-time whirlpool sign, a key sign of torsion. J Ultrasound Med. 2006;25:563-574.

34. Dieckmann KP, Boeckmann W, Brosig W, Jonas D, Bauer HW. Bilateral testicular germ cell tumors. Report of nine cases and review of the literature. Cancer. 1986;57:1254-1258.

54. Horstman WG, Middleton WD, Melson GL, Siegel BA. Color Doppler US of the scrotum. Radiographics. 1991;11: 941-957.

35. Adham WK, Raval BK, Uzquiano MC, Lemos LB. Best cases from the AFIP: bilateral testicular tumors: seminoma and mixed germ cell tumor. Radiographics. 2005;25:835-839.

55. Burks DD, Markey BJ, Burkhard TK, Balsara ZN, Haluszka MM, Canning DA. Suspected testicular torsion and ischemia: evaluation with color Doppler sonography. Radiology. 1990;175:815-821.

36. Winter TC 3rd, Keener TS, Mack LA. Sonographic appearance of testicular sarcoid. J Ultrasound Med. 1995;14:153-156. 37. Carucci LR, Tirkes AT, Pretorius ES, Genega EM, Weinstein SP. Testicular Leydig’s cell hyperplasia: MR imaging and sonographic findings. Am J Roentgenol. 2003; 180:501-503. 38. Dogra V, Nathan J, Bhatt S. Sonographic appearance of testicular adrenal rest tissue in congenital adrenal hyperplasia. J Ultrasound Med. 2004;23:979-981. 39. Avila NA, Premkumar A, Shawker TH, Jones JV, Laue L, Cutler GB Jr. Testicular adrenal rest tissue in congenital adrenal hyperplasia: findings at Gray-scale and color Doppler US. Radiology. 1996;198:99-104. 40. Nagamine WH, Mehta SV, Vade A. Testicular adrenal rest tumors in a patient with congenital adrenal hyperplasia: sonographic and magnetic resonance imaging findings. J Ultrasound Med. 2005;24:1717-1720. 41. Ors F, Lev-Toaff A, O’Kane P, Qazi N, Bergin D. Paraovarian adrenal rest with MRI features characteristic of an adrenal adenoma. Br J Radiol. 2007;80:e205-e208.

56. Deurdulian C, Mittelstaedt CA, Chong WK, Fielding JR. US of acute scrotal trauma: optimal technique, imaging findings, and management. Radiographics. 2007;27:357-369. 57. Phillips G, Kumari-Subaiya S, Sawitsky A. Ultrasonic evaluation of the scrotum in lymphoproliferative disease. J Ultrasound Med. 1987;6:169-175. 58. Harris RD, Chouteau C, Partrick M, Schned A. Prevalence and significance of heterogeneous testes revealed on sonography: ex vivo sonographic–pathologic correlation. Am J Roentgenol. 2000;175:347-352. 59. Casalino DD, Kim R. Clinical importance of a unilateral striated pattern seen on sonography of the testicle. Am J Roentgenol. 2002;178:927-930. 60. Dogra VS, Rubens DJ, Gottlieb RH, Bhatt S. Torsion and beyond: new twists in spectral Doppler evaluation of the scrotum. J Ultrasound Med. 2004;23:1077-1085. 61. Patel MD, Silva AC. MRI of an adenomatoid tumor of the tunica albuginea. AJR Am J Roentgenol. 2004;182:415-417.

42. Hamm B, Fobbe F, Loy V. Testicular cysts: differentiation with US and clinical findings. Radiology. 1988;168:19-23.

62. Akbar SA, Sayyed TA, Jafri SZH, Hasteh F, Neill JSA. Multimodality imaging of paratesticular neoplasms and their rare mimics. Radiographics. 2003;23:1461-1476.

43. Chen SS, Chou YH, Hsu CC, Tiu CM, Chang TE. Simple testicular cyst [in Chinese]. Zhonghua Yi Xue Za Zhi. (Taipei) 1990;46:285-288.

63. Choyke PL, Glenn GM, Wagner JP, et al. Epididymal cystadenomas in von Hippel-Lindau disease. Urology. 1997;49:926-931.

44. Rubenstein RA, Dogra VS, Seftel AD, Resnick MI. Benign intrascrotal lesions. J Urol. 2004;171:1765-1772.

64. Uppuluri S, Bhatt S, Tang P, Dogra VS. Clear cell papillary cystadenoma with sonographic and histopathologic correlation. J Ultrasound Med. 2006;25:1451-1453.

45. Brown DL, Benson CB, Doherty FJ, et al. Cystic testicular mass caused by dilated rete testis: sonographic findings in 31 cases. Am J Roentgenol. 1992;158:1257-1259. 46. Smith SJ, Vogelzang RL, Smith WM, Moran MJ. Papillary adenocarcinoma of the rete testis: sonographic findings. Am J Roentgenol. 1987;148:1147-1148. 47. Gabriel H, Marko J, Nikolaidis P. Cystadenoma of the rete testis: sonographic appearance. Am J Roentgenol. 2007;189:W67-W69. 48. Cho CS, Kosek J. Cystic dysplasia of the testis: sonographic and pathologic findings. Radiology. 1985;156:777-778. 49. Das KM, Prasad K, Szmigielski W, Noorani N. Intratesticular varicocele: evaluation using conventional and Doppler sonography. Am J Roentgenol. 1999;173:1079-1083. 50. Kessler A, Meirsdorf S, Graif M, Gottlieb P, Strauss S. Intratesticular varicocele: grayscale and color Doppler sonographic appearance. J Ultrasound Med. 2005;24: 1711-1716. 51. Caesar RE, Kaplan GW. Incidence of the bell-clapper deformity in an autopsy series. Urology. 1994;44:114-116.

65. Alleman WG, Gorman B, King BF, Larson DR, Cheville JC, Nehra A. Benign and malignant epididymal masses evaluated with scrotal sonography: clinical and pathologic review of 85 patients. J Ultrasound Med. 2008;27:1195-1202. 66. Yang DM, Kim SH, Kim HN, et al. Differential diagnosis of focal epididymal lesions with grayscale sonographic, color Doppler sonographic, and clinical features. J Ultrasound Med. 2003;22:135-142. 67. Saginoya T, Yamaguchi K, Toda T, Kiyuna M. Fibrous pseudotumor of the scrotum: MR imaging findings. AJR Am J Roentgenol. 1996;167:285-286. 68. Beccia DJ, Krane RJ, Olsson CA. Clinical management of nontesticular intrascrotal tumors. J Urol. 1976; 116:476-479. 69. Frates MC, Benson CB, DiSalvo DN, Brown DL, Laing FC, Doubilet PM. Solid extratesticular masses evaluated with sonography: pathologic correlation. Radiology. 1997;204:43-46. 70. Cornud F, Belin X, Amar E, Delafontaine D, Helenon O, Moreau JF. Varicocele: strategies in diagnosis and treatment. Eur Radiol. 1999;9:536-545.

770

Diagnostic Abdominal Imaging

71. Aizenstein RI, Wilbur AC, O’Neil HK, Gerber B. Clinical image. MRI of scrotal hemangioma. J Comput Assist Tomogr. 1996;20:888-889. 72. Hwang S, Aronoff DR, Leonidas JC. Case 82: polyorchidism with torsion. Radiology. 2005;235:433-435. 73. Amodio JB, Maybody M, Slowotsky C, Fried K, Foresto C. Polyorchidism: report of 3 cases and review of the literature. J Ultrasound Med. 2004;23:951-957. 74. Bhosale PR, Patnana M, Viswanathan C, Szklaruk J. The inguinal canal: anatomy and imaging features of common and uncommon masses. Radiographics. 2008;28:819-835. 75. Aso C, Enriquez G, Fite M, et al. Gray-scale and color Doppler sonography of scrotal disorders in children: an update. Radiographics. 2005;25:1197-1214. 76. Schnall M. Magnetic resonance imaging of the scrotum. Semin Roentgenol. 1993;XXVIII:19-30. 77. Turkvatan A, Kelahmet E, Yazgan C, Olcer T. Sonographic findings in tuberculous epididymo-orchitis. J Clin Ultrasound. 2004;32:302-305.

91. Miller FN, Sidhu PS. Does testicular microlithiasis matter? A review. Clin Radiol. 2002;57:883-890. 92. Bennett HF, Middleton WD, Bullock AD, Teefey SA. Testicular microlithiasis: US follow-up. Radiology. 2001;218:359-363. 93. Kim B, Winter TC 3rd, Ryu JA. Testicular microlithiasis: clinical significance and review of the literature. Eur Radiol. 2003;13:2567-2576. 94. Backus ML, Mack LA, Middleton WD, King BF, Winter TC 3rd, True LD. Testicular microlithiasis: imaging appearances and pathologic correlation. Radiology. 1994;192:781-785. 95. Cast JEI, Nelson WM, Early AS, et al. Testicular Microlithiasis: Prevalence and tumor risk in a population referred for scrotal sonography. Am J Roentgenol. 2000;175:1703-1706. 96. Teefey SA. Ask the Expert: Microlithiasis to follow or not to follow. Soc Radiol Ultrasound Newslett. 2007;17:12. 97. Kane CJ, Nash P, McAninch JW. Ultrasonographic appearance of necrotizing gangrene: aid in early diagnosis. Urology. 1996;48:142-144.

78. Kim SH, Pollack HM, Cho KS, Pollack MS, Han MC. Tuberculous epididymitis and epididymo-orchitis: sonographic findings. J Urol. 1993;150:81-84.

98. Peterson AC, Bauman JM, Light DE, McMann LP, Costabile RA. The prevalence of testicular microlithiasis in an asymptomatic population of men 18 to 35 years old. J Urol. 2001;166:2061-2064.

79. Eraso CE, Vrachliotis TG, Cunningham JJ. Sonographic findings in testicular sarcoidosis simulating malignant nodule. J Clin Ultrasound. 1999;27:81-83.

99. Middleton WD, Teefey SA, Santillan CS. Testicular microlithiasis: prospective analysis of prevalence and associated tumor. Radiology. 2002;224:425-428.

80. Reinecks EZ, MacLennan GT. Sarcoidosis of the testis and epididymis. J Urol. 2008;179:1147.

100. Bach AM, Hann LE, Hadar O, et al. Testicular microlithiasis: what is its association with testicular cancer? Radiology. 2001;220:70-75.

81. Koyama T, Ueda H, Togashi K, Umeoka S, Kataoka M, Nagai S. Radiologic manifestations of sarcoidosis in various organs. Radiographics. 2004;24:87-104. 82. Reddy NM, Gerscovich EO, Jain KA, Le-Petross HT, Brock JM. Vasectomy-related changes on sonographic examination of the scrotum. J Clin Ultrasound. 2004;32:394-398. 83. Kousei Ishigami MMA-YYE-Z. Tubular ectasia of the epididymis: a sign of postvasectomy status. J Clin Ultrasound. 2005;33:447-451. 84. Leung ML, Gooding GA, Williams RD. High-resolution sonography of scrotal contents in asymptomatic subjects. AJR Am J Roentgenol. 1984;143:161-164.

101. Lam DL, Gerscovich EO, Kuo MC, McGahan JP. Testicular microlithiasis: our experience of 10 years. J Ultrasound Med. 2007;26:867-873. 102. DeCastro BJ, Peterson AC, Costabile RA. A 5-year followup study of asymptomatic men with testicular microlithiasis. J Urol. 2008;179:1420-1423; discussion 1423. 103. Siegelman ES. Body MRI. Philadelphia, PA: Elsevier; 2004. 104. Pretorius ES, Siegelman ES, Ramchandani P, Banner MP. MR imaging of the penis. Radiographics. 2001;21 Spec No: S283-298; discussion S298-S299. 105. Kirkham AP, Illing RO, Minhas S, Minhas S, Allen C. MR imaging of nonmalignant penile lesions. Radiographics. 2008. 28(3):837-853.

85. Boyum J, Wasserman NF. Malignant mesothelioma of the tunica vaginalis testis: a case illustrating Doppler color flow imaging and its potential for preoperative diagnosis. J Ultrasound Med. 2008;27:1249-1255.

106. Dunnick NR, Sandler CM, Newhouse JH, Amis ES. Textbook of Uroradiology. 2007.

86. Tasu J-P, Faye N, Eschwege P, Rocher L, Blery M. Imaging of burned-out testis tumor: five new cases and review of the literature. J Ultrasound Med. 2003;22:515-521.

107. Bertolotto M, Pavlica P, Serafini G, Quaia E, Zappetti R. Painful penile induration: imaging findings and management. Radiographics. 2009;29:477-493.

87. Frauscher F, Klauser A, Stenzl A, Helweg G, Amort B, zur Nedden D. US findings in the scrotum of extreme mountain bikers. Radiology. 2001;219:427-431.

108. Middleton WD, Kurtz AB. Ultrasound: The Requisites. 2nd ed. 2003.

88. Turgut AT, Kosar U, Kosar P, Karabulut A. Scrotal sonographic findings in equestrians. J Ultrasound Med. 2005;24:911-917. 89. van der Sluijs JW, den Hollander JC, Lequin MH, Nijman RM, Robben SG. Prenatal testicular torsion: diagnosis and natural course. An ultrasonographic study. Eur Radiol. 2004;14:250-255. 90. Bushby LH, Miller FN, Rosairo S, Clarke JL, Sidhu PS. Scrotal calcification: ultrasound appearances, distribution and aetiology. Br J Radiol. 2002;75:283-288.

109. Pavlica P, Barozzi L, Menchi I. Imaging of male urethra. Eur Radiol. 2003:13(7): 1583-1596. 110. Bhatt S, Kocakoc E, Rubens DJ, Seftel AD, Dogra VS. Sonographic evaluation of penile trauma. J Ultrasound Med. 2005;24(7):993-1000; quiz 1001. 111. Ryu J, Kim B. MR imaging of the male and female urethra. Radiographics. 2001;21:1169-1185. 112. Sica GT, Teeger S. MR imaging of scrotal, testicular, and penile diseases. Magn Reson Imaging Clin N Am. 1996;4(3): 545-563.

Chapter 12 Imaging of the Scrotum and Penis 771 113. McCance DJ, Kalache A, Ashdown K, et al. Human papillomavirus types 16 and 18 in carcinomas of the penis from Brazil. Int J Cancer. 1986;37(1):55-59.

132. Forstner R, Hricak H, Kalbhen CL, Kogan BA, McAninch JW. Magnetic resonance imaging of vascular lesions of the scrotum and penis. Urology. 1995;46(4):581-583.

114. Mostofi FK, Davis CJ Jr, Sesterhenn IA. Carcinoma of the male and female urethra. Urol Clin North Am. 1992;19(2):347-358.

133. Kim SH, Lee SE, Han MC. Penile hemangioma: US and MR imaging demonstration. Urol Radiol. 1991;13(2):126-128.

115. Lucia MS, Miller GJ. Histopathology of malignant lesions of the penis. Urol Clin North Am. 1992;19(2):227-246.

134. Niku SD, Mattrey RF, Kalota SJ, Schmidt JD. MRI of pelvic neurofibromatosis. Abdom Imaging. 1995;20(2):176-178.

116. Oto A, Meyer J. MR appearance of penile epithelioid sarcoma. AJR Am J Roentgenol. 1999;172(2):555-556.

135. Jalkut M, Gonzalez-Cadavid N, Rajfer J. Peyronie’s Disease: A Review. Rev Urol. 2003;5(3):142-148.

117. Isa SS, Almaraz R, Magovern J. Leiomyosarcoma of the penis. Case report and review of the literature. Cancer. 1984;54(5): 939-942.

136. Hauck EW, Hackstein N, Vosshenrich R, et al. Diagnostic value of magnetic resonance imaging in Peyronie’s disease—a comparison both with palpation and ultrasound in the evaluation of plaque formation. Eur Urol. 2003;43(3):293-299; discussion 299-300.

118. Agrons GA, Wagner BJ, Lonergan GJ, Dickey GE, Kaufman MS. From the archives of the AFIP. Genitourinary rhabdomyosarcoma in children: radiologic-pathologic correlation. Radiographics. 1997;17(4):919-937. 119. Barnholtz-Sloan JS, Maldonado JL, Pow-sang J, Giuliano AR. Incidence trends in primary malignant penile cancer. Urol Oncol. 2007;25(5):361-367. 120. Agrawal A, Pai D, Ananthakrishnan N, Smile SR, Ratnakar C. Clinical and sonographic findings in carcinoma of the penis. J Clin Ultrasound. 2000;28(8):399-406. 121. Lont AP, Besnard AP, Gallee MP, van Tinteren H, Horenblas S. A comparison of physical examination and imaging in determining the extent of primary penile carcinoma. BJU Int. 2003;91(6):493-495. 122. Horenblas S, Kröger R, Gallee MP, Newling DW, van Tinteren H. Ultrasound in squamous cell carcinoma of the penis; a useful addition to clinical staging? A comparison of ultrasound with histopathology. Urology. 1994;43(5):702-707. 123. Bertolotto M, Serafini G, Dogliotti L, et al. Primary and secondary malignancies of the penis: ultrasound features. Abdom Imaging. 2005;30(1):108-112. 124. Vossough A, Pretorius ES, Siegelman ES, Ramchandani P, Banner MP. Magnetic resonance imaging of the penis. Abdom Imaging. 2002;27:640-659. 125. Bermejo C, Busby JE, Spiess PE, Heller L, Pagliaro LC, Pettaway CA. Neoadjuvant chemotherapy followed by aggressive surgical consolidation for metastatic penile squamous cell carcinoma. J Urol. 2007;177(4):1335-1338. 126. Ozsahin M, Jichlinski P, Weber DC, et al. Treatment of penile carcinoma: to cut or not to cut? Int J Radiat Oncol Biol Phys. 2006;66(3):674-679. 127. Bouchot O, Rigaud J, Maillet F, Hetet JF, Karam G. Morbidity of inguinal lymphadenectomy for invasive penile carcinoma. Eur Urol. 2004;45(6):761-765; discussion 765-766. 128. d’Ancona CA, de Lucena RG, Querne FA, Martins MH, Denardi F, Netto NR Jr. Long-term followup of penile carcinoma treated with penectomy and bilateral modified inguinal lymphadenectomy. J Urol. 2004;172(2):498-501; discussion 501. 129. Brown CT, Minhas S, Ralph DJ. Conservative surgery for penile cancer: subtotal glans excision without grafting. BJU Int. 2005;96(6)911-912. 130. Burgers JK, Badalament RA, Drago JR. Penile cancer. Clinical presentation, diagnosis, and staging. Urol Clin North Am. 1992;19(2):247-256. 131. Busby JE, Pettaway CA. What’s new in the management of penile cancer? Curr Opin Urol. 2005;15(5):350-357.

137. Moemen MN, Hamed HA, Kamel II, Shamloul RM, Ghanem HM. Clinical and sonographic assessment of the side effects of intracavernous injection of vasoactive substances. Int J Impot Res. 2004;16:143-145. 138. Horger DC, Wingo MS, Keane TE. Partial segmental thrombosis of corpus cavernosum: case report and review of world literature. Urology. 2005;66(1):194. 139. Zandrino F, Musante F, Mariani N, Derchi LE. Partial unilateral intracavernosal hematoma in a long-distance mountain biker: a case report. Acta Radiol. 2004;45(5): 580-583. 140. Ptak T, Larsen CR, Beckmann CF, Boyle DE Jr. Idiopathic segmental thrombosis of the corpus cavernosum as a cause of partial priapism. Abdom Imaging. 1994;19(6):564-566. 141. Al Saleh BM, Ansari ER, Al Ali IH, Tell JY, Saheb A. Fractures of the penis seen in Abu Dhabi. J Urol. 1985; 134(2):274-275. 142. Mydlo JH, Harris CF, Brown JG. Blunt, penetrating and ischemic injuries to the penis. J Urol. 2002;168(4 Pt 1): 1433-1435. 143. Bertolotto M, Neumaier CE. Penile sonography. Eur Radiol. 1999;9 Suppl 3:S407-S412. 144. Choi MH, Kim B, Ryu JA, Lee SW, Lee KS. MR imaging of acute penile fracture. Radiographics. 2000;20(5):1397-1405. 145. Beysel M, Tekin A, Gürdal M, Yücebaş E, Sengör F. Evaluation and treatment of penile fractures: accuracy of clinical diagnosis and the value of corpus cavernosography. Urology. 2002;60(3):492-496. 146. Bertolotto M, Mucelli RP. Nonpenetrating penile traumas: sonographic and Doppler features. AJR Am J Roentgenol. 2004;183(4):1085-1089. 147. Kervancioglu S, Ozkur A, Bayram MM. Color Doppler sonographic findings in penile fracture. J Clin Ultrasound. 2005;33(1):38-42. 148. El-Bahnasawy MS, Gomha MA. Penile fractures: the successful outcome of immediate surgical intervention. Int J Impot Res. 2000;12(5):273-277. 149. Fedel M, Venz S, Andreessen R, Sudhoff F, Loening SA. The value of magnetic resonance imaging in the diagnosis of suspected penile fracture with atypical clinical findings. J Urol. 1996;155(6):1924-1927. 150. Maubon AJ, Roux JO, Faix A, Segui B, Ferru JM, Rouanet JP. Penile fracture: MRI demonstration of a urethral tear associated with a rupture of the corpus cavernosum. Eur Radiol. 1998;8(3):469-470.

772

Diagnostic Abdominal Imaging

151. Uder M, Gohl D, Takahashi M, et al. MRI of penile fracture: diagnosis and therapeutic follow-up. Eur Radiol. 2002;12(1): 113-120. 152. Forman HP, Rosenberg HK, Snyder HM 3rd. Fractured penis: sonographic aid to diagnosis. AJR Am J Roentgenol. 1989;153(5):1009-1010. 153. Schmidt BA, Schwarz T, Schellong SM. Spontaneous thrombosis of the deep dorsal penile vein in a patient with thrombophilia. J Urol. 2000;164(5):1649. 154. Shapiro RS. Superficial dorsal penile vein thrombosis (penile Mondor’s phlebitis): ultrasound diagnosis. J Clin Ultrasound. 1996;24(5):272-274.

155. Shweta Bhatt HG, Vikram Dogra. Sonographic evaluation of scrotal and penile trauma. Ultrasound Clin. 2007;2:45-56. 156. James Halls GB, Uday Patel. Erectile dysfunction: the role of penile Doppler ultrasound in diagnosis. Abdom Imaging. 2009;34:712-725. 157. Hossein Sadeghi-Nejad DB, Vikram Dogra. Male erectile dysfunction. Ultrasound Clin. 2007;2:57-71. 158. Greene FL, Page, DL, Felming ID, Fritz, AG, Balch, CM, Haller DG, Morrow M, eds. AJCC Cancer Staging Handbook. 6th ed. New York: Springer-Verlag; 2002;331-335 [chapter 33].

CHAPTER

13

Imaging of the Lymph Nodes and Lymphatic Ducts Narainder K. Gupta, MD, DRM, MSc, FRCR Wallace T. Miller Jr., MD

I. NORMAL LYMPH NODES ANATOMY AND DRAINAGE PATTERNS a. Parietal Abdominal Lymphatics b. Visceral Abdominal Lymphatics i. Preaortic lymph nodes ii. Celiac lymph nodes iii. Superior mesenteric nodes iv. Inferior mesenteric nodes v. Lateral aortic lymph nodes c. Pelvic Lymphatics and Lymph Nodes d. Parietal Pelvic Lymphatics e. Visceral Pelvic Lymphatics i. External iliac lymphatics and lymph nodes ii. Internal iliac lymphatic and lymph nodes iii. Common iliac lymphatics and lymph nodes f. Drainage Patterns of the Abdominal Organs II. LYMPH NODE CHARACTERISTICS INDICATING DISEASE a. Increased Size and/or Increased Number b. Increased Metabolic Activity c. Unique Lymph Node Characteristics i. Lymph node calcification ii. Hyperenhancing lymph nodes iii. Low-attenuation lymph nodes iv. Lymph nodes with decreased MRI signal III. DISEASES AFFECTING LYMPH NODES a. Causes of Lymphadenopathy Affecting More Than 1 Drainage Distribution i. Lymphoma

NORMAL LYMPH NODES ANATOMY AND DRAINAGE PATTERNS Lymph nodes can be found throughout the body and normal lymph nodes may be identified noninvasively. Abnormal lymph nodes are found in many disease processes, including but not limited to malignancy, inflammation, and infection. Abnormalities of lymph node size, numbers, distribution, and imaging characteristics

ii. Leukemia iii. Systemic mastocytosis b. Metastatic Carcinoma and Sarcoma c. Inflammatory Diseases Causing Adenopathy in More than 1 Drainage Distribution i. Disseminated granulomatous infections ii. Systemic inflammatory disorders d. Amyloidosis e. Causes of Lymphadenopathy Affecting a Single Drainage Distribution i. Lymphoma ii. Metastatic tumor iii. Regional lymphadenopathy due to inflammatory conditions f. Focal Abdominopelvic Lymphadenopathy Without Other Abdominal Findings g. Castleman Disease IV. ANATOMY AND IMAGING OF THE LYMPHATIC DUCTS a. Anatomy b. Imaging Studies i. Lymphography ii. Lymphoscintigraphy c. Disorders of the Lymphatic System i. Lymphedema ii. Lymphatic trauma

can be an important indicator of disease in the region of the abnormality. Careful review of the presence of lymph nodes is an important part of the evaluation of abdominal disease. Many inexperienced readers overemphasize the mere presence of lymph nodes. Small lymph nodes between 2 and 10 mm in short axis are normally observed in the abdomen. Faster scanning and thin collimation helps one distinguish between lymph nodes and vessels, which previously was more challenging. Scrolling 773

774

Diagnostic Abdominal Imaging

through images also greatly aids the ability to distinguish lymph nodes from vessels. Lymph nodes appear as round or oval soft-tissue masses that appear and disappear over several images. These can be found in a variety of locations including behind the diaphragmatic crura (retrocrural lymph nodes); adjacent to the aorta; inferior vena cava; common, internal, and external iliac arteries and veins (retroperitoneal and pelvic lymph nodes); adjacent to the celiac axis (celiac axis lymph nodes); along the course of the portal vein (periportal lymph nodes); in the small bowel mesentery (mesenteric lymph nodes); and in the inguinal region. In the majority of cases, lymph node abnormalities are secondary to diseases in the organs that the lymph nodes drain and therefore, discovery of enlarged or otherwise abnormal lymph node should lead to a search for pathology in the drainage distribution of the abnormal lymph nodes. Although the arrangement of abdomen and pelvic lymphatics and draining lymph nodes is complicated, there

is a common principle of drainage, which will be explained in the text below. Abdominal lymphatics are divided into parietal and visceral lymphatic vessels and nodes and drain the abdominal wall and abdominal viscera, respectively (see Figure 13-1). These lymphatics and lymph nodes follow the course of parietal and visceral branches of abdominal aorta and drain into venous blood via thoracic duct. Some of the lymphatics from liver drain via alternate routes.1,2 Before draining into the thoracic duct, most of the lymphatics of the abdominal cavity are interrupted by retroperitoneal nodes called terminal lumbo-aortic nodes around the inferior vena cava and abdominal aorta.3,4

Parietal Abdominal Lymphatics The superficial parietal lymphatics from the anterior and posterior abdominal skin and subcutaneous tissues drain cranially into the pectoral and subscapular axillary nodes.

Abdominal lymphatics & lymph nodes

Visceral lymph nodes

Parietal lymph nodes

Superficial

Above umblicus

Pectoral & subscapular axillary lymph nodes

Lymph nodes close to organ

Deep

Intermediate lymph nodes

Below umblicus

Lumbar peri-aortic lymph nodes Superficial inguinal lymph nodes

Anterior lymphatics

Subdiaphragmatic nodes

External lnguinal lymph nodes

Median group

Right peri-aortic

Left peri-aortic

Posterior aortic

Posterior lymphatics

Lateral lumbo-aortic

Posterior lumbo-aortic

Figure 13-1 Patterns of Abdominal Lymphatic Drainage Basic scheme and principle of abdominal lymphatic drainage is illustrated in this diagram. Note that the drainage of the superficial parietal abdominal tissues above the umbilicus is superior to the diaphragm. Green = drainage from the viscera; gray = drainage from the parietes.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 775 However, below the umbilicus, the parietal drainage is to superficial inguinal lymph nodes (see Figure 13-2).5 The deep lymphatic vessels from the muscles and fasciae of abdomen run in the subperitoneal adipose tissue and superiorly drain into subdiaphragmatic nodes and inferiorly follow deep inferior epigastric vessels to drain  into

A

external iliac lymph nodes. From the posterior abdominal wall, the lymphatics drain into lateral or posterior lumboaortic lymph nodes.1

Visceral Abdominal Lymphatics Figure 13-3 shows the normal distribution of abdominal lymph nodes. Lymphatic drainage from various abdominal viscera drain firstly to lymph nodes close to particular viscera, then to lymph nodes of peritoneal ligaments and mesentery and drain into the lymph nodes of paired or unpaired branches of the abdominal aorta, which finally drain to periaortic nodes in the lumbar area (see Figure 13-3). The lumbar periaortic lymph nodes are divided into preaortic, postaortic, and left and right lateral aortic lymph nodes.4 The median preaortic lymph nodes drain the lymphatics of the gastrointestinal (GI) tract that are supplied by the ventral branches of the abdominal aorta. These drain into the cisterna chyli.1 The left and right lateral aortic lymph nodes receive lymphatics from the common iliac lymph nodes and lymphatics running along the lateral branches of the aorta, from the kidneys and adrenal glands, as well as male and female gonads. Hence, these lateral aortic nodes are main groups of drainage from the urogenital viscera in abdominopelvic cavity. These lateral aortic nodes drain into the cisterna chyli via paired left and right lumbar lymphatic trunks.1,5 The posterior aortic lymph nodes do not directly drain any viscera and usually serve as communications between left and right lateral aortic lymph nodes along with draining posterior deep parietes and then drain into the thoracic duct.2,3

Preaortic lymph nodes B

The preaortic groups of lymph nodes is located anterior to the aorta and are named after 3 unpaired vessels originating from the abdominal aorta named celiac, superior mesenteric, and inferior mesenteric nodes (see Figures 13-3 and 13-4).3

Celiac lymph nodes

C Figure 13-2 Lymphatic Drainage of the Skin and Subcutaneous Tissues of the Trunk The CT sections illustrating the draining lymph nodes from the anterior and posterior abdomen skin and subcutaneous tissue above the umbilicus. A and B. Above the umbilicus, the drainage is into subpectoral and subscapular nodes. C. Below the umbilicus, the drainage is into superficial inguinal lymph nodes. Green = pectoral lymph nodes; red = subscapular lymph nodes; blue = inguinal lymph nodes

The celiac lymph nodes drain lymphatics from the stomach, duodenum, most of the liver, gallbladder, spleen, and pancreas. The intermediate nodes for these organs are lined close to their supplying arteries and are called gastric, hepatic, and pancreaticosplenic nodes.5,6 Gastric nodes are found along the arterial vessels along the lesser and greater curvature of the stomach and are called right and left gastric nodes. On the lesser curvature side of the stomach, the lymph nodes lie in the lesser omentum, as right and left gastroepiploic nodes lying in the greater omentum in the lower portion of the greater curvature of the stomach. Cranial lymph nodes of the left gastric chain lie near the cardia and receive drainage from the abdominal portion of the esophagus. These predominantly drain into the celiac lymph nodes; however part of them is drained into the posterior lower mediastinal lymph nodes. A group of pyloric nodes 4 to 5 in number and lying close to the division of gastroduodenal artery receive

776

Diagnostic Abdominal Imaging

A

B

C

D

E

F

G

H

I

Figure 13-3 Normal Lymph Nodes of the Abdomen and Pelvis A-I. Axial CT of the abdomen demonstrates multiple small normal lymph nodes in the abdomen and pelvis. The celiac axis lymph nodes represent a type of preaortic lymph node; the para-aortic and interaortocaval lymph nodes are types of right

and left lateral aortic lymph nodes. gh = gastrohepatic; ca = celiac axis; pp = periportal, iac = interaortocaval; pa = para-aortic; m = mesenteric; ei = external iliac; ob = obturator; ii = internal iliac; ing = inguinal.

afferent lymphatics from the pylorus, first part of the duodenum and pancreatic head and receive efferents from the right gastroepiploic nodes. These pyloric nodes usually drain into the celiac group of lymph nodes. Alternatively, these can drain into the superior mesenteric group of lymph nodes. Hepatic nodes not only drain liver, bile ducts, and gallbladder but also receive efferents from stomach, duodenum, and pancreas. These nodes, 3 to 6 in number, are situated along the hepatic artery. The first hepatic nodes are situated at the origin of the hepatic artery and correspond to the superior border of the pancreas. Middle hepatic nodes are located along the anterior surface of the portal vein. The superior hepatic nodes are situated in the hepatic hilum and randomly distributed along left and right hepatic arteries. One of these

hepatic nodes is constant in location at the junction of cystic and common bile duct and called the Quénu cystic node. Splenic lymph nodes associated with the splenic artery are called pancreaticosplenic nodes and therefore situated along the superior and posterior part of the pancreas. The largest of these nodes lie behind the body of the pancreas. Lateral smaller pancreaticosplenic nodes lie near the hilum in the pancreaticosplenic ligament. These nodes receive lymphatics from the spleen, body, and tail of the pancreas and gastric fundus. The efferent drainage from the pancreaticosplenic nodes is into the celiac group.7

Superior mesenteric nodes This group of lymph nodes surrounds the origin of the superior mesenteric artery, anterior to the aorta. They are

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 777

A

B

Figure 13-4 Preaortic and Mesenteric Lymph Nodes A. Labeled illustration shows preaortic lymph nodes named after unpaired branches of abdominal aorta. Posterior aortic, left lateral aortic, and right lateral aortic lymph nodes are not shown in this picture. Major pelvic lymph node groups are shown. Purple = common and external iliac lymph nodes; blue = presacral lymph nodes; yellow = superior mesenteric (SM) artery

lymph nodes; green = inferior mesenteric (IM) artery lymph nodes; pink = internal iliac lymph nodes; red= celiac axis lymph nodes. B. Coronal maximum-intensity projection showing group of celiac, superior mesenteric, and inferior mesenteric lymph nodes shown in red, yellow, and green, respectively. Different subtle shades of the same color represent the drainage territory and lymph nodes.

situated behind the pancreas and anterior to the aorta at the level of L1 lumbar vertebra. These lymph nodes are almost contiguous with the mesenteric lymph nodes at the root of mesentery. These lymph nodes drain the mesenteric and ileocolic lymph nodes, thus draining the distal duodenum, small intestine, and right-sided colon.4,5 Mesenteric lymph nodes approximately 100 to 150 in numbers are located in the mesenteric fat. Most peripheral of mesenteric nodes are located close to the intestinal wall between the terminal jejunal and ileal arteries and these are called juxtaintestinal mesenteric nodes. The intermediate or second group of mesenteric lymph nodes lie within the mesentery between the primary and secondary loops of the superior mesenteric artery and the last central mesenteric nodal group or the third group, which are slightly larger than the other 2 groups of mesenteric lymph nodes, lie along the main stem of superior mesenteric artery near the mesenteric root where these are indistinguishable from superior mesenteric lymph nodes.1 Ileocolic nodes are situated around the ileocolic artery and there are approximately 20 in number; 1 of these nodes

is usually found in the mesoappendix (see Figures 13-3 and 13-4).1

Inferior mesenteric nodes Inferior mesenteric nodes are usually a group of 2 lymph nodes situated on either side of origin of inferior mesenteric artery. These are usually at the level of L3 lumbar vertebra and drain the lymphatics from the upper rectum and left side of colon via epicolic, paracolic, and intermediate colic nodes. The epicolic nodes are situated in the walls of colon itself, paracolic nodes are located along the mesenteric borders of the colon and intermediate colic nodes are located along the middle and left colic arteries. Similarly, the ascending colon and part of transverse colon have similar arrangement of drainage but end in superior mesenteric nodes instead of inferior mesenteric nodes.1,2,5

Lateral aortic lymph nodes Lateral aortic lymph nodes on the left are a continuous vertical chain of lymph nodes along the left side of abdominal

778

Diagnostic Abdominal Imaging

aorta. These lymph nodes lie on vertebral attachments of the psoas muscle and left crus of diaphragm. Left renal vessels and sympathetic nervous trunks cross this chain anteriorly.1 Right lateral aortic lymph nodes are located in the front of the inferior vena cava or behind it. Few of these can be lateral to the inferior vena cava or lie between the aorta and inferior vena cava. According to their relationship with the inferior vena cava, these are named as precaval, laterocaval, or postcaval lymph nodes. Their relationship otherwise is similar as the left counterpart.2,5 The lateral aortic lymph nodes through tortuous networks not only receive lymphatic drainage from the structures supplied by the posterior and lateral paired branches of the abdominal aorta but also drain pelvic lymphatics through the common iliac nodes. These also drain adrenal glands and lymphatics running along the renal, suprarenal, and diaphragmatic vessels. Lateral aortic lymph nodes also drain lymph from the kidneys, perirenal fat, renal capsule, and abdominal ureter. Gonadal lymphatics from the ovaries and testes as well as the fallopian tubes lymphatics drain in these lymph nodes.8

Functionally, the proximal lymphatics from the digestive tract cannot be easily demarcated because of complex 3-dimensional drainage; however, distally the lymphatics drain into preaortic pathways of celiac, superior, and inferior mesenteric lymph nodes. However, drainage from genitourinary lymphatic pathways is more precise to the lateral aortic ascending chains.

Pelvic Lymphatics and Lymph Nodes Similar to the abdominal lymphatics and lymph nodes, the pelvic lymphatics and lymph nodes are also divided into parietal and visceral networks. All lymphatics drain in the successive group of lymph nodes located at the level of pelvic inlet along the arcuate line of pelvis and L5 lumbar vertebra. These are mostly associated with iliac vessels and their branches and combine to form the ascending chains and then drain on the lateral aortic chains on the respective sides. Ascending chains are named after their location and called external iliac, internal iliac, common iliac, and sacral groups of lymph nodes (see Figures 13-3 and 13-5).9

Pelvic lymphatics & lymph nodes

Visceral lymph nodes

Parietal lymph nodes

Superficial

Juxta-visceral lymph nodes

Deep

Pre-, lateral-, post-, sub-vesical Para-vaginal and para-uterine Para-rectal

Superomedial group of superficial inguinal nodes

Intermediate lymph nodes

Inferior epigastric lymph nodes

Circumflex iliac lymph nodes

Sacral group of lymph nodes

Lower internal iliac lymph nodes

External iliac, internal iliac and sacral lymph nodes

Common iliac lymph nodes External iliac lymph nodes

Internal iliac lymph nodes

Figure 13-5 Patterns of Pelvic Lymphatic Drainage Basic scheme and principle of pelvic lymphatic drainage is illustrated in this diagram. Note that the drainage principle is similar to the abdominal drainage principle and lymph nodes groups follow the names of accompanying blood vessels.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 779

Parietal Pelvic Lymphatics Parietal lymph vessels and nodes drain the lymph from all the pelvic walls and are arranged in superficial and deep lymphatic network that drain the superficial softtissue of perineum and the muscles covering the pelvis respectively.2,3,9 Superficial network on the pelvic floor runs from the coccyx region to pubic region to drain in the superomedial group of superficial inguinal nodes. Functionally, these drain all soft-tissues of perineum below the external fascial sheath of urogenital diaphragm, distal vagina below the hymen, and the inferior part of the anal canal below the anocutaneous line.1 Deep parietal lymphatics follow the parietal branches of the external and internal iliac vessels and first drain into the inferior epigastric, circumflex iliac, and sacral nodal groups. Deep inferior nodes drain the lower anterior abdominal wall and retropubic region of the anterior pelvic wall. Up the hierarchy, these drain into lateral groups of external iliac nodes. The deep circumflex iliac nodes are located around the deep circumflex iliac artery and receive afferent vessels arising from the iliac muscle and the parietal peritoneal lining of the iliac fossa and then drain into external iliac nodes. The sacral groups of lymph nodes are situated around the lateral and median sacral arteries and constitute 3 ascending lymph chains running, respectively, one each along the lateral borders of the sacrum and third in front of its anterior aspect on the midline. The sacral group of lymph nodes drain the presacral space between the fascia recti anteriorly and the sacrum posteriorly. The largest of the median sacral nodes is known as the promontorial node as it sits anterior to disc space at L5/S1.8 Lateral pelvic wall lymphatics run on endopelvic fascia and drain to external and internal iliac nodes above the plane of levator ani and coccygeal muscles. Below the plane of the levator ani, the lymphatic vessels, which follow the internal pudendal artery along the surface of the obturator internus, drain the muscles and fasciae. These deep lymphatics originating in the prevesical space also collect the lymph vessels from the ischiorectal fossa, and then drain into the internal iliac chain.9

Visceral Pelvic Lymphatics Similar to the abdominal viscera, the pelvic viscera first drain to closely located juxtavisceral lymph nodes, then along the vascular pedicles of each organ, and finally along the iliac vessels. At iliac levels, rich and extensively developed lymphatic plexus form ascending pathways draining toward the lateral aortic chains. The juxtavisceral nodes are named in relation to the organ being drained.4 For the urinary bladder, these are called prevesical, lateral vesical, posterior vesical, and subvesical corresponding to the respective bladder surfaces. Pararectal lymph nodes are named left and right and situated in the pararectal fat of posterior digestive pelvic

compartment. Paravaginal and parauterine nodes correspond to the lymph nodes found on the lateral edges of vagina and cervix and situated in the parametrial fibrous tissue of the female pelvis. After receiving drainage from the neighboring organs, these nodes drain to the external iliac, internal iliac, or presacral chain of lymph nodes.9

External iliac lymphatics and lymph nodes External iliac lymph nodes are a group of approximately 10 lymph nodes arranged around the external iliac vessels. These lymph nodes form 3 distinct groups called the lateral, medial, and middle group of external iliac lymph nodes. The lowest lymph node of the lateral group is located under the inguinal ligament. This group drains from the deep inferior epigastric and deep circumflex iliac lymph nodes situated along corresponding arteries. The medial group of the external iliac lymph nodes are situated medially to external iliac vessels, predominantly drain the lower limb, and there are very few draining lymphatics from the pelvis itself. This group is also called obturator nodes and should not be confused with the isolated small obturator nodes that sit in the obturator foramen of the obturator canal in the obturator fossa. These small lymph nodes are linked to the internal iliac lymph nodes rather than the external iliac group.2,5,9 The external iliac lymph nodes group receives lymph from the lower limb through the superficial and deep inguinal lymph nodes. These also drain the subumbilical part of the abdominal wall, glans penis or clitoris, muscles of the medial compartment of the thigh, lateral lobes of the prostate, vesical fundus, cervix uteri, and upper vagina.3,9 Because of fetal positioning of prostate, vagina, and cervix uteri at the level of pelvic inlet, these organs develop drainage into nearby medial nodes of the external iliac chain rather than primarily draining in the internal iliac group of nodes. Similarly, this process explains the lymphatics of ovaries and testes, which drain in the lower lateral aortic nodes because of their lumbar origin fetally.5 The lymphatic efferents from the external iliac chain drain to corresponding lower lymph nodes of the common iliac chain (see Figures 13-3 and 13-6).

Internal iliac lymphatic and lymph nodes These lymph channels and lymph nodes are also called hypogastric lymphatics and lymph nodes. These surround the internal iliac artery and their corresponding branches. These are named superior-gluteal lymph nodes, uterine, internal pudendal, inferior gluteal, and middle rectal arteries.1,4 The internal iliac lymph nodes drain lymph from all the pelvic organs including the posterior prostate, lateral and lower part of the bladder, membranous and prostatic urethra, seminal vesicles, lower two-thirds of the vagina, the uterine body, and midrectum.4,6,10 Superior gluteal lymph nodes also draining the deep gluteal regions including glutei muscles. Internal pudendal lymph nodes drain to the internal iliac after receiving lymph from the deep

780 Diagnostic Abdominal Imaging abdomen in order to prevent overcalling the presence of disease. It is also important for the reader to be aware of the normal lymphatic drainage patterns for the organs of the abdomen. Malignancies and inflammatory conditions of each organ will commonly result in mild or moderate lymphadenopathy in the drainage distribution of the organ. These drainage distributions should be inspected for the presence of disease when an abnormality within the organ is discovered or vice versa to look for a diseased organ when lymphadenopathy is found. A list of the drainage distributions for many of the organs in the abdomen is shown (Table 13-1).

Table 13-1. Common Lymphatic Drainage Patterns

Figure 13-6 Pelvic Lymph Node Chains Volume-rendered image of the pelvis illustrates the major groups of pelvic lymph nodes and named after the vessels adjacent to which these lie.

perineum, ischioanal fossa, and lower parts of the vagina, prostrate, and rectum.5,7 Internal iliac lymphatics then travel upwards and drain into the intermediate group of common iliac lymph nodes.

Common iliac lymphatics and lymph nodes Four to 7 in numbers, these nodes depending on their location are divided into medial, lateral, and intermediate groups.4 Lateral common iliac lymph nodes lie between the medial border of psoas and lateral to the common iliac artery, and merges with the lateral aortic lymph nodes without clear demarcation. The middle or intermediate common iliac lymph nodes lie posteromedial to artery. The medial chain lies inner to the common iliac artery and travels upwards to meet its counterpart from the other side to form an uneven group just below the aortic bifurcation at the level of the L5 vertebra. The lateral and intermediate common iliac chains do not receive any direct lymphatics from the pelvic viscera. However, some lymphatics originating from the bladder neck, cervix uteri, and posterior rectum drains directly into the median subaortic group of lymph nodes.

Drainage Patterns of the Abdominal Organs Importance of proper evaluation of lymph nodes cannot be stressed enough and, therefore, it is crucial that an inexperienced reader acquires a feel for the normal variation in the distribution, number, and size of lymph nodes in the

Liver

Periportal – upper para-aortica Right paracardiac

Stomach

Celiac axis – upper para-aortic Gastrohepatic Perigastric

Pancreas Duodenum

Celiac axis – upper para-aortic

Right kidney Right adrenal Right ovary Right testicle

Right mid para-aortic

Left kidney Left adrenal Left ovary Left testicle

Left mid para-aorticb

Jejunum Ileum

Small bowel mesentery – upper para-aortic

Cecum Ascending colon

Pericecal – small bowel mesentery – upper para-aortic

Transverse colon

Transverse mesocolon – upper para-aortic

Descending colon Pericolonic fat LN – inferior Sigmoid colon para-aorticc Rectum Prostate Uterus Cervix

Retroperitoneal fat LN – internal iliac/obturator – inferior para-aortic

Anus Perineum Skin of scrotum

Inguinal

aUpper para-aortic = from diaphragm to renal arteries bMid para-aortic = 1-2 cm above and below the origins of the renal

arteries cLower para-aortic = from renal arteries to the iliac bifurcation

Note: Dashes between lymph node groups indicate sequential spread from first to last groups listed.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 781

LYMPH NODE CHARACTERISTICS INDICATING DISEASE Lymph node characteristics that can indicate disease include increased size, increased number, increased metabolic activity, increased or decreased attenuation, hyperenhancement, and decreased magnetic resonance imaging (MRI) signal.

Imaging Notes 13-1. The One-Centimeter Rule Use of a short axis diameter of 1 cm as cutoff to distinguish normal lymph nodes from diseased lymph nodes results in very poor clinical accuracy and should be avoided.

Increased Size and/or Increased Number The 2 most important characteristics of lymph nodes indicating the presence of disease is an increase in size or number of lymph nodes. Enlarged lymph nodes can be reactive, because of benign or malignant infiltration or lymphoproliferative disorders (Table 13-2). Many radiologists rely on the “1-cm” rule, a short axis diameter of greater than 1 cm, as a marker for “pathologic” enlargement of lymph

nodes. The attraction of this rule is its simplicity of use. Unfortunately, numerous studies of a wide range of diseases throughout the body have shown the 1-cm rule to be seriously flawed. When used as a marker for lymphatic spread of cancer, wide ranges in sensitivity and specificity have been reported for the variety of cancers.11-30 The poor sensitivity of the 1-cm rule is not surprising because

Table 13-2. Causes of Lymphadenopathy Infectious causes

Reactive lymphadenopathy

Viral

Inf. Mononucleosis, Rubella

Bacterial

Pyogenic, cat-scratch disease

Mycobacterial

Tuberculosis and atypical mycobacterium

Spirochetal

Treponema pallidum, leptospirosis

Chlamydial

Lymphogranuloma venereum

Parasitic

Toxoplasmosis

Fungal

Coccidioidomycosis

Noninfectious causes Sarcoidosis Connective tissue disorders [Systemic lupus erythematosus (SLE), RA, mixed connective tissue disease] Kawasaki disease Rosai-Dorfman disease Kikuchi disease Castleman disease Drug reaction/hypersensitivity (phenytoin) Malignant Metastatic carcinoma Infiltrative diseases

Metastatic melanoma Leukemia Germ cell tumor Lymphomas - Hodgkin, Non-Hodgkin

Primary lymphoproliferative diseases

Lymphomatoid granulomatosis Angioimmunoblastic lymphadenopathy Malignant histiocytosis

Nonmalignant Lipid storage disease eg, Gaucher disease Amyloidosis

782

Diagnostic Abdominal Imaging

microscopic spread of tumor will not change lymph node size. Some normal lymph nodes can be larger than 1 cm and the presence of inflammation in the drainage distribution of the lymph node can also result in an increase in diameter. Consequently, the specificity of the 1-cm rule for malignant spread is also relatively poor. Some researchers have attempted to increase sensitivity for spread by lowering the “pathologic cut-off” to a variety of subcentimeter values. This approach consistently increases sensitivity by reducing specificity and rarely results in a clinically useful accuracy in the diagnosis of lymph node metastasis. With the widespread availability of positron emission tomography (PET)–computed tomography (CT), there is a paradigm shift in how the lymph nodes are evaluated at present. It has been documented that PET is generally more sensitive and specific than CT, MRI, or other imaging methods for the detection of cancer.31-34 In this text, we would like to offer a new paradigm in the prediction of lymph node pathology that relies on a combination of lymph node size and the clinical pretest probability of disease in any lymph node distribution. The 1-cm rule as a gross rule of thumb is the starting point. The majority of small lymph nodes, those with a short axis less than 10 mm, will be normal and the majority of lymph nodes with a short axis greater than or equal to 10 mm will indicate the presence of pathology either in the drainage distribution of the lymph node or systemically within the lymph nodes and the majority of lymph nodes. However, there are important caveats: (1) When there is a known malignancy in the drainage distribution of a lymph node, any size lymph node can potentially harbor malignancy, and any small changes in lymph node number, location, or appearance should be viewed with suspicion. (2) One or 2  intermediate lymph nodes, those with a short axis between 10 and 20 mm, are indeterminate in significance.

Imaging Notes 13-2. New Paradigm for the Determination of Lymph Node Pathology Short axis ≤10 mm No cancer in drainage distribution Cancer in drainage distribution

Usually normal Indeterminate significance Use other imaging criteria to determine significance

Short axis 11-20 mm

Indeterminate significance Use clinical history and other imaging criteria to determine significance

Short axis >20 mm

Nearly always indicated disease Usually lymphoma, CLL, or metastasis

In the absence of a known disease in the drainage distribution of the lymph node. These intermediate lymph nodes will most often represent a normal variant. (3) Large lymph nodes, those with a short axis of greater than 20 mm, will virtually always indicate an underlying disease, most often a malignancy. Lymph nodes with short axis less than or equal to 10 mm we will call “small lymph nodes” and will usually represent a normal variant. However, any lymph node found in a location where lymph nodes are not normally present should we evaluated for possible disease. For example, small lymph nodes in the perirectal fat are not seen in normal individuals and will usually indicate a malignancy of the nearby tissues such as the rectum, cervix, or prostate (see Figure 13-7). Therefore, it is imperative for the reader to become familiar with the normal distribution of abdominal and pelvic lymph nodes. Furthermore, all lymph nodes, regardless of size, are potentially significant when they are found in the drainage distribution of a known cancer. In the setting of a cancer, any small variation from normal should be suspected to harbor metastasis. This includes an increase in number of small lymph nodes or lymph nodes in a location where they are not normally encountered. Large lymph nodes, those greater than 20 mm in diameter, will in the majority of cases indicate the presence of malignancy in the drainage distribution of the lymph node or a lymphoproliferative malignancy such as Hodgkin lymphoma (HL), non-Hodgkin lymphoma (NHL), and chronic lymphocytic leukemia (CLL) (see Figure 13-8). The inflammatory disorders that can also result in lymph nodes with short axis greater than 20 mm are sarcoidosis; granulomatous infections with tuberculous or nontuberculous organisms and endemic fungi such as histoplasma, coccidioides, and cryptococcal species; and connective tissue disorders, especially systemic lupus erythematosus and rheumatoid arthritis (Table 13-3). Lymph nodes with short axis between 11 and 20 mm we will call “intermediate lymph nodes.” The significance of intermediate lymph nodes is dependent on the patient’s clinical history and other characteristics of the lymph nodes, most importantly the number of intermediate lymph nodes. If the patient has a malignancy or inflammatory condition in the drainage distribution of the intermediate lymph nodes, then this mild lymph node enlargement will often be a result of the disorder. For example, it is common for patients with cirrhosis to have mildly enlarged, that is, intermediate, lymph nodes in the porta hepatis (see Figure 13-9). Similarly, intermediate lymph nodes in the celiac axis distribution of a patient with gastric carcinoma will have a high probability of harboring metastasis. On the other hand, 1 or 2 intermediate lymph nodes are commonly found in normal individuals and in the absence of a known disorder in the drainage distribution of the lymph node will usually represent a normal variant (see Figure 13-10). The number of intermediate lymph nodes is also an important characteristic in identifying the clinical

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 783

A

B

Figure 13-7 Perirectal Lymph Node Metastasis in Rectal Cancer This 55-year-old woman had blood in her stool. A. Enhanced CT of the pelvis shows asymmetric thickening of the rectal wall indicating the presence of a rectal carcinoma. Note the streaking of the perirectal fat, a finding suggesting local soft-tissue invasion.

B. CT images 1 cm higher than in A shows 5 mm × 9 mm lymph node (arrow) in the perirectal fat. Although this lymph node has a short axis less than 10 mm, lymph nodes are not normally seen in this location. The presence of any perirectal lymph nodes will usually indicate the presence of disease in the region.

significance of intermediate lymph nodes. More than a few intermediate lymph nodes in any nodal group is usually abnormal and will indicate pathology in the drainage distribution of the nodes or a systemic disorder of lymph nodes

(see Figure 13-11). There are also some locations where intermediate lymph nodes are a relatively common normal variant. This is especially true of inguinal lymph nodes and para-aortic lymph nodes between the celiac axis and the

A

B

Figure 13-8 Lymph Node Metastasis From Testicular Carcinoma This 35-year-old man had testicular carcinoma. A-C. Contrastenhanced CT shows infiltration of the fat of the right inguinal region typical of a previous right orchiectomy (arrows) and two markedly enlarged retroperitoneal lymph nodes (arrowheads) measuring 31 mm × 35 mm and 36 mm × 48 mm anterior to

C the inferior vena cava at the level of the kidneys. Lymph nodes of this size, even if few in number, will nearly always indicate malignancy. This is the characteristic location for lymphatic metastasis from testicular cancer because the lymphatic drainage travels with the venous supply to the gonads.

784

Diagnostic Abdominal Imaging

Imaging Notes 13-3. Clinical Significance of Intermediate Lymph Nodes (Long axis 11 mm-20 mm) 1-2 intermediate lymph nodes + No clinical disease in drainage distribution 1-2 intermediate lymph nodes + Known disease in drainage distribution >2 intermediate lymph nodes

Table 13-3. Disorders Commonly Causing Large Lymph Nodes (Short Axis >20 mm) 1. Metastatic tumor

Usually normal variant

2. Lymphoma 3. Chronic lymphocytic leukemia 4. Sarcoidosis

Lymph nodes usually a result of known disease Usually indicates disease Disease should be determined

5. Granulomatous infections a. Tuberculosis b. Nontuberculous mycobacterial infection c. Histoplasmosis d. Coccidioidomycosis e. Cryptococcosis

renal hilum, and therefore intermediate lymph nodes in these locations are often clinically not significant. In addition to increase in lymph node diameter, increased numbers of lymph nodes, more than is usually seen, can be a sign of disease. Increased numbers of large and intermediate-sized lymph nodes will usually indicate

the presence of lymphoma or metastatic tumor but can occasionally be due to inflammatory diseases. Increased numbers of small lymph nodes that involve multiple nodal groups is usually a sign of an inflammatory disorder but can also be a manifestation of malignancy (see Figures 13-12 and 13-13).

A

B

Figure 13-9 Periportal Lymphadenopathy due to Hepatitis This 51-year-old man with hepatitis C was being evaluated for the risk of hepatocellular carcinoma. A and B. Gadoliniumenhanced T1-weighted MRI sequences through the porta hepatis shows enlarged periportal lymph nodes (arrowheads) measuring

14 mm × 15 mm and 11 mm × 26 mm. No mass was found in the liver. Inflammatory diseases of the liver such as hepatitis and cirrhosis will commonly cause mild adenopathy in the periportal and celiac axis regions.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 785

A

B

C

D

E

F

Figure 13-10 Normal Variant Intermediate Lymph Node This 49-year-old woman complained of abdominal pain. A-F. Contrast-enhanced axial CT images show multiple normal-sized lymph nodes in the gastrohepatic (gh), periportal (pp), para-aortic (pa), mesenteric (m), external iliac (ei), and obturator (o) locations. However, there is also an intermediate-

sized superior mesenteric artery (SMA) lymph node (large arrow) in (C), measuring 12 mm × 15 mm. This was unchanged in size from a study 2 years previously. No other pathology was noted in the abdomen. This intermediate lymph node is likely to represent a normal variant.

Increased Metabolic Activity

used in the evaluation of the spread of malignancy and many individuals make the incorrect assumption that all increased metabolic activity indicates the spread of malignancy. Patients with a malignancy can have an unrelated inflammatory condition that results in the uptake of FDG. Similar to the interpretation of lymph node size, the significance of increased activity should be interpreted in the light of the patient’s clinical history. False-positive nonmalignant FDG uptake is common in patients with malignancy who have undergone surgery or received radiotherapy or a combination of the 2. Lymph node activity in the expected distribution of disease will have a high positive predictive value for the spread of malignancy. However, increased activities in areas of unexpected spread of disease have an indeterminate significance and warrants further workup or accelerated follow-up.

Metabolic activity, as measured by PET scan, is increasingly being used as a measure of lymph node disease. Numerous studies have demonstrated increased sensitivity of PET scan or PET-CT over CT scans in the detection of both neoplastic and inflammatory disorders.35 This is because PET scans have the opportunity to detect lymph node involvement by disease prior to enlargement of the lymph node (see Figure 13-14). Many studies have also suggested that PET scans have a higher specificity than CT scans.35 Causes of false-positive PET scans include muscular activity, brown fat, peristalsis of bowel, and synchronous inflammatory conditions (see Figure 13-15). With more prevalent use of PET-CT hybrid scanners, the false-positive rate has decreased.35 When using PET and PET-CT scans to diagnose diseases, it is important to understand that the imaging agent, fluorodeoxyglucose (FDG), is a glucose analogue and measures metabolic activity. This is a nonspecific measure of disease and does not indicate the cause of the increased metabolic activity. In most cases, PET and PET-CT is

Unique Lymph Node Characteristics Most normal and abnormal lymph nodes will have imaging characteristics similar to skeletal muscle with x-ray

786

Diagnostic Abdominal Imaging

A

B

C

D

Figure 13-11 Follicular Lymphoma This 54-year-old man presented with weight loss and diarrhea. A-D. Contrast-enhanced axial CT demonstrates multiple small and intermediate lymph nodes in the gastrohepatic (arrow in A, 7 mm × 12 mm), right para-aortic (arrow in B, 10 mm × 24 mm),

and mesenteric (arrow in C, 13 mm × 17 mm and arrow in D) 10 mm × 17 mm). Any one of these intermediate lymph nodes could be a normal variant; however, this many intermediate lymph nodes will usually signal disease. Lymph node biopsy was diagnostic of low-grade, follicular lymphoma.

A

B

Figure 13-12 Multiple Small Lymph Nodes due to Chronic Lymphocytic Leukemia (CLL) This 58-year-old woman had chronic lymphocytic leukemia. A-C. Contrast-enhanced axial images demonstrate 1 intermediatesized gastrohepatic lymph node (14 mm × 16 mm) (arrow) and

C many small para-aortic lymph nodes. This is an abnormal pattern of lymph node distribution and will usually be a manifestation of inflammatory disease but can occasionally be due to a malignancy such as CLL as in this case.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 787

A

B

C

Figure 13-13 Pericecal Adenopathy Due to Crohn Ileitis-Colitis This 31-year-old woman presented with nausea, vomiting, and diarrhea. A-C. Contrast-enhanced CT through the right lower quadrant demonstrates a thickened cecal wall (arrows)

and increase in number and size of pericecal lymph nodes (arrowheads). There is also streaking of the perirectal and mesenteric fat. Colonoscopic biopsy was diagnostic of Crohn disease.

attenuations similar to skeletal muscle on both enhanced and unenhanced CT and with T1-weighted, T2-weighted, and gradient echo signal characteristics similar to skeletal muscle. However, rarely lymph node pathology can be identified because of the presence of altered x-ray attenuation or signal intensity. Occasionally special techniques in MRI can be employed to characterize the tissues involved with malignancy and to measure response to therapy. These techniques include

MR spectroscopy to detect altered metabolism, diffusion MRI to detect cell proliferation, and dynamic MRI to detect angiogenesis and to identify hypoxia.36,37 Detailed description of these techniques is beyond the scope of this chapter.

A

B

Figure 13-14 Perirectal Small Lymph Node Metastasis A. Unenhanced CT from a PET-CT shows a tiny perirectal lymph node (arrow). In most cases this will represent a normal variant. B. However, PET-CT fused image shows this lymph node to be

FDG positive. In this 67-year-old patient with a known rectal carcinoma, this is highly likely to represent microscopic lymph node metastasis.

Lymph node calcification High attenuation within lymph nodes will usually be a result of dystrophic calcification within the lymph node. Dystrophic calcification, calcification of tissues due to cell death, is most

788

A

Diagnostic Abdominal Imaging

B

Figure 13-15 PET Activity Due to Brown Fat in 3 Patients A. PET-CT and (B and C) PET images demonstrate increased activity in the soft-tissues of the neck and in the paraspinal

C location (arrows). This is the typical appearance of brown fat and should not be confused with metabolically abnormal tissues.

A

B

C

D

E

F

Figure 13-16 Lymph Node Calcification in 3 Patients A-C. This 70-year-old woman was being evaluated for breast carcinoma. Contrast-enhanced CT demonstrates multiple calcified right lower quadrant mesenteric lymph nodes (arrows). Most lymph node calcifications will be secondary to prior granulomatous infection. The exact cause for these lymph nodes is not known but they have remained stable for many years. D. This 46-year-old woman is a known case of lymphoma and previously treated with chemotherapy. Calcified

mesenteric (arrowhead) and noncalcified right psoas (arrow) lymph nodes are seen. E and F. This 54-year-old woman had mucinous cystadenoarcinoma of the ovary. Contrast-enhanced CT demonstrates multiple calcified lymph nodes (arrow) in the upper retroperitoneum. These will most often be due to prior granulomatous disease. However, this patient’s primary tumor demonstrated coarse regions of calcification and these lymph nodes have slowly enlarged over 5 years, indicating widespread calcified lymph node metastasis.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 789 Table 13-4. Causes of Hyperattenuating Lymph Nodes A. Calcification 1. Granulomatous infection 2. Sarcoidosis 3. Treated lymphoma 4. Metastasis from calcified neoplasms B. Hyperenhancement 1. Hypervascular metastasis 2. Castleman disease C. Lymphangiographic contrast

often a result of prior granulomatous infection, prior sarcoidosis, or treated lymphoma and will typically appear as flocculent areas within the lymph node with very high attenuation similar to cortical bone (see Figure 13-16). Rarely lymph nodes can be high attenuation due to calcified metastasis. Very few tumors exhibit macroscopic calcification; however, notable exceptions include some tumors of bone and cartilage such as osteosarcomas and chondrosarcomas and some mucinous adenocarcinomas (Table 13-4) see Figure 13-16.

Hyperenhancing lymph nodes High attenuation within lymph nodes on CT can also be due to hyperenhancement of the lymph node following injection of intravenous contrast or the accumulation of lymphangiographic contrast. Hyperenhancing lymph nodes will most

A

B

Figure 13-17 Hypervascular Lymph Node Metastasis This 69-year-old man had a pancreatic neuroendocrine carcinoma. A-C. Contrast-enhanced axial CT images show a large hypervascular liver metastasis (arrowhead) and multiple markedly enlarged hyperenhancing celiac axis lymph node

often be seen in lymph node metastasis from hypervascular tumors such as neuroendocrine tumors, melanoma, renal cell carcinoma, papillary thyroid carcinoma, and Kaposi sarcoma (see Figure 13-17).38 Rarely, hyperenhancing lymph nodes will be due to Castleman disease (see Figure 13-18).39-41 Lymphangiographic contrast is a lipid based iodinated compound injected into the lymphatics of the feet and which accumulates in the lymph nodes. This has a characteristic attenuation and imaging appearance on plain films and CT scans. On plain films, the lymph nodes appear hyperdense. On CT scans lymphangiographic contrast is more attenuating than calcium and resemble the attenuation of metallic structures (see Figure 13-19).

Low-attenuation lymph nodes Attenuation of lymph nodes less than the normal attenuation of skeletal muscle is an uncommon finding that can be due to the presence of fluid or lipid material within the lymph node. Causes of low-attenuation lymph nodes include necrotic metastasis, lymphoma, granulomatous infections, and several other unusual disorders (Table 13-5). Recognition of low attenuation of lymph nodes will most often indicate central necrosis of the lymph node as a result of necrotic metastasis (see Figure 13-20). Neoplasms reported to cause low-attenuation lymph nodes include metastatic carcinoma from lung, germ cell tumors, ovary, and lymphoma.42 Granulomatous infections including tuberculosis, nontuberculous mycobacterial infection, and histoplasmosis can also cause low-attenuation lymph nodes.42 This

C metastasis (arrows). Note how the lymph nodes enhance as much as the normal liver parenchyma and more than the parasonal muscles. This indicates the presence of hypervascular disease, in this case metastasis.

790

A

Diagnostic Abdominal Imaging

B

C

Figure 13-18 Castleman Disease This 40-year-old woman had a palpebal pelvic abnormality. A. T2-weighted, (B) T1-weighted pregadolinium, and (C) T1weighted postgadolinium axial MRI images of the pelvis show

an enlarged right obturator lymph node (arrows). This node enhances (C) more brightly than the gluteus muscles following gadolinium enhancement. Surgical biopsy was diagnostic of Castleman disease.

finding was especially common in HIV-positive patients prior to highly active antiretroviral therapy (HAART). However, since HAART therapy has become the standard of care for HIV-infected individuals, recognition of lowattenuation lymph nodes due to granulomatous infections

has become rare in industrialized countries. However, in developing countries, tuberculosis remains an important cause of low-attenuation lymph nodes (see Figure 13-21). Whipple disease, a rare enteric infection caused by the bacterium Tropheryma whippelii causing bowel wall

A

B

C

D

E

F

Figure 13-19 Lymphangiographic Contrast on CT Exams A-C. Soft-tissue windows from an axial contrast-enhanced CT show multiple hyperattenuating lesions (arrowheads) in the distribution of abdominal and pelvic lymph nodes. Note

these appear brighter than cortical bone. D-F. Bone windows show these to have fine curvilinear shapes outlining the cortex and hilum of individual lymph nodes. This appearance is characteristic of lymphangiographic contrast.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 791 Table 13-5. Causes of Low-Attenuation Cystic Lymph Nodes A. Necrotic metastasis B. Whipple disease C. Infections (primarily in HIV-positive patients) 1. Tuberculosis 2. Mycobacterium avium-intracellulare 3. Histoplasma capsulatum

thickening is also a rare cause of low-attenuation lymph nodes. This disorder is a cause for malabsorption and can present with weight loss, diarrhea, joint pain, and arthritis and is described in greater detail in Chapter 2. However, in this case, the low attenuation is not due to lymph node necrosis but is due to lipid accumulation in the lymph nodes related to accumulation of the Whipple organism within the lymph nodes.42 Patients with celiac disease can rarely present with lowattenuation lymph nodes, a phenomenon that has been called “cavitating mesenteric lymph node syndrome.”43,44 Histologic evaluation of the lymph nodes demonstrates a rim of atrophic lymphocytes with central cavitation containing milky fluid and lipid droplets. It is believed that cavitation is a result of mesenteric lymphoid depletion and the crossing of antigenic material over abnormal intestinal

mucosa. Imaging examinations demonstrate enlarged lowattenuation lymph nodes with either fluid or fat attenuation (see Figure 2-53). Only Whipple disease and celiac disease are known to cause fat attenuation lymph nodes. Rarely, lymph nodes in celiac disease will contain fat-fluid levels, a finding that appears to be unique to celiac disease.43 Lymphangioleiomyomatosis (LAM) is a rare idiopathic disorder resulting in cystic interstitial lung disease. Although the clinically significant manifestations of LAM are primarily in the thorax, there are a variety of abdominal manifestations of LAM including angiomyolipomas of the kidney, LAM of the lymphatic ducts, chylous ascites, and abdominal lymphadenopathy.45 Some studies suggest that lymphadenopathy can be found in up to 40% of cases and can be as large as 4 cm in diameter. Lymph nodes can contain water-attenuation regions because of the accumulation of lymph fluid.

Lymph nodes with decreased MRI signal As a result of the paramagnetic effects of iron, iron deposition within lymph nodes will result in loss of signal on virtually all imaging sequences but is most pronounced on gradient echo sequences. Excessive iron deposition within lymph nodes is virtually diagnostic of hemosiderosis and will usually cause loss of signal in the liver and spleen because of similar iron deposition in those organs. These organs are part of the reticuloendothelial system that phagocytizes damaged red blood cells. Excessive red cell

A

B

Figure 13-20 Necrotic Lymph Node Metastasis This 83-year-old man with prostate cancer presented with right upper quadrant pain, anorexia, and weight loss. A and B. Contrast-enhanced axial CT images show multiple rimenhancing, centrally low-attenuation liver lesions (arrowheads) typical of liver metastasis. There are also low-attenuation gastrohepatic (arrow in A) and celiac axis (arrow in B) lymph

nodes. The central low attenuation of the liver metastasis and lymph nodes represents central necrosis of lymph node metastasis. This distribution and appearance of metastatic disease is unusual for prostate cancer. Further evaluation revealed an occult poorly differentiated esophageal adenocarcinoma.

792

Diagnostic Abdominal Imaging

A

B

C

Figure 13-21 Disseminated Tuberculosis This 37-year-old woman presented with fever, chills, and a 10-pound weight loss. A-C. Contrast enhanced axial CT images show centrally low attenuation, rim enhancing lymph nodes in the para-aortic (arrows in A and B) and external iliac (arrow in C)

chains. There is also a low attenuation lesion in the right retroperitoneum that either represents a necrotic lymph node or psoas abscess. Thoracic imaging (not shown) demonstrated innumerable micronodules. This patient was ultimately diagnosed with disseminated tuberculosis.

destruction as seen in hemoglobinopathies and related to blood transfusions can lead to the deposition of iron in the tissues of the reticuloendothelial system. Deposition in the lymph nodes is much less common than that seen in the liver and spleen.

2010, a total of 74,030 cases of lymphoma were diagnosed in the United States, approximately 8490 of which were HL.46 Further, HL and NHL are among the most common causes of abdominopelvic lymphadenopathy. Para-aortic lymphadenopathy is one of the most common findings in HL

DISEASES AFFECTING LYMPH NODES Metastatic tumor and lymphoma are the most common diseases to cause imaging-identifiable lymph node abnormalities. Hepatitis, infectious colitis, inflammatory bowel disease, and sarcoidosis are the most common infectious or inflammatory conditions to cause imaging identifiable lymphadenopathy but there are a variety of other causes of imaging identifiable lymphadenopathy.

Causes of Lymphadenopathy Affecting More Than 1 Drainage Distribution Lymphadenopathy affecting more than 1 drainage distribution will usually indicate a systemic disorder that has manifestations in the lymphatic system. This is most often due to lymphoma or CLL but can occasionally be due to disseminated granulomatous infections or some systemic inflammatory conditions such as systemic lupus erythematosus and rheumatoid arthritis (Table 13-6).

Lymphoma Lymphoma is cancer of lymphatic system and the most fifth most common malignancy in adults, the most common malignancy among teenagers and young adults and is the third most common neoplasm among children.46 Lymphoma is typically divided into HL and NHL. In year

Table 13-6. Causes of Lymphadenopathy Affecting More Than 1 Drainage Distribution A. Neoplasms 1. Non-Hodgkin lymphomaa 2. Hodgkin disease 3. Chronic lymphocytic leukemia 4. Systemic mastocytosis 5. Metastatic tumor B. Infectious diseases (usually in HIV patients) 1. Tuberculosis 2. Mycobacterium avium-intracellulare 3. Histoplasmosis C. Inflammatory disorders 1. Connective tissue disorders a. Systemic lupus erythematosus b. Rheumatoid arthritis 2. Henoch-Schönlein purpura 3. Sarcoidosis 4. Kikuchi disease D. Other causes 1. Amyloidosis aItems in bold are most common.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 793 and NHL. Moreover, HL and NHL can present as lymphadenopathy of 1 or more groups of lymph nodes or often involving isolated organ or can present as widely disseminated disease. Lymph node involvement due to lymphoma usually causes displacement of structures without their invasion. This is an important finding on imaging that distinguishes lymphomas from the carcinomas. HL and NHL usually present with similar radiological features; however, there are some significant differences in their radiographic presentations.

years, likely because of HAART-related improvements in immunity. Patients with lymphoma will usually present with either (1) a variety of systemic symptoms such as weight loss, fever, malaise, fatigue, night sweats, and pruritis; (2) painless lymphadenopathy; or (3) a combination of both.60

Hodgkin lymphoma: HL has a bimodal incidence, with peaks occurring between 15 and 35 years and after age 50 (see Figure 13-23).60 Further, HL accounts for approximately 10% of lymphomas and will most often present in the neck and thorax.60 The histologic diagnosis of HL is dependent on demonstration of large, bilobed or double-nuclei cells with prominent inclusion-like nucleoli giving an “owl’seye” appearance. These are known as Reed-Sternberg cells and with their mononuclear variants are thought to be the malignant cell in HL.61 The many other cells that surround the Reed-Sternberg cells in HL are believed to be reactive populations of nonneoplastic lymphocytes, histiocytes, fibroblasts, eosinophils, and plasma cells. HL can be subdivided into 4 histological subtypes. The nodular sclerosing subtype accounts for approximately 40% to 60% of cases.62,63 The remainder represent lymphocytedepleted, lymphocyte-predominant, and mixed-cellularity subtypes. Patients with mixed cellularity and lymphocyte depletion histology have a poorer prognosis than those with other histological subtypes.64 Seroepidemiologic studies and localization of EBV DNA in Reed-Sternberg cells suggests a role of EBV infection in HL. Up to 40% of HL cases in developed countries can be associated with EBV infection.65 In most cases, HL will spread contiguously from 1 lymph node region to adjacent lymph node regions.66,67 This contiguous pattern of spread makes staging prognostically and therapeutically important, with clinical stage being the most important determinant of prognosis in HL. Further, HL is commonly staged using the Ann Arbor classification system (Table 13-7). Most patients with Stage I or Stage II disease will be cured of their disease with appropriate radiation

Non-Hodgkin lymphoma: Non-Hodgkin lymphomas (NHLs) are a diverse set of diseases with varying histology, natural histories, and responses to therapy. At least 15 types of NHLs are recognized as distinct clinical entities.47,48 About 85% of NHLs are B-cells. The others have T cell–type or NK cell–type histology. In general, NHL can be clinically subdivided into low-grade, intermediate-grade, and high-grade lymphomas, with survival without therapy measured in, respectively, years, months, and weeks.47-49 As a group, they are the fifth most common malignancy in the United States in men and women.46 The incidence of NHL progressively increases with age from childhood through advanced old age (see Figure 13-22).50 The cause of NHL remains unknown but roles for infections, autoimmune disorders, ionizing radiation, and generalized immunodeficiency have been proposed. Epstein-Barr virus (EBV) infection has been associated with Burkitt and other lymphomas.51 Helicobacter pylori infection has been associated with GI maltomas.52,53 Therapeutic irradiation in patients with HL has been associated with an increase in the incidence of secondary NHL.54-56 Patients with HIV/ AIDS, organ transplantation, autoimmune diseases, and some congenital immunodeficiencies have also been associated with increased incidence of NHL.57-59 The incidence of NHL has doubled in the United States from the 1980s to 2000, largely as a result of HIV/AIDS and the increasing use of immunosuppressive therapies in organ transplants and other chronic illnesses. However, incidence of NHL and Kaposi sarcoma have declined markedly in recent

Incidence rates by age groups for non-Hodgkins lymphoma (2003–07) 140 119.4 106.7 103.4

120 100

82.5

80 64 60 46.3 40

31.4 0.6 0.9 1.2 1.8 2.4 3.3 4.8

15.1

22.3

1– 4 5– 10 9 –1 4 15 –1 9 20 –2 4 25 –2 9 30 –3 4 35 –3 40 9 –4 4 45 –4 9 50 –5 4 55 –5 9 60 –6 4 65 –6 9 70 –7 4 75 –7 9 80 –8 4

0

7 10.3

Age groups

85 +

20

Figure 13-22 Age-Specific Incidence Rates for NHL (2003-2007) Bar chart showing age-specific incidence rates for NHL (2003-07). On the x-axis, the incidence rates are shown per 100 000 population. Data from Altekruse SF, Kosary CL, Krapcho M, et al, eds. SEER Cancer Statistics Review, 1975-2007, National Cancer Institute. Bethesda, MD.

794

Diagnostic Abdominal Imaging

Imaging Notes 13-4. Spread of Hodgkins Disease

Table 13-7. Ann Arbor Staging System of Hodgkin and Non-Hodgkin Lymphoma

Hodgkin lymphoma typically spreads in a contiguous fashion along lymph node chains.

therapy and/or chemotherapy.64 Factors associated with poorer outcome include the presence of “B” symptoms, older age, bulky disease, and mixed cellularity and lymphocyte depletion histologies of the tumor.64 HL will typically present with asymptomatic enlargement of lymph nodes, especially of the cervical and axillary chains. Patients with symptoms will most often complain of dry cough or chest discomfort.60 Approximately one-third of patients will complain of systemic or “B” symptoms, including fever, night sweats, and weight loss. Weight loss and loss of appetite, and dragging pain due to splenomegaly are the usual abdominal symptoms. Bone pain, especially that which is worsened by alcohol consumption, is an unusual but interesting symptom associated with HL. HL is a chemotherapy-sensitive malignancy with an excellent response and cure rate. Five-year survival rates are approximately 80% to 85% for all races in the United States, and relapse will usually occur within the first 2 years following completion of therapy.54,60,64 Development of lymphadenopathy or masses within organs more than a few years following completion of therapy should raise the suspicion of a second malignancy, a phenomenon that will occur in approximately 13% of patients.54-56,68 Secondary malignancies are most frequently breast cancer, lung cancer, leukemia, and NHL.54,68 Increased risk of a second malignancy is associated with combined chemotherapyradiation therapy and is correlated with the size of the radiation portal. Breast cancer is at especially increased incidence when women are treated with radiotherapy at a young age.54,68 The majority of patients with HL will have presentations in the cervical, axillary, and thoracic lymph node

Stage I

Involvement of a single lymph node region (I) or a single extralymphatic organ or site (IE).

Stage II

Involvement of 2 or more lymph node regions on the same side of the diaphragm (II) alone or with localized involvement of an extralymphatic organ or site (IIE).

Stage III

Involvement of lymph node regions on both sides of the diaphragm (III) alone or with localized involvement of an extralymphatic organ or site (IIIE) or spleen (IIIS) or both (IIISE).

Stage IV

Diffuse or disseminated involvement of 1 or more extralymphatic organs with or without associated lymph node involvement.

All patients are subclassified A or B to indicate the absence or presence, respectively, of unexplained weight loss of more than 10% body weight, unexplained fever with temperatures more than 38°C, and night sweats.

chains.69,70 Abdominal involvement was found in approximately 15% and 27% of patients in 2 series.70,71 Lymph node involvement was found most often in retroperitoneal lymph nodes, followed by intraperitoneal and pelvic locations. In most cases, the PET and CT imaging will be concordant in their detection of specific disease loci. However, most studies have indicated that PET scans have greater sensitivity and specificity than CT scans in the staging of HL.72-74 The combination of PET imaging and CT scanning has been shown to be complementary because in some instances disease will only be detected by one or the other of the imaging studies.29 Further, PET-CT delivers the advantages of both examinations and is now an integral part of the staging and follow-up of HL.

Incidence rates by age groups for Hodgkins lymphoma (2003–07) 6 4.9 5.1

5

4.4 4.4

4

3.8

4 2.9

3

3.1

4.3

4

3.1

3 2.4 2.4 2.5

2 1.2

Age groups

+ 85

10 9 –1 4 15 –1 9 20 –2 4 25 –2 9 30 –3 4 35 –3 40 9 –4 4 45 –4 9 50 –5 4 55 –5 9 60 –6 4 65 –6 9 70 –7 4 75 –7 9 80 –8 4

4

0.1

1–

0

0.5

5–

1

Figure 13-23 Age-Specific Incidence of Hodgkin Lymphoma (2003-2007) Bar chart showing age-specific incidence rates for Hodgkin Lymphoma (2003-2007). On x-axis, the incidence rates are shown per 100 000 population. Data from Altekruse SF, Kosary CL, Krapcho M, et al, eds. SEER Cancer Statistics Review, 1975-2007, National Cancer Institute. Bethesda, MD.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 795 Older studies have suggested that approximately 60% to 75% of cases of HL will leave residual fibrotic soft-tissue masses following completion of chemotherapy and/or irradiation.75-77 Some studies have suggested that patients with residual masses are twice as likely to have recurrence than those without residual masses, especially those who were treated with chemotherapy without radiation therapy.75,77 However, other studies have failed to show this association.76 It is the author’s belief that residual masses are now much less common than previously reported. When found, residual masses typically appear as a non-descript soft-tissue mass at the sites of original disease. When noncalcified, these masses are indistinguishable from residual tumor on CT and MRI exams. However, cross-sectional imaging will show them to remains stable on serial follow-up exams and they will be cold on PET scans. A PET-CT study is very useful in the assessment of therapeutic response because of its discriminating power between benign fibrosis presenting as absent or low-grade FDG uptake and residual active lymphoma presenting as elevated FDG uptake. Persistent FDG activity in posttreatment scans is associated with early relapse and poor clinical outcome.78,79 The larger the initial tumor, the greater the likelihood of a residual mass on CT scans. Most residual fibrotic masses will slowly decrease in volume over many years and often subsequently calcify.80 In past decades, it was common to see dystrophic calcification developing in lymph nodes of previously treated HL.75-77,80 However, this is now a relatively uncommon phenomenon.

Imaging features of Hodgkin lymphoma and nonHodgkin lymphoma: The imaging characteristics of HL and NHL are similar and cannot be reliably distinguished. One of the distinguishing features between NHL and HL is that HL spreads in a contiguous fashion along the lymph

A

B

Figure 13-24 Non-Hodgkin Lymphoma This 57-year-old man presented with fevers and weight loss. A-C. Contrast-enhanced axial CT images demonstrate enlarged para-aortic and mesenteric lymph nodes (arrowheads). Extensive

Imaging Notes 13-5. Lymph Node Size and Lymphoma Lymph nodes larger than 4 cm in diameter are uncommon in Hodgkin lymphoma but are not infrequent in non-Hodgkins lymphomas.

node chains. Very large lymph nodes, greater than 4 cm in diameter, are uncommon in HL but are not infrequent in NHL. Also extranodal involvement in HL is much less common than NHL. The imaging role of CT scans in the evaluation of lymphoma is multifold. It not only defines the full extent of disease for accurate staging but also assists in treatment planning, to evaluate response to therapy, to monitor progression of disease, and to detect possible relapse. Care should be made in attributing small changes in lymph node size with treatment response. Studies of interobserver variability in the measurement of lymph node diameters have suggested that observers will have an average of 3-mm difference in measurements of lymph node diameters.81 CT has traditionally been the imaging modality of choice in the evaluation of lymphomas; however, it is limited in its accuracy because small lymph nodes can harbor lymphoma and large lymph nodes can be benign.82 Increasingly, PET-CT is being employed in the surveillance of lymphoma because of its greater accuracy in identifying the sites of active disease. Lymphoma can present as unifocal adenopathy, multifocal adenopathy, or as a diffusely infiltrating mass involving multiple lymph node sites (see Figures 13-11, 13-24, and 13-25). Any nodal chain within the abdomen and pelvis can

C multichain adenopathy like this will most often be due to lymphoma. The arrow in (B) points to the superior mesenteric artery. Node biopsy was diagnostic of non-Hodgkin lymphoma.

796

A

Diagnostic Abdominal Imaging

B

C

Figure 13-25 Hodgkin Lymphoma This 19-year-old man presented with weight loss. A-C. Contrastenhanced axial CT images demonstrate markedly enlarged mediastinal lymph nodes in (A) and increased numbers of smalland medium-sized lymph nodes (arrows) in the upper abdomen.

The spleen is moderately enlarged and contains multiple hypoattenuating lesions. This constellation of finding is highly suggestive of lymphoma. Lymph node biopsy was diagnostic of Hodgkin disease.

be involved, but para-aortic and iliac chains are the most frequently involved. Mesenteric involvement is the predominant finding in 4% to 5% cases of HL and 30% to 50% cases of NHL. Mesenteric lymphoma can present as round, oval, or irregular masses within the fat of the mesentery. In some cases, the lymph nodes will form 2 conglomerate homogeneous masses on either side of the mesenteric arteries and veins. In cross section, the flat conglomerate masses resemble the bread and the mesenteric vessels the filling of a sandwich, a phenomenon called the “sandwich sign” (see Figure 13-26).83

Lymphoma is a hypercellular malignancy and therefore can be mildly hyperattenuating on unenhanced CT scans. After contrast enhancement, lymphomatous lymph nodes will usually show homogenous low-level enhancement. If there is involvement of the extranodal organs of the abdomen, the organ involved typically shows greater enhancement than the lymphoma. Larger nodal masses can develop central necrosis. Calcification in lymphomas is nearly always a response to treatment and is uncommon as a primary manifestation of the malignancy. Accuracy of MR in detection of lymph node and organ involvement due to lymphoma is similar to that of CT exams. Lymphomas are usually isointense to skeletal muscle on T1-weighted imaging. On T2-weighted imaging, lymphomas have a moderately high signal intensity because of increased water content in untreated lymphomas.84,85 In general MRI is significantly more sensitive than CT for detecting bone marrow involvement. Nakayama et al recently showed that the lymphomas’ diffusion coefficient of retroperitoneal lymphoma is significantly lower than those of the malignant and benign mesenchymal tumors of the retroperitoneum.86 This principle can be exploited to distinguish between retroperitoneal lymphoma and nonlymphomatous involvement of the retroperitoneal structures. Further, PET scanning has become an essential tool in the imaging evaluation of patients with both Hodgkin and NHLs. In general, PET scans will often identify a higher stage than CT scans, in the initial staging of HL.72-74 Nearly one-third of patients with HL will be upstaged by PET over CT staging.72 Foci of increased activity will usually indicate the presence of lymphoma (see Figure 13-27). However, low-grade lymphomas and foci of disease less than 1 cm in diameter can result in false-negative examinations.74 False-positive foci of activity are frequently found in the neck and paraspinal locations because muscular activity, as a result of anxiety, can lead to metabolic activity at these

Figure 13-26 The Sandwich Sign This 61-year-old man had a non-Hodgkin Lymphoma. Contrastenhanced CT image shows massive mesenteric lymphadenopathy (small arrows) surrounding the mesenteric vessels (arrowheads). With the lymph nodes serving as the bun and the vessels serving as the filling, the combination resembles a sandwich. There are also multiple enlarged retroperitoneal lymph nodes (large arrows).

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 797 Imaging Notes 13-6. PET scanning and Hodgkins Disease PET or PET-CT is the preferred modality for staging Hodgkin lymphoma because nearly one-third of patients with Hodgkin lymphoma will be upstaged by PET over CT staging.

US has little role in the evaluation of the lymphomas and is predominantly used in real-time biopsies of the focal lesions of the liver and some other approachable lesions in lymphoma. Involvement of abdominal organs other than the spleen is uncommon in patients with HL and NHL. Details of lymphomatous involvement of extranodal sites in the abdomen are detailed in the chapters dedicated to each organ.

Leukemia sites. This has been especially recognized in children and adolescents.87 Increased metabolic activity can also be seen in the soft-tissues of the sites of brown fat. Uptake in the brown fat in the head and neck areas can result in falsepositive results in 2.3% to 4% patients.88,89 In the fetal life and neonates, brown adipose tissue is in abundance and plays an important role in thermoregulation. This is found usually in the cervical, axillary, paravertebral, mediastinal, and abdominal areas (see Figure 13-15).90 Most experienced readers are aware of these phenomena and discount activity at these locations. False-positive activity can also occur as a result of unrelated inflammatory conditions. These are the most likely areas of increased activity to be a cause of a false-positive interpretation. Further, PET-CT is not only useful in the initial assessment of lymphoma but is now being extensively used in the treatment follow-up of PETavid lymphoma.82

Leukemia is a malignant disease of the bone marrow and blood characterized by uncontrolled accumulation and proliferation of blood cells. Two major categories of leukemia are myelogenous or lymphocytic and each can be subdivided into acute and chronic types. Acute leukemias can present with tiredness, shortness of breath on exertion, fever, night sweat, jaundice, excessive bleeding, slow healing, ecchymosis, and joint pains. Most leukemias have no or few abnormalities detected by imaging exams, with the exception of CLL that commonly causes widespread lymphadenopathy and splenomegaly. Chronic myeloid leukemia can show enlarged spleen on plain abdominal radiograph or CT of abdomen.

Chronic lymphocytic leukemia: CLL occurs as a result of a mutation to the DNA of lymphocytes. In 95% of the cases of CLL, the changes involve B-lymphocytes, but in 5% it can involve T-lymphocytes or natural killer (NK) cells. With time, the CLL cells replace normal lymphocytes in

A

B

Figure 13-27 PET-CT of Non-Hodgkin Lymphoma This 50-year-old woman presented for staging with known non-Hodgkin lymphoma before treatment. A and B. Low-dose CT portion of PET-CT examination showed a subtle mantle of

retroperitoneal soft-tissue (arrows) (A) that showed moderately increased FDG activity on the combined PET-CT image (arrows) (B) consistent with active non-Hodgkin lymphoma.

798

Diagnostic Abdominal Imaging

the bone marrow and lymph nodes, thereby causing lower immunity and inability to fight infection. CLL is typically seen in patients older than age 60 years. The etiology of CLL is unknown but U.S. Department of Veterans Affairs has associated use of herbicides with CLL. Many CLL patients will be asymptomatic and are identified as a result of incidentally discovered lymphadenopathy. Those with symptoms will typically present with tiredness, shortness of breath during exertion, lymphadenopathy, splenomegaly, weight loss, and/or infections. Faster-growing CLL is the type that can present with lymphadenopathy, and this lymphadenopathy can compress the neighboring structures, causing GI and urinary symptoms in the abdomen. As with other leukemias, the diagnostic tests for CLL primarily includes blood examination, bone marrow examination, immunophenotyping, and immunoglobulin-level measurements. Incidental lymphadenopathy

and splenomegaly can be found on CT scans being performed for other reasons than CLL. It has been suggested that CT scans are not necessary to routinely evaluate disease extension in the CLL.91 However, CT scans are being used increasingly in CLL to assess disease extension and treatment response. Recently, it has been shown that CT was a strong predictor of progression in patients with early-stage CLL.92 In general, CT and MRI examinations will demonstrate innumerable small to moderate lymph nodes, widely distributed throughout the body. These are especially prevalent in the thorax but can also be seen in the abdomen (see Figures 13-12 and 13-28).

A

B

C

D

Figure 13-28 Chronic Lymphocytic Leukemia This 68-year-old man had an abdominal aortic aneurysm for which he had received a stent graft. He was otherwise asymptomatic. A-D. Contrast-enhanced axial CT images demonstrate many intermediate and a few large lymph nodes

(arrowheads) in the para-aortic, mesenteric, external iliac, internal iliac, and inguinal chains. Multidistribution moderate lymphadenopathy will usually indicate a lymphoprolipherative malignancy. This patient had chronic lymphocytic leukemia.

Systemic mastocytosis Systemic mastocytosis is a rare lymphoproliferative disease, characterized by monoclonal proliferation of mast

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 799 cells in organs throughout the body.93 Systemic mastocytosis is almost always associated with KIT D816V mutation resulting in this unusual lymphoproliferative neoplasm. GI symptoms are common and include abdominal pain, diarrhea, nausea, and vomiting and can infiltrate any of the GI organs from the esophagus to the rectum. In a report by Barete et al, 74% (43/57) of the patients of systemic mastocytosis presented with abdominal symptoms. Lymphadenopathy and splenomegaly are common physical findings in systemic mastocytosis.94 Abdominal lymphadenopathy has been found in 12 of 18 patients with systemic mastocytosis.95 Hepatomegaly and/or splenomegaly has been found in approximately one half of patients with systemic mastocytosis.

Disseminated granulomatous infections

Focal inflammatory diseases will typically cause lymphadenopathy in the drainage distribution of the affected organ. However, there are some systemic inflammatory conditions that can cause widespread lymphadenopathy across many drainage systems.

Abdominal tuberculosis is the most common abdominal granulomatous infection. It typically involves the terminal ileum and cecum and gives rise to localized adenopathy in the drainage distribution of the terminal ileum and cecum. However, immunocompromised patients can develop disseminated granulomatous infections that can cause widespread adenopathy throughout the body (see Figure 13-21). Prior to HAART therapy, this was most often seen in HIVpositive patients infected with tuberculosis, mycobacterium avium complex (MAC), or histoplasmosis. Although HIV is a relatively common cause of peripheral lymphadenopathy, it virtually never gives intrathoracic or abdominal adenopathy. Therefore, discovery of abdominal lymphadenopathy in an HIV-positive patient will usually indicate a superimposed disease, usually lymphoma or a disseminated granulomatous infection. The granulomatous infections most common to cause abdominal adenopathy are tuberculosis and MAC but less commonly histoplasmosis.96 Disseminated infection due to MAC usually presents with night sweats, abdominal pain, diarrhea, and weight loss. The most common CT findings are enlarged retroperitoneal and mesenteric lymph nodes in 42% of cases.96 Other findings include hepatomegaly in 50%, splenomegaly in 46%, and small bowel thickening in 14% of patients.96 Abdominal CT in more than 80% patients shows large, bulky, retroperitoneal, and mesenteric lymphadenopathy which can be soft-tissue or have low attenuation (see Figure 13-29).97 The lymph nodes due to MAC are typically smaller and of uniform attenuation in comparison

A

B

Figure 13-29 Lymphadenopathy Due to Mycobacterium avium intracellulare Infection This 42-year-old HIV-positive man complained of abdominal pain, malaise, and weakness. A and B. Contrast-enhanced axial CT images show extensive low-attenuation lymph nodes in the

subcarinal and right hilar locations in the thorax (arrowheads) and the anterior and lateral para-aortic locations (arrows) in the abdomen. Cultures of lymph node biopsies grew Mycobacterium avium intracellulare.

Metastatic Carcinoma and Sarcoma In the vast majority of cases, malignancies will metastasize in an orderly distribution along the lymphatic drainage path of the organ involved. However, rarely malignancies will cause widespread adenopathy across multiple drainage distributions.

Inflammatory Diseases Causing Adenopathy in More than 1 Drainage Distribution

800 Diagnostic Abdominal Imaging to the tuberculosis where the lymph nodes are typically larger and have low central attenuation. However, there is considerable overlap in the appearance of the 2 disseminated infections. In 1993, tuberculosis was listed as AIDS-defining disease by the centers for disease control (CDC).98 Necrotic lymph nodes showing low attenuation centers can be seen associated with tuberculosis.99 Koh et al showed that lymph nodes with low attenuation are more common and lymph nodes are usually larger.100 In non-HIV patients, abdominal tuberculosis will most frequently involve the ascending colon and terminal ileum.101-103 Characteristics of tuberculous colitis are detailed in Chapter 2. In tuberculous colitis, there is usually an increased number of small, focal mesenteric lymphadenopathy, especially in the pericecal regions. In acute progressive histoplasmosis, hepatosplenomegaly and lymphadenopathy may be seen. Progressive disseminated histoplasmosis occurs mostly in immunocompromised individuals, especially in AIDS patients with CD4 counts of less than 150 cells/μL. GI dissemination may produce diarrhea and abdominal pain. Abdominal CT findings in the disseminated histoplasmosis included abdominal homogenous lymphadenopathy in 44% patients, lymphadenopathy with diffuse or low central density in 13%, or both in 19% patients.104

1 drainage distribution. These include some connective tissue disorders, vasculitis, and sarcoidosis.

Connective tissue disorders: Connective tissue disorders including progressive systemic sclerosis, systemic lupus erythematosus, or rheumatoid arthritis can give rise to lymphadenopathy.105 The overall incidence of lymphadenopathy in active rheumatoid arthritis was 82%, and axillary lymph nodes were involved more than cervical, supraclavicular, and inguinal lymph nodes.106 In the active systemic lupus erythematosus, 69% patients showed lymphadenopathy, and similar to rheumatoid arthritis, there was more involvement of axillary lymph nodes than inguinal lymph nodes. With the exception of inguinal lymphadenopathy, abdominal and pelvic lymphadenopathy is rare in patients with connective tissue disorders (see Figure 13-30). These lymph nodes do not cavitate and are rarely larger than 2 cm. Lymphadenopathy will typically regress with the corticosteroid therapy.

Henoch-Schönlein purpura: Henoch-Schönlein pur-

There are a variety of systemic inflammatory disorders that can result in lymphadenopathy that affects more than

pura (HSP) is a systemic, generalized vasculitis and is a rare but reported cause of abdominal lymphadenopathy. Abdominal symptoms are common and sometimes precede the other manifestations of HSP. Disease predominantly involves the second part of duodenum but can also frequently involve other portions of the small intestine. Further, CT scans characteristically show multifocal bowel wall thickening with intervening normal segments.107

A

B

C

D

Figure 13-30 Systemic Lupus Erythematosis This 46-year-old woman had persistent fevers and rash. A-D. Unenhanced axial CT images demonstrate multiple large axillary (arrows in A), obturator and external iliac (arrows in C) lymph

nodes, and multiple moderate para-aortic (arrow in B) and inguinal (arrow in D) lymph nodes. Extensive adenopathy like this will usually indicate lymphoma but in this case was due to the patient’s SLE.

Systemic inflammatory disorders

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 801 Associated findings include mesenteric lymphadenopathy, mesenteric vascular engorgement, mesenteric fat stranding, ascites, pleural effusions, renal infarcts, and splenic infarcts.107 Jeong et al showed mesenteric lymphadenopathy of less than 1.5 cm on CT in 6 of 7 patients.107,108

Sarcoidosis: Sarcoidosis is a multisystem disorder primarily involving the lungs and lymphoid system. In the majority of cases, adenopathy due to sarcoidosis will be confined to the thorax. However, occasionally abdominal involvement can be seen with or without intrathoracic sarcoidosis and include disease of the lymph nodes, liver, spleen, and gut.109 In patients with hepatosplenic sarcoidosis, there is concomitant abdominal lymphadenopathy in 76%.110 Lymph node enlargement can be seen without imaging evidence of disease in other abdominal organs. In most cases, lymphadenopathy will be confined to the upper abdominal chains above the level of the renal arteries. Retrocrural lymph nodes were seen in approximately 30% of patients.110 In most cases, the lymph nodes will only be moderately enlarged between 10 and 20 mm in the short axis diameter. However, rarely sarcoidosis can cause bulky lymphadenopathy that can be confused with lymphoma (see Figure 13-31).111 In 10% of patients, there was involvement of 4 or more sites with lymph node size greater than 2 cm.110

of similar clinical and radiological findings.113-115 Patients typically present with fever and painless lymphadenopathy. Rarely patients will complain of weight loss, diarrhea, chills, and/or sweats. Histology of the excised lymphadenopathy demonstrates geographic foci of central necrosis with extensive karyorrhectic debris surrounded by immunoblasts, histiocytes, and plasmacytoid monocytes. Necrosis is not very extensive. Unilocation lymphadenopathy in Kikuchi disease has been reported in 83% of cases.116 Cervical lymphadenopathy is the commonest site but can involve any lymph node chains. Isolated intraabdominal lymphadenopathy has been reported rarely.116 A larger than normal number of small or mildly enlarged lymph nodes was observed by 1 of the authors.112 The size of the affected lymph nodes is typically less than 4 cm (see Figure 13-32).

Amyloidosis

inflammatory condition that can cause lymphadenopathy.112 Kikuchi disease can be misdiagnosed as malignant lymphoma or some other inflammatory diseases because

Amyloidosis is an uncommon systemic disorder that can involve a wide variety of organs giving diverse clinical features at presentation.117 The most common presenting symptoms are weakness, fatigue, and weight loss. Other symptoms include pedal edema, paraesthesias, abdominal pain, periorbital, facial purpura, syncope, and orthostatic hypotension. In the abdomen, hepatosplenomegaly is often present. Lymph node enlargement is a less common manifestation and can be isolated or part of multisystem disease.118-120 Generalized lymphadenopathy occurs in 8% of patients with systemic amyloidosis.121 Hilar, mediastinal,

A

B

Figure 13-31 Abdominal Adenopathy Due to Sarcoidosis This 52-year-old woman had a long history of thoracic sarcoidosis. A and B. Contrast-enhanced CT of the abdomen and pelvis demonstrates enlarged left para-aortic (small arrowhead), inter-aortocaval (arrow), and right common iliac adenopathy

(large arrowhead). Note that several of the lymph nodes have focal areas of calcification typical of chronic granulomatous diseases. These lymph nodes have remained stable for 8 years and are presumed to be secondary to sarcoidosis.

Kikuchi disease: Kikuchi disease is a rare, self-limited

802

Diagnostic Abdominal Imaging

A

B

C

Figure 13-32 Kikuchi Disease This 63-year-old man with diabetes mellitus presented with 3-week history of fatigue, myalgias, fever, and rigors. A-C. Contrast-enhanced axial CT images demonstrate increased

numbers and slight increase in size of retrocrural (white arrowhead), gastrohepatic (black arrowheads), celiac axis (black arrows), and para-aortic (white arrows) lymph nodes. Lymph node biopsy was diagnostic of Kikuchi disease.

and/or para-aortic lymph nodes are the most common sites of involvement; however, abdominal lymphadenopathy has also been reported.122

these will be briefly mentioned individually for major abdominal-pelvic organs.

Gastrointestinal malignancies: Distal esophageal carci-

Causes of Lymphadenopathy Affecting a Single Drainage Distribution Adenopathy in a single drainage distribution will usually be due to diseases of the organs that drain into that distribution, and recognition of focal adenopathy should lead to an evaluation of the organ that supplies that distribution. In many cases, focal adenopathy will indicate lymph node metastasis from a cancer in the organ in question. However, infectious and inflammatory diseases of the organ can also cause lymphadenopathy in the drainage distribution of the organ.

Lymphoma In most cases, lymphoma will cause lymph node abnormalities that involve multiple lymph node drainage distributions. However, occasionally, imaging abnormalities will be limited to a single drainage distribution.

Metastatic tumor Lymphatic metastases from abdominal malignancies are a common mode of cancer spread. In the majority of cases, the distribution of lymphatic metastasis will follow the typical drainage pattern of the organ of origin. In most cases, the typical lymphatic distribution will parallel the vascular drainage of the organ (Table 13-1). Details of the lymphatic drainage and nodal staging for each organ will be reviewed in the chapters detailing diseases of that organ; however,

noma will typically involve small periesophageal nodes first and then spread to the larger celiac axis lymph nodes.123 Gastric carcinoma will initially involve small lymph nodes in the gastrohepatic ligament and perigastric fat and then also spread to the celiac axis lymph nodes (see Figure 13-33). Tumors of the small bowel and colon will initially involve small lymph nodes in the fat surrounding the primary tumor and then spread to lymph nodes in the small bowel or colonic mesentery. Tumors of the small bowel and proximal colon to the level of the sigmoid flexure will then spread to the periaortic lymph nodes adjacent to the origin of the superior mesenteric artery (see Figure 13-33). Tumors of the descending and sigmoid colons and proximal rectum will involve the para-aortic lymph nodes at the level of the inferior mesenteric lymph nodes, and distal rectal cancers will spread to the internal iliac chains.

Hepatobiliary and pancreatic malignancies: Nodal drainage of malignancy from the liver, gallbladder, and bile ducts is usually to hepatoduodenal, peripancreatic and aortocaval nodes (see Figure 13-34).124 Pancreatic carcinoma frequently results in local lymph node spread in the retroperitoneum and peripancreatic distribution, but mesenteric lymph nodes involvement is not infrequent.

Prostate, cervix, uterine, and bladder malignancies: Tumors of the prostate, cervix, and uterus will typically first involve the pelvic fat surrounding each of the organs, followed by the internal iliac, obturator, and external iliac

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 803

A

B

C

D

Figure 13-33 Gastrointestinal Malignancy Lymph Node Metastasis A and B. This 47-year-old man with gastric carcinoma (small arrowhead) has a large hepatic metastasis (white arrow) and gastrohepatic ligament lymph node metastasis (large arrowhead). The gastrohepatic ligament nodes and celiac axis nodes are the

typical sites of lymph node metastasis from gastric malignancies. C and D. This 64-year-old woman with cecal carcinoma (arrow) has multiple enlarged para-aortic and SMA lymph nodes (arrowheads) typical of lymph node metastasis from a malignancy of the colon proximal to the splenic flexure.

nodes. Further spread is to the common iliac and retroperitoneal lymph nodes. Adenopathy outside of this distribution, for example, in the mesenteric lymph nodes of the pelvis, is rare and should prompt a search for another cause of lymph node enlargement (see Figure 13-35).125 In most cases, the relative likelihood of lymphatic metastasis increases as the tumor invades into the muscles and soft-tissues of the cervix, uterus, and bladder and the surrounding pelvic fat.126 For example, lymph node involvement has been seen in approximately 30% of the cases

where bladder carcinoma has invaded beyond the muscle of the bladder wall (see Figure 13-35).127

Ovarian cancer: Ovarian cancer has 3 routes of lymphatic spread: (1) via lymphatics that ascend with the ovarian vessels to the retroperitoneal nodes of the upper abdomen; (2) via lymphatics draining laterally in the broad ligament to reach the internal iliac and obturator nodes in the pelvic sidewall; (3) via lymphatics in the round ligament to external iliacs and inguinal lymph nodes.

804 Diagnostic Abdominal Imaging contralateral para-aortic and aortocaval lymph nodes can be affected. Testicular cancer should always be considered in the differential diagnosis for the para-aortic lymphadenopathy of unknown etiology. Residual masses following chemotherapy are a relatively common phenomenon with nonseminomatous germ cell tumors. Further, PET-CT has been reported to show a higher accuracy for evaluating the residual tumor compared with CT.128 In the American College of Radiology (ACR) appropriateness criteria, use of FDG-PET-CT is mentioned as appropriate and possibly indicated for the follow-up of residual or recurrent disease. It has been said that whole body FDG-PET has no clear benefit in initial staging over CT.

Vaginal carcinoma: Vaginal carcinomas spread to obtu-

Figure 13-34 Cholangiocarcinoma and Metastatic Portocaval Lymph Nodes A coronal fused PET-CT scan shows FDG-avid cholangiocarcinoma marked with the vertical arrow and metastatic portocaval lymph nodes marked with the horizontal arrow.

Testicular cancer: Nodal metastases are seen mostly in the para-aortic nodes between the renal and inferior mesentery arteries and are ipsilateral to the primary tumor (see Figure 13-8). Because of abundant collaterals, the

rator and internal iliac nodes. Tumors from the posterior wall tend to spread to the superior and inferior gluteal nodes. Carcinomas from the lower third of the vagina usually spread to the pelvic and/or inguinofemoral nodes.

Penis, vulva, and anus: The lymph node drainage from these organs is typically to the superficial inguinal lymph nodes (see Figure 13-36). Subsequent spread to the deep pelvic lymph nodes is less common.

Accuracy of abdominal lymph node staging for abdominal malignancies: None of the imaging modalities has been shown to be highly accurate in the diagnosis of metastatic adenopathy from intra-abdominal

A

B

Figure 13-35 Pelvic Lymph Node Metastasis from Prostate Carcinoma and Bladder Cancer in 2 Patients A. This 66-year-old man had prostate carcinoma. Contrastenhanced CT through the pelvis shows enlarged obturator (small arrowhead), external iliac (large arrowhead), and external

iliac (arrow) lymph nodes. This is the typical lymphatic drainage distribution for a pelvic malignancy including prostate cancer. B. This 89-year-old had bladder carcinoma. Fused PET-CT image shows an enlarged FDG-positive right internal iliac lymph node (arrow) which likely represents lymphatic metastasis.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 805

A

B

Figure 13-36 Inguinal Lymph Node Metastasis from Vulvar Cancer This 52-year-old woman had surgical resection of a vulvar carcinoma. A and B. Gadolinium-enhanced, fat-saturated T1-weighted MRI images of the inguinal regions demonstrate enlarged bilateral inguinal lymph nodes (arrows). There is

also a subcutaneous metastasis to the anterior abdominal wall (arrowhead). These findings were not present on an MRI examination 6 months previously and are the typical lymphatic drainage for a perineal malignancy. Lymph node biopsy confirmed a diagnosis of metastatic vulvar carcinoma.

malignancies. For example, studies of colorectal carcinoma have shown accuracies of CT and MRI of between 66% and 83% in the diagnosis of lymphatic metastasis.11,129,130 Studies of pancreatic, renal, prostate, bladder, endometrial, cervical, and ovarian malignancies have demonstrated similarly intermediate accuracies.12-30 In general, when using the 1-cm rule, anatomic imaging has had moderately high specificity in the range of 75% to 95%, with lower sensitivity for metastasis in the range of 50% to 75%.11-30 Therefore, the presence of enlarged lymph nodes will in most cases indicate the presence of metastasis. However, microscopic metastases occur with moderate frequency and lower the sensitivity of anatomic imaging for metastatic lymphadenopathy. Addition of functional imaging with FDG-PET has only slightly increased the sensitivity for lymph node metastasis for GI malignancies but has been shown to increase sensitivities in the range of 80% to 90% and specificities at levels greater than 90% for other neoplasms such as endometrial carcinoma and cervical cancer.131-137 Researchers have attempted to increase accuracy in lymph node staging with a variety of novel MR and nuclear medicine contrast agents. It has been shown that using lymphotropic super-paramagnetic nanoparticles results in higher sensitivity in lymph node involvement with prostate carcinoma. A sensitivity of 82% to 100%, a specificity of 93% to 96%, and a negative predictive value of 96% to

100% has been reported.138,139 Capromab pendetide scintigraphy is useful in prostate carcinoma staging and provides a sensitivity of 67% to 94% and a specificity of 42% to 80% in the detection of lymph node metastasis.140-142 However, the reported accuracies for these newer agents are still not sufficiently high to obviate the need for histologic staging of lymph node metastasis.

Regional lymphadenopathy due to inflammatory conditions Besides malignancy, lymph nodes respond to a wide variety of internal and external stimuli, including infectious, noninfectious inflammatory, systemic, and immunologic

Imaging Notes 13-7. Causes of Lymphadenopathy Affecting a Single Drainage Distribution Lymphadenopathy in a single drainage distribution will usually indicate a disease in the organ drained by the abnormal lymph nodes. Although this is often attributed to malignancies, inflammatory conditions, especially those of the bowel and liver, are also frequent causes of lymphadenopathy in a single drainage distribution.

806 Diagnostic Abdominal Imaging disorders. In general, an inflammatory condition within a given organ will cause mild lymphadenopathy in the drainage distribution of the organ. This can be seen as mild enlargement of 1 to 2cm or as increases in numbers of lymph nodes in the drainage distribution of the organ. Therefore, recognition of a focal or regional lymphadenopathy should engender a search of the organs in that drainage distribution for a malignant or inflammatory cause of the lymphadenopathy. Size, location, contour, density, relationship to aorta/vena cava, and presence of mass effect are of no value in distinguishing the malignant from benign causes of lymphadenopathy.143,144 The significance of lymphadenopathy is therefore based on recognition of the primary lesion causing the lymphadenopathy. When no cause can be identified, either short-term clinical and CT follow-up or percutaneous or other biopsies may be necessary. Appendicitis, diverticulitis, other causes of intraabdominal abscess, and perforation of any abdominal viscera can give rise to mesenteric lymphadenopathy. In most cases, the adenopathy will be present in the mesenteric nodes that drain the region of affected bowel and in each of

these disorders, and the mesenteric lymphadenopathy will be associated with the characteristic features of the primary bowel disorder. Many causes of bowel wall thickening will also be associated with mesenteric lymphadenopathy (see Figure 13-37). These include (1) inflammatory bowel diseases: Crohn disease and ulcerative colitis; (2) bowel infections: tuberculosis, Yersinia enterocolitica, Campylobacter species, Giardia lamblia; and (3) autoimmune disorders: celiac disease (see Figure 13-13).145-147 Abdominal tuberculosis is an important infection in developing countries. In most cases, the disease will involve the cecum and terminal ileum, causing thickening of the walls of the cecum and terminal ileum. Pericecal and mesenteric adenopathy is a common associated finding (see Figure 13-38). Multiple reports have suggested that these lymph nodes often have low attenuation centers and enhance peripherally with the administration of intravenous contrast.148 Biliary diseases and hepatitis of any cause, infectious, autoimmune, alcoholic, drug-induced, and others, will commonly give rise to mild periportal and celiac

A

B

C

D

Figure 13-37 Mesenteric Adenopathy Due to Celiac Disease This 37-year-old man had testicular carcinoma. A-D. Contrastenhanced CT demonstrates an increased number and size of mesenteric lymph nodes (arrowheads). With the history of testicular carcinoma, it might be tempting to attribute these nodes to metastatic cancer. However, this is the wrong drainage

pattern for metastatic testicular carcinoma. Furthermore, the wall of the small bowel appears much thickened, suggesting diffuse edema due to inflammation. Further evaluation lead to a diagnosis of celiac sprue, which had not been known before this CT examination.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 807

A

B

C

D

Figure 13-38 Abdominal Tuberculosis This 43-year-old man presented with abdominal pain and hiccoughs. A-D. Unenhanced, axial CT images show thickening of the wall of the cecum (arrowheads) with a large matted area of

pericecal adenopathy (arrow in D) and mesenteric (arrow in B) and para-aortic (arrows in A) adenopathy. This might have represented a colonic lymphoma or carcinoma; however, cultures were positive for Mycobacterium tuberculosis.

axis lymphadenopathy (see Figure 13-9). For example, lymphadenopathy, typically involving portohepatic and/ or portocaval nodes, is seen in 80% cases of primary biliary cirrhosis (PBC).149 The liver also drains to the cardiophrenic lymph nodes in the thorax and so hepatitis can also lead to adenopathy in that distribution. Other causes of cirrhosis have been shown to cause lymphadenopathy in 37% to 49% patients, ranging from approximately 50% in cases of hepatitis C–related hepatitis to less than 10% in cases of alcoholic cirrhosis and hepatitis B–related liver diseases.149,150 Interestingly, inflammatory diseases of the genitourinary tract such as pyelonephritis, prostatitis, and pelvic inflammatory disease do not usually cause abdominal or pelvic lymphadenopathy.

mesentery.151,152 It is found more commonly in men than women and can be seen at any age but with an average age of 60 years. Some patients are discovered as an incidental asymptomatic finding on abdominal imaging exam. Symptoms are variable and include abdominal pain, fever, nausea, vomiting, and weight loss.151,152 The appearance of the lesion can range from a well-defined soft-tissue mass to ill-defined areas of higher attenuation in the mesenteric fat.151,152 There may be preservation of fat around the mesenteric vessels, a phenomenon that is referred to as the “fat ring sign,” which can be an important distinguishing feature from other mesenteric processes such as lymphoma and carcinoid tumor.151,152 Local mesenteric lymphadenopathy is also often present in cases of mesenteric panniculitis (see Figure 13-39).151,152

Sclerosing mesenteritis (panniculitis): Sclerosing mesenteritis is a nonspecific, idiopathic inflammatory condition affecting the fatty tissue of the mesentery.151,152 The disorder is often associated with other idiopathic inflammatory disorders such as retroperitoneal fibrosis, sclerosing cholangitis, Riedel thyroiditis, and orbital pseudotumor. These associations suggest that sclerosing mesenteritis could be an autoimmune phenomenon, but infection, trauma, and ischemia have also been suggested as potential etiologies. Histologically, varying degrees of inflammation, fat necrosis, and fibrosis are found within the fat of the small bowel

Focal Abdominopelvic Lymphadenopathy Without Other Abdominal Findings There are a few rare causes of focal lymphadenopathy that do hot have associated intra-abdominal findings. These include mesenteric lymphadenitis, Castleman disease, and familial Mediterranean fever.

Mesenteric lymphadenitis: Mesenteric lymphadenitis is a well-established clinical entity and was first reported

808 Diagnostic Abdominal Imaging

A

B

C

D

E

F

Figure 13-39 Sclerosing Mesenteritis (Paniculitis) in 2 Patients This 61-year-old man complained of dyspepsia. A-C. Contrastenhanced axial CT images show a faint area of increased attenuation (arrows in A-C) within the central mesenteric fat. There are also mildly enlarged mesenteric lymph nodes (arrowheads in A-C). D. Coronal reconstruction also shows increased fat attenuation (arrow) and enlarged lymph nodes (arrowheads). This is a typical appearance of sclerosing

mesenteritis. E and F. This 59-year-old man was being evaluated for a renal mass (not shown). Axial CT images demonstrate a spiculated, infiltrating mass (arrows) within the small bowel mesentery (arrows in E and F). Differential diagnosis would include a mesenteric sarcoma, desmoid tumor, fibrosis associated with carcinoid tumor, and sclerosing mesenteritis. Biopsy was diagnostic of sclerosing mesenteritis.

in 1913 and can occur at any age but is primarily reported in children.153,154 Its clinical presentation mimics acute appendicitis without perforation, with symptoms of abdominal pain, fever being most prominent but also

including nausea, vomiting, diarrhea, and malaise. Mesenteric lymphadenitis represents inflammation of the mesenteric lymph nodes, usually as a result of infection. Organisms that have been cultured include Escherichia

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 809 coli, Bacteroides, Clostridia, Enterococci, β-hemolytic streptococcus, Staphylococcus, and Yersinia.155,156 Viruses, including coxsackie viruses (A and B), rubeola virus, and adenovirus are also believed to be causes of mesenteric lymphadenitis. These organisms originate in the bowel and in many cases the primary bowel source can be identified. However, in some patients, 30% in 1 series, no bowel abnormality is identified and the only imaging abnormality is lymph node enlargement.157 This latter group is called “primary mesenteric lymphadenitis” and may account for 7% to 14% of patients who are evaluated for suspected appendicitis.158,159 Surgical evaluations usually show intense inflammatory reaction in the ileocecal region with omental adhesions and periappendicitis. Cross-sectional imaging can play an important role in the diagnosis of mesenteric lymphadenitis and can help avoid unnecessary surgery since the disorder can be confused with acute appendicitis, lymphoma, and regional enteritis as well as other entities. Diagnostic features of mesenteric lymphadenitis include enlarged mesenteric lymph nodes with or without associated ileocecal or ileal wall thickening, in the setting of normal appendix.160 Rao et al showed that using their criteria of 3 or more nodes with a short axis diameter of at least 5 mm allowed a correct diagnosis of mesenteric adenitis in all patients.158 Both US and CT have been used to diagnose this disorder (see Figure 13-40).157,159

Castleman Disease Castleman disease, also known as giant lymph node hyperplasia and angiofollicular lymph node hyperplasia, is a rare disorder affecting the lymph nodes with noncancerous growths and is typically divided into unifocal and

A

B

Figure 13-40 Mesenteric Lymphadenitis This 19-year-old man presented with right lower quadrant pain and fever. A-C. Contrast-enhanced axial CT shows an increased number of small pericecal lymph nodes (arrows), with normal or minimally increased number of lymph nodes in the

multifocal forms of disease. The cause is unknown but probably represents either a response to chronic inflammation or a hamartoma of the lymphatic system.161,162 Pathologically, this disease is classified as hyaline, vascular, plasmacytic, or mixed-cellularity types. Castleman disease of the abdomen will typically present as abdominal pain, trouble eating, or abdominal fullness. Multicentric disease will also frequently have systemic symptoms such as fever, weakness, night sweats, weight loss, loss of appetite, nausea, vomiting, and neuropathy, whereas unicentric disease rarely presents with systemic symptoms. Imaging exams will most often demonstrate a solitary, enhancing mass in the retroperitoneum, porta hepatis, mesentery, or pancreas.39-41 Lesions less than 5 cm in diameter will usually appear as a uniformly enhancing mass; however, lesions greater than 5 cm in diameter will often  display inhomogeneous enhancement with lowattenuation areas consistent with necrosis. Calcification can be seen in some cases (see Figures 13-18 and 13-41).163

Familial mediterranean fever: Familial Mediterranean fever is a recessive genetic disease also known as recurrent polyserositis.164 It usually occurs in families and is more common in individuals of Mediterranean descent. It usually presents as fever with brief recurrent episodes of peritonitis, pleuritis, and arthritis.165 The disease is commonly confused with appendicitis and cholecystitis often leading to unnecessary surgery. Pleural and pericardial effusions are common. Arthritis can last longer than abdominal symptoms and in between attacks, joints are normal, with permanent damage rare. Rash and muscle pain can be other symptoms.

C mesentery (arrowheads). The appendix (not shown) appeared normal. This appearance was presumed to represent mesenteric lymphadenitis, and the patient was observed with serial clinical exams. The patient’s symptoms resolved 2 days later without therapy.

810 Diagnostic Abdominal Imaging

A

B

Figure 13-41 Castleman Disease Appearing as a Retroperitoneal Mass This 51-year-old man complained of dyspepsia. A. T2-weighted and (B) T1-weighted postgadolinium axial MRI images show a

heterogenous mass in the left retroperitoneum that has regions that densely enhance. This appearance would be consistent with a soft-tissue sarcoma, but biopsy was diagnostic of Castleman disease.

Imaging studies have shown that mesenteric lymphadenopathy is the predominant finding, present in 86% of patients as with familial Mediterranean fever.166 Other findings include engorged mesenteric vessels, thickened mesenteric folds, ascites, focal peritonitis, splenomegaly, dilated small bowel loops, and mural thickening of the ascending colon. CT can be an important tool in the evaluation of patients with familial Mediterranean fever because of its ability to exclude other causes of abdominal pain, including acute appendicitis and bowel obstruction.167-169

Imaging Studies

ANATOMY AND IMAGING OF THE LYMPHATIC DUCTS Anatomy Interstitial lymph is drained via afferent lymphatics, which converge toward the outer nodal surfaces of lymph nodes before exiting via efferents at centrally located medullary sinuses. Numerous small lymph channels join to form a small tubular structure in the superior retroperitoneum called the cisterna chyli. This is the only lymphatic duct structure that can be seen by conventional cross-sectional imaging. On transaxial images, it appears as a small, round or oval fluid-filled structure anterior to the vertebral bodies at the level of the diaphragmatic crux (see Figure 13-42). The cisterna chyli then empties into 1 or several lymphatic channels that run anterior to the thoracic vertebra and collectively are called the thoracic duct. At the level of the thoracic inlet, if multiple, these channels rejoin to form a single duct that drains lymph in the venous circulation by opening usually at the angle between the left jugular vein and the left subclavian vein.

Very few tools are available at present for imaging the lymphatics. There are 2 primary techniques to evaluate the lymphatic channels, lymphography that utilized radiopaque contrast to anatomically define the lymphatics and functionally define the transport of lipid materials and lymphoscintigraphy, which utilizes a radioactive agent to functionally evaluate the lymph flow. In the past few years, new macromolecular agents, gadoliniumlabeled dendrimers, fluorescent quantum dots, fluorescent labeled immunoglobulins have been used to image lymphatics and sentinel lymph node with MRI and optical imaging.170 With the exception of the cisterna chyli, the lymph channels cannot be seen with conventional crosssectional imaging.

Lymphography Lymphographic studies are useful in the detection of lymphatic fistulas or lymphatic leakage. Lymphography is diagnostically more accurate than CT in normal-sized lymph nodes or small lymph nodes by CT criteria, as this demonstrates internal architecture. Since the advent of crosssectional imaging, especially CT, the use of lymphographic studies has significantly decreased because of the difficulty in performing the procedure and its moderately invasive nature. First, methylene blue is injected into the softtissues between the toes. Then an incision is made into the dorsum of the foot, allowing for direct visualization and cannulation of the methylene blue–filled lymphatic ducts. Fatty (Lipiodol, Ethiodol) or water-soluble contrast is injected into the lymphatics and radiographs of the pelvis, abdomen, and thorax demonstrate the contrast-filled

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 811

A

B

C

D

E

F

Figure 13-42 Cisterna Chyli in 2 Patients A-C. Axial T2-weighted MRI images show a bright oval structure (arrows) behind the crux of the diaphragm. Note how this is narrow at the top and bottom and wider in the middle image, indicating a tubular structure. This is the typical axial appearance of the cisterna chyli, a small reservoir of lymph fluid as the lymph passes from the lower extremities and trunk into the thorax. D. The tubular character of the cisterna chyli (arrow) is

seen better on this longitudinal T2-weighted MRI sequence. E and F. Contrast-enhanced axial CT images show an oval lowattenuation structure (arrows) behind the crux of the diaphragm, typical of the CT appearance of the cisterna chyli. This can be confused with a retroperitoneal lymph node. However, note how the cisterna chyli is lower attenuation than the crux of the diaphragm. A lymph node would have similar attenuation to the diaphragm crux.

lymphatic ducts and lymph nodes (see Figure 13-43). Oily lymphatic contrast will remain in lymph nodes for years and will be seen on all subsequent x-ray and CT examinations of the pelvis, abdomen, thorax, and neck. The contrast is very strongly attenuating and appears nearly metallic on CT scans of the pelvis, abdomen, and thorax (see Figure 13-19). Complications of the procedure are not rare and most often consist of pulmonary oil embolization but also include pulmonary infarction, allergy to methylene blue and Ethiodol, intra-alveolar hemorrhage, systemic embolization, and hypothyroidism.171-176 Lymphography is still an important part of the diagnosis of chylous ascites,

chylothorax, chyluria, external genital lymphedema, and preplanning for thoracic duct embolization. In spite of the decline of lymphography worldwide, its role in diagnosis and localization of damage in lymph vessel is indisputable.177

Lymphoscintigraphy Lymphoscintigraphy has recently replaced the lymphography for the assessment of lymphedema. Lymphoscintigraphy is done by injecting the 99-m-Tc-rhenium sulfide nanocolloids (50-100 nm is average diameter) in a subcutaneous space of the toes and follow its progression with the help of

812

Diagnostic Abdominal Imaging

A

B

C

Figure 13-43 Lymphangiogram in Two Patients A. Image of the pelvis following injection of lymphangiographic contrast show innumerable, fine parallel channels extending along the iliac lymph node chains bilaterally. B. Image of the abdomen in a second patient shows lymphangiographic contrast in fine parallel lymphatics in the left para-aortic chain. The larger

tubular structure (arrow) in the superior abdomen represents the cisterna chyli and beginnings of the thoracic duct. C. Image of the thorax in the same patient as (A) shows continuation of fine lymphatic channels until a single large thoracic duct (arrow) forms in the superior thorax and then empties into the left subclavian vein.

a gamma camera, including static images. In the last 5 to 10 years, renewed interest in the use of lymphoscintigraphy for sentinel node mapping has been seen. The sentinel node is defined as the first lymphatic relay in the drainage territory of a primary tumor and is thought to be the first site of metastasis. This technique is well established for melanoma and breast carcinoma sentinel node mapping (see Figure 13-44).178

Lymphedema can be inherited (primary) or result from injury to the lymphatic vessels (secondary). Primary lymphedema is an inherited disease with incidence of less than 1% of live births. Lymphatic vessels are missing or maldeveloped and will typically present as swelling of 1 limb or several limbs and/or internal organs. Secondary causes are most often a result of prior surgical lymphadenectomy or irradiation for cancer, especially breast cancer, but can also be seen with accidental trauma and neoplastic obstruction of the lymph channels. In tropical climates, lymphatic filariasis, 90% of which is caused by the organism Wucheria bancrofti, is an important cause of lymphedema.

Disorders of the Lymphatic System Disruption or obstruction of the lymphatic system can lead to lymphedema and or lymphatic fistulas and chylous collections in the thorax or abdomen.

Lymphedema Lymphedema also known as lymphatic insufficiency is characterized by the accumulation of lymphatic fluid in the interstitium causing soft-tissue swelling, primarily of the legs or arms, but occasionally involving other parts of the body. Skin affected by chronic lymphedema can become fibrotic with a loss of the normal skin architecture, function, and mobility and can lead to life-threatening infection.

Lymphatic trauma The lymphatic system has a remarkable regenerating ability, and very rarely does injury cause significant morbidity. However, surgical or accidental traumatic injury to lymphatics or obstructing neoplasms can cause lymphatic fistula/lymphocele, or chylous ascites.179 Lymphatic fistula can be diagnosed by lymphangiography or lymphoscintigraphy, showing that the abnormal drainage or fluid collection contains lymph fluid.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 813

A

B

Figure 13-44 Sentinel Node Mapping in Melanoma This 47-year-old male is a known patient of melanoma of the left knee. A. After the injection of 99-m-Tc-sulfur nanocolloid

subcutaneously at the left knee, (B) the draining nodes were mapped, and a sentinel node is highlighted by “(S).”

SUMMARY A wide array of imaging techniques is available for imaging of the lymph nodes and lymphatic ducts. Differentiation of normal and abnormal lymph nodes on cross-sectional imaging has become progressively better, but further refinements in the techniques, especially use of molecular imaging, is necessary so that pathologically small lymph nodes can be detected more accurately. Fundamental knowledge of anatomy of lymphatic drainage pathways is crucial for the staging of abdominopelvic malignancies as prognosis is linked to the initial staging in every malignancy. There is a wide differential diagnosis for lymphadenopathy; however, disease-specific clues can be demonstrated on imaging to reach the diagnosis or at least for narrowing the differential diagnosis. REFERENCES 1. Poirier P, CuneoB, Delamere G. The Lymphatics. Westminster: Archibald Constable & Co, Ltd; 1903. 2. Sappey P. Des vaisseaux lymphatiques. Paris: Delahaye A, Lecrosnier E; 1888:731-842. 3. Warwick R, Williams P. Topography of the lymph nodes and vessels. In: Gray’s Anatomy. Edinburgh: Longman; 1973: 727-744. 4. Whitmore I. Lymphoid system. In: Federative Committee on Anatomical Terminology – Terminologica Anatomica. Stuttgart: Thieme Verlag; 1998:100-103. 5. Testut L. Des lymphatiques. In: Testut L, ed. Traite´ d’anatomie humaine. Paris: Doin; 1893:267-308.

6. Lengele B. The lymphatic system. In: Gregoire V, Scalliet P, Ang K, eds. Clinical Target Volume in Conformal and Intensity Modulated Radiotherapy A Clinical Guide to Cancer Treatment. Berlin: Springer Verlag; 2004:1-36. 7. Rouviere H, Tobias M. Anatomy of the Human Lymphatic System. Ann Arbor: Edwards; 1938. 8. Plentl A, Friedman E. Lymphatic System of the Female Genitalia. The Morphologic Basis of Oncologic Diagnosis and Therapy. Philadelphia, PA: WB Saunders; 1971. 9. Lengelé B, Nyssen-Behets C, Scalliet P. Anatomical bases for the radiological delineation of lymph node areas. Part II: upper limbs chest and abdomen. Radiother Oncol. 2007;84: 335-347. 10. Henrikson E. The lymphatic spread of carcinoma of the cervix and of the body of the uterus. A study of 420 necropsies. Am J Obstet Gynec. 1949;58:924. 11. Kwok H, Bissett I, Hill G. Preoperative staging of rectal cancer. Int J Colorectal Dis. 2000;15:9-20. 12. Bluemke D, Cameron J, Hruban R, et al. Potentially resectable pancreatic adenocarcinoma: spiral CT assessment with surgical and pathologic correlation. Radiology. 19956;197:381-385. 13. Diehl S, Lehmann K, Sadick M, Lackmann R, Georgi M. Pancreatic cancer: value of dual-phase helical CT in assessing resectability. Radiology. 1998;206:373-378. 14. Gmeinwieser J, Feuerbach S, Hohenberger W, et al. Spiral CT in diagnosis of vascular involvement of pancreatic cancer. Hepatogastroenterology. 1995;42:418-422. 15. Studer U, Scherz S, Scheidegger J, et al. Enlargement of regional lymph nodes in renal cell carcinoma is often not due to metastases. J Urol. 1990;144(2 pt 1):243-245.

814 Diagnostic Abdominal Imaging 16. Perrotti M, Kaufman RJ, Jennings T, et al. Endo-rectal coil magnetic resonance imaging in clinically localized prostate cancer: is it accurate? J Urol. 1996;156:106-109. 17. Bezzi M, Kressel H, Allen K, et al. Prostatic carcinoma: staging with MR imaging at 1.5 T. Radiology. 1988;169:339-346. 18. Jager G, Barentsz J, Oosterhof G, et al. Pelvic adenopathy in prostatic and urinary bladder carcinoma: MR imaging with a three-dimensional TI-weighted magnetization-prepared-rapid gradient- echo sequence. AJR Am J Roentgenol. 1996;167:1503-1507. 19. Kier R, Wain S, Troiano R. Fast spin-echo MR images of the pelvis obtained with a phased-array coil: value in localizing and staging prostatic carcinoma. AJR Am J Roentgenol. 1993;161: 601-606. 20. Tuzel E, Sevinc M, Obuz F, et al. Is magnetic resonance imaging necessary in the staging of prostate cancer? Urol Int. 1998;61:227-231.

34. Bar-Shalom R, Yefremov N, Guralnik L, et al. Clinical performance of PET/CT in evaluation of cancer: additional value for diagnostic imaging and patient management. J Nucl Med. 2003;44:1200-1209. 35. Mikosch P, Gallowitsch HJ, Zinke-Cerwenka W, et al. Accuracy of whole-body 18F-FDP-PET for restaging malignant lymphoma. Acta Med Austriaca. 2003;30:41-47. 36. Al-Hallaq H, River J, Zamora M, Oikawa H, Karczma G. Correlation of magnetic resonance and oxygen microelectrode measurements of carbogen induced changes in tumor oxygenation. Int J Radiat Oncol Biol Phys. 1998;41:151-159. 37. Rosen M, Schnall M. Dynamic contrast-enhanced magnetic resonance imaging for assessing tumor vascularity and vascular effects of targeted therapies in renal cell carcinoma. Clin Cancer Res. 2007;13(2 pt 2):770s-776s.

21. Hall T, MacVicar A. Imaging of bladder cancer. Imaging. 2001;13:1-10.

38. Herts B, Megibow A, Birnbaum B, Kanzer G, Noz M. Highattenuation lymphadenopathy in AIDS patients: significance of findings at CT. Radiology. 1992;185:777-781.

22. Barentsz J, Jager G, van Vierzen P, et al. Staging urinary bladder cancer after transurethral biopsy: value of fast dynamic contrast-enhanced MR imaging. Radiology. 1996;20:185-193.

39. Rahmouni A, Golli M, Mathieu D, Anglade M-C, Charlotte F, Vasile N. Castleman disease mimicking liver tumor: CT and MR features. J Comput Assist Tomogr. 1992;16:699-703.

23. Barentsz J, Ruijs S, Strijk S. The role of MR imaging in carcinoma of the urinary bladder. AJR Am J Roentgenol. 1993;160:937-947.

40. Lepke R, Pagani J. Pancreatic Castleman disease simulating pancreatic carcinoma on computed tomography. J Comput Assist Tomogr. 1982;6:1193-1195.

24. Tavares N, Demas B, Hricak H. MR imaging of bladder neoplasms: correlation with pathologic staging. Urol Radiol. 1990;12:27-33.

41. Garber S, Shaw D. Case report: the ultrasound and computed tomography appearance of mesenteric Castleman disease. Clin Radiol. 1991;43:429-430.

25. Creasman W, Morrow C, Bundy B, et al. Surgical pathologic spread patterns of endometrial cancer: a Gynecologic Oncology Group study. Cancer. 1987;60:2035-2041.

42. Glazer H, Semenkovich J, Gutierrez F. Mediastinum. In: Lee J, Sagel S, Stanley R, Heiken J, eds. Computed Body Tomography With MRI Correlation. 3rd ed. Philadelphia, PA: LippincottRaven Publishers; 1998:261-349.

26. Manfredi R, Mirk P, Maresca G, et al. Local-regional staging of endometrial carcinoma: role of MR imaging in surgical planning. Radiology. 2004;231:372-378.

43. Reddy D, Salomon C, Demos TC, Cosar. E. Mesenteric lymph node cavitation in celiac disease. AJR Am J Roentgenol. 2002;178:247.

27. Kim S, Choi B, Kim J, et al. Preoperative staging of uterine cervical carcinoma: comparison of CT and MRI in 99 patients. J Comput Assist Tomogr. 1993;17:633-640.

44. Huppert B, Farrell M. Case 60: cavitating mesenteric lymph node syndrome. Radiology. 2003;228:180-184.

28. Yang W, Lam W, Yu M, et al. Comparison of dynamic helical CT and dynamic MR imaging in the evaluation of pelvic lymph nodes in cervical carcinoma. AJR Am J Roentgenol. 2000;175:759-766.

45. Pallisa E, Sanz P, Roman A, Majó J, Andreu J, Cáceres J. Lymphangioleiomyomatosis: pulmonary and abdominal findings with pathologic correlation. RadioGraphics. 2002;22:S185-S198.

29. Tempany C, Zou K, Silverman, et al. Staging of advanced ovarian cancer: comparison of imaging modalities—report from the Radiology Oncology Group. Radiology. 2000;215: 761-767.

46. Facts 2009-2010. White Plains: The Leukemia and Lymphoma Society; 2010.

30. Ricke J, Sehouli J, Hach C, et al. Prospective evaluation of contrast-enhanced MRI in the depiction of peritoneal spread in primary or recurrent ovarian cancer. Eur Radiol. 2003;13: 943-949. 31. Antoch G, Saoudi N, Kuehl H, et al. Accuracy of whole-body dual-modality fluorine-18-2-fluoro-2-deoxy-d-glucose positron emission tomography and computed tomography (FDG-PET/ CT) for tumor staging in solid tumors: comparison with CT and PET. J Clin Oncol. 2004;22:4357-4368. 32. von Schulthess G, Steinert H, Hany T. Integrated PET/CT: current applications and future directions. Radiology. 2006; 238:405-422. 33. Gambhir S, Czernin J, Schwimmer J, Silverman D, Coleman R, Phelps M. A tabulated summary of the FDG PET literature. J Nucl Med. 2001;42(5 suppl):1S-93S.

47. Skarin A, Dorfman D. Non-Hodgkin’s lymphomas: current classification and management. CA Cancer J Clin. 1997;47: 351-372. 48. Harris N, Jaffe ES, Stein H, et al. A revised European-American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood. 1994;84: 1361-1392. 49. NCI Non-Hodgkin’s Classification Project Writing Committee: National Cancer Institute sponsored study of classification of non-Hodgkin’s lymphomas: summary and description of a Working Formulation for Clinical Usage. Cancer. 1982;49: 2112-2135. 50. Altekruse S, Kosary C, Krapcho M, et al. SEER Cancer Statistics Review, 1975-2007. Bethesda: National Cancer Institute; 2010. 51. Stewart S, King JB, Thompson TD, Friedman C, Wingo PA. Cancer mortality surveillance—United States, 1990-2000. MMWR Surveill Summ. 2004;53:1-108.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 815 52. Bouzourene H, Haefliger T, Delacretaz F, Saraga E. The role of Helicobacter pylori in primary gastric MALT lymphoma. Histopathology. 1999;34:118-123.

72. Munker R, Glass J, Griffeth L, et al. Contribution of PET imaging to the initial staging and prognosis of patients with Hodgkin’s disease. Ann Oncol. 2004;15:1699-1704.

53. Ratner L. Adult T cell leukemia lymphoma. Front Biosci. 2004;1:2852-2859.

73. Kumar R, Maillard I, Schuster S, Alavi A. Utility of fluorodeoxyglucose-PET imaging in the management of patients with Hodgkin’s and non-Hodgkin’s lymphomas. Radiol Clin North Am. 2004;42:1083-1100.

54. Wooldridge J, Link B. Post-treatment surveillance of patients with lymphoma treated with curative intent. Semin Oncol. 2003;30:375-381. 55. Coleman C, Williams C, Flint A, Glatstein E, Rosenberg S, Kaplan H. Hematologic neoplasia in patients treated for Hodgkin’s disease. N Engl J Med. 1977;297:1249-1252. 56. Canellos G, Arseneau J, DeVita V, Whang-Peng J, Johnson R. Second malignancies complicating Hodgkin’s disease in remission. Lancet. 1975;26:947-949. 57. Killebrew D, Shiramizu B. Pathogenesis of HIV-associated non-Hodgkin lymphoma. Curr HIV Res. 2004;2:215-221. 58. Adami J, Gabel H, Lindelof B, et al. Cancer risk following organ transplantation: a nationwide cohort study in Sweden. Br J Cancer. 2003;89:1221-1227. 59. Fisher S, Fisher R. The epidemiology of non-Hodgkin’s lymphoma. Oncogene. 2004;232:6524-6534. 60. Facts 2008-2009. White Plains: The Leukemia and Lymphoma Society, 2009:1-21. 61. Banks P. The pathology of Hodgkin’s disease. Semin Oncol. 1990;17:683. 62. Colby T, Hoppe R, Warnke R. Hodgkin’s disease: a clinicopathologic study of 659 cases. Cancer. 1982;49:1848-1858. 63. Shankar A, Ashley S, Radford M, Barrett A, Wright D, Pinkerton C. Does histology influence outcome in childhood Hodgkin’s disease? Results from the United Kingdom Children’s Cancer Study Group. J Clin Oncol. 1997;15:2622-2630. 64. Smolewski P, Robak T, Krykowski E, et al. Prognostic factors in Hodgkin’s disease: multivariate analysis of 327 patients from a single institution. Clin Cancer Res. 2000;6:1150-1160. 65. Alexander F, Jarrett R, Cartwright R, et al. Epstein-Barr virus and HLA-DPB1-*0301 in young adult Hodgkin’s disease: evidence for inherited susceptibility to Epstein-Barr virus in cases that are EBV(+ve). Cancer Epidemiol Biomarkers Prev. 2001;110:705-709. 66. Cobby M, Whipp E, Bullimore J, et al. CT appearances of relapse of lymphoma in the lung. Clin Radiol. 1990;41:232-238. 67. Rosenberg S, Kaplan J. Evidence for an orderly progression in the spread of Hodgkin’s disease. Cancer Res. 1966;26:1225-1231. 68. Tucker M, Coleman C, Cox R, Varghese A, Rosenberg S. Risk of second cancers after treatment for Hodgkin’s disease. N Engl J Med. 1988;318:76-81. 69. Bangerter M, Moog F, Buchmann I, et al. Whole-body 2-18F-fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET) for accurate staging of Hodgkin’s disease. Ann Oncol. 1998;9:1117-1122.

74. Jerusalem G, Beguin Y, Najjar F, et al. Positron emission tomography (PET) with 18F-fluorodeoxyglucose (18F-FDG) for the staging of low-grade non-Hodgkin’s lymphoma (NHL). Ann Oncol. 2001;12:825-830. 75. Radford J, Cowan R, Flanagan M, et al. The significance of residual mediastinal abnormality on the chest radiograph following treatment for Hodgkin’s disease. J Clin Oncol. 1988;6:940-946. 76. Jochelson M, Mauch P, Balikian J, Rosenthal D, Canellos G. The significance of the residual mediastinal mass in treated Hodgkin’s disease. J Clin Oncol. 1985;3:637-640. 77. Orlandi E, Lazzarino M, Brusamolino E, et al. Residual mediastinal widening following therapy in Hodgkin’s disease. Hematol Oncol. 1990;8:125-131. 78. Jerusalem G, Beguin Y, Fassotte M, et al. Whole-body positron emission tomography using 18F-fluorodeoxyglucose for posttreatment evaluation in Hodgkin’s disease and nonHodgkin’s lymphoma has higher diagnostic and prognostic value than classical computed tomography scan imaging. Blood. 1999;94:429-433. 79. Zijlstra J, van der Werf G, Hoekstra O, et al. 18F-fluorodeoxyglucose positron emission tomography for post-treatment evaluation of malignant lymphoma: a systematic review. Haematologica. 2006;91:522-552. 80. Nyman R, Forsgren G, Glimelius B. Long-term follow-up of residual mediastinal masses in treated Hodgkin’s disease using MR imaging. Acta Radiol. 1996;37:323-326. 81. Buerke B, Puesken M, Muter S, et al. Measurement accuracy and reproducibility of semiautomated metric and volumetric lymph node analysis in MDCT. AJR Am J Roentgenol. 2010;195: 979-985. 82. Rademaker J. Diagnostic imaging modalities for the assessment of lymphoma with special emphasis on CT, MRI and US. PET Clin. 2006;1:219-230. 83. Mueller P, Ferrucci JJ, Harbin W, et al. Appearance of lymphomatous involvement of the mesentery by ultrasound and body computed tomography: the sandwich sign. Radiology. 1980;134:467-473. 84. Moog F, Bangerter M, Diederchs C, et al. Lymphoma: role of whole-body 2-deoxy-2-{F-18} fluoro-D-glucose (FDG) PET in nodal staging. Radiology. 1997;203:795-800. 85. Rahmouni A, Divine M, Lepage E, et al. Mediastinal lymphoma: quantitative changes in gadolinium enhancement at MR imaging after treatment. Radiology. 2001;219:621-628.

70. Jerusalem G, Beguin Y, Fassotte MF, et al. Whole-body positron emission tomography using 18F-fluorodeoxyglucose compared to standard procedures for staging patients with Hodgkin’s disease. Haematologica. 2001;86:266-273.

86. Nakayama T, Yoshimitsu K, Irie J. Usefulness of the calculated apparent diffusion coefficient value in the differential diagnosis of retroperitoneal masses. Magn Reson Imaging. 2004;20: 735-742.

71. Kabickova E, Sumerauer D, Cumlivska E, et al. Comparison of 18F-FDG-PET and standard procedures for the pretreatment staging of children and adolescents with Hodgkin’s disease. Eur J Nucl Med Mol Imaging. 2006;33:1025-1031.

87. Dobert N, Menzel C, Hamscho N, Wordehoff W, Kranert W, Grunwald F. Atypical thoracic and supraclavicular FDG-uptake in patients with Hodgkin’s and non-Hodgkin’s lymphoma. Q J Nucl Med Mol Imaging. 2004;48:33-38.

816

Diagnostic Abdominal Imaging

88. Cohade C, Osman M, Pannu H, Wahl R. Uptake in supraclavicular area fat (“USA-Fat”): description on 18F-FDG PET/CT. J Nucl Med. 2003;44:170-176.

107. Jeong Y, Ha H, Yoon C, et al. Gastrointestinal involvement in Henoch-Schönlein syndrome: CT findings. Am J Roentgenol. 1997;168:965-968.

89. Yeung H, Grewal R, Gonen M, Schoder H, Larson S. Patterns of [18]F-FDG uptake in adipose tissue and muscle: a potential source of false-positives for PET. J Nucl Med. 2003;44: 1789-1796.

108. Siskind B, Burrell M, Pun H, Russo RJ, Levin W. CT demonstration of gastrointestinal involvement in HenochSchönlein syndrome. Gastrointest Radiol. 1985;10:352-354.

90. Johansson B. Brown fat: a review. Metabolism. 1959;8:221-240. 91. Cheson B, Bennett J, Grever M, et al. National Cancer Institutesponsored Working Group guidelines for chronic lymphocytic leukemia: revised guidelines for diagnosis and treatment. Blood. 1996;87:4990-4997. 92. Muntañola A, Bosch F, Arguis P, et al. Abdominal computed tomography predicts progression in patients with Rai stage 0 chronic lymphocytic leukemia. J Clin Oncol. 2007;25:1576-1580. 93. Vardiman J, Thiele J, Arber D, Brunning R, Borowitz M, Porwit A. The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: rationale and important changes. Blood. 2009;114:937-951. 94. Ben Romdhane K, Ben Romdhane N, Ben Younes M, Ayadi S, Ben Ayed M. Systemic mastocytosis: a case report. Arch Anat Cytol Pathol. 1990;38:100-103. 95. Avila NA, Ling A, Worobec AS, Mican JM, Metcalfe DD. Systemic mastocytosis: CT and US features of abdominal manifestations. Radiology. 1997 Feb;202(2):367-372. 96. Pantongrag-Brown L, Krebs T, Daly B, et al. Frequency of abdominal CT findings in AIDS patients with M. avium complex bacteraemia. Clin Radiol. 1988;53:816-819. 97. Nyberg D, Federle M, Jeffrey R, Bottles K, Wofsy C. Abdominal CT findings of disseminated Mycobacterium avium-intracellulare in AIDS. AJR Am J Roentgenol. 1985;145:297-299. 98. 1993 Revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Recomm Rep. 1992;41(RR-17):1-19. 99. Yee J, Wall S. Gastrointestinal manifestations of AIDS. Gastroenterol Clin North Am. 1995;24:413-434. 100. Koh D, Burn P, Mathews G, Nelson M, Healy J. Abdominal computed tomographic findings of Mycobacterium tuberculosis and Mycobacterium avium intracellulare infection in HIV seropositive patients. Can Assoc Radiol J. 2003;54:45-50. 101. Bhargava D, Tandon H, Chawla T, Shriniwas, Tandon B, Kapur B. Diagnosis of ileocecal and colonic tuberculosis by colonoscopy. Gastrointest Endosc. 1985;31:68-70. 102. Shah S, Thomas V, Mathan M, et al. Colonoscopic study of 50 patients with colonic tuberculosis. Gut. 1992;33:347-351. 103. Singh V, Kumar P, Kamal J, Prakash V, Vaiphei K, Singh K. Clinicocolonoscopic profile of colonic tuberculosis. Am J Gastroenterol. 1996;91:565-568. 104. Radin D. Disseminated histoplasmosis: abdominal CT findings in 16 patients. AJR Am J Roentgenol. 1991;157:955-958. 105. Kitsanou M, Andreopoulou E, Bai M, Elisaf M, Drosos A. Extensive lymphadenopathy as the first clinical manifestation in systemic lupus erythematosus. Lupus. 2000;9:140-143. 106. Calguneri M, Ozturk M, Ozbalkan Z, et al. Frequency of lymphadenopathy in rheumatoid arthritis and systemic lupus erythematosus. J Int Med Res. 2003;31:345-349.

109. Akinyemi E, Rohewal U, Tangorra M, et al. Gastric sarcoidosis. J Natl Med Assoc. 2006;98:948. 110. Warshauer D, Molina P, Hamman S, et al. Nodular sarcoidosis of the liver and spleen: analysis of 32 cases. Radiology 1995; 195:757-762. 111. Fazzi P, Solfanelli S, Morelli G, et al. Sarcoidosis: single bulky mesenteric lymph node mimicking a lymphoma. Sarcoidosis. 1995;12:75-77. 112. Miller WTJ, Perez-Jaffe LA. Cross-sectional imaging of Kikuchi disease. J Comp Assist Tomogr Issue. 1999;23:548-551. 113. Kim C, Hyun O, Yoo I, Kim S, Sohn H, Chung S. Kikuchi disease mimicking malignant lymphoma on FDG PET/CT. Clin Nucl Med. 2007;32:711-712. 114. Chamulak G, Brynes R, Nathwani B. Kikuchi-Fujimoto disease mimicking malignant lymphoma. Am J Surg Pathol. 1990;14:514-523. 115. Jayaraj S, Lloyd J, Frosh A, Patel K. Kikuchi-Fujimoto’s syndrome masquerading as tuberculosis. J Laryngol Otol. 1999;113:82-84. 116. Dorfman R, Berry G. Kikuchi’s histiocytic necrotizing lymphadenitis: an analysis of 108 cases with emphasis on differential diagnosis. Semin Diagn Pathol. 1988;5: 329-345. 117. Falk R, Comenzo R, Skinner M. The Systemic Amyloidoses. N Engl J Med. 1997;337:898-909. 118. Kyle R, Gertz M. Primary systemic amyloidosis: clinical and laboratory features in 474 cases. Semin Hematol. 1995;32:45-59. 119. Spitale L, Jimenez D, Montenegro R. Localised primary amyloidosis of inguinal lymph node with superimposed bone metaplasia. Pathology. 1998;30:321-322. 120. Kahn H, Strauchen J, Gilbert H, Fuchs A. Immunoglobulinrelate d amyloidosis presenting as recurrent isolated lymph node involvement. Arch Pathol Lab Med. 1991;115:948-950. 121. Kyle R, Bayrd E. Amyloidosis: review of 236 cases. Medicine. 1975;54:271-299. 122. Takebayashi S, Ono Y, Sasaki F, et al. Computed tomography of amyloidosis involving retroperitoneal lymph nodes mimicking lymphoma. J Comput Assist Tomogr. 1984;8:1025-1027. 123. Suga K, Shimizu K, Kawakami Y, et al. Lymphatic drainage from esophagogastric tract: feasibility of endoscopic CT lymphography for direct visualization of pathways. Radiology. 2005;237:952-960. 124. Efremidis S, Vougiouklis N, Zafiriadou E, et al. Pathways of lymph node involvement in upper abdominal malignancies: evaluation with high-resolution CT. Eur Radiol. 1999;9: 868-874. 125. Coakley F, Lin R, Schwartz L, Panicek D. Mesenteric adenopathy in patients with prostate cancer: frequency and etiology. AJR Am J Roentgenol. 2002;178:125-127. 126. Creasman W, Morrow C, Bundy B, et al. Surgical pathologic spread patterns of endometrial cancer: a Gynecologic Oncology Group study. Cancer. 1987;60:2035-2041. 127. MacVicar A. Bladder cancer staging. BJU Int. 2000;86(suppl 1): 111-122.

Chapter 13 Imaging of the Lymph Nodes and Lymphatic Ducts 817 128. Kumar R, Zhuang H, Alavi A. PET in the management of urologic malignancies. Radiol Clin North Am. 2004;42:1141-1153. 129. Low RN, McCue M, Barone R, Saleh F, Song T. MR staging of primary colorectal carcinoma: comparison with surgical and histopathologic findings. Abdom Imaging. 2003;28:784-793. 130. Brown g, Richards C, Bourne M, et al. Morphologic predictors of lymph node status in rectal cancer with use of high-spatialresolution MR imaging with histopathologic comparison. Radiology. 2003;227:371-377. 131. Lowe V, Booya F, Fletcher J, et al. Comparison of positron emission tomography, computed tomography, and endoscopic ultrasound in the initial staging of patients with esophageal cancer. Mol Imaging Biol. 2005;7:422-430. 132. Pfau P, Perlman S, Stanko P, et al. The role and clinical value of EUS in a multimodality esophageal carcinoma staging program with CT and positron emission tomography. Gastrointest Endosc. 2007;65:377-384. 133. McAteer D, Wallis F, Couper G, et al. Evaluation of 18F-FDG positron emission tomography in gastric and oesophageal carcinoma. Br J Radiol. 1999;72:525-529. 134. Yun M, Lim J, Noh S, et al. Lymph node staging of gastric cancer using (18)F-FDG PET: a comparison study with CT. J Nucl Med 2005;46:1582-1588.

146. Nagamata H, Inadama E, Arihiro S, Matsuoka M, Torii A, Fukuda K. The usefulness of MDCT in Crohn’s disease. Nippon Shokakibyo Gakkai Zasshi. 2002;99:1317-1325. 147. Trommer G, Bewer A, Kosling A. Mesenteric lymphadenopathy in Yersinia enterocolitica infection. Radiologe. 1998;38:37-40. 148. Yang Z, Min P, Sone S, et al. Tuberculosis versus lymphomas in the abdominal lymph nodes: evaluation with contrastenhanced CT. AJR Am J Roentgenol. 1999;172:619-623. 149. Blachar A, Federle M, Brancatelli G. Primary biliary cirrhosis: clinical, pathologic, and helical CT findings in 53 patients. Radiology. 2001;220:329-336. 150. del Olmo J, Esteban J, Maldonado L, et al. Clinical significance of abdominal lymphadenopathy in chronic liver disease. Ultrasound Med Biol. 2002;28:297-301. 151. Horton KM, Lawler LP, Fishman EK. CT findings in sclerosing mesenteritis (panniculitis): spectrum of disease. Radiographics. 2003;23:1561-1567. 152. Sabate JM, Torrubia S, Maideu J, Franquet T, Monill JM, Perez C. Sclerosing mesenteritis: imaging findings in 17 patients. AJR Am J Roentgenol. 1999;172:625-629. 153. Mitchell OWH. Acute suppurative lymphadenitis abdominal due to diplostreptococcus; autopsy. Am J Med Sci. 1913;145:721-723.

135. Tian J, Chen L, Wei B, et al. The value of vesicant 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET) in gastric malignancies. Nucl Med Commun. 2004;25:825-831.

154. Asch MJ, Amoury RA, Toulukian RJ, Santulli TV. Suppurative mesenteric lymphadenitis. Am J Surg. 1968;115:570-573.

136. Chen J, Cheong J, Yun M, et al. Improvement in preoperative staging of gastric adenocarcinoma with positron emission tomography. Cancer. 2005;103:2383-2390.

156. Domingo T, Alvear T, Kain M. Suppurative mesenteric lymphadenitis, a forgotten clinical entity: report of two cases. J Pediatr Surg. 1975;19:969-970.

137. Saga T, Higashi T, Ishimori T, et al. Clinical value of FDG-PET in the follow up of post-operative patients with endometrial cancer. Ann Nucl Med. 2003;17:197-203.

157. Macari M, Hines J, Balthazar E, Megibow A. Mesenteric adenitis: CT diagnosis of primary versus secondary causes, incidence, and clinical significance in pediatric and adult patients. Am J Roentgenol. 2002;178:853-858.

138. Heesakkers RAM, Hövels AM, Jager GJ, et al. MRI with a lymph-node-specific contrast agent as an alternative to CT scan and lymph-node dissection in patients with prostate cancer: a prospective multicohort study. Lancet Oncol. 2008;9:850-856. 139. Harisinghani M, Barentsz J, Hahn P, et al. Noninvasive detection of clinically occult lymph-node metastases in prostate cancer. N Engl J Med. 2003;348:2491-2499. 140. Lange P. ProstaScint scan for staging prostate cancer. Urology. 2001;57:402-406. 141. Bermejo C, Coursey J, Basler J, et al. Histological confirmation of lesions identified by ProstaScint scan following definitive treatment. Urol Oncol. 2003;21:349-352. 142. Polascik T, Manyak M, Haseman M, et al. Comparison of clinical staging algorithms and 111Indium-capromab pendetide immunoscintigraphy in the prediction of lymph node involvement in high risk prostate carcinoma patients. Cancer. 1999;85:1586-1592. 143. Harris R. Computerized tomography of retroperitoneal lymphadenopathy: benign or malignant? Comput Tomogr. 1979;3:73-80. 144. Subramanyam B, Balthazar E, Homii S, Hilton S. Abdominal lymphadenopathy in intravenous drug addicts: sonographic features and clinical significance. AJR Am J Roentgenol. 1985; 144:917-920. 145. Goldberg H, Gore R, Margulis A, Moss A, Baker E. Computed tomography in the evaluation of Crohn disease. AJR Am J Roentgenol. 1983;140:277-282.

155. Collins DG. Mesenteric lymphadenitis in adolescents simulating appendicitis. Can Med Assoc J. 1936;34:402-405.

158. Rao P, Rhea J, Novelline R. CT diagnosis of mesenteric adenitis. Radiology. 1997;202:145-149. 159. Puylaert J. Mesenteric adenitis and acute terminal ileitis: US evaluation using graded compression. Radiology. 1986;161: 691-695. 160. Borgia G, Ciampi R, Nappa S, et al. Tuberculous mesenteric lymphadenitis clinically presenting as abdominal mass: CT and sonographic findings. J Clin Ultrasound. 1985;13:491-493. 161. Abell M. Lymphoid hamartoma. Radiol Clin North Am. 1868; 6:15-24. 162. Gloviczki P, Lowell R. Lymphatic complications of vascular surgery. In: Rutherford R, ed. Vascular Surgery. Philadelphia, PA: WB Saunders; 2001:781-789. 163. Meador TL, McLarney JK. CT features of Castleman disease of the abdomen and pelvis. AJR Am J Roentgenol. 2000;175:115-118. 164. Gershoni-Baruch R, Brik R, Shinawi M, Livneh A. The differential contribution of MEFV mutant alleles to the clinical profile of familial Mediterranean fever. Eur J Hum Genet. 2002; 10:145-149. 165. Ben-Chetrit E, Levy M. Familial Mediterranean fever. Lancet. 1998;351:659-664. 166. Zissin R, Rathaus V, Gayer G, Shapiro-Feinberg M, Hertz M. CT findings in patients with familial Mediterranean fever during an acute abdominal attack. Br J Radiol. 2003;76:22-25.

818 Diagnostic Abdominal Imaging 167. Rao P, Rhea J, Novelline R, Mostafavi A, McCabe C. Effect of computed tomography of the appendix on treatment of patients and use of hospital resources. N Engl J Med. 1998;15: 141-146.

174. Fein D, Hanlon A, Corn B, Curran WJ, Coia L. The influence of lymphangiography on the development of hypothyroidism in patients irradiated for Hodgkin’s disease. Int J Radiat Oncol Biol Phys. 1996;36:13-18.

168. Burkill G, Bell J, Healy J. The utility of computed tomography in acute small bowel obstruction. Clin Radiol. 2001;56: 350-359.

175. Kusumoto S, Imamura A, Watanabe K. Case report: the incidental lipid embolization to the brain and kidney after lymphography in a patient with malignant lymphoma—CT findings. Clin Radiol. 1991;44:279-280.

169. Boudiaf M, Soyer P, Terem C, Pelage J, Maissiat E, Rymer R. CT evaluation of small bowel obstruction. Radiographics. 2001;21:613-624. 170. Lucarelli R, Ogawa M, Kosaka N, Turkbey B, Kobayashi H, Choyke P. New approaches to lymphatic imaging. Lymphat Res Biol. 2009;7:205-214. 171. Vogl T, Bartjes M, Marzec K. Contrast-enhanced lymphography: CT or MR imaging? Acta Radiol Suppl 1997;412:47-50.

176. Winterer J, Blum U, Boos S, Konstantinides S, Langer M. Cerebral and renal embolization after lymphography in a patient with non-Hodgkin lymphoma: case report. Radiology. 1999;210:381-383. 177. Guermazi A, Brice P, C H, Sarfati E. Lymphography: an old technique retains its usefulness. Radiographics. 2003;23: 1541-1558.

172. Mortazavi S, Burrows B. Allergic reaction to patent blue dye in lymphangiography. Clin Radiol. 1971;22:389-390.

178. Morton D, Wen D, Wong J, et al. Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg. 1992;127:392-399.

173. Dupont H, Timsit J, Souweine B, Gachot B, Bedos J, Wolff M. Intra-alveolar hemorrhage following bipedal lymphography. Intensive Care Med. 1996;22:614-615.

179. Gloviczki P, Lowell R. Lymphatic complications of vascular surgery. In: Rutherford R, ed. Vascular Surgery. Philadelphia, PA: WB Saunders; 2001:781-789.

CHAPTER

14

Imaging of the Spleen Lauren Ehrlich, MD Wallace T. Miller Jr., MD

I. ANATOMY OF THE SPLEEN II. IMAGING OF THE SPLEEN III. UNIFOCAL SPLENIC LESIONS a. Cystic-Appearing Nodules and Masses of the Spleen i. Neoplasms ii. Focal infections and abscesses of the spleen iii. Splenic trauma iv. Other causes of cystic nodules and masses b. Solid-Appearing Nodules and Masses of the Spleen i. Lymphoma ii. Metastasis iii. Hemangioma iv. Angiosarcoma v. Hamartoma vi. Inflammatory pseudotumor c. Nonspherical Focal Lesions of the Spleen i. Trauma ii. Infarction IV. MULTIFOCAL SPLENIC LESIONS a. Multifocal Neoplasms i. Lymphoma ii. Metastasis iii. Multifocal vascular tumors b. Multifocal Infections i. Granulomatous infections of the spleen ii. Fungal microabscesses iii. Pneumocystis infection iv. Hepatosplenic cat scratch disease

ANATOMY OF THE SPLEEN The spleen is a functionally diverse organ with active roles in immunosurveillance, red blood cell breakdown, splenic contraction for blood volume augmentation during hemorrhage, and hematopoiesis. The spleen lies within the left upper quadrant of the peritoneal cavity and abuts ribs 9 to 12, the fundus of the stomach, the upper pole of the left kidney, the splenic

c. Infiltrative Diseases i. Sarcoidosis ii. Gaucher disease iii. Peliosis V. SPLENOMEGALY AND OTHER DIFFUSE DISORDERS a. Splenomegaly i. Venous congestion ii. Infiltrative malignancies iii. Nonneoplastic infiltrative diseases iv. Infectious causes of splenomegaly v. Noninfectious inflammatory conditions causing splenomegaly vi. Hemoglobinopathies and splenomegaly vii. Extramedullary hematopoiesis and splenomegaly b. Splenic Atrophy: Autosplenectomy c. Unique Anomalies of the Spleen i. Wandering spleen ii. Heterotaxy syndromes iii. Splenic calcifications iv. Iron deposition in the spleen VI. POSTOPERATIVE FINDINGS RELATED TO SPLENIC SURGERY a. Normal Postoperative Findings i. Splenosis b. Complications of Splenic Surgery i. Intra-abdominal abscess ii. Portal or splenic vein thrombosis iii. Injuries to adjacent structures

flexure of the colon, and the tail of the pancreas. During childhood, the size of the spleen is based on age and weight, and tables of normal splenic size are available in the literature.1 A normal adult spleen weighs approximately 150 g and is approximately 11 cm in craniocaudal length. A craniocaudal measurement of 11 to 13 cm is frequently used as the upper limit of normal for splenic size in imaging studies. The spleen is composed of lymphatic follicles and

819

820

Diagnostic Abdominal Imaging

reticuloendothelial cells (“white pulp”) and vascular sinusoids (“red pulp”). The relative percentage of white pulp increases with age because of accumulated antigenic exposure. The spleen has a lenticular shape, convex superolateral, and concave inferomedial. Clefts and lobules are common. Accessory spleens, also known as supernumerary spleens or splenules, are the most common congenital anomalies of the spleen, detected in 10% to 30% of patients at autopsy, and in approximately 16% of CT studies.2 Their location is variable, but most often are near the hilum and tail of the pancreas.3 The vast majority of accessory spleens are asymptomatic and discovered incidentally. However, recognizing them is important in order not to misdiagnose an accessory spleen for lymphadenopathy or for a tumor arising from adjacent organs. Also, before splenectomy in a patient with a hematologic or autoimmune disorder, the presence of an accessory spleen must be known in order to remove all functional splenic tissue. Accessory spleens are usually less than 4 cm in diameter, are round or ovoid, have well-demarcated margins, and are characterized by their similar imaging characteristics to normal splenic parenchyma (see Figure 14-1). These characteristics should enable the reader to distinguish splenules from other causes of left upper quadrant masses.

Figure 14-1 Splenule Contrast-enhanced CT shows a small uniformly enhancing nodule (arrow) medial to the body of the spleen. This is a typical appearance of a splenule or small accessory spleen, a common normal variant. Splenules will have imaging characteristics that match the main body of the spleen on all imaging modalities.

IMAGING OF THE SPLEEN On abdominal and chest radiographs, the spleen will appear as a lenticular-shaped opacity in the left upper quadrant of the abdomen beneath the diaphragm. In some individuals, the medial edge of the spleen will be sharply defined because of the different attenuation of adjacent abdominal fat (see Figure 14-2). In other individuals, the size and location of the spleen can only be inferred by the displacement of adjacent gas-containing intestines. Splenomegaly is 1 of the few abnormalities of the solid abdominal viscera that can be reliably demonstrated on abdominal x-rays. In addition, calcified splenic abnormalities may be detected by plain film, including splenic artery aneurysms, calcified cysts, calcified granulomas, and calcified hematomas. Sonography is frequently the first imaging modality employed to evaluate the spleen, because it is noninvasive and relatively inexpensive. Ultrasonography (US) of the spleen is performed with the patient in the supine or right lateral decubitus position, with the spleen visualized beneath the left costal margin or between rib interspaces. A 3.0- to 5.0-MHz transducer is generally used. On US, the spleen is homogeneous, slightly more echogenic than normal kidney, and isoechoic or slightly greater in echogenicity than liver. The spleen is relatively hypervascular on Doppler interrogation (see Figure 14-3). On unenhanced CT, the spleen is homogeneous, with attenuation typically measuring 40 to 60 Hounsfield units (HUs). Unenhanced splenic attenuation is normally 5 to 10 HU less than that of liver. With contrast, spleen enhancement is initially heterogeneous, likely reflecting variation in the flow rate through red pulp in the spleen.4 Bizarre enhancement patterns are the norm, particularly during the arterial and early venous phases of enhancement. Later venous-phase images or delayed images

Figure 14-2 Spleen on Abdominal X-ray This 57-year-old man was being evaluated for nephrolithiasis. The presence of abdominal fat makes the outline of the spleen (arrows) and liver (arrowheads) visible on this abdominal x-ray.

Chapter 14 Imaging of the Spleen 821 slightly greater than that of skeletal muscle. The spleen appears high signal on T2-weighted imaging, greater than the liver. Inhomogeneous and bizarre enhancement patterns of gadolinium enhancement of the spleen are similar to those seen with iodinated contrast on CT (see Figure 14-5).5 104mTc-sulfur colloid is now only occasionally used to clarify splenic anatomy. 104mTc-labeled heat-damaged red cells can be used to assess splenic function and anatomy. The advantage of this technique is uptake only by the spleen, not the liver as also occurs with sulfur colloid. 18F-FDG-PET/CT is now used as a noninvasive imaging modality, predominantly in assessing solid splenic masses. Physiologic uptake of 18F-FDG by normal splenic tissue is generally uniformly low.6

UNIFOCAL SPLENIC LESIONS Figure 14-3 Normal Splenic Ultrasonography US image of the normal spleen shows the normal crescentic shape and uniform echogenicity.

greater than 70 seconds after the initiation of contrast injection will show homogeneous enhancement of the normal spleen (see Figure 14-4). Normally, the liver and spleen densities are within 25 HU on dynamic contrast-enhanced CT scan images. On MR, the spleen is usually evaluated with both T1and T2-weighted spin-echo sequences. On T1-weighted imaging, the spleen is hypointense to hepatic tissue and

A

B

Figure 14-4 CT of the Spleen A. Unenhanced image of the spleen shows the typical crescentshaped spleen with attenuation similar to the kidney, liver, and other soft-tissue organs. B. Arterial-phase image shows the

Unifocal splenic lesions can be cystic or solid appearing and spherical or nonspherical in shape. Our differential diagnosis of solitary splenic lesions will revolve around these imaging characteristics.

Cystic-Appearing Nodules and Masses of the Spleen A solitary cystic-appearing nodule is the most common imaging abnormality identified on routine cross-sectional imaging. Hemangiomas before contrast enhancement are the most common cause, with other causes including lymphangiomas, splenic abscesses, and posttraumatic or infarction cysts (Table 14-1).

C normal inhomogeneous enhancement of the spleen. C. Late venous-phase image now shows uniform enhancement of the spleen similar to liver and hyperattenuating relative to paraspinal muscle.

822

Diagnostic Abdominal Imaging

A

B

C

D

Figure 14-5 A Splenic MRI A. T2-weighted image shows the spleen to be slightly hyperintense to liver. B. T1-weighted image shows the spleen to be slightly hypointense to liver. C. T1-weighed arterial-phase

image shows the typical inhomogeneous enhancement of the spleen. D. T1-weighted venous-phase image shows the typical uniform enhancement of the spleen on late-phase images.

Table 14-1. Solitary Cystic Appearing Splenic Nodules or Masses

Neoplasms

A. Neoplasms 1. Hemangiomaa 2. Lymphangioma 3. Cystic metastasis B. Infections 1. Pyogenic abscess 2. Amebic abscess 3. Echinococcus C. Hematoma D. Posttraumatic and postinfarction cysts E. Epidermoid cysts F. Pancreatic pseudocyst aDisorders in bold are most common.

Cystic-appearing neoplasms of the spleen are common and include hemangiomas, lymphangiomas, and rarely cystic metastasis.

Hemangioma: Hemangiomas are the most common benign splenic neoplasm and can occur as either solitary, multiple, or diffuse lesions.7 They have a prevalence at autopsy ranging from 0.3% to 14%.8 Histologically, they are composed of endothelial-lined blood-filled spaces of varying size and can be characterized by the size of these spaces as capillary or cavernous lesions. The latter are more common. Infrequent associations include Kasabach-Merritt syndrome (anemia, thrombocytopenia, and consumptive coagulopathy) and Klippel-Trénaunay-Weber syndrome (cutaneous hemangiomas, venous varicosities, and softtissue and bony hypertrophy of an extremity). In most cases, hemangiomas of the spleen are small asymptomatic lesions that are discovered accidentally.

Chapter 14 Imaging of the Spleen 823 Imaging Notes 14-1. Hemangioma Hemangioma is the most common cause of an incidentally detected focal lesion of the spleen.

Splenic hemangiomas have a variety of imaging appearances depending on the proportion of cavernous and capillary features. They are the most common incidentally discovered focal lesion in the spleen. Calcification can occur as either scattered, punctuate, curvilinear, or radiating densities but is unusual. US may show both hypoechoic and hyperechoic regions.9 Focal hemangiomas typically appear as well-defined hypoattenuating lesions on noncontrast CT, resembling small cysts but can be isoattenuating to the spleen and be missed entirely (see Figure 14-6).8 On US examinations, the multiple vascular walls of hemangiomas will usually make them uniformly echogenic and are, therefore, rarely confused with cystic lesions (see Figure 14-7). Although, occasionally they will appear as a complex cystic and solid mass where the larger cavernous channels appear as cystic regions.10 On MRI, hemangiomas typically appear hypointense or isointense with the remainder of the spleen on T1-weighted images. They are usually hyperintense relative

A

B

Figure 14-7 Splenic Hemangioma This 85-year-old man had chronic renal disease. A. Axial T1-weighted MRI sequence demonstrates a faintly hypointense 1.8-cm nodule (arrow) in the spleen. B. The nodule (arrow) is very intense on this coronal T2-weighted MRI sequence.

Figure 14-6 Hemangioma on Unenhanced CT This 38-year-old man had end-stage renal disease. Unenhanced CT shows a large hypoechoic mass in the spleen that is slightly more hyperattenuating than the ascites surrounding the liver. This is a nonspecific mass but is statistically most likely to represent a hemangioma. US examination (not shown) had the typical appearance of a hemangioma.

C This is most likely to represent a hemangioma but could indicate a small lymphangioma. C. Longitudinal US of the spleen shows the nodule (arrow) to be very echogenic, a feature characteristic of hemangioma but would be atypical for a lymphangioma.

824

A

Diagnostic Abdominal Imaging

B

C

Figure 14-8 CT of Splenic Hemangioma This 75-year-old woman was being evaluated for an abdominal aortic stent graft. A. Unenhanced CT shows a normal-appearing spleen. B. Arterial-phase image taken at 35 s after injection shows a small focal low-attenuation lesion. C. Late venous-

phase image taken at 2 min after injection shows the lesion to be isoattenuating with the spleen. Most standard enhanced CT scans are taken between 35 and 60 s after injection, and therefore hemangiomas of the spleen will typically appear as a hypoattenuating mass. Note the incidental left adrenal adenoma.

to normal spleen on T2-weighted images, although slight heterogeneity is sometimes seen, reflecting the variation in the size of blood-filled spaces and the varying amounts of internal fibrosis and hemorrhage (see Figure 14-7).8 Following intravenous (IV) contrast administration, the appearance of hemangiomas also shows considerable variation. In the arterial phase and of enhancement, approximately 35 seconds after contrast administration, they will appear as hypoattenuating/hypointense nodules relative to the enhancing spleen and can resemble a cyst (see Figure 14-8). During the early venous phase, at approximately 60 seconds after contrast administration, the lesions will demonstrate centripetal enhancement from the periphery with persistence on delayed images. Unlike the classic peripheral nodular discontinuous enhancement characteristic of hepatic hemangiomas, this pattern of enhancement is rare in splenic hemangiomas, with most lesions showing a continuous peripheral rim of enhancement of variable thickness with an irregular inner margin (see Figure 14-9).8 Others can show immediate homogeneous enhancement (see Figure 14-10). Mottled enhancement has also been described, with areas remaining hypoattenuating relative to normal spleen, particularly in hemangiomas containing fibrotic components. In the late venous phase, 90 to 120 seconds after contrast administration, the mass will typically become isoattenuating to the spleen.

When the appearance is atypical, Tc-99m-labeled RBC scintigraphy can be utilized to confirm a diagnosis by demonstrating increased activity within the lesion on delayed images.11 With Tc-99m radiocolloid scanning, both photopenic areas and areas of increased activity have been reported with splenic hemangiomas.12

Lymphangioma: Splenic lymphangioma is a rare benign lesion usually diagnosed in childhood.13 It can occur as solitary or multiple splenic lesions, or as diffuse involvement replacing most of the splenic parenchyma, called lymphangiomatosis.14 An association with lymphangiomas in other sites (most commonly mediastinum, axilla, and neck) has also been noted.15 Although typically asymptomatic, lymphangiomas can grow quite large, causing splenomegaly and pain, or compressing adjacent structures. Lesions can be characterized as capillary, cavernous, or cystic depending on the size of the lymphatic channels. On imaging, the typical cystic lymphangioma is a welldefined, multilocular cystic lesion with thin septations. Cross sectional imaging typically reveals cysts of various sizes, ranging from a few millimeters to several centimeters in diameter.8 They are often found in a subcapsular location or around larger trabeculae.16 Lesions are anechoic or hypoechoic on US, and hypoattenuating on CT.17

Chapter 14 Imaging of the Spleen 825

A

B

C

D

Figure 14-9 Centripetal Enhancement of Splenic Hemangioma This 55-year-old man had colon cancer. A. T2-weighted MRI sequence demonstrates a 2-cm lobulated nodule in the spleen that most likely represents a hemangioma or lymphangioma. Contrast-enhanced T1-weighted sequences in the (B) arterial

phase, (C) early portal phase, and (D) late portal phase of enhancement show the nodule to progressively decrease in size, typical of centripetal enhancement seen in hemangiomas. Note the faint nodular enhancement (arrow) in the periphery of the nodule in (C).

Smaller lymphatic spaces, as can occur in capillary or cavernous lesions, can appear solid on CT and hyperechoic on US.18 Calcification is occasionally noted. No enhancement of the cystic spaces is noted following IV contrast administration, but moderate enhancement of the septa can be seen. On MRI, the lesions show typical hyperintense signal on T2-weighted images and are usually hypointense on T1-weighted images.19 The presence of proteinaceous fluid or hemorrhage may result in increased T1 signal (see Figure 14-11).

instances, a metastasis could appear as a thin-walled cyst within the substance of the spleen.

Cystic metastasis: The majority of metastasis will appear as solid nodules or masses within the spleen. Some will demonstrate central cavitation but will still show solid elements in the periphery of the nodule. In very rare

Focal infections and abscesses of the spleen Focal infections of the spleen are rare and include bacterial, mycobacterial, fungal, and protozoal causes. Bacterial and echinococcal infections are typically solitary and will be discussed here; however, mycobacterial and fungal abscesses typically cause multiple small cystic lesions and will be discussed in the subsequent section called Multifocal Splenic Lesions.

Pyogenic abscesses of the spleen: Pyogenic or bacterial abscesses involving the spleen are rare. In 1 study of hospitalizations in Denmark, only 20 splenic abscesses were

826

Diagnostic Abdominal Imaging

A

B

Figure 14-10 Uniformly Enhancing Splenic Hemangioma This 72-year-old man had bladder cancer. A. Unenhanced CT shows no splenic abnormality. B. Contrast-enhanced CT image shows a focal hyperenhancing lesion (arrow) in the posterior aspect of the spleen. Although most hemangiomas will have

slow-flowing blood and will appear as a hypointense nodule resembling a cyst on early-phase images and will fill in slowly over time, occasionally, some flash-filling hemangiomas will appear uniformly hyperattenuating on early-phase images, like this example.

seen over a 5-year period, accounting for 0.05% of hospitalizations and 0.005% of hospital deaths.20 The most common mechanism of splenic abscess is via hematogenous dissemination of organisms from another site of infection. This is most often a result of endocarditis but can be due to a wide variety of sites, including urinary tract infections, gastrointestinal infections, pneumonia, mycotic aneurysms, and osteomyelitis.21-23 Superinfection of necrotic tissue following splenic infarction or splenic trauma is also a well recognized cause for splenic abscess.23-28 Direct extension of infection from an adjacent organ can also occasionally occur from diverticulitis, subphrenic abscesses, pancreatic abscess, and other retroperitoneal infections. A variety of conditions have been associated with splenic abscesses, including, trauma, malignancies, corticosteroid use, AIDS, IV drug abuse, cirrhosis, alcoholism

and diabetes mellitus.28-31 Immunocompromised patients account for an increasing proportion of patients with splenic abscess, between one-fourth and one-third of patients.30,31 Patients typically present with fever, left upper quadrant pain, splenomegaly, and a left pleural effusion.22,28,32 Most will have a leukocytosis and approximately 2/3rds of patients will have positive blood cultures.23 Imaging, primarily CT and to a lesser extent US and MRI, is the primary diagnostic method for diagnosing splenic abscesses. CT has been shown to have a 92% to 98% sensitivity and US has been shown to have an approximately 87% sensitivity for splenic abscess.28,30,31 Splenic abscess carries a 15% to 35% mortality rate.20,28,32 Standard therapy, as with most abdominal abscesses, is systemic antibiotics and percutaneous or surgical abscess drainage or splenectomy.20,22,28,33

Chapter 14 Imaging of the Spleen 827

A

D

B

E

C

F

Figure 14-11 Two Splenic Lymphangiomas A-C. This 65-year-old woman had small cell lymphoma. (A) Coronal T2-weighted MRI sequence demonstrates a lobulated hyperintense mass in the spleen. (B) Arterial-phase and (C) venous-phase T1-weighted MRI images show the nodule to remain low attenuation without enhancement. This could represent a hemangioma with very slow flow but probably represents a small lymphangioma. D-F. The

second patient was an 83-year-old woman with breast cancer. (D) Venous-phase (60 s post injection) contrast-enhanced CT shows a hypoattenuating mass. (E) T2-weighted MRI image shows a septated high-signal mass indicating a complex cyst. (F) Late venous phase (90 s post injection) gadolinium-enhanced MRI shows a nonenhancing mass. These findings are most consistent with a diagnosis of a splenic lymphangioma.

Bacterial infections are the most common cause of splenic abscesses, accounting for between 56% and 80% of abscesses in 4 large series.28-31,34 No organisms are discovered in 11%-29% of cases. A wide variety of bacteria have been cultured from splenic abscess including gram negative and gram positive, aerobic and anaerobic organisms.22,28 A metaanalysis of reported cases from 1987 to 1995 showed Staphylococcus, Salmonella, and Escherichia coli to be the most common organisms cultured.31 The majority, approximately 70%, of bacterial abscesses will appear as a solitary fluid filled mass.28,35

Patients with multiple abscesses will often have underlying immunosuppression as a result of malignancies, corticosteroid use, AIDS, cirrhosis, alcoholism, or diabetes mellitus.28 Most fungal and mycobacterial abscesses will appear as multiple abscesses and so the majority of unilocular abscesses will be bacterial in origin.36 On US, lesions are typically internally avascular and may show either low echogenicity or complex internal echoes from debris.37 A solitary splenic abscess appears as a hypodense (20-40 HU) area on CT with low signal on T1-weighted MR images and intermediate or increased

828 Diagnostic Abdominal Imaging

A

B

Figure 14-12 Pyogenic Splenic Abscesses in 2 Patients A. This 63-year-old man with non-Hodgkin lymphoma developed splenic infarctions (not shown) as a result of splenomegaly. One month later, he developed fevers and Clostridium septicum septicemia. Contrast-enhanced CT shows a large low-attenuation

mass with multiple small bubbles of gas indicating development of a pyogenic splenic abscess complicating splenic infarction. B. The second patient is a 42-year-old man with Staphylococcus aureus endocarditis. CT of the spleen without contrasts shows multiple hypoattenuating masses in the spleen indicating septic emboli.

signal on T2-weighted images. The margins may be smooth or irregular.37 Gas is infrequently seen within the abscess. Following IV contrast administration, peripheral enhancement may be seen (see Figure 14-12).36

cells, fibroblasts, giant cells, and eosinophils, which create a few-millimeter-thick, rigid protective layer. The middle laminated membrane is a fibrous 2-mm, acellular structure that permits the passage of nutrients but is impervious to bacteria. Disruption of the laminated membrane predisposes to dissemination of the infection within the host. Outpouchings of the thin translucent inner germinal membrane, also called the brood capsule, produce scolices, the infectious embryonic tapeworms. Freed scolices together with brood capsules form a granular, sometimes calcified, material called hydatid sand that settles in the dependent part of the cyst. The clear cyst fluid is a transudate of serum, containing echinococcal proteins. This fluid is highly antigenic and, if released into the circulation of the host, can cause eosinophilia or anaphylaxis.41 Hydatid cysts are usually solitary. The cysts can be unilocular, in which case they are indistinguishable from other cysts, or they can contain multiple internal “daughter cysts,” a finding that is pathognomonic for echinococcal infection.41 Many cysts will demonstrate partial or complete rim calcification. Some cysts will develop fine psammomatous calcifications within the fluid of the cyst, a finding that represents hydatid sand. In some cases, the membranes of the cyst can collapse and float centrally within the fluid-filled exocyst. This appearance is essentially diagnostic of echinococcal cysts. Most of the various findings of echinococcal cysts can be demonstrated by US, CT, and MRI.

Amebic abscess: Amebic abscesses of the spleen are rare but have been reported to represent between 0% and 12% of splenic abscesses.28-31,34 Their appearance is indistinguishable from the more common bacterial abscess.

Echinococcal infection of the spleen: Although rare in the United States, echinococcal cysts historically have been the most common non-neoplastic cause worldwide for a cystic lesion in spleen.38 Echinococcus has a complex life cycle that requires development within both a canine animal, usually the dog, and a herd animal, usually a sheep. Humans are accidental hosts and become infected with the organism through contact with canine feces, usually in the form of contaminated, raw vegetables. Echinococcal infection results in the formation of complex cysts in a variety of organs, most commonly the liver and lung. Splenic involvement with Echinococcus granulosa is unusual, occurring in only between 1% and 3% of echinococcal infections.39 Isolated splenic involvement is very uncommon.40 Echinococcal cysts have 3 layers: an outer layer, or pericyst; a middle laminated membrane; and an inner germinal membrane.41 The pericyst represents the response of the host to the parasite and consists of modified host

Chapter 14 Imaging of the Spleen 829 MRI is capable of adequately demonstrating all features of hydatid disease with the exception of calcifications. On MRI, the cysts are hyperintense on T2-weighted images and hypointense on T1-weighted images.45 Daughter cysts are typically hypointense on T1-weighted images relative to fluid in the parent cyst.45 A continuous low-intensity rim surrounding the cyst corresponding to the dense fibrous capsule encasing the parasitic membranes is frequently seen.46

Splenic trauma

Figure 14-13 Echinococcal Cyst of the Spleen This 38-year-old woman had emigrated from Uzbekistan. She had a family history of polycystic kidney disease. Contrastenhanced CT scan shows the characteristic cyst within cyst appearance of echinococcal infection.

In most cases, plain films of the abdomen and chest will fail to detect echinococcal cysts of the spleen. In some cases, radiographs of the abdomen will show a soft-tissue mass in the left upper quadrant of the abdomen displacing the stomach, left kidney, and/or splenic flexure of the colon. Rarely, rim calcification of the cyst will indicate the cystic nature of the left upper quadrant mass. Chest radiograph can demonstrate an elevated left hemidiaphragm or pleural effusion in some cases.42 Various sonographic patterns of hydatid cysts have been described. A simple anechoic cyst is observed most often—a nonspecific finding with the differential diagnosis of all causes of splenic cysts.43 Presence of collapsed membranes within a cystic lesion is pathognomonic for hydatid disease.44 Demonstration of daughter cysts is also diagnostic of echinococcal infection. Echinococcal cysts appear as well-circumscribed, lowdensity lesions on CT.41 Lesions are often large and can be unilocular or contain daughter cysts distributed either peripherally or throughout the lesion (see Figure 14-13).41 The multiple cysts within larger cysts are diagnostic of hydatid disease and in most cases are easily distinguished from loculated cysts. The daughter cysts appear as perfect spheres within the larger cyst, rather than multiple septations as are seen in multiloculated cysts. In most cases, the daughter cysts are slightly less dense on CT relative to fluid in the parent cyst.41 A “serpent” or “snake” sign is occasionally noted, representing collapsed parasitic membranes within the cyst.41 Cyst wall calcification is noted commonly, and when extensive, suggests that the cyst is dead. Rarely, small calcified granules can be found in the dependent portions of the cysts and represent hydatid sand. Typically, no enhancement is noted following IV contrast administration.

Splenic trauma can result in a variety of focal lesions, including lacerations, hematomas, and vascular injuries resulting in pseudoaneurysms or active bleeding. Splenic trauma is discussed in detail under the heading Nonspherical Focal Lesions of the Spleen. However, splenic hematomas can appear as a cystic mass in the spleen and, in some cases, will leave a posttraumatic cyst and will be discussed briefly here.

Hematoma: Hematomas are among the most common causes of a cystic mass in the spleen. Subcapsular hematomas are more common than intraparenchymal hematomas and will appear lenticular or crescentic shaped adjacent to the capsule of the spleen. These are discussed in detail in a subsequent section of this chapter titled: Nonspherical Focal Lesions of the Spleen. Intraparenchymal hematomas will appear as round, oval, or irregularly shaped lesions within the substance of the spleen. In the early phases, the margins will often be indistinct and the contents of the hematoma will appear hyperattenuating to nonenhanced spleen on CT, will have multiple internal echoes on US, and will appear hyperintense on T1- and T2-weighted MRI sequences. As the hematoma ages, it will become progressively more sharply marginated with the splenic parenchyma and the contents will become progressively more simpleappearing: water attenuation on CT, anechoic on US, and hypointense on T1 to hyperintense on T2 MRI sequences (see Figure 14-14).

Posttraumatic and postinfarction cysts: Traumatic and ischemic damage to the spleen can evolve into a cyst that will remain for the lifetime of the individual. These posttraumatic and postinfarction cysts are the most common cyst of the spleen accounting for approximately 80% of splenic cysts.47,48 In some cases, the wall of these cysts will become calcified (see Figure 14-15).

Other causes of cystic nodules and masses The remaining causes of solitary cystic lesions of the spleen are epidermoid cysts and pancreatic pseudocysts.

Epidermoid cysts: Epidermoid or mesothelial cysts are uncommon congenital lesions of the spleen. They constitute less than 10% of nonparasitic splenic cysts and were incidentally found in 32 of 42 327 autopsies at 1 institution.49

830 Diagnostic Abdominal Imaging

A

B

Figure 14-14 Small Intrasplenic Hematoma This 86-year-old woman was in a motor vehicle accident. A. Enhanced CT scan demonstrates an irregularly shaped hypoechoic mass in the spleen. This could represent a

small intrasplenic hematoma or an incidentally discovered hemangioma. B. US examination shows the lesion to be hypoechoic, a finding consistent with a small hematoma and not a hemangioma that would be echogenic.

A

B

Figure 14-15 Postinfarction Splenic Cyst This 74-year-old woman had a history of non-Hodgkin lymphoma many years previously. A. The current examination shows a 3.5-cm cystic mass in the anterior aspect of the spleen with multiple small punctate rim calcifications seen as high

attenuation foci in the rim of the cyst. It is likely that this woman had lymphoma-related splenomegaly and this cyst formed from a splenic hematoma caused by infarction or trauma to the enlarged spleen. B. As a result of the rim calcification, the cyst could be seen on the lateral view of this patient’s chest radiograph.

Chapter 14 Imaging of the Spleen 831 These are true cysts with a fibrous wall and stratified squamous epithelium filled with serous fluid.50 They appear as simple cysts on all cross-sectional imaging.

Pancreatic pseudocyst: Rarely, pancreatic pseudocysts can dissect along the splenorenal ligament into the splenic parenchyma resulting in an intrasplenic pseudocyst.51 This will appear as a simple or complex cyst within the substance of the spleen. In most cases, there will be other imaging findings of pancreatitis evident in the pancreas and peripancreatic tissues.

Solid-Appearing Nodules and Masses of the Spleen All solitary solid-appearing nodules and masses of the spleen are neoplasms. They are most often due to lymphoma or metastasis but can occasionally be due to several other benign or malignant neoplasms of the spleen (Table 14-2).

Lymphoma The combination of Hodgkin disease and non-Hodgkin lymphoma is the most common malignancy affecting the spleen.52 Isolated or primary splenic involvement occurs in 1% to 2% of all lymphomas, the majority of which will represent Hodgkin disease.53 Lymphoma of the spleen can have a variety of imaging appearances, most commonly as splenomegaly. However, lymphoma can also appear as solitary or multiple round or oval lesions in the spleen. These lesions can have a miliary appearance or appear as larger nodular lesions (see Figure 14-16).

Figure 14-16 Lymphoma Appearing as a Solitary Splenic Mass This 73-year-old woman had right upper quadrant pain when this unsuspected mass in the spleen was discovered. Follow-up examinations showed progression of the mass and development of new abdominal adenopathy. Lymph node biopsy was diagnostic of non-Hodgkin lymphoma.

or multiple in number and can range from a few millimeters to several centimeters in diameter (see Figure 14-17). Splenic metastases are discussed most completely under the heading Multifocal Splenic Lesions later in this text.

Metastasis

Hemangioma

Although metastases to the spleen are uncommon, they represent 1 of the most common causes of solitary and multiple solid nodules in the spleen. When splenic metastases are present, multiple other metastatic sites such as lymph nodes, liver, and lungs are usually present.54 Common primary cancers that metastasize to the spleen include breast, lung, melanoma, ovary, stomach, pancreas, liver, and colon cancer. Splenic metastatic lesions may be solitary

Hemangiomas are composed of tangles of blood vessels ranging from small capillaries to large venules. Because the majority of the mass of a hemangioma represents blood coursing through the lesion, most hemangiomas will appear as a small round or oval lesion mimicking a simple cyst on MRI and CT examinations, appearing low attenuation on CT and low signal on T1-weighted sequences and very high signal on T2-weighted MRI sequences. However, some hemangiomas can have a solid appearance on CT, US, and MRI imaging because they will show homogenous or heterogenous enhancement on CT and MRI examinations (see Figure 14-10). On US examinations, the multiple vascular walls of the hemangioma make them uniformly echogenic and are therefore rarely confused with cystic lesions (see Figure 14-7). Hemangiomas are discussed in detail in the previous section titled: Cystic-Appearing Nodules and Masses of the Spleen.

Table 14-2. Solitary Solid-Appearing Splenic Nodules or Masses 1. Lymphomaa 2. Metastasis 3. Hemangioma 4. Angiosarcoma

Angiosarcoma

5. Hamartoma

Splenic angiosarcomas are rare highly aggressive lesions with nearly 80% of patients dead at a median interval of 6 months following diagnosis.55 Metastases are frequently present at diagnosis. The most common symptoms of

6. Inflammatory pseudotumor aDisorders in bold are most common.

832

Diagnostic Abdominal Imaging

A

B

Figure 14-17 Solitary Splenic Metastasis from Lung Cancer This 49-year-old man had lung cancer. A. Image from a contrast-enhanced CT shows a solitary focal low-attenuation mass in the anterior aspect of the spleen. There is also vague heterogeneity of the remainder of the spleen typical of the early

phase of contrast enhancement of the spleen. B. CT image from 2 years prior shows no evidence of the mass. These findings are indicative of a splenic metastasis from lung cancer. Also note the enlarged left adrenal gland in (A) indicating an adrenal metastasis.

angiosarcoma are abdominal pain and weight loss. Splenomegaly is common, with splenic rupture also noted. Splenic angiosarcoma has a heterogeneous appearance on imaging. US may show a large mass with areas of both increased and decreased echogenicity. CT, most commonly, demonstrates an enlarged spleen in which the normal parenchyma is almost entirely replaced by a heterogeneously attenuating mass or masses. Less common is a discrete solitary mass. Lesions range in size considerably and may show areas of necrosis. Areas of increased density may be noted from acute hemorrhage, hemosiderin deposits, or calcification. At MRI, lesions are heterogeneously hypointense on T1-weighted images and hyperintense on T2-weighted images. Areas of increased T1 signal can also be noted and may correspond to regions of hemorrhage. Enhancement with IV contrast is variable, with lesions showing both hypo- and hyperenhancement relative to normal spleen.56

can range in size up to 20 cm in diameter.57 They are discovered incidentally or due to mass-related symptoms. Typically, a splenic hamartoma appears as a wellcircumscribed mass. On US, lesions typically are uniform solid masses, however, rarely they can have a heterogeneous echotexture from cystic regions.8 Calcification is occasionally noted. On noncontrast CT, the lesions are usually isoor hypodense relative to spleen and can be missed entirely or only recognized by the contour abnormality produced by the mass (see Figure 14-18).8 As a consequence, both US and MRI are more sensitive for their detection.8 On MRI, hamartomas are usually isointense on T1-weighted images and heterogeneously hyperintense on T2-weighted images. A mildly hypointense appearance on T2-weighted imaging has also been reported and may reflect an increased fibrous component.58 After IV contrast administration, there is usually diffuse heterogeneous enhancement on both CT and MRI.9 The diffuse nature of early enhancement may be useful in distinguishing this lesion from the typical peripheral enhancement noted with hemangiomas. On delayed images, persistent hyperenhancement has been noted and may help distinguish hamartoma from lymphoma. Persistent areas of hypodensity or hypointensity may be seen and correspond to areas of necrosis within the lesion.

Hamartoma Splenic hamartomas are rare benign lesions composed of an anomalous mixture of normal splenic red pulp elements and have an incidence at autopsy of 0.024% to 0.13%.8 They are, in part, distinguished from normal splenic tissue by the absence of organized lymphoid follicles. Splenic hamartomas are usually solitary lesions and

A

B

C

D

E

F

Figure 14-18 Splenic Hamartomas in 2 Patients A-D. This 69-year-old man had a renal cell carcinoma of the kidney. Contrast-enhanced CT (not shown) of the spleen appeared normal. (A) T2-weighted MRI examination shows a faintly hyperintense smoothly marginated mass in the spleen. (B) T1-weighted image shows the mass to be nearly isointense with spleen. Coronal T1-weighted images in (C) arterial phase and (D) portal venous phase of enhancement shows initial heterogenous enhancement followed by uniform enhancement. These features are essentially diagnostic of a splenic hamartoma. E and F. Patient 2 is an 80-year-old man who had right upper quadrant pain. (E) US of the spleen in the longitudinal and transverse plane demonstrates a large solid, homogenous mass (arrowheads) that is slightly hypoechoic relative to the spleen. (F) On CT scan, the mass is isoattenuating to the spleen and is only recognizable because it deforms the surface of the spleen (arrow). MRI showed similar findings to the first patient.

834

Diagnostic Abdominal Imaging

Reticuloendothelial activity, either increased or somewhat decreased relative to normal spleen, has been noted on scintigraphy with either Tc-99m stannous phytate or sulfur colloid or with heat-treated Cr-51-labeled RBCs but is not invariably present. Such activity has also been noted with hemangioma, but when present may help distinguish hamartoma from lymphoma or metastasis where it is generally absent.59

Inflammatory pseudotumor Inflammatory pseudotumor consists of a polymorphous population of inflammatory and spindle cells with varying amounts of granulomatous reaction, fibrosis, and necrosis.60 Inflammatory pseudotumors are rare, benign, and of uncertain etiology. Patients can be asymptomatic or present with a mass and vague systemic symptoms of fever and malaise. These lesions are more common in adults than children. Inflammatory pseudotumor typically appears as a well-circumscribed solitary mass that ranges in size from a few centimeters to greater than 12 cm.61 On noncontrast CT, the lesions are generally heterogeneously hypodense. Peripheral and stippled calcification has been noted.62 Lesions have been reported as hypoechoic on US. On T1- and T2-weighted MRI, the lesions have been described as both slightly hypo- as well as slightly hyperintense relative to background spleen. Following IV contrast, mild-to-moderate enhancement has been noted with the lesion remaining hypo- or isodense and intense relative to normal spleen.63

Nonspherical Focal Lesions of the Spleen Posttraumatic lesions and infarction are the 2 most common nonspherical lesions of the spleen.

Trauma The spleen is the intra-abdominal organ most often injured as a result of blunt trauma, and is the second most commonly injured solid organ in penetrating trauma. The spleen is the most vascular organ in the body, and therefore bleeding from splenic injury is potentially life-threatening. Furthermore, delayed hemorrhage days and weeks after splenic injury is a known complication of splenic injury. When other abdominal injuries are excluded from analysis, the mortality rate for isolated splenic injury is still substantial at approximately 8.3% of patients, and the most severe splenic injury, a completely shattered spleen or one with a hilar vascular injury has a mortality of approximately 17.4%.64 Since the 1970s, management of splenic injuries has progressively moved from operative to nonoperative management. Between 53% and 77% of patients with blunt splenic injury meet currently accepted criteria for selective nonoperative management.65-69 Of those managed nonoperatively, 2% to 11% will require subsequent surgical

intervention. Management decisions are based on the clinical presentation, especially hemodynamic stability and on the cross-sectional imaging appearance of the splenic injury. Therefore, imaging plays a critical role in management decisions of splenic injuries. Splenic injuries include intraparenchymal and subcapsular hematomas, lacerations, acute bleeding, and vascular injuries. Focused assessment with sonography for trauma (FAST) is a quick and noninvasive method used by the majority of level 1 trauma centers in the United States for detecting intraperitoneal blood as a marker of blunt intraperitoneal injury.70 Four areas are surveyed for the presence of blood: the pericardial cavity, right upper quadrant, left upper quadrant, and pelvis. Studies have shown that US can detect as little as 100 mL of fluid in the most dependent areas of the peritoneal cavity.71 However, a high number of significant abdominal organ injuries occur without associated hemoperitoneum, and US detection of these injuries has been inconsistent such that other diagnostic tests are often necessary.72-74 The availability of multidetector-row CT (MDCT) at most trauma centers in the United States has led to increased detection and diagnostic accuracy in splenic injury. In general, MDCT is the imaging modality of choice in the patient who is hemodynamically stable. It is important to image for splenic injury during the portal venous phase of enhancement.75 Theoretically, MRI would have similar diagnostic accuracy; however, its use is impractical in most settings. Since the early 1980s, angiography with splenic artery embolization has been used as an adjunct to increase the number of patients managed nonoperatively.76 Multiple techniques can be used, including proximal main splenic artery embolization, selective distal embolization, and a combination of these techniques.76 No significant differences in outcomes or complication rates have been reported between proximal and distal splenic artery embolization.77 Indications for splenic arteriography reported in the literature include CT evidence of active bleeding, vascular injury, high-grade injury, and large-volume hemoperitoneum.78 However, 1 cohort study of 154 patients comparing splenic artery embolization with splenectomy demonstrated a significantly higher incidence of ARDS in the embolization group and demonstrated a 22% failure rate of nonoperative management requiring subsequent surgery and questioned the use of splenic artery embolization.17

Grading system for splenic trauma: The American Association for the Surgery of Trauma has developed a detailed organ injury scaling (OIS) system for the evaluation of blunt organ injuries (Table 14-3). This is based on the most accurate assessment of the injury at imaging examinations, laparotomy, or autopsy. Grade I injuries are small subcapsular hematomas or small lacerations. Grade II lesions represent intermediate subcapsular and small intraparenchymal hematomas and intermediate lacerations. Grade III lesions are large subcapsular

Chapter 14 Imaging of the Spleen 835 Table 14-3. American Association for the Surgery of Trauma Splenic Injury Scale Grade Injury Type

Description of Injury

I

Hematoma subcapsular Laceration

Surface area < 10% < 1 cm

II

Hematoma subcapsular Hematoma intraparenchymal Laceration

Surface area 10% to 50% ≤ 5 cm 1-3 cm, trabecular vessel not involved

III

Hematoma subcapsular Hematoma intraparenchymal Hematoma ruptured Laceration

Surface area > 50% or expanding > 5 cm or expanding Either subcapsular or intraparenchymal Depth > 3 or involving trabecular vessels

IV

Laceration

Involving segmental or hilar vessels or producing devascularization > 25% of spleen

V

Hematoma Laceration

Completely shattered spleen Hilar vascular injury devascularizes spleen

Note: Organ injury scale (OIS) for the spleen, based on the most accurate assessment at autopsy, laparotomy, or radiologic study (1994 revision)198 Source: American Heart Association.

or intraparenchymal hematomas or large lacerations. Grade IV lesions are lacerations that produce infarction of greater than 25% of the spleen. Grade V lesions represent either a completely shattered spleen or one that is completely infarcted. In a review of 1 130 093 patients, there was a statistically significant increase in mortality between grade II-III and grade IV-V injuries of 14.1% and 29.3%.64 Mortality rates for patients with isolated splenic injury are listed in Table 14-4. Thus, the primary determinant of mortality is related to significant vascular injury to the spleen. This vascular injury is predicted by 3 imaging findings: (1) evidence of active extravasation of contrast on venous phase imaging, (2) laceration involving the splenic hilum and causing significant infarction of the spleen, and (3) evidence of hemoperitoneum.79

Laceration: A laceration represents a focal linear defect in the surface of the spleen. The OIS grading system differentiates between small (3 cm) lacerations and whether the laceration

involves the trabecular vessels in the splenic hilum. Of these features, only vascular involvement has been shown to be a significant predictor of increased risk of mortality.64 In particular, vascular injuries that lead to infarction of greater than 25% of the spleen have an increased risk of mortality. Therefore, evaluation for the presence of associated infarction is an important feature that should be evaluated by imaging (see Figure 14-19). Lacerations of the spleen can be difficult to identify on unenhanced CT scans but can appear as linear or branching areas of low attenuation with sharply defined margins. Following contrast enhancement, the nonenhancing laceration will typically become more distinct from the enhancing splenic parenchyma (see Figure 14-20). With time, lacerations decrease in size and number, margins become less well defined, and the area becomes isodense to normal splenic parenchyma. On US examinations, a laceration will appear as a linear or branching hypoechoic defect in the splenic parenchyma.

Table 14-4. Mortality Rates for Splenic Injury by Gradea Imaging Notes 14-2. Imaging Features That Predict Need for Surgical or Angiographic Intervention in Splenic Trauma 1. Active extravasation of contrast 2. Splenic laceration causing infarction of >25% to 50% of the spleen 3. Shattered spleen 4. Hemorrhagic acites associated with splenic injury

Grade

Mortality with Other Injuries

Mortality Isolated Injury

I-II

9.9%

6.9%

III

10.4%

7.1%

IV

14.1%

9.4%

V

29.8%

22.7%

aData from Tinkoff G, Esposito TJ, Reed J, et al. American Association

for the Surgery of Trauma Organ Injury Scale I: Spleen, Liver, and Kidney, validation based on the National Trauma Data Bank. J Am Coll Surg 2008;207(5):646-655.

836

Diagnostic Abdominal Imaging

A

B

Figure 14-19 Splenic Laceration with Associated Infarction This 17-year-old woman was an unrestrained passenger in an automobile accident. A and B. Contrast-enhanced CT shows multiple linear and branching defects (arrow) in the spleen typical

of lacerations. These extend into the hilum of the spleen and cause infarction of the anterior portion (arrowheads) of the spleen in (B). Involvement of hilar vessels and infarction of greater than 25% of the spleen are both associated with increased mortality.

Intraparenchymal and subcapsular hematoma: Sple-

Figure 14-20 Splenic Laceration with Subcapsular Hematoma This 67-year-old woman was in an automobile accident. Contrast-enhanced CT shows multiple linear defects (small arrow) in the spleen characteristic of lacerations. There is also a crescentic hypoattenuating region surrounding the spleen (white arrowheads) typical of a subcapsular hematoma. Compare the attenuation of the subcapsular hematoma with the ascites surrounding the liver (black arrowheads). Note that the acute hematoma has a higher attenuation than the simple fluid surrounding the liver.

nic hematoma represents a focal collection of blood within the spleen. These are most often subcapsular in location but can less commonly be found within the spleen parenchyma. Although the OIS grading system distinguishes among small, medium, and large subcapsular hematomas and small and large intraparenchymal hematomas, the presence of a hematoma of any size or location without active extravasation of contrast will not alter the management of the patient with the exception of a completely shattered spleen (see Figure 14-21).64 On unenhanced CT, acute subcapsular hematomas are typically hyperdense compared with adjacent normal parenchyma (see Figure 14-20). After contrast administration, subcapsular hematomas are seen as low-attenuation collections between the splenic capsule and enhancing splenic parenchyma. The subcapsular hematoma will compress the underlying splenic parenchyma. With time, the attenuation of the hematoma will decrease and will appear hypoattenuating relative to unenhanced spleen. Intraparenchymal hematomas appear as irregular highor low-attenuation areas within the spleen parenchyma. In some individuals, a splenic hematoma will persist as a simple cyst for years later. How commonly this occurs is not known.

Chapter 14 Imaging of the Spleen 837

A

B

C

Figure 14-21 Shattered Spleen This 23-year-old man was involved in a motor vehicle accident. A-C. Contrast-enhanced CT demonstrates a fragmented spleen with disruption of a major vessel in the hilum (white arrow)

and active extravasation of contrast (black arrows). There is also extensive subcapsular and retroperitoneal hematoma surrounding the spleen. Active extravasation is a surgical emergency to prevent exsanguination.

Subcapsular hematomas on US examinations will typically appear as a crescentic hypoechoic structure with through transmission, adjacent to the surface of the spleen. In most cases, the hematoma will contain the lowlevel echoes typical of hematomas in any location (see Figure 14-20). Intraparenchymal hematomas will typically appear as complex irregularly shaped cystic lesions within the spleen parenchyma. Internal echoes due to blood products will usually be present.

Figure 14-21). The attenuation of the extravasated contrast material (85-350 HU) will be significantly higher than that of clotted blood (40-70 HU).80 On delayed imaging, the area of active extravasation remains high in attenuation and increases in size.

Vascular injuries: Posttraumatic splenic vascular injuries include active bleeding, acute pseudoaneurysms, and arteriovenous fistulae. These, especially active bleeding, are the most clinically significant findings associated with splenic trauma. The hypotension and/or the presence of active bleeding as demonstrated by contrastenhanced CT are the most common indications for surgical or angiographic intervention of patients with splenic injuries.17,64-69 Active bleeding: Active bleeding is the most significant imaging finding of splenic injury. This finding has been used as the primary feature identifying surgical or angiographic intervention of splenic injuries.17,65-69 It is identified by the extravasation of IV contrast during enhanced CT scanning. This is seen as a nodular, irregular, or linear area of contrast material extravasation into the splenic parenchyma or perisplenic tissues (see

Splenic pseudoaneurysm and arteriovenous fistula: These lesions have a similar appearance on CT examinations and can be differentiated only on splenic angiography. On enhanced CT examinations, both types of vascular lesions appear as well-circumscribed focal nodular areas of increased attenuation in the region of the splenic hilum. They are often surrounded by indistinctly marginated low-attenuation rims of inflammatory tissue (see Figure 14-22). A pseudoaneurysm is formed by an incomplete tear of the arterial wall. The defect in the intima and possibly media results in a weakening of the tensile strength of the vascular wall. The vessel then dilates in response to the arterial pressure, resulting in a pseudoaneurysm. The weakened wall is predisposed to rupture of the wall and subsequent high-volume extravasation of blood that can lead to hypotension, hypovolemia, and shock. Splenic arteriovenous fistulae develop as a result of injury to both the artery and the adjacent vein, resulting in a direct communication. On diagnostic angiography, arteriovenous fistulae can be differentiated from pseudoaneurysms by the characteristic early filling of veins.

838

Diagnostic Abdominal Imaging

Figure 14-22 Splenic Artery Pseudoaneurysm This 70-year-old woman was involved in a motor vehicle accident and had evidence of deep splenic lacerations on a CT scan (not shown). Selective digital subtraction angiogram of the splenic artery demonstrates a small focal nodular area of contrast (circle) typical of a posttraumatic pseudoaneurysm.

Infarction Splenic infarction is caused by occlusion of the splenic artery (global) or one of its branches (segmental). Splenicportal-mesenteric venous thrombosis may also cause splenic infarcts because of venous stasis and resultant ischemia. There are numerous causes of splenic infarcts, which are summarized in Table 14-5.81 Generally, however, the etiology varies with age. In older patients, an embolic event is the most frequent cause, whereas in patients younger

Table 14-5. Etiologies of Splenic Infarction 1. Hematological disorders Sickle cell disease Lymphoma Leukemia Myelofibrosis Gaucher disease 2. Thromboembolic disorders Embolism Atherosclerosis Pancreatic disease with vascular involvement Splenic artery aneurysm Vasculitis Hypercoagulable state 3. Mechanical disorders Splenic torsion Wandering spleen Portal hypertension

than age 40 years, the etiology is most often an underlying hematologic disorder.82 On US examinations, acute splenic infarcts classically appear as wedge-shaped, hypoechoic, and well-demarcated lesions, with absent flow in the infarcted area on color Doppler.83 This appearance is diagnostic of a splenic infarction. However, the presence of coexisting edema, bleeding, or necrosis can lead to different sonographic appearances, including round or irregular-shaped lesions that can be confused with other focal lesions of the spleen. As they age, infarcts become progressively hyperechoic due to fibrosis and scarring (see Figure 14-23). If large areas of the spleen become infracted, the volume decreases, resulting in a small irregularly-surfaced spleen. On noncontrast CT, infarcts can be poorly visualized because of the small attenuation difference between viable and infarcted spleen. Hemorrhagic infarcts can be more easily detected on unenhanced examinations because of the presence of scattered areas of increased attenuation representing acute blood.84 After contrast administration, infarcted areas become more distinct, usually appearing as peripheral, wedge-shaped, sharply marginated defects (see Figures 14-23 and 14-24). This classic appearance, however, is present in less than half of all acute splenic segmental infarcts.83 Many splenic infarctions will appear as round or irregular hypoattenuating regions within the spleen that can be difficult to differentiate from other splenic lesions such as tumors, hematomas, or abscesses. Global splenic infarction will show complete nonenhancement of the spleen with or without a “cortical rim sign.” The “cortical rim sign” consists of a thin layer of peripheral enhancement, representing residual capsular flow surrounding the nonenhancing splenic parenchyma (see Figure 14-25). In the chronic phase, infarcts may disappear completely or, more commonly, will appear as small, peripheral, linear, or wedge-shaped areas of low attenuation in the splenic parenchyma. In many cases, there will also be a focal contour deformity of the peripheral surface of the spleen because of retraction of the fibrous tissue in the splenic scar (see Figure 14-26). When large regions of the spleen have been infarcted, the overall volume of the spleen will decrease. In some instances, there can be calcifications from repeated infarctions. This is especially common in infarctions related to hemoglobinopathies. The end-stage appearance of a globally infarcted spleen is a very small and calcified lenticular or sickle-shaped structure beneath the left hemidiaphragm. This phenomenon is often called an “autosplenectomy.” Autosplenectomy is most commonly a manifestation of sickle cell disease (see Figure 14-27). On MRI, the signal intensity of the infarct depends on its age, the degree of hemorrhagic necrosis, and the amount of different blood products within the infarcted area.85 Recent hemorrhagic areas are increased in T1 signal intensity. Chronic infarcts are decreased in signal intensity on all pulse sequences. After gadolinium administration, most infarcts appear as wedge-shaped perfusion defects.85

Chapter 14 Imaging of the Spleen 839

A

C

B

Figure 14-23 Splenic Infarction Due to Emboli This 26-year-old man had a complex congenital heart disease. A and B. Contrast-enhanced CT of the spleen shows a focal wedge-shaped hypoattenuating region in the midspleen (black arrow) and a second larger infarction of the inferior aspect

of the spleen (white arrow). These infarcts were a result of thromboemboli from intracardiac thrombi because of slow flow in a dilated chamber. C. Transverse US 1 year later shows a focal wedge-shaped hyperechoic lesion (arrowheads) in the midbody of the spleen, characteristic of a chronic infarct.

A

B

Figure 14-24 Venous Infarction A and B. Enhanced CT images through the upper abdomen demonstrate a large mass in the tail of the pancreas (arrowheads) that causes occlusion of the splenic vein by direct invasion. The spleen is diffusely enlarged because of the venous obstruction

and has a small wedge-shaped area of hypodensity (arrow), characteristic of an infarction. In this case, the infarction is probably a result of the venous hypertension. The examination also shows multiple irregularly marginated hepatic lesions representing liver metastasis.

840 Diagnostic Abdominal Imaging 1 to 2% of all lymphomas, the majority of which will represent Hodgkin disease.53 Lymphoma of the spleen can have a variety of imaging appearances most commonly as splenomegaly. However, lymphoma can also appear as solitary or multiple round or oval lesions in the spleen. These lesions can have a miliary appearance or appear as larger nodular lesions (see Figure 14-28).

Metastasis

Figure 14-25 The Cortical Rim Sign of Infarction with Secondary Pyogenic Abscess This 54-year-old man recently underwent a distal pancreatectomy and now has persistent fevers. Enhanced CT shows the majority of the spleen to be mixed fluid attenuation representing a large region of infarction with only a small amount of residual normal splenic tissue (large arrow). Note the enhancing splenic capsule (arrowheads) surrounding the infarcted spleen, the “cortical rim sign.” There is also a small dot of gas (small arrow) indicating secondary bacterial abscess.

MULTIFOCAL SPLENIC LESIONS Multifocal splenic lesions can be due to malignant and benign tumors, hematogenous infections, sarcoidosis, and the rare disorder peliosis (Table 14-6).

Multifocal Neoplasms Multifocal splenic masses will most often be due to lymphoma or hematogenous metastasis to the spleen. However, rarely they can be due to multiple hemangiomas, lymphangiomas, or littoral angiomas.

Lymphoma The combination of Hodgkin disease and non-Hodgkin lymphoma is the most common malignancy affecting the spleen.52 Isolated or primary splenic involvement occurs in

The spleen is an infrequent site of tumor metastasis despite its vascularity. When splenic metastases are present, multiple other metastatic sites such as lymph nodes, liver, and lungs are often seen, indicating a worse prognosis.54 Common primary cancers that metastasize to the spleen include breast, lung, ovary, melanoma, stomach, pancreas, liver, and colon cancer. Calcification in splenic metastases is rare unless the primary tumor is a mucinous adenocarcinoma.85 Melanoma can cause cystic metastases. Metastatic ovarian cancer produces cystic implants along the peritoneal surfaces of the spleen. Splenic metastatic lesions may be solitary, multiple, or diffuse and vary in number and size from a few millimeters to several centimeters. On US, splenic metastases predominantly appear as small hypoechoic target lesions or may be well defined as either cystic or solid masses. A cystic or solid mass may show contrast enhancement at the periphery and within the septa of the lesion on CT and MRI (see Figure 14-29). Because metastatic disease often follows the MR signal of normal spleen, nonhemorrhagic splenic metastases may be difficult to see or the size may be underestimated on noncontrast T1- or T2-weighted MR sequences.86 Internal hemorrhage, necrosis, or contour deformity are often clues to the presence of splenic metastasis. In patients with transfusional iron overload, metastatic disease can be seen against a background of low T2 signal in surrounding abnormal spleen.

Multifocal vascular tumors Vascular tumors are the most common nonhematolymphoid tumors of the spleen.87 Several of these— hemangiomas, lymphangiomas, and littoral cell angiomas—can appear as multifocal masses within the spleen. Hemangiomatosis, multiple hemangiomas widely

Imaging Notes 14-3. Distinguishing Features of the Common Focal Lesions of the Spleen Hemangioma

Small, round, or oval

Incidentally discovered

Infarction

Wedge shaped

Hematoma

Features of blood products

History of trauma

Metastasis

Usually multiple

History of a malignancy

Lymphoma

Usually multiple

History of lymphoma

Chapter 14 Imaging of the Spleen 841

A

B

Figure 14-26 Focal Splenic Atrophy due to Prior Infarction or Trauma Unenhanced CT demonstrates a focal defect in the surface of the spleen (arrowheads) with associated dystrophic calcification.

This is typical of focal scarring and atrophy due to prior infarction or trauma.

involving the spleen, is usually a manifestation of a generalized angiomatosis such as Klippel-Trénaunay syndrome.87 Similarly, lymphangiomatosis, multiple lymphangiomas widely involving the spleen, is usually a manifestation of lymphangiomatosis, a syndrome where lymphangiomas involve multiple organs (see Figure 14-30). However, littoral cell angiomas are usually multiple in number and not associated with other syndromes.87

Littoral cell angioma: Littoral cell angioma is a rare tumor of the spleen that is derived from the littoral cells of the splenic red pulp sinuses and has features intermediate between those of endothelial cells and macrophages. These tumors can contain foci of extramedullary hematopoiesis, hemosiderin pigment, or calcification.87 To date, many cases have been found because of anemia, thrombocytopenia, or during evaluations for abdominal malignancies. It is possible that this is a result of meaningful association between these symptoms and the presence of the tumor, or

Table 14-6. Causes of Multiple Splenic Nodules A. Neoplasms 1. Lymphomaa 2. Metastasis 3. Hemangioma 4. Lymphangioma 5. Littoral cell angioma

Figure 14-27 Autosplenectomy in Sickle Cell Disease This 51-year-old woman with sickle cell disease presented with chest pain. Axial CT image through the upper abdomen shows a diminutive spleen (arrow) with multiple punctate calcifications. This calcified atrophy is typical of the autosplenectomy of sickle cell disease caused by multiple infarcts of the spleen.

B. Infections 1. Granulomatous infections a. Tuberculosis b. Histoplasmosis c. Cryptococcosis 2. Disseminated candidiasis 3. Pneumocystis jiroveci (HIV only) 4. Hepatosplenic cat scratch disease (Bartonella henselae) C. Sarcoidosis D. Peliosis aDisorders in bold are most common.

842

Diagnostic Abdominal Imaging

A

B

Figure 14-28 Multifocal Nodules Due to Lymphoma This 38-year-old man presented with nausea, vomiting, and a 15-pound weight loss. A and B. Axial contrast-enhanced images demonstrate multiple large nodular masses in the spleen and

innumerable small nodules in the liver. Images of the lower abdomen also demonstrated multiple retroperitoneal lymph nodes. Lymph node biopsy was diagnostic of a non-Hodgkin lymphoma.

it is possible that these tumors are accidentally discovered because of imaging related to these conditions. Like hemangiomas, the US appearance of littoral cell angiomas has been reported to range from hypoechoic to isoechoic to hyperechoic, depending on the size of the vascular channels created by the tumor, but they are probably most commonly hyperechoic.87 On unenhanced CT

examinations, littoral cell angiomas will typically appear as multiple hypoechoic masses within the spleen. With contrast enhancement, the tumors will initially enhance in an inhomogeneous fashion and on delayed contrast-enhanced images, littoral cell angiomas homogeneously enhance and become isoattenuating relative to the remaining splenic parenchyma.87 In most cases, both T1- and T2-weighted MRI sequences will have markedly low signal intensity. This reflects the presence of hemosiderin in the lesions, a finding that can be an important clue to the diagnosis (see Figure 14-31).87

Multifocal Infections Multifocal infections of the spleen are usually a result of hematogenous dissemination of organisms. Intravascular foci of infection such as bacterial endocarditis can result in multiple septic emboli to the spleen and have been discussed previously under the heading Cystic-Appearing Nodules and Masses of the Spleen (see Figure 14-12). However, granulomatous infections of the spleen and fungal abscesses typically appear as multiple small nodular lesions of the spleen and are discussed here. Rarely, infections with Pneumocystis jiroveci and hepatosplenic cat scratch disease can also produce multiple small splenic lesions.

Granulomatous infections of the spleen Figure 14-29 Splenic Metastasis from Breast Cancer This 79-year-old woman had a history of breast cancer. Enhanced CT image through the upper abdomen demonstrates many small hypoattenuating nodules within the spleen and liver. In this clinical setting, this is most likely to represent multiple metastases from the patient’s breast cancer.

Granulomatous infections, including tuberculosis, nontuberculous mycobacteria, and histoplasmosis are uncommon causes of splenic infection and have been reported to account for 0% to 5% of splenic abscesses.88-92 In the majority of cases, the splenic aspect of these infections will

Chapter 14 Imaging of the Spleen 843

A

B

Figure 14-30 Splenic Lymphangiomas in Lymphangiectasia This 32-year-old woman had congenital lymphangiectasia. A. Axial and (B) coronal contrast-enhanced CT images of the spleen demonstrate multiple small splenic lesions

with an attenuation similar to the right pleural effusion. Although unproven, these likely represent multiple splenic lymphangiomas.

be clinically silent and require no specific therapy other than general systemic antituberculous or antifungal medications when indicated. In one study of 57 patients with clinically diagnosed abdominal tuberculosis, splenic involvement was detected in 7 (12%).93 Most patients who have clinically detected splenic tuberculosis will have severe disseminated disease elsewhere and will often be immunocompromised.94,95 Focal splenic lesions have been detected in 30% of patients with tuberculosis and AIDS.96 In the active phase of disease, TB will most often appear as multiple small, few-millimeter to few-centimeter, lesions in the spleen that appear hypoechoic on US and hypoattenuating on CT examinations (see Figure 14-32).94,95,97,98 These are thought to represent small granulomas because of hematogenous dissemination of the infection.99 CT is more sensitive than US for the presence of these small lesions.95 In up to one-third of cases, tubercular infection will appear as a large unilocular fluid collection with or without rim enhancement and mimic the more common bacterial abscess.95,100 Tubercular abscesses may be more common in HIV-infected patients.95,101 In some cases, the spleen will appear enlarged with or without the presence of cystic lesions in the spleen. Once the infection is controlled, the small hypoechoic nodules will frequently leave multiple small splenic calcifications that persist for the lifetime of the patient.95 Histoplasmosis is the most common granulomatous fungal infection to involve the spleen. In most cases, the infection is clinically silent and so multiple small punctate calcified granulomas is the most common imaging manifestation of splenic histoplasmosis.98 Like tuberculosis,

during the acute infection, cross-sectional imaging will, in some cases, demonstrate multiple small nodular lesions that appear hypoechoic on US studies, hypoattenuating on CT scans, and hyperintense on T2 MR sequences.96,98 Splenomegaly, with or without the small nodular lesions, can also be seen in some patients.98,102 One study noted diffuse splenic hypoattenuation in 2 of 6 HIV-positive patients with histoplasmosis.102 This may be a rare but specific manifestation of histoplasmosis in this population. Splenic involvement with nontuberculous mycobacteria, Coccidioides and Cryptococcus, only occurs in HIVpositive individuals who remain severely immunosuppressed with diminished CD4 counts. Imaging manifestations mimic those of tuberculosis and include small nodular lesions that appear low attenuation on CT scans and splenomegaly.98 Small low attenuation nodular lesions have been detected in 7% and splenomegaly in 20% of patients with MAI and AIDS.96,103

Fungal microabscesses Fungal infection in the spleen is uncommon and accounts for between 1% and 25% of all splenic abscesses.28-31,34,96 Splenic fungal abscess most often occurs as a result of hematogenous dissemination in an immunocompromised patient.30,96,104 Patients at risk include those with hematologic malignancies, those that have received organ or bone marrow transplants, and those with cell-mediated immunodeficiencies, including AIDS.30,98,104 The typical patient presents with fevers, malaise, and weight loss and laboratory evaluations show neutropenia. The large majority of cases will be caused by Candida species; however,

844

Diagnostic Abdominal Imaging

A

B

C

D

E

F

Figure 14-31 Littoral Cell Angiomas of the Spleen in 2 Patients A-C. This 65-year-old woman had a renal cell carcinoma of the left kidney. (A) Unenhanced, (B) arterial-, and (C) venous-phase CT images demonstrate multiple irregularly marginated masses in the spleen. They demonstrate irregular predominantly peripheral enhancement on arterial-phase images and nearly completely fill in with contrast on the venous-phase images. These lesions remained stable over multiple years. This combination of findings indicates that these represent primary vascular tumors of the spleen, either multiple splenic

hemangiomas or littoral cell angiomas. D. T2-weighted, (E) T1-weighted, and (F) T1-weighted postgadolinium images in a 56-year-old man with prostate carcinoma demonstrate multiple irregularly shaped lesions distributed throughout the spleen. These are poorly seen on the T1-weighted sequence and filled in with contrast on delayed images. They have a remarkably similar appearance to those of first patient and likely represent multiple littoral cell angiomas of the spleen. These have remained stable on MRI examinations for several years.

occasionally Aspergillus or Cryptococcus organisms are discovered.96,105 Patients at risk include those that have received organ transplants or bone marrow transplants and patients with cell-mediated immunodeficiencies, including AIDS.30,98

Fungal infections of the spleen can only be seen on cross-sectional imaging studies. This hematologically disseminated infection causes multiple small splenic and hepatic abscesses that are sometimes imperceptible by imaging examinations.96,106 When detectable, US, CT,

Chapter 14 Imaging of the Spleen 845

A

B

Figure 14-32 Splenic Tuberculosis This 33-year-old African man was HIV-positive and febrile with cough and hemoptysis. Chest CT (not shown) demonstrated multiple low attenuation lymph nodes. A, B. Contrast enhanced

CT exam demonstrates multiple small hypoattenuating lesions of the spleen. Subsequent lymph node biopsy was diagnostic of tuberculosis.”

and/or MRI will typically demonstrate innumerable small, 5- to 10-mm, cysts widely distributed across the spleen.96,98,105,107,108 Occasionally, the cyst can be as large as 20 mm in diameter. In most cases, similar microabscesses will be seen in the liver. In many cases, these individuals will have a recent CT of the abdomen in which these small lesions were absent. The acute presentation of many tiny cysts in the liver and spleen of an immunocompromised patient is virtually diagnostic of disseminated fungal

infection, usually with Candida species (see Figure 14-33). Occasionally, US examinations will demonstrate a central focus of higher echogenicity, or a wheel-within-a-wheel pattern can be seen within the small cystic lesions.106

A

B

Figure 14-33 Candida Microabscesses This 22-year-old woman with AML was neutropenic and febrile. A. US of the spleen shows multiple tiny hypoechoic lesions scattered throughout the spleen. B. Enhanced CT image confirms the presence of multiple low-attenuation lesions.

Pneumocystis infection Extrapulmonary P jiroveci infection is a rare phenomenon that has primarily been detected in patients who are severely immunocompromised by HIV infection. With the

C Note the similar lesions in the left lobe of the liver. In this clinical setting, these findings are highly likely to represent multiple candidal abscesses. Blood cultures confirmed the presence of disseminated candidiasis.

846

Diagnostic Abdominal Imaging

advent of HAART therapy, fewer and fewer HIV-positive individuals remain immunocompromised and so splenic infection with Pneumocystis rarely occurs. Splenic involvement by Pneumocystis is typically asymptomatic and discovered incidentally on imaging examinations. In the acute phase of the infection, the spleen will be enlarged and contain multiple small nodular lesions that resemble fungal microabscesses on imaging examinations.96,109,110 With time, these lesions will become diffusely calcified and resemble multiple calcified granulomas within the spleen.111

Hepatosplenic cat scratch disease Cat scratch disease is an unusual lymphoreticular infection due to Bartonella henselae primarily seen in patients with T-cell immunodeficiency, such as AIDS.112 Imaging findings include splenomegaly with or without multiple nodular lesions that appear hypoechoic on US examinations and hypoattenuating on CT scans.112-114

Infiltrative Diseases Less commonly mulitiple splenic lesions can be due to the infiltrative diseases: sarcoidosis, Gaucher disease and peliosis.

Sarcoidosis Splenic involvement occurs in approximately 7% of patients with sarcoidosis and is usually asymptomatic.115,116  On imaging, there may be splenomegaly or the presence of multiple splenic nodules. Splenomegaly is the most common finding, occurring in approximately one-third of

patients with splenic sarcoidosis.117 Nodules occur more often when the spleen is enlarged. On US, splenic nodules are hypoechoic relative to background splenic parenchyma.118 They may also produce a diffuse heterogeneous pattern. Splenic nodules are visible on CT in approximately 6% to 33% of sarcoid patients.119 With contrast-enhanced CT, nodules appear hypodense relative to normal background spleen, are diffusely distributed through the spleen, and range in size from 1 mm to 2 cm in diameter.120 With increasing size, nodules tend to become confluent (see Figure 14-34). With MRI, splenic nodules associated with sarcoidosis are hypointense relative to background spleen on T1- and T2-weighted sequences.  Following IV gadolinium administration, nodules show no substantial enhancement.121

Gaucher disease Gaucher disease is an autosomal recessive lysosomal disorder in which lack of an enzyme results in accumulation of glucocerebrosides in the cells of the reticuloendothelial system, causing hepatosplenomegaly. On T1-weighted MR images, signal intensity is low relative to normal spleen secondary to glucocerebroside. On T2-weighted images, signal intensity is indeterminate except for nodal clusters of Gaucher cells, which appear hypointense to spleen (see Figure 14-35).122

Peliosis Splenic peliosis is a rare condition of unknown cause, characterized by multiple, variously sized, blood-filled cysts distributed throughout the spleen. The liver is more often involved than the spleen. Its clinical significance lies in the

A

B

Figure 14-34 Multifocal Nodules Due to Sarcoidosis This 44-year-old man had unexplained weight loss. A. Contrastenhanced CT demonstrates multiple low-attenuation nodules in the spleen and more subtle, smaller nodules in the liver. There

is also moderate splenomegaly. B. CT image of the lungs shows small nodules in a bronchovascular pattern. Lung biopsy was diagnostic of sarcoidosis.

Chapter 14 Imaging of the Spleen 847 potential of peliotic lesions on the splenic surface to rupture and cause intraperitoneal hemorrhage.87 Hepatosplenomegaly and multiple small hepatosplenic lesions are seen on all imaging modalities. On US, these lesions are hypoechoic. On unenhanced CT, lesions are hypodense but may contain fluid-fluid levels reflecting hematocrit effect.87 Following administration of IV contrast, lesions may not enhance if thrombosed or may demonstrate central enhancing foci if the thrombus becomes recanalized. On MRI, signal intensity within lesions varies depending on the stage of intralesional blood products.

SPLENOMEGALY AND OTHER DIFFUSE DISORDERS Diffuse abnormalities of the spleen are characterized by diffuse enlargement (splenomegaly) and diminished sized (atrophy, autosplenectomy). A

Splenomegaly

B

C Figure 14-35 MRI of Gaucher Disease This 58-year-old woman had a history of Gaucher disease. A. Coronal T1-weighted sequence demonstrates massive splenomegaly that fills the left upper quadrant of the abdomen and pelvis. B. Axial T1- and (C) T2-weighted sequences show splenomegaly and multiple small, low-signal lesions distributed throughout the spleen. These are nodal clusters of Gaucher cells.

There is no universally accepted definition of splenomegaly. Furthermore, several studies have shown that splenic size is directly correlated with the size of the individual and is inversely correlated with age.123-126 The most accurate means of identifying splenic size is to determine splenic volume by measuring the cross-sectional area of the spleen on individual CT or MRI slices, multiplying by the thickness of the slice and then summing the total number of slices. Using this method, 2 studies of 140 and 149 consecutive adult patients who underwent abdominal CT for indications not related to splenic disease identified the 95% confidence limit of the upper limit of normal splenic volume of 314.5 to 378 cm3.127,128 This method of determining splenic size is not generally practical, and fortunately studies have shown that measurement of the maximal length of the spleen is strongly correlated with splenic volume (r = 0.81 to 0.86).129,130 A study of 783 Chinese patients not known to have any condition likely to be associated with splenic enlargement recommended a length of 12 cm as the upper limits of normal.126 However, 2 studies of college athletes of 631 subjects and 129 subjects demonstrated mean splenic length of 10.65 ± 1.55 cm and 11.4 ± 1.7 cm.123 Therefore, 75% of normal young adult spleens will be less than 12.2 to 13.1 cm and 95% of normal young adult spleens will be less than 13.75 to 14.8 cm. This study corroborates

Imaging Notes 14-4. Splenomegaly Splenic length of 15 cm indicates splenomegaly Splenic length of 12-15 cm is a gray zone between normal and mild splenomegaly

848

Diagnostic Abdominal Imaging

the generally accepted criteria that a spleen with a maximal length of less than 12 cm is within the range of normal variation, that a spleen of greater than 15 cm maximal length is indicative of splenomegaly, and a spleen between 12 and 15 cm in length is within a gray zone between normal and mild splenomegaly. On physical examination and abdominal plain films, some large focal splenic lesions can be confused with splenomegaly. However, cross-sectional imaging can distinguish uniform splenic enlargement from enlargement of a focal portion of the spleen. We will reserve the term splenomegaly for uniform splenic enlargement without focal abnormality. Plain-film findings suggestive of splenomegaly are mass effect in the left upper quadrant of the abdomen, elevation of the left hemidiaphragm, medial and anterior displacement of the stomach, and inferomedial displacement of bowel (see Figure 14-36). In addition to length and volume measurements, cross-sectional imaging findings suggestive of splenomegaly include extension beyond the lower pole of the left kidney, extension medial to the aorta, and loss of inferomedial concavity. Splenomegaly can result from several pathophysiologies: (1) venous congestion, (2) infiltrative diseases, and (3) increased splenic function. Infiltrative diseases can be subdivided into neoplastic and nonneoplastic conditions. Increased splenic function can be due to infections, noninfectious inflammatory diseases (autoimmune disorders), removal of defective red cells (hemoglobinopathies), and extramedullary hematopoiesis (Table 14-7). Occasionally, the etiology for splenomegaly can be suggested based on other nonsplenic imaging findings such as liver cirrhosis, occlusion of the splenic vein, and retroperitoneal lymphadenopathy (usually indicating lymphoma), but often the cause of splenomegaly cannot be determined by imaging.

Table 14-7. Causes of Splenomegaly A. Splenic congestion 1. Cirrhosisa 2. Splenic vein obstruction a. Pancreatic carcinoma b. Pancreatitis 3. Portal or hepatic vein obstruction 4. Right heart failure B. Infiltrative diseases 1. Neoplasms a. Leukemia b. Lymphoma c. Malignant histiocytosis 2. Nonneoplastic a. Gaucher disease b. Niemann-Pick disease c. Alpha-mannosidosis d. Hurler syndrome e. Other mucopolysaccharidoses f. Amyloidosis g. Histiocytosis C. Increased splenic function 1. Infections a. Infectious mononucleosis b. Malaria c. Leishmaniasis (Kala Azar) d. Other infections 2. Noninfectious inflammatory conditions a. Rheumatoid arthritis b. Systemic lupus erythematosus c. Autoimmune hemolytic anemia d. Sarcoidosis 3. Hematologic disorders a. Thalassemia b. Hereditary spherocytosis c. Early sickle cell disease d. Autoimmune hemolytic anemia e. Other causes of anemia f. Polycythemia vera 4. Extramedullary hematopoiesis a. Myelofibrosis b. Marrow infiltration by neoplasms aDisorders in bold are most common.

Venous congestion

Figure 14-36 Splenomegaly due to Portal Hypertension This 77-year-old man with a history of cirrhosis had abdominal pain. Abdominal plain film demonstrates a small liver shadow (black arrowheads) and an enlarged spleen (white arrowheads), which measured 18 cm in craniocaudad length. These findings are typical of cirrhosis and splenomegaly due to portal hypertension.

Occlusion or hypertension of the splenic vein leads to edema of the spleen and splenomegaly. In many cases, cross-sectional imaging will also reveal secondary findings of portal hypertension such as perisplenic varices and dilation of the splenic vein.

Cirrhosis: Cirrhosis with portal hypertension is the most common cause of splenomegaly in the United States. Although the splenic enlargement is nonspecific,

Chapter 14 Imaging of the Spleen 849

A

B

Figure 14-37 Splenomegaly due to Cirrhosis This 22-year-old woman had cirrhosis due to autoimmune hepatitis. A and B. The liver is small and nodular in appearance, typical of cirrhosis. There are multiple abdominal wall

varices due to a recanalized umbilical vein indicating portal hypertension. Contrast-enhanced axial CT images demonstrate an enlarged spleen that extends into the lower abdomen and displaces the left kidney.

imaging will usually demonstrate findings of cirrhosis, including hepatic nodularity and relative caudate and left hepatic lobe hypertrophy (see Figure 14-37). On MRI, splenic foci of hemosiderin deposition are seen in 9% to 12% of patients with portal hypertension. These foci are called Gamna-Gandy bodies and are due to small areas of intrasplenic hemorrhage. They appear as multiple tiny foci of decreased signal intensity with all pulse sequences, and exhibit “blooming” on gradient echo sequences secondary to iron deposition.82 Gamna-Gandy bodies are discussed in greater detail under the heading: Many Small Splenic Calcifications or Hemosiderin Foci.

Lymphoma: Lymphoma is the most common malignancy

Venous obstruction: Obstruction of the splenic, portal, or hepatic vein can also result in splenomegaly. Causes of portal and hepatic vein obstruction are discussed in Chapter 3. Splenomegaly due to splenic vein obstruction is uncommon and can be due to surrounding inflammation, compression, or invasion by an adjacent mass or by in situ thrombosis of the splenic vein. The most common causes are acute and chronic pancreatitis and pancreatic cancer (see Figure 14-24).131,132 However, rarely mass effect or invasion from a renal cyst or cancer, direct injury from trauma, or thrombosis from blood dyscrasia can cause splenic vein occlusion.

Infiltrative malignancies Many forms of lymphoma and leukemia can cause splenomegaly. Malignant histocytosis is another rare infiltrative malignancy of the spleen.

affecting the spleen, most often, secondary involvement by disseminated disease but occasionally as the primary focus.52 Splenic involvement is common in both Hodgkin disease and non-Hodgkin lymphoma. Although splenic involvement in disseminated Hodgkin disease and non-Hodgkin lymphoma is common, isolated or primary splenic involvement is rare, occurring in approximately 1% to 2% of all lymphomas at presentation.53 The vast majority of primary splenic lymphomas are non-Hodgkin type. Patients with AIDS have an increased risk of splenic involvement in lymphoma, in addition to a more aggressive course of disease. The imaging features of splenic lymphoma include (1) a normal-appearing spleen in which the involvement is only microscopic, (2) splenomegaly in which the spleen is diffusely infiltrated without focal mass lesion, and (3) focal lesions of the spleen in which there are solitary or multiple discrete regions of lymphoma within the splenic parenchyma (see Figure 14-38). Focal and multifocal lymphomatous lesions will typically appear as round or oval nodules or masses within the normal spleen (see Figure 14-16). Multifocal involvement can be further subdivided into miliary/micronodular configuration, with nodules less than 1 cm in diameter or large masses between 1 and 10 cm (see Figure 14-28).53 Focal involvement can occur in both normal and enlarged spleens. Splenic size alone is an unreliable predictor of lymphomatous involvement. Although massive splenomegaly, in the setting of a patient with known lymphoma, is invariably indicative of splenic involvement by lymphoma, mild-moderate splenomegaly without lymphomatous

850 Diagnostic Abdominal Imaging

A

B

Figure 14-38 Splenomegaly due to Lymphoma This 65-year-old man had a low-grade B-cell lymphoma. A and B. Coronal reconstruction from an enhanced CT demonstrates

massive enlargement of the spleen, the most common splenic manifestation of lymphoma.

infiltration occurs in approximately 30% of patients with Hodgkin lymphoma and 70% of patients with nonHodgkin lymphoma.133 Accepting these limitations, it has been shown that alterations in serial splenic volume on CT correlate with changes in disease. Consequently, alterations in serial CT measurements may be used to assess treatment response or disease progression.134 CT is the most widely used imaging modality for lymphoma assessment. However, it has a low accuracy for detecting splenic involvement. Quoted sensitivities vary between 22% and 65%, but have increased with advances in multidetector imaging.135 This low sensitivity is primarily due to poor lesion-to-spleen contrast for lymphomatous deposits on CT. Consequently, even gross infiltrates that are clearly evident on US can be missed on CT. When focal splenic involvement is detected, deposits are of lower attenuation than adjacent normal splenic parenchyma on unenhanced CT and demonstrate little or no enhancement on enhanced images. Deposits may be markedly hypodense, of near water density, which can represent areas of liquefactive necrosis or hypovascular solid tumor. Rim enhancement has also been documented.136 US has limited accuracy in detecting splenic involvement in lymphoma. Sensitivities vary widely but have also improved over time, likely reflecting technical advances. US is more sensitive than CT for detecting splenic involvement, approaching 63% as compared to 37% for

CT in 1 study, with US preferentially demonstrating inhomogeneities as well as small nodular infiltrates.137,138 On US, the majority of detectable foci are hypoechoic relative to normal splenic tissue and have scattered and penetrating vascularity on color Doppler. Focal deposits may have smooth or indistinct margins.  Solid lymphoma deposits can appear deceptively cystic on B-mode US, although color Doppler can be used to confirm internal vascularity. Hypoechoic splenic deposits may become isoechoic as they regress with treatment or become hyperechoic secondary to fibrosis. Conventional T1- and T2-weighted MR sequences have poor sensitivity for splenic involvement in lymphoma, because of similar relaxation times of lymphoma and normal splenic tissue using these sequences.16 Identifiable splenic deposits are hypointense on T1-weighted images and hyperintense on T2-weighted images. Improvements in MRI have improved lesion detection. In 1 study using gradient echo sequences resulting in predominantly T2* and proton density-weighted images, MRI demonstrated splenic lymphoma with greater lesion-to-spleen contrast than US or CT.139 Gadolinium-enhanced MRI further increases conspicuity of lymphoma deposits as a consequence of their relatively poor enhancement compared to normal splenic parenchyma, although this is dependent on the timing of acquisition. Studies using dynamic-enhanced MR suggest that the optimal timing to demonstrate splenic

Chapter 14 Imaging of the Spleen 851 lesions is between 20 seconds and 1 minute from commencement of bolus injection.5 In general, CT, US, and MRI have similar limited specificity, with difficulty differentiating lymphomatous involvement from leukemic infiltrates, sarcoid deposits, healed infarcts, and secondary infection. Although currently limited in its availability, FDGPET imaging has established a definite role in imaging of patients with lymphoma at all stages in management. It is the radionuclide of choice for the evaluation of lymphoma. In terms of splenic assessment, benign and malignant splenic pathologies may be separated in patients with or without known malignancy based on their differing metabolic activities, which can be described quantitatively by standardized uptake values. Recent reports have documented greater accuracy of PET compared to CT for detecting splenic involvement in lymphoma.140 Inherent limitations of PET include false positives produced by other processes with high metabolic activity such as infection and false negatives that are a recognized feature of indolent, low-grade lymphomas.141

Leukemia: Both acute and chronic forms of leukemia are common causes of splenomegaly. They typically result in uniform enlargement of the spleen without focal defect. Some leukemias, specifically, chronic lymphocytic, acute myelocytic, and hairy cell leukemia are predisposed to massive splenomegaly (Table 14-8). Massive splenomegaly is traditionally defined as a weight greater than 1000 g (pathologic definition) or a spleen palpable more than 8 cm below the costal margin (clinical definition). The imaging criterion for massive splenomegaly is typically a maximal length of greater than 20 cm.

Malignant histiocytosis: Malignant histiocytosis is a rare neoplastic transformation of macrophages characterized by a syndrome of systemic symptoms such as fever wasting and malaise, pancytopenia, adenopathy, and

Table 14-8. Causes of Massive Splenomegaly 1. Leukemia (CLL, CML, hairy cell) 2. Lymphoma 3. Myelofibrosis

hepatosplenomegaly.142 Splenomegaly without systemic symptoms has been rarely reported.143

Nonneoplastic infiltrative diseases There are a variety of rare metabolic disorders that can cause infiltration of the spleen and resultant splenomegaly. These include Gaucher disease, Niemann-Pick disease, α-mannosidosis, Hurler syndrome and other mucopolysaccharidoses, amyloidosis, and Tangier disease. Discussion of these entities is beyond the scope of this text. We have briefly discussed Gaucher disease under the heading MULTIFOCAL SPLENIC LESIONS because it can sometimes be differentiated from the other etiologies of splenomegaly because of multiple, small low-signal lesions seen on MRI examinations (see Figure 14-35).

Infectious causes of splenomegaly Increased functioning of the spleen as a response to systemic infection can result in splenic enlargement. Infections that have been reported to cause splenomegaly include Epstein-Barr virus (mononucleosis), cytomegalovirus, HIV, viral hepatitis, malaria, typhoid fever, brucellosis, leptospirosis, tuberculosis, histoplasmosis, leishmaniasis, trypanosomiasis, Bartonella henselae (hepatosplenic cat scratch disease), and ehrlichiosis (see Figure 14-39). The most common of these are Epstein-Barr virus and malaria and will be discussed briefly.

Infectious mononucleosis: Infectious mononucleosis is a common clinical syndrome of pharyngitis, fever, and lymphadenopathy that can occur at any age; it is most frequently encountered in adolescents and young adults.144,145 Infectious mononucleosis is most often due to primary infection with Epstein-Barr virus but is occasionally a result of primary infection with cytomegalovirus. These organisms are transmitted through bodily secretions, primarily saliva in the case of Epstein-Barr virus and primarily via semen or vaginal secretions in the case of cytomegalovirus (CMV). Transmission can be subclinical in many cases, but can also cause the syndrome of infectious mononucleosis in other individuals. Epstein-Barr virus remains latent in human lymphocytes and CMV remains latent in human bone marrow–derived myeloid progenitors. The infection remains suppressed by the activity of human T cells and, therefore, T-cell immunosuppression, as seen in AIDS and organ transplantation, can result in reactivation of

4. Sarcoidosis 5. Malaria 6. Thalassemia 7. Polycythemia vera 8. Autoimmune hemolytic anemia 9. Schistosomiasis 10. Visceral leishmaniasis (Kala Azar)

Imaging Notes 14-5. Causes of Splenomegaly In the industrialized nations, the most common causes of splenomegaly are cirrhosis, leukemia, and lymphoma followed by sarcoidosis and myelofibrosis. In developing countries, the most common causes of splenomegaly are malaria and anemia144,150

852 Diagnostic Abdominal Imaging

A

B

Figure 14-39 Splenomegaly and Rupture in Cat Scratch Disease This 37-year-old man was HIV-positive and had bilateral lower quadrant pain and a falling serum hematocrit. A and B. Contrastenhanced CT images demonstrate moderate splenomegaly with multiple splenic fractures (small black arrowheads), acute

contrast extravasation (black arrow), and a large subcapsular hematoma (white arrows). Patients with splenomegaly have an increased incidence of rupture probably due to minor trauma. The surgical specimen was diagnostic of bacillary angiomatosis or cat scratch disease.

the viruses and recurrence of a mono-like illness. Palpable splenomegaly is detected on physical examination in approximately half of patients with the clinical syndrome.145 This disorder is usually diagnosed by a combination of clinical history, physical examination, and serologic testing for the presence of EBV and CMV infection. As a consequence, imaging is only rarely performed.

Noninfectious inflammatory conditions causing splenomegaly

Malaria: Malaria is among the most common infectious diseases in tropical and subtropical regions of the world, including Asia, Africa, and the Americas. It is caused by the protozoa of the genus Plasmodium of which 5 species, falciparum, vivax, ovale, malariae, and knowlesi, can cause human illness.146 There are approximately 350 to 500 million cases of malaria annually, causing the death of 1 to 3 million people, the majority of whom are young children in subSaharan Africa.147,148 Mosquito bites act as the vector from an infected individual to a new uninfected host. After a short incubation period in the liver, the parasite infects circulating red blood cells. Symptoms typically include fever and headache but can also include anemia, arthralgias, vomiting, anemia, and coma. In the developing world, malaria is among the most common causes of splenomegaly and accounted for 25% of cases in a study of 1400 patients in Pakistan and for 53% of cases of children in the Ivory Coast.149,150 In most cases, this is a direct result of the infection and the clearance of infected red blood cells. However, tropical splenomegaly syndrome, a rare complication of recurrent malarial infection thought to be secondary to an abnormal immunologic response to repeated infection, can also be a cause for splenomegaly.151

The autoimmune disorders rheumatoid arthritis, systemic lupus erythematosus, and autoimmune hemolytic anemia are all associated with splenomegaly as is the idiopathic granulomatous condition sarcoidosis. In these conditions, the splenomegaly is thought to be a result of increased lymphoreticular splenic function and subsequent hypertrophy of these splenic elements.

Sarcoidosis: Asymptomatic splenic involvement occurs in approximately 7% of patients with sarcoidosis.115,116 On imaging, there may be splenomegaly or the presence of multiple splenic nodules. Splenomegaly is the most common finding, occurring in approximately one-third of patients, but is a nonspecific finding (see Figure 14-40).117

Hemoglobinopathies and splenomegaly As a response to excessive hemolysis of red blood cells, the spleen can hypertrophy in order to process the extra damaged red cells. As a consequence, hemoglobinopathies, including thalassemia, hereditary spherocytosis, sickle cell disease, autoimmune hemolytic anemia, and some other causes of anemia can be a cause of splenomegaly. In patients with homozygous sickle cell disease, splenomegaly is seen only early in childhood because of the propensity for autoinfarction of the spleen in patients with sickle cell disease. However, patients with heterozygous sickle cell disease will commonly have splenomegaly.152

Chapter 14 Imaging of the Spleen 853

A

Figure 14-40 Splenomegaly due to Sarcoidosis This 38-year-old man with a history of sarcoidosis had abdominal pain. Abdominal x-ray reveals a massively enlarged spleen (arrowheads) due to sarcoidosis.

B

Extramedullary hematopoiesis and splenomegaly Disorders that lead to the destruction of the normal functioning of the bone marrow can result in extramedullary hematopoiesis in the spleen and other sites around the body. This activity in the spleen can be a cause for splenomegaly. Conditions causing this phenomenon include myelofibrosis, marrow infiltration by leukemia and lymphoma and marrow destruction by irradiation, medications, and other toxins.

Myelofibrosis: Myelofibrosis is a disorder of the bone marrow, in which the marrow is replaced by fibrous (scar) tissue resulting in anemia. Symptoms include fatigue, bruising, easy bleeding, bone pain, and increased susceptibility to infections. The cause of myelofibrosis is unknown and there are no known risk factors. Myelofibrosis typically develops after the age of 50 but can occur at any age. Splenomegaly due to extramedullary hematopoiesis is a characteristic finding on physical examination and in imaging of patients with myelofibrosis (see Figure 14-41).

C Figure 14-41 Splenomegaly in Myelofibrosis This 50-year-old man had a history of myelofibrosis. A and B. Contrast-enhanced axial CT shows a spleen that has a rounded bulbous appearance, extends inferior to the liver tip, and displaces the left kidney medially. These are all transaxial imaging features indicating splenomegaly, in this case as a result of the myelofibrosis. C. US of the spleen also shows the bulbous nature of the spleen and confirms the increased length of nearly 22 cm.

854 Diagnostic Abdominal Imaging

Splenic Atrophy: Autosplenectomy

Unique Anomalies of the Spleen

The term autosplenectomy is used to denote the atrophy of the spleen secondary to spontaneous widespread infarction, leading to loss of splenic function. In the vast majority of cases, this is a complication of homozygous sickle cell disease but has rarely been reported as a complication of pneumococcal septicemia and SLE.153,154 The spleen is prone to infarction in sickle cell disease because the slow flow and relative hypoxemia of the spleen results in sickling of red cells leading to microvascular occlusion and infarction. Complete infarction of the spleen has typically occurred by the age of 8.155 In most cases, patients are asymptomatic but are predisposed to bacterial infections with encapsulated organisms because of the loss of splenic function. Repeated episodes of infarction result in atrophy of the spleen. Chronically, the spleen can become diffusely calcified. Imaging examinations demonstrate diminished size of the spleen in all cases and will often demonstrate diffuse calcifications as increased opacity on plain films, increased attenuation on CT, and shadowing on US.152,156,157 Calcification can diffusely involve the spleen, be seen as multiple punctate foci or as curvilinear calcification (see Figures 14-27 and 14-42). Calcifications are common, demonstrated on plain films in 31% of 182 subjects, with an increasing prevalence with increasing age of the subject.156 On MRI examinations, the spleens of patients with sickle cell disease are typically decreased in signal intensity on both T1- and T2-weighted sequences because of the iron deposition in the spleen.158

There are several congenital anomalies of the spleen that can result in an abnormal appearance, location, or number of splenic tissue.

Wandering spleen Wandering or ectopic spleen refers to migration of the spleen from its normal site in the left upper quadrant to a more caudal location in the abdomen as a result of laxity or maldevelopment of the supporting splenic ligaments.159 It is a rare entity, found incidentally in less than 0.2% of patients undergoing splenectomy.160 The long pedicle renders the spleen hypermobile, predisposing it to torsion. Patients can be asymptomatic; present with a mobile mass in the abdomen; or present with acute, chronic, or intermittent abdominal pain due to torsion of the wandering spleen.161 The most characteristic imaging finding is absence of the spleen in its normal position and a soft-tissue mass resembling the spleen located somewhere else in the abdomen or pelvis.159 The most common location of the spleen is in the left midabdomen. The whirl sign of the splenic pedicle, representing the twisted splenic vessels and surrounding fat, is reported to be specific for splenic torsion.160 Twisting of the tail of the pancreas along with torsion of the splenic pedicle has been reported in cases with acute splenic torsion and can cause clinical signs of acute pancreatitis.162,163 An additional finding in acute torsion is total or partial splenic infarction, the imaging findings of which are described in detail earlier in this chapter.

Heterotaxy syndromes Heterotaxy is defined as the abnormal placement of organs. There are 2 defined heterotaxy syndromes, polysplenia and asplenia, that are due to failure to establish the normal leftright patterning during embryonic development. Although abnormalities of the spleen are some of the most recognizable features of these syndromes, clinical outcomes of these patients are primarily related to the severity of congenital heart disease.

Polysplenia syndrome: Polysplenia syndrome is a com-

Figure 14-42 Sickle Cell Autosplenectomy This 21-year-old woman with sickle cell disease presented with fever and jaundice. The spleen (arrow) is diminutive and minimally increased in attenuation. This splenic atrophy is a result of repeated splenic infarction.

plex congenital anomaly characterized by partial visceral heterotaxia (situs ambiguous) and concomitant levo isomerism (bilateral left-sidedness). It is usually diagnosed in childhood because of the various and often severe cardiac anomalies that are a part of the syndrome.164 Most patients with polysplenia syndrome die by the age of 5 years due to severe cardiac anomalies.165 Five percent to 10% of patients with polysplenia syndrome have a normal heart or only minor cardiac defects and reach adulthood without symptoms, and the syndrome may then be incidentally discovered by US, CT, or MRI of the abdomen.164 As the name of the syndrome implies, multiple discrete spleens are considered the hallmark of the syndrome.

Chapter 14 Imaging of the Spleen 855

A

B

C

D

Figure 14-43 Polysplenia This 52-year-old woman had congenital heart disease. A-D. Axial T1-weighted contrast-enhanced images show a left-sided heart (large arrow), right-sided stomach (small arrow), and multiple

splenules (small arrowheads) and a large midline liver. Features are characteristic of polysplenia. Note the dilated hemiazygous vein (large arrowhead in A) indicating interruption of the IVC and azygous continuation.

However, there is a wide variation ranging from many very small spleens to a multilobular spleen with tiny accessory spleens and even an undivided spleen.164 The spleens may be left-sided or right-sided and are always on the same side as the stomach and almost always along the greater curvature (see Figure 14-43).166 The liver is most often located in the midline extending symmetrically to both sides of the upper abdomen. The gallbladder is usually in a central location, and anomalies may also affect the branching pattern of the biliary tree. A shortened pancreas, consisting only of the pancreatic head, may be associated with polysplenia syndrome.167 Mirror image location of the bowel and mesenteric vessels is frequently seen in polysplenia.168 The most common venous anomaly associated with polysplenia syndrome is interruption of the inferior vena cava with azygos or hemiazygos continuation, occurring in 65% to 80% of individuals with polysplenia.169 Caudal to

the caval interruption, the inferior vena cava may lie to the right or the left of the midline, or may be duplicated. The hepatic segment of the inferior vena cava is often absent, and the hepatic veins drain directly into the right atrium.170

Asplenia syndrome: Asplenia is generally characterized by an abnormal arrangement of the abdominal organs and absence of the spleen. As in polysplenia, there is partial visceral heterotaxia (situs ambiguous) but in this case there is generally dextro isomerism (bilateral right-sidedness). Congenital heart disease complicates this anomaly in up to 99% to 100% of patients, accounting for a very high mortality rate of up to 95% of patients in the first year of life.168  As the name implies, the spleen is absent or rudimentary. The liver and gallbladder are often midline in location.169 The pancreas can be shortened, consisting only of the pancreatic head.168 Venous anomalies can include bilateral inferior vena cava (see Figure 14-44).

856 Diagnostic Abdominal Imaging Table 14-9. Causes of Multiple Calcified Splenic Nodules 1. Granulomatous infection a. Histoplasmosisa b. Tuberculosis 2. Other infections a. Pneumocystis jiroveci b. Schistosomiasis 3. Amyloidosis 4. Sarcoidosis 5. Anthrasilicosis 6. Gamna-Gandy bodies (cirrhosis) aDisorders in bold are most common.

Figure 14-44 Asplenia This 50-year-old woman was being evaluated for a neck mass. Contrast-enhanced CT shows a large midline liver with the stomach (arrow) in the right upper quadrant and absence of all splenic tissue. The oval structure in the left upper quadrant is the top of the left kidney (arrowhead). This patient had situs inversus of the thorax but had no congenital heart disease and is one of a tiny minority of patients with tiny minority of patients with asplenia to survive to adulthood.

under the heading: Splenic Atrophy: Autosplenectomy. Small punctate calcifications are most often due to granulomatous infections but can be a result of a variety of other disorders (Table 14-9).

Splenic granulomas: One or many small spherical cal-

Calcifications will appear as increased opacity on plain films, increased attenuation on CT scans, and as echogenic structures on US examinations. Diffuse calcification is usually seen with splenic atrophy and is a manifestation of sickle cell disease and has been discussed previously

cifications in the spleen will most often be a manifestation of granulomatous disease, most often tuberculosis or histoplasmosis.171,172 These are a common incidental finding on imaging examinations and have no clinical significance other than as a marker for previous granulomatous infection (see Figure 14-45). The presence of more than a few calcified nodules will usually indicate histoplasmosis as the cause, whereas smaller numbers of granulomas will often be due to tuberculosis.173,174 In areas where histoplasmosis is endemic, the majority of splenic

A

B

Splenic calcifications

Figure 14-45 Calcified Granulomas of the Spleen A. Abdominal x-ray and (B) CT scan both demonstrate multiple oval calcifications in the spleen, typical of splenic granulomas probably secondary to histoplasmosis.

Chapter 14 Imaging of the Spleen 857 calcifications will be secondary to histoplasmosis. In regions of the world where histoplasmosis is uncommon, tuberculosis will be the most common cause of small focal splenic calcifications.

Rare causes of multiple small splenic calcifications: Rarely primary and secondary amyloidosis, sarcoidosis, anthrasilicosis, Pneumocystis jiroveci infection, and schistosomiasis can result in multiple small calcifications in the spleen and liver.111,175-178 Gamna-Gandy bodies due to liver cirrhosis will occasionally calcify and appear as multiple punctate calcifications in the spleen on CT and US examinations; however, their primary manifestation is signal loss on MRI and will be discussed in detail under the subsequent heading: Iron deposition.

Iron deposition in the spleen Hemosiderin deposits in the spleen in most cases do not cause abnormalities detectable by plain films, CT, or US. However, the paramagnetic properties of hemosiderin result in T1 and T2 shortening and cause loss of signal of the spleen on both T1- and T2-weighted images. In hemosiderosis, this results in diffuse signal loss of the spleen; however, in the case of cirrhosis, small petechial hemorrhages called Gamna-Gandy bodies result in characteristic multiple small foci of signal loss.

Hemosiderosis: Iron overload can be due to 1 of 2 underlying causes: (1) multiple transfusions or (2) increased gut resorption of iron.179 Excess iron from blood transfusions is deposited in the reticuloendothelial system and in the parenchymal cells of the liver, spleen, and other organs, including the heart and lymph nodes and is known as hemosiderosis.180 Hemosiderosis can be distinguished from hemochromatosis, the other primary cause of iron overload, by the presence of iron overload in the spleen. Iron deposition in hemochromatosis will spare the spleen and bone marrow but involve the liver and pancreas, whereas hemosiderosis will involve all 4 organs. The amount of iron overload can be quantified by MRI examinations when necessary (see Figure 14-46).181

6% of patients with chronic liver disease.182 The ferritin in hemosiderin causes susceptibility artifact, leading to loss of signal in the surrounding tissues. This appears as multiple small low-signal foci within the spleen. Gradient echo sequences are more sensitive to susceptibility artifact and therefore the lesions will appear larger on gradient echo sequences.182 Further, US and CT examinations can detect Gamna-Gandy bodies through the presence of small dystrophic calcifications that appear as small echogenic foci on US examinations and as punctate hyperattenuating dots on CT examinations.182-185 The small areas of fibrosis can also appear as tiny hypoattenuating nodules on contrast-enhanced CT (see Figure 14-47).

POSTOPERATIVE FINDINGS RELATED TO SPLENIC SURGERY There are various normal and abnormal imaging findings related to splenectomy that can be detected by abdominal imaging, including long-term regeneration of small residual amounts of splenic tissue, called splenosis.

Normal Postoperative Findings Accumulation of a small amount of sterile, reactive peritoneal effusion of water density is a common normal finding in the immediate postoperative period. Most of these collections are not encapsulated, are confined to the surgical bed, and resolve on follow-up studies. A small amount of left-sided pleural effusion and posterior displacement of the stomach into the subphrenic space is also expected.186 Pneumoperitoneum is another common finding following recent intra-abdominal surgery, which usually disappears within days. A persistent or increasing amount of free air in the peritoneum beyond the first postoperative week in the absence of an abdominal drain or dehiscent incision is suggestive of perforation of the gastrointestinal tract.187

Gamna-gandy bodies: Gamna-Gandy bodies, also called

Splenosis

siderotic nodules, are small fibrotic nodules that occur within the spleen as a result of small petechial hemorrhages. These will contain hemosiderin, the chronic byproduct of hemorrhage and can also contain small areas of dystrophic calcification.182-185 These small lesions are typically a result of portal hypertension but can rarely be due to hemolytic anemia and leukemia. Gamna-Gandy bodies can be detected on crosssectional imaging because of the imaging characteristics of hemosiderin and calcium. In general, MRI examinations are the most sensitive study for the detection of Gamna-Gandy bodies which can be seen in 10% to 15% of patients with portal hypertension and approximately

Splenosis is defined as autotransplantation of splenic tissue in various abnormal sites after splenic injury that can be either traumatic or iatrogenic during surgery. Iatrogenic splenosis occurs more often during laparoscopic splenectomy. Abdominal splenosis is the most common form, and although it usually has no clinical significance, may cause relapse of hematologic disorders.188 Thoracic splenosis is less common, usually occurring following combined diaphragmatic and splenic injury, leading to the formation of left-sided, pleural-based pulmonary nodules or masses.189 Residual splenic function following splenectomy may occur due to an unidentified accessory spleen in

858 Diagnostic Abdominal Imaging

A

B

C

D

Figure 14-46 Hemosiderosis of the Spleen This 33-year-old woman had undergone stem cell transplant for Hodgkin lymphoma. A. T1-weighted and (B) T2-weighted MRI sequences show the spleen to be diffusely low signal. The liver is also lower signal than normal, although to a lesser degree than the spleen. This is the typical finding of hemosiderosis.

C. Unenhanced CT shows the spleen to be slightly higher attenuation than the liver, also indicating the greater deposition of hemosiderin in the spleen. D. The difference in attenuation between liver and spleen is masked by the presence of IV contrast in this contrast-enhanced image.

the abdominal cavity that was left in situ or from aforementioned splenosis auto-transplants left behind (see Figure 14-48). Such residual functioning splenic tissue helps preserve the host defense mechanism. However, when splenectomy was performed for the management of hematologic disease, residual functioning splenic tissue may induce relapse and is considered failure of surgical management. The diagnosis of splenosis or residual splenic tissue can be confirmed using nuclear scintigraphy with radiolabeled heat-damaged red blood cells or 99mTc-sulfur colloid, which detects functioning splenic tissue foci as small as 2 cm or smaller when using SPECT.190 Further localization of functioning tissue can then be demonstrated with CT prior to repeat surgery.

Complications of Splenic Surgery Complications of splenic surgery include intra-abdominal abscess, portal or splenic vein thrombosis, and injury to adjacent structures.

Intra-abdominal abscess Intra-abdominal abscess remains the main cause of morbidity in the postsplenectomy patient. CT is highly accurate in diagnosing postoperative abscess and is the preferred imaging modality in assessing its presence, location, and size. The CT features of an abscess include a wellcircumscribed fluid collection, occasionally with peripheral enhancing rim, which may contain gas bubbles or an airfluid level. The abscess may, however, appear as a mass of

Chapter 14 Imaging of the Spleen 859

A

B

C

D

E

F

Figure 14-47 Gamna-Gandy Bodies This 55-year-old woman had alcoholic cirrhosis. A. T1-weighted and (B) T1*- and T2-weighted MRI sequences show innumerable small low-signal lesions throughout the spleen. Note how the lesions appear larger on the T1* sequence. This is a susceptibility artifact and indicates the presence of a paramagnetic substance, most often hemosiderin. C. The low-attenuation regions are much less apparent on this T2-weighted sequence. D. US image

shows innumerable small echogenic foci in the spleen, most likely due to many small splenic calcifications. E. Unenhanced and (F) contrast-enhanced CT images show a combination of multiple tiny hyperattenuating foci typical of calcifications, associated with multiple small hypoattenuating foci, in this case due to small foci of fibrosis. These are all imaging findings of Gamna-Gandy bodies due to small foci of hemorrhage in this patient with portal hypertension.

soft-tissue density within inflamed peritoneal planes. In addition, CT plays a therapeutic role by guiding percutaneous drainage of an abscess.

Injuries to adjacent structures

Portal or splenic vein thrombosis Thrombosis of the portal venous system is an infrequent complication after splenectomy, but occurs more often following laparoscopic than open splenectomy.191 On contrastenhanced CT, a filling defect within the portomesenteric vein is diagnostic of venous thrombosis. If unrecognized and left untreated, this may appear later as cavernous transformation of the portal vein.

Intraoperative manipulation or improper use of electrocautery during splenectomy may lead to direct pancreatic injury and postoperative pancreatitis that may become complicated with pseudocyst formation.10 Gastric injury, including perforation and mural necrosis, may be caused during dissection and ligation of the short gastric vessels in the splenogastric ligament.192 On CT, gastric wall injury and perforation present as a localized fluid collection with gas bubbles and/or ingested oral contrast in the splenic bed, or with large-volume hydropneumoperitoneum. Injury to adjacent colon is a rare additional possible complication.

860 Diagnostic Abdominal Imaging

A

B

Figure 14-48 Splenosis This 53-year-old man had a splenectomy for splenic trauma several years previously. A. Contrast-enhanced CT of the left upper quadrant shows several surgical clips (arrows) and absence

of the spleen consistent with the prior splenectomy. B. Image a few centimeters lower shows 2 rounded masses (arrowheads) in the inferior aspect of the surgical bed. These represent regenerated splenic nodules: splenosis.

REFERENCES 1. Megremis S, Vlachonikolis I, Tsilimigaki A. Spleen length in childhood with US: normal values based on age, sex, and somatometric parameters. Radiology. 2004;231:129-134. 2. Dodds W, Taylor A, Erickson S, et al. Radiologic imaging of splenic anomalies. AJR Am J Roentgenol. 1990;155: 805-810. 3. Paterson A, Frush D, Donnelly L, et al. A pattern-oriented approach to splenic imaging in infants and children. Radiographics. 1999;19:1465-1485. 4. Donnelly L, Foss J, Frush D, et al. Heterogeneous splenic enhancement patterns on spiral CT images in children: minimizing misinterpretation. Radiology. 1999;210: 493-497. 5. Mirowitz S, Brown J, Lee J, et al. Dynamic gadoliniumenhanced MR imaging of the spleen: normal enhancement patterns and evaluation of splenic lesions. Radiology. 1991;179:681-686. 6. Gallagher B, Ansari A, Atkins H, et al. Radiopharmaceuticals XXVII: 18F-labeled 2-deoxy-2-fluoro-D-glucose as a radiopharmaceutical for measuring regional myocardial glucose metabolism in vivo—tissue distribution and imaging studies in animals. J Nucl Med. 1977;18:990-996.

9. Ramani M, Reinhold C, Semelka R, et al. Splenic hemangiomas and hamartomas: MR imaging characteristics of 28 lesions. Radiology. 1997;202:166-172. 10. Targarona E, Espert J, Bombuy E, et al. Complications of laparoscopic splenectomy. Arch Surg. 2000;135:1137-1140. 11. Wijaya J, Kapoor R, Roach P. Tc-99m-labeled RBC scintigraphy and splenic hemangioma. Clin Nucl Med. 2001; 26:1022-1023. 12. Gulenchyn K, Dover M, Kelly S. Splenic hemangioma presenting as a hot spot on radiocolloid scintigraphy. J Nucl Med. 1986;27:804-806. 13. Solomou E, Patriarheas G, Mpadra F, et al. Asymptomatic adult cystic lymphangioma of the spleen: case report and review of the literature. Magn Reson Imaging. 2003;21:81-84. 14. Morgenstern L, Bello J, Fisher B, et al. The clinical spectrum of lymphangiomas and lymphangiomatosis of the spleen. Am Surg. 1992;58:599-604. 15. Wadsworth D, Newman B, Abramson S, et al. Splenic lymphangiomatosis in children. Radiology. 1997;202:173-176. 16. Warnke R, Weiss L, Chan J, et al. Tumors of the lymph nodes and spleen. 3rd series ed. Washington, DC: Armed Forces Institute of Pathology; 1995.

7. Abbott R, Angela D, Aguilera N, et al. Primary vascular neoplasms of the spleen: radiologic-pathologic correlation. Radiographics. 2004;24:1137-1163.

17. Duchesne JC, Simmons JD, Schmieg RE Jr, McSwain NE Jr, Bellows CF. Proximal splenic angioembolization does not improve outcomes in treating blunt splenic injuries compared with splenectomy: a cohort analysis. J Trauma. 2008;65:1346-51.

8. Abbott R, Angela D, Aguilera N, et al. Primary vascular neoplasms of the spleen: radiologic-pathologic correlation. Radiographics. 2004;24:1137-1163.

18. Bader T, Ranner G, Kimpfinger M. Case report: CT appearance of capillary and cavernous lymphangiomatosis of the spleen in an adult. Clin Radiol. 1998;53:379-387.

Chapter 14 Imaging of the Spleen 861 19. Bezzi M, Spinelli A, Pierleoni M, et al. Cystic lymphangioma of the spleen: US-CT-MRI correlation. Eur Radiol. 2001;11: 1187-1190.

43. Franquet T, Montes M, Lecumberri F, et al. Hydatid disease of the spleen: imaging findings in 9 patients. AJR Am J Roentgenol. 1990;154:525-528.

20. Westh H, Reines E, Skibsted L. Splenic abscesses: a review of 20 cases. Scand J Infect Dis. 1990;22:569-573.

44. von Sinner W, Stnidbeck H. Hydatid disease of the spleen: ultrasonography, CT, and MR imaging. Acta Radiol. 1992;33:459-461.

21. Fotiadis C, Lavranos G, Patapis P, et al. Abscesses of the spleen: report of three cases. World J Gastroenterol. 2008;14: 3088-3091.

45. von Sinner W, te Strake L, Clark D, et al. MR imaging in hydatid disease. AJR Am J Roentgenol. 1991;157:741-745.

22. Losanoff J, Basson M. Splenic abscess. emedicine.medscape. com/article/194655-overview.

46. Marani S, Canossi G, Nicoli F, et al. Hydatid disease: MR imaging study. Radiology. 1990;175:701-706.

23. Lee C, Leu H, Hum T, Liu J. Splenic abscess in southern Taiwan. J Microbiol Immunol Infect. 2004;37:39-44.

47. Ito K, Mitchell K, Honjo K, et al. MR imaging of acquired abnormalities of the spleen. AJR Am J Roentgenol. 1997;168: 697-702.

24. Tung C, Chen F, Lo C. Splenic abscess: an easily overlooked disease? Am Surg. 2006;72:322-325. 25. Ulhaci N, Meteoglu I, Kacar F, Ozbas S. Abscess of the spleen. Pathol Oncol Res. 2004;10(4):234-236. 26. Villamil-Cajoto I, Lado F, Van den Eynde-Collado A, DíazPeromingo J. Splenic abscess: presentation of nine cases. Rev Chilena Infectol. 2006;23:150-154. 27. Al-Salem A, Qaisaruddin S, Al Jam’a A, Al-Kalaf J, El-Bashier A. Splenic abscess and sickle cell disease. Am J Hematol. 1998;58:100-104.

48. Garvin D, King F. Cysts and nonlymphomatous tumors of the spleen. Pathol Annu. 1981;16:61-80. 49. Robbins F, Yellin A, Lingua R, Craig J, Turrill F, Mikkelsen W. Splenic epidermoid cysts. Ann Surg. 1978;187:231-235. 50. Maskey P, Rupakheti S, Regmi R, Adhikary S, Agrawal C. Splenic epidermoid cyst. Kathmandu Univ Med J. 2007;5: 250-252. 51. Heider R, Behrns K. Pancreatic pseudocysts complicated by splenic parenchymal involvement: results of operative and percutaneous management. Pancreas. 2001;23:20-25.

28. Chang K, Chuah S, Changchien C, et al. Clinical characteristics and prognostic factors of splenic abscess: a review of 67 cases in a single medical center of Taiwan. World J Gastroenterol. 2006;12:460-464.

52. Georg C, Schwerk W, Georg K, et al. Sonographic patterns of the affected spleen in malignant lymphoma. J Clin Ultrasound. 1990;18:569-574.

29. Chun C, Raff M, et al. Splenic Abscess. Medicine (Baltimore). 1990;59:50-65.

53. Ahmann D, Kiely J, Harrison EG Jr, Payne WS. Malignant lymphoma of the spleen. A review of 49 cases in which the diagnosis was made at splenectomy. Cancer. 1966;19:461-469.

30. Nelken N, Ignatius J, Skinner M, Christensen N. Changing clinical spectrum of splenic abscess. Am J Surg. 1987;154:27-34. 31. Ooi L, Leong S. Splenic abscess from 1987-1995. Am J Surg. 1997;174:87-93. 32. Alvi A, Kulsoom S, Shamsi G. Splenic abscess: outcome and prognostic factors. J Coll Physicians Surg Pak. 2008;18:740-743. 33. Zerem E, Bergsland J. Ultrasound guided percutaneous treatment for splenic abscesses: the significance in treatment of critically ill patients. World J Gastroenterol. 2006;12: 7341-7345. 34. Ferraioli G, Brunetti E, Gulizia R, Mariani G, Marone P, Filice C. Management of splenic abscess: report on 16 cases from a single center. Int J Infect Dis. 2009;13:524-530. 35. Chiang I, Lin T, Chiang I, Tsai M. Splenic abscesses: review of 29 cases. Kaohsiung J Med Sci. 2003;19:510-515. 36. Urrutia M, Mergo P, Ros L, et al. Cystic masses of the spleen: radiologic-pathologic correlation. Radiographics. 1996;16: 107-129. 37. Ng K, Lee T, Wan Y, et al. Splenic abscess: diagnosis and management. Hepatogastroenterology. 2002; 49:567-571. 38. Fowler R. Hydatid cysts of spleen. Int Abstr Surg. 1953;96:105-116. 39. Manterola C, Vial M, Losada H, et al. Uncommon locations of abdominal hydatid disease. Trop Doct. 2003;33:179-180. 40. Durgun V, Kapan S, Kapan M, et al. Primary splenic hydatidosis. Dig Surg. 2003; 20:38-41. 41. Polat P, Kantarci M, Alper F, et al. Hydatid disease from head to toe. Radiographics. 2003;23:475-494; quiz 536-537. 42. Atmatzidis K, Papaziogas B, Mirelis C, et al. Splenectomy versus spleen-preserving surgery for splenic echinococcus. Dig Surg. 2003;20:527-531.

54. Schon C, Gorg C, Ramaswamy A, et al. Splenic metastases in a large unselected autopsy series. Pathol Res Pract. 2006;202: 351-356. 55. Falk S, Krishnan J, Meis J. Primary angiosarcoma of the spleen. Am J Surg Pathol. 1993;17:959-970. 56. Thompson W, Levy A, Aguilera N, et al. Angiosarcoma of the spleen: imaging characteristics in 12 patients. Radiology. 2005; 235:106-115. 57. Krishnan J, Frizzera G. Two splenic lesions in need of clarification: hamartoma and inflammatory pseudotumor. Semin Diagn Pathol. 2003;20:94-104. 58. Ohtomo K, Fukuda H, Mori K, et al. CT and MR appearances of splenic hamartoma. J Comput Assist Tomogr. 1992;16:425-428. 59. Gulenchyn K, Dover M, Kelly S. Splenic hemangioma presenting as a hot spot on radiocolloid scintigraphy. J Nucl Med. 1986;27:804-806. 60. Safran D, Welch J, Requke W. Inflammatory pseudotumor of the spleen. Arch Surg. 1991;126:904-908. 61. Monforte-Mujnoz H, Ro J, Manning J. Inflammatory pseudotumor of the spleen: report of two cases with a review of the literature. Am J Clin Pathol. 1991;96:491-495. 62. Franquet T, Montes M, Aizcorbe M, et al. Inflammatory pseudotumor of the spleen: ultrasound and computed tomographic findings. Gastrointest Radiol. 1989;14:181-183. 63. Glazer M, Lally J, Kanzer M. Inflammatory pseudotumor of the spleen: MR findings. J Comput Assist Tomogr. 1992;16:980-983. 64. Tinkoff G, Esposito T, Reed J, et al. American Association for the Surgery of Trauma Organ Injury Scale I: Spleen, Liver, and Kidney, Validation Based on the National Trauma Data Bank. J Am Coll Surg. 2008;207:646-655.

862

Diagnostic Abdominal Imaging

65. Peitzman A, Heil B, Rivera L, et al. Blunt splenic injury in adults: multiinstitutional study of the Eastern Association for the Surgery of Trauma. J Trauma. 2000;49:177-189.

85. Rabushka L, Kawashima A, Fishman E. Imaging of the spleen: CT with supplemental MR examination. Radiographics. 1994; 14:307-332.

66. Haan J, Scott J, Boyd-Kranis R, Ho S, Kramer M, Scalea T. Admission angiography for blunt splenic injury: advantages and pitfalls. J Trauma. 2001;51:1161-1165.

86. Hahn P, Weissleder R, Stark D, et al. MR imaging of focal splenic tumors. AJR Am J Roentgenol. 1988;150:823-827.

67. Pachter H, Guth A, Hofstetter S, Spencer F. Changing patterns in the management of splenic trauma: the impact of nonoperative management. Ann Surg. 1998;227:708-719. 68. Bee T, Croce M, Miller P, et al. Failures of splenic nonoperative management: is the glass half empty or half full? J Trauma. 2001;50:230-236. 69. Velmahos G, Chan L, Kamel E, et al. Nonoperative management of splenic injuries: have we gone too far? Arch Surg. 2000;135:674-681. 70. Boulanger B, Kearney P, Brenneman F, et al. Utilization of FAST (Focused Assessment with Sonography for Trauma) in 1999: results of a survey of North American trauma centers. Am Surg. 2000;66:1049-1055. 71. Goldberg B, Goodman G, Clearfield H. Evaluation of ascites by ultrasound. Radiology. 1970;96:15-22. 72. Shanmuganathan K, Mirvis S, Sherbourne C, et al. Hemoperitoneum as the sole indicator of abdominal visceral injuries: a potential limitation of screening abdominal US for trauma. Radiology. 1999;212:423-430. 73. Röthlin M, Näf R, Amgwerd M, Candinas D, Frick T, Trentz O. Ultrasound in blunt abdominal and thoracic trauma. J Trauma. 1993;34:488-495. 74. Goletti O, Ghiselli G, Lippolis P, et al. The role of ultrasonography in blunt abdominal trauma: results in 250 consecutive cases. J Trauma. 1994;36:178-181. 75. Becker D, Metha G, Terrier F. Blunt abdominal trauma in adults: role of CT in the diagnosis and management of visceral injuries Part 1: liver and spleen. Eur Radiol. 1998;8:553-562.

87. Abbott R, Levy A, Aguilera N, Gorospe L, Thompson W. From the archives of the AFIP primary vascular neoplasms of the spleen: radiologic-pathologic correlation. Radiographics. 2004;24:1137-1163. 88. Chang KC, Chuah SK, Changchien CS, Tsai TL, Lu SN, Chiu YC, et al. Clinical characteristics and prognostic factors of splenic abscess: a review of 67 cases in a single medical center of Taiwan. World J Gastroenterol. Jan 2006;12(3):460–464. 89. Chun CH, Raff MJ, Contreras L, Varghese R, Waterman N, Daffner R, et al. Splenic abscess. Medicine (Baltimore). 1990;59:50–65. 90. Nelken N, Ignatius J, Skinner M, Christensen N. Changing clinical spectrum of splenic abscess. Am J Surg. 1987;154:27–34 91. Ooi LL, Leong SS. Splenic abscess from 1987-1995. Am J Surg. 1997;174:87–93. 92. Ferraioli G, Brunetti E, Gulizia R, Mariani G, Marone P, Filice C. Management of splenic abscess: report on 16 cases from a single center. Int J Infect Dis. Jul 2009;13(4):524–530 93. Tan K, Chen K, Sim R. The spectrum of abdominal tuberculosis in a developed country: a single institution’s experience over 7 years. J Gastrointest Surg. 2009;13:142-147. 94. Topal U, Savci G, Yurtkuran Sadikoglu M, Parlak M, Tuncel E. Splenic involvement of tuberculosis: US and CT findings. Eur Radiol. 1994;4:577-579. 95. Sharma S, Smith-Rohrberg D, Tahir M, Mohan A, Seith A. Radiological manifestations of splenic tuberculosis: a 23-patient case series from India. Indian J Med Res. 2007;125:669-678.

76. Sclafani S. The use of angiographic hemostasis in salvage of the injured spleen. Radiology. 1981;141:645-650.

96. Rabushka L, Kawashima A, Fishman E. Imaging of the spleen: CT with supplemental MR examination. Radiographics. 1994;14:307-332.

77. Haan J, Biffl W, Knudson MM, et al. Splenic embolization revisited: a multicenter review. J Trauma. 2004;56:542-547.

97. Malik A, Saxena N. Ultrasound in abdominal tuberculosis. Abdom Imaging. 2003;28:574-579.

78. Shanmugnathan K, Mirvis S, Boyd-Kranis R, et al. Nonsurgical management of blunt splenic injury: use of CT criteria to select patients for splenic arteriography and potential endovascular therapy. Radiology. 2000;217:75-82.

98. Radin R. HIV infection: analysis in 259 consecutive patients with abnormal abdominal CT findings. Radiology. 1995;197: 712-722.

79. Thompson B, Munera F, Cohn S, et al. Computed tomography scan scoring system predicts the need for intervention after splenic injury. Trauma. 2006;60:1083-1086. 80. Shanmuganathan K, Mirvis S, Sover E. Value of contrastenhanced CT in detecting active hemorrhage in patients with blunt abdominal or pelvic trauma. AJR Am J Roengenol. 1993;161:65-69. 81. De Schepper A, Vanhoenacker F, de Beeck BO, et al. Vascular pathology of the spleen, part II. Abdom Imaging. 2005;30: 228-238. 82. Jaroch M, Broughan T, Hermann R. The natural history of splenic infarction. Surgery. 1986;100:743. 83. Goerg C, Schwerk W. Splenic infarction: sonographic patterns, diagnosis, follow-up, and complications. Radiology. 1990;174:803. 84. Robertson F, Leander P, Ekberg O. Radiology of the spleen. Eur Radiol. 2001;11:80-95.

99. Kapoor R, Jain A, Chaturvedi U, Saha M. Ultrasound detection of tuberculomas of the spleen. Clin Radiol. 1991;43:128-129. 100. Hasan M, Sarwar J, Bhuiyan J, Islam S. Tubercular splenic abscess. Mymensingh J Med. 2008;17:67-69. 101. Reichel C, Theisen A, Rockstroh J, Muller-Miny H, Spengler U, Sauerbruch T. Splenic abscesses and abdominal tuberculosis in patients with AIDS. Z Gastroenterol. 1996;34:494-496. 102. Radin D. Disseminated histoplasmosis: abdominal CT findings in 16 patients. AJR Am J Roentgenol. 1991;157(5): 955-958. 103. Buckner C, Leithiser R, Walker C, et al. The changing epidemiology of tuberculosis and other mycobacterial infections in the United States: implications for the radiologist. AJR Am J Roentgenol. 1991;156:255-264. 104. von Eift M, Essink M, Roos N, Hiddemann W, Buchner T, van de Loo J. Hepatosplenic candidiasis, a late manifestation of Candida septicaemia in neutropenic patients with haematologic malignancies. Blut. 1990;60:242-248.

Chapter 14 Imaging of the Spleen 863 105. Cho J, Kim E, Varma D, et al. MR imaging of hepatosplenic candidiasis superimposed on hemochromatosis. J Comput Assist Tomogr. 1990;14:774-776.

125. Kaneko J, Sugawara Y, Matsui Y, Makuuchi M. Spleen size of live donors for liver transplantation. Surg Radiol Anat. 2008;30:515-518.

106. Pastakia B, Shawker T, Thaler M, O’Leary T, Pisso P. Hepatosplenic candidiasis: wheels within wheels. Radiology. 1988;166:417-421.

126. Loftus WK, Metreweli C. Normal splenic size in a Chinese population. J Ultrasound Med. 1997;16:345-347.

107. Chew F, Smith P, Barboniak D. Candidal splenic abscesses. AJR Am J Roentgenol. 1991;156:474.

127. Prassopoulos P, Daskalogiannaki M, Raissaki M, Hatjidakis A, Gourtsoyiannis N. Determination of normal splenic volume on computed tomography in relation to age, gender and body habitus. Eur Radiol. 1997;7:246-248.

108. Semelka R, Shoenut J, Greenberg H, Bow E. Detection of acute and treated lesions of hepatosplenic candidiasis: comparison of dynamic contrast-enhanced CT and MR imaging. J Magn Reson Imaging. 1992;2:341-345.

128. Geraghty EM, Boone JM, McGahan JP, Jain K. Normal organ volume assessment from abdominal CT. Abdom Imaging. 2004;29:482-490.

109. Lubat E, Megibow A, Balthazar E, et al. Extrapulmonary Pneumocystis carinii infection in AIDS. Radiology. 1990;174:157-160.

129. Bezerra AS, D’Ippolito G, Faintuch S, Szejnfeld J, Ahmed M. Determination of splenomegaly by CT: is there a place for a single measurement? AJR Am J Roentgenol. 2005;184:1510-1513.

1110. Fishman E, Magid D, Kuhimanj E. Pneumocystis carinii involvement of the liver and spleen: CT demonstration. J Comput Assist Tomogr. 1990;14:146-148.

130. Lamb PM, Lund A, Kanagasabay RR, Martin A, Webb JAW, Reznek RH. Spleen size: how well do linear ultrasound measurements correlate with three-dimensional CT volume assessments? Br J Radiol. 2002;75:573-577.

1111. Radin DR, Baker EL, Klatt EC, et al. Visceral and nodal calcification in patients with AIDS-related Pneumocystis carinii infection. AJR Am J Roentgenol. 1990;154:27-31. 112. Kahr A, Kerbl R, Gschwandtner K, et al. Visceral manifestation of cat scratch disease in children. A consequence of altered immunological state? Infection. 2000;28:116-118. 113. Ishikawa T, Suzuki T, Shinoda M, et al. A case of hepatosplenic cat scratch disease [in Japanese]. Nippon Shokakibyo Gakkai Zasshi. 2006;103:1050-1054. 114. Luciano A, Rossi F, Bolognani M, Trabucchi C. Hepatic and splenic micro-abscess in cat scratch disease. Report of a case [in Italian]. Pediatr Med Chir. 1999;21:89-91. 115. Baughman R, Teirstein A, Judson M, et al.; Case Control Etiologic Study of Sarcoidosis (ACCESS) research group. Clinical characteristics of patients in a case control study of sarcoidosis. Am J Respir Crit Care Med. 2001;164:1885-1889.

131. Weber SM, Rikkers LF. Splenic vein thrombosis and gastrointestinal bleeding in chronic pancreatitis. World J Surg. 2003;27:1271-1274. 132. Koklu S, Koksal A, Yolcu O, et al. Isolated splenic vein thrombosis: an unusual cause and review of the literature. Can J Gastroenterol. 2004;13:173-174. 133. Castellino R. Hodgkin disease: practical concepts for the diagnostic radiologist. Radiology. 1986;159:305-310. 134. Rueffer U, Sieber M, Stemberg M, et al.; German Hodgkin’s Lymphoma Study Group (GHSG). Spleen involvement in Hodgkin’s lymphoma: assessment and risk profile. Ann Hematol. 2003;82:390-396. 135. Breiman R, Castellino R, Harrell G, et al. CT-pathologic correlations in Hodgkin’s disease and non-Hodgkin’s lymphoma. Radiology. 1978;126:159-166.

116. Judson M. Hepatic, splenic, and gastrointestinal involvement with sarcoidosis. Semin Respir Crit Care Med. 2002;23:529-541.

136. Dachman A, Buck J, Krishnan J, et al. Primary non-Hodgkin’s splenic lymphoma. Clin Radiol. 1998;53:137-142.

117. Warshauer D, Dumbleton S, Molina P, et al. Abdominal CT findings in sarcoidosis: radiologic and clinical correlation. Radiology. 1994;192:93-98.

137. Munker R, Stengel A, Stabler A, et al. Diagnostic accuracy of ultrasound and computed tomography in the staging of Hodgkin’s disease. Verification by laparotomy in 100 cases. Cancer. 1995;76:1460-1466.

118. Kessler A, Mitchell D, Israel H, et al. Hepatic and splenic sarcoidosis: US and MR imaging. Abdom Imaging. 1993;18: 159-163.

138. Siniluoto T, Tikkakoski T, Lahde S, et al. Ultrasound or CT in splenic diseases? Acta Radiol. 1994;35:597-605.

119. Folz S, Johnson C, Swensen S. Abdominal manifestations of sarcoidosis in CT studies. J Comput Assist Tomogr. 1995;19: 573-579.

139. Hess C, Griebel J, Schmiedl U, et al. Focal lesions of the spleen: preliminary results with fast MR imaging at 1.5 T. J Comput Assist Tomogr. 1988;2:569-574.

120. Warshauer DM, Molina PL, Hamman SM, et al. Nodular sarcoidosis of the liver and spleen: analysis of 32 cases. Radiology. 1995;195:757-762.

140. Rini J, Leonidas J, Tomas M, et al. 18F-FDG PET versus CT for evaluating the spleen during initial staging of lymphoma. J Nucl Med. 2003;44:1072-1074.

121. Warshauer D, Molina P, Hamman S, et al. Nodular sarcoidosis of the liver and spleen: analysis of 32 cases. Radiology. 1995; 195:757-762.

141. Hicks R, Mac Manus M, Seymour J. Initial staging of lymphoma with positron emission tomography and computed tomography. Semin Nucl Med. 2005;35:165-175.

122. Poll L, Koch J, vom Dahl S, et al. Gaucher disease of the spleen: CT and MR findings. Abdom Imaging. 2000;25:286-289.

142. Egeler RM, Laura S, Pieter S, Carlos M, Mark EN. Malignant histiocytosis: a reassessment of cases formerly classified as histiocytic neoplasms and review of the literature. Med Pediatr Oncol. 1995;25:1-7.

123. Hosey RG, Mattacola CG, Kriss V, Armsey T, Quarles JD, Jagger J. Ultrasound assessment of spleen size in collegiate athletes. Br J Sports Med. 2006;40:251-254. 124. Spielmann AL, DeLong DM, Kliewer MA. Sonographic evaluation of spleen size in tall healthy athletes. AJR Am J Roentgenol. 2005;184:45-49.

143. James WV, Gerald EB Jr, Henry R. Malignant histiocytosis with massive splenomegaly in asymptomatic patients. A possible chronic form of the disease. Cancer. 1975;36: 419-427.

864

Diagnostic Abdominal Imaging

144. Hirsch MS. Cytomegalovirus infection. In: Braunwald E, Isselbacher KJ, Petersdorf RG, Wilson JD, Martin JB, Fauci AS, eds. Harrison’s Principles of Internal Medicine. New York: McGraw-Hill; 1987:697-699. 145. Schooley RT. Epstein-Barr virus infections, including infectious mononucleosis. In: Braunwald E, Isselbacher KJ, Petersdorf RG, Wilson JD, Martin JB, Fauci AS, eds. Harrison’s Principles of Internal Medicine. New York: McGraw-Hill; 1987: 699-703. 146. Singh B, Kim Sung L, Matusop A, et al. A large focus of naturally acquired Plasmodium knowlesi infections in human beings. Lancet. 2004;363:1017-1024. 147. Snow R, Guerra C, Noor A, Myint H, Hay S. The global distribution of clinical episodes of Plasmodium falciparum malaria. Nature. 2005;434:214-217. 148. Malaria Facts. Centers for Disease Control and Prevention. 149. Nadeem A, Ali N, Hussain T, Anwar M. Frequency and etiology of splenomegaly in adults seeking medical advice in Combined Military Hospital Attock. J Ayub Med Coll Abbottabad. 2004;16:44-47. 150. Timite-Konan M, Kouame K, Konan A, et al. Etiology of splenomegaly in children in the tropics. 178 cases reviewed at the university hospital center of Abidjan-Cocody (Ivory Coast) [in French]. Ann Pediatr (Paris). 1992;39:136-141. 151. Betticher D, Nicole A, Pugin P, Regamey C. The hyperreactive malarial splenomegaly syndrome in a European: has the treatment of a modulatory effect on the immune system? J Infect Dis. 1990;161:157-159. 152. Magid D, Fishman EK, Siegelman SS. Computed tomography of the spleen and liver in sickle cell disease. AJR Am J Roentgenol. 1984;143:245-249. 153. Eshel Y, Sarova-Pinhas I, Lampl Y, Jedwab M. Autosplenectomy complicating pneumococcal meningitis in an adult. Arch Intern Med. 1991;151:998-999. 154. Leipe J, Hueber AJ, Kallert S, Rech J, Schulze-Koops H. Autosplenectomy: rare syndrome in autoimmunopathy. Ann Rheum Dis. 2007;66:566-567. 155. Claster S, Vichinsky EP. Managing sickle cell disease. BMJ. 2003;327:1151-1155.

163. Deux J, Salomon L, Barrier A, et al. Acute torsion of wandering spleen: MRI findings. AJR Am J Roentgenol. 2004;182:1607-1608. 164. Winer-Muram H, Tonkin I. The spectrum of heterotaxic syndromes. Radiol Clin North Am. 1989;27:1147-1170. 165. Peoples W, Moller J, Edwards J. Polysplenia: a review of 146 cases. Pediatr Cardiol. 1983;4:129-137. 166. Lee FT Jr, Pozniak M, Helgerson R. US case of the day. Polysplenia syndrome. Radiographics. 1993;13:1159-1162. 167. Soler R, Rodriguez E, Comesana M, et al. Agenesis of the dorsal pancreas with polysplenia syndrome: CT features. J Comput Assist Tomogr. 1992;16:921-923. 168. Fulcher A, Turner M. Abdominal manifestations of situs anomalies in adults. Radiographics. 2002;22:1439-1456. 169. Applegate K, Goske M, Pierce G, et al. Situs revisited: imaging of the heterotaxy syndrome. Radiographics. 1999;19:837-852; discussion 853-854. 170. Marx M, Van AR. SIR 2005 film panel case: heterotaxia with polysplenia. J Vasc Interv Radiol. 2005;16:1055-1059. 171. Serviansky B, Schwarz J. The incidence of splenic calcifications in positive reactors to histoplasmin and tuberculin. Am J Roentgenol Radium Ther Nucl Med. 1956;76:53-59. 172. Grey EF. Calcifications of the spleen. AJR Am J Roentgenol. 1944;51:336-351. 173. Topin J, Mutlu GM. Splenic and mediastinal calcifications in histoplasmosis. N Engl J Med. 2006;354:179. 174. Okudaira M, Straub M, Schwarz J. The etiology of discrete splenic and hepatic calcifications in an endemic area of histoplasmosis. Am J Pathol. 1961;39:599-611. 175. Jacobs JE, Birnbaum BA, Furth EE. Abdominal visceral calcification in primary amyloidosis: CT findings. Abdom Imaging. 1997;22:519-521. 176. Vanhoenacker F, Van den Brande P, De Schepper A. Hepatosplenic anthracosilicosis: a rare cause of splenic calcifications. Eur Radiol. 2001;11:1184-1186. 177. Radhakrishnan S, al Nakib B, Sivanandan R, Menon N. Hepatosplenic and small bowel calcification due to Schistosoma mansoni infection. Dig Dis Sci. 1988;33: 1637-1640.

156. McCall I, Vaidya S, Serjeant G. Splenic opacification in homozygous sickle cell disease. Clin Radiol. 1981;32: 611-615.

178. Ishizaki T, Kuroda H, Kuroda T, Nakai T, Miyabo S. Sarcoidosis with multiple calcification. Jpn J Med. 1988;27:191-194.

157. Walker T, Serjeant G. Focal echogenic lesions in the spleen in sickle cell disease. Clin Radiol. 1993;47:114-116.

179. Fischer R, Harmatz PR. Non-invasive assessment of tissue iron overload. Hematology. 2009;2009:215-221.

158. Adler DD, Glazer GM, Aisen AM. MRI of the spleen: normal appearance and findings in sickle-cell anemia. AJR Am J Roentgenol. 1986;147:843-845.

180. Salo S, Alanen A, Leino R, Bondestam S, Komu M. The effect of haemosiderosis and blood transfusions on the T2 relaxation time and 1/T2 relaxation rate of liver tissue. Br J Radiol. 2002;75:24-27.

159. Gayer G, Zissin R, Apter S, et al. CT findings in congenital anomalies of the spleen. Br J Radiol. 2001;74:767-772. 160. Raissaki M, Prassopoulos P, Daskalogiannaki M, et al. Acute abdomen due to torsion of wandering spleen: CT diagnosis. Eur Radiol. 1998;8:1409-1412. 161. Taori K, Ghonge N, Prakash A. Wandering spleen with torsion of vascular pedicle: early diagnosis with multiplaner reformation technique of multislice spiral CT. Abdom Imaging. 2004;29:479-481. 162. Desai D, Hebra A, Davidoff A, et al. Wandering spleen: a challenging diagnosis. South Med J. 1997;90:439-443.

181. Wood JC, Enriquez C, Ghugre N, et al. MRI R2 and R2* mapping accurately estimates hepatic iron concentration in transfusion-dependent thalassemia and sickle cell disease patients. Blood. 2005;106:1460-1465. 182. Sagoh T, Itoh K, Togashi K, et al. Gamna-Gandy bodies of the spleen: evaluation with MR imaging. Radiology. 1989;172: 685-687. 183. Bhatt S, Simon R, Dogra V. Gamna-Gandy bodies: sonographic features with histopathologic correlation. J Ultrasound Med. 2006;25:1625-1629.

Chapter 14 Imaging of the Spleen 865 184. Luo T, Itai Y, Yamaguchi M, Kurosaki Y, Saida Y. Gamna-Gandy bodies of the spleen depicted by unenhanced CT: report of two cases. Radiat Med. 1998;16:473-476. 185. Selçuk D, Demirel K, Kantarci F, Mihmanli I, Oğüt G. Gamna-Gandy bodies: a sign of portal hypertension. Turk J Gastroenterol. 2005;16:150-152. 186. Ghahremani G, Gore R. CT diagnosis of postoperative abdominal complications. Radiol Clin North Am. 1989;27: 787-804. 187. Gayer G, Hertz M, Zissin R. Postoperative pneumoperitoneum: prevalence, duration, and possible significance. Semin Ultrasound CT MR. 2004;25:286-289. 188. Khosravi M, Margulies D, Alsabeh R, et al. Consider the diagnosis of splenosis for soft tissue masses long after any splenic injury. Am Surg. 2004;70:967-970.

189. Osadchy A, Zissin R, Shapiro-Feinberg M. Thoracic splenosis. Isr Med Assoc J. 2001;3:547. 190. Sharma R, Mondal A, Kashyap R, et al. Radiolabeled denatured RBC scintigraphy in autologous splenic transplantation. Clin Nucl Med. 1996;21:534-536. 191. Winslow E, Brunt L. Perioperative outcomes of laparoscopic versus open splenectomy: a meta-analysis with an emphasis on complications. Surgery. 2003;134:647-653. 192. Martinez C, Waisberg J, Palma R, et al. Gastric necrosis and perforation as a complication of splenectomy. Case report and related references. Arq Gastroenterol. 2000;27:227-230. 193. Moore EE, Cogbill TH, Jurkovich GJ, Shackford SR, Malangoni MA, Champion HR. Organ injury scaling: spleen and liver (1994 revision). J Trauma. 1995;38(3):323-324.

This page intentionally left blank

CHAPTER

15

Imaging of the Arteries and Veins of the Abdomen and Pelvis Saurabh Jha, MBBS, MRCS Wallace T. Miller Jr., MD

I. IMAGING MODALITIES a. CT Angiography b. MR Angiography c. Ultrasonography d. Catheter Angiography II. ANATOMY OF THE ABDOMINAL ARTERIES III. DISEASES OF THE ABDOMINAL ARTERIES a. Wall Thickening i. Atherosclerotic plaque ii. Mural thrombus iii. Intramural hematoma iv. Vasculitis v. Mycotic aneurysm b. Periadventitial Fat Stranding i. Retroperitoneal fibrosis (chronic periaortitis, inflammatory AAA, perianeurysmal retroperitoneal  fibrosis) c. Wall Calcification i. Atherosclerosis and intimal calcification ii. Aging, end-stage renal disease, diabetes mellitus, and medial calcification iii. Vasculitis d. Stenosis and Occlusion i. Atherosclerosis ii. Takayasu arteritis iii. Connective tissue disease iv. Thromboembolism v. Spontaneous thrombosis vi. Median arcuate ligament syndrome

Given that arteries and veins of a caliber detectable by modern-day imaging permeates virtually every organ system, imaging of any organ system can be considered imaging of the vascular tree. However, the vascular tree is an organ in its own right afflicted by a range

e. Aneurysms i. Degenerative AAAs (Atherosclerotic aneurysm, AAA) ii. Inflammatory aneurysms (chronic periaortitis, retroperitoneal fibrosis) iii. Penetrating atherosclerotic ulcer iv. Intramural hematoma and dissection v. Mycotic aneurysm vi. Trauma and surgery vii. Vasculitis viii. Connective tissue disorders ix. External factors leading to abdominal aneurysms f. Wall Disruption i. Penetrating atherosclerotic ulcer ii. Dissection iii. Rupture g. Gas in the Arterial Lumen or Wall of the Vessel i. Mycotic aneurysm and infected aortitis ii. Aortoenteric fistula IV. DISEASES OF THE ABDOMINAL VEINS a. Thrombus b. Narrowing of Veins i. Nutcracker syndrome ii. May-Thurner syndrome (iliocaval compression syndrome) c. Dilation of Veins i. Venous stenosis or occlusion ii. Pelvic congestion syndrome iii. Arteriovenous fistula/malformation d. Reversal of Flow i. Congenital abnormalities of the abdominal veins

of unique pathology and amenable to unique forms of treatment. There are differences in principle between arterial pathology and its management and management of pathology in other organ systems. These differences must 867

868

Diagnostic Abdominal Imaging

be appreciated by imagers in order to provide the most insightful and clinically useful reports. Imaging of the arteries carries with it 2 major responsibilities. The first is to aid in the diagnosis of arterial disease, the predominant etiology of which is atherosclerosis. The second purpose of imaging is that it helps the clinician decide the best course of intervention which, broadly speaking, falls into 3 groups: (1) do nothing surgically, either because nothing needs to be done or because nothing can be done; (2) open surgical intervention; and (3) percutaneous surgical intervention. What distinguishes imaging of the arteries from other organs is that it is generally assumed that information for surgical planning will be needed. There are a narrow range of morphologic end points that can result from a range of pathologic processes. However, both the morphology of the abnormality and the underlying pathology can affect the type of intervention. For example, there is no single management strategy for atherosclerosis. The management depends on what exactly atherosclerosis does to the artery and is a different strategy, if the result is occlusion of the artery from or if the result is an aneurysm of the artery. As a corollary, aneurysms caused by infection are managed differently to aneurysms caused by atherosclerosis in that the size threshold for surgical intervention is reduced, and aggressive medical management in the form of organism-specific antibiotics is also instituted. The abdominal arteries, particularly the aorta, are seldom afflicted in isolation and the systemic nature of the arterial pathology obliges a search for disease in unsuspected arterial beds. Also, extrinsic pathology can involve the aorta or the arteries of the lower extremities, a finding that can affect the type of intervention.

IMAGING MODALITIES The mainstay of imaging for arterial disease includes computed tomographic angiography (CTA) and magnetic resonance angiography (MRA), with ultrasonography (US) and conventional angiography (CA) having a less central role.

CT Angiography The combination of helical CT technology and multiple detectors have enabled interrogation of a large field of view within an achievable breath hold. From the vantage point of arterial imaging, this combination has been revolutionary, and now high-quality arterial contrast is all but routinely assumed. The typical vascular CT protocol is, of course, tailored to the clinical circumstance but generally comprises precontrast, arterial-phase and delayed (venous phase) imaging. Precontrast images are not mandatory but are useful particularly in a patient post surgical intervention or with suspected acute rupture. This is to distinguish between in situ high-attenuating materials (calcium, graft, stent) from iodinated contrast, particularly if such distinction makes a difference to the interpretation. Similarly, a venous phase is

valuable in certain situations such as the search for endoleaks post stent, but is not required routinely. The timing of the acquisition is the key to a diagnostic arterial phase. To best synchronize the acquisition of the images with the peak maximal enhancement of iodinated contrast in the artery of interest, guesswork should be kept to a minimum. Two recognized methods of achieving this are the bolus tracking and the test bolus methods. The former relies on placing a tracker on a certain part of the arterial tree, for example the suprarenal aorta, and performing low-dose single-slice images until a certain attenuation threshold, typically on the order of 130 Hounsfield units (HUs), is attained; thereafter, the scan commences after a fixed delay, typically 6 seconds. The ease of the bolus tracking method accounts for its greater popularity than the test bolus in which a smaller volume of contrast, approximately 20 mL, is injected at the same injection rate as used in the subsequent diagnostic scan and the time at which peak maximal enhancement occurs for a particular section of the aorta is graphed. The time to peak maximal enhancement is noted to guide the main injection. Nonionic low or iso-osmolar iodinated contrast agents are used for CT angiograms. The slices are reconstructed at a thickness between 0.6 to 3 mm to achieve near-isotropic resolution, in order that the images are processed with a variety of  3-dimensional (3D) rendering technologies (see Figure 15-1).

MR Angiography Arterial contrast can be achieved without the administration of exogenous contrast using flow-dependent methods, although many are still evolving and have yet to make a widespread clinical impact. The main method of achieving arterial contrast is the administration of gadolinium chelates and subsequent exploitation of the T1 shortening properties of gadolinium. With current hardware and pulse sequence design, high-quality arterial-phase images can be achieved with near isotropic resolution within an achievable breath hold. The availability of step table and moving table technology means that multiple contiguous body parts can be imaged with a single injection of contrast (see Figure 15-2). Advantages of MR over CT include (1) absence of ionizing radiation and (2) easier postprocessing because it is easier to remove nonarterial structures from MRI images than CT images. There are several disadvantages of MR relative to CT: (1) MR does not assess calcification—this is an important blind spot as the amount and location of calcification affects the suitability of the vessel for percutaneous interventions. (2) Stents cause a signal drop on MR due to the susceptibility artifact, making it impossible to measure stent stenosis and/or occlusion. (3) Finally, MR takes a longer time than CT to acquire images.

Ultrasonography US has been proposed as a screening modality for asymptomatic abdominal aortic aneurysms (AAAs). Its role in

Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis 869

A

B

Figure 15-1 Normal CTA A. Axial CTA shows intense opacification of the abdominal aorta. The 3 layers of the normal wall are imperceptible. B. Coronal-

oblique MIP shows the right renal artery extending from the origin to the hilum. Note the smooth, tapered appearance with the absence of luminal irregularity, typical of normal vessels.

diagnosis is not surprising as it is relatively inexpensive, available, and free of ionizing radiation and the use of exogenous contrast agents. However, it is not comparable to CT and MR for definitive diagnosis or surgical planning because it is unable to interrogate the entire vascular tree with the same clarity, precision, and reproducibility (see Figure 15-3). Ultrasound is not only operator-dependent but is also disadvantaged by nonoperator issues such as body habitus and the presence of bowel gas. It is for these reasons that barring the renal vasculature, US has not made any headway into the interrogation of vascular trees. However, duplex US has certain advantages. It is better able to appreciate the hemodynamic significance of narrowing than a purely anatomical modality such as CT and MR. It does so by measuring velocities across stenosis, quantifying a gradient, and by measuring resistive indices of stenosis such as in the renal arteries. Ultrasound has much to offer in certain clinical situations and in patients in whom it can be predicted a priori that the operator is less likely to be encumbered by technical challenges.

provide information about the arterial wall, including the presence of luminal thrombus or calcification, or extravascular structures. Catheter angiogram is less sensitive to contrast enhancement than CTA or MRA. Projectional images from catheter angiograms are less accurate at quantifying asymmetric stenosis than 3D images from CTA and MRA. Additionally, 3D images do not suffer from the problems due to overlap of vessels seen with projectional images. Catheter angiogram is invasive with puncture site complications, and it is more resource intensive than cross-sectional imaging. Despite its limitations, catheter angiograms have some advantages over cross-sectional imaging. (1) The temporal resolution of catheter angiogram is very high and multiple frames can provide information about the directionality of blood flow and (2) catheter angiogram has a higher spatial resolution than cross-sectional imaging. (3) Catheter angiogram is not susceptible to metallic artifact and assessment of in-stent stenosis is superior to CTA and MRA. (4) Finally, catheter angiogram has a role during intraprocedural guidance of intravascular intervention.

Catheter Angiography

ANATOMY OF THE ABDOMINAL ARTERIES

Biplane 2-dimensional (2D) catheter angiograms were the initial imaging method to evaluated the vascular tree but have been sidelined by CTA and MRA for a variety of reasons. Catheter angiogram offers a luminogram but does not

The aorta enters the abdominal cavity through an opening in the diaphragm opposite T12. The aorta bifurcates into the common iliac arteries opposite L4 into the right and left common iliac arteries.

870 Diagnostic Abdominal Imaging

A

B

C Figure 15-2 Normal MRA A. Axial T2-weighted, (B) axial reformation from 3D postgadolinium MRA, and (C) axial 2D postgadolinium gradient echo shows the normal aorta. Note the imperceptible aortic wall, which is composed of the intima, media, and the adventitia. Note the absence of detectable mural enhancement on the delayed images.

The abdominal aorta provides blood supply to the viscera of the abdomen and pelvis. The major visceral arteries include the celiac trunk, the superior mesenteric artery (SMA), the inferior mesenteric artery (IMA), and the bilateral renal arteries. In addition, the aorta gives rise to the

inferior phrenic artery, adrenal and lumbar arteries, the gonadal artery, and the median sacral artery. The celiac divides into the left gastric, common hepatic, and splenic arteries. The celiac provides blood supply to the upper abdominal viscera, the descendants of the foregut. There is a rich anastomotic network about the stomach and pancreas from branches of the celiac trunk and between the celiac trunk and the SMA. The common hepatic artery gives off the gastroduodenal artery (GDA) and right gastric artery and continues as the proper hepatic artery. The proper hepatic artery gives rise to the cystic artery and then divides into the right and left hepatic arteries. The GDA divides into the superior pancreaticoduodenal and right gastroepiploic artery. The superior pancreaticoduodenal artery then divides into an anterior and posterior branch that anastomoses with their counterparts from the inferior pancreaticoduodenal branch of the SMA. The right gastroepiploic artery anastomoses with the left gastroepiploic artery, which is a terminal branch of the  splenic artery. The splenic artery, in addition, gives rise to several short gastric arteries and several branches to the pancreas including the dorsal pancreatic, transverse pancreatic, caudal pancreatic, and pancreatic magna arteries. The SMA gives rise to the inferior pancreaticoduodenal artery, jejunal arteries, the ileocolic artery, ileal arteries, and the right and middle colic arteries. The SMA provides blood supply to the descendants of the midgut as far as the splenic flexure. There is a rich intramesenteric anastomotic network along the transverse colon between branches of the SMA and the IMA. The IMA gives rise to the left colic artery, several sigmoid arteries, and terminates as the superior hemorrhoidal artery that anastomoses with the middle hemorrhoidal branch of the anterior division of the internal iliac artery. Renal arteries divide at the renal hilum into anterior and posterior division. The anterior division supplies the apical, upper, middle, and lower segments of the kidney whereas the posterior division supplies the middle segment. Further subdivision into the lobar and thence interlobar arteries occurs. The interlobar arteries terminate at the corticomedullary junction into arcuate arteries. The arcuate arteries generally give rise to the interlobular artery, which becomes the afferent arteriole. The common iliac artery bifurcates into the internal and external iliac arteries opposite the pelvic brim. The internal iliac artery divides into an anterior and posterior branch. The posterior branch of the internal iliac arteries gives rise to the lateral sacral, iliolumbar, and superior gluteal arteries. The anterior division gives off the inferior gluteal, obturator, vesicle, middle hemorrhoidal, internal pudendal, and in females the uterine and in males the deferential arteries. The external iliac artery becomes the common femoral artery below the inguinal ligament. Before that it gives rise to the circumflex iliac, external pudendal, and inferior epigastric arteries.

Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis 871

A

B

Figure 15-3 Normal US Abdominal Aorta A. Longitudinal and (B) transaxial US images of the abdomen demonstrate a smooth, anechoic tubular structure (arrows),

with increased through transmission, measuring approximately 2.7 cm in diameter typical of the normal abdominal aorta.

Branches of the external and internal iliac arteries become important pathways for collaterals during aortoiliac occlusion. For example, blood in the inferior epigastric artery flows retrograde to reconstitute the external iliac artery in upstream occlusion, through its anastomosis with the superior epigastric artery. Also, the middle hemorrhoidal branch of the anterior division of the internal iliac artery anastomoses with the superior hemorrhoidal branch of the IMA to provide blood to the distal left colon in IMA occlusion. Noting the rich anastomotic network about the pelvic arteries, it is important for the imager to comment on the health of the parent arteries, so that a sense of potential for collateral recruitment may be gained, in the event of deliberate or accidental occlusion of the arteries during vascular intervention. A major congenital variant to be aware of is the persistent sciatic artery. This artery normally regresses but when persistent is associated with diminutive or absent ipsilateral external iliac and femoral arteries. The classic presentation is that of patient with absent femoral pulses but palpable pedal pulses. The persistent sciatic artery travels with the sciatic nerve through the greater sciatic foramen into the posterior thigh.

The arteries in the abdomen and pelvis can be affected by a range of acute and chronic conditions. These disorders begin with an abnormality of the arterial wall that can be radiographically occult or result in wall thickening or calcification or both. Structural abnormalities in the aortic wall can further lead to vascular stenosis, dilation (aneurysm formation), disruption, and perivascular soft tissue abnormalities (Table 15-1). The imaging features, pattern of vascular involvement, and the age of the patient can often indicate the underlying cause of vascular diseases. Because the disease begins with structural abnormalities in the vessel wall, we will begin our discussion of vascular diseases with a discussion of causes of wall thickening and calcification and then continue on to the complications of vascular diseases: stenosis, aneurysm formation, wall disruption, and perivascular soft-tissue changes.

DISEASES OF THE ABDOMINAL ARTERIES Arterial disease can be classified according to etiology (eg, atherosclerosis) or morphological end point (eg, aneurysm). In the following section, grouping of disease will be made according to morphologic end points. This is because such classification is most favorable to an imager seeking pattern recognition and applying a differential diagnosis based on pattern.

Table 15-1. Imaging Descriptors for Abdominal Arterial Pathology 1. Wall thickening 2. Wall calcification 3. Stenosis/occlusion 4. Aneurysm 5. Wall disruption 6. Perivascular abnormality 7. Intraluminal abnormalities a. Clot b. Gas

872

Diagnostic Abdominal Imaging

Table 15-2. Causes of Wall Thickening

Atherosclerotic plaque

The arterial wall is histologically composed of the intima, media, and adventitia. The arterial wall, although comprising 3 layers, is normally imperceptible on imaging examinations (see Figures 15-1 and 15-2). Pathologic conditions can occur in subintimal, intramural, and periadventitial locations (Table 15-2). Atherosclerotic plaque and mural thrombus are the principal causes of subintimal deposits and result in focal wall thickening. Calcification can be superimposed on either. Subintimal thickening tends to be irregular, noncircumferential, and discrete. Intramural thickening is thickening of the medial layer of the aorta and can be a result of edema and inflammatory expansion of the tissues or due to bleeding within the wall of the vessel. Inflammatory thickening of the media can be a result of vasculitis and bacterial infection (mycotic aneurysm). Intramural thickening typically is uniform, circumferential, and diffuse or segmental.

Atherosclerosis is the leading cause of death in the western world. Atherosclerosis is a chronic inflammatory process and is responsible in the vast majority of systemic arterial disease. It results from the combination of vessel wall injury, angiogenesis, and inflammation, leading to the deposition of lipid, fibrous tissue, and calcification in varying amounts, which constitutes the atherosclerotic plaque. The plaque affects both the lumen and the arterial wall, leading to luminal narrowing, stenosis, and occlusion. In addition, plaque can undergo ulceration, which can progress beyond the plaque and disrupt the intimal layer of the aorta leading to a penetrating atherosclerotic ulcer (PAU). Atherosclerosis is chronic and progressive although the rate of progression can be retarded by appropriate therapy, including lipid lowering and antiplatelet drugs and attention to certain risk factors. Atherosclerosis is a multifactorial phenomenon with a familial predisposition augmented by risk factors many of which are reversible or can be modified, such as smoking, hypertension, diabetes, serum lipid profile, obesity, and inactivity, and some which cannot be reversed such as male gender and age. Focal wall thickening due to both calcified and noncalcified plaque are easily visualized on contrast-enhanced CT and MRI. Noncalcified plaque typically appears as softtissue attenuation, plateau-shaped protuberance into the lumen of the vessel in question, and can extend over a few millimeters to many centimeters in length (see Figure 15-4). The surface can be smooth or irregular. Some plaques will contain foci of lipid attenuation/signal within them. Occasionally, plaque can be polypoid in morphology and attached to the wall with a narrow neck. This morphologic

A

B

Figure 15-4 Atherosclerotic Plaque A. Sagittal and (B) oblique-axial reformations show calcified and noncalcified plaque in the right external iliac artery in this

81-year-old female with rest pain. The plaques cause narrowing of the iliac vessels.

A. Subintimal 1. Atherosclerotic plaque 2. Mural thrombus B. Intramural and adventitial 1. Intramural hematoma 2. Vasculitis 3. Bacterial infection (mycotic aneurysm) 4. Aneurysm with a chronic leak 5. Graft repair of the aorta

Wall Thickening

Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis 873 feature predisposes the plaque to becoming dislodged during catheter angiography, leading to distal emboli and the potential for distal ischemia. Therefore, it is important for these lesions to be specifically noted to alert physicians performing catheter angiography of the potential for emboli. Much research has focused on detection of plaque vulnerability morphologically on CT and MR. Vulnerable plaque is plaque likely to rupture, leading to vessel thrombosis. Such plaques have a large amount of lipid core and a thin fibrous capsule. With improving spatial and lowcontrast resolution, in vivo histological characterization of plaque is becoming increasingly possible. As atherosclerotic plaque ages, it develops regions of dystrophic calcification. As a consequence, the extent of

plaque calcification increases with the age of the individual. Calcification can be detected on abdominal radiographs, US, and CT (see Figure 15-4). Calcification is not reliably detected on MRI.

Mural thrombus According to Virchow triad, thrombogenicity is increased when there is an abnormality of (1) the vessel wall, (2) blood flow, or (3) the nature of blood itself. Mural thrombus is the name given to the thrombus that forms in areas of arterial ectasia/aneurysm because of the altered flow patterns (see Figure 15-5). This thrombus is attached to the wall of the artery. Mural thrombus can also form in nondilated arterial

A

B

C

Figure 15-5 Abdominal Aortic Aneurysm With Mural Thrombus and Intimal and Neointimal Calcifications A and B. Axial CTA shows an infrarenal aneurysm of the aorta (A) and right common iliac artery (B). The peripheral, curvilinear calcification (arrows) marks the intimal layer of the aorta and iliac arteries. The low-attenuation material between the aortic wall and the contrast-enhanced lumen represents mural thrombus that is subintimal in location. The medial and adventitial layers of the

aorta remain imperceptible. Curvilinear calcification is a landmark for the intimal layer of the aorta in patients with atherosclerosis. However, chronic mural thrombus can also calcify, and this is known as neointimal calcifications, which is seen as small flecks of high attenuation within the mural thrombus (arrowhead). C. Volume-rendered image shows the distribution of contrast and calcification. Note how this underestimates the diameter of the aneurysm because of the presence of mural thrombus.

874

Diagnostic Abdominal Imaging

segments if the flow dynamics are altered, such as in areas of atherosclerotic plaque. As a consequence, plaque and thrombus frequently coexist. The imaging differentiation of plaque from thrombus can be difficult. Both mural thrombus and plaque tend to have an irregular surface. Thrombus tends to have a higher attenuation than plaque and will not have the areas of fat attenuation seen in plaque. As mentioned, plaque can ulcerate and the ulceration can form a fissure, which extends to the intimal layer, a feature that is not seen as frequently in mural thrombus but can happen. Like plaque, thrombus can occasionally have a polypoid morphology that is predisposed to dislodgement and distal embolization. As is evident from this discussion, the imaging distinction between plaque and mural thrombus is not always possible. Fortunately, the distinction between plaque and thrombus is usually clinically irrelevant.

Intramural hematoma Rupture of the vaso vasorum of the aorta and other great vessels will result in an intramural hematoma.1 Intramural hematomas most commonly occur spontaneously in association with systemic hypertension but can also be a result of penetrating atherosclerotic ulcers (PAUs).2,3 Spontaneous aortic intramural hematomas are typically discovered in elderly patients, with a mean age of presentation of 74 years and a range of 63 and 87  years.1,4,5 Systemic hypertension is seen in the majority of patients and many will have a variety of comorbidities such as chronic obstructive pulmonary disease (COPD), heart disease, and chronic renal insufficiency.1,5,6 Symptoms mimic an aortic dissection such as severe, midscapular, substernal, or anterior chest or back pain.1,6 On histologic examination, there is degeneration of the aortic media, and the hematoma will typically be present in the outer third of the media only a few cells from the adventitia.1 This is in comparison with aortic dissections, which will normally involve the more central aspects of the media, closer to the intima. The natural history of intramural hematomas is quite variable. Approximately one-half of medically treated intramural hematomas will regress within 6 months and will ultimately resolve within 1 year.1,7,8 Twelve percent to 24% will evolve into dissections and 13% to 30% evolve into aneurysms or and will usually require surgical or endovascular intervention.4,6,9 Some series have identified a 35% to 51% incidence of acute rupture of intramural hematomas either spontaneously or associated with progression to dissections or aneurysms.1,6 Involvement of the ascending aorta increases the likelihood of rupture.1,5,6,8,10-13 The majority of complications of intramural hematomas, approximately 90%, will happen within a few weeks of presentation but rarely, others will occur months to years later.6 The majority, approximately 70%, of intramural hematomas will occur in the descending thoracic aorta.1 Intramural hematoma of the abdominal aorta is rare in

isolation and is usually an extension of an intramural hematoma that originates in the thoracic aorta. On CT scans, intramural hematoma will appear as high attenuation thickening of the aortic wall on precontrast images.14-16 MRI will most often reveal increased signal intensity on T1-weighted images, indicating subacute blood.14,17 Following contrast injection, the attenuation of intramural hematoma will remain unchanged (see Figure 15-6). Intramural hematoma must be distinguished from mural thrombus, which can also contain blood products. Intramural hematoma tends to be circumferential, smooth and continuous whereas mural thrombus is irregular and discrete. Intimal calcification, when present, can aid in distinguishing mural thrombus from intramural hematoma. Calcification generally tends to be intimal/ subintimal. The consequence of this is that intramural hematoma tends to surround the calcification whereas mural thrombus is on the luminal side of the calcification.

Vasculitis Vasculitis is typically classified according to the size of vessel involved. Large vessel vasculitis, such as Takayasu and giant cell arteritis, affects the aorta and major branches. Medium vessel vasculitis, such as polyarteritis nodosa (PAN), affects the primary arteries to major organs such as the celiac artery and the SMA and their branches. Small vessel vasculitis, such as microscopic polyarteritis, Churg-Strauss vasculitis, and Wegener granulomatosis affects capillaries. In addition to the primary vasculitides, vessel wall inflammation can also be secondary to medications, infection, and malignancy. From the perspective of imagers it is best to think of vasculitides as those that are occult on cross-sectional imaging (small vessel vasculitis) and those that are seen on imaging (large and medium vessel vasculitis). In a patient suspected of vasculitis, a negative CTA or MRA does not rule out small vessel vasculitis. The primary imaging abnormality of medium and large vessel vasculitis is inflammatory thickening of the media.18-27 On precontrast CT and MRI images, the thickened media have an attenuation and signal intensity similar to or less than that of muscle. This is an important distinguishing feature from intramural hematoma, which should be of higher attenuation than skeletal muscle on CT and high signal on T1-weighted MR sequences. Further, following contrast administration, the thickened vessel wall on both CT and MRI may enhance, whereas intramural hematoma will have an unchanged attenuation. In the acute phase of the vasculitis, wall thickening is a result of mural edema, which can be detected by T2-weighted MRI as increased signal of the vessel wall or as a hypoechoic halo on sonography.26,28 As with intramural hematoma, the thickened vessel wall in vasculitis is smooth and the thickening is circumferential, features that help distinguish it from atherosclerotic plaque and most cases of mural thrombus. The thickening also tends to be more symmetrically circumferential in vasculitis than intramural

Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis 875

A

B

C

D

E

F

Figure 15-6 Intramural Hematoma in 2 Patients A. A crescenteric area of high attenuation can be seen in the ascending and descending aorta. This is the typical appearance of an intramural hematoma on unenhanced CT. B-F. MRI exam in a second patient. (B) Axial T1 through the abdomen shows high-signal crescent along the levolateral wall of the abdominal aorta suggestive of IMH. (C) Axial reformation of 3D gadolinium MRA does not show any enhancement of the area

of high signal. (D) Axial postgadolinium delayed image reveals 2 different signal intensities. The inner aspect of the crescent has a markedly low signal and is thought to represent calcified intima, which is displaced by an intramural hematoma. (E) Axial 3D pregadolinium and (F) postgadolinium shows high signal along the levolateral wall of the aorta, which does not enhance in keeping with an intramural hematoma.

876

Diagnostic Abdominal Imaging

hematoma. In comparison with intramural hematoma that is typically a single long focus of wall thickening adjacent to the site of primary injury, vasculitis tends to be uniformly diffuse across the entire extent of the involved vessel or multifocal with thickening of noncontiguous arterial segments/beds.18-25,28 Vasculitis tends to have luminal narrowing associated with the wall thickening (see Figure 15-7).

In addition to the primary abnormality, wall thickening, vasculitis can lead to the gamut of other vascular morphologic disturbances including stenosis, occlusion, aneurysm formation, and arterial rupture, which will be discussed later in this chapter. Further, vasculitis, like other systemic vascular pathologies, can present on imaging with end organ stigmata of ischemia and hemorrhage. These

A

B

C

D

Figure 15-7 Wall Thickening due to Takayasu Arteritis This 47-year-old man presented with lower extremity edema. A-C. Axial contrast-enhanced T1-weighted images through the abdomen demonstrate extensive periaortic soft-tissue enhancement (arrows). The caliber of the iliac arteries is small, and careful observation of the iliac arteries in (B and C) shows modest wall thickening. Note the low signal ring around the iliac

arteries bilaterally, which is believed to represent nonenhanceing intima. These features suggest an inflammatory process involving the abdominal great arteries, probably a large vessel vasculitis. D. Coronal MIP shows tight stenosis (arrowhead) of the proximal iliac arteries bilaterally. Further evaluation lead to a diagnosis of Takayasu arteritis.

Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis 877 findings are discussed in many other chapters as imaging features of the affected organ.

Takayasu arteritis: Takayasu arteritis is a rare idiopathic inflammatory disorder of the large vessels primarily involving the aorta, the aortic branch vessels, and pulmonary arteries.19,20,29 Takayasu arteritis has a worldwide distribution, but is seen most frequently in Asian populations.19,20,29 It is primarily a disorder of young and middle-aged women, with median age of presentation of 36  years (range 19-57  y).18 Chronic inflammation of the aortic wall can lead to arterial stenosis and occlusion or weakening of the wall and aneurysm formation.19,20,29-31 Stenosis is the more common manifestation of Takayasu arteritis; however, aortic aneurysm and subsequent rupture is the most common fatal complication.32,33 Histologic examination of the acute phases of the disease will demonstrate extensive inflammation and thickening of the media and adventitia.19,34 With disease progression, there is marked thinning of the media, disruption of elastic fibers and pronounced fibrotic thickening of the adventitia leading to vascular stenosis.19,35 Aneurysms appear to be a result of degeneration and weakening of the media.19,36 Some researchers suggest that the development of aortic aneurysms is most closely associated with elevations in blood pressure.37 Patients typically present with the insidious onset of vague systemic symptoms such as fever, malaise, and weakness. With progressive disease, arterial stenoses can result in diminished or absent peripheral pulses. It causes inflammation of the aortic wall, which can lead to stenosis, occlusion, and aneurysm formation.19,20,29-31 In general, CT and MRI examinations will demonstrate extensive wall thickening of 1-mm to 4-mm thickness in the acute phases of disease.18-25 Arterial-phase imaging and delayed-phase imaging following intravenous contrast administration will demonstrate extensive wall enhancement, in patients with active arteritis.18 In some patients, a thin low attenuation ring is seen between the enhancing lumen and the enhancing aortic wall and is thought to represent nonenhancing intima (see Figure 15-7).18 Further, CT and MRI can be of value in monitoring for active disease. Studies have shown decreases in vascular wall thickening on CT and MRI examinations following administration of corticosteroids.24,25 This is especially important because clinical markers of disease in some cases may fail to identify patients with histologically active disease.38 With progression of disease, regions of stenosis, aneurysm formation or both are common.18-20 The presence of wall thickening is associated with greater likelihood of aneurysm progression and rupture.37 In the late phases of disease, chronic inflammation can also lead to dystrophic calcification within the wall of the aorta. Vascular stenosis, aneurysm formation, and wall calcifications as a complication of Takayasu arteritis are discussed more completely in the subsequent section titled: Aneurysms.

Giant cell arteritis (temporal arteritis): Giant cell arteritis is a granulomatous arteritis of large vessels. It primarily involves the arteries of the head, especially the extracranial branches of the carotid artery, but can also involve the aorta and other aortic branch vessels.39 Giant cell arteritis is also called “temporal arteritis” because of predisposition to involving the superficial temporal artery. Giant cell arteritis is an autoimmune disorder resulting in T-cell and macrophage-driven inflammation of the media of large arteries.39 The triggers leading to development of giant cell arteritis are not certain but appear to include genetic factors (HLA–DRB1*04 alleles and others), infectious agents, and other environmental factors.39 Giant cell arteritis is the most common arteritis in Europe and the United States and primarily affects the elderly, with an average age of onset of 70 years, and it rarely presents in patients younger than age 50 years.39 It primarily affects individuals from northern European descent, especially those with Scandinavian heritage.39 Patients often present cranial symptoms such as headache, scalp tenderness, jaw claudication, and visual disturbances (double vision, visual loss, blindness) or systemic symptoms such as fatigue, weight loss, and malaise.40 Ocular ischemic complications are the primary source of chronic disability in patients with giant cell arteritis. There is an association with the rheumatologic condition polymyalgia rheumatica, and approximately 10% to 15% of patients with polymyalgia rheumatica will develop giant cell arteritis.39,40 The disease primarily causes wall thickening and occlusion of the branch arteries originating from the aortic arch but can occasionally involve any of the large vessels of the body. The superficial temporal, vertebral, ophthalmic, and posterior ciliary arteries are most commonly affected followed by the internal and external carotid arteries. The inflammation is typically not continuous and therefore regions of thickening can be interspersed with areas of normal wall thickness.41 With active inflammation, US examinations show a hypoechoic halo of wall thickening due to edema of the arterial wall.26 The sensitivity and specificity of this finding are not established. With fibrosis, the arterial wall becomes hyperechoic.42,43 In general, MRI examinations will demonstrate increased thickness of the wall of the affected vessels and will also demonstrate wall enhancement following gadolinium administration.27 In a study of 20 patients with suspected giant cell arteritis, MRI was able to identify 16 of 17 patients with disease.27 One incidentally discovered case of giant cell arteritis was on 18F fluorodeoxyglucose (FDG) positron emission tomography (PET)-CT imaging.44 This finding may suggest that PET-CT could be used to identify patients with active inflammation due to giant cell arteritis. The major features that distinguish giant cell arteritis from Takayasu arteritis are the age (elderly females), predisposition to involve the temporal arteries, dramatically elevated ESR (>100), and disease disproportionately centered on the branches of the aortic arch.

878

Diagnostic Abdominal Imaging

Imaging Notes 15-1. Differentiating Features of Arterial Wall Thickening Distribution

Surface

BPa

Aneurysm

Stenosis

Atherosclerotic plaque

multifocal

irregb

no

sometimes

sometimes

Mural thrombus

mutlifocal

irreg

yes

sometimes

sometimes

Intramural hematoma

diffuse

smooth

yes

rarely

rarely

Vasculitis

diffuse

smooth

no

often

often

Bacterial infection

diffuse

smooth

no

always

never

Inflammatory aneurysm

diffuse

smooth

no

always

never

Chronic leak

diffuse

smooth

no

always

never

Cause

aBlood products. bIrregular.

Polyarteritis nodosa: PAN is a necrotizing, medium vessel vasculitis due to autoimmune-mediated vessel wall inflammation. Because the primary abnormality of PAN is inflammation of the arterial wall, it is expected that this disease would also cause thickening of the walls of the affected arteries. However, because of the small size of these vessels, vascular wall thickening is difficult to appreciate. Imaging examinations will most often identify multiple aneurysms of the medium arteries supplying the abdominal viscera, especially the kidneys, bowel, liver, and spleen. PAN is discussed most completely under the heading: Aneurysm.

Mycotic aneurysm Infection of the wall of the great vessels will virtually always lead to inflammatory thickening of the vessel wall. However, vascular infections will virtually always result in an associated aneurysm of the vessel involved. Therefore, mycotic aneurysms will be discussed most completely in the subsequent section titled: Aneurysms.

Periadventitial Fat Stranding Fat stranding in the soft tissues surrounding an artery will in most cases indicate inflammation of the periadventitial soft tissue. In many cases, the stranding will indicate the presence of an inflammatory process in the adjacent vessel. In most cases, the differential diagnosis for periadventitial fat stranding will be similar to the inflammatory causes of wall thickening such as vasculitis and mycotic aneurysm. Chronic leak of an aneurysm and a chronic reaction to graft repair of an aneurysm are other causes of soft tissue stranding. Rarely periadventitial thickening can be due to the condition retroperitoneal fibrosis (RPF) (Table 15-3).

Retroperitoneal fibrosis (Chronic periaortitis, inflammatory AAA, perianeurysmal retroperitoneal fibrosis) The 3 rare retroperitoneal fibrosing conditions RPF, inflammatory AAA, and perianeurysmal RPF are now believed to be differing manifestations of the same disorder.45 It has been suggested that these can be grouped under the umbrella term chronic periaortitis. All 3 conditions are characterized by advanced atherosclerosis, thinning of the media, and chronic inflammation and fibrosis of the adventitia of the abdominal aorta and periaortic soft tissues.45 This inflammation and fibrosis can be associated with formation of an AAA or can strangle adjacent structures, most commonly the ureters. Inflammation and fibrosis in isolation has been called “retroperitoneal fibrosis,” but when associated with aneurysm formation has been called “inflammatory abdominal aortic aneurysms” and when spreading to involve adjacent structures has been called “perianeurysmal retroperitoneal fibrosis.” The majority of cases are idiopathic; however, occasionally disease can be associated with a variety of factors,

Table 15-3. Causes of Periadventitial Inflammation 1. Vasculitis 2. Bacterial infection (mycotic aneurysm) 3. Aneurysm with a chronic leak 4. Graft repair of the aorta 5. Retroperitoneal fibrosis

Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis 879 including drugs (methysergide, bromocryptine, B-blockers, etc), malignancies (Hodgkin and non-Hodgkin lymphoma, cancer of the prostate breast and stomach), infections (tuberculosis, histoplasmosis, actinomycosis), prior surgery (lymphadenectomy, colectomy, hysterectomy, aortic aneurysmectomy), and miscellaneous other disorders (amyloidosis).45 Most cases associated with malignancies and infections appear to be an exaggerated desmoplastic response to retroperitoneal lymph node involvement by the process. Some data have suggested that idiopathic RPF is an inflammatory reaction to antigens from atherosclerotic plaques of the abdominal aorta.46,47 However, more recent research has suggested that the idiopathic forms of chronic periaortitis are probably an autoimmune disorder. Findings suggesting an autoimmune etiology include frequent systemic symptomatology, frequent association with other autoimmune disorders, and preliminary evidence suggesting that autoantibodies to fibroblasts can be seen in up to one-third of patients with chronic periaortitis.45,48,49 Patients will most often complain of side, back, or abdominal pain and some will develop lower extremity edema because of compression of the vena cava.45 Constitutional symptoms such as fatigue, fever, weight loss, nausea anorexia, and myalgias will often precede the localizing symptoms. Elevations in acute phase reactants (erythrocyte sedimentation rate, C-reactive protein) are commonly discovered. It is seen more commonly in men with a frequency of 2:1 to 3:1.45 The mean age of presentation is between 50 and 60 years but cases have been reported in children and the elderly. There are no standardized criteria for the diagnosis of chronic periaortitis but the diagnosis is typically based on a typical CT or MRI appearance associated with characteristic symptoms and elevated acute phase reactants or based on the histologic evaluation of tissues.45 On CT and MRI examinations, RPF appears as an infiltrative soft tissue mass, typically surrounding the infrarenal abdominal aorta and iliac arteries and occasionally enveloping the ureters and inferior vena cava.45,50,51 Ureteral involvement can also result in ureteral obstruction. The ureters are typically drawn into the fibrotic mass and are, therefore, medially displaced, a finding that can be seen on intravenous urogram.51 The aorta is variably dilated or narrowed. On CT scans, the mass is typically isoattenuating with skeletal muscle.50,51 MRI examinations show the mass to be isointense to muscle on T1-weighted sequences. On T2-weighted sequences, the signal from the mass is variable depending on the stage of disease. In the early phases, the RPF will typically be hyperintense because of the presence of edema and inflammatory cells but late in disease will appear hypointense as a result of mature fibrosis.50,51 Inhomogeneous T2 signal has been associated with RPF due to malignancies.52 In comparison with lymphoma and other

retroperitoneal neoplasms, RPF will usually not elevate the aorta away from the spine. In general, FDG-PET and PET-CT scans have been used to assess for disease activity (see Figure 15-8).53

Wall Calcification Vascular calcification occurs at 2 distinct sites within the vessel wall: the intima and the media. Intimal calcification is a manifestation of atherosclerosis, whereas medial calcification can exist independently of atherosclerosis and is associated with aging, chronic renal failure, and diabetes mellitus.54 Rarely, Takayasu arteritis and syphilitic aortitis will also be a cause of large vessel calcification.

Atherosclerosis and intimal calcification Atherosclerosis results in focal calcification of the intima. Intimal calcification only occurs within atherosclerotic plaques and can be seen pathologically as early as the second decade of life but is typically first discovered on imaging examination of patients older than age 50 years.54 Calcification appears to be a result of apoptotic cell death of intimal vascular smooth muscle cells, the presence of lipids and lipoproteins within the intima of the vessel, and mechanisms that increase calcium and phosphate extracellular concentrations.54 Calcified plaque typically appears as high-attenuation, plateau-shaped protuberance into the lumen of the vessel in question on CT scans and can extend over a few millimeters to many centimeters in length (see Figures 15-4, 15-9, and 15-10). Calcification typically occurs at the base of the plaque. Abdominal plain films can also demonstrate vascular calcifications as linear opacities in the distribution of the aorta and iliac vessels. On US examinations, atherosclerotic plaque calcifications will be seen as areas of shadowing along the surface of the aorta.

Aging, end-stage renal disease, diabetes mellitus, and medial calcification Unlike atherosclerosis, aging, end-stage renal disease (ESRD), and diabetes mellitus cause calcification of the media of the vascular wall.54,55 Medial calcification histologically appears as linear deposits along the elastic lamina and at its most severe can form dense circumferential calcification of the vessel wall. Atherosclerotic calcifications typically involve the large and medium arteries of the body such as the aorta and its main branch vessels. Whereas medial calcification will typically involve the small branch arteries of the periphery of the abdominal organs or lower extremities. This phenomenon is most commonly a complication of diabetes mellitus and chronic renal failure.56 It also occurs in the small vessels of the foot and lower legs in otherwise healthy elderly patients, where it is called Mönckeberg sclerosis. Medial calcifications are associated with an increased incidence of neuropathy and

A

B

C

D

E

F

G

H

Figure 15-8 Retroperitoneal Fibrosis This 52-year-old woman presented with renal insufficiency. A-C. Unenhanced CT shows bilateral hydroureteronephrosis and a poorly defined soft tissue mass (arrows) surrounding a heavily calcified aorta and iliac arteries. D. T1-weighted, (E) T2-weighted, and (F) postgadolinium MRI also shows a vague retroperitoneal soft-tissue mass, with intermediate T1, high T2 signal that moderately enhances. G. Longitudinal and (H) transaxial US also shows amorphous soft tissue surrounding the aorta and a diffusely thickened aortic wall. Biopsy was diagnostic of retroperitoneal fibrosis without evidence of malignancy.

Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis 881

A

B

C

D

Figure 15-9 Intimal and Neointimal Calcifications in 2 Patients A. Axial unenhanced CT in the first patient shows an abdominal aortic aneurysm with both peripheral “rim” calcification and nonperipheral calcification. Rim calcification indicates intimal calcification. Nonperipheral calcification can represent displaced intimal calcifications or neointimal calcification, where neointimal calcification represents calcification of thrombus. B-D. CT images in a second patient with an abdominal aneurysm demonstrates similar peripheral and nonperipheral calcifications. The peripheral rim calcification are intimal

calcification. In this case, we can prove that the nonperipheral calcifications are calcifications of thrombus, ie, neointimal calcification. Careful observation of (B) shows there to be a region of decreased attenuation surrounding the nonperipheral calcification, which is confirmed by the attenuation measurements in(C). Thrombus has a lower attenuation than blood in the lumen on unenhanced CT. Contrast-enhanced image (D) confirms the presence of thrombus within the lumen of the aneurysm with several areas of neointimal calcification.

are associated with increased cardiovascular mortality in patients with diabetes.57,58 Calcifications are typically identified in small and medium arteries throughout the body and can be seen in the extremities on x-rays of the hands, forearms, feet, and lower leg as paired linear opacities, which often demonstrate a branching pattern. Medial calcifications can also be seen in small vessels of the abdominal organs such as the kidneys, liver, and spleen on CT scans and appear as smooth, usually continuous, linear calcifications within the walls of small vessels with the involved organ (see Figures 15-11 and 15-12).59 Small vessel calcification in most cases will indicate the presence of diabetes mellitus or chronic renal failure.

Vasculitis Calcification as a manifestation of vasculitis is relatively uncommon, but is a known phenomenon in patients with Takayasu arteritis.

Takayasu arteritis: As noted previously, Takayasu arteritis is an uncommon idiopathic inflammatory condition of the large arteries of the body, seen predominantly in young and middle aged women. The most characteristic features of Takayasu arteritis are wall thickening, vascular stenosis, and aneurysm formation. However, as the disease becomes quiescent, the walls of involved vessels can develop extensive wall calcification.18 Histologically, the calcification begins within the intima of the diseased vessel but can

882 Diagnostic Abdominal Imaging

B

A

C Figure 15-10 Calcified Atherosclerotic Plaque in 2 Patients A. Abdominal radiograph shows tubular calcifications outlining the wall of the abdominal aorta, iliac arteries, and femoral arteries (arrows). In most cases, this will represent calcification

of atherosclerotic plaque. B and C. Axial images of CTA through the abdominal aorta in a second patient shows calcification of the intima and noncalcified plaque deposited subintimally.

extend diffusely throughout the aortic wall. Diffuse calcification is not seen in patients with atherosclerosis and may be a specific finding of Takayasu arteritis.18 Further, atherosclerosis is predominantly a disease of the elderly. The presence of extensive aortic calcification in a patient younger

than age 40 years can suggest a diagnosis of Takayasu arteritis (see Figure 15-13). Some authors believe that the presence of mural calcification is protective against aneurysm formation.37 They have found that most aneurysms form in regions of the aorta relatively devoid of wall calcification.

A

B Figure 15-11 Small Vessel Calcifications on an Abdominal Radiograph This 36-year-old woman had childhood diabetes mellitus and chronic renal failure. Magnified views of the (A) upper abdomen

and (B) pelvis from an abdominal radiograph demonstrate extensive vascular calcifications of the small arteries of the kidneys (arrows in A) and the branches of the internal iliac arteries (arrows in B).

884

Diagnostic Abdominal Imaging

A

B

C

D

Figure 15-12 Small Vessel Calcification This 77-year-old woman had both diabetes mellitus and chronic renal insufficiency. A-D. Axial CT images demonstrate multiple intermediate and small vessel calcifications of the splenic artery

(arrow in A), the renal arteries (arrow in B), the SMA (arrow in C), and branches of the external iliac (arrow in D) and uterine arteries (arrowhead in D). Small vessel calcifications are often a manifestation of diabetes mellitus, renal dysfunction, or both.

Stenosis and Occlusion

Patients will typically present with symptoms of end organ ischemia such as abdominal pain, renal insufficiency, and claudication. In general, CTA and MRA will show narrowing of the involved vessel. Percentage of luminal narrowing is a predictor of end organ ischemia, with studies in some arterial beds suggesting that diameter narrowing greater than 70% is likely to be associated with hemodynamic compromise.39 However, there is no universally accepted dividing line largely because each vascular bed is unique and there may be varying degrees of contribution from collateral vessels (Table 15-4). An abrupt transition from enhanced vessel to unenhanced vessel, the meniscal sign, at the site of occlusion suggests the presence of acute thromboembolism whereas a smooth tapered narrowing will typically indicate stenosis

The progressive narrowing of an arterial lumen produces end-organ ischemia. The most common cause of arterial narrowing is atherosclerosis. Nonatherosclerotic entities such as vasculitis and connective tissue disease should be considered in the younger individuals, in the absence of findings of atherosclerotic involvement, and with certain patterns of wall thickening or intimal disruption. Occasionally, vascular stenosis can be a result of external compression of an artery. This is best typified by the arcuate ligament syndrome. Progressive narrowing will eventually occlude the artery. An artery can also be acutely occluded by an embolus or, more rarely, in situ thrombus. It is important to distinguish between an acute occlusion and chronic narrowing, as the former requires emergent intervention.

Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis 885

A

B

C

D

Figure 15-13 Abdominal Aneurysm due to Takayasu Arteritis This 34-year-old man had been diagnosed with Takayasu arteritis for several years. A and B. Contrast-enhanced axial CT images and MIP projection images in the (C) coronal and (D) sagittal planes show dilation of the infrarenal abdominal aorta and both

internal iliac arteries (arrows). There are multiple plaquelike areas of calcification within the walls of the great vessels. In an elderly individual, this would most often be a result of degenerative aneurysms but in this young man are due to Takayasu arteritis.

(see Figures 15-14 and 15-15). The presence of enlarged collateral vessels will indicate a more long-standing stenosis. It can be difficult to distinguish between acute and chronic narrowing in cases of in situ thrombosis where there is a background of chronic narrowing with ischemia. In the presence of potential arterial compromise, the most important point to ascertain is the presence or risk of end-organ ischemia, in other words, whether the lesion is hemodynamically significant. Imaging features of end-organ ischemia on imaging ranges from asymmetry in the degree

Imaging Notes 15-2. Arterial Occlusion • Where possible, a distinction between an acute occlusion and chronic occlusion of the arteries should be made • The presence of collaterals suggests a chronic process • The surrogate of the significance of arterial compromise is the organ itself

886

Diagnostic Abdominal Imaging

A

B

C

D

Figure 15-14 SMA Embolus This 62-year-old woman with the recent onset of atrial fibrillation complained of 2 days of abdominal pain. A. Axial CT image through the midabdomen demonstrates thickening of the wall (large arrowheads) of the ascending and transverse colon. In this clinical setting, the possibility of ischemic colitis should be considered. B and C. Axial CTA confirms contrast

in the proximal SMA (small arrowhead) but absence of contrast in the distal SMA (arrow). D. This abrupt transition between contrast-enhanced (arrowhead) and unenhanced (arrow) SMA is seen better on this sagical reconstruction and indicates the presence of an acute embolus (arrow) to the SMA causing colonic ischemia.

of vascular enhancement of the organ to an infarct whose morphology can take the classical wedge shape with the apex pointing centrally. The bowel shows some sign with ischemia/infarction with varying specificity. These include wall thickening with mural stratification, narrowing,

pneumatosis coli, mesenteric venous air, and portal venous air. Although these findings are individually not specific for ischemia, the distribution, in particular of the wall thickening and location of the luminal narrowing, can be highly suggestive of an ischemic process if it conforms to

Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis 887 Table 15-4. Causes of Vascular Stenosis or Occlusion 1. Atherosclerosis 2. Emboli 3. Vasculitis 4. Connective tissue disease 5. Spontaneous thrombosis

a vascular territory. Features of end-organ ischemia have been discussed in detail in prior chapters under the organ of interest.

Atherosclerosis Atherosclerosis is a chronic degenerative process causing the proliferation of smooth muscle cells that results in the production of fatty intimal lesions known as atherosclerotic plaques.60 Plaques and plaque complications such as plaque rupture and associated arterial thrombosis can cause acute or chronic narrowing of arteries. As a result, many of the complications of atherosclerosis are a consequence of arterial stenosis and subsequent tissue ischemia and infarction. Both calcified and noncalcified plaque can lead to narrowing of an artery, creating flow-limiting stenosis and occlusion.

The degree of atherosclerosis does not usually correlate with the degree of narrowing. This is because plaque can cause wall thickening without concomitant luminal narrowing, an observation known as the Glagov phenomenon.61 Stenosis due to atherosclerosis is identified by the presence of calcified and noncalcified plaque within the stenotic vessel. As noted previously, plaque typically appears as a calcified or noncalcified, plateau-shaped protuberance into the lumen of the vessel in question and can extend over a few millimeters to many centimeters in length (see Figure 15-16). In addition to in situ narrowing and occlusion, atherosclerotic plaque can be a source of emboli. Friable plaque can be located in the descending aorta or the aortic arch. The presence of plaque ulceration is an important marker of an increased risk of embolization. Emboli as a cause for vascular occlusion are discussed later in this section.

Takayasu arteritis One of the important complications of the large and medium vessel vasculitides is the development of vascular stenosis leading to end-organ stenosis. Vascular stenosis is the most common complication of Takayasu arteritis. Histologically, the chronic inflammation of Takayasu arteritis leads to marked thinning of the media with disruption of elastic fibers and pronounced fibrotic thickening of the adventitia.19,35 These changes result in

A

B

Figure 15-15 Stenosis due to Takayasu Arteritis A. Sagittal MPR from CTA shows multisegment narrowing (arrows) of the abdominal aorta and branch vessels culminating in occlusion (arrowhead). B. Axial image at the level of the left

renal vein shows an unidentifiable abdominal aorta (arrow) but a vascular structure anterior to the left renal vein (arrowhead) which is an extra-anatomical bypass graft connecting the proximal aorta to the prebifurcation abdominal aorta.

888 Diagnostic Abdominal Imaging

A

B

C

D

Figure 15-16 Stenosis due to Atherosclerosis in 2 Patients This 75-year-old woman had systemic hypertension. A. Maximum intensity projection (MIP) image from a CTA shows an approximately 50% ostial stenosis of the right renal artery (arrowhead) and what appears to be complete occlusion of the left renal artery (arrow) caused by large atherosclerotic plaques. B. Source axial image through the left renal artery shows a tight

stenosis (arrow) but not occlusion due to aortic and renal artery plaque. C. Sagittal MIP through the SMA also shows plaque (arrow), causing an approximately 50% ostial stenosis. D. Coronal MIP in a 75-year-old man with claudication demonstrates a long noncalcified stricture of the left renal artery (arrow) due to atherosclerosis. He also had strictures of his femoral arteries (not shown), which were the cause of his claudication.

narrowing of the lumen of the vessel. The disease is seen primarily in young and middle-aged women, with median age of presentation of 36 years (range 19-57 y) and will usually present with systemic symptoms of fever, malaise, and weakness.18 Arterial stenoses can also cause diminished or absent peripheral pulses and symptoms of ischemia in the affected territory (see Figures 15-7 and 15-16).

arterial wall that most commonly affects the renal and common carotid arteries.62 The etiology of fibromuscular dysplasia remains a mystery. However, the disease has an increased incidence in individuals who are current or former smokers and in those with systemic hypertension.62 An increased incidence of fibromuscular dysplasia in firstdegree relatives suggests that there are heritable factors in the development of disease.62 Fibromuscular dysplasia has also been associated with Ehlers-Danlos syndrome type IV. Fibromuscular dysplasia is subclassified depending on involvement of the layers of the vessel, the media, intima or adventitia.62 Medial fibromuscular dysplasia is the most common type accounting for nearly 80% of all cases of fibromuscular dysplasia and is further subclassified into medial fibroplasia, perimedial fibroplasia, and medial hyperplasia. Medial fibroplasia is the most common

Connective tissue disease A variety of connective tissue diseases can have vascular manifestations. The most common connective tissue disease that is associated with abdominal arterial stenosis is fibromuscular dysplasia.

Fibromuscular dysplasia: Fibromuscular dysplasia is a nonatherosclerotic, noninflammatory disorder of the

Chapter 15 Imaging of the Arteries and Veins of the Abdomen and Pelvis 889 subtype of medial fibromuscular dysplasia and is characterized by a hypertrophied media and a thin or absent internal elastic lamina. Weakening of the wall leads to multiple adjacent aneurysms in medium-sized arteries and causes the classic “string of pearl” appearance on imaging examinations.62 In perimedial fibroplasias, the outer layer of the media undergoes fibrosis. This phenomenon primarily causes stenosis without aneurysm formation. The rarest subtype is the medial hyperplasia and appears as smooth concentric stenosis of the involved vessel. Intimal fibroplasia accounts for