Segmental Enhancement Inversion of Small Renal Oncocytoma: Differences in Prevalence According to Tumor Size

Transcription

1 Genitourinary Imaging Original Research Woo et al. Renal Oncocytoma Genitourinary Imaging Original Research Sungmin Woo 1 Jeong Yeon Cho 1,2 Seung Hyup Kim 1,2 Sang Youn Kim 1 Hak Jong Lee 3 Sung Il Hwang 3 Min Hoan Moon 4 Chang Kyu Sung 4 Woo S, Cho JY, Kim SH, et al. Keywords: CT, pathologic change, renal oncocytoma, segmental enhancement inversion, tumor size DOI: /JR Received May 23, 2012; accepted after revision July 5, Department of Radiology, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul , Korea. ddress correspondence to J. Y. Cho (radjycho@snu.ac.kr). 2 Institute of Radiation Medicine and Kidney Research Institute, Seoul National University Medical Research Center, Seoul, Korea. 3 Department of Radiology, Seoul National University undang Hospital, undang, Seongnam, Gyeonggi-do, Korea. 4 Department of Radiology, Seoul Metropolitan oramae Medical Center, Seoul, Korea. JR 2013; 200: X/13/ merican Roentgen Ray Society Segmental Enhancement Inversion of Small Renal Oncocytoma: Differences in Prevalence ccording to Tumor Size OJECTIVE. The purpose of this study was to retrospectively assess the prevalence of segmental enhancement inversion of small renal oncocytomas according to tumor size. MTERILS ND METHODS. Thirty-three patients (19 men, 14 women; mean age, 61 years; range, years) with 33 oncocytomas diagnosed at surgical resection who had undergone contrast-enhanced biphasic CT between January 2000 and December 2011 were included. CT scans were analyzed by two radiologists blinded to the specifics of the pathology report for size, presence of segmental enhancement inversion, enhancement pattern, and homogeneity. Segmental enhancement inversion was present when a renal mass was divided into two differently enhanced segments in the corticomedullary phase (30 40 seconds after contrast injection) with the degree of enhancement reversed in the nephrographic phase ( seconds after contrast injection). The masses were further assessed for fibrous septa, cystic change, hemorrhage, and necrosis. For statistical analysis, the Pearson chi-square test and linear regression were used to evaluate the relation between the prevalence of segmental enhancement inversion and tumor size or pathologic changes. RESULTS. The mean diameter of 33 renal oncocytomas was 2.65 cm (range, cm). There was no significant linear trend according to size (p = 0.762), although segmental enhancement inversion was significantly (p = 0.006) more common (10/12) in tumors measuring cm. Pathologic change was present in 14 oncocytomas. There was no significant linear trend according to size (p = 0.068), but 2.5-cm and larger tumors had a significantly higher prevalence (57.9%) (p = 0.036). Segmental enhancement inversion was more common (13/19) in tumors without pathologic change (p = 0.024). CONCLUSION. Segmental enhancement inversion was a characteristic finding in our series of small renal oncocytomas and was more common in tumors measuring cm. Pathologic changes such as central scar were more common in oncocytomas larger than 2.5 cm and may be related to the low occurrence of segmental enhancement inversion. R enal oncocytoma is the second most common benign renal tumor. It originates from epithelial cells of the proximal renal tubule and constitutes approximately 3 7% of all renal tumors [1, 2]. Oncocytoma is a generally benign tumor, unlike renal cell carcinoma, which is characterized by renal vein invasion, postoperative recurrence, and distant metastasis [3]. Radiologic characteristics that could be used to noninvasively differentiate oncocytoma from renal cell carcinoma would be invaluable. Reports of imaging findings have characterized oncocytoma with well-differentiated margins and homogeneous enhancement without hemorrhage, calcification, or necrosis; the presence of a central stellate scar; and a spoke-wheel pattern of arterial enhancement [4 7]. These findings, however, overlapped considerably with those of renal cell carcinoma [7 9]. We previously reported [10] that the presence of segmental enhancement inversion may be helpful in identifying renal oncocytoma. We defined segmental enhancement inversion as the contrast-enhanced biphasic renal CT finding of two distinct regions of enhancement of a renal mass in the corticomedullary phase in which the degree of enhancement reverses in the nephrographic phase. We found segmental enhancement inversion in 8 of 10 (80%) oncocytomas with a diameter of 4 cm or less. In a more recent article, McGahan et al. [11] disagreed with our report. In their retrospective study, only 1 of 16 (6.25%) renal onco JR:200, May 2013

2 Renal Oncocytoma TLE 1: Prevalence of Segmental Enhancement Inversion in Renal Oncocytomas by Size Tumor Size (cm) Segmental Enhancement Inversion Prevalence < 1 0 0/ / / / /4 Total 17 17/33 Note Values are numbers of tumors. cytomas exhibited segmental enhancement inversion. We conducted this study to elucidate the cause of the differing results. In our more recent clinical practice, we noticed that the prevalence of segmental enhancement inversion was lower in larger renal oncocytomas. The purpose of this study was to retrospectively assess the different prevalence of segmental enhancement inversion of small renal oncocytomas according to tumor size. Materials and Methods Patients This retrospective study was approved by the institutional review board; the requirement for informed consent in this records-based study was waived. computerized search was performed of medical records at two institutions from January 2000 to December 2011 for surgically confirmed oncocytomas 5 cm or smaller for which the patient underwent preoperative biphasic contrast-enhanced MDCT. total of 33 patients (19 men, 14 women; mean age, 61 years; range, years) with 33 renal oncocytomas were included. Of these 33 patients, 10 patients had been enrolled in our previous study [10]. CT Examinations CT examinations were performed with the following scanners: one with a single-detector scanner (Genesis HiSpeed RP, GE Healthcare), four with a 4-MDCT scanner (MX8000, Philips Healthcare; LightSpeed Plus, GE Healthcare; Somatom Plus 4, Siemens Healthcare), one with a 6-MDCT scanner (rilliance 6, Philips Healthcare), seven with an 8-MDCT scanner (LightSpeed Ultra, GE Healthcare), 11 with a 16-MDCT scanner (MX8000 IDT 16, Philips Healthcare; Sensation 16, Siemens Healthcare; LightSpeed 16 and LightSpeed 16 Pro, GE Healthcare), eight with a 64-MDCT scanner (rilliance 64, Philips Healthcare; Sensation 64, Siemens Healthcare), and one with a 320-MDCT scanner (quilion ONE, Toshiba). pproximately 2 ml/kg and up to a maximum volume of 150 ml of IV contrast material (iopromide, Ultravist 350 or 370, ayer Schering Pharma) was injected at a rate of 3.0 ml/s with a power injector. ll patients underwent biphasic contrast-enhanced CT that included unenhanced, corticomedullary phase, and nephrographic phase scans. Corticomedullary phase images were acquired seconds and nephrographic phase images seconds from the start of the injection. The scanning parameters varied for the different scanners, but all had a tube voltage of 120 kvp (for all phases), tube current of ms (depending on patient size), and section thickness of mm. Image nalysis One genitourinary radiologist with 19 years experience and one radiologist in training who were aware of the diagnosis of renal oncocytoma but blinded to the specifics of the pathology report reviewed the CT images in consensus at a PCS workstation that had a flat-panel monochrome 3-megapixel monitor (ME 511 L, Totoku Electric). The radiologists measured tumor size according to maximum diameter and categorized the oncocytomas into five groups: < 1 cm, cm, cm, cm, and cm. The presence of segmental enhancement inversion also was evaluated. efore reaching a consensus, the two radiologists independently evaluated whether segmental enhancement inversion was present in the tumor. Images were then reviewed together for final decision about the presence of segmental enhancement inversion. Segmental enhancement inversion was determined to be present when the renal oncocytoma was divided into two differently enhanced segments on corticomedullary phase images but the relative degree of enhancement was reversed on the nephrographic phase images [10]. For each CT image, the window width and level were adjusted for better visualization of the two differently enhancing regions, if present, in each renal oncocytoma. For measurement of attenuation, regions of interest (ROIs) were set in areas having the greatest degree of enhancement difference on axial or coronal images. Each ROI cursor, round or elliptic, was placed carefully to include most of the area of each segment but without involving adjacent renal parenchyma, encompassing a mean area of 1.97 cm 2 (range, cm 2 ). The size and location of the ROIs were identical on images acquired in all scan phases. If segmental enhancement inversion was not clearly visualized, the radiologists checked the enhancement pattern and homogeneity. The enhancement pattern was categorized as suggested by Kim et al. [10]. Early washout was defined as peak enhancement in the corticomedullary phase and washout of at least 20 HU in the nephrographic phase, gradual enhancement as more than 20 HU of increased attenuation in the nephrographic phase than in the corticomedullary phase, and Fig year-old woman with 2.1-cm right renal oncocytoma that exhibited segmental enhancement inversion., Corticomedullary phase CT image obtained 30 seconds after contrast injection shows triangular region (arrowhead) that is less enhanced than rest of tumor (arrows)., Nephrographic phase CT image obtained 180 seconds after contrast injection shows triangular region (arrowhead) is more enhancing than rest of tumor (arrows). JR:200, May

3 Woo et al. prolonged enhancement as an attenuation difference between the corticomedullary and nephrographic phases of 20 to 20 HU. Homogeneity of tumor enhancement was subjectively evaluated by means of visual inspection on both corticomedullary and nephrographic phase images separately. The tumor was defined as having a homogeneous enhancement pattern when most of the mass had relatively similar degrees of enhancement. Otherwise, heterogeneous enhancement was considered present. When a heterogeneous enhancement pattern was seen, the two radiologists further evaluated the images for scar and necrosis. Scar and necrosis were defined as well-demarcated areas having decreased attenuation, as suggested by McGahan et al. [11]. C Fig year-old woman with 3.8-cm left renal oncocytoma in which segmental enhancement inversion was better depicted on coronal CT images. and, xial corticomedullary () and nephrographic () phase images do not clearly show segmental enhancement inversion. C and D, Coronal corticomedullary (C) and nephrographic (D) phase images clearly show two segments (arrow, arrowhead) with inversion of enhancement. In placement of the ROI cursor for measurement of degree of enhancement, the area that appeared as an enhancing solid area on corticomedullary phase images was selected. Consistent round or elliptic ROIs of approximately 1.85 cm 2 (range, cm 2 ) were placed on enhancing solid areas on unenhanced, corticomedullary phase, and nephrographic phase scans. D Pathologic nalysis Two pathologists (urologic pathologists with 8 and 22 years experience) cut the surgically resected kidneys sequentially in 5- to 8-mm sections and stained them with H and E. Immunohistochemical staining was performed with renal cell carcinoma antigen, c-kit, epithelial membrane antigen, vimentin, cytokeratin 7, cytokeratin 20, E- cadherin, CD10, and S-100. The pathologists evaluated the masses for the presence of the following four specific pathologic changes: fibrous septa formation, cystic change, hemorrhage, and necrosis. Statistical nalysis Statistical analysis was performed with SPSS software (version 17.0, SPSS). The Pearson chisquare test and linear regression were used to evaluate the relation between the prevalence of segmental enhancement inversion and tumor size or pathologic changes. value of p < 0.05 was considered to indicate a significant difference. Results Imaging Findings The mean diameter of the 33 renal oncocytomas was 2.65 ± 1.18 [SD] cm (range, cm). The number of tumors and the corresponding prevalence of segmental enhancement inversion in each size group are shown in Table 1. Seventeen of 33 (51.5%) renal oncocytomas exhibited segmental enhancement inversion (Fig. 1). In two cases (Fig. 2), the inversion pattern was difficult to perceive on axial images but was clearly visualized on coronal images. In six cases the two segments with enhancement inversion were arranged in a targetlike pattern (Fig. 3). There was only one case in which consensus was required because of discrepancy between the two radiologists regarding the presence of segmental enhancement inversion at initial evaluation. In this case (Fig. 4), the mass was divided into two differently enhancing segments in the corticomedullary phase and became relatively homogeneous in the nephrographic phase. However, at visual inspection under careful adjustment of window width and level and by quantitative measurement with an ROI, the enhancement degree of corresponding segments was proved to be reversed and was finally determined to be segmental enhancement inversion. lthough differences between the attenuation of more- and less-enhanced segments in all cases were positive during the corticomedullary phase (54.88 ± HU), they were negative in the nephrographic phase ( ± HU). In seven oncocytomas, although one segment exhibited a progressive enhancement pattern, the other exhibited a persistent enhancement pattern. In six oncocytomas, one segment had a progressive enhancement pattern, and the other exhibited washout. In one oncocytoma, although one segment exhibited washout, the other was persistent. oth segments exhibited 1056 JR:200, May 2013

4 Renal Oncocytoma washout in two oncocytomas and progressive enhancement in one oncocytoma. There was no significant linear trend in the prevalence of segmental enhancement inversion according to tumor size (p = 0.762). However, comparison of oncocytomas with a diameter of cm (10/12 [83%]) with those smaller than 1.5 cm or larger than 3 cm (7/21 [33%]) showed a significant difference (p = 0.006). In renal oncocytomas without segmental enhancement inversion, the following features were observed: six oncocytomas with persistently homogeneous enhancement in the corticomedullary and nephrographic phases, seven Fig year-old woman with 1.2-cm left renal oncocytoma with targetlike segmental enhancement inversion., Corticomedullary phase CT image shows central dotlike lesion (arrowhead) is more enhancing than rest of tumor (arrow)., Nephrographic phase CT image shows central dotlike lesion (arrowhead) is less enhancing than rest of tumor (arrow). with heterogeneous enhancement in the corticomedullary phase but closer to homogeneous enhancement in the nephrographic phase, and three with scar or necrosis (Fig. 5). Pathologic Findings With Radiologic Correlation The type and prevalence of pathologic changes in each oncocytoma by size group are summarized in Table 2. Fourteen of 33 (42%) oncocytomas had at least one type of pathologic change. In total, hemorrhage was the most common pathologic change (eight cases) followed by cystic change (five cases) and fibrous septa formation (five cases). There was no significant linear trend of segmental enhancement inversion prevalence by tumor size (p = 0.068). However, oncocytomas larger than 2.5 cm in diameter had a significantly higher prevalence (57.9%) than those smaller than 2.5 cm (21.4%) (p = 0.036). Stellate scar was not found in oncocytomas smaller than 2.5 cm but was observed in 5 of 19 (26.3%) oncocytomas larger than 2.5 cm. These differences were significant (p = 0.049). Moreover, segmental enhancement inversion was significantly more common (68%) in the 19 tumors without pathologic changes than in the 14 tumors with pathologic changes (28%) (p = 0.024). Discussion It has been previously reported that % of small renal masses are found to be benign at histologic analysis [12 14]. Furthermore, many reports have suggested that parenchyma-sparing surgery such as partial nephrectomy is sufficient for such benign renal masses and that radical nephrectomy would only increase the risk of renal impairment [12, 15]. Therefore, CT characteristics that differentiate oncocytoma, the second most common benign renal tumor, from renal cell carcinoma would be invaluable. Several authors have reported that renal oncocytoma has the following classic angiographic characteristics: a spokewheel pattern, a homogeneous nephrogram, and a sharp, smooth rim [6, 16 18]. They have also reported, however, that renal cell carcinomas may have similar findings [6]. Many reports have suggested a characteristic CT finding of oncocytoma. Some reports suggest that homogeneous renal masses larger Fig year-old man with 3.4-cm left renal oncocytoma. Subtle segmental enhancement inversion was perceived only after careful window adjustment., Corticomedullary phase CT image at usual window width and level shows mass composed of less-enhancing central segment (arrowhead) and more-enhancing peripheral segment (arrow)., Nephrographic phase CT image at usual window width and level shows mass is almost homogeneous. C, Nephrographic phase CT image at narrowed width shows subtle segmental enhancement inversion between center (arrowhead) and periphery (arrow). C JR:200, May

5 Woo et al. than 3 cm with a well-defined stellate area of hypoattenuation, termed a central scar, may be indicative of oncocytoma [4, 19]. Others suggest that there is considerable overlap between oncocytoma and renal cell carcinoma and that imaging characteristics alone are not sufficient for diagnosis [7 9]. We reported [10] that the CT segmental enhancement inversion pattern is sensitive (80%) and specific (99%) for the diagnosis of renal oncocytomas smaller than 4 cm. We also found correlation between differentially enhanced segments on contrast-enhanced CT images and microscopic architectural patterns in pathologic specimens. The more enhanced segment correlated with compactly arranged tumor cells with scarcely intervening stroma, and the less enhanced segments correlated with abundant hyalinized stroma. McGahan et al. [11] disagreed with our report, finding segmental enhancement inversion in 1 of 16 renal oncocytomas. The most common feature they found was slight heterogeneous enhancement in the corticomedullary phase and closer to homogeneous enhancement in the nephrographic phase in 8 of 16 cases. Two readers in a study by Rosenkrantz et al. [20] found segmental enhancement inversion in eight and 12 of 28 oncocytomas but saw no significant difference from chromophobe renal cell carcinoma. Given this background, we performed our study with a larger sample size to see whether segmental enhancement inversion is indeed a good indicator of oncocytoma and to clarify differences TLE 2: Prevalence of Pathologic Change in Renal Oncocytomas by Size Tumor Size (cm) Fig year-old woman with 3-cm left renal oncocytoma without segmental enhancement inversion but with central necrosis., Transverse corticomedullary phase CT image shows tumor is relatively homogeneously enhancing except for tiny nonenhancing cystic area (arrow)., Transverse nephrographic phase CT image shows tumor has continuously homogeneous enhancement except for tiny nonenhancing cystic area (arrow). Pathologic Change Fibrous Septa Cystic Change Hemorrhage Necrosis among studies. mong the 33 patients with renal oncocytoma, 10 patients had been enrolled in our previous study. In the current study, 17 of 33 (51.5%) renal oncocytomas had segmental enhancement inversion, in agreement with our previous finding that segmental enhancement inversion is characteristic of small renal oncocytoma [10]. Most common (10/12 [83%]) were those measuring cm (p = 0.006). There are several possible explanations for the discrepancy between our findings and those of McGahan et al. [11]. First, the definitions of segmental enhancement inversion differed. McGahan et al. limited segmental enhancement inversion to cases with distinct enhancing segments and a noncentral pattern. In six cases in our study, however, the two segments with enhancement inversion were arranged in a targetlike pattern. Second, there was a CT protocol difference. Our delay time for the corticomedullary phase (30 40 seconds) was shorter than that used by McGahan et al. (40 52 seconds). Moreover, the delay for the nephrographic phase ( seconds) was longer than that used by McGahan et al. ( seconds). The longer interval between the corticomedullary and nephrographic phases in our series may have increased the prevalence of segmental enhancement inversion. Third, in one case in our study, although the segmentation of enhancement in the corticomedullary phase was definite, the mass became relatively homogeneous in the nephrographic phase. However, with careful adjustment of window width and level, we visualized a subtle degree of reversal in enhancement, and measurement with an ROI confirmed the presence of segmental enhancement inversion. These findings may have been classified as heterogeneous masses that became closer to homogeneous in the study by McGahan et al. Fourth, we noted in two cases that the segmental enhancement inversion was not observed on axial images but was clearly visualized on coronal images. Change Rate < / / / / /4 Total /33 Note Values are numbers of tumors JR:200, May 2013

6 Renal Oncocytoma lthough McGahan et al. performed CT examinations with reconstruction in the axial, sagittal, and coronal planes, they did not comment on multiplanar image interpretation. ny of these factors may have contributed to their lower prevalence of segmental enhancement inversion compared with ours. Rosenkrantz et al. [20] reported a higher prevalence ( %) of segmental enhancement inversion than that found by Mc- Gahan et al. (6.3%) and closer to that (51.5%) in our study. That difference can be partially explained by the difference in size of the tumors. The tumors in the study by Rosenkrantz et al. (3.51 ± 1.86 cm) were larger than those in our study (2.65 ± 1.18 cm). Furthermore, our results suggest that oncocytomas larger than 2.5 cm tend to have more pathologic changes (especially central scar) and that the presence of pathologic changes is inversely correlated with that of segmental enhancement inversion. Therefore, the larger tumors in the study by Rosenkrantz et al. may have been associated with a larger proportion of tumors with pathologic changes. Regarding the low prevalence of segmental enhancement inversion in tumors smaller than 1.5 cm, we speculate that the small size of the tumor may have limited the visual segmentation into a more- or less-enhancing region. There were several limitations to this study. First, the number of renal oncocytomas evaluated was relatively small. However, the number of cases in our series, 33, was larger than those in the studies by McGahan et al. [11], 16, and Rosenkrantz et al. [20], 28. To our knowledge ours is the largest study of small renal oncocytomas. Second, although CT image evaluation was done only in the axial, coronal and sagittal planes, pathologic examination was done in numerous planes, limiting radiologicpathologic correlation. However, we found in a previous study [10] that segments more and less enhanced in the corticomedullary phase correlate with compactly arranged tumor cells and hyalinized stroma, respectively. Third, owing to the retrospective approach of our study, there was variation in CT scanners and protocols. However, to include a sufficient number of patients with renal oncocytoma, a long study period, during which there was remarkable improvement in technology, was inevitable. Fourth, although the radiologists were blinded to the four types of pathologic changes, they were aware that the lesions were pathologically proven renal oncocytoma. ecause this may have led to the possibility of overclassification of renal oncocytomas as positive for segmental enhancement inversion, we analyzed segmental enhancement inversion in a quantitative manner, which is an objective method of evaluating segmental enhancement inversion. fifth limitation was that we did not perform a comparative study with renal cell carcinoma. Reevaluating the sensitivity and specificity of segmental enhancement inversion was not the aim of this study because we had evaluated it previously [10]. Finally, we did not analyze interobserver concordance because all image interpretation was based on consensus. However, consensus was reached after independent evaluation, and only one case had discrepant results at initial evaluation. Moreover, this discrepancy was due to the level of perceiving segmental enhancement inversion, which we speculate was attributed to the differing results among previous studies rather than limited value of segmental enhancement inversion as a finding of small renal oncocytoma. Conclusion Segmental enhancement inversion was a characteristic CT finding in our series of small renal oncocytomas and was more common in tumors measuring cm. Pathologic changes such as central scar were more common in oncocytomas larger than 2.5 cm and may explain the lower prevalence of segmental enhancement inversion in larger tumors. We believe that our results may contribute to establishing guidelines for the management of small renal tumors in that preoperative identification of a benign tumor such as renal oncocytoma may help avoid unnecessary surgical resection through the use of observation after pathologic confirmation in a specimen obtained at percutaneous biopsy. Future prospective studies of small renal oncocytoma with respect to tumor characterization of CT findings of segmental enhancement inversion, the diagnostic capability of percutaneous biopsy, and long-term follow-up of patients who undergo observation are warranted. References 1. Perez-Ordonez, Hamed G, Campbell S, et al. Renal oncocytoma: a clinicopathologic study of 70 cases. m J Surg Pathol 1997; 21: Lieber MM. Renal oncocytoma. Urol Clin North m 1993; 20: Kovacs G, khtar M, eckwith J, et al. The Heidelberg classification of renal cell tumours. J Pathol 1997; 183: Quinn MJ, Hartman DS, Friedman C, et al. Renal oncocytoma: new observations. Radiology 1984; 153: Jasinski RW, mendola M, Glazer GM, ree RL, Gikas PW. Computed tomography of renal oncocytomas. Comput Radiol 1985; 9: mbos M, osniak M, Valensi QJ, Madayag M, Lefleur RS. ngiographic patterns in renal oncocytomas. Radiology 1978; 129: Davidson J, Hayes WS, Hartman DS, McCarthy WF, Davis CJ Jr. Renal oncocytoma and carcinoma: failure of differentiation with CT. Radiology 1993; 186: Choudhary S, Rajesh, Mayer NJ, Mulcahy K, Haroon. Renal oncocytoma: CT features cannot reliably distinguish oncocytoma from other renal neoplasms. Clin Radiol 2009; 64: Zhang J, Lefkowitz R, Ishill NM, et al. Solid renal cortical tumors: differentiation with CT. Radiology 2007; 244: Kim JI, Cho JY, Moon KC, Lee HJ, Kim SH. Segmental enhancement inversion at biphasic multidetector CT: characteristic finding of small renal oncocytoma. Radiology 2009; 252: McGahan JP, Lamba R, Fisher J, et al. Is segmental enhancement inversion on enhanced biphasic MDCT a reliable sign for the noninvasive diagnosis of renal oncocytomas? JR 2011; 197:[web]W674 W Duchene D, Lotan Y, Cadeddu J, Sagalowsky I, Koeneman KS. Histopathology of surgically managed renal tumors: analysis of a contemporary series. Urology 2003; 62: Neuzillet Y, Lechevallier E, ndre M, Daniel L, Coulange C. ccuracy and clinical role of fine needle percutaneous biopsy with computerized tomography guidance of small (less than 4.0 cm) renal masses. J Urol 2004; 171: Jaff, Molinie V, Mellot F, Guth, Lebret T, Scherrer. Evaluation of imaging-guided fineneedle percutaneous biopsy of renal masses. Eur Radiol 2005; 15: McKiernan J, Yossepowitch O, Kattan MW, et al. Partial nephrectomy for renal cortical tumors: pathologic findings and impact on outcome. Urology 2002; 60: Weiner SN, ernstein RG. Renal oncocytoma: angiographic features of two cases. Radiology 1977; 125: Sos T, Gray GF Jr, altaxe H. The angiographic appearance of benign renal oxyphilic adenoma. JR 1976; 127: Levine E, Huntrakoon M. Computed tomography of renal oncocytoma. JR 1983; 141: Eiss D, Larousserie F, Mejean, et al. Renal oncocytoma: CT diagnostic criteria revisited. J Radiol 2005; 86: Rosenkrantz, Hindman N, Fitzgerald EF, Niver E, Melamed J, abb JS. MRI features of renal oncocytoma and chromophobe renal cell carcinoma. JR 2010; 195:[web]W421 W427 JR:200, May