Year : 2005 | Volume
: 15 | Issue : 1 | Page : 117--125
Spectrum of US and CT findings in renal neoplasms with pathologic correlation
P Hatimota, S Vashist, K Aggarwal, A Kapoor, NP Gupta
Department of Radio-diagnosis, Department of Urology, All India Institute of Medical Sciences, New Delhi - 29, India
Department of Radio-diagnosis, AIIMS, New Delhi - 29
Objective: The purpose of this study was to characterize the image morphology of renal neoplasms on US (including CDFI) and CT and to stage these tumors and correlate the imaging findings with operative and/or histopathological findings. Materials and methods:
a total of 50 patients were studied. These included renal cell carcinoma (n=38), angiomyolipoma (n=5), renal metastases (n=3), oncocytoma (n=1), transitional cell carcinoma (n=1), multilocular cystic nephroma (n=1) and nephroblastomatosis with Wilms tumour (n=1). All patients were evaluated by US and CT. Imaging finding were correlated with both operative and histopathological findings in 31 patients. Diagnosis was confirmed by percutaneous biopsy in 14 patients. In 5 patients final diagnosis was made on classical imaging features. Results : Most renal cell carcinomas had a solid appearance. On US, the most common appearance of RCC was predominantly isoechoic to normal renal parenchyma (78%) followed by hypo & hyperechoic lesions. Only 1 patient had cystic RCC. Solid renal cell carcinomas showed presence of peripheral and intralesional vascularity. The peak systolic Doppler frequency shift ranged from 0.5 KHz to 3.6 KHz. On CT, most of the lesions of RCC were heterogeneous in attenuation with inhomogeneous contrast enhancement less than that of normal renal parenchyma. Calcification was seen in 51%, necrosis in 94%, and haemorrhage in 20% lesions of RCC. Cystic RCC had thick walls, internal septations and solid components, all showing enhancement. Most of the angiomyolipomas (n=4) were homogenously hyperechoic on US, showed peripheral vascularity with peak systolic Doppler frequency shift less than 2.5 kHz. On CT, AML most commonly appeared as well defined, heterogeneous mass lesions showing internal fat density and variable contrast enhancement. Metastatic renal lesions were seen as small, homogenously hypoechoic on US, hypo to isodense on CT and showed minimal or no contrast enhancement. One case of oncocytoma showed a well defined, homogenous mass with central echogenic stellate scar and calcification on US. CT showed homogenous contrast enhancement with central hypodense stellate scar. TCC appeared as an intermediate echogenic mass on US, located in the dilated renal pelvis and extending into the upper ureter. On CT the mass was located centrally having density greater than that of urine, showing minimal contrast enhancement with centrifugal extension and invasion of renal parenchyma. Multilocular cystic nephroma was seen as a well-defined lesion with multiple cystic spaces, separated by septae which were echogenic on US and showed mild to moderate enhancement on CT. Conclusion: US and CT show many of the key imaging features used to characterize most of the renal neoplasms. However, some lesions remain indeterminate and require percutaneous biopsy to confirm the diagnosis.
|How to cite this article:|
Hatimota P, Vashist S, Aggarwal K, Kapoor A, Gupta N P. Spectrum of US and CT findings in renal neoplasms with pathologic correlation.Indian J Radiol Imaging 2005;15:117-125
|How to cite this URL:|
Hatimota P, Vashist S, Aggarwal K, Kapoor A, Gupta N P. Spectrum of US and CT findings in renal neoplasms with pathologic correlation. Indian J Radiol Imaging [serial online] 2005 [cited 2019 Aug 17 ];15:117-125
Available from: http://www.ijri.org/text.asp?2005/15/1/117/28761
The imaging features of various imaging modalities in renal neoplasms have been well described in the literature. US and CT being the most commonly used imaging modalities in evaluation of renal neoplasms, this study was undertaken to evaluate their role in assessment of renal neoplasms, with respect to lesion characterization and staging and to correlate the imaging findings with operative and / or histopathology findings.
MATERIALS AND METHODS
The study included a total of 50 cases (41 evaluated prospectively and 9 retrospectively).
The prospective study comprised of 41 patients clinically suspected or incidentally diagnosed on US or CT to have renal neoplasms. A relevant clinical history was taken and clinical examination was performed prior to imaging. Presenting symptoms were hematuria, abdominal lump, pain, weight loss and fever. Six patients were asymptomatic, they were incidentally detected to have renal mass on US which was done for some other reason. In two patients the renal lesions were detected incidentally on CT performed for routine metastatic work up of known primary. After the clinical work up, patients were subjected to the US examination. US was performed using ATL Ultramark-9 USA or Sonoline versa pro Siemens,Germany or ATL-apogee 800 plus USA. Abdomen and pelvis were examined by sonography and sonographic characteristics of renal masses were evaluated viz.. size, location, extension, echotecture,margins and vascularity (using color Doppler).Renal vein and inferior vena cava were evaluated by colour Doppler for any thrombus or invasion.Search was made for retroperitoneal lymphadenopathy or metastasis.
CT was carried out on Somatom plus-4 or AR star (Siemens,Germany) helical CT scanner.Preliminary unenhanced scans were obtained to localize the kidneys, detect calcification and evaluate the attenuation of mass before contrast administration. This was done using 5mm collimation in the helical mode at a pitch of 1 using 120 KV and 200 mA.
For contrast enhanced scans 80-120ml of contrast material was injected in the forearm vein by rapid bolus hand injection technique. Either ionic or nonionic contrast material was used with choice of agent dictated by the patient's history, condition and ability to afford. All patients at risk for an adverse reaction (asthmatics, with history of allergy or those who have had a prior idiosyncratic contrast reaction) were given nonionic agents. Images were acquired helically from diaphragm to upper pole of kidney using (8mm collimation images at a pitch of 1). Renal area was scanned with technique similar to that used for the pre contrast scan. From the lower pole of kidneys to pelvis, 8mm collimated images at a pitch of 1 were obtained. Renal images of excretory phase were taken helically (5mm collimated images at a pitch of 1) after about 5 minutes of beginning of contrast injection.
study In retrospective study, the records of the 9 patients of renal cell carcinoma who had complete data of US, CT, operative and histopathological findings, carried out during the last five years were analysed. US and CT of both prospective and retrospective study were reviewed to evaluate the location and size of renal neoplasms, image characteristics on US and CT including pattern of enhancement after contrast administration, presence of calcification and necrosis, perinephric extension, infiltration of adjacent organs, presence of thrombus in renal vein and inferior vena cava (IVC), retroperitoneal lymphadenopathy and distant metastasis. In addition, staging was done in patients with renal cell carcinoma according to Robson's staging criteria . The findings obtained from imaging were correlated with operative and/or histopathological findings.
Renal cell carcinoma:
In the present study RCC was the commonest renal neoplasm detected and evaluated. A total of 38 cases of RCC were studied. The age at presentation ranged between 19 years to 79 years with mean age of 51 years. Maximum number of patients were seen in the age group of 41-60 years (58%). The male to female ratio was 4.42:1. The presenting complaints in the descending order of frequency were haematuria 71%, pain abdomen 26%, palpable lump 37% and weight loss 13%. Fever, lymphadenopathy, skin nodules and varicocele were noted in one patient each (2.6%) at the time of presentation. The classical triad of haematuria, pain and flank mass was seen only in one patient. The RCC were nearly equally distributed between the right (57.89%) and the left (42.10%) kidney. Thirty percent of the renal masses were in the upper pole, 51.2% in the lower pole and 5.7% in the middle portion of the kidney. In 12.8% of cases almost the entire kidney was replaced by the mass. Lesions of RCC varied in size ranging from 3cm-16cm. Of all the masses representing RCC, 64% were greater than 8cm in size, 33% were 4-8cm and 3% were less than 4cm in size. Sonographically, 37 of the 38 renal cell carcinomas were primarily solid, but 89% had internal cystic components probably due to areas of hemorrhage, necrosis or tumour vascularity. Of the 37 primarily solid tumours, the level of internal echoes was similar to the level of echoes found in liver or normal renal parenchyma in 78.9% [Figure 1]a. Of the primarily solid carcinomas, 13.1% had a level of internal echoes less than that of normal renal parenchyma [Figure 1]b. The level of internal echoes was similar to that of the renal sinus or pararenal fat in 7.8% of the carcinomas [Figure 1]c. Only one of the 38 renal cell carcinomas was primarily cystic but showed a solid component [Figure 1]d. In addition, it had thick irregular wall and septations. On US the perinephric extension of lesion was assessed by the presence of ill defined, indistinct margins of lesion. It was very difficult to evaluate the infiltration of adjacent organs. Liver was seen to be infiltrated by the tumour in 4 patients, spleen in one patient and psoas in three patients. Enlarged lymphnodes were considered significant and likely to have tumour infiltration if they were more than 1cm in size. They were looked for in retroperitoneum and renal hilar location. In two patients retroperitoneum could not be evaluated on US because of bowel gas/obesity. Retroperitoneal lymphadenopathy was present in 15 patients on US.
Colour Doppler flow imaging was undertaken in 29 lesions of RCC. Twenty eight solid renal cell carcinomas showed presence of peripheral and intralesional vascularity. One cystic RCC showed blood flow in the solid nodule within the cyst. The peak systolic Doppler frequency shift was noted for all 29 lesions. The shift ranged from 0.4 KHz to 3.6 KHz., a shift of >2.5 KHz was demonstrated in 34.4% of the RCCs [Figure 1]e.
The CT appearance of RCC vary considerably depending on their pathological features. On non contrast CT 37 of the 38 tumours (94%) appeared solid with necrotic areas. One was cystic in nature. On contrast enhanced scans solid parts of the tumour enhanced markedly but the cystic or necrotic areas showed minimal or no enhancement resulting in inhomogenous appearance [Figure 2]a, b.
There were a total of 5 patients with angiomyolipoma, of these one patient had bilateral renal involvement, 3 patients (60%) had no obvious stigmata of tuberous sclerosis. Two patients (40%) had typical cutaneous manifestations of tuberous sclerosis. There were 3 female patients and two male patients with female to male ratio of 1.5:1. On US the most common pattern of angiomyolipoma was well defined solid hyperechoic mass (80%) [Figure 3]a. One lesion was of heterogenous echogenicity but with a large hyperechoic area. Out of 5 lesions only peripheral vascularity was seen in 3 patients, both peripheral and internal vascularity was seen in 2 lesions. Peak systolic Doppler frequency shift was less than 2.5 KHz in all lesions [Figure 3]b.
Angiomyolipomas were predominantly of heterogenous attenuation with internal fat density areas on unenhanced CT (80%) and showed inhomogenous pattern of contrast enhancement less than normal renal parenchyma [Figure 4]. One lesion was small (1cm), seen as hypodense non-enhancing lesion.
Three patients with renal metastases were seen. One patient was a 60 year old male with primary carcinoma of the lung and bilateral renal metastases.The lesions were hypoechoic on US, hypodense and minimally enhancing after contrast administration on CT [Figure 5].
Second patient was a 40 year old male with a primary malignant nerve sheath tumour. US revealed a large well defined hypoechoic lesion in lower pole of left kidney. On CT, the lesion was well defined isodense to normal kidney on NCCT and showed only minimal enhancement with contrast.
Third patient was a 24 year old female with primary choriocarcinoma of the ovary showing multiple small lesions in the left kidney detected on CT. The patient was otherwise asymptomatic. On US the lesions were well defined and hypo to isoechoic to normal renal parenchyma. The lesions were Oncocytoma:
There was one case of oncocytoma. This patient presented with abdominal pain. He had a solitary mass in the right kidney. The mass was well-defined and homogenous in echotexture with central echogenic stellate scar and central calcification as seen on US. On CT the mass was well-defined, lobulated, homogenous with central hypodense radiating stellate scar and calcification, showing marked homogenous enhancement [Figure 6].
Transitional cell carcinoma
This was seen in 65 year old male patients presenting with haematuria. On US there was a mixed echogenic mass in the dilated renal pelvis of right kidney. The mass was extending into the upper ureter and there was also evidence of selective upper polar caliectasis. On CT the mass was in central location with density greater than that of urine showing minimal enhancement after contrast, centrifugal extension, invasion of renal parenchyma but maintaining the shape of the kidney and extending into upper ureter [Figure 7]. There was no evidence of calcification, lymphadenopathy, renal vein invasion or distant metastases.
Multilocular cystic nephroma
Multilocular cystic nephroma was seen in a 40 year old male patient. The patient was asymptomatic and the mass was detected incidentally on US. The size of the mass was 6.5x9cm. US revealed a well-defined mass with multiple cystic areas separated by echogenic septae [Figure 8]a. On CT the mass appeared as a well marginated, cystic mass with trabeculations and septae [Figure 8]b. There was mild to moderate enhancement of the capsule and septae on post contrast study.
Nephroblastomatosis with Wilm's tumour:
A 25 year old male had bilateral multifocal nephroblastomatosis. The patient was aymptomatic and the lesions were detected incidentally on US. There were multiple solid lesions in both kidneys.
One lesion in right kidney was large about 8cm in size and turned out to be wilm's tumour on fine needle aspiration biopsy, other lesions were smaller (<3cm). On US the lesions were homogenously isoechoic or slightly hypoechoic compared to the normal renal cortex [Figure 9]a. On CT the foci were predominantly isodense to the cortex. After contrast administration they were seen as homogenous hypodense lesions because of poor contrast enhancement [Figure 9]b. Correlation of US and CT findings with histopathological findings in diagnosing perinephric extension.
Out of 38 patients with renal cell carcinoma both operative and histopathological correlation was available in 26 patients. Other patients could not be operated because of advanced disease or some other reason. In these 26 patients, perinephric extension was diagnosed on US in 5 lesions and on CT in 18 lesions. However histopathologically it was present only in 11 lesions. The sensitivity of US in detection of perinephric extension was 45.4% and specificity was 100%. Sensitivity of CT in detection of perinephric extension was 100% and specificity was 53.33%. It was found very difficult to evaluate infiltration of adjacent organs on US. However US was found very useful in evaluation of upper pole lesions for assessing hepatic and splenic infiltration. Infiltration of liver in 2 patients and of spleen in 1 patient by the primary renal lesions were diagnosed on CT however on US there were well maintained fat planes between the lesions and liver and spleen. Histopathologically also in these patients liver and spleen were free of tumour.
CT was found to be 100% sensitive but less specific (specificity=72.7%) in diagnosing the infiltration of adjacent organs with significant yield of false positive results. The predictive value of positive test was only 40%. Only lesions of renal cell carcinoma showed evidence of renal vein and IVC thrombosis. On US renal vein and IVC thrombi were diagnosed when echogenic soft tissue suggestive of thrombus was seen intraluminally [Figure 10]a. Addition of color Doppler was helpful in evaluation of renal vein in patients with equivocal results on B scan ultrasound. On CT the criteria for renal vein and IVC thrombus was presence of intraluminal filling defect with or without dilatation [Figure 11].
Renal vein could not be well evaluated in 2 patients on US because of bowel gas/obesity. In both the patients the lesion was on right side. On CT renal vein could not be well evaluated in 3 patients because of either partial volume averaging with adjacent enlarged lymphnodes or distorted anatomy by large mass lesion. However US was able to demonstrate renal vein in these 3 cases. Renal vein thrombi were detected in total of 5 patients, with renal cell carcinoma. US was 100% sensitive and 100% specific for detection of renal vein thrombus. CT was 80% sensitive and 95% specific for detection of renal vein thrombus. Inferior vena cava thrombi were detected in 5 patients with renal cell carcinoma. Sensitivity and specificity both were 100% for US and CT respectively in detection of IVC thrombi, however upper extent was better demonstrated on US.
US had an advantage over CT in detection of nature of lesion (solid/cystic) and evaluation of renal vein invasion by the lesion. However CT was found more sensitive in evaluation of lesion size and detection of calcification and necrosis within the lesion. The sensitivity of US in detection of calcification was only 40%. CT also had advantages over US in evaluation of perinephric extension, adjacent organ infiltration and regional lymphadenopathy. Both US and CT were equally sensitive in detection of IVC thrombus, however upper extent of thrombus was better demonstrated on US. US was also more sensitive and specific for evaluation of renal vein invasion by the tumour.
Staging was done according to Robson's staging for renal cell carcinoma. Most patients had stage IVa renal cell carcinoma. Out of the 38 patients with renal cell carcinoma, the CT staging was compared with operative staging in 26 patients who were treated surgically.
CT overstaged the lesions in 8 patients because false positive results were high in assessing perinephric extension and infiltration of adjacent organs by the lesion. In one patient CT understaged the lesion because retroperitoneal lymphadenopathy was mistaken as perilesional collateral vessels. Accuracy of CT in staging of renal cell carcinoma in the present study was 65.0%. CT overstaged the neoplasm in 30% of cases.
With the explosion of technologies in diagnostic radiology, more and more asymptomatic renal lesions are being detected at an early stage thereby increasing cure rate and patient survival. Many renal cell carcinomas are now being discovered incidentally during abdominal US and CT done for reason other than suspected renal tumour. When a renal mass is discovered, the next step is to characterize the lesion. Excretory urography is rarely useful for this purpose, but US, CT and percutaneous biopsy all may help in diagnosing renal neoplasms. However, because these studies individually may be negative or indeterminate, various combinations may be necessary for lesion characterization. If a renal neoplasm is discovered by US or IV urography, CT should be performed next for confirmation and staging ,. Diagnosis of renal neoplasms by CT depends mainly on demonstration of significant contrast enhancement ,,,,,. Only significant increases in attenuation value (at least 20 H) indicate true enhancement, because some difference in attenuation value may result from partial volume averaging even on thin CT sections (4,8). Other criteria for diagnosis of neoplasms include central calcification, margin irregularity and inhomogenicity ,,,,,,. Although the aforementioned CT criteria strongly suggest a renal neoplasm, a further imaging study is sometimes necessary for diagnosis, because of the possibility of misinterpretation on CT between high density cysts and neoplasms . US is usually an excellent complementary procedure when CT findings are indeterminate. If US shows a sonolucent lesion with a smooth wall and distal acoustic enhancement, a high density cyst can be diagnosed confidently. Conversely, if the mass shows internal echoes, mural nodules and no distal acoustic enhancement, the diagnosis of neoplasm is likely. Occasionally, a haemorrhagtic cyst is echogenic on US and differentiation from a neoplasm is then not possible. Fine needle aspiration biopsy may be confirmatory in such patients .
As in other studies RCC was the most common lesion of our series of 50 neoplasms accounting for 76% of cases. Sonographically nearly 95% of the RCCs were primarily solid. 78.9% of the solid tumours were isoechoic to normal renal parenchyma, 13.1% were hypoechoic and 7.8% were hyperechoic. Our findings were essentially similar with the previous reports ,,,,,,,,,,,,. While none of the echogenic carcinomas was as echogenic as the typical angiomyolipoma, particular care must be taken not to misinterpret these as angiomyolipoma. A peripheral hypoechoic halo, intratumoral cyst and absence of posterior shadowing are some of the important characteristics US features described in the literature to differentiate hyperechoic RCC from angiomylipoma ,,. A CT scan of these echogenic masses should differentiate low density fat of an angiomyolipoma from the higher density of RCC ,. Other masses that have been reported to be highly echogenic are renal cavernous haemangioma, renal infarct, angiosarcoma, undifferentiated sarcoma and metastatic neoplasms ,,. Four of the five angiomyolipomas in our study were highly echogenic. Several possibilities have been suggested for this high echogenicity, including the high fat content, the heterogenous architecture, and the numerous blood vessels within these tumors.
Renal oncocytomas are benign tumours that arise from the proximal tubular epithelium and have characteristic histologic and electron microscopic features . Distinction between oncocytomas and renal cell carcinomas has potential therapeutic implications because an oncotytoma may be managed by partial nephrectomy, whereas radical nephrectomy is usually preferred for renal cell carcinoma . A central radiating stellate scar which is echogenic on US and hypodense and non-enhancing on CT is one of the important characteristic feature to differentiate oncocytoma from RCC. However, very small oncocytomas usually cannot be distinguished from small renal carcinomas by CT, angiography, or percutaneous biopsy ,.
US has an advantage over CT in detection of nature of the lesion (solid/cystic) and evaluation of renal vein invasion by the lesion. CT is more sensitive in evaluation of lesion size and detection of calcification and necrosis within the lesion. CT also has an advantage over US in evaluation of perinephric extension, adjacent organ infiltration and regional lymphadenopathy. Both US and CT are equally sensitive in detection of IVC thrombus, however upper extent of thrombus is better demonstrated on US.
To conclude, it is most often possible to characterize a variety of renal neoplasms on US, CDFI and CT, based on the image morphology. It is recommended that US should be the screening modality in patients suspected of renal neoplasms and CT should be used for further characterising the masses and for staging of the malignant tumours. However CT has its limitation in differentiating stage I lesions from stage II and stage IVa lesions. US has a definite advantage in evaluation of renal vein invasion by the lesion and it shows better demonstration of relationship of upper polar lesions with liver and spleen. Percutaneous biopsy may be helpful in the diagnosis of suspicious lesions.
|1||Beahrs OH, Henson DE, Hutter RVP et al. Manual for staging of cancer -American Joint Committee on Cancer, 4th ed Philadelphia. JB Lippncott, 201-202, 1992. |
|2||Asher WM, Leopold GR. A streamlined diagnostic approach to renal mass lesions with renal echogram. J Urol 1972; 108 : 205-208. |
|3||Einstein DM, Herts BR, Wearer R, Obuchowski N, Zepp R, Singer A. Evaluation of renal masses detected by excretory urography : cost-effectiveness of sonography versus CT. AJR 1995; 164 : 371-375.|
|4||Cohan HK, Sherman LS, Korobkin M, Bass JC, Francis IR. Renal masses: assessment of corticomedullary-phase and nephrographic-phase CT scans. Radiology 1995; 196 : 445-451. |
|5||Curry NS, Schabel SI, betsill WL. Small renal neoplasms : diagnostic imaging, pathologic features and clinical course. Radiology 1986; 158 : 113-117.|
|6||Pollack HM, Banner MP, Arger PH, Goldberg BB, Mulhern CB. Comparison of computed tomography and ultrasound in the diagnosis of renal masses. Clin Diagn Ultrasound 1979; 2 : 25-72.|
|7||Fien AB, Lee JKT, Balfe DM, Heiken JP, Ling D, Glazer HS, McClennan BL. Diagnosis and staging of renal cell carcinoma: A comparison of MR imaging and CT. AJR 1987; 148 : 749-753|
|8||Silverman SG, Lee BY, Seltzer SE, Bloom DA, Corless CL, Adams DF. Small (<3 cm) renal masses: correlation of spiral CT features and pathologic findings. AJR 1994; 163 : 597-605. |
|9||Amendola MA, Bree RL, Pollack HM et al. Small renal cell carcinomas: resolving a diagnostic dilemma. Radiology 1988; 166 : 637-641.|
|10||Hartman DS, Davidson AJ, Davis CJ, Goldman SM. Infiltrative renal lesions: CT-sonographic-pathologic correlation. AJR 1988; 150 : 1061-1064.|
|11||Curry NS. Imaging the small solid renal mass. Abdom Imaging 2002; 27:629-636. |
|12||Herts BR. Imaging for renal tumours. Curr Opin Urol. 2003 May;13(3):181-186. Review. |
|13||Israel GM, Bosniak MA. Renal imaging for diagnosis and staging of renal cell carcinoma. Urol Clin North Am. 2003 Aug;30 (3):499-514. Review.|
|14||Walter C, Kuressell M, Grindele A, Brochhagen HG, et all. Imaging of renal lesions: evaluation of fast MRI and helical CT. Br J Radiol. 2003; Oct; 76(910): 696-703.|
|15||Hauser M, Krestin GP, Hagspiel KD. Bilateral solid multifocal intrarenal and perirenal lesions: differentiation with ultrasonography, computed tomography and magnetic resonance imaging. Clinical Radiology 1995; 50 : 288-294.|
|16||Foster WL, Roberts L, Halvorsen RA, Dunnick NR. Sonography of small renal masses with indeterminant density characteristics of computed tomography. Urol Radiol 1988; 10 : 59-67. |
|17||Levine E, Huntrakoon M, Wetzel LH. Small renal neoplasms : Clinical, pathologic and imaging features. AJR 1989; 153 : 69-73.|
|18||Barnett E, Morley P. Ultrasound in the investigation of space-occupying lesions of the urinary tract. Br J Radiol 1971; 44 : 733-742.|
|19||Romeiser RS, Walls WJ, Valk WL. B-scan ultrasound in the evaluation of renal mass lesions. J Urol 1974; 112 : 8-12.|
|20||Charboneau JW, Hattery RR, Ernst EC, James EM, Williamson B, Hartman GW. Spectrum of sonographic findings in 125 renal masses other than benign simple cyst.AJR 1983; 140 : 87-94. |
|21||Bloom JN, Mattey WE, Arevalo FL, DelGuercic LRM. B-mode ultrasound scanning in the diagnosis of renal lesions. Am J Surg 1975; 129 : 636-641.|
|22||Lee TG, Henderson SC, Freeny PC, Raskin MM, Benson EP, Pearse HD. Ultrasound findings of renal angiomyolipoma. JCU 1978; 6 : 150-155. |
|23||Siegel CL, Middleton WD et al. Angiomyolipoma and renal cell carcinoma: US differentiation. Radiology 1996; 198 : 789-793.|
|24||Hartman DS, Goldman SM, Friedman AC, Davis CJ Jr, Madewell JE, Sherman JL. Angiomyolipoma: ultrasonic-pathologic correlation. Radiology 1981; 139 : 451-458.|
|25||Stanley RJ, Cubilo E, Mancilla-Jimenez R. Geisse G, Melson GL. Cavernous hemangioma of the kidney. AJR 1976; 125 : 682-687. |
|26||Lieber MM, Tomera KM, Farrow GM. Renal oncotyoma. J Urol 1981; 125 : 481-485. |
|27||Cohan RH, Dunnick NR, Degesys GE, Korobkin M. Computed tomography of renal oncocytoma. J Comput Assist Tomogr 1984; 8 : 284-287.|