Paul & Juhl’s Essentials of Radiologic Imaging
7th Edition

Chapter 20
The Urinary Tract
Fred T. Lee Jr.
John R. Thornbury
F. T. Lee, Jr. and J. R. Thornbury: Department of Radiology, University of Wisconsin Hospital and Clinics, Madison, Wisconsin 53792-3252.
Introduction of new methods and enhancements of traditional radiologic methods have greatly influenced the use of imaging to diagnose and treat patients who have urinary tract disease. In the past, plain films of the abdomen and excretory urography (EXU) were the starting point in the diagnostic imaging process. Today, computed tomography (CT), ultrasonography, or magnetic resonance imaging (MRI) may be requested initially. Choosing the appropriate method has become more complex because of the variety that confronts the physician.
If physicians think critically about the selection of patients before requesting an imaging examination, they can improve their use of such examinations.60 First, the physician must hypothesize a differential diagnosis. Then, based on personal experience and knowledge of the literature and before deciding whether to request the examination, the physician should answer two questions: (1) Is this examination going to affect my diagnostic certainty about the differential diagnosis I am considering, and, if so, how much? (2) Will the information I expect to receive from this examination change my diagnostic thinking enough to affect my choice of treatment? Particularly important is the action of linking the use of the diagnostic test to the choice of treatment.
The following paragraphs present the most frequently used (or most useful) examinations for the specific diagnostic problem situations that are discussed subsequently.
Plain-Film Roentgenography
Roentgenographic examination of the urinary tract may begin with a plain film of the abdomen, exposed with the patient in a supine position, that includes the kidneys and the ureteral and bladder areas. This “scout” film, which must be obtained before contrast medium is given for EXU, reveals the renal shadows and permits assessment of the size, shape, and position of the kidneys. The presence of calcium in cysts, tumors, or stones can be detected along with vascular or lymph node calcifications in the area. Psoas muscle shadows usually are well outlined, and asymmetry or other abnormalities can be noted. The ureters cannot be defined, but radiopaque calculi may be detected along the course of the ureter. The shadow cast by the urinary bladder often can be identified. Vesical calculi can be outlined. Vascular calcifications, including phleboliths and arterial plaques, frequently are seen in the pelvis and must be differentiated from urinary calculi. Doing so may require other examinations, such as intravenous pyelography, CT, or ultrasound.
Excretory Urography
Preparation of the Patient
EXU, variably known an intravenous pyelography (IVP) or intravenous urography (IVU), is a frequently used imaging examination for general assessment of the urinary tract, which requires intravenous injection of radiopaque contrast material. Serial films are then obtained over 15 to 25 minutes as the contrast agent is excreted by the kidneys for visualization of the renal collecting systems, ureters, and bladder. Patient preparation before an elective examination often involves bowel cleansing, with use of cathartics such as castor oil, senna preparations (X-Prep), or bisacodyl (Dulcolax). Catharsis is particularly helpful in bedridden patients to remove gas and fecal matter from the colon, both of which obscure the renal areas. In ambulatory patients, gas and feces are not as much of a problem.
Many variations of the approach just described are in use. Satisfactory urography often can be obtained with no preparation, particularly in ambulatory outpatients. There are situations in which adequate hydration is important. In patients who have multiple myeloma, renal failure, or insulin-dependent diabetes mellitus (IDDM), and in those who are critically

ill (including neonates), the preparation is altered to fit the patient’s needs and avoid dehydration.
Contrast Material
The contrast media are organic iodides that depend on their iodine content for radiopacity. Currently there are two types of contrast material in use: ionic and nonionic. The former, represented by diatrizoate- or iothalamate-based media, has been standard for more than 40 years. In the early 1980s, a low-osmolar ionic agent, ioxaglate, was introduced for intravascular use.120 Nonionic contrast media of lower osmolality were introduced in Europe for general use in the late 1970s. In 1986, after the initial European experience and subsequent testing in the United States documented much lower toxicity with decreased reaction rates (including deaths), the US Food and Drug Administration approved two new nonionic media, iopamidol and iohexol, for intravascular and myelographic use. In 1996, the first low-osmolar radiopaque dimer was introduced (iodixanol). These compounds are virtually iso-osmolar to blood and are well tolerated by patients during rapid intravenous injection.
Clinical experience58, 94 indicates that the nonionic media have overall reaction rates about one third to one fourth those of the ionic materials (3.13% versus 12.66%). Severe reactions are reported in 0.22% of patients given ionic agents and in 0.04% of those given nonionic agents. The death rate generally considered representative by most experts for traditional ionic contrast agents is estimated to be about 1 in 40,000.4 Estimates for the nonionic media are as low as 1 in 168,000.94
Given the favorable statistics for nonionic contrast media versus traditional ionic materials in terms of adverse reaction rates, why do some institutions continue to use ionic agents? The answer is complex and involves local issues, but the choice generally is driven by the greater cost of the new, nonionic agents. For most institutions, a selective-use policy for nonionic agents can represent a significant cost savings at a locally acceptable reaction rate. Nonionic agents are reserved for patients who qualify under the established American College of Radiology (ACR) criteria, including previous reactors, patients with a history of allergy or asthma, those with known cardiac dysfunction, and those who are severely debilitated.35 Nonetheless, some institutions have continued a universal nonionic contrast material policy for all intravenous injections. As the cost of nonionic agents decreases as a result of competitive factors, more institutions are likely to convert to universal use.54, 144
For ionic contrast agents, a meglumine diatrizoate medium is often used in a dose of 0.5 mL per pound of body weight. This method delivers 0.34 mg of iodine per kilogram of body weight, which is usually a satisfactory dose for patients with reasonably good renal function. These contrast agents are excreted almost entirely by glomerular filtration with very little, if any, tubular resorption. In children, dose ranges have been recommended that are based on body surface area.46 The upper limit of dose for a contrast agent providing 300 mg/mL of iodine is 4 mL per kilogram of body weight for infants weighing less than 2.5 kg. The dose may be decreased in thin patients and increased in obese patients. A nomogram for dose determination has been provided by Diament and Kangerloo.46
In premature and newborn infants, relatively larger doses of contrast media are required (up to 4 mL per kilogram of body weight) because of the relatively decreased ability of the kidney to concentrate contrast agents in this age group. Currently, however, ultrasonography is the most common modality used initially to evaluate abdominal masses in premature and newborn infants.
Contraindications to injection of intravenous contrast material include (1) hypersensitivity to the contrast agent, (2) the presence of combined renal and hepatic disease, (3) oliguria, (4) a serum creatinine level higher than 2.5 to 3.0 mg/100 mL, (5) IDDM in combination with renal insufficiency (serum creatinine greater than 1.5 mg/dL), (6) multiple myeloma (unless the patient can be kept well hydrated during and after the study), (7) history of severe allergy, and (8) use of the oral hypoglycemic agent metformin (Glucophage) within the previous 48 hours. Patients who are using metformin are at risk for severe lactic acidosis if they are put into renal failure, and, given the high mortality that results (approximately 50%), most authors recommend stopping metformin for 48 hours before elective injection of contrastmedia. Emergency studies should be weighed on a case-by-case basis, and if the decision is made to proceed, renal function should be monitored for 48 hours before the patient is restarted on the drug. All of these contraindications are relative, and the value of potential information to be obtained must be weighed against the risk in each patient.
Adverse Reactions to Contrast Material
Iodinated, intravenously injected contrast media can produce reactions of varying severity.4, 89, 130 Minor reactions are the most common, having an incidence of about 5% to 10% for injections of ionic agents. The incidence of minor reactions to nonionic agents is less by a factor of approximately 6.94, 120 The most common signs and symptoms are urticaria, itching, nausea, and vomiting. These are usually self-limited, but on occasion antihistamine treatment may be required for more comfortable recovery. Minor reactions are more common in patients who have a history of allergy. It is not clear whether a history of minor reaction to a contrast medium causes a patient to be at significantly greater risk for a life-threatening major reaction from a subsequent injection.
Severe major reactions are rare, with a reported incidence ranging from about 3 to 5 times the death rate. Reported death rates for ionic agents range from about 1 in 30,000 to 1 in 75,000. One study of nonionic agents reported an even lower rate of severe reactions, 0.045%.94 Major reactions most often feature sudden onset of cardiovascular collapse,

which can rapidly progress to cardiac arrest if not promptly and successfully treated. Less frequently, respiratory system collapse or central nervous system disaster is the predominant feature initially and also can progress rapidly to death.
The precise mechanism of these major reactions remains obscure. They clearly are not classic antigen-antibody–type allergic reactions. There are no pretesting methods (e.g., use of an intravenous test dose) that can identify patients who are likely to have a major reaction.57 However, some types of patients seem to have an increased risk of severe reaction and death. Such patients include those with (1) prior severe reaction to contrast media, (2) asthma, (3) severe cardiac or renal disease, (4) hyperviscosity conditions (e.g., macroglobulinemia, multiple myeloma), (5) advanced dehydration, or (6) anxiety states.102
The radiologist should have a plan for response to a serious reaction. The equipment, the medication, and the personnel trained to manage severe reactions must be immediately available whenever and wherever contrast media are injected. If a major reaction appears to be developing, treatment should begin at once and an emergency call for skilled help should be made. Maintenance of an adequate airway is essential, and oxygen should be given in all major reactions. Table 20-1 indicates the various types of reaction and the methods of treatment along with the representative drugs and their doses.
Table 20-1. Treatment of acute reactions to contrast media
The prophylactic use of drugs to decrease reaction rates in high-risk patients is becoming more widely accepted. A blinded, randomized study demonstrated the protective effect of 32 mg of methylprednisolone given at 12 hours and at 2 hours before contrast administration, as well as the lack of protection from a single dose given 2 hours before injection.105 One regimen that yields a low reaction rate (0.5%) in patients who have had a previous adverse reaction includes the use of a nonionic contrast agent, administration of 50 mg of prednisone at 13, 7, and 1 hour before intravascular contrast administration, and administration of 50 mg of diphenhydramine 1 hour before contrast injection.68 It is important to realize, however, that prophylactic steroid treatment does not guarantee nonreaction.31, 105
Radiopaque contrast media also have a potential toxic effect on the kidney. The risk is very low in normal, healthy, young adults but begins to increase slightly in elderly patients with otherwise normal (for age) renal function. By age 65, about one fourth of normally functioning nephrons have been destroyed by the aging process. IDDM; chronic renal parenchymal disease (e.g., glomerulonephritis); shock from trauma or sepsis; renal ischemia; and other clinical problems with renal components (e.g., heart failure) considerably increase the risk of contrast-induced renal failure.119 Such risk factors must be balanced against expected diagnostic value when one decides to request urography.
Ultrasonography has replaced EXU in the imaging of patients with renal failure to exclude hydronephrosis.93 Renal failure caused by obstruction implies that both kidneys are blocked, because a unilateral obstruction would leave a functioning contralateral kidney. Obviously, unilateral obstruction can cause acute renal failure in the presence of a poorly functioning or absent contralateral kidney. Ultrasonography is very sensitive and specific in detection of chronic hydronephrosis. However, serial examinations may have to be performed to exclude acute hydronephrosis, because dilatation of the renal collecting system may take up to several days to develop after acute obstruction. Renal size and cortical thickness also can be evaluated by ultrasonography. These parameters are important to the clinician, because normal-sized kidneys may have a reversible cause, whereas small kidneys usually imply more chronic disease. The echogenicity of both kidneys is diffusely increased in medical renal disease, regardless of the specific cause, although the sensitivity of this finding has been questioned.150
Nuclear medicine renal scintigraphy is also valuable in the workup of patients with renal failure. Renal blood flow, acute tubular necrosis, glomerular filtration rate, renal function, and obstructive or reflux nephropathy can all be evaluated with the appropriate radionuclide. When the cause and level of obstruction are still unclear, retrograde pyelography may be necessary; this requires cystoscopy for ureteral catheter placement.
Technique of Examination
EXU (IVP) requires injection of intravenous contrast material. After injection, it is desirable to obstruct the ureters slightly with an abdominal compression band. Doing so retains some opacified urine in the kidneys and produces better visualization of the pelvis and calyces. Compression is not advised in patients with urinary obstruction (e.g., suspected acutely impacted ureteral calculus). In patients with known aortic aneurysm, the use of compression is unwise.
The first roentgenogram is obtained about 1 minute after injection, and a second is obtained at 5 minutes. The decision about compression is made on review of the 5-minute roentgenogram. The compression band is then applied, and small frontal and oblique views of the kidneys are obtained at approximately 10 minutes. The compression band is released, and a postrelease view of the abdomen is taken at about 15 minutes. We use tomography on most patients, usually between 1 and 5 minutes after injection of a bolus of contrast medium. A minimum of three tomographic cuts are obtained at 1-cm intervals. Additional tomograms are obtained as needed, and each film is monitored as the examination progresses.
Alterations in the procedure are made when indicated. Additional exposures of the bladder area may be necessary in some instances. We obtain a postvoiding film of the bladder in patients older than 40 years of age and in those with incontinence. Oblique films are of value in patients with suspected ureteral calculi and in those in whom questionable calyceal abnormalities are observed on anteroposterior films. When it is important to visualize the ureters, a film with the patient in right or left prone oblique position is useful, because

the ureters fill better in the prone than in the supine position.
If excretion of contrast material is delayed, it may be necessary to obtain roentgenograms for periods up to several hours after injection. In patients with acute ureteral obstruction (e.g., a ureteral calculus being passed), there is often a delay in excretion on the involved side. Delayed roentgenograms may show opacification of the renal pelvis and ureter down to the level of obstruction even when the immediate roentgenograms reveal only increased density of the renal

parenchyma. In patients with decreased renal function, preliminary tomograms are necessary to provide a baseline with which to compare the faint opacification of renal parenchyma (nephrogram) or calyces and pelves. Calyceal appearance time is normally within 3 minutes after the injection of contrast medium. Several variations of this technique are used, such as hypertensive urography (see Renovascular Hypertension).
The kidneys usually can be imaged adequately with real-time, high-resolution sector scanners. The right kidney can be scanned with the patient in a supine or decubitus position (left side down) with longitudinal, transverse, and coronal images. Similarly, the left kidney can be imaged with similar views and the right side down. Occasionally, a prone position may prove useful. The best images are obtained with the patient’s respiration suspended; frequently, the end of partial or full inspiration brings the kidney into better view. A 5-MHz transducer is preferable to optimize resolution. If this does not provide adequate acoustic penetration, a 3.5-MHz transducer can be used. Duplex and color Doppler technology can be used to assess renal vasculature. This is particularly important in the evaluation of the renal transplantation patient (see later discussion).
In cases of hydronephrosis, it may be difficult to follow dilated ureters to the point of obstruction because of overlying bowel gas. Varying the patient’s position may provide a better view along with imaging in coronal, longitudinal, or oblique planes. The distal ureters often can be visualized through a full bladder. The distended bladder is optimally viewed from a suprapubic approach with the patient supine. Transrectal transducers also can be used, especially to view the bladder neck, distal ureters, and prostate. When the suprapubic approach is used, the gain should be adjusted so that there are no echoes within the bladder.
Computed Tomography
CT examination of the kidneys is tailored to the specific clinical indication. In general, there are three broad categories in which CT examination is used, and the examination is different for each of the three.
The most common indication for CT of the kidneys is a morphologic examination for potential renal mass after an ultrasound, IVP, or other examination. This protocol can also be applied when searching for potential renal trauma or infection. CT examinations for these indications should include an initial noncontrast scan through the kidneys with narrow collimation (3 to 5 mm). This is necessary to search for calcifications, blood, and fluid collections before administration of contrast material. After the kidneys have been imaged by noncontrast scans, a bolus of contrast material (100 to 150 mL of 60% iodinated contrast) is administered, and the kidneys are rescanned. We start our contrast-enhanced sequence at the diaphragm, scan down to the kidneys with 1-cm collimation, scan the kidneys with 5-mm collimation, and proceed through the remainder of the abdomen and pelvis by 1-cm increments. The advent of helical CT now allows the kidney to be scanned at multiple time intervals if desired, including the so-called cortical phase (40 seconds after contrast injection) and the medullary phase (160 seconds after contrast injection).34
Another indication for CT of the abdomen and pelvis is suspicion of urinary tract stones; this examination is called CT urography.172 CT urography has been shown to be highly accurate in the evaluation of suspected ureteral and renal stones in the setting of acute flank pain. In addition, this technique often can determine a source for the patient’s pain when renal calculi have been excluded. We perform CT urography with no oral contrast material and no intravenous contrast material. The patient is scanned from the top of the kidneys through the bladder using helical technique: 2:1 pitch and 5-mm collimation during several breath holds.
Another indication for CT of the kidneys is CT angiography for visualization of renal vasculature. Although MRI is probably superior for the evaluation of renal artery stenosis (RAS), this examination has great utility in the workup of renal transplant donors and has now replaced preoperative renal angiography with lower morbidity and substantial cost savings. In CT angiography for renal donors, noncontrast sections are first obtained through the kidneys to rule out renal calcifications. A large bolus of contrast material (150 mL) is then given rapidly (5 mL/second) through an antecubital vein, and the kidneys are rapidly scanned using helical technique.161 Axial source images and two-dimensional and three-dimensional reconstructions are filmed depending on the local surgeon’s preference. An immediate post-CT plain abdominal film is obtained to evaluate the renal collecting system, ureters, and bladder.
The bladder can be evaluated by using the low attenuation of urine as an in situ contrast agent. To obtain better distention of the bladder, a Foley catheter can be inserted and sterile saline, water, or contrast material introduced. Alternatively, air or carbon dioxide can be introduced through the Foley catheter with the patient placed in a position so that the air is adjacent to the region of abnormality. Intravenous contrast material should still be given at some point to identify the ureters.
Retrograde Pyelography
Retrograde pyelography, another method employed in the examination of the upper urinary tract, is generally used when the EXU has been unsatisfactory or inconclusive for visualization of the renal collecting system and ureters. Cystoscopy and catheterization of the ureters are necessary for this examination. Roentgenograms are obtained after direct instillation of contrast material into the pelves through the catheters. Roentgenograms are obtained after 3 to 5 mL of contrast material (intravenous contrast agent diluted to 20% to 30%) are slowly introduced through the ureteral catheter into the renal pelvis. The catheters are withdrawn, and another

roentgenogram is obtained. Oblique views and delayed frontal views also may be necessary in some patients. The contrast medium may be injected by syringe or introduced by gravity with the vessel containing the medium no higher than 45 cm above renal level.
Care should be taken to avoid overdistention of the collecting system, because the high pressure may produce backflow into the renal tubules, interstitium, lymphatics, or veins. The chief advantage of retrograde pyelography is that contrast material can be injected directly under controlled pressure into the ureters and collecting system. If performed correctly, this technique provides unsurpassed visualization of the ureter and collecting system in patients whose renal function is impaired.
Renal Angiography
There are several methods for the radiographic study of the renal arteries. The use of a vascular catheter, introduced percutaneously into the femoral artery by the Seldinger technique, permits wide variability in techniques. A midstream aortic injection is made first, using 40 to 60 mL of an organic iodide. This examination indicates the location and number of renal arteries and may define abnormal lumbar vessels in patients with metastatic tumor (Fig. 20-1).
FIG. 20-1. A: Angiogram in the study of a renal transplant. The percutaneous catheter technique was used demonstrate the renal artery, which is anastomosed to the internal iliac artery. B: Digital subtraction transfemoral catheter abdominal aortogram. The abdominal aorta is normal. There is excellent fill of the celiac axis and of the hepatic and splenic arteries. There are single renal arteries bilaterally. Some branches of the superior mesenteric arteries overlie the left renal arteries.
Selective renal arteriograms are then obtained by manipulation of the catheter tip under fluoroscopic control into the desired renal artery and injection of small amounts (10 to 15 mL of the opaque medium) into the artery. The advantage is dense opacification of the renal artery and its branches, which is needed in a detailed study of the vessels. The disadvantages are that a small accessory vessel may be missed, and multiple injections are required when multiple vessels are present. Simultaneous study of both renal arteries can be obtained by midstream aortic injection with the catheter tip above the orifices of the renal arteries. This method fills accessory arteries that may be present and provides simultaneous visualization, allowing comparison of the renal vessels on the normal and abnormal sides.
Rapid film changers are used, programmed to radiographically record the arterial, nephrographic, and venous phases of the examination. Magnification techniques may be used to study vascular detail in one or both kidneys. Digital subtraction angiography (DSA) has begun to replace traditional cut-film angiography and has the advantage that lower doses of contrast agents can be used. In general, renal angiography is used only in selected cases where cross-sectional imaging is unable to give enough information for treatment decisions or when endovascular treatment is being considered, including renal angioplasty/stenting and renal embolization. Major renal arterial and venous anatomy can be directly imaged noninvasively with the use of helical CT angiography, magnetic resonance angiography, and, to a lesser degree, ultrasound.
Percutaneous Antegrade Pyelography (Percutaneous Nephrostomy)
A needle is placed percutaneously into the renal pelvis from a posterolateral approach, and either fluoroscopic or

ultrasonic guidance is used. After a sample of urine is obtained for analysis, a contrast medium can be injected to evaluate the pelvocalyceal system and ureter. Subsequently, conventional percutaneous nephrostomy, brush biopsy, stent placement, stone dissolution or extraction, physiologic pressure-flow study, or dilatation of a stenosis can be performed using the access gained to the collecting system from the initial needle placement.
Percutaneous nephrostomy is most often used for emergency drainage of an obstructed, infected upper urinary tract. For temporary, acute drainage, a standard percutaneous catheter with external drainage is effective. For chronic drainage, a locking loop nephrostomy or ureteral stent can be used. Percutaneous nephrostomy can also be used to introduce solutions such as sodium bicarbonate to dissolve uric acid stones. Using the percutaneous route, calculi can be treated by basket removal8 or by the use of an ultrasonic lithotriptor. Percutaneous nephrostomy can provide palliation in patients with terminal neoplastic disease and is used in renal transplantation for diagnosis and therapy.
Voiding Cystography or Cystourethrography
This examination is used in the study of patients with suspected lower urinary tract obstruction or vesicoureteral reflux and in children with persistent or recurrent urinary tract infection in whom vesicoureteral reflux is suspected. The examination consists of filling the bladder with radiopaque material to the point of urge to void, so that the voiding process can be imaged.
The patient is examined before, during, and after urination. Voiding cystography should be performed on unanesthetized patients who have not had recent instrumentation. Bladder filling is monitored as contrast is instilled. Image-intensification fluoroscopy is necessary to obtain an adequate examination. In addition, spot films (70- or 90-mm or conventional spot films) are used to record the findings. The examination can be recorded on videotape if desired, because no additional radiation is used. Films are exposed with the patient in a lateral or an extreme posterior oblique position for best visualization of the bladder neck and urethra. If vesicoureteral reflux is present, films are obtained to show the amount and level of ascent of reflux, as well as the size of the ureters and renal collecting systems.
Retrograde cystography is another method of studying the bladder. After voiding, a urethral catheter is inserted and the bladder is filled with opaque contrast material. Rarely, a small amount of contrast medium is absorbed from the bladder. Among the indications for cystography are suspected bladder rupture in a trauma patient and suspected bladder tumors, diverticula, or calculi. In many situations, direct cystoscopy has supplanted cystography.
CT cystography is increasingly being used to evaluate patients after trauma or after pancreas or kidney transplantation to rule out bladder leak.
Renal Scintigraphy
The most common radionuclides used to image the kidney are technetium-99m pentetate (99mTc-DTPA) and iodine-131 ortho-iodohippurate (131I-Hippuran). These are used to evaluate overall renal function. Renal perfusion and glomerular filtration can be assessed by 99mTc-DTPA, and renal plasma flow and tubular function by 131I-Hippuran. A newer tubular agent, 99mTc-mertiatide, can be used much like 131I-Hippuran. The collecting systems can be evaluated by all of these radionuclides, although DTPA and mertiatide are the preferred agents because of better imaging properties. 99mTc-Gluceptate is an agent that can evaluate both glomerular and tubular function. Radionuclides such as Tc-99m dimercaptosuccinic acid (99mTc-DMSA), which permit evaluation of renal cortical morphology, are used only occasionally, such as for the evaluation of renal scarring in children with recurrent urinary tract infections or for the evaluation of the function of a renal mass that may represent fetal lobulation or a dromedary hump.
intravenous injection of 10 to 20 mCi of 99mTc-DTPA is given, and images are obtained with a gamma camera at 1-second intervals for approximately one-half minute. These dynamic images allow evaluation of renal perfusion. Subsequent serial static images over the next 30 minutes, with delayed views as needed, allow assessment of renal function and the collecting systems.
For the 131I-Hippuran renogram, 100 to 150 μCi is injected intravenously with simultaneous gamma camera and computer acquisition over the next 30 minutes. Using a region of interest over renal parenchyma, one can generate a time-activity curve (renogram) for each kidney. The renogram has three phases: the vascular, cortical transit, and excretory phases. Similar data can be obtained after injection of 99mTc-mertiatide. The usual dose is 5 to 10 mCi injected intravenously.
A diuretic (Lasix) renogram is frequently obtained, especially in children, to distinguish functional from mechanical obstruction. Intravenous Lasix administration of 1 mg per kilogram of body weight in children, or 20 mg in adults, is given when retention of activity is demonstrated within the renal pelvis. The renogram is best performed with mertiatide, although Lasix can also be used with DPTA.
A nonobstructed pelvocalyceal system (such as an extrarenal pelvis) will respond with a brisk diuresis and a decrease in radionuclide activity in the region of the pelvis. If a true mechanical obstruction is present, there will be no diuresis or change in radionuclide activity.
Renal transplants can be evaluated with a 99mTc-DTPA perfusion study and an 131I-Hippuran or 99mTc-mertiatide renogram with static images. Usually, a baseline study is obtained within the first day after transplantation. Serial

studies are then obtained as needed to evaluate transplant status.
The Kidney
The normal kidneys are bean-shaped structures that lie on either side of the lower thoracic and upper lumbar spine, usually between the upper border of the 11th thoracic and the lower border of the 3rd lumbar vertebrae. In the upright position, the kidney descends 2 or 3 cm. The right kidney lies approximately 2 cm lower than the left. Both move moderately with respiration and with change in position. The long axis is directed downward and outward, parallel to the lateral border of the psoas muscle on either side. In the lateral plane, the axis is directed downward and anteriorly, so that the lower pole is 2 to 3 cm anterior to the upper pole. When the patient is supine, the renal pelvis and proximal ureters lie well posterior to the anterior edge of the vertebral bodies (shown well on CT). At L3, the average ureter is three fourths of a vertebral width from the posterior vertebral margin. It then curves anteriorly to the level of the anterior vertebral border at L4; at L5, it is anterior to the vertebral body by about one fourth of the anteroposterior diameter of this vertebral body.
Normal renal size varies. The normal range of renal length in adults is 11 to 15 cm. The right kidney is usually shorter than the left, with the upper limit of variation in length being 1.5 cm. There is a relation between vertebral body height and renal length. As a rule, the kidney length is 3.7 ± 0.37 times the height of the second lumbar vertebra measured on the same film using the posterior margins of the vertebral body. According to Batson and Keats,12 97% of normal kidneys are within the range between the height of L1 through L3 and the height of L1 through L4. In children between 1½ and 14 years of age, the renal length is about equal to the length of the first four lumbar bodies including the three intervening disks plus 1 cm. In infants, the renal size is relatively greater. In children, the normal difference in renal length on the two sides may be up to 1 cm.
There is some variation in renal shape, particularly on the left side. Fetal lobulations may persist on one or both sides, producing rather clearly defined indentations or notches along the lateral aspect of the kidney. The left kidney may be generally triangular in shape with a local bulge or convexity along the left midborder, sometimes termed a dromedary hump. This may be related to the position of the spleen, or it may be a form of fetal lobulation, or both. The kidney is visualized in roentgenograms mainly because of the presence of perirenal fat. The increased radiolucency of fat makes the outline of the kidney stand out from the surrounding soft tissues. If there has been much wasting caused by chronic illness or malnutrition, the loss of perirenal fat may make the renal outlines very indistinct or completely invisible. The kidneys are contained within the renal capsule and surrounded by perirenal fat, which is enclosed within Gerota’s (perirenal) fascia. Perirenal hemorrhage, pus, or urine tends to be contained within this fascia and can be detected with CT or ultrasonography.
The anatomy of the retroperitoneum can be quite complex, and a detailed discussion is beyond the focus of this chapter. Readers are referred to the classic work by Dr. Morton Meyers, Dynamic Radiology of the Abdomen.124 There are three anatomic spaces around each kidney: perirenal, anterior pararenal, and posterior pararenal (Fig. 20-2). The perirenal space is bounded by the anterior and posterior portions of the renal (Gerota’s) fascia. The leaves of fascia fuse superiorly, laterally, and medially, enclosing the kidney, adrenal gland,

renal vasculature, and emerging portion of the proximal ureter. This fascial envelope is functionally open caudally to just above the pelvic brim. At that point it communicates with the caudal extent of the anterior and posterior pararenal spaces. The ureter emerges from the perirenal space about midlumbar spine level to course caudad in the anterior pararenal space and ultimately to reach the bladder.
FIG. 20-2. Anatomic spaces around the kidney. (From
Radiol Clin North Am 17:323-324, 1979
, with permission of the author and WB Saunders Co.)
The anterior pararenal space is bound posteriorly by the anterior portion of the renal fascia, anteriorly by the posterior parietal peritoneum, and laterally by the lateral conal fascia. It contains the pancreas; the second, third, and fourth portions of the duodenum; the ascending and descending colon; and the vascular supply to the spleen, liver, pancreas, and duodenum. The posterior pararenal space is bound posteriorly by the transversalis fascia and anteriorly by the posterior portion of Gerota’s fascia. It contains only fat and scattered vessels and nerves. All three spaces potentially communicate caudally near the bony pelvic brim. Therefore, blood or an infected fluid collection in the perirenal space can track caudally and then involve either or both ipsilateral pararenal spaces (see Fig. 20-2).
A theory proposed by Molmenti and associates128 suggests that fluid collections are more likely to collect in spaces between tissue planes rather than in the perirenal and pararenal spaces. The space anterior to the kidney, between the anterior pararenal space and anterior to the anterior renal fascia, is termed the retromesenteric space. The area between the posterior renal fascia and the posterior pararenal space is the retrorenal space.
The Ureters
The ureters normally course downward from the most dependent portion of the pelves to the midsacral region, then turn posterolaterally and course in an arc downward and then inward and anteriorly to enter the trigone of the bladder on either side of the midline. Slight redundancy is common, and alteration in size is frequently noted. Therefore, it is necessary to exercise care in making the diagnosis of ureteral stricture, displacement, or dilatation. There are three areas where normal narrowing of the ureter can be observed when it is filled with radiopaque material: the ureteropelvic junction (UPJ), the ureterovesical junction, and the bifurcation of the iliac vessels. These are sites where calculi often lodge in the course of passage. A common normal variant is symmetric deviation medially of the ureters as they enter the bony pelvis. The narrower the bony pelvis, the more medial the position of the ureters.
The Bladder
The normal urinary bladder is transversely oval or round; the inferior aspect normally projects 5 to 10 mm above the symphysis pubis. Its floor parallels the superior aspect of the pubic rami, and its dome is rounded in the male and flat or slightly concave in the female owing to the presence of the uterus above it. The size and shape of the normal bladder vary considerably. The internal aspect of the wall of the normal bladder is smooth as outlined by opaque material used in urography or cystography. The bladder is in a higher position in children than in adults and is slightly higher in males than in females. The bladder is relatively larger in children than in adults. A common normal variant is the anterior prolongation type, which results in a pear-shaped appearance.
The Normal Urogram
The renal pelvis varies considerably in size and shape but is usually roughly triangular, with the base parallel to the long axis of the kidney (Figs. 20-3 and 20-4). It may be conical with the apex contiguous to the upper ureter. The range of normal is wide; some pelves are long, narrow tubes and others are large and globular. There is also a considerable variation in position of the pelvis in relation to the kidney. It may be almost completely within the renal outline (intrarenal) or almost completely extrarenal. In the former position it is usually small, whereas in the latter it is large. The average normal pelvis is partially intrarenal and partially extrarenal. Bifurcation or duplication of the pelvis is very common and is considered an anatomic variant rather than a congenital anomaly.
FIG. 20-3. Intravenous urogram of a normal person showing good filling of the pelves, calyces, and ureters down to about the level of the compression device, the superior portion of which overlies the lower fourth lumbar vertebra.
FIG. 20-4. Intravenous urogram of a normal person. The use of compression has resulted in a slight degree of blunting of the calyceal fornices. Note that there is some asymmetry on the two sides.
The calyceal system consists of major calyces that begin at the pelvis and extend into the kidney to the junction with

the minor calyces. Each major calyx may be divided into a base (adjacent to the pelvis) and an infundibulum that is more or less tubular and extends from the base to the apex, or distal portion, from which one or more minor calyces project. The minor calyx consists of the body or calyx proper, beginning at the junction with the major calyx, and the fornix, which surrounds the conical renal papilla and into which the latter appears to project. The anatomic shape of the minor calyx is fairly constant, but because this structure is projected in various planes in the urogram there is considerable apparent variation. When viewed en face, it resembles a circular life preserver with a dense periphery and a relatively radiolucent center. In profile the appearance of a minor calyx is somewhat triangular, with the apex of the triangle pointing toward the major calyx; the base is pointing away from it and is sharply concave or cupped. By contrast, there is marked variation in the shape of the major calyces, which can be long and narrow or short and broad. There are usually two major calyces and 6 to 14 minor calyces, but the number can vary widely. The calyceal system is not always bilaterally symmetrical, which makes interpretation difficult in some instances.
Coordinated peristalsis begins in the calyceal system of the kidneys. The collecting systems alternately fill and contract, activity that accounts for the variable appearance of the collecting system during intravenous urography. The discharge of urine from the pelvis into the ureter is accompanied by ureteral peristalsis. This occurs as broad waves at variable intervals (from 4 to 12 per minute). Ureteral peristalsis causes the ureter to have a variable caliber in different portions at the same time and a variation in contour in serial roentgenograms. The waves are visible as smooth areas of constriction or complete absence of filling that may separate one or more areas of slight dilatation. The effects of calyceal and ureteral peristalsis must be taken into account in the interpretation of the EXU.
Renal Backflow
The term backflow was initially applied to the escape of contrast material from the renal pelvis and calyces during retrograde pyelography as a result of an increase in intrapelvic pressure. The pressure is increased in EXU owing to osmotic diuresis and use of compression devices. Acute ureteral obstruction also results in increased intrapelvic pressure. Because similar phenomena occur in these instances, the term backflow has been carried over to describe changes observed in EXU. Backflow occurs in the normal kidney, and its recognition and differentiation from changes caused by disease of the kidney are therefore important.
There are two major types of backflow, pyelotubular and pyelointerstitial (pyelosinus). Pyelolymphatic and pyelovenous backflow are merely stages of the pyelointerstitial form. Pyelotubular backflow is the most common type; when it occurs during EXU, it represents stasis in the tubules in the papilla rather than actual backflow. Roentgenographic findings consist of a brush-like tuft of opacity radiating into the papilla from the minor calyx (Fig. 20-5). Pyelointerstitial (pyelosinus) backflow begins with minute (painless) rupture of the fornix of a calyx; this permits the escape of contrast material or urine into the renal sinus, which is the loose adipose and connective tissue surrounding the pelvis and calyces and supporting a venous plexus. When the amount of extravasation increases it extends medially into the peripelvic area, into the perirenal fat within Gerota’s fascia, and downward along the ureter. The extravasated material may enter the lymphatics to produce pyelolymphatic backflow. A much less common occurrence is pyelovenous backflow, in which, presumably, the material enters the arcuate and other veins. Some investigators believe that the arcuate shadows observed in this condition are produced by perivascular extension of pyelosinus extravasation, not by filling of the veins. All forms of backflow may be observed at one time (Fig. 20-5).
FIG. 20-5. Backflow. This retrograde pyelogram shows a marked amount of pyelolymphatic backflow (upper arrow). Pyelotubular backflow is outlined (lower arrow). There is also some extravasation in the vicinity of the ureteropelvic junction, representing interstitial backflow.
The roentgenographic findings in the early extravasation of the pyelointerstitial backflow consist of a horn-like projection of opaque medium extending from the fornix away from the papilla into the renal substance. As more material is extravasated, it extends medially to the hilum and along the upper ureter, producing poorly defined densities in these areas. Pyelolymphatic backflow is manifested by opacification of lymphatic channels that extend from the hilum of the kidney medially toward the para-aortic nodes. These channels tend to be redundant, somewhat tortuous, and branched.

Extravasation of medium into the renal parenchyma also results when the catheter penetrates a calyx in retrograde pyelography. The roentgenographic appearance is variable, depending on the amount and distribution of the extravasated material.
Arterial and Venous Impressions
Arterial impressions or indentations on the renal pelvis and infundibula were found in 18% of 150 patients studied by Nebesar and colleagues.136 They occur three times more often on the right than on the left. The most common site is the superior infundibulum on the right. The impressions consist of smooth transverse or oblique indentations on the infundibulum or pelvis. Most of the involved vessels are ventral to the collecting system. They usually cause no symptoms and are significant only in that they must be differentiated from pathologic processes (Figs. 20-6 and 20-7). Rarely, partial infundibular obstruction is produced, leading to dilatation of calyces and pain and occasionally to infection. Oblique as well as frontal projections are needed to make the diagnosis. Confirmation by angiography may be necessary if other causes are suspected. Occasionally a slightly tortuous renal artery, renal artery aneurysm, or bulbous renal vein simulating a renal sinus mass (pseudotumor) is observed, particularly on a tomogram. The appearance is that of a round or oval mass in the renal sinus that is usually recognized as vascular, but angiography may be needed for differentiation in some instances.
FIG. 20-6. A: Urogram showing arterial indentations producing a vertical lucency on the lateral aspect of the pelvis (upper arrow) and horizontal pelvic indentation (lower arrow). B: Selective renal arteriogram, on the same patient as in A, demonstrates the relationship of arteries to the indentations noted.
FIG. 20-7. A: Unusual vascular indentation causing a persistent elongated defect in the upper pole infundibulum on the left (arrow). B: Closeup view of the defect, which was persistent. A subsequent selective arteriogram showed a renal arterial branch causing the defect.
Venous impressions of the superior infundibulum are not as common as those produced by arteries. Urographic findings are quite characteristic123 and include a wide, smooth filling defect of the proximal part of the superior infundibulum that is usually best shown on the prone film. Venography can be used to confirm the diagnosis but is seldom necessary. Ultrasound or CT is also frequently used to rule out the possibility of a renal mass mimicking a venous impression.
The Normal Cystogram
The normal cystogram outlines the contrast-opacified urine in the smooth-walled, rounded, or oval bladder. The urinary bladder is usually filled to some extent during EXU, and this examination is often sufficient to outline gross lesions. When additional study of the bladder is needed, especially to rule out bladder leak, cystography is used (Fig. 20-8). Films are exposed in frontal, lateral, and oblique projections; if necessary, upright and postvoiding roentgenograms may be obtained.
FIG. 20-8. Cystogram showing bilateral vesicoureteral reflux. The bladder outline appears normal.
A procedure termed CT cystography has been shown to be useful to detect small anastomotic leaks in pancreas transplantation patients with bladder drainage. The bladder is filled through a Foley catheter with up to 500 mL of iodinated contrast material and 60 mL of air. Scans are obtained before and after voiding. This approach may be useful to detect bladder rupture not detected by conventional techniques in trauma patients.17
Computed Tomography, Ultrasonography, MRI, and Radionuclide Anatomy of the Kidney
On CT, the kidneys appear as elliptical or round structures of soft-tissue density, with the central renal sinus composed predominantly of fat density. Because of the surrounding perirenal fat, the margin of each kidney is visible and should be smooth (Fig. 20-9). The cortex and medullary portions cannot be distinguished on unenhanced scans but can be demarcated with a rapid intravenous bolus. The collecting system is best seen on enhanced scans because of the high attenuation values of the contrast agent. The renal vessels are best seen with dynamic or helical scans; the renal veins are usually larger than the arteries. The anterior pararenal, posterior pararenal, and perirenal compartments usually can be distinguished (see Fig. 20-2). Normally, Gerota’s fascia is imperceptible or is seen as a thin fascial band. With certain pathologic processes (e.g., pancreatitis) Gerota’s fascia becomes thickened and is more easily seen. Coronal and sagittal reformations allow estimation of renal size and volume.
FIG. 20-9. Normal left renal anatomy. Enhanced computed tomography shows smooth contour of left kidney.
The most prominent feature of the normal kidney on ultrasonography


is the central renal sinus. This is quite echogenic, mainly because of the fat surrounding the pelvocalyceal system (Fig. 20-10). The amount of fat varies with the individual, and there is an increase in fat and echogenicity of the renal sinus with age.157 There may be mild dilatation of the pelvis as a normal variant, but dilated calyces do not usually accompany this condition. Normal calyces usually are not seen. Vascular branching may mimic hydronephrosis. A vessel should not directly abut the renal pyramid as a dilated calyx in hydronephrosis would. Color Doppler examination can more definitively identify vessels in questionable cases. The peripheral cortex contains low-level echoes, whereas the pyramids are hypoechoic or anechoic. The bright, small, circular echoes in the region of the corticomedullary junction represent the arcuate arteries. The perinephric fat and capsule are seen as a variably hypoechoic or echogenic region surrounding the kidney. The normal ureter usually is not visible. The renal veins are readily visualized, but the arteries are more difficult to see.
FIG. 20-10. Normal renal anatomy. Longitudinal ultrasound of right kidney in which the echogenic central renal sinus is visible. The renal parenchyma is isoechoic or hypoechoic to adjacent normal liver. (Courtesy of Deborah Krueger, RDMS.)
MRI anatomy of the kidney is similar to that described for CT in the axial plane, but MRI has the advantage of multiplanar display (Fig. 20-11). On T1-weighted images, the cortex has medium to high signal intensity, and the medulla has low signal intensity. Because of this, the corticomedullary junction is usually well seen. T2-weighted images reveal less signal contrast between the cortex and medulla.113 Vessels in the renal hilum are easily identified, and associated flow void ensures vascular patency.
FIG. 20-11. Normal renal anatomy. Coronal T1-weighted magnetic resonance image. Multiplanar capability allows wide range of imaging planes.
The normal perfusion image with 99mTc-DTPA shows counting rates that are almost equal over both kidneys. Static images at 0, 5, and 10 minutes, and later as needed, reveal activity in the renal parenchyma and excretion into the collecting systems (Fig. 20-12). The normal ureters are only occasionally visualized. Activity can be demonstrated in the bladder as early as the 5-minute image, although more often it is seen on the 10-minute image. The normal 131I-Hippuran renogram demonstrates an initial rise in counts owing to activity in the extrarenal and renal vessels during the vascular phase. The counts continue to increase gradually during the cortical transit phase because of the accumulation of 131I-Hippuran or 99mTc-mertiatide by the renal tubular cells. The peak occurs when the rate of uptake equals the rate of excretion into the collecting system. The counts then gradually decrease during the excretory phase because of the excretion of 131I-Hippuran into the collecting system (Fig. 20-13).
FIG. 20-12. A: Normal perfusion scan of aorta and kidneys, also showing spleen and lung bases, at 3 seconds per frame. B: Immediate static image from 99mTc-DTPA scan. C: Static image at 10 minutes from 99mTc-DTPA scan. Note excretion into collecting systems (arrows).
FIG. 20-13. A: Gamma camera scans of 131I-Hippuran renogram. B: Normal time-activity curve for 131I-Hippuran (x axis represents minutes).
Anomalies of the kidney and ureter result from errors in development. The kidneys arise from a mass of renal mesenchyme

at the upper end of the ureteral buds, which in turn rise from the lower end of the mesonephric (wolffian) ducts. The mesonephron is the excretory organ lower in the phylogenetic scale, and in the human it functions for a short time in early embryologic development before becoming part of the male genital system. The ureteral buds grow dorsally, lying close together as the renal mesenchyme differentiates. Each bud bifurcates into an upper and lower sprout to form the major calyces. The ureter is anterior to the kidney as the latter ascends from the upper sacral area to its position in

the lower-thoracic–upper-lumbar region. As it ascends, the kidney rotates to bring it lateral to the ureter in the midlumbar region. The renal blood supply is attained after the kidney reaches its normal adult position. The lower end of the ureter loses its relation to the wolffian duct and opens into the bladder in a higher and more lateral position. The wolffian duct migrates distally, and its orifices are eventually situated in the distal portion of the floor of the prostatic urethra to become the ejaculatory ducts in the male. The orifices of the wolffian duct become vestigial structures in the female.
Anomalies in Number
Renal Agenesis (Single Kidney)
The occurrence of a single kidney is a rare anomaly. Care must be taken when making a radiographic diagnosis of unilateral renal agenesis, because a contralateral nonfunctioning or malpositioned kidney may not be readily visible. The single kidney tends to be larger in patients with agenesis of one kidney than in patients with secondary compensatory renal hypertrophy. Radiographic signs are an absence of a renal shadow on one side with an unusually large kidney on

the other side. The trigone is usually deformed, with the ureteral orifice missing on the involved side, so that cystoscopy may confirm the diagnosis. At times, however, a portion of the lower ureter may be present in renal agenesis; in these cases, the trigone may have no deformity. Angiography confirms the absence of the renal artery, but renal venography is said to be more reliable than arteriography in making the diagnosis of renal agenesis. Other anomalies, such as congenital heart disease and a neuromuscular deficit accompanied by a small pelvic outlet, sacral agenesis, and bladder hypoplasia (caudal regression), may be associated with renal agenesis. With the advent of CT, MRI, and ultrasonography, the diagnosis of renal agenesis has become much easier, and angiography is no longer routinely used.
Supernumerary Kidney
Supernumerary kidney is a rare anomaly. The usual finding is that the anomalous kidney is small and rudimentary, and the other kidney on the same side is often smaller than the normal kidney on the opposite side. Demonstration of the presence of a separate pelvis, ureter, and blood supply is necessary to make the diagnosis. EXU can be used to outline the collecting system of the supernumerary kidney if it is functioning. Aortography can show the blood supply if that is necessary to confirm the diagnosis. CT, MRI, and ultrasonography are less invasive and may be helpful.
Anomalies in Size and Form
Anomalies of renal size and form are more common than anomalies in number. Hypoplasia on one side is usually associated with hyperplasia on the other. The hypoplastic or infantile kidney functions normally and can be seen on EXU. It must be differentiated from the acquired atrophic kidney, which is small and contracted because of vascular or inflammatory disease. In congenital hypoplasia, the calyceal system and pelvis are small, and there is a normal relation between the amount of parenchyma and the size of the collecting system (Fig. 20-14). In the secondarily contracted kidney the pelvis and calyces tend to be normal in size, so the decrease in renal size is caused by a parenchymal deficit. Furthermore, the function of the kidney in the latter case tends to be impaired. Despite these differences, it is often very difficult to distinguish between the two conditions without the use of renal arteriography. The size of the orifice of the renal artery is important: in hypoplasia it is small, and in an atrophic kidney it is normal but may taper to a very small size near the orifice.
FIG. 20-14. Hypoplasia of right kidney. Note the marked difference in size of the two kidneys. Function is present in the right kidney despite its small size; its limits outlined (arrows).
The other anomaly in size, hyperplasia, is associated with agenesis or hypoplasia on the opposite side. Enlargement of the kidney is usually caused by conditions other than agenesis or hypoplasia, however, and it is then more properly termed compensatory hypertrophy. Several disorders can cause renal enlargement; these include obstructive hydronephrosis, polycystic disease, other cystic or dysplastic disorders, neoplasm, renal vein thrombosis, acute infection, Waldenström’s macroglobulinemia, hemophilia, acute arterial infarction, and duplication of the renal pelvis.76 Often the enlargement is bilateral, however, and there are clinical, laboratory, and urographic findings that help to make the differentiation.

Conditions that characteristically cause bilateral renal enlargement include (1) acute glomerulonephritis, (2) lymphoma, (3) leukemia in children, (4) systemic lupus erythematosus, (5) polycystic disease, (6) bilateral renal vein thrombosis, (7) amyloidosis, (8) sarcoidosis, (9) sickle cell disease, (10) lipoid nephrosis, (11) lobular glomerulonephritis, (12) glycogen storage disease, (13) hereditary tyrosinemia, and (14) total lipodystrophy.
Fusion Anomalies
Fusion anomalies represent an alteration in form of the kidneys and can often be recognized or at least suspected on plain roentgenograms of the abdomen. CT provides more complete information about these anomalies than does EXU. Ultrasonography likewise provides better assessment of the renal parenchyma.
Horseshoe Kidney
The horseshoe kidney is the most common type of fusion anomaly. In this condition, the lower poles of the kidney are joined by a band of soft tissue, the isthmus, which varies from a thick parenchymatous mass as wide as the kidneys themselves to a thin, string-like band of fibrous tissue. The upper poles are rarely involved. The long axis of the kidney is reversed in this anomaly, so that the lower pole is nearer the midline than the upper. There is also an associated rotation anomaly on one or both sides that varies in degree, usually more on the left. The calyces are directed backward or posteromedially rather than laterally. As a result they are seen on end or obliquely, which alters their appearance considerably (Fig. 20-15A). The ureters tend to be somewhat stretched over the isthmus, and partial obstruction on one or both sides is not unusual. This leads to dilatation of the pelvis and calyces and may also lead to chronic inflammatory disease

and the formation of calculi. The roentgenographic diagnostic features on plain film are (1) alteration in the axis of the kidneys, (2) mass observed connecting the lower poles, (3) renal enlargement if present, and (4) calculi if present. Urography confirms these findings; in addition, the following are present: (1) malrotation, with the pelves anterior or anterolateral in position; (2) nephrographic demonstration of the parenchymal isthmus connecting the lower poles (if present); (3) often, varying degrees of dilatation of the collecting system on one or both sides; (4) possible nonfunction of one kidney because of massive obstructive hydronephrosis; (5) possible partial obstruction of both kidneys, usually at or near the UPJ; and (6) upper ureteral displacement, which varies with the amount of malrotation. These findings are often particularly striking on CT (Fig. 20-15B) or MRI. Horseshoe kidneys are frequently supplied by multiple arteries, and the isthmus is often supplied by anomalous branches of the common iliac artery on one or both sides. Complications of horseshoe kidneys are common and include UPJ obstruction, stones, Wilms’ tumors in children, and susceptibility to trauma.
FIG. 20-15. Horseshoe kidney. A: Note the reversal of the long axis of the kidneys. Rotation anomaly is present. The fusion inferiorly is faintly visualized on this reproduction. B: Enhanced computed tomogram in a separate patient shows isthmus crossing the midline.
Crossed Ectopy
Crossed fused ectopy is an anomaly of form that is much less common than the horseshoe kidney. It consists of fusion of the kidneys on the same side; the lower one is ectopic and its ureter crosses the midline to enter the bladder normally on the opposite side. Both kidneys are often lower in position than normal, and various rotation anomalies as well as a wide variation in shape and type of fusion are noted. This anomaly is also frequently associated with partial obstruction, which results in inflammation and often in calculus formation. The “pancake” kidney is a variation in which there is fusion of both upper and lower poles, with failure of rotation, and the calyces are directed posteriorly. The renal mass lies in or near the midline and is low in position, often overlying the sacrum. The ureters enter the bladder normally. Several descriptive terms have been applied to other rare forms of fusion. All these forms tend to result in obstruction, which is turn causes hydronephrosis, infection, and calculus formation. These ectopic kidneys usually have an aberrant blood supply, often with multiple arteries.
Extrarenal calyces occur rarely if the portion of the ureteric bud fails to invaginate the ectopic nephrogenic mass. The extrarenal calyx is large, probably because there is no supporting parenchyma, and mimics the blunt calyx resulting from obstruction or infection. CT and ultrasonography usually are more informative than urography.
Anomalies in Position
Anomalies of renal position are common. Malrotation has been described as being almost constantly present in fusion anomalies, but it also occurs as a single anomaly (Fig. 20-16). It results from incomplete or excessive rotation, and

urographic study indicates the degree of anomaly. Rotation anomalies are usually of little clinical significance unless they are associated with obstruction, but it is important to recognize these as innocuous anatomic variations that do not produce symptoms. Retroperitoneal tumor masses may displace the kidneys and produce an alteration in rotation that must be differentiated from congenital rotation anomalies. Crossed ectopy can occur without fusion, and the findings are similar to those described in the preceding section except for the lack of fusion. The ectopic kidney is lower than the normal one in position and is usually described as a sacral or pelvic kidney, depending on its position. Failure to visualize the kidney in its normal position should lead one to suspect ectopy and to look for it, because agenesis of a kidney is rare. In many instances the kidney can be visualized only when contrast material outlines it, so that CT, EXU, radionuclide scanning, or retrograde pyelography may be necessary to indicate its position. A simple way to assess ectopic kidneys is by ultrasonography; however, overlying bowel gas may obscure visualization of renal parenchyma. If it is nonfunctioning, aortography may be used to identify the aberrant artery (or arteries) to a pelvic kidney, which may or may not appear as a pelvic mass on plain film. The pelvic mass representing an ectopic kidney may be discovered on study of the small bowel or colon as an extrinsic mass displacing bowel. Characteristically, the ureter of an ectopic kidney is only long enough to reach from the renal pelvis to the bladder, and this aids in distinguishing displacement of a normal kidney downward from development of the kidney in an abnormally low position. Superior ectopia of the kidney (“intrathoracic kidney”) is probably more common than reports in the literature would indicate. The possibility of intrathoracic kidney should be considered in the differential diagnosis of masses of appropriate size projecting into the posterior thorax from below the diaphragm. An intrathoracic kidney is usually unilateral. It may be associated with herniation through the foramen of Bochdalek or a congenital eventration of the diaphragm posteriorly. EXU or ultrasonography readily identifies the position of the kidney in these cases.
FIG. 20-16. Ectopic and malrotated right kidney. The calyces are dilated in comparison with those in the normal left kidney.
Nephroptosis is the term applied to downward displacement and more mobility of the kidney than usual. It is of doubtful clinical significance because obstruction ordinarily is not produced and surgical intervention is rarely, if ever, indicated. Roentgenographic demonstration of this condition can be accomplished by obtaining an additional exposure during urography with the patient in an upright position.
Other Renal Anomalies
Aberrant Papilla
Aberrant papilla occurs occasionally. The papilla projects directly into the lumen of the infundibulum as a smooth, conical mass that appears round or oval when viewed en face. It bears no resemblance to a minor calyx in which a normal papilla projects. Other anomalies include multiple papillae entering a single calyx, which may simulate a blood clot or nonopaque stone.16
The term megacalyces describes an anomaly consisting of enlargement of calyces in one or both kidneys associated with underdeveloped renal pyramids. There is no evidence of obstruction, and function is normal. Because of calyceal size, there may be stasis with a tendency for stone formation, which may result in infection that can alter the urographic findings.180
Benign Cortical Nodule
Cortical nodules are a normal variation resulting from the presence of more cortical tissue than usual in a portion of the kidney. Based on location, there are three types of cortical nodules: subcapsular, hilar lip, and septa of Bertin. The patterns of urographic appearance of cortical nodules have been described by Thornbury and colleagues185 based on the elegant anatomic correlation done by Hodson. When the question of the differential diagnosis (cyst, tumor, or cortical nodule) is raised on EXU, ultrasonography is the most direct way to resolve the question if simple cyst is the most likely diagnosis. If tumor or cortical nodule seems more likely, CT examination of the kidney usually provides the most definitive information to distinguish tumor from normal variant.
A focal prominence of one or more columns of Bertin can mimic a mass lesion caused by a tumor or inflammation. However, it has a fairly easily recognized roentgenographic appearance in most cases.185 It is caused by a variant of normal renal development in which there is more cortical tissue in an area than usual. Depending on its location this cortical nodule can distort the adjacent calyces and the adjacent surface of the kidney.
Solitary Renal Calyx
A solitary renal calyx is an extremely rare anomaly in which one or both kidneys have a single calyx that drains the entire kidney into a somewhat bulbous tube that represents the pelvis. Kidneys in several other mammals have a solitary calyx. This anomaly does not necessarily indicate renal disease, but other congenital anomalies may be associated with it.
Anomalies of the Renal Pelvis and Ureter
Ureteropelvic Junction Anomalies
UPJ dysfunction or obstruction, the most common congenital anomaly of the urinary tract and the most common cause of neonatal hydronephrosis, is usually bilateral but not

always symmetrical. The left side is often more severely involved than the right. The amount of hydronephrosis depends on the severity of obstruction. In the neonate, marked obstruction may be the cause of massive unilateral or bilateral renal enlargement. There is some controversy as to the cause of this anomaly. Most cases appear to result from of an intrinsic wall abnormality that is functional rather than anatomic. The peristaltic wave may fail to pass normally across the abnormal area. Occasionally the cause may be an extrinsic abnormality such as an aberrant vessel or band of fibrous tissue, either of which may angulate the ureter and tend to hold it in place while the pelvis dilates. Rarely, an intrinsic mucosal fold or web may be present.1 If infection occurs, secondary fibrosis may aggravate the condition.
Urographic findings vary with the severity of the condition. Caliectasis and pyelectasis are observed, along with a somewhat rectangular extrarenal type of pelvis that is rather characteristic. Often the UPJ is not dependent as in the normal individual, so the insertion of the ureter is high and posterior. Diuretic-influenced radionuclide study often helps determine whether the ureteropelvic narrowing is functionally significant.190 It is important to carefully examine the contralateral kidney when a UPJ obstruction is diagnosed, because of the strong association with other renal anomalies.
Duplication of the Pelvis and Ureter
Incomplete double ureter is formed when the renal bud divides too early or the division extends into the ureter.74 The division varies from an exaggeration of the length of major upper- and lower-pole calyces to duplication of the ureter for most of its length.
Complete duplication of the ureter can also occur. Each ureter has its own vesical orifice; the upper ureter usually drains the upper third of the kidney, whereas the ureter that drains the lower pelvis drains the lower two thirds of the kidney (Fig. 20-17). The ureter that drains the upper pole is ventral to the lower one but crosses over and empties into the bladder into a ureterocele in a lower, more medial ectopic location (Weigert-Meyer rule); when one of the ureters empties in an extravesical location, it is the one that drains the upper renal pelvis. The upper pole moiety is prone to obstruction at the ureterovesical junction and may be associated with an ectopic ureterocele. The lower pole moiety in a duplicated system is prone to vesicoureteral reflux caused by distortion of the orthotopically located ureteral orifice.
FIG. 20-17. Duplication of the pelvis and ureter. The upper pelvis drains the upper pole of the kidney, whereas the lower pelvis drains the central portion of the kidney as well as the lower pole.
These anomalies of the pelvis and ureter may be unilateral or bilateral, with a tendency to be asymmetrical. Multiple budding occasionally results in multiple short upper pelves and ureters that are extrarenal in type. In this anomaly, each of several major calyces has its own pelvis and upper ureter, which usually joins with the others to form a common lower ureter. It is not uncommon for half of a double ureter to be obstructed.
The radiographic findings of ureteral duplication are varied depending on the degree of obstruction and reflux that are present. If both ureters fill with contrast material, the diagnosis is usually simple. Commonly, however, the upper pole becomes obstructed and enlarged, and a nonfunctioning upper pole renal mass effect is the result.
The lower opacified calyces may be displaced by the large upper pole mass (“drooping lily sign”). There may also be some rotation of the kidney, the amount depending on the size of the mass. If the mass is very large, the entire kidney and upper ureter may be displaced laterally. Calculi may occur in the obstructed or infected upper pole.
When obstruction and infection result in nonfunctioning of the upper pole, the roentgenographic findings are varied. If there is a nonobstructive inflammatory lesion resulting in nonfunction, the findings are those of a calyceal system that drains only the central and lower pole of the kidney, so the calyces are fewer than normal in number and the most superior calyx does not extend into the upper pole of the kidney. To make this determination, one must obtain clear visualization of the outline of the upper pole.

Anomalies in Position of Ureteral Orifice
There are several possible anomalies in position of the ureteral orifice. This variation is usually better studied by cystoscopy than by radiographic means. In the male, the ureter may open into the seminal vesicles, the vas deferens, the ejaculatory duct, or the posterior urethra. In the female, the abnormal ureter may open into the urethra; beneath the urethral orifice near the hymen; or into the lateral vulvar wall, the uterus, the vagina, or, rarely, the rectum. Although ectopic ureteral insertion is present in all patients with ureteral duplication, it also occurs in patients with a single ureter. The sites, symptoms, and radiographic findings are similar except that there is no duplication.
Ectopic ureteral insertion usually is associated with urinary incontinence in females resulting from insertion of the ureteral orifice below the urinary sphincteric mechanism. This is not the case with males because of the insertion of the ureteral orifice above the urinary sphincter. However, ectopic insertion in boys can cause urinary tract infection or prostatitis at a young age.
Ureteral Jet Phenomenon
The ureteral jet phenomenon can be seen on CT or IVP and is caused by a jet of opaque medium propelled by ureteral peristalsis, which may occasionally extend across the base of the bladder to the opposite side. The jet maintains the caliber of the ureter and simulates an anomalous ureter that opens on the opposite side of the trigone (Fig. 20-18). When present, it excludes the possibility of significant vesicoureteral reflux or ureteral obstruction.99 If there is a question about the cause of an apparent anomaly, another film will reveal a normal lower ureter in these patients. It is also possible to visualize a ureteral jet on color Doppler ultrasound examination of the bladder. Some authors believe that presence of a ureteral jet excludes functionally significant ureteral obstruction.45
FIG. 20-18. Ureteral jet phenomenon. Note the apparent extension of the ureter across the midline. Cystoscopy revealed a normal position of the ureteral orifice on the left.
Retrocaval Ureter
Postcaval or retrocaval ureter is limited to the right side except in situs inversus. It is caused by failure of the right subcardinal vein to atrophy and its persistence as the adult vena cava. Normally, the right supracardinal vein persists as the vena cava. The abnormal relationship may cause partial obstruction, leading to hydronephrosis, infection, and calculus formation. The ureter passes to the left, behind the inferior vena cava, then turns toward the right and courses downward in its normal position. In some cases there is redundancy of the ureter proximally, so that an S-type, fish-hook, or inverted J-type of deformity is produced. The site of narrowing or obstruction, if present, is proximal to the vena cava and at the lateral edge of the psoas, and it is caused by the pressure of the retroperitoneal fascia over the muscle. In other cases of retrocaval ureter with no redundancy, obstruction is less common and, when present, coincides with the lateral margin of the inferior vena cava.
The diagnosis can usually be made on urography. In addition to the abnormal course of the right ureter in the frontal projection, its posterior position in the lateral view can be observed. The diagnosis can be confirmed by inferior vena cavography with an opaque catheter in the ureter but is more easily demonstrated by CT. The medial swing is usually maximum at L4-L5 and occasionally as high as L3. There may be partial obstruction at the level of the lateral wall of the vena cava. Medial deviation of the ureter (medial to the vertebral pedicles) may also be related to psoas muscle prominence associated with a narrow pelvic inlet. Retroperitoneal fibrosis and abdominal aortic aneurysm may also cause medial deviation. The deviations in the retroperitoneal fibrosis are bilateral, and there is no S- or J-shaped redundancy. Retroperitoneal masses (e.g., lymphoma) with ureteral displacement must also be differentiated, and CT is particularly helpful for this determination.
There are two types of ureterocele, simple and ectopic.189 The simple ureterocele consists of an intravesical dilatation of the ureter immediately proximal to its orifice in the bladder. It usually results from a combination of ureteral orifice stenosis and a deficiency in the connective tissue attachment of the ureter to the bladder. It varies in size from a scarcely perceptible dilatation to one that is moderately large and resembles a cobra head (or spring onion) in shape. There may be partial obstruction resulting in ureterectasis. In general, the simple ureterocele is smaller than the ectopic ureterocele. It occurs with equal frequency in males and females and is usually discovered incidentally. A calculus in the intramural

portion of the ureter may produce dilatation simulating ureterocele, but the calculus produces pain and is usually visible on plain films. A tumor of the bladder, either primary or secondary, may also cause dilatation simulating simple ureterocele, a so-called pseudoureterocele.
The ectopic type of ureterocele is usually discovered in childhood and is much more common in girls (6:1 or 7:1) than in boys. It is more likely to be associated with severe hydronephrosis, ureterectasis, or infection than the simple type. Both types tend to occur in the presence of duplication of the ureter; the ectopic type is almost always associated with this anomaly. The ectopic ureterocele consists of the submucosal passage of the distal portion of the involved ureter within the vesical wall to terminate in the urethra rather than in the bladder, as in the simple type. The submucosal portion of the ureter dilates and bulges anteromedially into the bladder to form the ureterocele. It may prolapse through the urethra to form a vulvar “cyst,” and it usually extends posteriorly to the vesical neck and proximal urethra. It invariably involves the ureter from the upper pole of the kidney (Fig. 20-19).
FIG. 20-19. Ectopic ureterocele. A: Note the large, rounded mass encroaching on the bladder, chiefly on the left side. The collecting system on the left drains the lower left kidney. B: Retrograde pyelogram in which a ureteral orifice in the normal position was catheterized shows drainage of the lower kidney. The ectopic ureter draining into the ureterocele could not be catheterized. It drained the upper pole in the left kidney and was obstructed so that no function was present at the time of the urogram. C: Upper pole of the kidney on the left (arrows).
The roentgenographic appearance of the simple type depends on whether the opaque medium fills the ureterocele. If it is filled, the lesion is outlined by a radiolucent wall that stands out in contrast to the filled bladder and to the filled, dilated, distal ureter. When the ureterocele is not filled with opaque material, it appears as a radiolucent mass within the opacified bladder in the region of the ureteral orifice. The shape may be somewhat fusiform with a narrow lower end resembling a cobra’s head, but the larger ones tend to be more rounded in shape. When a calculus is present in the ureterocele, it is noted to lie on one side of the midline and remains there despite changes in the patient’s position.
Ectopic ureteroceles are larger than the simple ones and often extend to the anterior bladder wall when viewed in the lateral projection. The contact with the floor of the bladder is broad and extends to the internal urethral orifice. Obstruction of the other ureter is frequent, and the extravesical portion may distort the bladder. Several conditions can simulate ectopic ureterocele. These include hydrometrocolpos and “cyst” of the seminal vesicle produced when an ectopic ureter inserts into the seminal vesicle. The condition is less common in males than in females, but in males the incidence of infection is higher, the malformation is more complex, and the frequency of a single collecting system is greater. Eversion is also more common in males, and the tendency to prolapse into the posterior urethra, causing bladder outlet obstruction, is greater.
EXU is the roentgenographic method of choice in diagnosis of ureterocele. An eccentric mass encroaching on the bladder floor in a patient with duplication of the ureter is virtually pathognomonic.
There is an acquired condition, called a pseudoureterocele, that simulates a simple ureterocele.187 It is usually encountered on EXU. The contrast-delineated distal ureter simulates a simple ureterocele. On closer inspection, however, this dilatation usually is slightly asymmetrical. There may even be a very small eccentric beaked appearance of the tip of the contrast column in the ureter. It is important to distinguish a pseudoureterocele from a true ureterocele, because the pseudoureterocele may be the first indication of a neoplasm. Usually this is a transitional cell carcinoma of the bladder or invasion of the trigone area by carcinoma of the cervix. Benign conditions can also cause a similar appearance and include fibrosis of the ureteral orifice secondary to transient impaction of a ureteral calculus or injury from previous transurethral resection of bladder pathology. Cystoscopy and re-examination of the patient after discovery of the pseudoureterocele usually reveal its cause.
Ureteral Diverticula
A single ureteral diverticulum is probably a congenital anomaly and may represent a dilated rudimentary branched ureter.38 When the diverticulum is filled with contrast medium, the diagnosis is easily made because the appearance is similar to that of a diverticulum elsewhere. Some of these diverticula have the appearance of a blind-end duplication without much dilatation and almost certainly are rudimentary or partially duplicated ureters. They are best demonstrated by retrograde pyelography but may be apparent on EXU.
Most authorities believe that multiple diverticula are almost always acquired and are indicative of previous infection. They appear as ureteral outpouchings of various sizes and numbers with associated ureteral strictures, best seen on a retrograde pyelogram, but with good ureteral filling they may also be clearly defined on EXU. Importantly, ureteral diverticulosis is associated with an increased incidence of ureteral metaplasia and ureteral tumors.33
Other Ureteral Anomalies
Transverse Ureteral Folds
In infants, a corkscrew appearance may be demonstrated in the upper ureters on EXU. This appearance is caused by thin, transverse folds that represent inward projections of the full thickness of the ureteral wall. They appear as horizontal folds measuring about 1 mm in thickness on the urogram, probably represent persistence of fetal tortuousity of the ureter, are of no clinical significance, and represent a minor anatomic variant that occasionally persists into adolescence.
Vertical Ureteropelvic Striations
The vertical striations occasionally observed in the pelvis and upper ureter usually are associated with reflux and are probably secondary to infection and mucosal edema. In rare instances, however, they appear to be a minor anatomic variant. They may be observed on EXU or retrograde pyelography.
Ureteral Valves
A ureteral valve is a very unusual anomaly that is manifested by the following: (1) anatomically

demonstrable transverse folds of ureteral mucosa containing bundles of smooth muscle fibers; (2) obstruction above the valve and a normal ureter below it; and (3) no other evidence of mechanical or functional obstruction. A ureteral valve is usually unilateral; it may be annular with a pinpoint opening, or it may be cusp-like in appearance. It may occur anywhere in the ureter, although it is slightly more common in the lower ureter than elsewhere. The cause of this anomaly is uncertain.1
Patent Urachus and Urachal Cyst
The urachus represents the intra-abdominal remnant of the allantoic duct or caudal extension of it, which is continuous with the vesical portion of the urogenital sinus in embryologic development. Normally it constitutes the middle umbilical ligament. The allantois extends from the primitive urinary bladder through the umbilicus to the placenta. Four types of anomalies are possible: (1) complete patency (patent urachus), (2) patency at umbilical end or blind external type (urachal sinus), (3)

patency at vesical end or blind internal type (urachal diverticulum), and (4) a urachal remnant that is obstructed at both ends (urachal cyst). Urachal prominence or urachal remnant is observed fairly often in patients with high intravesical pressure dating from birth or before birth. Examples are patients with myelomeningocele or posterior urethral valves. The blind external type and complete patency are usually recognized on inspection when the umbilical cord sloughs off but may be suspected earlier because of leakage of material from the umbilicus. Urachal cysts often go undetected unless they become infected, in which case the patient may present with fever, leukocytosis, and signs of sepsis (Fig. 20-20). Long-term complications of urachal remnants include urachal carcinoma. This is an adenocarcinoma located at the dome of the bladder.
FIG. 20-20. Infected urachal cyst. A: Midline longitudinal view of the pelvis demonstrates a complex mass immediately superior to the bladder in a septic 18-month-old patient. B: Computed tomographic scan through the level of the umbilicus confirms the presence of a complex mass in the region of the urachus.
Roentgen visualization can be obtained by using contrast materials that can be injected into the umbilical end of the urachus. Cystography is needed to demonstrate the effect of a urachal cyst or patent urachus. The findings are those of a smooth-walled tubular structure lying in the anterior midline that extends into the plane of a line between the umbilicus and the bladder. The bladder may be distorted and elevated. The cyst may extend from the bladder to the umbilicus or end blindly when it begins at either end. When a cyst of the urachus is present without internal or external communication, roentgenographic findings depend on its size. If large, it may be noted as a midline, soft-tissue mass lying between the bladder and the umbilicus in the anterior abdominal wall. Gas-filled small intestine may be displaced, and study of the small bowel by means of barium meal will show comparable displacement. Rarely, calculi may form in a patent urachus or urachal cyst. CT and ultrasonography provide more complete assessment of the extent of this urachal anomaly.101
Regardless of its cause, chronic obstruction of the urinary tract leads to hydronephrosis, which is dilatation of the pelvis and calyces with potential progressive destruction of renal parenchyma. The terms pyelectasis, caliectasis, ureterectasis, and hydroureter are more accurate in designating the location of the dilatation. The obstruction that produces hydronephrosis may be unilateral or bilateral, depending on the site of the lesion producing it. Unilateral obstruction is caused by a lesion at or above the ureterovesical junction, whereas bilateral obstruction may be caused by a lesion distal to that point. Enlargement of the urinary collecting system, including pelves, calyces, ureters, and bladder, may also result from causes other than obstruction.
Nonobstructive Hydronephrosis (Urinary Stasis)
Several nonobstructive conditions can cause dilatation of the renal pelvis, calyces, and ureters. Diabetes insipidus may be associated with relatively moderate hydronephrosis. Nephrogenic diabetes insipidus tends to cause more severe dilatation, often with tortuousity of the ureters in addition to the dilatation. In this condition there is a tubular abnormality with insufficient absorption of water, leading to a large volume of hypotonic urine.125 Urinary tract infection tends to cause segmental or generalized dilatation of the ureter, with poor or reversed peristalsis leading to pyelectasis and caliectasis. This may be augmented by vesicoureteral reflux, which is commonly found in association with urinary infections. The changes may decrease or disappear after the infection is successfully treated. Dilatation with stasis in the absence of urinary tract abnormality may also be caused by

intra-abdominal inflammatory disease such as appendicitis or peritonitis, a finding similar to that of adynamic ileus involving the gut in patients with peritonitis. Excessive fluid intake (overhydration) may cause some dilatation. A variety of neurologic disorders are also associated with dilatation without obstruction. An adynamic, short segment of upper ureter can cause some dilatation of the pelves and calyces; this appears on a urogram as a short, narrow ureteral segment with dilatation above it.
Congenital Hydronephrosis
Congenital hydronephrosis is the most common cause of an abdominal mass in neonates. It is caused by a variety of lesions. Because many of these can be found in combination with other genitourinary anomalies, discovery of one anomaly should prompt close examination of the remainder of the genitourinary tract. The most common cause of congenital hydronephrosis is usually an obstruction at the UPJ. However, vesicoureteral reflux, congenital ureteroceles with obstruction, urethral valves, congenital strictures, and bands also are known causes of hydronephrosis. In addition, there are instances of congenital hydronephrosis in which the cause is obscure; many of these are “neurogenic” in that they are associated with lesions of the spinal cord and with congenital megacolon. The dilatation is usually bilaterally symmetrical in patients with congenital megacolon. Therefore, congenital abnormalities may result in either obstructive or nonobstructive uropathy.
In neonates, ascites at birth may indicate obstructive uropathy, often secondary to posterior urethral valves, but a variety of lesions may cause the obstruction.69 Obstruction of the bladder outlet, ureteral atresia, presacral neuroblastoma, complex caudal anomalies including urethral and anorectal atresia, ureterocele, vesical neck valve, and myelomeningocele have also been reported as rare causes of neonatal ascites secondary to obstructive uropathy. If congenital hydronephrosis is severe, oligohydramnios may develop. This can lead to pulmonary hypoplasia, which may cause death depending on the severity. Mechanical ventilation can lead to pneumothorax or pulmonary interstitial emphysema (PIE) because of the elevated ventilatory pressures required to provide sufficient gas exchange.
Rarely, UPJ obstruction may produce intermittent hydronephrosis related to overhydration in patients who have an extrarenal-type pelvis. In patients with symptoms of intermittent hydronephrosis, urography at the time of acute dilation after overhydration may confirm the diagnosis. As indicated previously, duplication with ectopic ureteral insertion into the urethra often results in hydronephrosis of the upper collecting system. This is a congenital anomaly, but acquired disease such as infection may be the major presenting sign when obstruction is observed in the adult. Rarely, there is lower-pole hydronephrosis in a duplicated kidney.
Acquired Hydronephrosis (Obstructive Uropathy)
Acquired hydronephrosis is caused by a variety of lesions, among which are tumors, calculi, strictures, radiation therapy, operative procedures, and prostatic enlargement. UPJ obstruction is the most common type of bilateral obstruction above the bladder. It may be asymmetrical. Congenital narrowing appears to be the most common cause of the obstruction. Pregnancy in the third trimester is often associated with hydronephrosis that tends to be more severe on the right than on the left. Ureters are dilated to the pelvic brim. The most likely cause is mechanical pressure from the enlarging uterus. Hydrocolpos and hydrometrocolpos also tend to cause ureteral obstruction. Abdominal aortic aneurysm may compress the ureter, or retroperitoneal bleeding (associated with aneurysm) may cause fibrosis leading to ureteral stricture and hydronephrosis. Granulomatous (Crohn’s) disease of the small intestine or colon occasionally causes distal ureteral obstruction or ureterointestinal fistulae. There are all degrees of dilatation, and progression of the changes can be noted on serial examinations if the obstruction is not relieved.
Imaging Findings
Ultrasound is considered the test of first choice to evaluate patients with suspected hydronephrosis. Ultrasound is particularly sensitive to small amounts of fluid and therefore is well suited to detection of a dilated collecting system. Hydronephrosis is graded at ultrasound as mild, moderate, or severe depending on the morphologic findings, which may not parallel the degree of obstruction or the functional significance. A potential pitfall in the ultrasound diagnosis of hydronephrosis is the parapelvic cysts that may surround a nondilated renal pelvis. A contrast-enhanced study such as IVP or CT can confirm the presence of parapelvic cysts if necessary. Although controversial, work with Doppler ultrasound shows potential to be used to diagnose early renal obstruction.151, 191 Work by Platt and colleagues151 calculated the resistive index (RI), which is equal to the peak systolic frequency shift minus the minimum diastolic frequency shift, divided by the peak systolic frequency shift. RIs higher than 0.70, or an elevation of 0.10 above the asymptomatic side, are suggestive of acute renal obstruction. CT often yields more specific information than urography as to the cause of obstruction, particularly when the cause is extraureteral (e.g., metastatic tumor).
The earliest urographic change in hydronephrosis is a flattening of the normal concavity of the calyx and a blunting of the sharp peripheral angle produced by the papilla as it juts into the calyx. This early change is reversible and is readily produced by a small increase in pressure. The blunting of calyces found in early hydronephrosis is accompanied by a decrease in the rate of contrast material accumulating in the collecting system. Tomograms obtained immediately after injection of contrast material demonstrate an asymmetrical

delay of caliceal opacification on the partially obstructed side. As the obstruction becomes higher grade or more prolonged, the pelvis enlarges gradually, but pelvic and calyceal dilatation are not necessarily parallel. The next calyceal change is that of “clubbing,” in which the concavity produced by the papilla is reversed (Figs. 20-21 and 20-22). Calyces then gradually enlarge, with progressive destruction of parenchyma and enlargement of the collecting system. A prolonged and increasingly dense nephrogram is also quite characteristic of acute renal obstruction. Prolonged, high-grade obstruction eventually causes marked collecting system dilatation, until the kidney becomes a nonfunctioning hydronephrotic sac in which the normal anatomy is obliterated (Fig. 20-23).
FIG. 20-21. Minimal bilateral hydronephrosis. The pelves are not enlarged, but there is a little blunting of the calyces. Note the minimal pyelolymphatic backflow on the right (arrow).
FIG. 20-22. Bilateral hydronephrosis showing the value of delayed films. A: Intravenous urogram, obtained 15 minutes after injection of a contrast agent, shows dilatation of pelves and calyces with no definite ureteral opacification. B: This film, exposed 90 minutes after injection of the medium, shows dilatation of ureters extending down to stricture-like narrowing, which is a little higher on the right than on the left.
FIG. 20-23. Massive hydronephrosis in a child. The greatly dilated pelvis is opacified. It almost fills the entire left abdominal cavity. Physical findings were those of a large, somewhat fluctuant abdominal mass on the left.
Occasionally, acute obstruction leads to rupture of the collecting system, usually at a calyceal fornix. Extravasation of urine into the retroperitoneum occurs and can track along the psoas muscles. These patients do well clinically if the obstruction is promptly removed. Long-standing obstruction and urine leakage can lead to sizable urine collections (urinomas), which may require percutaneous or surgical drainage, particularly if complicated by superimposed infection.
Renal function may be greatly diminished in severe hydronephrosis, and there is accumulation of opaque material in

the parenchyma adjacent to the grossly dilated calyces. This forms crescentic areas of faint opacification, termed the crescent sign of hydronephrosis (Fig. 20-24). Later there may be faint opacification of the calcyces themselves. Infection can be a complicating factor; it tends to accelerate parenchymal destruction and cause signs and symptoms of sepsis. An obstructed, prefilled collecting system is a true medical emergency requiring rapid decompression, usually by a percutaneous nephrostomy tube. Radiographically, infection produces more irregularity in the dilated calyces than is seen in uncomplicated hydronephrosis. Also there may be bleeding with clots in the dilated collecting system, which may resemble intrapelvic tumor. Ultrasound may show echogenic material in a dilated collecting system, and CT images may also demonstrate material in the collecting system as well as a thick-walled renal pelvis.
FIG. 20-24. Crescent sign of hydronephrosis. A: Selective left renal arteriogram shows grossly stretched, narrow vessels with very few branches. Note the vessels stretched over the large renal pelvis medially. B: Later film shows the crescent sign caused by opacification in the thin rim of remaining renal tissue. Surgical removal confirmed the diagnosis of severe hydronephrosis. (Courtesy of Thomas L. Carter, M.D., and Richard Logan, M.D.)
In the evaluation of patients with dilatation of the pelvis and calyces, particularly children, it is important that the bladder be emptied before the urogram is obtained. A distended bladder may result in a false appearance of hydronephrosis.13 When the bladder is empty, the hydronephrosis disappears. This finding is also seen during ultrasound examinations of the kidneys, and therefore the bladder should be empty before the diagnosis of hydronephrosis is made. Vesicoureteral reflux may accentuate the dilatation of the

upper urinary tract in these patients, but it does not appear to be the major cause.
When severe obstruction persists, there is usually increasing hydronephrosis, with so-called hydronephrotic atrophy resulting in loss of renal parenchyma of varying degrees. In some instances, function returns after relatively long periods of obstruction, provided that a reasonable amount of renal parenchyma remains. A combination of obstruction and ischemia occasionally results in decreased renal size after the obstruction is relieved. Ultrasonography can be especially helpful when contrast material is not excreted by the kidney (Fig. 20-25). Ultrasound is also useful in assessing the amount of renal parenchymal atrophy or scarring in long-standing cases of hydronephrosis.
FIG. 20-25. Hydronephrosis in a renal transplant. Sonogram shows the dilated pelvocalyceal system and proximal ureter. The small anechoic space adjacent to the kidney is the bladder.
It is thought that upper urinary tract calculi originate as Randall’s plaques deep in the lining of the collecting ducts in the renal papillae.186 These may detach and pass into the renal collecting system. Calculi may lodge in the pelvis, often in the region of the papillae and calyces. The calculi may remain within the pelvis and gradually increase in size to form a cast of the pelvis and calyces, representing a staghorn calculus (Fig. 20-26). Multiple calculi may form within the calyceal system; they may be similar, or they may vary considerably in size. Urinary stasis and infection are important factors in promoting the formation of calculi, but the exact cause is not certain in many instances. Calculi tend to be asymptomatic until they cause obstruction. Then typical

renal or ureteral colic symptoms are produced. About 90% of upper tract calculi contain enough calcium to be visualized on plain-film roentgenograms. Calcium phosphate, calcium oxalate, and magnesium ammonium phosphate (struvite) stones are the most common. Stones usually are composed of a mixture of chemical compounds; pure stones are relatively rare. Diamonium calcium phosphate and magnesium phosphate stones are uncommon. Cystine, urate, and xanthine stones are rare and often of low density.
FIG. 20-26. Staghorn calculi. Note the calcification forming a cast of the pelvis and calyces on each side. Renal function was so poor that very little additional density was observed on these urograms. The ureters are faintly opacified, however.
Matrix calculi are a combination of about two thirds mucoprotein and one third mucopolysaccharide; they are radiolucent and usually form in the presence of Proteus infection. This amorphous mucoprotein is present in stone formers, and, since it is not present in normal urine, it probably plays a role in the formation of renal calculi. Hyperparathyroidism and other conditions with hypercalcemia, including some that cause dissolution of bone, may also be associated with calculi. These include osteolytic metastases, leukemia, multiple myeloma, and sarcoidosis. Gout and other conditions associated with high serum uric acid and hyperuricosuria increase the incidence of uric acid stones. Hyperoxaluria, whatever the cause, also tends to promote formation of renal calculi. There is some evidence to indicate that calculi may also result from RAS; the vascular insufficiency may cause parenchymal injury leading to calculus formation.
Roentgenographic Findings
The roentgenographic findings are those of an opacity of varying size and shape overlying the urinary tract. Often the plain-film diagnosis is easily made, particularly when the calculus forms a cast of the pelvis or calyces or both. Subsequent EXU is often used for localization and to determine the condition of the calyceal system. Oblique views may be necessary in addition to frontal projections to localize a calculus definitively. Urography, ultrasound, or CT may be necessary to find radiolucent low-density calculi. These calculi appear as negative shadows displacing the opaque contrast medium. On CT, radiolucent stones are easily visible as dense masses (more than 80 Hounsfield units) in the urinary tract. Stones of all types are readily apparent on ultrasound provided they are of sufficient size (larger than 5 to 10 mm), depending on the frequency of the transducer, the location in the urinary tract, and the patient’s body habitus.

Ureteral calculi cause a well-defined, echogenic focus with posterior acoustic shadowing on ultrasound.
Patients with renal or ureteral colic usually have delayed excretion by the involved kidney. With the acute obstruction produced by the passage of a ureteral stone, the intrapelvic pressure increases to the point at which there is decreased glomerular filtration of contrast material. Increasing density of the kidney (nephrogram) is caused by slowed flow of urine through the collecting ducts in the parenchyma, and ongoing obligatory tubular water resorption in the nephrons results in increasing concentration of opacified urine.181 Eventually there usually is some opacification of the calyces, pelvis, and ureter. It is, therefore, important to obtain films until opacification is adequate to make the diagnosis. If the urogram shows a prolonged density on the involved side, it is likely that serial films exposed at 30-minute or progressively longer intervals will show enough opacification to localize the site of the obstructing ureteral calculus and confirm its presence within the ureter (Fig. 20-27). Prone and upright films can also be obtained to aid in localizing the site of obstruction. Rarely, it may be necessary to do cystoscopy and place an opaque catheter. The calculus can be localized in relation to the ureteral catheter by means of frontal and oblique roentgenograms.
FIG. 20-27. Right ureteral calculus. A: Note the density (arrow) just above the right iliac crest in this right posterior oblique projection. B: Urogram shows slight dilatation of the ureter extending down to the site of the calculus. This roentgenogram was obtained at 90 minutes after intravenous injection of contrast material. Earlier films showed no excretion on the right side, demonstrating that delayed films were essential to confirm the diagnosis of ureteral calculus in this patient.
Nonenhanced helical CT urography is now more frequently used in the diagnosis of renal and ureteral calculi. This technique does not require use of intravenous or oral contrast material and is therefore quite rapid to perform. Signs of an obstructing ureteral calculus include an ipsilateral enlarged kidney with fluid and soft tissue stranding around it, an enlarged renal pelvis, a dilated ureter above the calculus, and a normal ureter below. In addition, direct visualization of a renal calculus is possible in more cases than with IVP, because of the superior contrast resolution of CT (Fig. 20-28). Evidence suggests that CT may be superior to IVP for the detection of calculi that cause minimal degrees of ureteral obstruction. An additional advantage of CT urography is the ability to detect other causes of flank pain unrelated to urinary calculi.172
FIG. 20-28. Radiolucent ureteral stone. A: Filling defect in left ureter (arrow) was not visible before contrast enhancement. B: Computed tomographic scan through the level of the filling defect demonstrates a high-attenuation mass in the ureteral lumen, confirming the diagnosis of a radiolucent stone. The term “radiolucent” applies only to the plain radiographic appearance of these stones.
The most common lodging site for ureteral calculi is at

or above the ureterovesical junction in the pelvic portion of the ureter. Occasionally a calculus is passed before the examination is completed and no obstruction is then visible. If the calculus has lodged at the ureterovesical junction for any length of time, it is not uncommon to note a localized radiolucent indentation on the bladder caused by edema of the trigone above the ureteral orifice, even though the calculus may have been passed. Ureteral calculi are usually small in size (1 to 3 mm in diameter). In general, these small stones pass quickly down the ureter to lodge at or near the ureterovesical junction. The great majority of stones pass in 72 to 96 hours.186 The stones tend to be parallel to the course of the ureter when they are oval or elongated. Most lie above a line drawn through the ischial spines. However, angulation of the roentgen-ray tube or alteration of position of the pelvis may project these calculi lower in position. Larger calculi are not as likely to leave the renal pelvis and become lodged in the ureter.
Ureteral calculi tend to be round or oval in shape. If the calculus remains within the ureter for a long time, it may become elongated and increased in size from deposition of urinary sediment. Large stones found within the ureter usually have been present for a considerable period of time (Fig. 20-29). In patients with acute colic caused by an obstructing ureteral calculus, urography may demonstrate the classic findings of hydronephrosis, as described previously. In addition, pyelointerstitial backflow may be seen. Occasionally, forniceal rupture can lead to urine extravasation and urinoma formation, as previously discussed.
FIG. 20-29. Multiple large ureteral calculi on the left. The ureteral catheter indicates the relationship of the calculi to the ureter. The bladder is outlined by air.
Differential Diagnosis
Suspected renal or ureteral calculus must be differentiated from other calcifications that can occur in the renal areas and along the course of the ureters. CT urography is a simple method for separating the various abdominal calcifications, although conventional radiographs and IVP in combination can virtually always perform this function as well. Gallstones are usually multiple, tend to be faceted, and often exhibit typical concentric rings of calcium. Oblique roentgenograms show their anterior position. Common duct and cystic duct stones may be opaque, but they also lie anterior to the kidney and ureter. Calcification of costal cartilages is common and usually readily identified. Oblique projections show the relationships of such shadows to the anterior lower thoracic wall if there is any doubt about their nature. Calcified mesenteric nodes and calcifications in the appendices epiploicae usually move enough from one position to another to be differentiated from urinary calculi. The same is true of opaque material in the gastrointestinal tract. Pancreatic calculi usually conform to the shape and location of the pancreas and can be identified readily. Calcifications in cysts and tumors of the kidney and elsewhere in the abdomen also must be differentiated. The contour of the cyst wall usually can be identified, and when a calcified tumor is present, it is usually large enough to be visualized as a soft-tissue mass. Occasionally the lateral tip of a transverse process of one of the lumbar vertebrae is easily visible in comparison with the remainder of the process and resembles a ureteral calculus; close inspection suffices to make the differentiation.
Vascular calcification, either in pelvic arteries or in veins (phleboliths), is generally the most difficult calcification to differentiate from ureteral calculi. Arterial calcification usually occurs along the course of a large artery and tends to be elongated and to outline the arterial walls, forming a ring-like density when seen in cross-section and parallel lines when seen in longitudinal section. Phleboliths often have a fairly typical appearance, with a radiolucent central area, and tend to be more rounded in contour than calculi; some have a central calcific nidus surrounded by a zone of lesser density, which in turn is surrounded by a denser periphery. Roentgenograms obtained in anteroposterior and oblique projections are often sufficient to exclude the possibility of urinary calculus as the cause for the density or densities present. If they are not conclusive, EXU with oblique and special views, including delayed roentgenograms or fluoroscopy, usually provides the diagnosis. If there still is doubt, CT urography, contrast-enhanced CT, or cystoscopy with introduction of a radiopaque catheter into the ureter usually solves the problem. The ureteral stone stays in relation to the contrast medium in the ureter in all projections.

In any patient having symptoms suggestive of ureteral colic, importance should be attached to any calcific density, no matter how small, occurring along the course of the ureter, particularly if it is found in the distal part of the ureter. Conversely, in a patient with no symptoms suggestive of ureteral colic, small, rounded calcifications in the lateral aspect of the pelvis can usually be disregarded because they probably represent phleboliths. Occasionally, impacted ureteral calculi do not cause dilatation of the ureter or collecting system.200
Renal Milk of Calcium
The term milk of calcium refers to a suspension of fine sediment containing calcium that is observed most often in a calyceal diverticulum or hydrocalyx with little or no drainage or in a so-called pyelogenic or calyceal cyst.135 Films exposed with the patient erect show a horizontal level, indicating that the calcium is in suspension. The appearance is similar to that observed in the milk-of-calcium gallbladder. Similar suspension of liquid or semisolid calcium has been observed in association with renal cysts. Rarely, it has been seen in association with hydronephrosis. On upright films, several levels of calcium are noted in calyces.
Findings on plain films that suggest the diagnosis include (1) a somewhat peripheral location, compared with the central location of stones in the collecting system; (2) an unusually large area; (3) circular, or almost circular, configuration; (4) faint calcification, particularly in relation to size; (5) diminishing density toward the periphery; and (6) indistinct margins. When these findings suggest the possibility, an upright or decubitus film can be obtained to confirm the diagnosis.
Nephrocalcinosis refers to multiple calcium deposits within the renal parenchyma. Two forms of nephrocalcinosis have been described: cortical and medullary. The cortical type, which is the more uncommon of these conditions, is associated with renal cortical necrosis. Causes of this rare condition include hypotension (often from obstetric complications), chronic renal transplant rejection, chronic glomerulonephritis, Alport’s syndrome (chronic hereditary nephritis), and oxalosis. The common pathogenesis of all these conditions is renal disease in which calcium is precipitated in damaged renal cortical tissue. Blood calcium levels usually are normal. Imaging findings of renal cortical necrosis are best shown by CT and ultrasound, although the MRI findings have been described. In acute renal cortical necrosis, the renal cortex becomes abnormally hypodense or hypoechoic. During the healing phase, crescentic or stippled calcifications are deposited in the renal cortex.
Medullary nephrocalcinosis is found in association with several diseases characterized by abnormally high concentrations of calcium or phosphorus which result in precipitation of calcium phosphate in healthy renal tissue. Primary hyperparathyroidism is the best example of this group, and nephrocalcinosis occurs in approximately 25% of patients with the disease. Renal lithiasis, however, is more common than nephrocalcinosis. When the latter occurs, tiny calcifications confined to the medulla are usually present, with occasional larger calcifications occurring in the renal pyramids. Hypercalciuria of undetermined cause, hyperchloremic acidosis, hypervitaminosis D, milk-alkali syndrome, sarcoidosis, renal tubular acidosis, hyperoxaluria, carcinoma metastatic to bone, regional enteritis with secondary enteric hyperoxaluria, and idiopathic hypercalcemia are other conditions producing this type of nephrocalcinosis. Medullary sponge kidney can also cause medullary nephrocalcinosis (see later discussion).
Roentgenographic findings depend on the extent of calcification, which varies from faintly visible granular densities to stippled calcification in the renal papilla and cortex (Fig. 20-30). The findings are relatively rare, and there are many instances of histopathologically proven renal calcification in which the calcium cannot be visualized radiographically in the living subject. Films of good quality in the low-kilovoltage range (70 to 76 kVp), exposed before a contrast medium is given, are necessary to demonstrate small amounts of calcium. Coned views that include only the renal area, as well as oblique views, are often necessary to localize the calcium within the kidney. Tomography, CT, and ultrasound are also very helpful in demonstrating and localizing calcifications in the kidney. At times, calcification is seen on tomograms or CT when it is not visible on the plain film. This is particularly true of low-density stones, which are best confirmed on CT.
FIG. 20-30. Nephrocalcinosis. Note the bilateral stippled renal calcific densities in this patient with renal tubular acidosis. There is overlying calcification of the costal cartilages.
Renal Tubular Acidosis
Nephrocalcinosis and nephrolithiasis are the radiographic findings in patients with renal tubular acidosis.36 The nephrocalcinosis is manifested by dense calcium deposits in the medullary portion of the kidney. Those patients who lose calcium also have osteomalacia. The calcifications occur chiefly in patients with distal tubular acidosis and not in those with only slight hypercalciuria. Radiographically, the dense medullary calcifications are somewhat similar to those observed in medullary sponge kidney, but the individual calcifications are larger in renal tubular acidosis and have less tendency to be oval or elongated than in medullary sponge kidney. Also, the calcification is somewhat more widespread in some patients.
Low-density or Nonopaque Calculi
Although 85% to 90% of upper urinary calculi are opaque and readily seen on the plain roentgenogram, the remainder are not. These calculi are most often predominantly cystine, uric acid, or xanthine stones. Stones reach the ureter, cause colicky pain, and become impacted, causing varying degrees

of obstruction. EXU reveals the effects of the obstruction, but the deformity of the end of the ureteral contrast column often simply indicates an intraluminal radiolucent defect, which could be caused by calculus, tumor, or blood clot. In the past, retrograde pyelography usually was necessary to make diagnostic distinctions. However, CT is now the most reliable method for depicting the anatomy and distinguishing tumor from calculus from clot (Fig. 20-31). CT detects even minimal calcifications, with the CT number measuring in the range of 75 to 140 Hounsfield units. Tumor or other soft-tissue-like materials measure less than 60 Hounsfield units. There is very little overlap, so tumor can usually be excluded with great confidence when the CT number exceeds about 70 Hounsfield units.168
FIG. 20-31. Stone at the left ureterovesical junction. A: Computed tomographic urogram demonstrates a distended left pelvocalyceal system, enlarged kidney, and stranding in the perirenal space. B: Slice through the lower abdomen in the same patient shows a dilated left ureter (arrowhead), implying that the point of obstruction is below this level. C: Slice through the pelvis reveals a small, dense calculus (arrow) lodged at the ureterovesical junction.

Acute Pyelonephritis
Acute nontuberculous pyelonephritis, which is among the most frequently encountered acute renal infections, is the least serious of a spectrum of related acute infectious problems. This spectrum extends from acute pyelonephritis, to acute renal abscess, to acutely infected pre-existing spaces such as renal cysts or hydronephrotic kidneys. The pathogenesis is similar in all these entities. Bacteria reach the kidney by the hematogenous route or by the ascending route from the bladder via the ureter. The course of the acute renal infection is then determined by the aggressiveness of the infectious agent, the immune response of the patient, and the predisposing conditions (e.g., urinary obstruction).
The ordinary case of uncomplicated acute pyelonephritis is readily diagnosed by the clinical presentation of acute onset of flank pain and tenderness accompanied by sudden onset of substantial fever. These findings, coupled with bacteriuria and pyuria on urinalysis, usually confirm the diagnosis. No imaging is necessary to make treatment decisions unless the infection does not respond promptly to the usual antibiotic therapy. In that event, imaging (IVP, ultrasound, or CT) is used to determine whether the disease has progressed beyond simple acute pyelonephritis. Specifically, predisposing and complicating conditions such as obstruction, congenital anomaly, occult calculus, or renal abscess need to be excluded.
Acute pyelonephritis shows positive urographic findings in about 25% of uncomplicated cases in which urography is done.188 Findings include renal enlargement, diminished intensity of the nephrogram, decreased calyceal contrast density, delayed calyceal appearance time, distortion and attenuation of the calyces and infundibula, and pyelocaliectasis. Renal enlargement is the most common finding, usually on the symptomatic side. Occasionally, the contralateral kidney is also enlarged.
CT better demonstrates positive findings. Unenhanced scans may be normal or may show regions of slightly decreased attenuation within the parenchyma. Contrast-enhanced scans demonstrate radially oriented or wedge-shaped regions of low attenuation that extend from the collecting system to the renal surface.85 CT is more sensitive than urography in detecting acute infectious parenchymal changes and is the imaging nodality of choice.188 Ultrasound is rarely useful in making the diagnosis of acute pyelonephritis, but it can help exclude significant complicating conditions such as hydronephrosis and renal stones. When present, ultrasound findings include renal enlargement and decreased echogenicity of the renal parenchyma, which may be diffuse or regional. Conventional color Doppler imaging and color Doppler energy have been investigated for the diagnosis of acute pyelonephritis, particularly the focal form. A focal nonperfused area of the kidney is suspicious for focal pyelonephritis in the appropriate clinical setting.
Occasionally, the acute pyelonephritic process may involve only a portion (or lobe) of the kidney. This is termed acute focal pyelonephritis. On urography, the only suggestive findings are a focal parenchymal contrast blush and focal renal enlargement. CT and ultrasonography demonstrate the findings previously described for pyelonephritis in a more focal distribution. Clinical information is very important in distinguishing acute focal pyelonephritis from solid renal tumors, which can have an identical imaging appearance. If doubt concerning the diagnosis still exists, follow-up imaging after an appropriate course of antibiotics usually excludes tumor diagnosis.
Emphysematous pyelonephritis is a rare special form of acute pyelonephritis affecting diabetics and patients with urinary tract obstruction. The finding of gas in and around the kidney in an acutely ill patient suggests the diagnosis. The affected kidney usually does not function well. Gas-forming organisms recovered include Escherichia coli and Proteus vulgaris. Emphysematous pyelonephritis should be regarded as a complication of a severe necrotizing infection, usually indicating extensive destruction of renal parenchyma (Fig. 20-32). Even with emergency surgical debridement, the mortality for patients with emphysematous pyelonephritis is very high.39
FIG. 20-32. Emphysematous pyelonephritis. Unenhanced computed tomogram shows bilateral collections of air in the kidneys (arrows). The right kidney is almost totally replaced with air.
Renal Abscess
Acute suppurative abscess of the renal parenchyma is rare, is usually hematogenous in origin, and begins in the cortex. Unless recognized and treated early, there is often extensive destruction of renal parenchyma. The most frequent causative organisms are ascending gram-negative bacteria. Skin infections and intravenous drug use can give rise to hematogenous abscesses caused by staphylococci, streptococci, and enterobacteria. When one or more small cortical abscesses develop in the parenchyma, no roentgenographic manifestations are present. If these small abscesses coalesce to form a large abscess, a plain-film roentgenogram often shows local enlargement of the kidney. The perirenal fat is blurred in the area of involvement so that the renal outline tends to be

indistinct. The involved kidney may be fixed during inspiration and expiration. The psoas muscle is often indistinct. There may be scoliosis with the concavity toward the involved side; this suggests the complication of perirenal abscess. EXU is of value if there is enough function to outline the calyceal system. The findings are those of compression and displacement or obliteration of the calyces by the tumor-like mass produced by the abscess. The cortical abscess may break through into the collecting system, to appear as a cavity communicating with a calyx and simulating tuberculosis. A peripheral abscess may also break through the renal capsule and produce a perirenal abscess (Fig. 20-33). Clinical signs of infection may not be present in some patients, particularly when the course is prolonged and the infection is chronic. Therefore, the differentiation from tumor may be difficult.
FIG. 20-33. Development of a renal abscess from acute focal pyelonephritis. A: Computed tomographic (CT) scan demonstrates an area of decreased attenuation in the left kidney caused by focal acute pyelonephritis. B: CT scan 2 weeks later, after the patient discontinued antibiotics, now shows areas of liquifaction from suppuration of the infected renal parenchyma.
The imaging diagnosis can best be established by use of CT.82 Renal ultrasonography is an alternative if contrast use is contraindicated or the patient’s condition does not warrant use of CT.191 On CT, the findings of acute renal abscess are (1) round or ovoid regions of low attenuation, (2) an irregular wall that may exhibit varying degrees of enhancement, (3) a central fluid component that shows little or no enhancement, (4) extension into the perirenal space or pelvis, (5) gas within the fluid collection, and (6) inflammatory changes and associated fascial thickening surrounding the kidney in the perirenal and pararenal spaces. Only the finding of gas is specific for an abscess. On ultrasonography, findings highly indicative of acute abscess are (1) irregular margin, (2) fluid component (anechoic or low-level internal echoes), and (3) bright echoes representing gas. The distinction between acute focal pyelonephritis and renal abscess is important, because an abscess must be drained either percutaneously or surgically. When findings are equivocal, selective renal arteriography may be used if a necrotic neoplasm is a reasonable possibility in the patient’s clinical context. The arteriographic clue of most reliability is demonstration of tumor vessels. However, inflammatory vessels on the periphery of an abscess can mimic tumor vessels. In that event, percutaneous fine-needle aspiration biopsy of the mass using CT or ultrasound guidance usually distinguishes a tumor from an abscess.149
Chronic Renal Abscess
Chronic renal abscess is simply a later stage in the development of an acute renal abscess that does not respond to treatment or the patient’s immune system. The abscess requires about 10 to 21 days to mature to a chronic state. The focal area of central inflammatory mass necrosis progresses to a more coalescent liquefaction state. The inflammatory parenchymal margin progresses to a definitive thickened “wall” composed predominantly of fibrotic tissue. The imaging approach is basically the same as in acute abscess, but imaging results often are less certain in distinguishing necrotic tumor from chronic abscess. Fine-needle aspiration biopsy and, on occasion, open surgery are required to make a definitive diagnosis.
Perirenal Abscess
Hematogenous infection in the renal parenchyma may also result in perirenal inflammatory disease and abscess formation. Rarely, the infection may actually arise in the perirenal area in addition to extending there as a complication of cortical abscess. Plain-film roentgenograms of perirenal abscess may show an absence of the perirenal fat shadow, causing an indistinctness of the renal margin. When the abscess is confined within the perirenal (Gerota’s) fascia, the posterior and inferior portion of the perirenal space fills with

pus; this may be outlined as a mass that is confined chiefly to the infrarenal area, since the perirenal space is largest in this area. The lower pole of the kidney is obscured. EXU may show upward, anterior, and either medial or lateral renal displacement, depending on the site of the abscess. There may be some compression of the collecting system if the abscess is large. Fixation of the kidney by the infection is demonstrated on roentgenograms obtained in inspiration and expiration, which show a failure of the normal movement of the kidney with respiration. The psoas muscle shadow is enlarged, and its margin is indistinct adjacent to the area of infection. Lumbar scoliosis with convexity away from the side of the lesion results from muscle splinting and is usually present. The diaphragm is often slightly elevated, with areas of linear subsegmental atelectasis in the basal lung manifested by small horizontal densities in the basal lung parenchyma. CT is the imaging modality of choice in evaluating for potential perirenal abscess (Fig. 20-34), and either CT or ultrasound can show the size and extent of the fluid collection. Psoas abscess may displace the kidney and ureter, but it does not ordinarily spread to involve the kidney (Figs. 20-35 and 20-36).
FIG. 20-34. Perirenal abscess. Contrast-enhanced computed tomographic scan shows a large perirenal abscess with involvement of the right kidney. Abscess also extends into the adjacent pararenal space.
FIG. 20-35. Psoas abscess. Note the large psoas mass (arrows), which displaces the left kidney and upper ureter. It also compresses the ureter. This is a chronic abscess that is so well localized that the psoas shadow is clearly defined.
FIG. 20-36. Psoas abscess. Unenhanced computed tomogram of a patient with Crohn’s disease and a left psoas abscess. Note the bubbles of air anteriorly (arrow).
Severe Diffuse Pyelonephritis
Davidson and Talner42 described the correlation of clinical setting and imaging findings in this rare complication of acute renal infection. This condition is characterized by rapid, aggressive hematogenous spread of infection in the kidney that overwhelms the patient’s immune response. The result is a generalized, life-threatening infection featuring renal enlargement owing to severe diffuse inflammatory edema, which severely decreases the parenchymal renal blood supply. The early onset of septicemia in this rare infection accounts for the high mortality rate (30% to 40%).
The clinical setting is almost pathognomonic. The usual acute pyelonephritis clinical signs and symptoms are exaggerated, and the patient often presents in septicemic shock. This occurs almost exclusively in patients with inhibited immune response caused by severe insulin-dependent diabetes, chemotherapy for cancer, or drug abuse. It is critical that the diagnosis be established immediately and that appropriate antibiotic and support treatment be initiated.
The imaging examinations used may begin with EXU but usually go on to CT or Doppler ultrasonography to make the distinction between acute bacterial nephritis and acute renal vein thrombosis, the usual differential diagnosis. The renal veins are normal in acute bacterial nephritis. Urographic findings are (1) generalized renal enlargement; (2) faint, diminished nephrogram; and (3) delayed and severely diminished calyceal opacification. CT demonstrates patchy or diffuse regions of low attenuation on enhanced scans. If the infection is advanced, there may be progression to frank abscess that can rupture into the subcapsular and perirenal spaces. Ultrasonography is not as sensitive as CT in the detection of acute bacterial nephritis. The involved kidney is enlarged, with decreased echogenicity of the renal parenchyma. If there is abscess formation, this can appear as an anechoic focal mass.29

Acute Infection of Pre-existing Renal Spaces
In acute infection of a simple renal cyst or dilated collecting system of a chronically obstructed kidney, the clinical presentation mimics that of severe acute pyelonephritis. The cyst most closely mimics acute renal abscess on imaging examinations. The term pyonephrosis is used to describe an infected hydronephrotic renal collecting system (Fig. 20-37). When severe acute infection clinically is associated with such obstruction, percutaneous needle aspiration to make the diagnosis and then to provide a route for percutaneous pyelostomy drainage is the usual approach. Often this is performed in an urgent clinical setting.148
FIG. 20-37. Pyonephrosis. Infection involving a left congenital ureteropelvic junction obstruction. Note thickening of renal pelvis (arrow) and thickening of Gerota’s fascia (arrowhead). These findings suggest the presence of an acute inflammatory process requiring emergency drainage.
Chronic Pyelonephritis (Atrophic Pyelonephritis)
The following criteria for the diagnosis of chronic atrophic pyelonephritis were suggested by Hodson83,84: (1) the disease is centered in the medulla, with scarring eventually affecting the whole thickness of renal substance; (2) there is an irregular surface depression over the involved area; (3) the involved papilla is retracted because of scarring, with secondary dilatation of its calyx; (4) the dilated calyx has a smooth margin but variable shape; (5) renal tissue adjacent to the involved area is normal or hypertrophied, with a sharp definition between normal and abnormal; (6) distribution is unifocal or multifocal, involving one or both kidneys; and (7) there is a decrease in size of the involved kidney.
Chronic bacterial infection of the kidney usually starts as a focal process in the medulla, which causes a localized area of fibrosis or scarring. As it progresses, the infection causes further scarring which results in loss of renal parenchyma, irregularity of the renal surface, and distortion of the calyx in the involved area as seen on EXU. The involved calyces become clubbed. Renal tissue between involved areas is normal or hypertrophied. Parenchymal loss may progress to the point that there are only a few millimeters of scar tissue between the capsule and the calyx. Unless there is obstruction or significant reflux, the distribution of the lesions is uneven (Fig. 20-38).
FIG. 20-38. A: Pyelonephritis. Chronic pyelonephritis in the left kidney. Note the blunted calyces and the parenchymal loss adjacent to the calyces (greater in the upper pole). B: In another patient with chronic pyelonephritis, the upper-pole calyces appear reasonably normal. Central- and lower-pole calyces are clubbed, and, on the initial film, marked decrease in lower-pole parenchyma could be observed.
The disease usually begins in childhood, but it may not be recognized until early adult life. The earliest roentgenographic sign is a decrease in the amount of renal parenchyma, often in one pole of the kidney. Later, the adjacent calyx or calyces exhibit clubbing. As the disease progresses, the findings become more generalized and are often bilateral but usually not symmetrical.
Ureteral reflux and bladder infection, as well as focal ischemia, probably play a part in the development of changes in the kidney. Scarring and atrophy are most severe in areas

in which there is intrarenal reflux in addition to the ureteral reflux into the collecting system.
Hydronephrotic atrophy or obstructive atrophy of the kidney also causes progressive blunting of the calyces and narrowing of the renal parenchyma. However, this tends to be very symmetrical, in contrast to the irregular distribution of calyceal clubbing and the scars of chronic pyelonephritis. A similar appearance may be observed in patients with vesicoureteral reflux. Infection may be present in both conditions and can cause the focal parenchymal scarring that is found in pyelonephritis. When the disease begins in adult life, there is less scarring of the parenchyma, but the calyceal blunting is similar.
EXU findings feature small, irregular renal shape; clubbed calyces approaching the scarred margin; and interposed focal parenchymal hypertrophy. CT reflects the same appearance. Ultrasonography demonstrates small, shrunken kidneys with increased echogenicity of the kidneys relative to the liver and spleen. The borders of the kidneys are difficult to visualize, because the irregular, scarred margins scatter the echoes so they do not return to the transducer.165
Xanthogranulomatous Pyelonephritis
Xanthogranulomatous pyelonephritis is a form of severe chronic inflammation of the kidney found predominantly in adult women with some degree of urinary obstruction, often from a staghorn calculus. The clinical findings consist of a history of easy fatigability and low-grade fever that may antedate urinary symptoms of dysuria, frequency, and a dull, aching flank pain sometimes associated with a palpable flank mass. Attacks may be recurring. Calculi are common, and there may be parenchymal calcification. The disease is usually unilateral, but the opposite kidney often is involved by pyelonephritis. The pathologic process consists of granulomatous involvement of the renal parenchyma associated with infiltration of foam cells (lipid-laden macrophages), cholesterol, extensive fibrotic changes, and atrophic glomeruli. Chronic obstruction at the ureteral, ureteropelvic, or major calyx level is almost always present. The process may be focal or diffuse; at times it may extend to produce a periureteric mass in the upper ureteral region. It may also extend to involve perirenal fat leading to the production of a fixed renal mass. P. vulgaris is commonly found in the urine but may not be the etiologic agent.
Urographic findings consist of a nonfunctioning or poorly functioning kidney with calyceal dilatation and blunting, irregularity of papillae, decreased cortical thickness, and ureteral deformity and stricture that may resemble changes caused by extensive tuberculosis. Obstruction, often caused by a staghorn calculus, is frequently present. The kidney and psoas outlines may be indistinct. Retrograde pyelography reveals dilatation and gross distortion of the pelvis and calyces, indicating obstruction. In some patients a local mass

resembling carcinoma is present; in others there is a poorly demarcated, diffuse mass, often associated with greatly diminished or no renal function. Angiography reveals displacement and stretching of intrarenal arteries with absence of small peripheral branches. Capsular and ureteric branches may be prominent. The nephrogram phase resembles that of hydronephrosis. In many instances the granulomatous mass cannot be differentiated from renal cell carcinoma by angiographic methods. CT better delineates the total parenchymal process, but often percutaneous biopsy is required to confirm the diagnosis.65 Characteristic features of xanthogranulomatous pyelonephritis on CT are low-density material (lipid-laden macrophages and debris) filling the collecting system and enhancement of the surrounding calyces. When these findings are present in combination with a staghorn calculus, the diagnosis of xanthogranulomatous pyelonephritis should be strongly considered. An additional characteristic feature is local infectious invasion, particularly of the psoas muscle (Fig. 20-39), but cases of xanthogranulomatous pyelonephritis invading areas as distant as the mediastinum have been reported.
FIG. 20-39. Xanthogranulomatous pyelonephritis. In this case of long-standing ureteral obstruction (note bilateral ureteral stents, arrows), a large area of low attenuation in the left kidney (asterisk) extends to and invades the left psoas muscle.
Pyelitis of Pregnancy
The term pyelitis is applied to renal infection that accompanies pregnancy. Most pregnancies are associated with some degree of dilatation of the collecting system and ureter. One study found that 90% of right kidneys and 67% of left kidneys showed at least mild hydronephrosis.146 The hydronephrosis is probably caused by both mechanical obstruction of the ureters resulting from increased uterine size and smooth muscle relaxation resulting from hormonal changes. The predominance of hydronephrosis in the right kidney has been attributed to the sharper angulation of the right ureter as it crosses the right iliac artery and ovarian vein at the level of the pelvic brim.51 The hydronephrosis usually clears 3 to 6 weeks after delivery.
When infection occurs, however, urinary symptoms result from the combination of obstruction and infection. Infection is more common during the last two trimesters. Before use of ultrasonography, urography showed dilated collecting systems and ureters down to the brim of the pelvis (Fig. 20-40). Infection, when present, is usually of recent origin, so that there are no anatomic changes directly related to it unless the patient has had repeated infections in the past or has chronic pyelonephritis.
FIG. 20-40. Hydronephrosis in pregnancy. The patient had an upper urinary tract infection. Combined with pregnancy, this condition is sometimes termed pyelitis of pregnancy. A urogram obtained 2 months after delivery showed a normal urinary tract. Hydronephrosis is present in the latter period of pregnancy in most gravid patients, but infection is relatively uncommon.
Ultrasonography provides a safe, easy, and noninvasive means of evaluating hydronephrosis associated with pregnancy. Although obstruction plays some role in the dilatation of the renal collecting system during pregnancy, there is no

accompanying increase in the intrarenal resistive index in normal pregnant patients.80 CT plays less of a role because of the radiation involved, but it can be valuable after delivery if hydronephrosis or infection persists.
Renal Papillary Necrosis
Renal papillary necrosis is characterized by infarction of renal papillae which results in necrosis and sloughing of the involved papillary tissue. The necrotic material may be passed in fragments or as a single mass, or it may remain in the calyx. When it remains, it may calcify peripherally to form a rather typical triangular concretion. The cause of the necrosis is not clear, but medullary ischemia probably can result from several causes. The condition is usually bilateral and may involve few or many papillae. It is more common in women than in men. The abuse of analgesics such as phenacetin over prolonged periods is associated with a chronic form of the disease. A chronic form also may be associated with sickle cell (homozygous-SS) disease; with heterozygous-SC hemoglobinopathy, minimal papillary necrosis develops without signs or symptoms. An acute fulminating form associated with infection occurs in patients with diabetes mellitus and in patients with obstructive uropathy, particularly when infected. In the acute fulminating form the diminished renal function may make EXU useless, but in most instances the diagnosis can be made with this examination. Therefore, retrograde pyelography is seldom necessary. Renal size is normal in the analgesic-abuse group of patients, but in those with the fulminant infectious form, the kidneys may be enlarged and kidney function decreased. Eventually there is enough destruction or atrophy to decrease renal size, so that the kidneys become small and smooth. Early papillary swelling may be very difficult to assess by EXU. The earliest urographic manifestations suggesting the diagnosis are those of necrosis with formation of tracts extending from the fornix into the parenchyma, paralleling the long axis of the papilla.
There are three forms of papillary sloughing.154 One is the central or partial type, in which there is a tract extending inward from the tip of the papilla. The shape of this cavity varies considerably from one calyx to another (Fig. 20-41). In the second form, the necrosis occurs at the base of the papilla, resulting in sloughing of the papilla. The papilla may remain in the kidney or be excreted, and it can occasionally become lodged in the ureter, causing obstruction. The third form of papillary sloughing is necrosis in situ, in which the papilla remains attached, decreases in size, and eventually may calcify; it usually cannot be recognized until calcification occurs. A triangular radiolucent shadow ringed with a dense opaque shadow, the “ring shadow,” may be observed when the separated necrotic papilla remains in the calyx. Eventually, a typical concretion may develop; it consists of a dense, calcified shell surrounding a radiolucent center. Late in the disease, scarring may result in some distortion. The diagnosis can be histopathologically confirmed if some of the sloughed material is passed and recovered from the urine.
FIG. 20-41. Papillary necrosis. Contrast collection within the papilla (arrow) is the hallmark of papillary necrosis.
Bilateral Acute Renal Cortical Necrosis
This disease is characterized by bilateral, symmetrical, ischemic necrosis of the renal cortex, sparing the subcapsular cortex. It is a cause of acute renal failure and may be associated with a number of antecedent conditions such as severe burns, multiple fractures, internal hemorrhage, severe infections, transfusions of incompatible blood, peritonitis, and others. It occurs frequently in pregnancy, often in association with abruptio placentae. With the advent of modern treatment, including hemodialysis, several patients have recovered partially from the disease and certain roentgenographic findings have been observed that suggest the diagnosis.
Initially, the kidneys usually are enlarged; this is followed by a decrease in size of varying degrees. Faint cortical calcification is seen on urography in the form of a thin, shell-like rim around the periphery of the kidney that appears in 50 to 60 days after onset. This is so faint that tomography or CT may be necessary for adequate visualization in patients suspected of having the disease. Tram-line or double-line calcification has also been reported. The calcification may extend into the interlobular septa, appearing as diffuse and punctate densities in the remaining cortical tissue.143 The renal contour may be irregular and the calcification interrupted, depending on distribution of the disease. The renal pelvis and calyceal system appear normal, but function is usually so decreased after recovery from the acute phase of the disease that retrograde pyelography is necessary to outline the collecting system.

Pyeloureteritis Cystica, Pyelitis Cystica, Ureteritis Cystica
Pyeloureteritis cystica and ureteritis cystica are manifested by small suburethelial cysts that elevate the epithelium of the ureteral wall and sometimes the wall of the renal pelvis in association with chronic urinary tract infection. These cysts appear as small radiolucent defects along the course of the ureter when it has been opacified by contrast substance. The appearance of multiple, small, mucosal filling defects is pathognomonic. The defects usually are more numerous in the upper ureter than elsewhere. They may become large enough to produce partial ureteral obstruction, and they range from microscopic size to 2 cm in diameter. Signs of active infection at the time of urography are often present in these patients, or there may be a history of previous urinary tract infection. Stones in the urinary tract are also common. The lesions may be unilateral (70%) or bilateral. The condition is relatively rare in the ureters and renal pelvis and is extremely rare in the infundibula and calyces.52
The kidney is involved by tuberculosis in a manner comparable to involvement of other organs. The infection is hematogenous. The organisms are filtered out by the glomerular capillary bed, where they may produce small tubercles, some of which heal. However, necrosis may occur and organisms may migrate from the cortex to the region of the renal papilla. There, new tubercles are formed in Henle’s loop, leading to destruction of medullary tissue and ulceration. These early lesions are often multiple but do not involve all the papillae. As the disease progresses, involvement of adjacent infundibula often leads to obstruction. Similar stricture formation leading to obstruction is found when there is ureteral involvement. If the disease does not heal spontaneously, the destruction continues, producing irregular cavities adjacent to the calyces. Eventually this leads to virtual destruction of the entire kidney. If ureteral obstruction is not a factor, the kidney may gradually decrease in size, or it may remain normal in size and gradually fill with caseous material along with some calcium to form the so-called putty kidney. If ureteral obstruction occurs before the kidney is destroyed and functionless, the result is a large hydronephrotic kidney in which there are irregular cavities adjacent to the calyces. The anatomic changes are visible on urograms and form the basis for the roentgenographic diagnosis of renal tuberculosis. This diagnosis should always be confirmed, as in pulmonary tuberculosis, by demonstration of the organisms in the urine from the involved kidney. Even though the disease is hematogenous, the initial source, usually lung or bone, may not be detected. Clinical evidence of renal involvement is unilateral in about 75% of patients, even though the organisms have presumably been disseminated to both kidneys. CT and ultrasound findings are similar to those found at urography.
Roentgenographic Findings
The roentgenographic findings on plain-film examination are those of alteration in size of the kidney and calcification within it owing to advanced disease. These are nonspecific findings but they may be suggestive, particularly if cloudy flocculent calcification outlines most of the renal shadow, indicative of extensive destruction of parenchyma, the so-called autonephrectomy. The calcification may be dense and irregular and may lie within the renal outline, often in the cortical area. In the early stages of cortical involvement, no urographic findings are present, and it is possible to have considerable parenchymal involvement without urographic change. The earliest finding is that of a slight irregularity of the involved calyx caused by ulcerative papillary lesions (Fig. 20-42). Further destruction is manifested by loss of the normal papilla and irregular, ragged cavity formation (Fig. 20-43). Often this is associated with a narrowing of the infundibulum to the affected calyx. The infundibulum may later become completely obstructed, so that the diseased area is not visible on retrograde pyelography. A careful evaluation of calyceal distribution in relation to the renal outline is necessary in all patients with suspected renal tuberculosis. Parenchymal destruction may result in cortical scarring with irregular narrowing of the parenchyma and irregularity of the renal outline. When the renal pelvis is involved, the mucosa is irregular because of ulceration. Later local constriction caused by fibrosis is also common, and dilatation results

when there is obstruction at or below the UPJ. Ureteral involvement may result in stricture formation, often multiple; mucosal infection can also produce small local nodules that appear as filling defects along the ureteral wall. The appearance is quite variable, ranging from a beaded pattern, to a corkscrew pattern, to single or multiple strictures in some instances. In advanced involvement of the ureter it is common to find the ureter unusually straight, extending in a direct line downward from the renal pelvis to the pelvic brim (“pipe stem” ureter) without the usual slight curves seen in the normal ureter. The bladder, seminal vesicles, and vas deferens may also be involved in patients with renal tuberculosis. The bladder wall may be thickened and its capacity diminished. Tuberculous granulation tissue projecting into the bladder may resemble carcinoma in some instances. Irregular mottled calcification in these structures suggests the diagnosis.
FIG. 20-42. Renal tuberculosis. The calyces of the upper pole are involved and are irregular as a result of adjacent parenchymal destruction.
FIG. 20-43. Renal tuberculosis. A: There is extensive involvement with cavity formation superiorly and centrally. There is also irregular narrowing of one of the upper central infundibula. B: The right kidney is relatively normal, but the small, dense, irregular kidney on the left did not change during urography. This left kidney represents the so-called putty kidney.
Urography is used in renal tuberculosis as a method of making an anatomic diagnosis, to be confirmed by bacteriologic study. It is also of value in monitoring the renal lesion during treatment, in detecting complications (e.g., ureteral obstruction, infundibular obstruction), and in outlining the opposite kidney.
Differential Diagnosis
Differential diagnosis of calcium deposits must include consideration of renal calculi and nephrocalcinosis as well as cyst and tumor calcification. Calculi usually are more discrete and rounded than the calcification seen in tuberculosis. Tumor calcification often extends beyond the border of the kidney and tends to appear less hazy and flocculent than that seen in tuberculosis. Calcification in cysts occurs in the wall and tends to outline it in an arcuate form of varying size. This calcification also tends to extend beyond the shadow of the normal kidney. Urographic changes in chronic pyelonephritis consist of calyceal abnormality that may resemble early tuberculous involvement, but the change is usually more general than in tuberculosis. The same is true in renal papillary necrosis, which is usually bilateral and tends to be more extensive than renal tuberculosis. In some patients, granulomatous changes predominate to the extent that a renal mass is formed, which must be differentiated from other renal masses. Usually other signs are present that aid in the diagnosis of these cases. Brucellosis may produce findings in the kidney identical to those caused by tuberculosis, but it is very rare.
Renal Candidiasis
Because renal candidiasis usually occurs in patients who have a chronic illness or whose immune system has been altered, the following factors are important in its development: acquired immunodeficiency syndrome (AIDS), antibiotic therapy, prolonged use of indwelling intravenous catheters, treatment with steroids or chemotherapy, therapy with

immunosuppressive agents, blood dyscrasias, diabetes mellitus, intravenous drug abuse, and chronic disease such as malignant neoplasm.32 In systemic candidiasis, involvement of the kidney is common. Renal candidiasis can assume three forms, which may be different stages of the same disease: (1) acute pyelonephritis, in which the fungi proliferate in the renal tubules to form cortical and medullary abscesses with interstitial edema and renal failure; (2) a more chronic process, with hydronephrosis and chronic pyelonephritis; and (3) disseminated candidiasis, which involves several organs including the kidneys. EXU may demonstrate multiple fungus balls in the renal pelvis and upper ureter in patients with pyelonephritis, but renal function may be so poor that retrograde studies, CT, or ultrasound may be needed to reveal their presence. The appearance is one of microabscesses in the renal parenchyma, and shaggy, irregular filling defects in the renal pelvis, often extending into the infundibula and upper ureter.171 Acute papillary necrosis resulting from candidiasis is similar to that caused by other acute fulminating infections, except that in candidiasis more debris may be present in the calyces and pelvis, representing the sloughed necrotic papilla plus the fungus balls (mycelial masses) (Fig. 20-44). CT findings are similar, although without corroborating clinical data the findings may be confused with other causes of luminal filling defects, such as transitional cell carcinoma.
FIG. 20-44. Renal candidiasis. A: Irregular filling defects are seen in the calyces of a patient with chronic candidiuria. B: Computed tomographic scan demonstrates a large soft-tissue filling defect (“fungus ball”) surrounding a ureteral stent in the right renal pelvis.
The kidney lies in a well protected area and is seldom injured. In patients with chronic renal disease, however, relatively minor trauma can cause considerable damage. Direct force over the renal area is the usual cause of injury. Trauma to the kidney is usually manifested by hematuria, which may be gross or microscopic. Hematuria after an injury indicates some type of damage to the kidney or injury to the lower urinary tract. EXU with tomography has been the traditional imaging modality to evaluate minor renal trauma, but now, particularly in major trauma, IVP has been largely replaced by CT, which is considered the imaging modality of choice to evaluate the kidneys after blunt abdominal trauma.
CT reveals contusions, incomplete and complete lacerations, intrarenal and extrarenal hematomas, and fractured or shattered kidneys.23, 28 It is superior to EXU and other nonivasive modalities in differentiating major from minor renal injuries.55, 164 The choice of surgical or medical therapy depends on the severity of injury to the kidney, and CT therefore has a significant impact. It offers the added value of detection of extrarenal injuries in trauma affecting other visceral organs such as the liver or spleen. The CT scan should be performed with the use of an intravenous contrast agent and dynamic or helical technique to optimize detection of lacerations, extravasated urine, and hematomas. Dynamic CT is superior to conventional axial CT in assessing parenchymal renal injuries. Dynamic CT correctly diagnosed parenchymal injuries in 129 of 130 cases, compared with 116 of 130 cases for conventional CT, in a study presented by Lang and colleagues.104
Injury to the renal pedicle should be suspected if there is delayed visualization or nonvisualization of the kidney on CT or EXU after a bolus injection of iodinated contrast material. CT has the advantage over EXU of being able to prove definitively the existence or absence of a nonfunctioning kidney in the event of nonvisualization after contrast injection (Fig. 20-45). Patients with renal aplasia, hypoplasia, or other congenital abnormalities may have false-positive studies at EXU and receive unnecessary angiography.
FIG. 20-45. Renal trauma. A: Enhanced computed tomogram (CT) of a laceration in the midportion of the left kidney (white arrow) with predominantly perirenal hematoma (black arrow). B: Enhanced CT of a different patient with perirenal, pararenal, and central sinus hematomas. C: Enhanced CT of a patient with a renal fracture and pedicle injury. Note lack of contrast enhancement of renal parenchyma on left.
Renal angiography is indicated in the post-traumatic period if there is suspicion of injury to the renal pedicle. Subintimal flaps, renal artery thrombosis, arteriovenous fistula, or laceration of the artery may occur. Arteriography is still considered the best method for demonstrating injuries to the

renal artery, although MRI may have a role in the future. Angiography also has the potential of being therapeutic. Transcatheter embolization or balloon occlusion can be used to treat extensive hemorrhage or arteriovenous fistula.104 If renal artery damage is demonstrated, immediate surgical correction of the vascular lesion may prevent permanent renal damage in some instances.
The severity of parenchymal injury can vary from rupture of a calyx with extravasation of blood or urine into the parenchyma to more extensive fracture of the parenchyma with subcapsular and parenchymal extravasation. With more extensive injury, the capsule can rupture, producing perirenal hemorrhage and extravasation of urine. These nonpenetrating injuries can be classified into the following categories: (1) contusion, (2) cortical laceration (often with intrarenal hematoma), (3) calyceal laceration, and (4) fracture with laceration of the renal capsule. A fractured kidney may have associated lacerations of the calyces, infundibula, or pelvis. Nonparenchymal injury, such as rupture of the renal pelvis or rupture of an anomalous extrarenal calyx, may occur. Immediate surgical extirpation of the kidney may be required in some instances when there is extensive fracture with retroperitoneal and intraperitoneal hemorrhage. The opposite kidney should always be studied by means of EXU or CT before nephrectomy. Conservative management of the traumatized kidney, whenever possible, is being used more often now than in the past, particularly if the patient is hemodynamically stable.
The radiographic findings of renal trauma depend on the extent of injury. A simple contusion is usually manifested by renal swelling, decreased density of the nephrogram in the affected portion, and decreased or delayed excretion of contrast material into the collecting system. If perirenal hemorrhage is present, the renal shadow, and sometimes the psoas shadow, is obliterated or enlarged on the plain film. At times the hemorrhage remains localized in the perirenal area and produces a localized or generalized enlargement of the kidney. CT nicely demonstrates perirenal and pararenal hematoma (Fig. 20-45). Kunin100 described bridging fibrous septa within the perinephric space, which can act to limit and stop the perirenal hemorrhage by tamponade.100 Rarely, calcification of a hematoma in or around the kidney may be seen as a late finding in renal trauma. Subcapsular hematoma produces a lenticular indentation on the renal parenchyma

(Fig. 20-46). Accessory signs on plain films are scoliosis (convexity to the opposite side indicating muscle spasm); dilated small-bowel loops in the vicinity of the injury, caused by local adynamic ileus; and fracture of an adjacent rib, vertebral body, transverse, or spinous process. Urography demonstrates the amount of extravasation and may show calyceal compression and distortion caused by parenchymal and subcapsular accumulations of blood, urine, or both. CT provides a better view of the extent of extravasation in many instances (Fig. 20-47). The amount of extravasation is not necessarily proportional to the parenchymal or vascular damage, however. Because post-traumatic distortion and stricture are important, urographic studies should also be carried out during or after convalescence to outline any residual deformity.
FIG. 20-46. Renal trauma. Enhanced computed tomogram shows subcapsular hematoma. Note the indentation on the renal parenchyma (arrow) from the pressure of the hematoma confined by the renal capsule.
FIG. 20-47. Renal trauma. Enhanced computed tomogram shows extravasation of contrast material (arrow) in addition to the hematoma around the left kidney.
The classification of renal cystic disease is difficult, and there is much confusion in the literature because of disagreement among pathologists. For our purposes the classification of Hartman72 is most useful for radiologists (Table 20-2).
Table 20-2. Classification of renal cysts
Simple Renal Cysts
The simple renal cyst is often a “silent” lesion of little or no clinical importance, but it is the most common unifocal renal mass. When cysts become large, they may cause pain and occasionally hematuria. Simple cysts rarely bleed, but they may become large enough to be noted as masses that can be palpated through the abdominal wall. They can cause renal damage because of their size, particularly if they are situated in a region where obstruction of the excretory system can occur. Lesions can be unilateral or bilateral. A cyst may be solitary, but often there are two or more in a kidney. They can be so numerous that differentiation from adult polycystic disease is difficult. The roentgenographic findings depend on the location. The chief importance of the simple cyst is that it may simulate a tumor in its appearance on plain-film roentgenography and EXU. Plain films may outline a smooth, local enlargement of the kidney. Occasionally (fewer than 1% of instances), there is a thin shell of calcium outlining the cyst wall or a portion of it. The cysts may reach massive size and actually dwarf the kidney. Urographic findings consist of crescentic defects and stretching of the infundibula and calyces when the lesion arises close to the calyces. If the cyst arises farther away there is less calyceal change, and if it is in a subcapsular position there is little or no pressure deformity on the pelvis or calyces.
The following urographic signs are present in the great majority of cysts: (1) the lesion is peripheral so that it bulges out of the kidney; (2) the wall, if visible, is very thin and

smooth; (3) the mass is quite radiolucent (compared with the adjacent parenchyma) and is sharply demarcated from the renal parenchyma. Often this appears as a beak-like deformity or “claw sign” (Fig. 20-48). If all these signs are present, the lesion most likely is a cyst. After a urographic examination suggesting the diagnosis of a cyst, CT or ultrasonography is usually the next imaging examination, provided the patient would be treated if the lesion turned out to be a tumor rather than a cyst.
FIG. 20-48. The claw sign of a renal cyst (arrows). Nephrotomogram clearly outlines the smooth wall of the radiolucent cyst adjacent to the density of the opacified parenchyma.
On CT cysts are round, have a thin smooth wall, and have no attenuation increase after administration of contrast material. Attenuation values should be near that of water (−10 to −2 Hounsfield units). By ultrasound, cysts should be round, be anechoic, have a smooth, thin wall, and have increased through-transmission.153 Minimally complicated cysts that have thin, curvilinear calcifications in the wall and high density material within them (in the absence of any other signs of malignancy) also have low malignant potential. For further discussion on the differentiation of cysts from tumors see Renal Cell Carcinoma).
Renal Cystic Disease Associated with Multiple Renal Neoplasms
Bilateral renal cystic disease that develops during dialysis treatment is known as acquired renal cystic disease. The radiographic features are consistent with a large number of small (less than 3 cm) cortical and medullary cysts in the face of a small kidney. Complications include development of renal cell carcinoma in approximately 7% of patients67 and retroperitoneal hemorrhage.
Von Hipple-Lindau disease is a phakomatosis of autosomal dominant inheritance. There are a number of extrarenal manifestations of this disease, including retinal angiomatosis, central nervous system hemangioblastomas, and pheochromocytomas. Renal manifestations are usually limited to multiple simple cysts, and approximately one third of patients have multiple renal cell carcinomas.112 Imaging findings demonstrate multiple solid and cystic masses in both kidneys in association with extrarenal findings.
In tuberous sclerosis, renal cysts are usually small and of tubular origin. Rarely, cortical cysts large enough to produce distortion have been reported. Angiomyolipoma of the kidney is found frequently in association with tuberous sclerosis (see description of this tumor or hamartoma in Benign Tumors).
Polycystic Kidney Disease
There are two general categories of polycystic kidney disease, autosomal recessive and autosomal dominant. Both are inherited diseases characterized by multiple cystic abnormalities within both kidneys. Beyond those similarities, the two categories differ greatly in many aspects. The microdissection research of Osathanondh and Potter142 in the mid-1960s established the pathogenesis and the embryologic defects that characterize the two basic categories of polycystic disease.
Autosomal Recessive Polycystic Kidney Disease
The pathologic-radiologic hallmark of this disease is the presence of grossly elongated, dilated collecting ducts throughout the renal parenchyma. The dilated ducts extend from the papillary tips to the surface of the cortex, and urine flow from nephrons into these ducts is slower than normal. When seen in profile, clusters of these ducts in each renal lobule look like straws seen from the side. When seen on end, the ducts appear like a cluster of straws seen on end. The kidneys are usually symmetrically and uniformly enlarged with correspondingly enlarged, but otherwise normal, collecting systems. The renal vasculature, nephrons, and ureters are normal. The surface of the kidneys is studded with a myriad of 1- to 2-mm vesicle-like protrusions representing the peripheral ends of clusters of dilated collecting ducts seen on end.
The embryologic defect is a result of abnormal development of the interstitial portion of the ureteral bud in the first trimester of gestation. Genetically this is an autosomal recessive disease. The only cystic involvement of other organs occurs in the liver. It features generalized proliferation of the intrahepatic biliary ducts with associated periportal fibrosis. Grossly, this appears as generalized ectasia of the ducts, and it occurs to a greater or lesser degree in all cases.

Caroli’s disease, characterized by segmental intrahepatic ductal dilatation, periportal fibrosis, and renal tubular ectasia, is now considered one end of the spectrum of autosomal recessive polycystic kidney disease.
The clinical course is determined by the severity of the degree of involvement of the kidneys compared with the liver. There are four clinical subgroups based on the time at which the disease becomes clinically manifest.18 The earliest is the perinatal group, in which about 90% of renal collecting ducts are involved. The disease is apparent at birth, with grossly and palpably enlarged kidneys. Patients usually die before the age of 6 weeks from rapidly progressive renal failure. The hepatic abnormality does not have time to become clinically manifest. The latest appearance is the juvenile group, in whom 10% or fewer renal collecting ducts are involved. The renal disease usually is not clinically manifest, but the liver disease is predominant and usually lethal. Clinical manifestations reflect severe progressive periportal hepatic fibrosis with portal hypertension, gastric and esophageal varices, and hematemesis. Age of clinical onset is usually between 4 and 8 years.
In between the perinatal and juvenile groups are two intermediate groups: the neonatal group, in whom about 60% of renal ducts are involved, and the infantile group, in whom about 25% of ducts are involved. The radiologist, on the basis of urographic findings, can sort the patients into two main groups, which Elkin has called polycystic disease of the newborn and polycystic disease of childhood.53 The former roughly encompasses the pathologic-based perinatal and neonatal groups; and the latter, the infantile and juvenile groups.
EXU has been replaced by ultrasonography to a large extent. Findings on urography in the newborn category include bilateral massive renal enlargement (with smooth margins), prolonged nephrogram density (up to 72 hours), accumulation of contrast material in dilated collecting ducts producing a streaky parenchymal pattern, and normal collecting systems and ureters (when seen). In the childhood category, the hepatic component predominates. Urography usually shows mildly enlarged kidneys, a mildly prolonged nephrogram phase, a streaky parenchymal contrast pattern that is predominantly medullary in location, and gross hepatic enlargement.
On ultrasound examination, the kidneys are enlarged and show a diffusely heterogeneous increased echogenicity. The cysts are too small to be seen on ultrasonography as fluid-filled regions but are large enough to cause echoes, mainly in the medulla. The periphery of the kidneys may be hypoechoic. Cysts may be seen in the liver. The hepatic parenchyma appears more echogenic because of fibrosis.
Autosomal Dominant Polycystic Kidney Disease
This disease is transmitted as an autosomal dominant trait with strong penetrance. Spherical, fluid-filled cysts, usually from 1 to 3 cm in size, are scattered throughout the renal parenchyma. Occasionally the cysts have curvilinear walls or intrarenal punctate calcifications. The superficial cysts produce a knobby appearance of the kidney surface. Between the cysts are scattered islands of normal parenchyma containing normal nephrons and collecting ducts. Kidneys are greatly enlarged, with correspondingly larger-than-normal collecting systems that are irregularly compressed by many adjacent cysts. In about one third of patients, large spherical cysts are scattered throughout the liver. These do not communicate with the biliary tree. Spherical cysts occasionally also involve the pancreas, spleen, lungs, and ovaries. There is an approximately 10% incidence of rupture of berry aneurysm of the arteries at the base of the brain with associated high mortality rates.121 The embryologic abnormality is a sporadic failure of both the ampullary and interstitial portions of the ureteral bud. This results in failure to form normal nephrons and collecting ducts in areas scattered throughout the renal parenchyma.
Clinical manifestations most often appear in adults in the 30- to 50-year-old age group. Rarely, it occurs in young infants. The most common presenting problems are hypertension, microscopic hematuria, and presence of palpable abdominal masses and pain as a result of renal enlargement. Complications of the large number of renal cysts include bleeding, infection, stone formation, and urinary tract obstruction. The disease may be diagnosed on EXU, ultrasound, or CT examination. Urographic findings include enlarged, knobby-surfaced kidneys; round, radiolucent nephrogram defects; irregularly distorted collecting systems; and, in some cases, lucent hepatic defects representing cysts as an unexpected nephrogram film finding.
Ultrasonography readily demonstrates the enlarged kidneys containing a multitude of fluid-filled, anechoic cysts. The hepatic cysts, likewise, are easily seen. CT shows to good advantage all the pathologic cystic findings described previously in the kidneys and the liver (Fig. 20-49). For follow-up of a known disease, ultrasonography is less expensive, noninvasive, and easy to perform.
FIG. 20-49. Adult polycystic disease. Enhanced computed tomogram shows hepatic and bilateral renal cysts.
Medullary Cysts
Medullary Sponge Kidney
Medullary sponge kidney is a form of cystic disease involving the medulla of the kidney. It is more common in men than in women (ratio of about 2:1); it has been reported in siblings but does not appear to be hereditary. The changes are confined to the renal medulla and consist of dilatation involving the collecting tubules in the renal pyramids. Calculi can develop in the dilated ducts. The condition may be limited to a single pyramid but is usually more extensive; it is usually bilateral but not necessarily symmetrical. Renal enlargement may be present when the lesions are general. The defect appears to be a developmental one involving the formation of the collecting ducts. There may be morbidity caused by infection or colic when calculi are passed. Microscopically, the elongated or irregular cysts present a varied

appearance. The epithelium varies from transitional to squamous or columnar, normal tubules are reduced or absent, and some degree of inflammatory change is usually present. The dilated ducts may contain calculi or masses of calcified debris.
The urographic findings usually are quite characteristic.145 The plain film demonstrates the calculi when present. Their medullary position and appearance are often diagnostic but can mimic stones in renal tubular acidosis. The calculi are usually multiple, small, and spindle-shaped, and they occur in clusters or in a fan-like arrangement in the renal pyramids. On EXU the dilated tubules are seen to be opacified unless infection has impaired renal function. Minimal dilatation produces a fine striated appearance; with increasing dilatation the appearance becomes more cyst-like, with rounded or elongated cavities enlarging and often distorting the papilla and minor calyx (Fig. 20-50). Adjacent calyces may show a considerable difference in the degree of involvement.
FIG. 20-50. Sponge kidney. A: Preliminary film shows mottled calcifications in the left kidney. B: Urogram of the patient shown in A demonstrates the relationship of the calcifications to several of the calyces. These calcifications are in dilated tubules in the renal papillae.
Medullary Cystic Disease
Nephronophthisis (familial juvenile nephronophthisis, medullary cystic disease of the kidney) is a rare disorder of unknown origin usually found in children and young adults. Anemia, polydipsia, polyuria, salt wasting, and progressive uremia develop insidiously. The childhood form is inherited as an autosomal condition and is associated with blonde or red hair, multiple ophthalmologic abnormalities, neurologic abnormalities, growth retardation, bone deformities, and hypocalcemic tetany. The adult form is an autosomal dominant condition and is not associated with extrarenal abnormalities. Urine has a low, fixed specific gravity with absence of protein or formed elements. Histopathologic findings consist of alternating areas of cystic dilatation and atrophy in the proximal and distal tubules with marked thickening of the basement membrane. Interstitial fibrosis with round-cell infiltration is prominent. Glomeruli show minor focal thickening early, progressing to sclerosis and periglomerular fibrosis.
Urographic study has been of limited value because of poor function. With high-dose urography, minor calyceal blunting, uniform contraction of the kidneys, and cyst-like areas of medullary lucency may be observed. Renal angiography shows marked cortical thinning and multiple cysts that spare the thin outer cortex, which is undulating because of the numerous cysts that also displace vessels. The cortex is best seen in the nephrogram phase of angiography.122 Ultrasonography shows bilateral small, echogenic kidneys.157 A few cysts may be visible as fluid-filled structures. CT may show small, smooth kidneys with cysts in the medulla or at the corticomedullary junction.
Multicystic Dysplastic Kidney
Congenital multicystic dysplastic kidney is a rare disorder usually considered to be a severe form of renal dysplasia related to ampullary dysfunction and urinary tract obstruction.15 Pelvoinfundibular atresia results in a number of small cysts with little or no renal parenchyma. The renal pelvis is absent or small and does not communicate with the cysts. The hydronephrotic form is similar in appearance to the pelvoinfundibular type, with the exception of an enlarged renal pelvis which communicates with the renal cysts.
The bilateral form results in renal nonfunction, whereas the unilateral form, which is more common, carries a good prognosis if uncomplicated by other anomalies. There is absence of normal renal parenchyma, the pelvis is small or absent, and the ureter is hypoplastic, stenotic, or atretic. The few nephrons that are present are hypoplastic with arrested development. Blood supply is variable. The kidney consists of a mass of cysts of varying size and is usually very large. The opposite, uninvolved kidney is often hypertrophied, and there is an association with contralateral UPJ obstruction. At presentation, the patient is usually a healthy-appearing infant with a unilateral flank mass. The mass may be visible on plain film.
The pattern on EXU of a large mass in the renal fossa with opacification of strands of vascularized dysplastic tissue and cyst walls surrounding radiolucent cysts, plus absent nephrogram and no identifiable collecting system or ureter, is virtually diagnostic. This is especially so if cystoscopy and retrograde pyelography have demonstrated absence of one half of the trigone or atretic ureter on the involved side. Shell-like calcification may outline some of the cysts. Occasionally there is delayed opacification of some irregular cystic spaces. There may be a small amount of functioning renal tissue scattered in the dysplastic kidney to account for this.

Ultrasonography demonstrates unilateral multiple cysts of varying shape and size in the renal fossa (Fig. 20-51). These cysts do not communicate (as opposed to the communicating appearance of hydronephrosis), and the atretic renal pelvis cannot be visualized. Occasionally, ultrasound-guided percutaneous cyst puncture is needed to distinguish renal dysplasia from hydronephrosis. Dysplasia is a benign lesion and need not be removed once the diagnosis is made.
FIG. 20-51. Multicystic dysplastic kidney. Longitudinal ultrasound image demonstrates multiple large, noncommunicating cysts in the renal fossa.
Segmental multicystic renal dysplasia may also occur and usually is associated with a duplicated collecting system. In the cases reported by Daughtridge,41 there was peripheral and central calcification resembling that sometimes observed in renal cell carcinoma. Arteriography often reveals a sharply demarcated, avascular mass with no neovascularity. Differentiation from avascular tumor is difficult in this segmental form of the disease.
Extraparenchymal Renal Cysts
Parapelvic Cysts
Parapelvic cysts are found in approximately 1.25% to 1.50% of autopsy cases.79 Unlike simple renal cysts, they do not lie within the renal parenchyma. They are located in, and probably originate in, the hilus of the kidney in close proximity to the pelvis and major calyces. Their origin is obscure. Most authors believe them to be of lymphatic origin, arising from lymphatic obstruction and subsequent ectasia.
Urographic findings are those of a mass in the renal hilus that causes compression and displacement of the pelvis and distortion and displacement of the major calyces and infundibula. Mild local caliectasis may result from partial obstruction caused by compression. They do not contain calcium. They resemble renal sinus lipomatosis when the latter results in a focal mass in the renal hilum.
Because there is no renal parenchymal interface with this type of cyst, the nephrographic phase of arteriography or nephrotomography is somewhat different from that seen in

the simple cyst. The parapelvic cyst appears as a spherical mass (of lesser density than the opacified renal parenchyma adjacent to it) surrounded by a halo of fat that is more radiolucent than the cyst (Fig. 20-52). Ultrasonography demonstrates an anechoic structure in the renal sinus that does not branch. However, some of these cysts are difficult to differentiate from hydronephrosis. In those cases, CT (especially with delayed images that demonstrate the cysts around the collecting system) can make a definitive diagnosis.
FIG. 20-52. Halo sign of a parapelvic cyst. A: Nephrotomogram shows ill-defined radiolucency caused by compressed renal sinus fat surrounding the smoothly rounded cyst in the renal hilus. B: Enhanced computed tomogram of parapelvic cyst. Note that cyst displaces contrast-filled collecting system. This differentiates cyst from hydronephrosis or extrarenal pelvis, in which contrast material would enter the fluid-filled cavity.
Calyceal Diverticulum (Pyelogenic or Calycine Cysts)
The term calyceal diverticulum refers to the small, cyst-like spaces that often communicate with a calyx but that occasionally are observed on urography to opacify despite no apparent connection with the adjacent calyx. This lesion may be a true congenital cyst. Cyst-like structures of similar appearance may result from inflammatory destruction of parenchyma adjacent to a calyx, or they may be associated with sickle cell disease. The diagnosis is made on EXU when a small, rounded space fills with opaque medium (Fig. 20-53).121 The cysts are filled by their own tubules and are visible despite lack of apparent communication with the calyceal system. In contrast to the cysts in the renal pyramids as found in sponge kidney, calycine cysts often arise from the fornix of the calyx and occur laterally rather than centrally in relation to the papilla. These cysts are of little clinical significance unless they become infected or are the site of calculus formation. Rarely, milk of calcium is observed, with a fluid level evident on an upright film (Fig. 20-54).
FIG. 20-53. Calyceal diverticulum (arrow). The cyst communicates with one of the calyces of the upper pole.
FIG. 20-54. Milk of calcium in a renal cyst. This film, exposed with the patient in the upright position, shows calcium, which is somewhat granular in appearance, in a cyst located in the central portion of the left kidney. There is also evidence of calcium in the cyst’s wall.
Pararenal Pseudocyst—Urinoma
The term pararenal pseudocyst is used to describe a complication of injury of the renal pelvis or proximal ureter. A rent in the renal collecting system or kidney can result in persistent extravasation of urine or blood into the perirenal space. This may produce compression of the pelvis or upper ureter and resultant hydronephrosis that may lead to eventual loss of renal function. Urographic findings consist of a mass effect that is medial and inferior to the kidney. The mass often displaces the kidney upward and rotates it laterally. Usually, a definite line of separation between the mass effect and the kidney is observed. EXU may opacify the mass (if contrast extravasates into it) and reveals the displacement of the ureter, kidney, or both. Retrograde study may be necessary

to visualize the upper tract distal to the obstructing urinoma. Ultrasonography or CT can better delineate the extent of the pseudocyst and its relation to the space around the kidney (Fig. 20-55).75
FIG. 20-55. Urinoma. A: Enhanced computed tomogram of renal transplant with fluid collection surrounding the kidney. B: Delayed scan reveals contrast within the urinoma.
Renal Vascular Abnormalities
Renal Artery Aneurysm
Aneurysm of the renal artery is not common but is well known to radiologists because the aneurysm wall can contain calcium. About 25% to 30% of these aneurysms contain enough calcium to be roentgenographically visible. The diagnosis usually can be made on plain-film studies in the calcified group. The rounded, ring-contoured, calcified aneurysm maintains a constant relationship to the renal pelvis in various projections. About two thirds of these lesions are located at the bifurcation of the renal artery and one third in the segmental branches. About half of those involving segmental arteries are intraparenchymal. They may be congenital, atherosclerotic, or post-traumatic (often false aneurysms). Systemic hypertension is present in about 15% of patients with aneurysm of the renal artery. Renal arteriography, CT, or MRI can be used for confirmation. Bilateral renal aneurysms are relatively common (about 20%), so arteriography is indicated on the opposite side when an aneurysm, either calcified or noncalcified, is found on one side. Calcified aneurysms generally do not rupture, but the incidence of rupture of uncalcified aneurysms is about 25%. Surgical repair should be considered for selected patients with noncalcified renal artery aneurysms.174 Occasionally, such aneurysms are found incidentally on CT or ultrasound examinations. If indicated by the clinical setting (e.g., hypertension, size of the aneurysm), arteriography should then be considered.
Polyarteritis Nodosa
Multiple, small, intraparenchymal renal aneurysms are commonly observed in polyarteritis nodosa. Similar microaneurysms are present in other viscera involved by this disease. In the kidney, the multiple small aneurysms involving the interlobar and arcuate arteries are associated with scarring, which results from thromboses and infarctions. The aneurysms may rupture to produce renal hemorrhage (Fig. 20-56). Clinically, the infarcts produce pain and hematuria. Renal arteriography outlines the aneurysms, and its use is

essential for making the diagnosis. The small aneurysms found in this disease sometimes regress spontaneously.
FIG. 20-56. Polyarteritis nodosa. A: Enhanced computed tomographic scan demonstrates spontaneous perirenal hemorrhage as measured by cursors. B: Angiogram in same patient. Note multiple microaneurysms.
Renal Arteriovenous Fistula
A renal arteriovenous fistula may be congenital or acquired. The etiologic background is undetermined in some, which may be termed idiopathic. Some congenital malformations are very large and may result in high-output cardiac failure, whereas others are very small and are of little clinical significance. Most arteriovenous fistulae in the kidney are acquired (usually after percutaneous biopsy), and they often heal spontaneously. Others occur after blunt or penetrating renal trauma or surgical procedures in the renal area, and some are secondary to renal neoplasms. Renal arterial disease occasionally may result in an arteriovenous fistula.

When the fistulae persist, hematuria and hypertension may occur as complications. Urograms and plain films usually are not very helpful in determining the diagnosis, but plain films may reveal calcification; in the case of a large arteriovenous fistula, the collecting system may be displaced by the mass produced by the fistula. Often, color Doppler ultrasound can make the diagnosis noninvasively by directly imaging the fistula. Duplex Doppler interrogation of the renal vein demonstrates pulsatile, arterial-type flow if the fistula is of adequate size. Arteriography reveals the fistula along with early venous filling and often demonstrates venous dilatation and tortuousity secondary to the arteriovenous shunt (Fig. 20-57). Renal function may be diminished on the side of involvement, depending on the severity of the shunt.
FIG. 20-57. Renal arteriovenous fistulas. A: Congenital fistula. Congenital fistulas typically demonstrate a “grape-like” sac of dilated vasculature along with an early draining vein. B through D: Acquired fistula from a stab wound is shown on computed tomogram as an area of contrast enhancement originating from the puncture site. Angiograms taken during arterial phase (C) and venous phase (D) demonstrate a more globular fistula than in A.

Renal Artery Occlusion
Renal arterial occlusion is most frequently caused by embolism in patients with cardiac disease. Thrombosis also occurs, most often secondary to atherosclerosis, but sometimes as a result of trauma. Regardless of cause, when a renal artery is occluded function is lost. Partial function may occasionally return after 1 year or longer. The roentgenographic findings in acute renal arterial occlusion consist of urographic evidence of a nonfunctioning kidney of normal size in which retrograde pyelography shows no abnormality. A peripheral rim of opacified cortex may be observed during the nephrographic phase. Presumably, this cortex is supplied by the collateral circulation of the renal capsule. Acute segmental infarction may be responsible for either complete nonvisualization on EXU or a local failure of calyceal filling. The cause for the complete loss of function in these patients is not certain. It may be that a shower of small emboli accompanies the segmental embolic block.134 After total renal infarction without infection, the kidney decreases in size and usually remains nonfunctioning. Retrograde study reveals decrease in size of the calyceal system consistent with renal size. The late findings in segmental infarction are those of local decrease in size, which may distort the kidney locally and cause an irregular contour. Renal arteriography can be used to confirm the diagnosis in all types of arterial occlusion. Both left and right sides should be studied, because involvement is frequently bilateral. Doppler ultrasound can suggest the diagnosis of renal arterial occlusion. The role of MRI angiography is under investigation. CT findings parallel those of urography, but the effects of hemodynamics are better shown.
Renal Vein Anomalies
Valves are occasionally found in renal veins, as indicated previously. Anomalies of the left renal vein, specifically circumaortic or retroaortic left-renal veins, are occasionally seen on CT. Radiographically, this is recognized as a bifurcation or retroaortic position of the renal vein as it courses from the kidney to the vena cava. It is important to recognize these vascular anomalies as incidental findings and not mistake them for adenopathy or other pathology. Surgeons need to know of the existence of these variants before undertaking retroperitoneal surgery.
Renal Vein Thrombosis
Thrombosis of the renal vein occurs more frequently in children than in adults. In adults, direct invasion or extrinsic pressure by tumor and thrombosis of the inferior vena cava are among the more frequent causes. Acute enteritis is considered the chief cause in children, but any condition that produces dehydration, acidosis, and hemoconcentration may be an inciting factor. In infants who have had intrauterine renal vein thrombosis, a faint, lace-like calcification corresponding to intrarenal vascular structures may be observed on plain film. Because diagnosis is difficult from a clinical standpoint, radiographic methods are of prime importance. Roentgenographic findings depend on the rapidity of the occlusion and its relation to the development of venous collaterals.
In acute renal vein thrombosis, the kidney is enlarged and EXU shows no excretion of contrast. The renal arteriogram shows delayed flow through narrow, stretched, interlobar arteries. Opacification of the parenchyma is poor, and the nephrogram phase is prolonged. Venous drainage cannot be identified. When the acute occlusion is partial, the kidney also becomes enlarged. Function as demonstrated by EXU is gradually regained in about 2 weeks as venous collaterals develop. In gradual occlusion, collateral circulation has time to develop, and the roentgenographic examination may show no abnormality whatsoever. When renal arteriography is carried out, the venous phase may show extensive venous collaterals. Dynamic CT also delineates the process well, when it is in an advanced stage. Direct visualization of thrombosis in the inferior vena cava and renal veins is possible if thin slices and dynamic or helical technique are used. The role of ultrasound in detection of renal vein thrombosis is unclear. Ultrasound signs include thrombus in the inferior vena cava, loss of corticomedullary junction, and hyperechoic streaks in the interlobar spaces surrounding the medullary pyramids.103 The inability to detect flow in the renal veins using pulsed Doppler or color Doppler interrogation is a known finding of renal vein thrombosis. However, intrarenal renal venous signal may still be present as a result of collateralization. Renal transplants with renal vein thrombosis do not have intrarenal venous flow because of the inability to recruit collateral vessels.77
When renal vein thrombosis is suspected, EXU should be performed. If the diagnosis is supported by absence or decrease of function and increase in renal size, renal arteriography should be the next step in the radiographic study of the patient. Evaluation of the renal arterial supply, intrarenal pathologic state, venous collaterals, or absence of renal vein filling and possibly demonstration of the actual obstructive site can all be accomplished with proper timing of the examination (Fig. 20-58). Inferior venacavography and selective renal venography can alternatively be performed, but catheter insertion adds to the risk of thromboembolism.
FIG. 20-58. Renal vein thrombosis. A: Renal arteriogram at 2 seconds shows a large, vascular mass in the upper pole of the right kidney. Note the renal vein below the artery (arrow). B: Five-second film shows a continued opacification of the tumor and clear definition of the renal vein. C: Renal venogram. The tumor thrombus is clearly defined in the inferior aspect of the renal vein at its junction with the vena cava, which is now outlined above the renal vein. Although there is partial obstruction of the renal vein, no collaterals are visible.
Renovascular Hypertension
Gifford63 defined renovascular disease as the presence of a stenotic lesion in the renal artery or its branches. Diagnosis depends on demonstrating the stenotic lesion on imaging. The patient may or may not be hypertensive. In contrast, renovascular hypertension connotes the presence of a stenotic renal artery lesion plus relief of the hypertension by revascularization or removal of the affected kidney. In Gifford’s experience, simultaneous occurrence of essential hypertension and coincidental renovascular disease is far more common than true renovascular hypertension.

The clinical consequences of renal artery stenosis (RAS) include renovascular hypertension and progressive renal failure. Historically, the suspicion of renovascular hypertension most often prompted efforts to diagnose RAS as the cause. More recently, RAS has been recognized as an important factor in the development of progressive renal failure.140 Atherosclerotic vascular disease affecting the renal arteries is commonly seen in patients with atherosclerotic vascular disease elsewhere.197 RAS is particularly prevalent in the large subgroup of patients with diabetes and peripheral vascular disease (50%). Atherosclerotic vascular disease affects both the large and the small vessels of the kidney. Small-vessel disease (nephrosclerosis) is not surgically treatable. On the other hand, macrovascular disease is a potentially treatable cause of rapidly progressive renal failure. Although the prevalence of RAS in atherosclerotic vascular disease is high, the proportion of patients in whom RAS causes progressive renal failure is unknown.
A stenotic lesion of the renal artery is particularly significant in those patients with pre-existing renal disease, in which there is an overall reduction in nephron mass and an altered ability for the existing nephrons to compensate for the reduced perfusion associated with a renal artery lesion. However, because contrast angiography is contraindicated in these patients with renal failure, no suitable screening technique is available to identify and directly image the renal vasculature.
With the advent of percutaneous transluminal angioplasty and improved surgical techniques for the correction of RAS, accurate diagnosis of this condition in order to preserve function or correct hypertension has become increasingly important.27 Screening arteriography is not a viable option, because the use of contrast agents is believed to hasten the progression of renal failure in these patients. A noninvasive test that requires no nephrotoxic contrast agent would have tremendous clinical impact. The impact of such a test would be further enhanced by accurate computations of the threshold sensitivity of the test using cost-benefit analysis.
Clinical Features
In order to establish functionally significant RAS, diagnostic tests must accurately identify renal artery lesions which, when corrected, result in normalization of blood pressure or stabilization in the decline of renal function. Certain clinical features suggest the presence of significant RAS. These include rapidly progressive renal failure, onset of hypertension at an age younger than 20 years or older than 50 years, uncontrollable hypertension, rapidly progressive hypertension, hypertension responsive to angiotensin-converting-enzyme inhibitors, and the presence of abdominal bruits. Diagnostic tests have focused on detecting renovascular disease in groups of patients selected to be at high risk for functionally significant renovascular disease based on

these clinical findings. Although in general these clinical findings are poor predictors of significant renovascular disease, prior studies have demonstrated that the prevalence of renovascular hypertension is increased in such high-risk patients. The prevalence of renovascular disease in these groups has been demonstrated at 5% to 39%.177 Similar selection strategies have not been implemented for detecting lesions in patients with renal insufficiency that might be caused by RAS, for example in diabetics, patients with peripheral vascular disease, and those with renal insufficiency. A diagnostic test that could identify those patients with RAS as a contributing factor to progressive renal failure would have important clinical impact when coupled with surgical techniques to improve renal perfusion.
Current Diagnostic Tests
Multiple strategies have been proposed for the diagnosis of RAS. In general, the strategies may be classified as those that demonstrate the anatomy of RAS and those that detect the secondary affects of stenotic lesions. Conventional arteriography, intra-arterial DSA, and intravenous DSA belong to the former category. Functional tests that measure the secondary effects of RAS may be divided into tests that measure renin production response to captopril stimulation and nuclear medicine renal perfusion studies. The most accepted measurements of renin production response to captopril stimulation include the captopril stimulation test and renal vein renin ratio measurements.30, 132 Indirect measurements of renal perfusion include rapid-sequence hypertensive urography, 131I-Hippuran and 99mTc-DTPA renography studies, and sonographic measurements of renal artery blood flow using duplex Doppler sonography.
Arteriography, either the cut film type or DSA, is the standard for detecting the anatomic lesions of RAS.81 Findings on arteriography that suggest a functionally significant renal artery lesion include stenosis of greater than 75% (Fig. 20-59), a pressure gradient of greater than 10 to 15 mm Hg, reduced size and perfusion of the affected kidney, and the presence of collateral circulation.3,19,44 Renal arteriography is the only currently available technique that can demonstrate peripheral or segmental RAS associated with fibromuscular dysplasia (Fig. 20-60).166 On the other hand, renal arteriography has not achieved wide acceptance as a screening technique for renovascular hypertension because it is invasive and expensive and is associated with use of a high contrast dose. Intra-arterial DSA has gained wider acceptance as a screening technique in groups of patients considered to be at high risk for renovascular hypertension for several reasons. Intra-arterial DSA can demonstrate lesions with similar anatomic resolution compared with conventional arteriography. In addition, peripheral and segmental vessels can be identified. Intra-arterial DSA has advantages over conventional arteriography because a smaller contrast dose can be administered and small catheters are used, which allows for the procedure to be performed with less risk. Therefore, the technique is considered to be much less invasive. However, because these techniques are invasive, require the use of contrast agents, and are expensive, they are not well suited for screening a population of patients with a low prevalence of RAS. As a result, many other diagnostic imaging screening techniques have been evaluated.
FIG. 20-59. Translumbar aortogram with good visualization of the renal artery. The site of a stenotic lesion (arrow) secondary to arteriosclerosis in a 63-year-old man with hypertension is shown. Note the poststenotic dilatation of the renal artery.
FIG. 20-60. Fibromuscular dysplasia in an 18-year-old woman with hypertension. A: This urogram shows very slight hyperconcentration on the left. There are a few minor indentations on the upper-left ureter, suggesting the possibility of collateral vessels. B: This arteriogram shows a normal right renal artery. On the left there are multiple constrictions with poststenotic dilatation. There is great delay in perfusion of the left kidney, compared with the right. Note the collateral arteries in and below the renal hilus.
Atherosclerotic plaque-type lesions are the most common cause of RAS. Usually these are short and vary from a smooth circumferential narrowing to an irregular eccentric focal defect in the opacified renal artery. Most often these involve the renal artery orifice or the proximal third of the main renal artery, and they can progress to complete occlusion.
Also common, but usually occurring in younger female patients, are varying types of fibromuscular dysplasia lesions. The right renal artery is involved more often than the left. The stenoses are located in the distal two thirds of the renal artery and extend into the segmental branches. The arteriographic appearance varies from a prominent string-of-beads appearance to tubular symmetrical or eccentric arterial lumen deformity. These lesions can progress in severity of luminal narrowing.
Other rare renal artery lesions that can cause renovascular hypertension include those associated with arterial dissections, neurofibromatosis, musculotendinous bands, embolus, thrombosis, encasement by tumors, polyarteritis nodosa,

scleroderma, aneurysms, arteriovenous malformations, and effects of renal trauma.
Intravenous Digital Subtraction Angiography
Intravenous DSA has been advocated as a less invasive method for directly visualizing the renal arteries and associated lesions.25 The procedure is associated with less morbidity, because only a femoral venous catheterization is required. A significant problem associated with intravenous DSA studies is related to the high dose of contrast material required during multi-injection procedures. The high contrast load is an especially significant problem in patients with renal insufficiency, including the patients in this study population. Sensitivities and specificities for intravenous DSA in identifying RAS range from 83% to 100%. Svetky and colleagues179 prospectively evaluated the accuracy for identification of various diagnostic stenoses causing renovascular

hypertension. In their report, intravenous DSA had a sensitivity of 100% but a specificity of only 71% for predicting significant RAS. Most importantly, intravenous DSA is associated with a 6% to 12% occurrence of nondiagnostic studies, which are usually created by superimposition of bowel gas, poor cardiac output, or overlying visceral circulation.81 In addition, branch or segmental renal arteries are poorly delineated, which limits the evaluation of peripheral stenoses and fibromuscular dysplasia. Few community diagnostic radiologists have been able to duplicate the results of academic centers, which typically have much greater experience with the use of digital techniques. In addition, although the procedure is associated with less morbidity, it is quite expensive.
Radionuclide Renography
In the absence of captopril stimulation, radionuclide renogram studies for the detection of RAS typically had fair to poor results. With the advent of captopril, which inhibits the angiotensin-converting enzyme, improved results of radionuclide renography have been noted29 in patients being evaluated for renovascular hypertension. Addition of captopril enhances differences in glomerular filtration rate associated with the reduced perfusion state of RAS. It is clear that segmental RAS is not identified by captopril techniques. In addition, authors have noted significant false-negative findings in patients with bilateral stenoses.116 A significant finding in captopril-stimulated renography is asymmetrical differences in perfusion and glomerular filtration rate. In symmetrical vascular disease (e.g., bilateral RAS), the incidence of false-negative results increases. This represents a significant problem for captopril-stimulated radionuclide renography in the diagnosis of RAS in patients with progressive renal failure, because the principal cause of renal failure is bilateral disease. In addition, the volatile volume status in patients with renal insufficiency further complicates radionuclide renography. Large, multicenter trials of radionuclide renography are currently underway and should determine the usefulness of this test in comparison with other imaging modalities.81
Several investigators initially claimed that duplex Doppler evaluation of main renal artery velocity had potential value in identifying patients with RAS. Taylor and associates183 used a renal-to-aortic peak velocity ratio to identify patients with RAS in a group of 29 patients with angiographic correlation. They found that a peak velocity ratio greater than 3.5 was associated with a sensitivity of 84% and a specificity of 97% for detecting RAS identified on angiography. More recent studies have generated conflicting results, including sensitivities and specificities ranging from 30% to 90%.14 Therefore, there is little consensus as to the role of duplex Doppler sonography in the evaluation of RAS. Overall, the sample sizes have been small and the studies not well controlled. There is a wide variation in technical success, with reported rates of 82% to 98%, for obtaining Doppler signals from the main renal artery. One disadvantage of sonographic evaluation is the inability to accurately identify accessory renal arteries, which are seen in more than 20% of those examined.14
Handa and coworkers71 studied the acceleration of systolic velocity in branches of the main renal artery. This study of 20 patients revealed a sensitivity of 100% and a specificity of 93% for detecting RAS. However, the number of patients was small and the examinations were unusually lengthy. These results are sufficiently promising to warrant further research, but the ultimate role of Doppler ultrasound for this diagnosis has not yet been completely settled.
Magnetic Resonance Angiography
Although magnetic resonance angiography has proved successful in the evaluation of circulation in the head and neck, renal artery imaging poses particular challenges because of artifacts from cardiac, respiratory, and bowel motion and problems with stationary tissue suppression (Fig. 20-61). Problems are encountered in evaluation of renal vasculature because of overlap from renal veins, the inferior vena cava, and the adrenal/gonadal vasculature. Finally, the inherent vessel tortuousity and complex flow patterns, as well as widely disparate flow velocities in the aorta and renal arteries, create problems in imaging the renal vessels by magnetic resonance angiography.175 These difficulties have translated into variable success in the assessment of renovascular disease using MRI. A report of 37 patients by Kim and colleagues95 demonstrated an overall sensitivity of 100% and a specificity of 94% for detecting stenoses greater than 50% by diameter. These authors used coronal and axial breath-held gradient-echo images (time-of-flight technique). As they pointed out, the spatial resolution of the technique was adequate only to evaluate the proximal renal arteries. The excellent results may therefore be explained at least in part by the atherosclerotic origin of all detected lesions. The proximal location of the lesions makes them amenable to diagnosis by magnetic resonance angiography. No branch vessel lesions, fibromuscular dysplasia, induced stenoses, or lesions in accessory renal arteries were among the detected lesions.
FIG. 20-61. Magnetic resonance angiography of the renal arteries. A: Normal axial magnetic resonance imaging made with phase-contrast technique. Aorta (asterisk) and left renal vein (arrowheads) are shown. B: Renal artery stenosis. Note stenosis in proximal left renal artery (arrow). Aorta (asterisk) and left renal vein (arrowheads) leading to inferior vena cava are shown. (Courtesy of Thomas M. Grist, M.D., Madison, Wisconsin.)
A renal mass may be seen first on plain film, urography, ultrasound, or CT. If a typical cyst is suspected, then ultrasonography should be performed to confirm this finding. If there is an atypical cyst or solid mass, then CT should be performed to evaluate the mass. Percutaneous fine-needle aspiration biopsies can be performed using CT or ultrasonography for guidance on indeterminate lesions and solid masses. Angiography should be reserved for equivocal findings

on CT or ultrasonography; angiography or MRI can also be used to assess the renal vascular anatomy. Aspiration with contrast injection of atypical cysts may be required in some cases, but cyst punctures are not performed as often as they were before the use of CT and ultrasonography. Although not all solid renal masses are adenocarcinomas, because of the lack of accurate distinguishing features by any imaging modality, solid renal masses should be treated as if they represented a malignant lesion.43
Benign Tumors
Most benign renal tumors are small and asymptomatic; they are rare, and most are discovered at autopsy. Histologic types include adenoma, fibroma, lipoma, leiomyoma, hemangioma, and hamartoma; the renin-secreting juxtaglomerular cell tumor is very rare and is usually small, but it may be seen on arteriography as an avascular mass surrounded by a denser rim of compressed parenchyma. A few dilated, tortuous arteries may be present, and, of crucial importance in the diagnosis, there may be elevation of venous renin from the affected kidney. If the benign renal tumors are small, they may not be detected on EXU, although they may be seen on CT or ultrasonography. If they attain sufficient size, a plain-film roentgenogram reveals enlargement of the renal shadow at the site of the tumor. Urography may then show enough distortion of the pelvocalyceal system to make the diagnosis of renal tumor. The chief importance of these benign tumors lies in their differentiation from malignant tumors, which usually cannot be done with certainty by any imaging modality. A leiomyoma arising in the renal capsule may rarely contain calcium resembling that observed in other leiomyomas. Most benign renal tumors, except for hemangiomas, are avascular on angiography. Malignant tumors may be avascular, and benign tumors may exhibit abnormal vascularity. In current practice, CT and ultrasonography should be the imaging modalities used to evaluate renal masses. Angiography can then be used if needed to assess a tumor’s vascularity and the renal arteries. Retroperitoneal masses arising near the kidney may be difficult to differentiate from renal tumors on EXU, since they may distort the kidney and displace the upper ureter (Fig. 20-62). CT, ultrasonography, or MRI allows more accurate differentiation of intrarenal from extrarenal masses.
FIG. 20-62. Huge radiolucent tumor distorting the right kidney and ureter. This is a large retroperitoneal lipoma.

Renal Angiomyolipoma
Renal angiomyolipoma (hamartoma) is one of the few benign renal tumors that can often be differentiated from other benign tumors and from malignant renal tumors. It is a mixed mesodermal tumor composed of adipose tissue, smooth muscle, and blood vessels in varying proportions. Angiomyolipomas may be associated with tuberous sclerosis, in which case they are usually multiple and bilateral (Fig. 20-63). Angiomyolipoma may also occur as a solitary unilateral tumor, usually in older women. Clinical manifestations include signs of infection, pain, hematuria, or an asymptomatic abdominal mass. This tumor is prone to bleed, and patients may present in hypovolemic shock.
FIG. 20-63. Bilateral angiomyolipomas associated with tuberous sclerosis. Bilateral masses of fat attenuation are present in both kidneys in a patient with tuberous sclerosis.
Urographic findings are those of a mass that enlarges the kidney and distorts and displaces the pelvis and calyces. If much fat is present, the radiolucent areas within the mass suggest the diagnosis. If the tumors are multiple and bilateral, they may simulate polycystic disease. Typically, the angiomyolipoma appears as a focal hyperechoic mass on ultrasonography because of the fatty elements. However, Hartman and colleagues found in a retrospective review that a mixed or hypoechoic pattern could also be found. Renal-cell carcinoma can also have a hyperechoic pattern, so that this ultrasonic finding can be suggestive but not pathognomonic of angiomyolipoma.59,73 The demonstration of fat within a renal tumor on CT is considered diagnostic of angiomyolipoma (Fig. 20-64), although a few renal cell carcinomas containing fat have been reported.78, 176 The ease of detection of fat on CT is proportional to the amount of fat in the angiomyolipoma, since partial-volume artifacts and bleeding into the tumor can alter density measurements.21
FIG. 20-64. A and B: Angiomyolipoma. Enhanced computed tomograms at two different levels. Note the very low attenuation areas, compatible with fat in the tumor.
The most striking angiographic finding is the presence of many peculiar, small, regular outpouchings of the interlobar and interlobular arteries resembling berry aneurysms. In some patients, the interlobular arteries terminate in these aneurysms resembling a cluster of grapes, in contrast to the irregular size and contour of the tumor vessels seen in renal cell carcinoma. This appearance is present in the arterial phase and is obscured by the nephrographic phase. Later, there is irregular puddling, indistinguishable from that caused by malignant tumor. The venous phase is normal, not early as in renal cell carcinoma. Differentiation from renal cell carcinoma by angiographic means alone may be difficult.
Multilocular Cystic Nephroma
This rare condition consists of unilateral solitary cysts that contain numerous loculi that neither intercommunicate nor connect with the renal pelvis.9 This abnormality is usually found in childhood. The remainder of the kidney is normal. Some consider it a form of renal dysplasia; others believe it to be a benign neoplasm, a multilocular cystic nephroma. Rarely, diffuse calcification is present, resembling that seen

in renal cell carcinoma. Abdominal mass is the usual major presenting complaint. Roentgenographic findings resemble those of simple cyst on the urogram. Arteriography shows the mass to be avascular; vessels around it may be stretched, but no neovascularity or “tumor stain” is demonstrated. Ultrasonography and CT show a multichambered, cystic-type lesion. Cyst puncture is occasionally needed to eliminate the possibility of malignancy.
Lipomatosis of the Renal Sinus
This condition is also called fibrolipomatosis, fatty replacement, fatty transformation, lipomatous paranephritis, and lipoma diffusum renis. Because it may resemble tumor on urographic study, it is considered here. Lipomatosis consists of an excessive amount of fat in the renal sinus that distorts the calyces, infundibula, and renal pelvis to varying degrees. It is usually found as a replacement process in renal atrophy, whatever the cause, but it may occur in simple obesity. It is found in older age groups, usually in patients older than 50 years of age. The urographic appearance may simulate renal tumor, parapelvic cyst, or polycystic disease. It is important that differentiation be made, because lipomatosis is not a surgical problem.
The condition can usually be suspected on EXU, but nephrotomography best outlines the changes and facilitates the diagnosis. The pelvis is flattened or irregularly indented on its lateral aspect; the infundibula are elongated and narrow and often appear stretched. The calyces may be relatively normal, but they may be dilated and blunted if pyelonephritis is superimposed.
At times, the deposition of fat is localized so that cyst or tumor may be simulated. The diffuse type can also resemble polycystic disease. The infusion method of urography in conjunction with tomography demonstrates in sharp relief the relationship of the radiolucent fat to the opacified pelvocalyceal system (Fig. 20-65). CT also nicely demonstrates the parapelvic fat in renal sinus lipomatosis. The fat can be recognized by its low attenuation values, with the site and extent of fatty replacement also delineated.178 On ultrasonography, renal sinus lipomatosis appears hyperechoic, similar to but more so than normal fat in the sinus.
FIG. 20-65. A and B: Lipomatosis of the renal sinus. These tomograms show radiolucent fat in the renal sinus with elongation and narrowing of infundibula.
This process may be unilateral or bilateral. When it is bilateral, it is not necessarily symmetrical. Involvement on the left is often somewhat greater than on the right. The involved kidney may be enlarged, with some thinning of the remaining renal tissue.
Renal Pseudotumor
The kidney is capable of hypertrophy and hyperplasia. A focal mass or pseudotumor can be difficult to differentiate from a tumor on urography. Compression of the pelvocalyceal system, splaying of the calyces, and local enlargement

with protrusion from the renal surface may be present. Because the calyces do not regenerate, the masses do not contain these structures. In addition to the demonstration of a mass, EXU may reveal the signs of renal disease, which results in the regeneration of renal parenchyma including nonobstructive caliectasis and irregular thinning of the cortex. Enlarged Bertin’s column is an example of invagination of the renal cortex into the medulla, simulating a tumor. It is often found at the junction of the middle and upper third of the kidney.
Angiography reveals spreading of the arteries, but no tumor vessels or arteriovenous shunts are present, so there is no early venous filling. The capillary blush equals or exceeds that of the remaining renal parenchyma, and there is no evidence of a wall or capsule. Ultrasonography, radionuclide studies, or MRI can be used to differentiate pseudotumors from true tumors. However, helical or dynamic CT is generally the procedure of choice. Pseudotumors should have similar attenuation values with other normal parenchyma on both precontrast and postcontrast scans. Tumors, in comparison, are generally hypoattenuating on contrast-enhanced images. Angiography is sometimes used for this indication. Some tumors on ultrasonography can have very similar echogenicity to normal renal parenchyma and be missed.
A number of other conditions may resemble renal hilar or parenchymal masses. These include fetal lobulations, aneurysm of the renal artery, dilated veins, renal abscess, hematoma, xanthogranulomatous pyelonephritis, renal tuberculosis, and, most importantly and most often, renal cyst.
Malignant Tumors
Malignant tumors of the kidney can be divided into the following types: (1) renal cell carcinoma (adenocarcinoma or hypernephroma); (2) embryonal tumors (Wilms’ tumor); (3) tumors of the renal pelvis (transitional cell carcinoma); (4) mesenchymal tumors (sarcoma); (5) lymphoma, including leukemia; and (6) metastases. The adenocarcinoma or renal cell carcinoma is the most common malignant tumor, and it can arise in any portion of the kidney. The tumor may attain great size before it causes symptoms.
In general, solid renal masses in adults are surgically removed because of the difficulty in excluding malignancy by any imaging findings. Additionally, needle-biopsy specimens can be difficult for the pathologist to interpret without examination of the entire specimen.
Renal Cell Carcinoma (Hypernephroma, Adenocarcinoma)
Renal cell carcinoma occurs more frequently in men than in women and is most commonly seen in 40- to 60-year-olds. Clinically, the patient may present with weight loss, flank pain, palpable mass, or hematuria. Renal cell carcinoma has a propensity to metastasize to the lungs, liver, and lymph nodes, with tumor extension into the renal vein and inferior vena cava. Plain-film findings consist of local or general enlargement of the kidney, which varies with the size of the tumor. The renal border may be preserved (although lobulated or distorted), or it may be irregular and disrupted. The lesions are usually limited by the renal capsule until they are far advanced. It is not uncommon to note calcification, which may be irregularly scattered or curvilinear, within the tumor. Calcification may also be rim-like or curvilinear, outlining the periphery of the tumor. Renal displacement or tilting of the axis may result when there is a large medial mass in the upper or lower pole, or the entire kidney may be displaced if the tumor is large. Displacement of neighboring organs occurs when the tumor attains sufficient size. This may be apparent on a plain film, but is best appreciated with CT or MRI.139,169
Urographic changes are caused by the distortion produced by the tumor mass. Calyces are elongated, distorted, narrowed, or obliterated. The renal pelvis may be altered similarly. Renal cell carcinoma usually produces more disruption of the calyces or pelvis than a cyst of similar size. This disruption is important, because although cysts can elongate and compress the pelvocalyceal system, they tend to cause less distortion. Large tumors can cause considerable displacement of the upper ureter and may partially obstruct the pelvis or upper ureter (Fig. 20-66). There is usually enough function to visualize the calyces and pelvis on EXU, in contrast

to the loss of function often noted in hydronephrosis (which can produce renal enlargement) or with transitional cell carcinoma of the renal pelvis. Occasionally, infiltration of the kidney by tumor may be so extensive that no function remains. Also, function may be decreased by invasion and thrombosis of the renal vein.
FIG. 20-66. A through E: These excretory urograms indicate various findings produced by renal carcinoma. Note the considerable distortion of the calyces sometimes associated with the tumor mass. Tumor site in each case is shown by arrows.
The malignant cystic renal tumor must be differentiated from a simple renal cyst. In the past, urographic criteria were used to make this distinction. Renal cysts had a “claw sign” in the nephrographic phase and a crowding together of the calyces. Calyces were rarely amputated, and the cyst was attached to the periphery of the kidney. Renal tumors, conversely, demonstrated invasion of the renal pelvis, separation of the calyces, calyceal amputation, and a mass contiguous with the body of the kidney.158 These criteria, however, were not accurate in distinguishing cyst from tumor, which is currently done with the use of ultrasonography and CT. Pollack and colleagues153 demonstrated an accuracy rate of 98% for diagnosing a simple cyst when certain criteria were fulfilled on ultrasonography. These criteria included (1) enhancement of sound transmission beyond the cyst, (2) absence of internal echoes, (3) sharp delineation of the far wall, and (4) a spherical or slightly ovoid shape (Fig. 20-67). CT criteria for a simple cyst include (1) homogeneous attenuation value near that of water density, (2) no enhancement with intravenous contrast, (3) no measurable thickness of the cyst wall, and (4) a smooth interface with renal parenchyma (Fig. 20-68).118
FIG. 20-67. Simple renal cyst. Longitudinal ultrasound of a large cyst in the upper pole of the right kidney. Note the absence of internal echoes and the through-transmission (arrows) beyond the cyst.
FIG. 20-68. Simple renal cyst. Dynamic enhanced computed tomogram of a cyst in the posterior aspect of the left kidney. Note the smooth interface with the renal parenchyma and the lack of enhancement within the cyst.
When cysts do not meet the criteria for a simple renal cyst, they are best categorized by the criteria of Bosniak.20 This system classifies renal cysts into four types. Type 1 is a simple renal cyst. Type 2 is a minimally complicated cyst with increased attenuation values, thin calcifications in the periphery, or thin septations (Fig. 20-69). Type 1 and 2 cysts have essentially no chance of malignancy. Type 3 cysts have thick septations or chunky calcifications, uniform thick wall, or nonenhancing nodules. Type 4 cysts are frankly malignant, with thick walls, enhancing components, and solid enhancing nodules. Type 3 and 4 cysts have a large chance of malignancy (57% and 100%, respectively).5,20 More recent literature has reaffirmed the value of the Bozniak classification but has suggested that type 2 lesions have a higher rate of malignancy than previously reported.198
FIG. 20-69. Hyperdense renal cyst. Unenhanced computed tomogram shows the homogeneous, high-attenuation cyst in the lateral portion of the left kidney.
Renal cell carcinoma on CT appears as a solid lesion (with or without cystic components) that deforms the renal contour. The interface between the tumor and normal renal parenchyma may be difficult to define on precontrast scans but is usually clear after injection of intravenous contrast material. Postcontrast scans usually show the tumor enhancing less than the normal renal parenchyma (Fig. 20-70).49 Low-attenuation regions within the tumor that do not change after addition of contrast material are compatible with necrosis or hemorrhage. CT also can be used in the staging of renal cell carcinoma (Fig. 20-71). More recently, MRI has

been used in the evaluation of renal cell carcinoma. It is generally agreed that, although most renal cell carcinomas can be detected by MRI, some may be isointense to normal kidney on T1- and T2-weighted sequences. Newer literature suggests that fat-suppression and breath-hold techniques may be as sensitive as CT for detection of renal masses. MRI may be more advantageous in evaluating vascular patency (Fig. 20-72), detecting perihilar lymphadenopathy, and assessing direct tumor invasion of adjacent organs.92 Staging of renal adenocarcinoma is now performed primarily by dynamic CT, helical CT, and MRI.200
FIG. 20-70. Renal cell carcinoma. Enhanced computed tomogram shows lack of enhancement of tumor (asterisk) relative to normal parenchyma. Tumor thrombus involves right renal vein and extends into inferior vena cava (arrows). (Courtesy of Phillip Murphy, M.D., Rochester, New York.)
FIG. 20-71. A: Helical computed tomography (CT) image during the arterial phase demonstrates a hypervascular mass (arrow). Note fluid collection surrounding kidney (asterisk), consistent with a spontaneous bleed from this tumor. B: Helical CT during the parenchymal phase. The renal mass is now lower in attenuation than the surrounding normal renal parenchyma.
FIG. 20-72. Renal cell carcinoma. A: Coronal T1-weighted magnetic resonance image (MRI) of a large right renal cell carcinoma. B: Coronal MRI in a different patient shows tumor thrombus (arrow) within the inferior vena cava. C: Axial time-of-flight image demonstrates dark clot in caval lumen (arrow).
Both ultrasonography and CT can be used to guide percutaneous biopsies of renal masses for cytopathology or histology. On ultrasonography, renal cell carcinoma can be isoechoic, hypoechoic, or hyperechoic compared with the normal renal parenchyma (Fig. 20-73).59 Ultrasonography can also be used to evaluate renal vein and inferior vena caval involvement.
FIG. 20-73. Renal cell carcinoma. Longitudinal ultrasound study demonstrates a solid exophytic mass in the midportion of the kidney (arrow), typical of renal cell carcinoma.
Angiography is infrequently used for the diagnosis of renal cell carcinoma. Angiography of renal cell carcinoma may show (1) increased vascularity with irregular pooling and arteriovenous communications with early venous filling (Fig. 20-74), (2) relative avascularity of the entire tumor or a portion of it, (3) abnormal circulation by way of capsular or extrarenal vessels, (4) venous collaterals and abnormal peripheral venous channels around the mass, and (5) lack of constrictor response to epinephrine in tumor vessels. Transcatheter embolization of renal cell carcinoma has been performed.61 It is done preoperatively to aid in surgical removal because the embolization causes tumor vessels to collapse, resulting in decreased operating time and decreased blood loss. In patients with inoperable tumors, it is undertaken to relieve symptoms and reduce tumor size. Materials that have been used for embolization include autologous clot, Gelfoam, Ivalon (a polyvinyl alcohol), microspheres, muscle tissue, steel coils, and ferromagnetic silicone. Balloon catheter occlusion has also been used.
FIG. 20-74. Renal cell carcinoma as seen on the selective renal arteriogram. A: At 1.5 seconds there are abnormal vessels within the lower pole mass while arteries surrounding it are stretched. B: At 3 seconds, more puddling and arteriovenous shunting is visible with many abnormal “tumor” vessels. C: At 16 seconds, the renal vein is opacified. Note also the abnormal veins lateral to the mass, outside the kidney.
Wilms’ Tumor
Wilms’ tumor, or nephroblastoma, is the most common abdominal neoplasm of infancy and childhood. Wilms’ tumor arises from embryonic renal tissue and tends to become very large. Most Wilms’ tumors arise in the first 5 years of life but are rarely present at birth, in contrast to neuroblastoma or the fetal renal hamartoma of the kidney. Nephroblastomatosis, or persistence of subcapsular deposits of primitive nephroblastic tissue, is considered to represent

a precursor of Wilms’ tumor. Wilms’ tumor is usually unilateral (at least 95% of cases) and appears as an abdominal mass. Hematuria is not common, and pain is present in about one fourth of cases. Scout roentgenograms show the outline of the mass with displacement of neighboring structures and elevation of the diaphragm on the side of the lesion. Wilms’ tumor may occasionally contain calcifications, in contrast to neuroblastoma, which also causes a large tumor mass in infants and children but contains calcifications on plain film in 50% of cases.
Urographic findings are those of a large intrarenal tumor that distorts the calyces and pelvis and often displaces and partially obstructs the ureter. The distortion of the calyces tends to be less than with renal cell carcinoma of similar size but is greater than with neuroblastoma, which often arises adjacent to the kidney and displaces it. Renal function may be impaired, but there is usually enough function to outline some of the calyces on urography and to differentiate this tumor from hydronephrosis causing massive renal enlargement. Ultrasonography usually shows a homogeneous, echogenic renal mass; there may be small hypoechoic regions that represent cysts within the tumor.37 CT is useful to confirm the intrarenal location of Wilms’ tumor (Fig. 20-75) and can also demonstrate necrosis or hemorrhage, the absence of vessel encasement (vessel encasement would suggest neuroblastoma), and the distortion of the renal calyces. Retrocrural lymphadenopathy, if seen on CT, would suggest neuroblastoma. In one study of 15 patients with Wilms’ tumor, no retrocrural lymphadenopathy was seen.104 MRI is particularly useful in cases in which the intrarenal

location of a retroperitoneal mass is in doubt because of the multiplanar reformat capability of MRI. Angiography is not often done. Arteriovenous shunts are not seen, nor is any puddling or pooling of contrast in venous lakes. Tumor vessels that are long and tortuous and that may resemble a creeping vine are seen. These vessels tend to be discrete, to be of large caliber, and to have an irregular diameter. Invasion, obstruction, or displacement of the inferior vena cava can be detected by cavography, although ultrasonography, MRI, or CT (Fig. 20-76) is more commonly used now. Wilms’ tumor tends to metastasize to the lungs and para-aortic lymph nodes and can also extend locally by direct invasion. Calcification in metastases has been reported but is extremely rare.
FIG. 20-75. Wilms’ tumor. A: Longitudinal ultrasound examination of the left kidney shows a large mass arising in the lower pole, causing upper pole hydronephrosis. B: Computed tomography scan demonstrates similar findings. In contrast to neuroblastoma, note that the aorta and inferior vena cava (arrow) are displaced rather than enveloped by the mass.
FIG. 20-76. Wilms’ tumor. Enhanced computed tomogram shows involvement of the inferior vena cave (arrow) by the low-attenuation tumor thrombus. The primary tumor is seen in the left kidney.
Tumors of the Renal Pelvis
Tumors of the renal pelvis are of epithelial origin and present a different urographic picture than renal cell carcinoma. There are two major types of malignant lesions: transitional cell carcinoma and the squamous cell carcinoma that arises from squamous metaplasia of urinary tract epithelium. Transitional cell tumors, which comprise almost 90% of malignant

tumors of the renal pelvis, tend to be somewhat less invasive than the squamous cell type. The latter is frequently associated with chronic infection, leukoplakia, or calculi. These tumors produce symptoms of hematuria, pain (obstructive type), and sometimes a palpable mass caused by obstructive hydronephrosis or a large tumor with perirenal extension. On the plain film, there may be no sign of tumor. Because hematuria is an early sign, the patients are usually examined when the lesion is small and therefore difficult to detect in the calyx or pelvis. The tumor causes a filling defect that may be irregular or smooth and may be small or large. These defects are outlined on urograms as radiolucent areas projecting into the opacified calyx or pelvis (Fig. 20-77). Malignant tumors are usually more irregular than benign papillomas of the pelvis, but urographic differentiation of the various cell types of renal pelvis tumors is not possible. The urographic evidence of an infiltrating type of tumor may be minimal, but it can also become very large and produce major alterations in the renal pelvis (Fig. 20-78). Blood clots and radiolucent calculi can produce similar defects. For this reason, it is common practice to follow EXU by retrograde pyelography when a filling defect is noted. If a blood clot or calculus caused the defect, the second examination will show disappearance or alteration in the size or position of the defect. Calcification may occur within this type of tumor but is rare. Ureteral and bladder implants occur frequently and produce small defects similar to those caused by the primary tumor in the kidney. Occasionally, a tumor may invade and infiltrate the adjacent parenchyma to simulate renal cell carcinoma.
FIG. 20-77. Carcinoma of the renal pelvis. Note the irregular filling defects of the upper-pole calyces and infundibula. The lesion was a transitional cell carcinoma.
FIG. 20-78. Infiltrating carcinoma (transitional cell) of the renal pelvis. Note gross distortion of infundibula and calyces. The pelvis does not opacify because it is full of tumor.
CT can be used to differentiate a tumor from a radiolucent calculus in the renal pelvis.168 Baron and colleagues11 recommended that CT be performed after EXU in patients with known or suspected transitional cell carcinoma for diagnosis,

preoperative staging, and evaluation of whether limited resection is possible. On CT, transitional cell carcinoma appears as a soft-tissue density within the renal pelvis (Fig. 20-79).152
FIG. 20-79. Carcinoma of the renal pelvis. A: Retrograde pyelogram demonstrates filling defect in the right renal pelvis. B: Enhanced computed tomographic scan of same patients as in A shows tumor in renal pelvis (arrow). Soft-tissue mass definitively excludes low-density renal calculus as cause of filling defect. (Courtesy of Patrick J. Fultz, M.D., Rochester, New York.)
Transitional cell carcinoma is relatively hypovascular on angiography. There is no pooling, arteriovenous shunting, or neovascularity. The residual parenchymal vessels are occluded or encased by tumor and narrowed. These angiographic findings, however, may also be found in metastases to the kidney.
Squamous Metaplasia of the Renal Pelvis
Leukoplakia or squamous metaplasia of the renal pelvis is probably caused by infection or chronic irritation from another source. It is included here because it may resemble carcinoma of the pelvis and often precedes squamous cell carcinoma of the renal pelvis. Infection is found in 80% of patients, and 40% have had renal calculi. The patient may describe passing tissue or gritty material, and the diagnosis is established by finding keratinized squamous epithelium in the urine. The condition is usually unilateral. Clinically, hematuria may be present with intermittent chills or fever also occurring.
Radiographic findings are varied. Irregular areas in the renal pelvis partially surrounded by contrast material, large laminated masses presenting an onion-skin appearance, irregular plaques or bands producing linear striations, and roughening or wrinkling of the renal pelvis may be found. Any of these findings should suggest the diagnosis, but lucent calculi, hematoma, and carcinoma of the renal pelvis must be considered in the differential diagnosis.
Sarcoma of the kidney may originate from fibrous elements in the renal parenchyma or renal capsule or from smooth muscle rests in the parenchyma or blood vessels. Retroperitoneal sarcoma arising in or near the kidney often becomes so extensive that it is not possible to determine the site of origin even at autopsy. Renal sarcomas are rare tumors. Fibrosarcoma, liposarcoma, leiomyosarcoma, rhabdomyosarcoma, and even osteosarcoma of the kidney have been reported. Roentgenographic findings are those of a mass in the renal area that is often difficult to outline clearly and may obliterate the psoas shadow in its more cephalad aspect. Urograms tend to show somewhat less distortion of the pelvocalyceal system than is noted in a renal cell carcinoma of similar size. Neifeld and colleagues138 stated that CT is valuable in suggesting more extensive involvement than may be appreciated with other diagnostic modalities or by clinical examination. CT can also be used to monitor patients for recurrence of the sarcoma. Liposarcoma is the most common retroperitoneal sarcoma.

Lymphoma and Leukemia
Involvement of the kidneys in patients with chronic leukemia may occur late in the disease. Renal involvement is more common in children with acute leukemia than in those with the chronic form. The leukemic infiltrate tends to be largely cortical in location. Renal enlargement, usually bilateral, is demonstrated on plain films. Urographic signs in addition to bilateral enlargement consist of enlargement of the renal pelvis (without dilatation or evidence of obstruction), stretching and elongation of the calyces and pelvis, and irregularity of the renal outline. CT is useful in demonstrating the renal enlargement, associated lymphadenopathy and splenomegaly, and complications such as hemorrhage.76,91 Breatnach and colleagues22 reported a case of intrarenal chloroma causing obstructive uropathy.
The kidneys can also be involved with lymphoma; most cases represent secondary involvement. Non-Hodgkin’s lymphoma involves the kidneys more frequently than Hodgkin’s lymphoma does. The distribution of disease is somewhat more varied than in the leukemias. Renal enlargement (caused by infiltration of the kidneys) is a common form of involvement. Other findings include distortion or elongation of the pelvocalyceal system, solitary or multiple tumor nodules, and perirenal masses that may engulf or displace the kidney (Fig. 20-80). The solitary tumor nodules cannot be differentiated from primary renal tumors by urography, CT, ultrasonography, or MRI. Multiplicity of tumor nodules should suggest the possibility of lymphoma. Some lymphomas with multiple renal masses may resemble polycystic disease on urography or angiography, but CT or ultrasonography can be used to demonstrate the solid nature of these masses. Retroperitoneal adenopathy, along with the CT or MRI findings of renal nodules, renal enlargement, or renal infiltration, can suggest the diagnosis of renal lymphoma.170 CT can also be used to evaluate the course of lymphoma and its response to therapy.90
FIG. 20-80. Renal lymphoma. Enhanced computed tomogram shows large soft-tissue masses involving both kidneys. (Courtesy of Bevan Bastian, M.D., Green River, Wyoming.)
Metastases to the Kidney
Metastatic tumors to the kidney are twice as common as primary renal tumors. However, most of these are small and are discovered at autopsy. If lymphoma is excluded, lung cancer is the most common source of metastases. Breast, stomach, colon, cervix, pancreas, and contralateral renal tumors can metastasize to the kidney, as can melanoma. These metastases usually are bloodborne but can occur from lymphatic spread or direct extension. In a series of nine patients studied by Mitnick and colleagues,127 the tumors were usually multiple and bilateral.
Metastases to the kidney may be detected with EXU, but the increasingly widespread use of CT and ultrasonography in oncology patients may allow for increased detection rates. The metastases are best seen on dynamic enhanced CT scans as regions of low attenuation relative to the normal parenchyma. The appearance on ultrasonography in the series by Mitnick and colleagues127 varied from hypoechoic to a mixed pattern of hypoechoic and more echogenic regions. The amount of vascularity and the histology of the primary tumor may determine the appearance on ultrasonography.
Renal amyloidosis may be primary or secondary to chronic inflammatory disease. Plain-film findings include bilateral, symmetrical renal enlargement with normal collecting systems. Late in the disease the kidneys may become small and have diminished function. Angiographic findings recorded in a few reports include slightly decreased renal artery size, pruning, tortuousity and irregularity of the distal interlobar arteries, a relatively homogeneous nephrogram, nonvisualization of the interlobar arteries, prominent extrarenal arteries, and uneven involvement of the kidneys. Renal vein thrombosis is common. When the disease is restricted to the renal pelvis, linear submucosal calcification outlining the pelvis may be noted. CT findings reflect the angiographic features of the venous thrombosis component.
Renal scleroderma may show changes similar to those found in the kidneys of patients with advanced malignant hypertension. In patients with scleroderma, the nephrogram phase of the roentgenographic examination is quite characteristic. Spotty lucencies are seen scattered throughout the kidney, with a delay in arterial flow manifested by persistent filling of arteries during the nephrogram phase.199
Chronic Glomerulonephritis
Renal cortical calcification somewhat similar to that observed in patients who survive acute renal cortical necrosis

is seen rarely in patients with chronic glomerulonephritis. The kidneys of patients with chronic glomerulonephritis are usually symmetrically small with relatively normal collecting systems.
Multiple Myeloma
In multiple myeloma, renal involvement is the result of precipitation of abnormal proteins in the tubules. Findings are those of bilateral, smooth renal enlargement with normal collecting systems. Parenchymal thickness is increased. Later in the disease, renal failure with oliguria may result in small kidneys. In this disease, it is essential that the patient be well hydrated during EXU to decrease the risk of precipitation of abnormal urinary protein in the tubules, causing renal failure.
Agnogenic Myeloid Metaplasia
When extramedullary hematopoiesis occurs in the renal hilus, it may simulate a parapelvic cyst or neoplasm.192
Nonobstructive renal enlargement has been noted in patients with hemophilia, primarily in children.
Sickle-Cell Hemoglobinopathy
Sickle-cell hemoglobinopathy may be a cause of renal papillary necrosis. It may occur in SS, SC, and SA disease. Papillary necrosis in these patients is similar to that in analgesic-abuse patients. Bilateral renal enlargement has also been reported in patients with sickle-cell hemoglobinopathies and in thalassemia but is not found in many of these patients.
The hypercalcemia frequently observed in patients with sarcoidosis may result in nephrocalcinosis and nephrolithiasis. In addition, the granulomas of sarcoidosis may involve the kidney. The granulomatous disease is usually not severe enough to cause any recognizable alteration, but the nephrocalcinosis and renal calculi can be observed. Usually there is no deformity of the collecting system. There may be some irregularity of the renal outline secondary to scarring related to tubular atrophy in patients with long-standing hyperuricemia.
Radiation Nephritis
Acute radiation nephritis develops after a period of 6 to 12 months after radiation treatment for malignancy. The urogram and CT may show only slight diminution in excretion. In chronic radiation nephritis, there is glomerular damage, tubular atrophy, and interstitial fibrosis as well as damage to smaller arteries and arterioles. Atrophy involves the area irradiated, whether it be a portion of the kidney or the entire kidney, which is then decreased in size. The other finding is diminution in renal excretion, which may be observed when comparison is made with the opposite, normal kidney. Hypertension often develops in patients with severe renal damage. In some, the condition progresses to malignant hypertension, which is relieved if the involved kidney is removed. The cause of hypertension in these patients is not entirely clear.
Nephrosclerosis refers to the alteration in renal parenchyma that results from decreased arterial blood flow due to diffuse arteriosclerosis; therefore, the condition is associated with renal ischemia. Some patients have this small-vessel disease as a result of predisposing disease such as diabetes mellitus. Nephrosclerosis can lead to local infarction. The imaging findings depend on the type of involvement. The infarcts produce a renal scar that appears as an irregularity indenting the renal cortical margin. The collecting system is usually normal. When the disease is uniform and widespread, the kidney decreases in size and there is also evidence of decrease in function.
Radiology in Renal Transplantation Patients
Arteriographic studies are obtained on potential kidney donors to identify the number of renal arteries, to demonstrate unsuspected disease involving the renal arteries, and to outline the anatomic variations. CT angiography has been shown to be both cost-effective and efficacious for this purpose, and in some centers it has replaced conventional angiography. In the post-transplantation period, both radionuclide scintigraphy and ultrasonography are valuable imaging modalities.
Radionuclide scintigraphy can be used to evaluate parenchymal failure in the renal transplant caused by rejection or acute tubular necrosis. 99mTc-DTPA and 131I-Hippuran are the radiopharmaceuticals most commonly used to evaluate the renal transplant, although use of 99mTC-mertiatide is becoming more common.182 Evaluation of renal perfusion, parenchymal accumulation, and transit time on serial scans can help to distinguish acute tubular necrosis from transplant rejection. Rejection usually is manifested by progressive deterioration of renal perfusion and function, whereas acute tubular necrosis shows improvement or a plateau in perfusion and function after an initial drop at about 24 to 48 hours after transplantation.48 Radionuclide scintigraphy can also help to evaluate mechanical injuries to the blood vessels or problems with urinary drainage in the transplant.
Ultrasonography may demonstrate several findings compatible with post-transplant rejection. Medullary edema, increase in transplant size, increase in cortical echoes, decreased

parenchymal echoes, indistinct corticomedullary boundary, and decreased renal sinus echoes have been reported as anatomic features of rejection (Fig. 20-81). Serial scans are again important to document changes from a baseline study. Ultrasonography can also evaluate hydronephrosis and perinephric fluid collections around the transplant (e.g., urinoma, hematoma, lymphocele).
FIG. 20-81. Renal transplant rejection. A and B: Normal renal allograft. Ultrasound demonstrates normal renal sinus and cortex. Duplex Doppler study has normal computed resistive index (RI) of 70.7%. C and D: Acute rejection. Allograft is now swollen with loss of differentiation between sinuses and parenchyma. Collecting system is more prominent. Computed RI is now 87.0% with noticeable lack of diastolic flow.
Doppler ultrasound, both duplex and color, is widely used to evaluate renal transplants. Because of its ability to distinguish flowing blood from other structures, Doppler technology greatly assists in locating and characterizing transplant vascularity. Complications such as arteriovenous fistulae, RAS, renal infarction, renal vein thrombosis, and pseudoaneurysms may be diagnosed by ultrasound, often reducing the necessity for angiography.173,184
Sampling of arcuate and segmental arteries with pulsed Doppler ultrasound in normal renal allografts yields characteristic low-impedance antegrade flow during diastole. Early studies showed that acute rejection could be diagnosed by demonstrating increased vascular impedance leading to decreased or reversed flow during diastole.137,160 This was quantified by the resistive index (RI) or the pulsatility index (PI).
A PI greater than 1.5 or an RI greater than 0.9 was thought to be a specific sign of acute rejection. A resistive index of 0.8 to 0.89 was thought likely to correspond to rejection.160

Later work demonstrated that other complications, such as acute tubular necrosis, pyelonephritis, external compression, chronic rejection, and renal vein obstruction, can also elevate these indices. Therefore, needle biopsy, often performed under ultrasound guidance, is needed to evaluate transplant dysfunction definitively after surgical complications have been excluded.47,155,156
EXU should play a limited role because of the possibility of renal damage and the availability of ultrasonography and radionuclide scintigraphy. Angiography also plays a more limited role currently because it is an invasive procedure and involves the use of a contrast agent. If an angiographic procedure must be performed, DSA may be preferable to evaluate RAS or an abnormal vascular pattern.
The role of MRI in renal transplantation is still being evaluated. One study61 was able to distinguish hematomas from lymphoceles. Patients with acute rejection showed a decrease in corticomedullary differentiation and a decrease in overall signal intensity compared with a baseline state. The corticomedullary junction could not be differentiated in chronic rejection.61 MRI spectroscopy may prove useful in making the specific diagnosis of rejection noninvasively, but this remains under investigation.70
Tumors of the Ureter
Benign tumors of the ureter are rare but include papilloma, hemangioma, fibroepithelial polyp, fibrolipoma, and leiomyoma. Papillomas are the most common benign ureteral tumor.167 They are epithelial in origin, and there is some controversy regarding the premalignant potential of these tumors. Some pathologists consider papillomas to be grade I malignancies. Radiographically, papillomas most often appear as a filling defect that can not be differentiated from transitional cell carcinoma by imaging. The fibroepithelial polyp is composed of a core of fibrous tissue that is covered by a layer of transitional epithelium. The polyp is often a long, branched, smooth, intraluminal structure that becomes quite large. A polyp reported by Howard86 measured 10 by 3 cm. Multiple polyps may be present. Polyps can cause obstruction or ureteral intussusception, but the obstruction may not be as severe as the size of the polyp might indicate. Fibroepithelial polyps can be quite mobile, with long, worm-like projections (Fig. 20-82). The nonpolypoid tumors are not as common. Endometriosis of the ureter may simulate tumor because the lesion can invade the ureter and penetrate the mucosa. Endometriosis more often is an extrinsic lesion that involves the adventitia. If the ureter is involved, this usually occurs below the pelvic brim. Obstruction may be severe enough to cause symptoms.
FIG. 20-82. Fibroepithelial polyp of the ureter. Note long stalk seen as a left ureteral filling defect (arrows) with eventual protrusion through the ureteral orifice into the bladder (arrowhead).
Most malignant tumors of the ureter are epithelial. Transitional cell carcinoma is the most common epithelial tumor and is usually papillary in form. The papillary tumors tend to be multiple. When multiple tumors are found, this is usually attributed to a multicentric origin. Roentgenographic findings do not differentiate the various types of epithelial tumors. Obstruction of the ureter is common and leads to hydronephrosis. An intraluminal tumor mass may be seen in addition to the ureteral obstruction on EXU. An infiltrating carcinoma occasionally may result in local narrowing of the lumen, simulating a benign ureteral stricture (Fig. 20-83). In some cases there is no ureterectasis above the tumor even if it is large. The ureter is apparently able to dilate locally to accommodate the tumor as it grows, so that obstruction does not occur. The entire length of the ureter must be seen in these instances. The only roentgenographic sign is an intraluminal mass appearing as a negative filling defect that must be outlined by contrast material within the ureter to be visualized. A localized dilatation immediately below the tumor has been described (Fig. 20-84). If retrograde pyelography is attempted, the catheter tip may coil in the region of localized dilatation, the intraluminal mass impeding its

upward progress; this is known as Bergman’s sign. The coiling of the catheter is then a sign of ureteral tumor, because local dilatation does not occur below a ureteral calculus unless it is large, calcified, and easily detected. It is difficult to demonstrate both the upper and lower borders of a negative filling defect on one examination. EXU may be required to demonstrate the upper border and retrograde pyelography to demonstrate the lower border in these cases. Tomography of the ureter at the time of urography is helpful in some instances. CT can be used to detect periureteral extension in patients with transitional cell carcinoma and can demonstrate metastases to enlarged lymph nodes. It cannot, however, distinguish tumors with muscle wall invasion from those limited to the mucosa. The tumors approach soft-tissue density, with some of the tumors noted to increase in attenuation value on the enhanced scans.11
FIG. 20-83. Carcinoma of the left ureter (arrow) has produced partial obstruction hydroureter and hydronephrosis above it.
FIG. 20-84. Ureteral carcinoma. The tumor produces a filling defect that is somewhat irregular superiorly and demonstrates dilatation below the tumor.
Metastases to the ureter from tumors outside the urinary tract may cause local involvement with little or no extrinsic mass. Alternatively, the ureter may be caught in a retroperitoneal mass, displaced, and narrowed. Metastases to the ureter on CT are demonstrated as a soft-tissue thickening or mass in the ureteral wall (Fig. 20-85). There may be associated ureterectasis and hydronephrosis proximal to the lesion. Although selective arteriography can be used to identify ureteral tumors and help differentiate them from strictures, it is rarely used.
FIG. 20-85. Ureteral metastasis. Enhanced computed tomogram shows the thickened right ureteral wall (white arrow) in this patient with breast cancer. Note contrast in the normal left ureter (black arrow).
Other Ureteral Abnormalities
Ureteral Displacement
The ureter may herniate into the inguinal canal, the femoral canal, or the sciatic notch or behind the inferior vena cava. The relationship of the ureter to these various structures usually determines the diagnosis, and there is often a portion of the bowel extending into the hernia. Ureters are

also displaced in patients with uterine prolapse or procidentia. Primary or secondary tumors or massive lymph nodes may also displace one or both ureters and may produce obstruction. Fecal impactions in the sigmoid colon may displace the ureter in a manner simulating a pelvic mass. In patients with Crohn’s disease, a ureter can be caught in an inflammatory mass and displaced or obstructed by the accompanying fibrosis. In patients with aortic aneurysm, there may be traction displacement medially of one of the ureters toward the aneurysm. Presumably this is an indication of retroperitoneal bleeding with resultant fibrosis causing the displacement. Pelvic lipomatosis may also result in alteration in the course of the ureters and in some instances may produce ureteral obstruction. More often there is alteration in the appearance of the bladder, which is elevated and elongated vertically (pear-shaped bladder).
Schistosomiasis can cause calcification in the ureteral wall as well as medial deviation, a straight lumbar course, or a bowed appearance in the pelvis with medial and upward displacement at the level of the trigone. There may be some stasis and dilatation of the upper urinary tract, owing mainly to fibrosis rather than to mechanical obstruction.
Polyarteritis can cause dilatation and nodular irregularity of the ureteral wall simulating a string of pearls.
Amyloidosis may involve the ureter, causing a stricture-like defect that can partially obstruct it, producing colicky pain and hematuria as well as ureterectasis above the stricture.108
Primary Megaureter (or Adynamic Distal Ureteral Segment)
Primary megaureter is the most common cause of ureteral obstruction in the presence of orthotopic ureters and is responsible for 20% of cases of neonatal hydronephrosis.24 It is characterized by varying degrees of dilatation of one or both ureters just above the bladder, often without evidence of anatomic obstruction or reflux.147 Usually there is no upper ureteral, pelvic, or calyceal dilatation unless infection is present. Maximal dilatation occurs immediately proximal to the aperistaltic segment near the ureterovesical junction, which averages 1.5 cm in length and is relatively narrow. It may not fill on EXU. Videotaped fluoroscopy may be used to demonstrate the disturbed peristalsis. The peristaltic activity is normal or vigorous in the proximal dilated ureter, but peristalsis is absent in the distal ureteral segment. Other genitourinary anomalies are seen in approximately 40% of cases.149
Retroperitoneal Fibrosis
Originally called periureteral fibrosis, this disease is more aptly termed retroperitoneal fibrosis because it may involve lymphatic and vascular structures in addition to the ureters. The cause of the disease is unknown, and it is characterized by a fibrosing inflammatory process in the retroperitoneum that may extend from the kidneys caudally to the pelvic brim and spread laterally to involve the ureters. In addition to the idiopathic cases (about 70%), several cases have been reported in patients receiving long-term methysergide (Sansert) therapy for migraine headaches. If the drug is withdrawn, the process may regress. However, if methysergide therapy is resumed, the process seems to recur. Other drugs such as phenacetin and methyldopa have also been implicated. Patients with retroperitoneal neoplasms (primary or metastatic) may also have a fibrotic reaction similar to that seen in the idiopathic form.
Fibrosis of the orbit, duodenum, rectosigmoid, common bile duct, and pancreatic duct have been reported in association with retroperitoneal fibrosis. Involvement of the splenic vein, vena cava, celiac axis, superior mesenteric artery, or iliac arteries may also be found. Males are affected twice as often as females, and the condition is usually bilateral. Retroperitoneal fibrosis affects patients 8 to 80 years old, with the peak incidence in the fifth and sixth decades.78 The patients can have back, flank, or abdominal pain. A palpable abdominal or rectal mass is noted in 30% of cases.
The radiographic findings may suggest the diagnosis. The normal fat lines may disappear so that the outlines of the psoas muscle are not visible on plain film. EXU may show delayed excretion with varying degrees of hydronephrosis or absence of excretion caused by obstruction. If the ureters are opacified, they gradually taper to the area of maximum stenosis. The involved segment is often 4 to 5 cm in length, usually with medial deviation of the tapered ureters at the level of the lower lumbar vertebrae. Some slight redundancy may be observed in the ureter above the stenotic segment. Retrograde pyelography can also reveal the site and length of obstruction, and there is often a paradoxical ease of retrograde passage of ureteral catheters. Lymphangiography and inferior vena cavography were also used in the past to demonstrate lymphatic or vascular obstruction but are rarely used at present.
CT can provide direct information about the fibrosis and also allow evaluation of vascular and urologic structures. The fibrosis is easy to recognize when it appears as a large, bulky mass. The appearance is nonspecific, and lymphoma, hematoma, sarcoma, and lymph node metastases can cause a similar appearance. The location may be helpful, because lymphomas are found higher in the retroperitoneum and the fibrosis is usually located caudal to the renal hilum. The attenuation values on unenhanced scans are in the range of

muscle. After injection of intravenous contrast material, the density of the fibrosis may markedly increase, simulating a vascular neoplasm.70 Ultrasonography also demonstrates a solid retroperitoneal mass, which usually envelops but does not displace the adjacent vascular structures and ureters. MRI is becoming increasingly useful in the diagnosis and follow-up of patients with retroperitoneal fibrosis. In particular, MRI may help assess vascular patency and distinguish malignant from nonmalignant varieties based on signal characteristics. T2-weighted images in the malignant form show increased signal intensity when compared with T1-weighted images; benign retroperitoneal fibrosis has low signal intensity on both T1- and T2-weighted images. Some reports have shown overlapping signal characteristics of the different forms, although it is distinctly unusual for malignant retroperitoneal fibrosis to have low signal intensity on T2-weighted images.2,6,133 Early recognition and surgical ureterolysis are important to preserve renal function in idiopathic disease. The vascular obstruction may be more important clinically than the ureteral obstruction in some instances.
Congenital Anomalies
Exstrophy of the bladder is of imaging interest because of the wide separation of the pubic bones anteriorly at the symphysis that accompanies this anomaly and the increased incidence of bladder adenocarcinomas in the repaired bladder. The symphysis is separated by approximately the width of the sacrum; this leads to a rather square appearance of the pelvis (Fig. 20-86). Exstrophy consists of an absence of the anterior wall of the bladder and of the lower anterior abdominal wall. The diagnosis is made on observation; roentgenographic examination is not necessary, but it is useful to study the kidneys and ureters, since ureteral obstruction is often associated. A characteristic dilatation of the distal ureters (“hurley-stick ureter”) is sometimes noted. Wide separation of the pubic bones is also found in some patients with epispadias.
FIG. 20-86. Bladder exstrophy. Note the wide symphysis pubis and characteristic “hurley-stick” dilatation of the distal ureters.
Duplication of the Urinary Bladder
Duplication of the urinary bladder is extremely rare and is usually associated with urethral duplication (Fig. 20-87). Incomplete duplication, in which a septum partially divides the bladder, may also occur, and a multilocular or multiseptated bladder has been described. There may occasionally be a partial horizontal septum, sometimes resulting in an hourglass appearance. The ureters empty into the lower compartment. Cystography is often necessary to outline these anomalies, but EXU with special films of the bladder may obviate the need for cystography.
FIG. 20-87. Duplication of the urinary bladder. Cystogram outlines the duplication with each bladder drained by its own urethra.
Agenesis of the Bladder
Agenesis of the bladder is very rare and is usually incompatible with life, largely because of associated anomalies. Congenital enlargement of the bladder with hydronephrosis

and ureterectasis is found in association with congenital absence or hypoplasia of the abdominal muscles, the prune-belly syndrome. This is a rare condition that occurs almost exclusively in boys. No obstruction can be demonstrated to account for the dilatation. Associated abnormalities include nondescent of the testes, malrotation of the intestine, and, more rarely, persistent urachus, dislocated hips, clubfoot, harelip, spina bifida, hydrocephalus, and cardiac malformations. The abdomen is distended and the skin wrinkled (prune belly).
Bladder Ears
Lateral protrusions of the bladder, caused by extraperitoneal herniations through the internal inguinal ring into the inguinal canal, have been observed. They are usually noted in infants and are associated with a high incidence of clinical inguinal hernia. This is not a true bladder anomaly but is rather a bladder deformity that occurs secondary to a large internal inguinal ring. The term bladder ears has been used in preference to hernia because the deformity does not usually persist beyond infancy. Roentgenographic findings at cystography or urography consist of anterolateral protrusion of the bladder into the inguinal canal, which is usually bilateral. The protrusion is most often observed in the partially filled bladder and tends to disappear when the bladder is completely filled. Oblique or lateral views show the anterior extent of the protrusion. It is important to be aware of this condition so that surgical misadventures are avoided during inguinal herniorrhaphy.
Pear-Shaped Bladder
Some patients have this appearance as a normal variant caused by anterior prolongation of the bladder. However, the teardrop or pear-shaped alteration in the shape of the bladder was first described in patients with pelvic hematoma. Other abnormalities that also may result in an elongated bladder with a narrow base include pelvic lipomatosis, large iliopsoas muscles, enlarged pelvic nodes, lymphoceles, lymphoma and other pelvic tumors, and inferior vena cava occlusion. When the inferior vena cava is occluded, large venous collaterals may compress the bladder, altering its shape to this configuration.
Vesical Calculi
Obstruction and infection are the chief causes of vesical calculi. About half of these calculi are radiopaque and can be easily seen on plain-film roentgenograms (Fig. 20-88). Others contain small amounts of calcium and are poorly visualized on plain films. The condition occurs largely in males.
FIG. 20-88. Vesical calculi. This roentgenogram of the pelvis, obtained without the use of contrast material, outlines five partially calcified bladder stones. Note their midline position.
Cystography with air or with diluted contrast medium can be used to outline radiolucent stones. CT and ultrasonography also readily delineate vesical calculi. Stones may be single or multiple, and they tend to lie in the midline except when contained in a bladder diverticulum. In such cases, the position of the calculus depends on the site of the diverticulum. Bladder calculi must be differentiated from calcification in lymph nodes, fecaliths, calcification in uterine fibroids, and prostatic and seminal vesicle calculi. Radiopaque bladder calculi are often laminated and very dense; when multiple, they may be faceted. Lymph node calcification usually is higher in position, and the nodes are mottled and not as uniformly dense as calculi. Uterine leiomyomas that contain calcium are often higher in position than bladder calculi and have a mottled appearance. Fecaliths of the sigmoid are rare but may simulate bladder calculi closely in texture and position. Oblique views and barium enema permit differentiation. Prostatic calculi are usually multiple and produce a mottled density, in contrast to the uniform or laminated appearance of vesical calculi; they are also lower in position. The same is true of calculi in the seminal vesicles.
Cystography and cystoscopy may be necessary to differentiate bladder calculi from other causes of calcification. A foreign body within the bladder can act as a nidus for deposition of calcium and other salts to form a calculus. Foreign bodies may be introduced by way of the urethra during treatment or by the patient. They also may be introduced through penetrating wounds or left in or near the bladder during surgery. The shape of the calculus depends on the foreign body, which can often be visualized on plain roentgenograms. Foreign bodies in the bladder almost invariably become encrusted with calcium.
Calculi in the prostate usually occur in the form of small granular deposits and are visualized overlying or directly above the level of the symphysis pubis in standard anteroposterior roentgenograms of the lower abdomen. As a rule, they offer little difficulty in differential diagnosis because of their position, characteristic small size, and multiplicity (Fig. 20-89).
FIG. 20-89. Prostatic calculi. The mottled densities above the symphysis pubis are characteristic of prostatic calculi. Their distribution indicates probable prostatic enlargement. The calcifications on the left, which tend to parallel the pubic ramus, are phleboliths (arrows).

Inflammation of the Bladder (Cystitis)
Acute inflammation of the urinary bladder usually does not produce changes that can be recognized and diagnosed on cystography. Chronic cystitis results in decreased bladder size. The wall may be smooth but sometimes is serrated; when serration is present along with major contraction at the dome, the so-called Christmas-tree bladder of chronic cystitis is formed. A bladder of this shape, also called pine-tree bladder, is also found in association with neurogenic bladder dysfunction although it may occur with any chronic bladder obstruction. Cystoscopy is more useful than cystography in the examination of bladder infections.
Of interest roentgenologically is cystitis emphysematosa (emphysematous cystitis), an inflammatory disease of the bladder in which there is gas in the vesical wall or lumen, or both (Fig. 20-90). This condition is caused by gas-forming bacteria, and almost 50% of the reported cases have occurred in patients with diabetes mellitus. The gas may be present for only a short time, which probably accounts for its low reported incidence. Roentgenographic findings are characteristic. A ring of radiolucency outlines the bladder wall or part of it. There is often gas within the bladder lumen as well. The zone of gas expands and contracts with the bladder and is a transient finding unless the infection fails to respond to therapy. The bladder infection is often benign and transient and may produce very few symptoms.
FIG. 20-90. Emphysematous cystitis. Note gas (arrows) in bladder wall in this diabetic patient with pyuria.
Schistosomiasis (bilharziasis) is caused by a group of blood flukes of the genus Schistosoma, S. mansoni, S. japonicum, and S. haematobium. The lower urinary tract is involved mainly by S. haematobium. Large numbers of ova are deposited in the submucosa of the bladder wall, which becomes thickened, ulcerated, and sometimes papillomatous. In chronic disease, the distal ureters may be involved, leading to stricture, hydronephrosis, and renal damage. Calcification in the bladder wall, which occurs in chronic cases, has a characteristic appearance. When the bladder is empty, thin parallel lines of density are observed. The appearance resembles that of the postvoiding bladder on EXU, in which a thin coating of opaque medium outlines the bladder wall. When the bladder is full of contrast medium, a very thin radiolucent line, representing the thickened mucosa, separates the opacified bladder lumen and the thin rim of submucosal calcification (Fig. 20-91). The ova in the bladder wall may calcify. Lower ureteral nodular involvement may produce the roentgenographic findings of ureteritis cystica. Bladder capacity eventually is reduced. The distal ureters may be calcified in a manner similar to that noted in the bladder. Calculi in the bladder, ureters, and kidneys are common.
FIG. 20-91. Schistosomiasis of the bladder. A: Plain films show parallel lines of calcific density in the wall of the empty bladder. B: Air cystogram shows a thin rim of calcium in the bladder wall. C: Renografin cystogram shows the thin rim of calcium separated from the lumen by a lucent line representing the thickened bladder wall.
Candidiasis may involve the urinary bladder in circumstances similar to those in which the kidney is affected. There may be gas within a fungus ball, giving a laminated appearance. Otherwise, the fungus balls appear as filling defects that must be differentiated from blood clots, radiolucent calculi, and tumors.
Cyclophosphamide Cystitis
Cyclophosphamide (Cytoxan), used in the treatment of leukemia and lymphoma, may produce a hemorrhagic cystitis with hematuria of varying severity. Blood clots within the bladder may then appear as filling defects. On cystography, minimal mucosal irregularity is noted in addition to the clots. Later, contraction and thumbprinting secondary to edema

and submucosal hemorrhage are evident. Ultimately, the bladder may be markedly contracted, and, very rarely, calcification may occur in its wall.
Radiation Cystitis
Radiation therapy may produce enough necrosis of the bladder wall to result in calcification that is similar radiographically to that observed in schistosomiasis. Lesser exposure can cause a small bladder with thick walls and lack of distensibility.
Cystitis Cystica
Cystitis cystica is a form of chronic disease of the bladder in which a number of small, cyst-like mucosal lesions are noted, mainly in the region of the trigone. In almost all instances, this condition is associated with infection, obstruction, tumor, calculi, or stasis. Its presence in children usually indicates that an associated chronic infection will be difficult to control. Radiographically, multiple filling defects are observed, chiefly in the region of the trigone. The irregularity and deformity, if severe, may resemble changes observed in bladder tumors. Although the lesions may be visible on cystography, cystoscopy is the best method of examination in patients with this condition.
Cystitis Glandularis
Cystitis glandularis represents metaplasia of the bladder epithelium induced by a variety of irritants. Most lesions occur in the region of the vesical neck and trigone. They appear as irregular, rounded elevations separated by deep ridges and are usually sharply demarcated from the normal mucosa. When they exist in the dome of the bladder, villus-like proliferations may occur, often several centimeters in size. The lesions are considered premalignant. Because their radiographic appearance is similar to that of cystitis cystica, cystoscopy and, often, biopsy are required for definitive diagnosis. The condition may also simulate bladder tumor.40
This rare chronic inflammatory disease is usually confined to the bladder, renal pelvis, and ureters. Malakoplakia may occasionally involve the renal parenchyma on one or both sides, and it causes marked enlargement of the involved kidneys. When renal infection with gram-negative organisms is an associated finding, xanthogranulomatous pyelonephritis and infected polycystic disease should be considered in the differential diagnosis. Malakoplakia is probably caused by an unusual histiocytic response to infection (usually E. coli) and may result in renal failure when bilateral. The soft plaques formed in the bladder may not produce any recognizable roentgenographic changes on cystography.
Obstruction of the Bladder
Bladder obstruction may be caused by congenital or acquired lesions. Benign prostatic hyperplasia is the most frequent cause. Prostatic enlargement is difficult to assess by urography. However, when elevation of the bladder floor is accompanied by a J-shaped or hockey-stick appearance of the distal ureters, prostatic enlargement can be inferred. Prostatic carcinoma, acquired urethral stenosis, urethral valves, and neurogenic dysfunction (cord bladder) are other causes. The first change in the bladder wall resulting from obstruction is hypertrophy of the bladder muscle, which can often be observed as a soft-tissue shadow, several millimeters thick, paralleling the opaque shadow of the inner bladder wall on EXU or cystography. The normal bladder wall does

not ordinarily produce a visible soft-tissue shadow. As the muscle bundles enlarge, they cause irregular interlacing bands known as trabeculae. The intervening outpouchings are called cellules (Fig. 20-92). Trabeculation becomes more prominent as obstruction continues, and the cellules may enlarge until diverticula are formed. There also may be reflux of urine into one or both ureters with development of hydronephrosis. It is more likely that reflux is caused by infection, however.
FIG. 20-92. Trabeculation of the bladder. Note the irregularity of the inferior and lateral bladder wall shown in this cystogram. The patient had a transurethral prostatectomy with opacification of the rounded prostatic bed after the operation.
As obstruction develops, the bladder may become decompensated, increasing in size and containing increasing amounts of residual urine, until it appears as a large lower abdominal mass on physical examination and on roentgenographic study. The plain roentgenogram demonstrates a large soft-tissue mass extending out of the pelvis, often displacing the bowel upward and posteriorly. Cystography outlines the large bladder with trabeculations standing out in a somewhat reticular manner, with cellule or small-diverticulum formation.
Cystourethrography is used to examine patients with suspected bladder or urethral obstruction as well as patients (usually children) with chronic or recurrent urinary tract infections. There are differing opinions regarding the mechanisms of roentgenographic findings in bladder-neck obstruction that is not obviously caused by a mechanical blockage such as prostatic enlargement. Such “functional” obstructing problems have been better understood since the advent of urodynamic methods for investigating the bladder and urethra. Better understanding of bladder physiology and recognition of the infrequent voiding syndromes have cleared up previous ambiguities.
Diverticulum of the Bladder
A diverticulum of the bladder wall is a localized herniation of mucosa, usually having a narrow neck. These protruding defects, which may be single or multiple, vary in size from a small cellule to a large sac having a capacity greater than the bladder itself. Chronic obstruction is a frequent cause, but some diverticula are of congenital origin. Infection is also a factor in many cases. If the diverticula are small and empty completely, they are usually of no clinical significance. Large diverticula that do not empty completely are often the site of infection that is fostered by stagnation. Calculus formation is also common in this type of large diverticulum. Roentgenographic findings are confined to those diverticula that are noted at cystography or EXU unless the diverticulum is large enough to produce an actual mass shadow on a plain roentgenogram. Then the nature of the mass must be determined by means of cystography, cystoscopy, or ultrasonography. When the bladder is examined by means of cystography, the diverticulum is outlined by the opaque substance, and its size, shape, and position, as well as the width of its neck, can be determined (Fig. 20-93). It is often important to assess the presence and amount of urinary retention in a large diverticulum, and a roentgenogram obtained after voiding is usually sufficient for this purpose. If the bladder still contains enough opaque material to obscure the diverticulum partially, a second roentgenogram may be obtained after catheterization of the bladder. A tumor may occasionally occur in a diverticulum. This lesion is often difficult to visualize, but it appears as a filling defect on the otherwise smooth wall of the diverticulum. CT enables more complete assessment of such a tumor.
FIG. 20-93. Multiple bladder diverticula. The large opacified diverticulum on the right has a smooth wall, in contrast to the trabeculation of the bladder wall. There is also a small diverticulum on the left.
Neurogenic Bladder Dysfunction
Disease or injury involving the spinal cord or peripheral nerves supplying the bladder results in changes in bladder

function that may produce either incontinence or retention of urine. The urographic appearance in neurogenic dysfunction is not related to the type of neurologic lesion producing it. Patients with small, spastic, trabeculated bladders often have upper motor neuron lesions, but simple outlet obstruction can cause the same appearance (pine-tree bladder). In theory, patients with lower motor neuron lesions have large, atonic bladders, but some have small, trabeculated bladders. The large, atonic bladder with little or no trabeculation often is found in association with tabes dorsalis, diabetes, or syringomyelia, but it may also be caused by infrequent voiding in patients with no neurologic disease. The following findings may be observed in patients with neurogenic dysfunction: a trabeculated bladder with a circular or pyramidal (pine-tree) pattern; an hourglass bladder; a small, hypertonic, trabeculated bladder; a large, dilated, hypotonic bladder without trabeculation; and variations in the contour of the vesical neck and prostatic urethra in which there may be saccular dilatation, funnel-shaped dilatation or contraction, and spasm of the bladder neck. The diagnosis of exact abnormality is based on urodynamic studies. Cystographic study delineates vesical size, presence or absence of trabeculation, reflux into the ureters, retention or lack of it, vesical-neck dilatation, and the presence of other associated gross anatomic changes.
Vesicoureteral Reflux
Vesicoureteral reflux in children is usually caused by abnormal anatomy of the vesicoureteral junction. Normally, the ureters enter the bladder at a shallow angle and proceed in the bladder submucosa before emptying into the bladder. This arrangement creates a valve mechanism that allows antegrade flow of urine without reflux. If abnormal anatomy is present, usually a shortened submucosal course of the ureter, vesicoureteral reflux is common. Reflux often spontaneously resolves as the child ages because of the lengthening of the submucosal portion of the ureter. In general, the worse the reflux at the time of diagnosis, the less likely it is to resolve spontaneously and therefore the more likely it is to require surgical intervention.83,110
Infection is the most common adult cause of vesicoureteral reflux, which is also found occasionally in patients with lower-urinary-tract obstruction. The obstructive lesions include posterior urethral valves, urethral stricture, and median bar enlargement of the prostate. Neurologic disorders that result in neurogenic bladder dysfunction, congenital anomalies such as ectopic ureter, and other anomalies of the distal ureter and trigone may also produce reflux.
The study used for detection of reflux is the voiding cystourethrogram. If reflux is present, it is manifested by retrograde filling of one or both ureters. The ureters may dilate considerably, and there may be marked hydronephrosis associated with reflux. Children with unexplained recurrent urinary tract infection should have a urologic study including cystourethrography. Because evidence of reflux is sometimes fleeting, fluoroscopic spot-film examination or videotaping is important. Radionuclide cystography now provides a more physiologic method that uses a lower radiation dose to assess reflux.
Renal parenchymal scarring, either local with a blunted calyx or extensive, may be seen by ultrasound or at EXU. About 60% of adult patients with reflux have extensive scarring. Mucosal striations in the pelvis and upper ureter are also observed in patients with reflux, probably as a result of edema and infection.88
Megaureter Megacystis Syndrome
The definition of this syndrome is somewhat controversial. However, the term megaureter megacystis refers to a large, thin-walled, smooth bladder. This is usually accompanied by severe vesicoureteral reflux with ureterectasis and recurrent or persistent urinary tract infection. It is usually discovered in childhood and is more common in girls than in boys. The trigone may appear large because the ureteric orifices are situated more laterally than normal. The intramural portions of the ureters are shortened and widened. The nature of the underlying disorder leading to the megacystis syndrome is unclear, but it may represent the most severe end of the vesicoureteral reflux spectrum. The enlarged bladder may be caused by the child’s resetting his or her voiding urge so as to void less frequently. The enlarged bladder can accommodate the large volume of refluxed urine that drains back down into the bladder after micturition plus the additional urine that the kidneys excrete.52 There may also be a congenital disproportionate increase in size of the vesical base leading to reflux and infection. Cystography demonstrates the large bladder with vesicoureteral reflux on one or both sides.
Vesical Tumors
Benign Lesions
Benign bladder tumors are rare and not as clinically important as malignant epithelial tumors. They include neurofibroma, leiomyoma, fibroma, fibromyoma, myxoma, hemangioma, lymphangioma, paraganglioma (pheochromocytoma), and nephrogenic adenoma. Other heterotopic types include dermoid cyst, rhabdomyoma, and chondroma. There are other conditions that may produce bladder changes, demonstrable by cystography, that resemble tumor. Endometriosis can involve the bladder wall or appear as an extravesicular mass. Hematuria may be present, although cyclic pain, dysuria, and frequency are more often present. Granulomatous disease can also involve the colon or small bowel adjacent to the bladder. Occasionally, localized cystitis glandularis or cystitis cystica may simulate bladder tumor. The radiographic findings are those of a mass extending into or indenting the bladder wall. There are no characteristics to distinguish the histologic type. Multiple and extensive

masses can be seen in neurofibromatosis; most patients also have cutaneous involvement. Leukoplakia is associated with chronic urinary infection and is thought to be precancerous. The diagnosis is best made by cystoscopy, because the lesions usually are not demonstrable by cystography.
Malignant Tumors
Some malignant tumors of the bladder arise in the region of the trigone and tend to obstruct the ureteral or urethral orifices. The so-called benign papilloma, the most common tumor, is often multiple. Because this epithelial tumor is malignant or has malignant potential, the term benign is a misnomer. Many consider it to be a grade I papillary transitional cell carcinoma. Radiographic detection of the papilloma depends on its size; the small tumors are very difficult to visualize with cystography.
Carcinoma of the bladder is usually of the transitional cell type; squamous cell carcinoma and adenocarcinoma are comparatively rare. Cystography demonstrates an irregular filling defect, usually at the base, often resulting in ureteral obstruction (Fig. 20-94). Calcification may occur in the primary tumor and in metastases from it. The size and shape of these tumors vary widely. Double-contrast cystography can be used to study the bladder mucosa in patients with intravesicular tumors, but this is time-consuming and is not warranted if the patient undergoes cystoscopy. CT demonstrates bladder cancer as a soft-tissue density involving the bladder wall. The appearance of the tumor varies depending on its size and whether it is sessile or polypoid (Fig. 20-95). CT cannot reliably differentiate involvement of the mucosa, lamina propria, and superficial or deep muscle; it is advantageous, however, in detecting extravesicular tumor spread. CT also is useful in evaluating extension of tumor to the pelvic sidewalls, lymphadenopathy, and hydronephrosis/ureterectasis caused by bladder cancer.
FIG. 20-94. Transitional cell carcinoma. A: Irregular filling defect represents tumor. Impression at bladder base is caused by enlarged prostate. B: Enhanced computed tomographic scan of the same patient as in A. Note rim of calcification involving tumor (arrows). (Courtesy of James P. Bronson, M.D., Laconia, New Hampshire.)
FIG. 20-95. Transitional cell carcinoma. Unenhanced computed tomogram shows a sessile tumor along the right posterolateral bladder wall. A Foley catheter balloon filled with air is in the center of the bladder.
MRI of the bladder is most useful for staging of known neoplasms (Fig. 20-96). MRI is probably superior to CT in staging of early disease, and both techniques are accurate in pelvic nodal involvement. Both CT and MRI have limitations in staging local bladder tumors, however, and there are problems of understaging tumors with both techniques. One study demonstrated understaging of bladder tumors in 32.5% by MRI, though several other studies have demonstrated better results.26,87 The use of pulse sequences and phased-array coils may increase the accuracy of MRI staging in the near future. Particularly intriguing is the development of rapid dynamic gradient-echo techniques that utilize bolus injections of contrast material.10
FIG. 20-96. A and B: Transitional cell carcinoma. Coronal and axial T1-weighted magnetic resonance images demonstrate invasion of the tumor through the perivesical fat to the pelvic sidewall. (Courtesy of Patrick J. Fultz, M.D., Rochester, New York.)
Rhabdomyosarcoma occasionally arises in the bladder and usually is discovered in the first 3 or 4 years of life or

in late adulthood. The tumor originates from remnants of the urogenital sinus and wolffian ducts; it may originate in the submucosal or superficial layers, usually at the base. The tumor tends to involve the trigone and can become large enough to displace the ureters laterally. This sarcoma can also bulge upward into the bladder to form a lobulated filling defect in it. The tumor nodules may also force their way downward into the urethra, forming a cone of dilatation in the posterior urethra. Some appear as rectal masses; others may protrude through the vulva. Urinary retention caused by bladder obstruction is the most common symptom. EXU often demonstrates ureteral displacement as well as deformity and displacement of the bladder. Ultrasound, CT, and MRI demonstrate the soft-tissue mass. Biopsy is necessary to make the specific diagnosis. Rhabdomyosarcomas comprise about 10% of the malignant tumors of childhood, and there is a slight male predominance. Most rhabdomyosarcomas arise in the bladder, but they may also arise in the prostate, vagina, spermatic cord, or broad ligament. The tumor is sometimes termed sarcoma botryoides because of its polypoid appearance (Fig. 20-97).
FIG. 20-97. Rhabdomyosarcoma of the prostate. Large tumor mass displaces bladder anteriorly.
Metastases to the Bladder
Three general types of metastases to the bladder may be observed: (1) bladder implant secondary to epithelial tumors of the kidney or ureter; (2) direct extension from primary tumors in the area such as prostatic, uterine, ovarian, and colonic neoplasms; and (3) hematogenous metastases from various sources such as breast, lung, or stomach or from melanoma arising at a distant site. Melanoma is the most common tumor that metastasizes to the bladder.
Trauma of the Bladder
Rupture of the bladder may result from a direct blow to a distended bladder as a single injury, or it may occur in association with more extensive injury such as pelvic fracture, penetrating wounds, or gunshot wounds. Instrumentation can also cause rupture of the bladder or urethra. Bladder rupture may be intraperitoneal or extraperitoneal. Intraperitoneal rupture is more common in children, because the bladder is mainly abdominal in location before maturity. Otherwise it is difficult to determine the incidence of intraperitoneal versus extraperitoneal rupture because the state of bladder distention is an uncontrollable variable affecting the type of injury and because in many series all

pelvic fractures are grouped together when the incidence of bladder injury is determined.163
Retrograde cystography is the preferred imaging modality to evaluate for bladder rupture. EXU is suboptimal because of dilution of the contrast agent and because small tears may not be demonstrated with the low resting intravesical pressure. Intraperitoneal rupture results in extravasation of contrast material into the peritoneal cavity, with outlining of the smooth outer walls of the pelvis and the lower abdominal and pelvic viscera. The actual site of rupture may not be visible on the roentgenogram because of overlapping shadows. Extraperitoneal bladder rupture usually is caused by blunt trauma associated with a fractured pelvis and produces a more varied pattern, depending on the site of rupture. The extravasated contrast material outlines the tissue planes of the pelvic floor and extends varying distances into the perivesical soft tissues in an irregular, streaky manner. CT offers the advantage over cystography of summary assessment of all pelvic structures and spaces (Fig. 20-98). Certain patients, such as those involved in motor vehicle accidents, also require evaluation of the upper abdomen, and CT of the abdomen and pelvis can be performed quickly. At times, the differentiation between intraperitoneal and extraperitoneal rupture is difficult. Rarely, spontaneous rupture of the bladder may occur, usually in patients with severe cystitis, extravesical infection, or malignant disease; it also may occur as a result of overdistention secondary to mechanical obstruction or neurogenic dysfunction. When evaluating patients with bladder-drained pancreas transplant, CT cystography (with injection of up to 500 mL of 10% iodinated contrast material and 60 mL of air before scanning) may now be the test of choice.17 Pelvic and lower abdominal trauma may also result in perivesical hematoma without rupture. Cystography or CT then shows displacement of the bladder, which varies with the size and location of the hematoma.
FIG. 20-98. Extraperitoneal bladder rupture. A: Computed tomography shows the site of bladder rupture (arrow) in the right bladder wall. B: A more caudal slice shows the marked extravasation of contrast medium into the surrounding soft tissues.
Foreign Bodies
Foreign bodies in the bladder can be seen on plain-film roentgenograms if they are radiopaque. Oblique and lateral views may be necessary to verify the position of the foreign body in relation to the bladder. Cystography outlines radiolucent foreign bodies and demonstrates associated changes in the bladder wall. Various oblique and lateral projections are usually necessary to ascertain the location. Ultrasonography or CT can be very useful in this situation. Cystoscopy is used for both diagnosis and treatment. A foreign body in the bladder usually has been introduced by the patient; the incidence is higher in adults with mental illness and in children. Occasionally, foreign bodies are introduced at the time of surgery or instrumentation, and they may also result from penetrating wounds. A foreign body can serve as a nidus for the deposition of calcium salts and the formation of a bladder calculus, as indicated previously.
Hernia of the Bladder
Bladder herniation is said to occur in 10% of all inguinal hernias in men older that 50 years of age, but large hernias with descent into the scrotum are unusual.64 Herniation must be differentiated from diverticulum; this is usually accomplished on the basis of the hernia’s location and the direction of its protrusion as well as its relatively wide mouth compared with a diverticulum. Films obtained with the patient in the erect and prone oblique positions are necessary to demonstrate these findings in most patients.
The retrograde urethrogram is the simplest radiographic method for examination of the urethra. It consists of retrograde injection of a radiopaque contrast agent into the urethra, after which films are exposed in various projections as the occasion demands. It is sometimes used in female patients to demonstrate diverticula that may be missed at cystoscopy. In the male, diverticula, strictures, abscess cavities, fistulas, and prostatic abnormalities and enlargement may be delineated. Videotaping may be used in selected cases when fleeting deformities are expected.
Seminal vesiculography is a specialized urologic-radiologic

method of examining the seminal vesicles. The technique and the normal seminal vesiculogram are described by Banner and Hassler.7
Anomalies of the Urethra
Posterior Urethral Valves
Posterior urethral valves produce varying degrees of obstruction leading to infection, vesicoureteral reflux, and hydronephrosis, followed by destruction of the kidneys unless the condition is corrected. Valves are found almost exclusively in males, most often in children. Enuresis is a common symptom. Other symptoms and signs are bladder distention, dribbling, a poor stream, and failure to thrive. In the male newborn, signs of flank mass caused by urinary ascites coupled with respiratory distress should suggest the diagnosis.129 The valves are located in the vicinity of the verumontanum. Voiding cystourethrography is the roentgenographic method used to demonstrate this lesion.
Roentgenographic findings of the most common type of valve consist of a thin membrane arising near the verumontanum and coursing anteriorly, laterally, and inferiorly. This partially obstructs the urethra during voiding. The posterior urethra must be filled in order to distend the valve; otherwise, it will not be visible. A true lateral projection is necessary to identify the position of the valve. The valve stretches in sail-like fashion to obstruct the urethra. The valve itself may not be visible, but the dilatation of the prostatic urethra and a constricting ring at the vesical neck are characteristic. Rarely, anterior urethral valves may be present; obstruction with proximal dilatation similar to that seen with posterior valves may be present. The lucent defect of the anterior valve may be visualized on cystourethrography.
Urethral Diverticula
In males, some urethral diverticula are congenital, but most occur after trauma or infection. The diverticula are visible on urethrography. In females most, if not all, urethral diverticula are acquired; they usually result from retention in periurethral glands. Infection and calculi can complicate urethral diverticula. Voiding urethrography usually outlines these abnormalities in females, in whom retrograde urethrography is a difficult technique.
Urethral Diseases
Calculi are almost always associated with diverticula and infection in females. In males, calculi occur proximal to obstruction, often in the prostatic or bulbous urethra. If the area is included on the plain film, the diagnosis can usually be made radiographically.
Trauma may result in complete or incomplete urethral rupture or in urethral laceration. Anterior urethral injuries (bulbous and cavernous urethra) are associated with direct blows or straddle injuries. Posterior urethral injuries (membranous and prostatic urethra) usually are caused by pelvic fractures or trauma. Urethrography demonstrates the abnormality by showing extravasation of opaque medium. As a rule, when there is complete urethral rupture, no medium injected into the urethra reaches the bladder because of retraction of the ruptured ends of the urethra.
The anterior urethra can be examined despite placement of an indwelling Foley catheter. A small plastic feeding tube is gently inserted through the external meatus, and its tip is placed about halfway up the anterior urethra, next to the Foley catheter. With the glans compressed adequately, contrast material injected gently through this accessory catheter can delineate urethral injuries.
Condyloma Acuminata (Venereal Warts)
Condyloma acuminata occasionally spread into the urethra. Radiographic findings consist of varying numbers of flat, verrucous filling defects. Retrograde urethrography occasionally can be used in examining patients with this condition.
Urethral Strictures
Urethral strictures may be caused by infection or trauma and rarely represent a congenital anomaly in males. The site, severity, length, and associated sinus or fistulous tracts can be outlined on urethrography.
Urethral Tumors
Urethral tumors are more common in females than in males. Urethrography is difficult technically and not very successful in demonstrating tumors in females, but is useful in outlining the irregularity and intraluminal masses found in the male with urethral carcinoma. Rarely, polyps may occur in the prostatic urethra of boys; they are demonstrated as small, rounded or oval filling defects on urethrography.
Vas Deferens Calcification
Calcification of the vas deferens is occasionally present in diabetic men, but also occurs in nondiabetic, elderly, and

hypercalcemic men. It probably represents a degenerative phenomenon in these patients. Roentgenographic findings are the presence of densely calcified, often bilaterally symmetrical, tubular shadows about 3 mm in diameter in the low midpelvis (Fig. 20-99), which have a distinctive appearance.
FIG. 20-99. Calcification in the vas deferens in a 45-year-old man with a history of diabetes of more than 20 years’ duration.
The right adrenal gland is cephalad and slightly medial to the upper pole of the right kidney. The right adrenal is also just posterior to the inferior vena cava, which serves as a good landmark on CT. The left adrenal gland is also cephalad to the upper pole of the left kidney, but it is located more medially relative to the aorta than the right adrenal is (Fig. 20-100). The right adrenal has an inverted V shape, with its two limbs roughly paralleling the diaphragmatic crus. The left adrenal is usually wider and shorter and can have a variety of shapes, but it may have a “seagull” shape. The adrenals are small, the combined average weight being 11 to 12 g.
FIG. 20-100. Normal adrenal glands (arrows). Computed tomogram of right adrenal posterior to the inferior vena cava and left adrenal just lateral to left diaphragmatic crus. The shape of the left adrenal can vary; this left adrenal has the “seagull” configuration.
The most common cause of calcification in the adrenal is believed to be hemorrhage, often associated with hypoxia or birth trauma, severe maternal infection, hypoprothrombinemia, or increased vascular fragility. Calcification is often found in infants of diabetic mothers. There is a high incidence of abnormal obstetric history, including prematurity, use of forceps, and breech deliveries, in children with adrenal calcification. Neonatal adrenal hemorrhage, which may be massive, unilateral, or bilateral, is seen as a radiolucent suprarenal mass in the total-body phase of EXU in neonates. Adrenal hemorrhage can have a variable appearance on ultrasonography, depending on when the hematoma is scanned. Acutely, the hematoma can appear echogenic, but as it liquefies it usually assumes a more anechoic or hypoechoic appearance.141 Calcification occurs rapidly around the periphery a few weeks after the hemorrhage, then contracts slowly to the original size and shape of the gland. If the hemorrhage is unilateral, the right adrenal is more frequently involved. Stippled adrenal calcifications may be found in adults without any signs or symptoms of adrenal insufficiency. The cause in these cases remains obscure. CT or ultrasonography can document calcifications within the adrenal. Tuberculosis of the adrenal gland is now a rare cause of adrenal insufficiency. About one third of tuberculosis patients develop calcification within the adrenal. The calcification may appear as amorphous granular density within the gland, or it may entirely outline the gland.
Adrenal cysts also can contain calcium. Peripheral calcification is present in about 15% of cases. Peripheral calcification is suggestive of a cyst, whereas calcification within an adrenal mass is more suggestive of tumor. Adrenal cysts are rare, and they occur equally on the right and left sides. Their incidence is 50% higher in women than in men. Lymphangiectatic cysts and pseudocysts are the most common types, although parasitic cysts, epithelial cysts, and those occurring secondary to hemorrhage or necrosis are also found. These cysts are usually asymptomatic. Cyst puncture is advocated by some physicians in the diagnostic evaluation of adrenal masses, and the procedure can be done under CT or ultrasound guidance. Cysts that do not contain calcium may simulate a tumor on plain film, but they are not visible on plain film unless they become large (Fig. 20-101). CT, ultrasonography, and MRI are all useful in the imaging of adrenal cysts.50
FIG. 20-101. Cyst of the left adrenal gland. Note the large mass above and partly overlying the upper pole of the left kidney. The kidney is displaced downward but is not significantly deformed or distorted.

Adrenal Cortical Tumors
Adrenal tumors are divided according to their origin into cortical and medullary types. The cortical lesions are glandular in type and mesodermal in origin. Benign adenoma and carcinoma are the two types that are found, and they can be either functioning or nonfunctioning. When these tumors produce a disturbance of function, they can affect both cortical and medullary function. Hyperplasia of the adrenal cortex can also disturb cortical and medullary function. The symptoms are varied and may be caused by an excess of androgens, estrogens, adrenocorticotropic hormone, or other hormones. Sex changes and Cushing’s syndrome may be present. About 75% of patients with Cushing’s syndrome have hyperplastic adrenals, and 25% have an adenoma or carcinoma. The hyperplastic gland is enlarged bilaterally and retains its normal shape (Fig. 20-102), whereas tumors tend to be round or oval and to produce an alteration in the contour of the adrenal gland (Fig. 20-103). EXU was used in the past to localize suspected adrenal tumors. CT scanning and ultrasonography are now the primary diagnostic methods. MRI is usually reserved for characterization of the known adrenal mass. Ultrasonography may be somewhat limited in the evaluation of patients with Cushing’s syndrome because of the increased adipose tissue. More invasive procedures such as arteriography, venography, or venous sampling may be necessary if CT or ultrasonography cannot demonstrate the adrenal mass. The angiographic signs are similar to those of tumor elsewhere and consist of dilated and displaced vessels, tortuous vascular patterns, and arteriovenous shunts.
FIG. 20-102. Adrenal hyperplasia. Left adrenal gland (arrow) is enlarged but maintains its normal shape. This appearance is in contrast to that of adrenal tumors, which tend to produce round masses.
FIG. 20-103. Cushing’s syndrome. Enhanced computed tomogram of a left adrenal adenoma (arrow) medial to the spleen.
Primary aldosteronism or Conn’s syndrome consists of hypertension, hypokalemia, hyperkaliuria, and low plasma renin. The renin production is suppressed by the excess aldosterone. Seventy percent to 80% of patients with Conn’s syndrome have a unilateral adenoma, which is often less than 2 cm in diameter.195 Nodular bilateral cortical hyperplasia accounts for most of the remaining patients. Because of the risks of angiographic procedures, CT is preferable for the initial imaging. CT can demonstrate the small adenoma causing the hyperaldosteronism. Venography and radionuclide techniques can also be used to detect these small adenomas. Scintiscanning with 131I-labeled 19-iodocholesterol has been used to demonstrate increased radioactivity in the abnormal adrenal gland and to differentiate hyperplasia from cortical adenoma and adenocarcinoma.
Nonfunctioning cortical adenomas and carcinomas also appear as solid masses on CT. Adenomas are usually small (less than 3 cm in diameter) and are unilateral. Because of high lipid content, adenomas often have density measurements that approach that of water (Fig. 20-104).96 This high lipid content can help distinguish adenomas from adrenal malignancies, which lack this material.109 In particular, unenhanced CT97 or CT obtained approximately 1 hour after contrast injection98 appears useful to help differentiate adenomas from malignancies. An unenhanced CT attenuation value of less than 18 Hounsfield units and a 1-hour postcontrast value of less than 30 Hounsfield units are strong predictors of adenomas. Carcinomas are usually larger than adenomas

and can be bilateral. Most nonfunctioning tumors large enough to produce symptoms are malignant (Fig. 20-105).50 Myelolipoma is a rare benign tumor of the adrenal gland. It is composed of fat and erythroid and myeloid elements. If fat is found in an adrenal mass, the diagnosis of myelolipoma can be made with confidence (Fig. 20-106).
FIG. 20-104. Nonhyperfunctioning adrenal adenoma. Incidentally found low-density adrenal adenoma (arrow). (Courtesy of Luke E. Sewall, M.D., Madison, Wisconsin.)
FIG. 20-105. Adrenal cortical carcinoma. Large right adrenal mass typical for adrenal cortical carcinoma is seen. This invades both the liver and inferior vena cava.
FIG. 20-106. Adrenal myelolipoma. Large right adrenal tumor is composed primarily of fat. (Courtesy of Donald R. Yandow, M.D., Madison, Wisconsin.)
MRI can detect adrenal abnormalities with approximately the same sensitivity as CT. Currently, the most widespread use of MRI in adrenal imaging is to separate patients with adrenal metastases from those with nonhyperfunctioning adenomas (Fig. 20-107). Because metastases to the adrenals and nonhyperfunctioning adenomas are both common, this can represent a significant clinical problem.
FIG. 20-107. Adrenal adenoma. Magnetic resonance imaging performed using an in-phase gradient echo sequence (left) and an opposed-phase sequence (right) demonstrates marked signal loss in an adrenal mass (arrows) with the opposed-phase sequence. This finding is typical for an adenoma containing lipid.
Recent work has shown the potential of several imaging sequences in differentiating adenomas from nonadenomas. In particular, chemical shift imaging using in-phase and opposed-phased pulse sequences has been used successfully for this purpose.117,126 These sequences also take advantage of lipids found in adenomas. Coexistence of both lipid and water components in a single volume causes a cancelling of signal at certain echo times (opposed phase) (Figs. 10-107 and 20-108). Therefore, decreased signal in an adrenal mass on opposed-phase images compared with standard in-phase images is strong evidence of the presence of an adenoma.
FIG. 20-108. Adrenal metastasis. In contrast to adrenal adenomas, there is no signal loss in this adrenal mass seen on computed tomogram (A) and on in-phase (B) and opposed-phase (C) magnetic resonance images. The increased signal intensity in the adrenal mass (arrows) on opposed-phase images makes the diagnosis of malignancy much more likely. (Courtesy of Todd Kennell, M.D., Madison, Wisconsin.)
Adrenal Medullary Tumors
Adrenal medullary tumors are ectodermal in origin. They include ganglioneuroma, ganglioneuroblastoma, neuroblastoma, and pheochromocytoma. Pheochromocytoma clinically is characterized by hypertension (secretion of epinephrine and norepinephrine), which is paroxysmal in nature, with flushing, sweating, tachycardia, and anxiety. The diagnosis usually is made by measuring elevated levels of urine catecholamines. About 10% of pheochromocytomas are bilateral, 10% are extra-adrenal, and 10% are malignant. Pheochromocytomas are usually solitary and are located more commonly in the right adrenal than in the left. CT is very accurate in locating pheochromocytomas. In one review study, CT demonstrated the tumor in 52 patients who initially presented and in 8 patients with evidence of recurrence.194 These tumors are usually greater than 2 cm in size and may be solid or cystic. When the adrenal glands do not show evidence of tumor by CT, then the remainder of the abdomen and pelvis should be scanned. The retroperitoneum in the periaortic region and the organ of Zuckerkandl (located near the aortic bifurcation) are the most common locations of pheochromocytomas outside of the adrenal. The neck and chest are other possible sites. Pheochromocytomas are associated with the multiple endocrine adenomatoses (both type I and type II).
MRI is also an effective way to localize suspected pheochromocytomas. CT should be used as the first method of

imaging, but MRI can be valuable when the tumor is in an extra-adrenal location and in the patient who has had previous retroperitoneal surgery.50,56
Pheochromocytoma is usually a very vascular tumor with arteriovenous lakes and early venous filling. Because of the danger of a precipitous rise in blood pressure in patients with suspected pheochromocytoma, the blood pressure and electrocardiogram are continuously monitored. Intravenous injection of iodinated contrast materials can precipitate a hypertensive crisis and therefore should be avoided if possible. Phentolamine should be available for immediate intravenous injection if blood pressure rises suddenly. An injection of 5 mg phentolamine usually is sufficient to control the blood pressure, but repeated injections may be necessary in some patients. Angiography usually demonstrates a hypervascular mass with an intense capillary stain (Fig. 20-109). Less frequently, a hypovascular mass can be seen. Adrenal venography can also be used if arteriography is not successful. Ganglioneuroma, which is benign, usually causes no symptoms; the tumor is probably the differentiated form of neuroblastoma. Large ganglioneuromas may be visible on scout radiographs. If not, CT, MRI, venography, or angiography can be used for diagnosis.
FIG. 20-109. Pheochromocytoma. A: Angiogram demonstrates hypervascularity typical of pheochromocytomas. B: T1-weighted coronal magnetic resonance image. Pheochromocytoma (asterisk) is seen in the left adrenal gland of a second patient. C: T2-weighted axial image shows high signal typical of pheochromocytoma (asterisk).
Neuroblastoma, the most common malignancy in infants and children, is a tumor of adrenal medullary origin. About 30% of neuroblastomas are diagnosed in children younger than 1 year of age, another 15% to 20% in children younger than 2 years, 25% in children between the ages of 2 and 5 years, and about 90% by 8 years of age. The tumor may arise in cells of the sympathetic nervous system as well as in the adrenal medulla; therefore, this type of tumor may be found either below or above the kidney. Neuroblastoma is highly malignant and can attain great size before discovery. Two thirds of children with neuroblastoma have distant metastases (usually osseous) at presentation. Calcification in the mass is common (about 40% to 50% on plain films and 85% on CT) (Fig. 20-110), whereas in Wilms’ tumor, from which neuroblastoma must be differentiated, calcium is rare. The calcification has a fine granular or stippled appearance. Urography serves to indicate that the mass is extrarenal, with the kidney typically being displaced laterally. The tumor may fill most of the abdomen. CT is more sensitive for detecting calcification, is superior to urography in defining

extent of disease, can evaluate possible involvement of the inferior vena cava, and can be used to evaluate recurrent disease. Ultrasonography is an alternative imaging modality that can provide similar information in the evaluation of neuroblastoma (Fig. 20-111).
FIG. 20-110. Neuroblastoma arising in the left adrenal gland. Note the mottled, somewhat granular calcification in the left upper abdomen, which is typical of the calcification observed in neuroblastoma.
FIG. 20-111. Neuroblastoma. Transverse ultrasound demonstrates the lateral displacement of the left kidney (K) by the solid retromantle of tumor (arrows). S, spine.
Metastases to the liver and lungs as well as bone are common. The osseous metastases are often very extensive and characteristic, being of mixed lytic-blastic character. Often there is extensive involvement of the calvarium, with separation of the sutures. Because the tumor may revert to benign ganglioneuroma, spontaneously or during treatment, localization and treatment of the primary tumor are important.
Other Adrenal Tumors
Other adrenal tumors are rare. They arise in the adrenal stroma and include neuromas, fibromas, lipomas, neurofibromas, hemangiomas, myomas, sarcomas, lymphangiomas, myelolipomas, osteomas, and melanomas. Tumors metastatic from other areas are common in the adrenal glands and should be considered in the differential diagnosis of every adrenal mass. Metastatic lung carcinoma is the most common; melanoma and breast, colon, and thyroid cancer can also metastasize to the adrenals. Metastases are often bilateral.50 CT is the preferred imaging modality for evaluation of adrenal metastases. The CT density is variable and can be solid or have low attenuation values (Fig. 20-112). In patients with adrenal masses in whom proof of metastatic disease will alter therapy, CT- or ultrasound-guided aspiration biopsy can be performed. Melanoma metastatic to the adrenals has been reported to cause curvilinear calcifications simulating those noted in benign adrenal cyst.193
FIG. 20-112. A: Enhanced computed tomogram of bilateral adrenal metastases in a patient with bronchogenic cancer. Upper poles of kidneys (arrows) are separated from adrenal glands by a fat plane. B: Coronal T1-weighted magnetic resonance image of a patient with non-Hodgkin’s lymphoma shows bilateral large adrenal masses (asterisk).
Miscellaneous Adrenal Conditions
Adrenal abscess is very rare, but it is a possible cause of adrenal enlargement in neonates. It must be differentiated from other adrenal masses such as hematoma and neuroblastoma. Total-body opacification was used in the past, but CT or ultrasonography is now preferable in evaluating abscess. It may be difficult to distinguish abscess from hematoma and from infected hematoma.


Adrenal milk of calcium is a rare occurrence in which a suspension of calcification, resembling the milk of calcium observed in the gallbladder, accumulates in an adrenal cyst.131
Recent innovations in ultrasound and MRI technology have led to increased interest in imaging of the prostate gland. Clinical interest has remained high because of the large numbers of deaths from carcinoma of the prostate, estimated at 39,200 in the United States in 1998.103A Several factors make diagnosis, screening, and treatment of this disease difficult.
Carcinoma of the prostate mostly affects older men. One autopsy series showed a prevalence of 30% in patients who died of unrelated causes. The question then arises, which patients have clinically significant tumors that will cause morbidity and mortality, and in which patients is the disease largely incidental? This question has not been completely resolved, although several authors believe that tumor volume predicts tumor behavior. Tumors 3 cc or larger are more likely to have extracapsular spread and aggressive histologic types. Therefore, even though tumors may be multicentric in a single patient, the largest tumor, or index cancer, will most likely have the most aggressive histologic changes, and will most closely predict tumor behavior.
A second confounding factor is the complexity of internal anatomy and the changes the aging prostate normally undergoes. The prostate can be divided into an inner gland (transition zone) and an outer gland (central and peripheral zone). The transition zone, which lies in a periurethral location, is the site of benign prostatic hyperplasia, which can occlude

the urethra when severe. The peripheral and central zones lie posterior and lateral to the transition zone. The peripheral zone is the primary tumor site in up to 70% of patients, whereas the transition zone is the primary site in approximately 15%. Tumor is more likely to escape the gland into the seminal vesicles, through the capsule, or into the neurovascular bundles when located in the peripheral zone; therefore, this area must be closely scrutinized when the prostate is imaged (Fig. 20-113).
FIG. 20-113. Prostate cancer. Enhanced computed tomographic scan demonstrates large prostate cancer with invasion of the bladder (arrowheads) and rectum (arrows).
Transrectal ultrasound of the prostate (Fig. 20-114) was introduced in the 1970s in Japan by Watanabe103A and others. Until 1985, it was commonly believed that cancer arose centrally in the gland and was hyperechoic or of mixed echogenicity. During the late 1980s, whole-mount pathologic studies were performed, and it became clear that cancer was primarily hypoechoic (Fig. 20-115). In addition, most tumors originated in the peripheral zone, a finding in accord with the pathologic literature. As tumors enlarged, the histology is more likely to become infiltrative, with tumor “fingers” extending into surrounding tissue. Because of this, echogenicity varies with increasing tumor size.
FIG. 20-114. Normal axial transrectal ultrasound study of the prostate. Transition zone (asterisks) is prominent in this patient with benign prostatic hypertrophy. Peripheral zone is external to surgical capsule (arrows).
FIG. 20-115. Prostate cancer. Axial transrectal ultrasound image shows large hypoechoic tumor mass (asterisk) extending into periprostatic tissues. A second small lesion (arrowheads) is present on contralateral side of the gland. Note peripheral location and deformity of prostatic capsule (open arrows).
Transrectal ultrasound–guided biopsy became widely available during the late 1980s. This technique is a useful tool in the diagnosis and staging of prostate cancer. Biopsy can now be performed simply and easily in an outpatient setting, and strategically guided biopsies can occasionally prove extraprostatic spread. The positive predictive value for cancer varies when biopsy of a hypoechoic nodule is

performed, with one report yielding 46%.106 This number is probably lower in an unselected population. The prostate-specific antigen (PSA) level may help select patients for biopsy.106
The role of MRI in prostate cancer is unclear. Imaging of the internal architecture of the gland with body coil technology has been largely unsuccessful, and attempts to predict local spread have had mixed results. Conventional MRI techniques therefore are limited largely to detection of gross extraprostatic spread and enlarged pelvic lymph nodes. Two technologic advances may prove fruitful in the diagnosis and staging of local disease: endorectal surface coils and phased-array coils (Fig. 20-116 and 20-117). Early work has been encouraging, but more study is needed. CT is largely limited to detection of nodal enlargement in the pelvis and metastatic disease elsewhere. CT does not play a significant role in the evaluation of intraprostatic or early extraprostatic disease (see Fig. 20-113).13,106,107,139,159
FIG. 20-116. Normal prostate gland in a 68-year-old man. Endorectal surface coil magnetic resonance T1-weighted image. TR, 600 ms; TE, 15 ms. The neurovascular bundles can be seen on each side (curved arrows). (Courtesy of Howard M. Pollack, M.D., Philadelphia, Pennsylvania.)
FIG. 20-117. Carcinoma of the prostate, stage C. Endorectal surface coil magnetic resonance T2-weighted image. TR, 3,000 ms; TE, 90 ms. The tumor involves the entire peripheral zone (asterisk), which normally should be of high signal intensity. On the right side, at the 7-o’clock position, the tumor has extended beyond the capsule to involve the neurovascular bundle (curved arrow). (Courtesy of Howard M. Pollock, M.D., Philadelphia, Pennsylvania.)
The issue of screening for prostate cancer remains controversial. The roles of PSA, ultrasound, and MRI continue to evolve for this indication. Studies are currently underway to evaluate the best strategy for screening and to decide whether large-scale efforts are warranted.
Scrotal contents include the testis, epididymis, and spermatic cord. The testicle is covered by a fibrous sheath called the tunica albuginea, and this in turn is surrounded by the tunica vaginalis. The epididymal head is superior to the testis, and the body and tail are located posteriorly and inferiorly. The blood supply to the testicle derives from the spermatic cord, which contains the testicular artery.
The testicle is imaged using both ultrasound and nuclear medicine techniques.13,111,115,162 Ultrasound should be performed with a high-resolution, linear-array transducer. Doppler technology is also useful, particularly to assess for torsion or hyperemia associated with injury or infection. Nuclear medicine imaging requires the injection of an adult dosage of approximately 15 to 20 mCi of 99mTc-pertechnetate intravenously. Flow and static images are obtained with the penis taped up to the pelvis.
Testicular (Spermatic Cord) Torsion
Testicular torsion usually is discovered because of testicular pain in the newborn or adolescent male. The process is caused by a twisting of the spermatic cord, which compromises venous and then arterial flow. Diagnosis must be made promptly, usually within 6 hours, to ensure testicular viability. Nuclear medicine imaging with 99mTc-pertechnetate has been the traditional method for diagnosis of testicular torsion. Acutely, decreased flow is seen in the expected position of the testis. If the condition remains untreated for more than 24 hours, a rim of increased activity with a photopenic central area may develop, the so-called halo sign. This sign of delayed torsion can also be seen with testicular abscess, hematoma, or tumor.

Doppler ultrasound has become widely accepted as a sole modality to diagnose torsion. Acquisition of an arterial signal from the central portion of the testicle effectively excludes complete torsion, although partial torsion is still possible. The diagnosis of partial torsion should be considered when a dampened arterial signal, compared with that from the asymptomic contralateral testicle, is present in a symptomatic patient. Before making this diagnosis, it is important to confirm that normal flow is seen in the contralateral testis. This verifies that the Doppler instrument has sufficient sensitivity to document slow flow for small vessels and that imaging parameters are set correctly. Ultrasound has the advantage over nuclear medicine of being able to image the remainder of the scrotum. Often, if testicular torsion is excluded, ultrasound can suggest an alternative diagnosis.
Infection of the epididymis is often associated with infection of other portions of the genitourinary tract. The main importance of this condition is differentiating it from testicular torsion, which is a surgical emergency.
Nuclear medicine imaging in epididymitis reveals increased blood flow to the affected side. No photopenic areas are encountered. Ultrasound shows an enlarged, heterogeneous-appearing epididymis with increased flow by Doppler (Fig. 20-118). If orchitis is also present, the testicle is usually swollen, hypoechoic, and hyperemic.
FIG. 20-118. Epididymitis. Transverse ultrasound image demonstrates an enlarged epididymal head (asterisk). Color Doppler examination demonstrated increased flow to the epididymis.
Testicular Tumors
The main job of the radiologist in evaluating a painless scrotal mass is to decide whether it arises from within the testicle or from elsewhere. Testicular abnormalities must be viewed with a great deal of suspicion and should be considered tumors until proved otherwise. Testicular tumors are a common neoplasm in men aged 25 to 35 years, with 7,600 new cases expected in the United States in 1998.103A Malignant testicular tumors may be seminomas, choriocarcinomas, embryonal carcinomas, or teratomas/teratocarcinomas (Fig. 20-119). Other tumors that can affect the testicles include lymphoma, leukemia, and metastases.62 In general, tumors are hypoechoic to normal testicle, and seminoma is relatively homogeneous in appearance. Predicting the tumor cell type by ultrasound findings has not proved accurate, however. If a testicular tumor is found, the retroperitoneum should be examined, usually with CT, for nodal spread. Involved lymph nodes are initially found at the sites of lymphatic drainage. Tumors from the right testis spread to nodes in the preaortic, precaval, and aortocaval areas. Tumors on the left drain to left para-aortic nodes. Trauma to the scrotum, in particular testicular hematoma, can mimic testicular tumor. It is important to monitor cases of suspected hematoma, which should undergo evolution and complete regression. Tumors stay stable or enlarge over time.
FIG. 20-119. Nonseminomatous testicular tumor. Transverse ultrasound image demonstrates heterogeneous, centrally placed testicular mass (arrows). Comparison with uninvolved testis (asterisk) shows size difference in testicles. (Courtesy of Kathleen Scanlan, M.D., Madison, Wisconsin.)
Hydroceles represent fluid collections between layers of the tunica vaginalis. They are often idiopathic but can be seen with almost any scrotal or testicular pathology, including trauma, infection, or tumor.
Varicoceles are abnormal collections of dilated veins of the spermatic cord. They are caused by an abnormal venous valvular mechanism or obstruction to blood return from the testis. The junction of the right spermatic vein and the inferior vena cava forms a valve-like mechanism owing to the acute angle of incidence. The left spermatic vein joins the left renal vein at a more obtuse angle and therefore is subject to increased retrograde venous pressure; for this reason, 90% of idiopathic varicoceles arise on the left. Varicoceles may also be found in association with venous obstruction secondary to tumor mass or other retroperitoneal process. If a patient presents with bilateral or unilateral right varicocele, a search of the retroperitoneum for a causative factor is strongly recommended. Varicoceles appear as dilated vascular structures near the upper pole of the testis. The Valsalva maneuver or scanning with the patient in the upright position causes an increase in size. Varicoceles may be a cause of male infertility because of increased scrotal temperature; they are amenable to surgical repair, or they can be thrombosed by percutaneous transcatheter placement of coils in the spermatic vein. In this approach, the spermatic vein is accessed by passage of a catheter through the left renal vein.

Spermatoceles, or dilatations of efferent ductules, usually are found in the head of the epididymis. They are generally asymptomatic and should be differentiated from varicoceles.
1. Albertson KW, Talner LW: Valves of the ureter. Radiology 103:91, 1972
2. Amis ES Jr: Retroperitoneal fibrosis. AJR Am J Roentgenol 157:321, 1991
3. Andersson I: Unilateral renal artery stenosis: II. Angiographic assessment of renal artery pathology. AJR Am J Roentgenol 141:1299-1303, 1983
4. Ansell G, Tweedie MCH, West CR: The current status of reactions to intravenous contrast media. Invest Radiol 15:S32, 1980
5. Aronson S, Frazier HA, Balwah JD, Hartman DS, Christenson PPJ: Cystic renal masses: Usefulness of the Bosniak classification. Urol Radiol 13:89–93, 1988
6. Arrive L, Hricak H, Tavares NJ, et al: Malignant vs. non-malignant retroperitoneal fibrosis: Differentiation with MR imaging. Radiology 172:139, 1989
7. Banner MP, Hassler R: The normal seminal vesiculogram. Radiology 128:339, 1978
8. Banner MP, Pollack HM: Fluoroscopically guided percutaneous extraction of upper urinary tract calculi. Radiol Clin North Am 22:415, 1984
9. Banner MP, Pollack HM, Chatten J, et al: Multilocular renal cysts: Radiologic-pathologic correction. AJR Am J Roentgenol 136:239, 1981
10. Barentz JO, Jager GJ, vanVierzen PBJ, et al: Staging urinary bladder cancer after transurethral biopsy: Value of fast dynamic contrast-enhanced MR imaging. Radiology 201:185–193, 1996
11. Baron RL, McClennan BL, Lee JKT, et al: Computed tomography of transitional-cell carcinoma of the renal pelvis and ureter. Radiology 144:125, 1982
12. Batson PG, Keats TE: The roentgenographic determination of normal adult kidney size as related to vertebral heights. AJR Am J Roentgenol 116:737, 1972
13. Benson CB, Doubilet PM, Richie JP: Sonography of the male genital tract. AJR Am J Roentgenol 153:705, 1989
14. Berland LL, Koslin DB, Routh WD, Keller FS: Renal artery stenosis: Prospective evaluation of diagnosis with color duplex US compared with angiography. Radiology 174:421–423, 1990
15. Bernstein J: The morphogenesis of renal parenchymal maldevelopment (renal dysplasia). Pediatr Clin North Am 18:395–407, 1971
16. Binder R, Korobkin M, Clark RE, et al: Aberrant papillae and other filling defects of the renal pelvis. AJR Am J Roentgenol 114:746, 1972
17. Bischof TP, Thoeni RF, Melzer JS: Diagnosis of duodenal leaks from kidney-pancreas transplants in patients with duodenovesical anastomoses: Value of CT cystography. AJR Am J Roentgenol 165:349–354, 1995
18. Blyth H, Ockendon BG: Polycystic disease of kidneys and liver presenting in childhood. J Med Genet 8:257, 1971
19. Bookstein JJ: Appraisal of arteriography in estimating the hemodynamic significance of renal artery stenosis. Invest Radiol 1:281–294, 1966
20. Bosniak MA: The current radiological approach to renal cysts. Radiology 158:1–10, 1986
21. Bosniak MA: Angiomyolipoma (hamartoma) of the kidney: A preoperative diagnosis is possible in virtually every case. Urol Radiol 3:135, 1981
22. Breatnach E, Stanley RJ, Carpenter JT Jr: Intrarenal chloroma causing obstructive nephropathy: CT characteristics. J Comput Assist Tomogr 9:822, 1985
23. Bretan PN Jr, McAninch JW, Federle MP, Jeffrey R Jr: Computerized tomographic staging of renal trauma: 85 Consecutive cases. J Urol 136:561–565, 1986
24. Brown T, Mandell J, Lefowitz RL: Neonatal hydronephrosis in the era of sonography. AJR Am J Roentgenol 148:959–963, 1987
25. Buonocore E, Meaney TF, Borkowsky GP, Pvalicek MS, Gallagher J: Digital subtraction angiography of the aorta and renal arteries. Radiology 139:281–286, 1981
26. Buy JB, Moss AA, Guinet C, et al: MR staging of bladder carcinoma: Correlation with pathologic findings. Radiology 169:695, 1988
27. Canzanello VJ, Millan VG, Spiegel JE, et al: Percutaneous transluminal renal angioplasty in the management of atherosclerotic renovascular hypertension: Result in 100 patients. Hypertension 13:163–172, 1989
28. Carroll PR, McAninch JW: Staging of renal trauma. Urol Clin North Am 16:193–201, 1989
29. Chen CC, Hoffer PB, Vahjen G, et al: Patients at high risk for renal artery stenosis: A simple method of renal scintigraphic analysis with Tc-99m-DTPA and captopril. Radiology 176:365, 1990
30. Chiarini C, Espositi ED, Losinno F, et al: Renal scintigraphy versus renal vein renin activity for identifying and treating renovascular hypertension. Nephron 32:8–13, 1982
31. Cho KJ, Thornbury JR: Severe reactions to contrast material by three consecutive routes: Intravenous, subcutaneous, and intra-arterial. AJR Am J Roentgenol 131:509, 1978

32. Clark RE, Minagi H, Palubinskas AJ: Renal candidiasis. Radiology 101:567, 1971
33. Cochran ST, Waisman J, Barbaris ZL: Radiographic and microscopic findings in multiple ureteral diverticula. Radiology 137:631–636, 1980
34. Cohan RH, Sherman LS, Korobkin M, Bass JC, Francis IR: Renal masses: Assessment of corticomedullary-phase and nephrographic phase CT scans. Radiology 196:445–451, 1995
35. Committee on Drugs and Contrast Media: Manual on iodinated contrast media. Reston, VA, American College of Radiology, 1991.
36. Courey WR, Pfister RC: The radiographic findings in renal tubular acidosis. Radiology 105:497, 1972
37. Cremin BJ: Wilms’ tumors: Ultrasound and changing concepts. Clin Radiol 38:465–474, 1987
38. Culp OS: Ureteral diverticulum: Classification of the literature and report of an authentic case. J Urol 58:309, 1947
39. Dacher JN, Pfister C, Monroe M, Eurin D, Dasseur PL: Power Doppler sonographic pattern of acute pyelonephritis in children: Comparison with CT. AJR Am J Roentgenol 166:1451–1455, 1996
40. Dann RH, Arger PH, Enterline HT: Benign proliferation processes presenting as mass lesions in the urinary bladder. AJR Am J Roentgenol 116:822, 1972
41. Daughtridge TG: Segmental, multicystic renal dysplasia. J Can Assoc Radiol 26:149, 1975
42. Davidson AJ, Talner LB: Urographic and angiographic abnormalities of adult-onset bacterial nephritis. Radiology 106:249, 1973
43. Davidson AJ, Hayes WS, Hartman DS, McCarthy WF, Davis CJ Jr: Renal oncocytoma and carcinoma: Failure of differentiation with CT. Radiology 186:693–696, 1993
44. Delin NA, Ekestrom S, Hoglung NO: Arteriographic appearance of renal artery stenosis compared to resistance measured at operation: Effect of artery reconstruction on flow, pressure gradient, and resistance. Acta Chir Scand Suppl 356B:150–162, 1966
45. Deyoe LA, Cronan JJ, Breslaw BH, Ridlen MS: New techniques of ultrasound and color Doppler in the prospective evaluation of acute renal obstruction: Do they replace the intravenous urogram? Abdom Imaging 20:58–63, 1995
46. Diament MJ, Kangarloo H: Dosage schedule for pediatric urography based on body surface area. AJR Am J Roentgenol 140:815, 1983
47. Don S, Kopecky KK, Filo RS, et al: Duplex Doppler ultrasound of renal allografts: Causes of elevated resistive index. Radiology 171:709, 1989
48. Dubovsky EV, Russell CD: Radionuclide evaluation of renal transplants. Semin Nucl Med 18:181–198, 1988
49. Dunnick NR, Korobkin M: Computed tomography of the kidney. Radiol Clin North Am 22:297, 1984
50. Dunnick NR: Adrenal imaging: Current status. AJR Am J Roentgenol 154:927, 1990
51. Dure-Smith P: Pregnancy dilatation of the urinary tract. Radiology 96:545, 1970
52. Elkin M: Radiology of the Urinary System, 1st ed. Boston, Little, Brown, 1980
53. Elkin M: Renal cystic disease: An overview. Semin Roentgenol 10:99, 1975
54. Ellis JH, Cohan RH, Sonnad SS, et al: Selective use of radiographic low-osmolality contrast media in the 1990s. Radiology 200:297–311, 1996
55. Federle MP, Kaiser JA, McAninch JW, et al: The role of computed tomography in renal trauma. Radiology 141:455, 1981
56. Fink IJ, Reinig JW, Dwyer AJ, et al: MR imaging of pheochromocytomas. J Comput Assist Tomogr 9:454, 1985
57. Fischer HW, Doust VL: An evaluation of pretesting in the problem of serious and fatal reactions to excretory urography. Radiology 103:497, 1972
58. Fischer HW, Spataro FR, Rosenberg PM: Medical and economic considerations in using a new contrast medium. Arch Intern Med 146:1717, 1986
59. Forman HP, Middleton WD, Melson GL, McClennan BL: Hyperechoic renal cell carcinomas: Increase in detection at US. Radiology 188:431–434, 1993
60. Fryback DG, Thornbury JR: Informal use of decision theory to improve radiological patient management. Radiology 129:385, 1978
61. Geisinger MA, Risius B, Jordan ML, et al: Magnetic resonance imaging of renal transplants. AJR Am J Roentgenol 143:1229, 1984
62. Geraghty MJ, Lee FT Jr, Bernsten SA, Gilchrist K, Pozniak MA, Yandow DJ: Sonography of testicular tumors and tumor-like conditions: A radiologic-pathologic correlation. Clin Rev Diagn Imaging (in press).
63. Gifford RW Jr: Epidemiology and clinical manifestations of renovascular hypertension. In Stanley JC, Ernst CB, Fry WJ (eds): Renovascular hypertension, pp 77–99. Philadelphia, WB Saunders, 1984
64. Goldin RR, Rosen RA: Effect of inguinal hernias upon the bladder and ureters. Radiology 115:55, 1975
65. Goldman SM, Hartman DS, Fishman EK, et al: CT of xanthogranulomatous pyelonephritis: Radiologic-pathologic correlation. AJR Am J Roentgenol 142:963, 1984
66. Goldstein HM, Medellin H, Beydoun MT, et al: Transcatheter embolization of renal cell carcinoma. AJR Am J Roentgenol 123:557, 1975
67. Grantham JJ, Levine E: Acquired renal cystic disease: Replacing the kidney disease with another. Kidney Int 28:99–105, 1985
68. Greenberger PA, Patterson R: The prevention of immediate generalized reactions to radiocontrast media in high risk patients. J Allergy Clin Immunol 87:867–872, 1991
69. Griscom NT, Vawter GF, Fellers FX: Pelvoinfundibular atresia: The usual form of multicystic kidney; 44 unilateral and two bilateral cases. Semin Roentgenol 10:125, 1975
70. Grist TM, Charles HC, Sostman HD: Renal transplant rejection: Diagnosis with 31P MR spectroscopy. AJR Am J Roentgenol 156:105, 1991
71. Handa N, Fukunaga R, Etani H, Yoneda S, Kimura K, Kamada T: Efficacy of echo-Doppler examination for the evaluation of renovascular disease. Ultrasound Med Biol 14:1–5, 1988
72. Hartman DS: An overview of renal cystic disease. In Hartman DS (ed): Renal Cystic Disease, p 2. Philadelphia, WB Saunders, 1989
73. Hartman DS, Goldman SM, Friedman AC, et al: Angiomyolipoma: Ultrasonic-pathologic correlation. Radiology 139:451, 1981
74. Hartman GW, Hodson CJ: The duplex kidney and related abnormalities. Clin Radiol 20:387, 1969
75. Healy ME, Teng SS, Moss AA: Uriniferous pseudocyst: Computed tomographic findings. Radiology 153:757, 1984
76. Heiberg E, Wolverson MK, Sundaram M, et al: CT findings in leukemia. AJR Am J Roentgenol 143:1317, 1984
77. Helenon O, Rody FE, Correas JM, et al: Color Doppler ultrasound of renovascular disease in native kidneys. Radiographics 15:833–854, 1995
78. Helenon O, Chretien Y, Paraf F, Melki P, Denys A, Moreau JF: Renal cell carcinoma containing fat: Demonstration with CT. Radiology 188:429–430, 1993
79. Henthrone JC: Peripelvic cysts of the kidney: A review of the literature on peripelvic cysts. Am J Clin Pathol 8:28, 1938
80. Hertzberg GS, Carroll BA, Bowie JD, et al: Doppler US assessment of maternal kidneys: Analysis of intrarenal resistivity indexes in normal pregnancy and physiologic pelvicaliectasis. Radiology 186:689–692, 1993
81. Hillman BJ: Imaging advances in the diagnosis of renovascular hypertension. AJR Am J Roentgenol 153:5–14, 1989
82. Hoddick W, Jeffrey RB, Goldberg HI, et al: CT and sonography of severe renal and perirenal infections. AJR Am J Roentgenol 140:517, 1983
83. Hodson CJ: Reflux Nephropathy: A personal historical review. AJR Am J Roentgenol 137:451, 1981
84. Hodson J: The radiological contribution toward the diagnosis of chronic pyelonephritis. Radiology 88:857, 1967
85. Hoffman EP, Mindelzun RE, Anderson RU: Computed tomography in acute pyelonephritis associated with diabetes. Radiology 135:691, 1980
86. Howard TL: Giant polyp of ureter. J Urol 79:397, 1958
87. Husband JES, Olliff JFC, Williams MP, et al: Bladder cancer: Staging with CT and MR imaging. Radiology 173:435, 1989
88. Hyde I, Wastie ML: Striations (longitudinal mucosal folds) in the upper urinary tract. Br J Radiol 44:445, 1971
89. Jacobsson BF, Jorulf H, Kalantar MS: Nonionic versus ionic contrast media in intravenous urography: Clinical trial in 1,000 consecutive patients. Radiology 167:601, 1988
90. Jafri SZH, Bree RL, Amendola MA, et al: CT of renal and perirenal non-Hodgkin lymphoma. AJR Am J Roentgenol 138:1101, 1982

91. Jeffrey RB: Computed tomography of lymphovascular structures and retroperitoneal soft tissues. In Moss AA, Gamsu G, Genant H (eds): Computed Tomography of the Body, 1st ed. Philadelphia, WB Saunders, 1983
92. Kallman DA, King BF, Hattery BR, et al: Renal vein and inferior vena cava tumor thrombus in renal cell carcinoma: CT, US, MRI, and venacavography. JCAT 16:240–247, 1992
93. Kamholtz RG, Cronan JJ, Dorfman GS: Obstruction and the minimally dilated renal collecting system: US evaluation. Radiology 170:51, 1989
94. Katayama H, Yamaguchi K, Kozuka T, Takashima T, Seez P, Matsuura K: Adverse reactions to ionic and nonionic contrast media: A report from the Japanese committee on the safety of contrast media. Radiology 175:621, 1990
95. Kim D, Edelman R, Kent K, Porter D, Skillman JJ: Abdominal aorta and renal artery stenosis: Evaluation with MR angiography. Radiology 174:727–731, 1990
96. Korobkin M, Giordano TJ, Brodeur FJ, et al: Adrenal adenomas: Relationship between histologic lipid and CT and MR findings. Radiology 200:743–747, 1996
97. Korobkin M, Brodeur FJ, Yutzy GG, et al: Differentiation of adrenal adenomas from nonadenomas using CT attenuation values. AJR Am J Roentgenol 166:531–536, 1996
98. Korobkin M, Brodeur FJ, Francis IR, Quint LE, Dunnick NR, Goodsitt M: Delayed enhanced CT for differentiation of benign from malignant adrenal masses. Radiology 200:737–742, 1996
99. Kuhns LR, Hernandez R, Koff S, et al: Absence of vesico-ureteral reflux in children with ureteral jets. Radiology 124:185, 1977
100. Kunin M: Bridging septa of the perinephric space: Anatomic, pathologic, and diagnostic considerations. Radiology 158:361, 1986
101. Kwok-Liu JP, Zikman JM, Cockshott WP: Carcinoma of the urachus: Role of computed tomography. Radiology 137:731, 1980
102. Lalli AF: Urographic contrast media reactions and anxiety. Radiology 112:267, 1974
103. Lalmand B, Avni EF, Nasr A: Perinatal renal vein thrombosis. J Ultrasound Med 9:437, 1990
103A. Landis SH, Murray T, Bolden S, Wingo PA: Cancer statistics, 1998. CA Cancer J Clin 48:6–29, 1998
104. Lang EK, Sullivan J, Frentz G: Renal trauma: Radiological studies. Comparison of urography, computed tomography, angiography, and radionuclide studies. Radiology 154:1, 1984
105. Lasser EC: Pretreatment with corticosteroids to prevent reactions to IV contrast material: Overview and implications. AJR Am J Roentgenol 150:257, 1988
106. Lee F, Littrup PJ, Loft-Christensen L: Predicted prostate specific antigen results using transrectal ultrasound gland volume: Differentiation of benign prostatic hyperplasia and prostate cancer. Cancer 70(Suppl 1):211–220, 1992
107. Lee F, Torp-Pedersen ST, Siders DB, et al: Transrectal ultrasound in the diagnosis and staging of prostatic carcinoma: State of the art. Radiology 170:609, 1989
108. Lee KT, Deeths TM: Localized amyloidosis of the ureter. Radiology 120:60, 1976
109. Lee MJ, Hahn PF, Papanicolan N, et al: Benign and malignant adrenal masses: CT distinction with attenuation coefficients. Radiology 179:415–418, 1991
110. Leonideas JC, McCauley RG, Klauber GC: Sonography as a substitute for excretory urography in children with urinary tract infection. AJR Am J Roentgenol 144:815, 1985
111. Lerner RM, Mevorach RA, Hulbert WC, et al: Color Doppler US in the evaluation of scrotal disease. Radiology 176:355, 1990
112. Levine E, Collins DL, Horton WA, et al: CT screening of the abdomen in von Hipple-Lindau disease. AJR Am J Roentgenol 139:505–510, 1982
113. LiPuma JP: Magnetic resonance imaging of the kidney. Radiol Clin North Am 22:925, 1984
114. Lowe RE, Cohn MD: Computed tomographic evaluation of Wilms tumor and neuroblastoma. Radiographics 4:915, 1984
115. Lutzker LG, Zuckier LS: Testicular scanning and other applications of radionuclide imaging of the genital tract. Semin Nucl Med 20:159, 1990
116. Mann SJ, Pickering TG, Sos TA, et al: Captopril renography in the diagnosis of renal artery stenosis: Accuracy and limitations. Am J Med 90:30–40, 1991
117. Mayo-Smith WW, Lee MJ, McNicholas MMJ, et al: Characterization of adrenal masses (<5 cm) by use of chemical shift MR imaging: observe performance vs. quantitative measures. AJR Am J Roentgenol 165:91–95, 1995
118. McClennan BL, Stanley RJ, Melson GL, et al: CT of the renal cyst: Is cyst aspiration necessary? AJR Am J Roentgenol 133:671, 1979
119. McClennan BL, Stolberg HO: Intravascular contrast media: Ionic versus nonionic. Current status. Radiol Clin North Am 29:437, 1991
120. McClennan BL: Ioxaglate (Hexabrix): A new low osmolality contrast medium. Invest Radiol 19(Suppl 6):5289–5292, 1984
121. Melson GL, Shackelford GD, Cole BR, McClennan BL: The spectrum of sonographic findings in infantile polycystic kidney disease with urographic and clinical correlations. J Clin Ultrasound 13:113–119, 1985
122. Mena E, Bookstein JJ, McDonald FD, et al: Angiographic findings in renal medullary cystic disease. Radiology 110:277, 1974
123. Meng C-H, Elkin M: Venous impressions on the calyceal system. Radiology 87:878, 1966
124. Meyers MA: Dynamic Radiology of the Abdomen, 3rd ed. New York, Springer-Verlag, 1988
125. Miller SS, Winston MC: Nephrogenic diabetes insipidus. Radiology 87:893, 1966
126. Mitchell DG, Crovello M, Mattencci T, Peterson RO, Miettinen MM: Benign adrenocortical masses: Diagnosis with chemical shift MR imaging. Radiology 185:345–351, 1992
127. Mitnick JS, Bosniak MA, Rothberg M, et al: Metastatic neoplasm to the kidney studied by computed tomography and sonography. J Comput Assist Tomogr 9:43, 1985
128. Molmenti EP, Balfe DM, Kanterman RY, Bennett HF: Anatomy of the retroperitoneum: Observations of the distribution of pathologic fluid collections. Radiology 200:95–103, 1996
129. Mooney JK, Berdon WE, Lattimer JK: A new dimension in the diagnosis of posterior urethral valves in children. J Urol 113:272, 1975
130. Moore RD, Steinberg EP, Powe NR: Frequency and determinants of adverse reactions induced by high-osmolarity contrast media. Radiology 170:727–732, 1989
131. Moss AA: Milk of calcium of the adrenal gland. Br J Radiol 49:186, 1976
132. Muller FB, Sealey JE, Case DB, et al: The captopril test for identifying renovascular disease in hypertensive patients. Am J Med 80:633–644, 1986
133. Mulligan SA, Holley HC, Koehler RE, et al: CT and MR imaging in the evaluation of retroperitoneal fibrosis. J Comput Assist Tomogr 13:277, 1989
134. Mundth ED, Shine K, Austen WG: Correction of malignant hypertension and return of renal function following late renal artery embolectomy. Am J Med 46:985, 1969
135. Murray RL: Milk of calcium in the kidney: Diagnostic features on vertical beam roentgenograms. AJR Am J Roentgenol 113:455, 1971
136. Nebesar RA, Pollard JJ, Fraley EE: Renal vascular impressions: Incidence and clinical significance. AJR Am J Roentgenol 101:719, 1967
137. Needleman L, Kurtz AB: Doppler evaluation of the renal transplant. J Clin Ultrasound 15:661, 1987
138. Neifeld JP, Walsh JW, Lawrence W Jr: Computed tomography in the management of soft tissue tumors. Surg Gynecol Obstet 155:535, 1982
139. Newhouse JH: Clinical use of urinary tract magnetic resonance imaging. Radiol Clin North Am 29:455, 1991
140. Novick AC: Management of renovascular disease. Circulation 83:167–171, 1991
141. Orazi C, Fariello G, Malena S, et al: Renal vein thrombosis and adrenal hemorrhage in the newborn: Ultrasound evaluation of 4 cases. JCU 21:163–169, 1993
142. Osathanondh V, Potter EL: Pathogenesis of polycystic kidneys: Type III due to multiple abnormalities of development. Arch Pathol 77:485, 1964
143. Palmer FJ: Renal cortical calcification. Clin Radiol 21:175, 1970
144. Palmisano SM: Low osmolality contrast media in the 1990’s: prices change. Radiology 203:309–315, 1997
145. Palubinskas AJ: Renal pyramidal structure opacification in excretory urography and its relation to medullary sponge kidney. Radiology 81:963, 1963
146. Peake SL, Roxburgh HB, Langlois SLP: Ultrasonic assessment of hydronephrosis of pregnancy. Radiology 146:167, 1983
147. Pfister RC, McLaughlin AP III, Leadbetter WF: Radiological evaluation of primary megaloureter. Radiology 99:503, 1971
148. Pfister RC, Newhouse JH: Interventional percutaneous pyeloureteral techniques: I and II. Radiol Clin North Am 17:341, 1979

149. Pitts WR Jr, Muecke EC: Congenital megaloureter: A review of 80 patients. J Urol 111:468–473, 1974
150. Platt JF, Rubin JM, Bowerman RA, et al: The inability to detect kidney disease on the basis of echogenicity. AJR Am J Roentgenol 151:317, 1988
151. Platt JF, Rubin JM, Ellis JH: Acute renal obstruction: Evaluation with intrarenal duplex Doppler and conventional US. Radiology 186:685–688, 1993
152. Pollack HM, Arger PH, Banner MP, et al: Computed tomography of renal pelvis filling defects. Radiology 138:645, 1981
153. Pollack HM, Banner MP, Arger PH: The accuracy of gray-scale renal ultrasonography in differentiating cystic neoplasms from benign cysts. Radiology 143:741, 1982
154. Poynter JD, Hare WSC: Necrosis in situ: A form of renal papillary necrosis seen in analgesic nephropathy. Radiology 111:69, 1974
155. Pozniak MA, Kelcz F, D’Alessandro A, et al: Sonography of renal transplants in dogs: The effect of acute tubular necrosis, cyclosporine nephrotoxicity, and acute rejection on resistive index and renal length. AJR Am J Roentgenol 158:791, 1992
156. Pozniak MA, Kelcz F, Stratta RJ, et al: Extraneous factors affecting resistive index. Invest Radiol 169:367, 1988
157. Resnick MI, Sanders RC: Ultrasound in urology, 2nd ed. Baltimore/London, Williams & Wilkins, 1984
158. Reynolds L, Fulton H, Snider JJ: Roentgen analysis of renal mass lesions (cysts and tumors). AJR Am J Roentgenol 82:840, 1959
159. Rifkin MD, McGlynn ET, Choi H: Echogenicity of prostate cancer correlated with histologic grade and stromal fibrosis: Endorectal US studies. Radiology 170:549, 1989
160. Rifkin MD, Needleman L, Pasto ME, et al: Evaluation of renal transplant rejection by duplex Doppler examination: Value of the resistive index. AJR Am J Roentgenol 148:759, 1987
161. Rubin GD, Alfrey EJ, Dake MD, et al: Assessment of living renal donors with spiral CT. Radiology 195:457–462, 1995
162. Ruzal-Shapiro C, Newhouse JH: Imaging of scrotal contents. In Taveras JM, Ferrucci JT (eds): Radiology: Diagnosis, Imaging, Intervention, vol 4, pp 1–10, 134. Philadelphia, JB Lippincott, 1988
163. Sandler CM, Phillips JM, Harris JD, et al: Radiology of the bladder and urethra in blunt pelvic trauma. Radiol Clin North Am 19:195, 1981
164. Sandler CM, Toombs BD: Computed tomographic evaluation of blunt renal injuries. Radiology 141:461, 1981
165. Sarti DA: Diagnostic ultrasound text and cases, 2nd ed. Chicago, Year Book Medical Publishers, 1987
166. Scott JA, Rake FE, Becker GJ, et al: Angiographic assessment of renal artery pathology. AJR Am J Roentgenol 141:1299–1303, 1983
167. Scott WW: Review of primary carcinoma of the ureter with report of case. J Urol 50:45, 1943
168. Segal AJ, Spataro FR, Linke CA, et al: Diagnosis of nonopaque calculi by computed tomography. Radiology 129:447, 1978
169. Semelka RC, Shoenut JP, Kroeker MA, et al: Renal lesions: Controlled comparison between CT and 1.5-T MR imaging with nonenhanced and gadolinium-enhanced fat-suppressed spin-echo and breath hold FLASH techniques. Radiology 182:425, 1992
170. Semelka RC, Kelekis NL, Burdeny DA, Mitchell DG, Brown JJ, Siegelman ES: Renal lymphoma: Demonstration by MR imaging. AJR Am J Roentgenol 166:823–827, 1996
171. Shirkhoda A: CT findings in hepatosplenic and renal candidiasis. JCAT 11:795–798, 1987
172. Smith RC, Verga M, McCarthy SM, et al: Diagnosis of acute flank pain: Comparison of non-contrast enhanced CT and intravenous urography. Radiology 194:789–794, 1995
173. Snider JF, Hunter DW, Moradian GA, et al: Transplant renal artery stenosis: Evaluation with duplex sonography. Radiology 172:1027, 1989
174. Stanley JC, Rhodes EL, Gewertz BL, et al: Renal artery aneurysms: Significance of macroaneurysms exclusive of dissections and fibrodysplasic mural dilations. Arch Surg 110:1327, 1975
175. Steinberg FL, Yucel EK, Dumoulin CL, Souza SP: Peripheral vascular and abdominal applications of MR flow imaging techniques. Magn Reson Med 14:315–320, 1990
176. Strotzer M, Lehner KB, Becker K: Detection of fat in a renal cell carcinoma mimicking angiomyolipoma. Radiology 188:427–428, 1993
177. Subcommittee on Definition and Prevalence of the 1984 Joint National Committee: Hypertension prevalence and the status of awareness, treatment and control in the United States (final report). Hypertension 7:457–468, 1985
178. Subramanyam BR, Bosniak MA, Horii SC, et al: Replacement lipomatosis of the kidney: Diagnosis by computed tomography and sonography. Radiology 148:791, 1983
179. Svetky LP, Himmelstein SI, Dunnie NR, et al: Prospective analysis of strategies for diagnosing renovascular hypertension. Hypertension 14:246–256, 1989
180. Talner LB, Gittis RF: Megacalyces. Clin Radiol 23:355, 1972
181. Talner LB: Urographic contrast media in uremia? Physiology and pharmacology. Radiol Clin North Am 10:421, 1972
182. Taylor A Jr, Ziffer JA, Eshima D: Comparison of Tc-99m-MAG3 and Tc-99m-DTPA in renal transplant patients with impaired renal function. Clin Nucl Med 15:371, 1990
183. Taylor DC, Kettler MD, Monetta GL, et al: Duplex-ultrasound scanning in the diagnosis of renal artery stenosis: A prospective evaluation. J Vasc Surg 7:363–369, 1988
184. Taylor KJW, Morse SS, Rigsby CM, et al: Vascular complications in renal allografts: Detection with duplex Doppler ultrasound. Radiology 162:31, 1987
185. Thornbury JR, McCormick TL, Silver TM: Anatomic/radiologic classification of renal cortical nodules. AJR Am J Roentgenol 134:1, 1980
186. Thornbury JR, Parker TW: Ureteral calculi. Semin Roentgenol 17:133, 1982
187. Thornbury JR, Silver TM, Vinson RK: Ureteroceles vs pseudoureteroceles in adults. Radiology 122:81, 1977
188. Thornbury JR: Acute renal infections. Urol Radiol 12:209, 1991
189. Thornbury JR: The roentgen diagnosis of ureterocele in children. AJR Am J Roentgenol 90:15, 1963
190. Thrall JH, Koff SA, Keyes JR: Diuretic radionuclide renography and scintigraphy in the differential diagnosis of hydroureteronephrosis. Semin Nucl Med 11:89, 1981
191. Tublin ME, Dodd GD III, Verdile VP: Acute renal colic: Diagnosis with duplex Doppler US. Radiology 193:697–701, 1994
192. Tuite MJ, Weiss SL: Ultrasound and computed tomographic appearance of extramedullary hematopoiesis encasing the renal pelvis. J Clin Ultrasound 19:238, 1991
193. Twersky J, Levin DC: Metastatic melanoma of the adrenal. Radiology 116:627, 1975
193A. Wantabe H, Igari D, Tanahashi Y, Harada K, Saitoh M: Transrectal ultrasonotomography of the prostate. J Urol 114:734–739, 1975
194. Welch TJ, Sheedy PF II, Van Heerden JA, et al: Pheochromocytoma: Value of computed tomography. Radiology 148:501, 1983
195. White EA, Schambelan M, Rost CR, et al: Use of computed tomography in diagnosing the cause of primary aldosteronism. N Engl J Med 303:1503, 1980
196. Wicks JD, Thornbury JR: Acute renal infections in adults. Radiol Clin North Am 17:245, 1979
197. Wilms G, Marchal G, Peene P, Baer AL: The angiographic incidence of renal artery stenosis in the arteriosclerotic population. Eur J Radiol 10:195–197, 1990
198. Wilson TE, Doelle EA, Cohan RH, Wojro K, Korobkin M: Cystic renal masses: A reevaluation of the usefulness of the Bosniak classification system. Acad Radiol 3:564–570, 1996
199. Winograd J, Schimmel DH, Palubinskas AJ: The spotted nephrogram of renal scleroderma. AJR Am J Roentgenol 126:734, 1976
200. Wolfman MG, Thornbury JR, Braunstein EM: Nonobstructing radiopaque ureteral calculi. Urol Radiol 1:97–104, 1979
Berdon WE, Baker DH: The significance of a distended bladder in the interpretation of intravenous pyelograms obtained on patients with “hydronephrosis.” AJR Am J Roentgenol 120:402, 1974
Bozniak MA, Ambos MA, Madayag MA, et al: Epinephrine-enhanced renal angiography in renal mass lesions: Is it worth performing? AJR Am J Roentgenol 129:647, 1977
Cohen RH, Dunnick NR, Bashore TM: Treatment of reactions to radiographic contrast material. AJR Am J Roentgenol 151:263, 1988
Dachman AH: New contraindication to intravascular iodinated contrast material [letter]. Radiology 197:545, 1995
Davidson AJ: Radiology of the Kidney, 2nd ed. Philadelphia, WB Saunders, 1985
Elkin M, Bernstein J: Cystic diseases of the kidney: Radiological and pathological considerations. Clin Radiol 20:65, 1969

Fein AB, Lee JKT, Balfe DM, et al: Diagnosis and staging of renal cell carcinoma: A comparison of MR imaging and CT. AJR Am J Roentgenol 148:749, 1987
Glazer GM: MR imaging of the liver, kidneys and adrenal glands. Radiology 166:303, 1989
Griscom NT, Colodny AH, Rosenberg KH, et al: Diagnostic aspects of neonatal ascites: Report of 27 cases. AJR Am J Roentgenol 128:961, 1977
Kim B, Semelka RC, Ascher SM, Chalpin DB, Carroll PR, Hricak H: Bladder tumor staging: Comparison of contrast-enhanced CT, T1- and T2-weighted MR imaging, dynamic gadolinium-enhanced imaging, and late gadolinium-enhanced imaging. Radiology 193:239–245, 1994
Koep L, Zuidema GD: The clinical significance of retroperitoneal fibrosis. Surgery 81:250, 1977
Krestin GP, Steinbrich W, Friedmann G: Adrenal masses: Evaluation with fat gradient-echo MR imaging and Gd-DTPA enhanced dynamic studies. Radiology 171:675, 1989
McClennan BL, Deyoe LA: Imaging evaluation of renal cell carcinoma: Diagnosis and staging. Radiol Clin North Am 32:55–69, 1994
Seltzer SE, Getty DJ, Tempany CMC, et al: Staging prostate cancer with MR imaging. A combined radiology-computer system. Radiology 202:219–226, 1997
Talner LB, Davidson AJ, Lebowitz RC, Dalla Palma L, Goldman SM: Acute pyelonephritis: Can we agree on terminology? Radiology 192:297–305, 1994
Tempany CMC, Zhon X, Zerhouni EA, et al: Staging of prostate cancer: Results of radiology diagnostic oncology group project comparison of three MR imaging techniques. Radiology 192:47–53, 1994
Weyman PJ, McClennan BL, Stanley RJ, et al: Comparison of computed tomography and angiography in the evaluation of renal cell carcinoma. Radiology 137:417, 1980
Zagoria RJ, Bechtold RE, Dyer RB: Staging of renal adenocarcinoma: Role of various imaging procedures. AJR Am J Roentgenol 164:363–370, 1995