Two basic imaging characteristics of solid renal masses are encountered on cross-sectional imaging studies: those that contain macroscopic fat and those that do not. Nearly all of the former lesions represent benign AMLs, while the majority of the latter represent renal cancers. Pre- and postcontrast thin-section multiphasic CT or MRI are the methods of choice for evaluating a patient with a suspected renal mass. When a renal mass is identified on a CT scan performed for another purpose, it may be prudent to perform a dedicated renal CT in order to assess the vascularity of the lesion and its enhancement characteristics, as well as to assess the status of the inferior vena cava, regional lymph nodes, and adjacent organs.

Although ultrasound can be helpful in distinguishing solid from cystic renal masses, it is not nearly as sensitive in detecting small renal masses as is CT or MRI. In one study of patients with VHL followed with US and CT examinations US failed to identify 80% of masses smaller than 1 cm seen on CT. When detected on US, solid renal masses are usually identified as such, although cystic regions representing areas of hemorrhage or necrosis may be seen. Microbubble contrast ultrasound agents have been shown to enhance vascular elements within soft tissues, including the kidney, and may be useful in evaluating patients with known or suspected renal cancers when contrast agents cannot be administered for CT or MRI.

Isoechoic solid renal tumors may be difficult to detect with ultrasound, especially if they are small and do not displace the collecting system or produce a contour deformity. Less echogenic primary renal tumors can simulate more homogeneous tumors such as lymphomas or may be confused with renal cysts. However, these solid tumors usually have poorly defined margins and do not have the increased sound transmission seen with cysts. A few renal adenocarcinomas are primarily cystic. However, these cystic tumors can usually be distinguished from simple cysts by their irregularly thick walls and the presence of some internal echoes. Duplex and color Doppler ultrasounds provide a noninvasive measure of the vascularity of a renal mass. High-velocity signals, which are presumably due to arteriovenous shunting, are frequently seen in renal carcinomas.

Dedicated renal mass CT or MRI should be obtained as the study of choice in any patient with a known or suspected solid renal mass. At a minimum, a two-phase study that includes a noncontrast examination of the liver and kidneys and postcontrast study made during the nephrographic phase (NP) should be performed. In some instances, the addition of arterial phase (e.g., to evaluate the arteries prior to planned surgery), corticomedullary phase (CMP) (to evaluate the veins and other organs), and/or excretory phase (to evaluate the renal collecting systems) series may also be helpful. Arterial phase images (in which the arteries enhance briskly, the renal cortex enhances briskly, but the renal medulla demonstrates little enhancement) are usually acquired beginning at about 20 to 30 seconds following the initiation of intravenous contrast material administration. CMP images (in which the arterial enhancement is not as pronounced, but the renal cortex continues to enhance briskly and the medulla enhances, but to a much lesser extent) are routinely acquired beginning at about 60 to 70 seconds, NP images (in which the kidney now enhances homogeneously) beginning at about 90 to 100 seconds, and excretory phase (EP) images (when the nephrogram remains homogeneous but begins to fade and when excreted contrast material is detected in the renal collecting systems) beginning at 120 to 180 seconds.

CT is best performed by obtaining narrowly collimated thinsection images (no more than 2.5 to 5.0 mm in thickness) both before and after intravenous contrast material administration. A comparison of the density on the enhanced series to that of the unenhanced image allows for assessment of contrast enhancement. While all findings can be made on the axial images, coronal reformatted images are particularly helpful in identifying renal masses, especially those that are small or that have a polar location.

NP images are important for the detection of solid renal masses for several reasons. Firstly, some solid renal masses will not demonstrate definite enhancement on CT until the NP images occur. If only CMP images are obtained in these patients, a renal cancer might be falsely diagnosed as a high-density renal cyst. Secondly, the persistent differential enhancement of the renal cortex and renal medulla on CMP images may result in failure to diagnose some hypervascular cortical renal masses (enhancing similarly to normal renal cortex) or some hypovascular medullary masses (enhancing similarly to normal renal medulla) (Fig. 6.2).

There are a few problematic issues related to CT imaging of solid renal masses. It has recently been shown that just under 10% of
renal cancers may measure water attenuation on noncontrast images (between -10 and 20 HU). This is problematic, as these lesions could be misdiagnosed as cysts if only unenhanced CT images are available (Fig. 6.3). The low attenuation of these lesions has primarily been seen in some patients with clear cell carcinomas, probably due to the fact that these tumors contain large amounts of intracellular lipid. While the vast majority of water attenuation masses seen on unenhanced CT are cysts, it is important to assess any such masses for even subtle heterogeneity, to minimize the likelihood that a renal cancer is not erroneously diagnosed as a benign lesion.

In general, a renal mass can be considered as being definitely solid (and enhancing) on CT if it increases in attenuation by 20 HU or more (between the unenhanced and enhanced series). Most investigators agree that an increase in attenuation of 10 to 20 HU should be considered equivocal, and follow-up (preferably with MRI, which is more sensitive in detecting enhancement) is warranted. In one recent study, 20 (17%) and 4 (3%) of 116 renal cancers did not enhance by 15 HU or more on CMP images alone and both corticomedullary and NP images, respectively. In this series, if the threshold was increased to 20 HU to define enhancement, the number of renal
cancers that did not enhance increased to 24 (21%) and 11(9%). Only masses that increase in attenuation by 10 HU or less should be considered as truly nonenhancing; however, even this threshold is not absolute. Some hypoenhancing renal cancers may increase in attenuation by <10 HU (Fig. 6.4). Once again, it is essential that any renal mass be assessed for heterogeneity on contrast-enhanced CT images. Most, but not all, nonenhancing or equivocally enhancing renal cancers will not be completely homogeneous. MRI is more sensitive for detecting enhancement. In some series, even those solid renal masses that did not demonstrate an increase in attenuation between nonenhanced and enhanced series on CT could be seen to enhance on subtraction MR images (Fig. 6.5).

Angiomyolipomas (AMLs) are tumors composed of a mixture of mature adipose tissue, thick-walled blood vessels, and sheets of smooth muscle. The amount of each component varies in each tumor. AMLs belong to a group of tumors classified as perivascular epithelioid cell tumors or PEComas. The vast majority of these tumors are benign. They are generally seen in two groups of patients: (1) as small isolated renal masses in middle-aged or older women and (2) in up to 80% of patients with tuberous sclerosis (in which case the tumors are often multiple and are equally likely to occur in men and women). The AMLs associated with tuberous sclerosis also occur at a younger age and tend to be larger in size than the sporadic lesions.

Manifestations of tuberous sclerosis include epilepsy, mental retardation, hamartomas (which may be cerebral), and retinal phakomas. Adenoma sebaceum may be seen in the malar areas of the face. Patients may also develop intrahepatic AMLs. In addition to renal AMLs, renal abnormalities in patients with tuberous sclerosis may also include the development of multiple renal cysts. Tuberous sclerosis is caused by an autosomal dominant inherited gene mutation, with variable expressivity. In some patients, the clinical syndrome is incompletely manifest.
Nonetheless, the presence of multiple AMLs with or without associated renal cysts on imaging studies should raise the possibility of tuberous sclerosis.

Renal AMLs are also found in approximately 15% of patients with lymphangiomyomatosis. Lymphangiomyomatosis, which is considered by many to be a forme fruste of tuberous sclerosis, is an idiopathic disease that occurs in young women. It consists of smooth muscle hamartomas along the lymphatic system. It most commonly involves intrathoracic lymphatics, but abdominal involvement can be extensive. The pulmonary findings are diffuse and include a reticular or reticulonodular pattern and multiple small cysts, with a honeycomb-like appearance.

Most patients with AMLs have no symptoms; however, patients with large tumors may present with abdominal fullness or pain. Patients who develop hemorrhage, the most common and potentially clinically significant complication, can present with sudden back or flank pain, hypotension, and/or hematuria. Hemorrhage is more likely to occur when AMLs exceed 4 cm in maximal diameter and/or in tumors that contain more and larger renal artery branch aneurysms.

About 95% of AMLs contain enough macroscopic fat to produce characteristic findings on imaging studies. On plain abdominal radiography, some large AMLs may contain enough fat to produce an area of increased radiolucency. Calcification is seldom seen with conventional radiography or on other imaging studies, but may rarely be present, usually as a result of previous hemorrhage.

On ultrasonography, most AMLs are very echogenic (Figs. 6.6 and 6.7). Hypoechoic areas are occasionally encountered within an echogenic mass, however, possibly resulting from prior hemorrhage. It should be noted, however, that small renal cancers may also be echogenic (Fig. 6.8). Therefore, hyperechogenicity cannot be used to establish a definite diagnosis of AML. Some AMLs may also demonstrate posterior shadowing, a finding not encountered posterior to renal cancers (Fig. 6.9). Conversely, an anechoic rim or halo is sometime present around hyperechoic renal carcinomas, but has not been seen surrounding AMLs.

Both CT and MRI can be used to make diagnoses of AML with certainty in most instances. On CT, the detection of any component of a solid renal mass that measures fat attenuation is considered to
be definitive. In general, macroscopic fat can be identified if any measured component of a solid renal mass measures -10 HU or less (Figs. 6.10, 6.11 and 6.12). Examination of a small fatty renal mass by CT must be performed carefully to avoid volume averaging of a solid mass not containing any macroscopic fat with adjacent perinephric fat and a false low-density reading. Additionally, volume averaging of an AML containing macroscopic fat with adjacent normal renal parenchyma can create a false high density reading (Fig. 6.13). Other approaches for identifying small amounts of fat (such as counting individual pixels or pixel distribution) remain controversial and have not been widely accepted.

CT is occasionally performed in symptomatic patients with AMLs. Spontaneous hemorrhage can produce large acute perinephric hematomas or even active extravasation of blood if arterial phase contrast-enhanced images are acquired (Fig. 6.14). In some cases, the perinephric hematoma may be sufficiently large enough to obscure visualization of the bleeding tumor, in which case identification of the AML may require repeat imaging after the hematoma has resolved. For this reason, repeat 4- to 6-week follow-up imaging is recommended in any patient who has spontaneous subcapsular or perinephric hemorrhage if no cause of the hemorrhage can be visualized on the initial imaging study.

MRI can detect fat within an AML, also allowing for a definitive diagnosis to be made. The macroscopic fat in AMLs produces high signal intensity that is seen on both T1- and T2-weighted images. Most characteristic is the loss of signal intensity of components of macroscopic fat on fat-suppressed images (Fig. 6.15C). Chemical shift imaging may also be helpful, as a thin dark line, often termed “India ink” artifact, is usually seen at the interface between AMLs and adjacent renal parenchyma on opposed-phase images, with this line representing signal loss at a fat-fluid interface (Fig. 6.15A,B). While an entire AML may also lose signal on opposed-phase images, this feature is not as helpful for two reasons: (1) some AMLs do not lose signal (since they may contain a great deal of fat, but little fluid, with both of these components required for signal loss), and (2) some clear cell renal cancers can lose signal on these images (due to their containing a great deal of intracellular lipid) (Fig. 6.16).

Although macroscopic fat has occasionally been identified in other renal neoplasms, including renal cancer (often due to osseous metaplasia, in which case calcification is also often present), oncocytomas, Wilms tumor, and metastases, these are all considered to be case reportable exceptions (Fig. 6.17). For this reason, any

mass that contains macroscopic fat on CT or MRI should be considered an AML, unless that mass contains calcification (a feature that can only be detected on CT). Additionally, sometimes, solid renal masses can engulf perinephric or renal sinus fat, creating the impression that the mass itself contains the fat. Only rarely, does this create confusion on CT.

Both lipomas and liposarcomas usually also contain CT-detectable macroscopic fat. These neoplasms can usually be distinguished from AMLs, because the former create smooth impressions on the kidney, while the latter create a defect in the renal contour (indicating the site at which they originated) (Figs. 6.18 and 6.19). Also, AMLs are usually hypervascular lesions, demonstrating large enhancing vessels on contrast-enhanced CT, while lipomas and liposarcomas are hypovascular.

Several variants of AMLs have been described. Up to 5% of all AMLs contain little or no fat. For this reason, the absence of fat does not exclude the diagnosis of AML. These tumors are referred to as lipid-poor or minimal fat-containing AMLs. On ultrasonography, lipid-poor AMLs are often isoechoic rather than echogenic (Fig. 6.20A). On CT, AML without detectable fat usually appear as homogenous tumors with attenuation higher than normal renal parenchyma (Fig. 6.20B). By definition, these lesions do not contain any identifiable macroscopic fat, as seen on either CT or MRI (Fig. 6.20C). They may demonstrate homogeneous enhancement after the intravenous administration of contrast material (Fig. 6.21). These lesions cannot be distinguished from other nonfat-containing renal masses, including renal cancers.

There are several rare histologic types of AML. Epithelioid AMLs are composed partially or entirely of epithelioid cells having abundant neoplasm. Most of these lesions behave as benign lesions; however, approximately one-third of these AML subtypes have aggressive features. On histology, these lesions contain two
or more mitotic figures per high-power field. Although epithelioid AMLs are probably less likely to contain identifiable macroscopic fat on cross-sectional imaging studies (Fig. 6.22), at least some of these tumors do have visible fat (Fig. 6.23), making distinction from other types of AML impossible. Clinically, these tumors can invade the renal veins or inferior vena cava and can even metastasize distantly. Thus, when a fatty mass appears to be behaving aggressively on CT or MRI (extending locally into other organs or growing into the inferior vena cava), a diagnosis of epithelioid AML should be considered (Fig. 6.24).

Another AML variant is the AML with epithelial cysts (AMLEC). These tumors, which are benign, contain both solid and cystic elements, both of which can be visualized on cross-sectional imaging
studies. Macroscopic fat has not been described in these lesions. For this reason, AMLECs can be confused with cystic renal cancers.

Treatment of isolated/solitary AMLs is unnecessary in asymptomatic patients with lesions that measure up to 4 cm in maximal diameter. Partial nephrectomy is recommended by many urologists for AMLs exceeding 4 cm in diameter, due to their increased propensity to bleed. As an alternative, selective arterial embolization may be performed in patients with large tumors, if they are not considered amenable to partial nephrectomy. Should hemorrhage occur, prompt catheter embolization is often effective. Surgical treatment is much more problematic in patients with tuberous sclerosis, when many AMLs may be present. Recently, it has been found that some chemotherapeutic agents can be effective in reducing AML size in such patients. Specifically, mammalian target of rapamycin (mTOR) inhibitors, such as temsirolimus, have been used in this setting. These agents
have been observed to reduce tumor volumes by 50% or more in nearly half of the treated patients, with response being greatest during the first year of treatment (beginning at about 3 months) (Fig. 6.25). Of course, mTOR inhibitors are not without side effects, which include stomatitis, hyperlipidemia, amenorrhea, and increased susceptibility to infection.

Many renal masses do not contain any identifiable fat. Most of these non-fat-containing renal masses represent renal cancers; however, up to 20% of solid renal masses measuring 4 cm or less and nearly 50% of solid renal masses measuring 1 cm or less in size have been found to be benign. The odds that a large renal mass is a cancer are much higher, with nearly 95% of solid renal masses exceeding 7 cm in diameter being malignant.

While there have been a large number of studies attempting to analyze both CT and MRI features of non-fat-containing masses in an effort to identify imaging features, which allow for their differentiation from one another (including noncontrast attenuation or signal intensity, degree of enhancement, rate of enhancement, and pixel distribution on region of interest analysis), overall, these have not been very successful, due to overlap. Occasionally, however, some combinations of features may be suggestive, as will be discussed in the paragraphs that follow.

It should be emphasized that, to date, there is also no proof that differences in renal mass growth rate can be utilized to differentiate malignant from benign renal masses. Several studies have demonstrated that malignant and benign renal masses frequently enlarge at similar rates. Some researchers have recommended that growth of a renal mass should only be considered suspicious when it exceeds 5 mm within a 12 month period of time.

In recent years, there has been a concerted movement toward performance of percutaneous biopsy to determine the nature of small non-fat-containing solid renal masses. Biopsies can be performed safely in the vast majority of patients, with a low complication rate and little risk of significant bleeding or seeding along the tumor tract.

Renal cell carcinoma (RCC) is twice as common in men as in women. This tumor may occur at any age, but the incidence peaks in the sixth decade. No specific etiologic agents are recognized, but there is an as much as 50% increased incidence in patients who use tobacco and an increased incidence in patients with obesity and uncontrolled hypertension. RCC has been increasing worldwide, partly owing to the more frequent detection by imaging, but also to an increase in the prevalence of risk factors. There is also a fourfold increase in the incidence of the disease in first-degree family members of patients with RCCs. This is largely because about 5% to 10% of renal cancers are believed to develop in patients with hereditary renal cancer syndromes, including von Hippel-Lindau (VHL) disease, hereditary papillary renal cancer, hereditary leiomyomatosis renal cancer syndrome, Birt-Hogg-Dube, hereditary succinate dehydrogenase-deficient renal cell cancer, and Lynch syndrome.

Also, patients on chronic hemodialysis or peritoneal dialysis nearly always eventually develop acquired renal cystic disease (see Chapter 8). These patients have an incidence of renal carcinoma of approximately 7%, although not all of these tumors behave in a biologically aggressive manner. Tumors in patients with acquired cystic disease of dialysis begin to appear in patients as early as 3 years after initiation of dialysis. They can be difficult to detect because these patients’ kidneys are not functioning and usually contain multiple cysts and dystrophic calcifications. The incidence of renal tumors seems to parallel the development of acquired cystic disease, such that more tumors are seen the longer the patients are maintained on dialysis. The effect of renal transplantation on the development of renal cancer is unclear. Although it is likely that successful transplantation reduces the risk of renal carcinoma, this is counterbalanced by the increased risk of malignancy with immunosuppression.

The classic clinical presentation of renal cancer, of a flank mass, pain, and hematuria occurs in only a minority of patients. As abdominal computed tomography (CT) and ultrasound examinations are commonly performed for a variety of nonrenal indications, the vast majority of renal tumors are now discovered incidentally in patients who are asymptomatic.

Among symptomatic patients, hematuria is the most common sign, occurring in more than 50% of patients. Flank pain, present in more than one-third of patients, is probably caused by distension of the renal capsule. A flank mass is palpable at presentation in approximately one-third of patients with large tumors.

Occasionally, patients with large renal tumors may first complain of nonspecific symptoms such as weight loss, fatigue, or even gastrointestinal or neurologic symptoms. Less common presenting complaints include fever or a new left-sided varicocele (Fig. 6.26). Fatigue may be caused by a normochromic normocytic anemia. A variety of hormones may be secreted by
RCCs in sufficient quantity to cause distinct clinical manifestations, including renin, erythropoietin, parathyroid hormone, adrenocorticotropic hormone, prolactin, and gonadotropin (Table 6.1).

It is difficult to predict the natural history of renal carcinoma. Some tumors demonstrate aggressive behavior by growing rapidly and metastasizing early. However, occasional patients may live for years with an untreated primary tumor. Small tumors, in particular, may show only minimal growth when followed over many years. Metastases generally occur when tumors are larger than 5 cm and are extremely rare when tumors are <3 cm in diameter. Metastases may be present at the time of initial presentation, but have also been reported to first appear as late as 31 years after nephrectomy. Spontaneous regression of metastases or the primary tumor may also occur.