Introduction: Renal cell carcinoma (RCC) accounts for 3% of all human malignant tumors [1,2]. In Europe alone, there were an estimated 88,400 new cases of RCC and 39,300 deaths from RCC in 2008 . With the exception of Denmark and Sweden, the incidence of this tumor type has been, until recently, increasing worldwide at an annual rate of 2% .
Currently, approximately 90% of renal malignant diseases are considered to be a form of RCC . RCC is recognized as a group of neoplasms with different cytogenetic abnormalities and histopathologic features [3,4]. As such, the World Health Organization (WHO) has proposed three major distinct histologic subtypes of RCC in the Classification of the Renal Tumors of the Adults (2004): clear cell (ccRCC), which is the predominant type accounting for 80’90% of RCCs; papillary (pRCC), which accounts for 10’15% of cases; and chromophobe (chRCC), which accounts for 4’5% of cases [3-7]. Papillary carcinoma can be further divided into two subtypes, type I and type II, the latter of which carries a worse prognosis [6, 7]. Aformentioned RCC subtypes have a diverse range of malignant potential, which is associated with significant differences in prognosis, e.g. pRCC is less malignant than ccRCC specifically. Consequently information about RCC subtype can be useful in the treatment planning [3-10].
Oncocytoma represents approximately 5% of all renal tubular epithelium neoplasms and 5 to 15% of surgically removed kidney tumors . Oncocytoma is considered the most common benign solid renal neoplasm, the majority is asymptomatic at presentation [3, 11]. Consequently, oncocytomas are usually incidental findings during diagnostic workup of unrelated conditions. On postcontrast CT images typical radiographic finding for oncocytoma is central hypodense zone, so-called central scar .
The purpose of this study was to evaluate the influence of different tumor enhancement measurement approaches in the form of tumor attenuation values and tumor enhancement ratios for the differentiation between solid forms of ccRCC subtypes, non-clear cell RCCs and onocytoma. The differences in attenuation values between solid forms of ccRCC and other histologic subtypes of RCC (papillary, chromophobe, and collecting duct RCC) and oncocytoma measured using two different techniques were retrospectively assessed. This research was performed in order to define the most accurate and simple method of tumor enhancement measurement that could be used as an auxiliary method for pretreatment differentiation of solid forms of the most frequent RCC subtypes and benign tumors in patients with malignant tumors who are not suitable for imaging-guided core biopsy. Results of research studies about differentiation between the most frequent RCC subtypes using minimally invasive imaging methods are not unimportant. The clinicians could use this technique as an auxiliary tool in decision making about the application of neoadjuvant therapy in patients who are not suitable for invasive diagnostics (core biopsy), surgical therapy, and/or invasive interventional methods because differences in RCC subtypes structure and consequently malignant potential have important influence on the treatment response.
Patients and methods
This study was performed at two institutions to ensure sufficient patient enrollment. The institutional review boards permission was granted for a retrospective review of imaging and pathologic data archived in hospital information systems .
A computerized search of the imaging and pathologic databases at both institutions revealed 146 consecutive patients with tumors of the renal parenchyma who underwent preoperative CT imaging of the kidneys and partial or radical nephrectomy between May 2001 and May 2011. Medical records of patients with cystic forms of RCC (2 patients), malignant mesenchymal renal tumors (one patient with primitive neuroectodermal tumor-PNET ) and patients who did not undergo multiphasic CT scanning (30 patients) were excluded from the study group. Medical records of 113 patients were included in the study group. Of these 113 patients, 65 were men (mean age, 62 years; range 23’80 years), of whom 110 patients had one tumor (one patient with RCC), one patient had two tumors, and two patient had three tumors at the time of detection (one patient with ccRCC and one patient with oncocytoma).
According to the pathologic findings in the 106 patients, 97 tumors (92%) were diagnosed as ccRCC (Fig. 1, Fig. 2, Fig. 3). 96 of 97 ccRCCs were graded as Fuhrman grade I (n = 9), Fuhrman grade II (n = 49), Fuhrman grade III (n = 23), and Fuhrman grade IV (n = 15). Other RCC subtypes (papillary, chromophobe, and collecting duct cell carcinoma) were diagnosed in nine patients (8.3%), of which five were ppRCCs (four type I, one type II; Fig. 4), two were chRCCs (Fig. 5), one was collecting duct renal carcinoma, and one was combination of ccRCC and pRCC. Oncocytoma was diagnosed in nine patients (Fig. 6). One patient with oncocytomas had three tumors on each kidney (Fig. 7).
A total of 118 tumors were diagnosed in 113 patients, with 110 patients with one tumor, one patient with two tumors (in the same kidney), and two patients with three tumors (one with two RCCs in the same kidney and one cRCC on the other kidney, other patient with three oncocytomas in the same kidney). The mean tumor size was 62 mm ( ?? 31), range 7’160 mm.
CT examination protocol
CT examinations were performed with the patient in a supine position and holding his/her breath. Sixty-seven patients were examined using a single-slice helical CT scanner (LXi HiSpeed, General Electric Medical Systems, Milwaukee, WI, USA), and 46 patients were examined using a 16-detector row helical scanner (SOMATOM Sensation 16 scanner, Siemens, Erlangen, Germany). CT images were obtained with the following parameters: 120 kVp, 250’300 mA, slice thickness of 5.0 mm, and reconstruction interval of 5.0 mm. A volume of 100 mL containing 300’320 mgI/mL of intravenous contrast material was injected into the antecubital vein using an automatic injector through an 18’19 gauge venous line with a flow rate of 3’4 mL/s. A group of nonionic iodine contrast materials (270’320 mgI/mL) was used, including iohexol (Omnipaque 300, GE Healthcare Inc., Princeton, NJ, USA), iopromide (Ultravist 300, Bayer Healthcare Pharmaceuticals Inc., Wayne, NJ, USA), and iodixanol (Visipaque 270 and Visipaque 320, GE Healthcare Inc., Princeton, NJ, USA).
For all CT examinations, three series of images covering the whole kidneys were obtained, including unenhanced images (Fig. 1) and post-contrast images during the arterial (15’25 s delay) (Fig. 2) and nephrographic (90’100 s delay) phases (Fig. 3).
Attenuation values measurements
The attenuation values of renal tumors were measured on three separate regions of interest (ROI) in one examination phase, and the mean of these three values was calculated, representing the tumor attenuation value in the particular scanning phase. An ROI cursor was placed in the corresponding position in images of all three scanning phases on areas that contained solid tissues of the highest attenuation as perceived by the radiologist. ROIs were not placed in areas with calcifications, cystic degeneration, or necrotic parts. A wide (round or elliptical) ROI with a diameter equal to the tumor maximal diameter was selected to cover as much of the enhanced solid tumor as possible.
Attenuation values of the aorta measured at the level of the renal arteries and normal renal parenchyma of the kidneys with a tumor were obtained. ROIs were placed in the corresponding position on the cortex in images of all three scanning phases. Tumor enhancement ratios were calculated according to formulas originally defined by Herts et al. (Table 1) . The tumor attenuation values used for calculating the tumor enhancement ratios were obtained using both of the ROI styles described above.
Pathologic findings were used as the gold standard.
Pearson’s ??2-test or Fisher’s exact test was used to determine the significance of the differences between two independent samples for categorical variables. Differences between groups of normally distributed numerical variables were tested with parametric or non-parametric tests, depending on the type of distribution. The normality of distributions were assessed by using Kolmogorov-Smirnov test. Correlations between variables were tested by Spearman’s correlation test. The strength of associations between categorical variables and the output predictor variables was estimated using univariate and multiple logistic regressions.
Receiver operating characteristics (ROC) curves were generated and analyzed to evaluate the diagnostic validity of the tumor attenuation values on contrast-enhanced scans and tumor enhancement ratios in the differentiation of ccRCC from non-clear cell renal carcinomas. The cut-off value for each approach to the tumor enhancement measurement was chosen in order to achieve appropriate level of sensitivity and specificity with emphasis on the specificity. The level of statistical significance was ?? = 0.05. Statistical analysis was performed using STATA/IC ver11.2 (StataCorp LP, Texas, USA).
All distributions of attenuation values of renal tumors were normal (Kolmogorov-Smirnov test, P > 0.05) (Table 2).
The most significant Spearman’s rank correlation coefficient was computed between tumor attenuation values of RCCs on arterial phase scans and the tumor-to-aorta enhancement ratio in the arterial phase (rs = 0.73, P < 0.001).
The univariate and multiple logistic regression analysis revealed that the tumor-to-aorta enhancement ratio in the arterial phase produced results with the most significant impact on the prediction of a renal cell carcinoma subtype (clear-cell versus non-clear cell) using CT (Table 3). The tumor-to-aorta enhancement ratio in the nephrographic phase produced significant results in the differentiation between of a clear-cell renal cell carcinoma and oncocytoma using CT (Table 4). Results of direct tumor attenuation value measurements and other tumor-enhancement-ratios (tumor-to-aorta in arterial phase, tumor-to-kidney in arterial and nephrographic phase) for differentiating the clear cell RCC from oncocytoma were not significant (Table 4).
The area under the ROC curve [Az value] for the differentiation of renal carcinomas subtypes with direct measurements of tumor attenuation values in the arterial phase was 0.73 (95% CI 0.53’0.93). The cut-off value with the highest level of sensitivity and specificity was 74 HU, with a sensitivity of 79.0% and a specificity of 77.8% (Fig. 8)(Fig. 9)(Table 5).
Analysis of ROC curves revealed that the differentiation of renal carcinomas subtypes by the tumor-to-aorta enhancement ratio during the arterial phase was significant (Az value 0.79, 95% CI 0.63’0.96). The cutoff value with the highest level of sensitivity and specificity was 0.29, with a sensitivity of 72.0% and a specificity of 77.8% (Fig. 8) (Fig. 10)(Table 5).
For differentiating the clear cell subtype from non-clear cell subtypes based on the directly measured tumor attenuation value in the nephrographic phase, the Az value was 0.68 (95% CI 0.53’0.83) with a cutoff value with the highest accuracy of 80 HU (sensitivity of 63.0% and specificity of 77.8%) (Fig. 8) (Fig. 11). The Az value for the tumor-to-aorta enhancement ratio during the nephrographic phase was 0.71 (95% CI 0.51’0.91), and the cutoff value with the highest level of sensitivity and specificity was 0.65 with a sensitivity of 65.0% and a specificity of 77.8% (Fig. 8)(Fig. 12) (Table 4).
The differences in the ROCs for differentiating the clear cell subtype from non-clear cell subtypes of RCCs from the different tests were significantly different from one another in the tumor-to-aorta enhancement ratio during the arterial phase when compared with direct measurements of tumor attenuation values in the arterial phase and the tumor-to-aorta enhancement ratio during the nephrographic phase, the parameters which had the second and the third highest AUC (Chi-square test, P=0.0195).
Analysis of ROC curves revealed that results of the differentiation of clear-cell RCCs from oncocytomas based on the directly measured tumor attenuation values were better than results of the differentiation with tumor-enhancement-ratios (Table 6) (Fig 13). For differentiating the clear cell subtype of RCC from oncocytoma based on the directly measured tumor attenuation value on the precontrast images, the Az value was 0.5 (95% CI 0.33’0.77) with a cutoff value with the highest accuracy of 30 HU (sensitivity of 51.0% and specificity of 55.6%). The area under the curve for the differentiation of ccRCC and oncocytoma with the direct measurements of tumor attenuation values in the arterial phase was 0.50 (95% CI 0.26’0.74), and the cutoff value with the highest level of sensitivity and specificity was 103 with a sensitivity of 53.0% and a specificity of 55.6% (Table 6) (Fig 13).
All ROC curves for differentiating the clear cell RCC from oncocytomas have AUC (area under the curve) too small (0.5 or less) to have chose cut-off value with sensitivity and specificity that could be applied in clinical work.
In recent years methods of targeted therapy and minimally invasive approaches, such as several forms of percutaneous radiofrequency ablation (RFA) and cryoablation, in the treatment of RCC have been developed . At present, patients with both small renal tumors with no signs of local tumor progression and a decreased risk of metastatic disease have the choice between active surveillance for tumor management or a more aggressive treatment such as minimally invasive interventional methods (e.g,. cryoablation) or partial nephrectomy [8, 9]. Histological subtyping of RCC prior to treatment has been proved to be extremely valuable in patients with comorbidities, advanced age and in poor general condition who are not suitable for surgical and interventional therapy. The aformentioned pretreatment determination of RCC subytpes is of particular significance when the application of targeted therapy is considered, as the efficacy of such therapy is dependent on the cellular structure and metabolism of certain types of cancer cells [9, 10].
The most accurate method for pretreatment determination of histologic subtypes of RCC is core biopsy under imaging guidance, which is usually ultrasound (US) . CT remains the most commonly used imaging method for the evaluation of patients with renal masses, although several studies have demonstrated the effectiveness of MR as a first-line imaging tool for the characterization and staging of renal tumors [8, 15, 16].
Several groups of authors have performed retrospective studies assessing the enhancement characteristics of different histologic subtypes of RCC on postcontrast CT images [13,17-23]. In number of previous CT studies, ccRCC showed significantly higher attenuation values compared with other RCC subtypes and oncocytoma[5,13,17-23]. The differentiation of other less frequent subtypes (i.e. pRCC, chRCC, and collecting duct carcinoma) was less consistent in these previous studies [13,17’24].
The results of previous studies of the use of CT for the differentiation of RCC subtypes indicated that solid forms of more frequent RCC subtypes are associated with certain imaging features [6,13,17-24]. For example, the degree of contrast enhancement of solid tumor parts differs significantly between ccRCC and chRCCs or pRCCs [17, 18-22]. These studies have also shown that the contrast enhancement of ccRCCs is significantly higher than that of chRCC, pRCCs, collecting-duct carcinoma, and benign tumors like oncocytoma and AML with minimal fat, as well [17-23, 25, 26].
The purpose of this study was to evaluate CT with contrast enhancement as a rapid and a simple method to differentiate between the most malignant RCC subtype ccRCC and other solid forms of RCCs and oncocytomas on a routine, daily basis in the clinic. Patients with oncocytomas were included in the study because differentiation from solid forms of RCCs using radiographic (contrast enhancement) pattern is not enough sufficient [23, 26]. Differentiation of ccRCCs from oncocytomas and other solid tumors such as AMLs with minimal fat is of great clinical importance because previously mentioned renal tumors are usually benign tumors which could be treated conservatively if the diagnosis is accurately established. AML with minimal fat were not included in this study because of small number for statistical analysis.
To this end, four different techniques were used for measuring tumor contrast enhancement. Direct measurements of tumor attenuation values were used as described by Kim et al., Zhang et al., and Yamada et al., but only for studying malignant tumors [18, 21, 24]. Similarly, Tumor-enhancement ratios were used, such as those described by Herts et al.. Additionally, the tumor enhancement measurements were only used as a tool for differentiating RCC subtypes to avoid subjectivity in the analysis of the morphological features of tumors.
The ROI method applied in this study with a ROI that covered as much as possible of the solid parts of the tumor was used to reduce (intra)observer/measurement variability.
Zhang et al. consider the narrow ROI (or small area ROI) a better approach because it minimizes the volume-averaging effects caused by the heterogeneous nature of renal carcinomas . Subsequently, the attenuation values of the most enhanced, hyperdense areas of the tumor measured with narrow ROI are assumed to represent an adequate measure of tumor enhancement . The opposite opinion is that a narrow ROI (small area ROI) does not cover all solid portion(s) of the tumor, which can result in the loss of attenuation values that provide information about tumor enhancement and indirectly vascularization [22, 24].
The attenuation values of tumors may vary depending on the quality of renal vascular supply, which in turn depends on intrinsic factors that define the patient’s clinical status (Ruppert-Kohlmayr et al. 2004), such as cardiac function, intravenous access, patient weight and size, and blood viscosity  . Therefore, tumor enhancement ratios were introduced by Herts et al. as a method that takes into account the quality of the renal vascular supply . Tumor enhancement ratios were used for the prediction of pRCC subtype and the nuclear grade of RCCs .
The degree of tumor contrast enhancement measured in the arterial phase using different methods allowed for significant differentiation of clear cell from non-clear cell RCCs in this study. The most significant results were obtained through the use of the tumor-to-aorta enhancement ratio in the arterial phase (measurements made with an ROI that covered as much as possible of the solid parts of the tumor) with a cutoff value 0.25 and an Az value of 0.79. The observed AUC in the ROC analysis in this study are suboptimal and indicate that the parameters used in RCC differentiation showing better performance (Az of about 0.7-0.8) are only “moderately accurate” according to Swets classification of ROC curve results  . The differences in the contrast-enhancement characteristics of clear cell renal carcinoma and non-clear cell types support the results of previous studies [18-25]. The contrast enhancement characteristics of clear cell renal carcinoma are a consequence of a rich vascular network and the alveolar architecture of the tumor .
In our study, special attention was given to the scanning protocol parameters of each patient. Application of recommended standardized scanning protocols is essential for this type of research in order to define the enhancement characteristics of different hystologic types(or subtypes) of tumors, like RCC subtypes . Arterial phase images were used in order to get the optimal depiction of the arterial supply of the kidneys and the tumor which was important for better planning and selecting the most appropriate form of surgical treatment for urologists [30 – 32]. According to the previous studies importance of arterial phase scans should not be neglected because the possibilities of mischaracterization of vascular changes, such as renal artery aneurysms, as a solid occupying kidney lesions on scans taken during later perfusion phases such are nefrographic and excretory phase [30 – 32]. The nephrographic phase was included in the imaging protocol because it is the most important phase of renal enhancement with contrast material for detecting renal expansive masses especially smaller than 3cm [33, 34]. Scanning delay for nephrographic phase according to the different authors lasts from 80 seconds to 180 seconds after the start of contrast material injection. [30, 33, 34].
The degree of tumor contrast enhancement measured in the arterial phase using different methods allowed the most siginificant results in differentiation of clear cell from non-clear cell RCCs in this study. The differences in the contrast-enhancement characteristics of clear cell renal carcinoma and non-clear cell types support the results of previous studies [18-22, 25, 26]. The contrast enhancement characteristics of clear cell renal carcinoma are a consequence of a rich vascular network and the alveolar architecture of the tumor . The results of our study suggest that a wider application of arterial phase CT of the kidney in renal tumor imaging would permit better surgical treatment planning (partial nephrectomy). This could allow the differentiation of clear cell type from non-clear cell types of RCC before invasive management.
In this study results of differentiation between ccRCCs and oncocytomas with CT using both methods, direct tumor attenuation values and tumor-enahancement-ratios, were insignificant unlike the results for differentiating the clear cell subtype from non-clear cell subtypes of RCC.
The highest Az values for differentiating the clear cell subtype from non-clear cell subtypes of RCCs was calculated for the tumor-to-aorta enhancement ratio during the arterial phase (AUC=0.791; 95%CI=0.626-0.956) which can be interpreted as fair to good validity. The results of our study indicate that this method is potentially a good diagnostic tool. Therefore, it could be further investigated in larger study or meta-analysis study.
All ROC curves for differentiating the clear cell RCC from oncocytomas have AUC (area under the curve) too small (0.5 or less) to have chose cut-off value with sensitivity and specificity that would be meaningful in clinical work. The probable reason is too few oncocytoma patients to have valid diagnostic instrument. Due to small number of patients with this tumor, this result could be easily biased or characteristic to our patient cohort .
The limitations of this study are its retrospective nature and the small number of patients in non-clear cell RCC group, which prevented the evaluation of the enhancement characteristics of each subtype in non-clear cell RCCs group and other benign tumor of interest like AML with minimal fat. Malignant mesenchymal tumors of renal parenchymal were not included in the study because of insignificant small number, one patient with PNET. The another limitation is application of different contrast materials, as well. The another confounding factor is the use of two different CT techniques, 16-slices MSCT and single-slice spiral CT. The small number of patients is responsible for the exclusion of the benign renal tumors from the study, as well. Above mentioned limitation of this study is important because one of the major issues of diagnostic imaging of renal tumors and probably the most demanding is differentiating RCC from benign lesions, such as benign cystic lesions, oncocitoma, and angiomiolipoma with minimal fat [16-18, 35] . Consequently, further investigations with more patients with less frequent subytpes of RCC (pRCC, chRCC) and benign renal tumors should be performed in the future because of the importance of CT in diagnostic evaluation of renal tumors [8, 15, 17 ‘ 24, 28, 35].
Previous reports on the enhancement characteristics of non-clear cell RCC and solid benign parenchymal tumors were not as consistent as those of ccRCC [19-24,25,26]. Solid forms of RCC subtypes can be differentiated by tumor enhancement characteristics on multiphase renal CT images, with statistically significant accuracy for the differentiation of ccRCC from non-clear cell RCCs. This is an important finding because differences between clear cell and non-clear cell RCCs affect therapy choice and prognosis. Currently, differentiating between tumor types offers the possibility of using a neoadjuvant treatment with targeted therapy, the apropriate choice of which depends on distinguishing between clear cell and non-clear cell RCCs [9, 10, 36].
The results of our study indicate that tumor enhancement measurements on CT images in the arterial phase via either the direct measurement of tumor attenuation values or tumor enhancement ratios are not a substitute for core biopsy but can be used as an easily implemented auxiliary method for the pretreatment differentiation of solid renal parenchymal neoplasms. The results of our study highlight the importance of enhancement on arterial phase images for distinguishing solid forms of ccRCC from solid forms of other histological types of RCC. It is expected, that such noninvasive imaging techniques may prove quite valuable for managing patients with comorbidities who are not candidates for imaging-guided biopsy and surgical treatment but who are suitable for minimally invasive treatment methods and/or targeted therapy.
 Motzer RJ, Bukowski RM (2006) Targeted therapy for metastatic renal cell carcinoma. J Clin Oncol 24:5601-8
 Ferlay J, Parkin DM, Steliarova-Foucher E (2010) Estimates of cancer incidence and mortality in Europe in 2008 Eur J Cancer 46:765-81. doi: 10.1016/j.ejca.2009.12.014.
 Eble JN, Togashi K, Pisani P, et al (2004) Chapter I:Tumours of the kidney In: Eble JN, Sauter G, Epstein JI, Sesterhenn IA (eds) Pathology and genetics of tumours of the urinary system and male genitale organs, World health Organization Classification of Tumours, IARC Press, Lyon, pp 7-88. http://www.iarc.fr/en/publications/pdfs-online/pat-gen/bb7/BB7.pdf Accessed: 4 July 2012.
 Lopez-Beltran A, Scarpelli M, Montironi R, Kirkali Z (2006) 2004 WHO Classification of the Renal Tumors of the Adults. Eur Urol 49:798-805 .
 Prasad SR, Humphrey PA, Catena JR, Narra VR, Srigley JR, Cortez AD, Dalrymple NC, Chintapalli KN (2006) Common and Uncommon Histologic Subtypes of Renal Cell Carcinoma: Imaging Spectrum with Pathologic Correlation. RadioGraphics 26:1795-1806.
 Vikram R, Ng CS, Tamboli Ph, Tannir NM, Jonasch E, Matin SF, Wood CG (2009) Papillary Renal Cell Carcinoma: Radiologic-Pathologic Correlation and Spectrum of Disease. RadioGraphics. 29(3):741-54; discussion 755-7. doi: 10.1148/rg.293085190.
 Pignot G, Elie C, Conquy S, Vieillefond A, Flam T, Zerbib M, Debr?? B, Amsellem-Ouazana D (2007) Survival analysis of 130 patients with papillary renal cell carcinoma: prognostic utility of type 1 and type 2 subclassification. Urology 69:230-5 .
 Ljungberg B, Cowan CC, Hanbury DC, Hora M, Kuczyk MA, Merseburger AS, Patard JJ, Mulders PFA, Sinescu IC (2010) EAU Guidelines on renal cell carcinoma: the 2010 update. Eur Urol 58(3):398-406. doi: 10.1016/j.eururo.2010.06.032.
 Facchini G, Perri F, Caraglia M, Pisano C, Striano S, Marra L, Fiore F, Aprea P, Pignata S, Iaffaioli RV (2009) New treatment aproaches in renal cell carcinoma. Anti-Cancer Drugs 20(10):893-900. doi: 10.1097/CAD.0b013e32833123d4.
 Schrader AJ, Steffens S, Schnoeller TJ, Schrader M, Kuczyk MA (2012) Neoadjuvant therapy of renal cell carcinoma: A novel treatment option in the era of targeted therapy? Int J Urol 19(10):903-7. doi: 10.1111/j.1442-2042.2012.03065.x.
 Schatz SM, Lieber MM (2003) Update on oncocytoma. Curr Urol Rep. 4(1):30-5.
 World Medical Association (2008) World Medical Association Declaration of Helsinki – Ethical Principles for Medical Research Involving Human Subjects; 59th WMA General Assembly, Oct 2008. http://.www.wma.net/en/30publications/10policies/b3/ Accessed: 4 July 2012
 Herts BR, Coll DM, Novick AC, Obuchowski N, Linnell G, Wirth SL, Baker ME (2002) Enhancement characteristics of papilary renal neoplasms revealed on triphasic helical CT of the kidneys. AJR Am J Roentgenol 178:367-372.
 Leveridge MJ, Finelli A, Kachura JR, Evans A, Chung H, Shiff DA, Fernandes K, Jewett MA (2011) Outcomes of small renal mass needle core biopsy, nondiagnostic percutaneous biopsy, and the role of repeat biopsy. Eur Urol 60(3):578-84. doi: 10.1016/j.eururo.2011.06.021.
 Catalano C, Fraioli F, Laghi A, Napoli A, Pediconi F, Danti M, Nardis P, Passarielo R (2003) High resolution multidetector CT in the preoperative evaluation of patients with renal cell carcinoma. AJR Am J Roentgenol 180:1271’7.
 ??pero M, Brklja??i?? B, Kolari?? B, Marotti M (2010) Preoperative staging of renal cell carcinoma using magnetic resonance imaging: comparison with pathological staging. Clin Imaging 34:441-7. doi: 10.1016/j.clinimag.2009.09.005.
 Jinzaki M, Tanimoto A, Mukai M, Ikeda E, Kobayashi S, Yuasa Y, Narimatsu Y, Murai M (2000) Double-phase helical CT of small renal parenchymal neoplasms: correlation with pathologic findings and tumor angiogenesis. J Comput Assist Tomogr 24:835-42.
 Kim JK, Kim TK, Ahn HJ, Kim CS, Kim K-R, Cho K-S (2002) Differentiation of subtypes of renal cell carcinoma on helical CT scans. AJR Am J Roentgenol 178: 1499-1506.
 Ruppert-Kohlmayr AJ, Uggowitzer M, Meissnitzer T, Ruppert G (2004) Differentiation of renal clear cell carcinoma and renal papillary carcinoma using quantitative CT enhancement parameters AJR Am J Roentgenol 183:1387-1391.
 Sheir KZ, El-Azab M, Mosbah A, El-Baz M, Shaaban AA. Differentiation of renal cell carcinoma subtypes by Multislice Computerized Tomography (2005) J Urol 174:451-55.
 Zhang J, Lefkowitz RA, Ishill NM, Wang L, Moskowitz ChS, Russo P, Eisenberg H, Hricak H (2007) Solid renal cortical tumors:differentiation with CT. Radiology 244(2):494-505.
 Jung SC, Cho JY, Kim SH (2012) Subtype differentiation of small renal cell carcinomas on three-phase MDCT: usefulness of the measurement of degree and heterogeneity of enhancement. Acta Radiol 53:112-8. doi: 10.1258/ar.2011.110221.
 Young JR, Margolis D, Sauk S, Pantuck AJ, Sayre J, Raman SS (2013). Clear Cell Renal Cell Carcinoma: Discrimination from Other Renal Cell Carcinoma Subtypes and Oncocytoma at Multiphasic Multidetector CT. Radiology. 267(2):444-53. doi: 10.1148/radiol.13112617.
 Yamada T, Endo M, Tsuboi M, Matsuhashi T, Takase K, Higano S, Takahashi S (2008) Differentiation of pathologic subtypes of papillary renal cell carcinoma on CT. AJR Am J Roentgenol 191(5);1559-63. doi: 10.2214/AJR.07.3181.
 Wildberger JE, Adam G, Boeckmann W, Munchau A, Brauers A, Gunther RW, Fuzesi L. (1997) Computed tomography characterization of renal cell tumors in correlation with histopathology. Invest Radiol. 32(10):596-601
 Kim JK, Park SY, Shon JH, Cho KS. (2004) Angiomyolipoma with minimal fat: differentiation from renal cell carcinoma at biphasic helical CT. Radiology 230;677-84 doi:10.1148/radiol.2303030003
 Swets, JA(1988) Measuring the accuracy of diagnostic systems. Science 240:1285-93
 Fujimoto H, Wakao F, Moriyama N, Tobisu K, Sakamoto M, Kakizoe T (1999) Alveolar architecture of clear cell renal carcinoma (<or = 5.0cm) show high attenuation on dynamic CT scanning. Jpn J Clin Oncol 29:198-203.
 Johnson PT, Horton KM, Fishman EK (2010) How not to miss or mischaracterize a renal cell carcinoma: protocols, pearls, and pitfalls. AJR Am J Roentgenol 194(4):W307-15. doi: 10.2214/AJR.09.3033.
 Herts BR, Coll DM, Lieber ML, et al. (1999) Triphasic helical CT of the kidneys: contribution of vascular phase scanning in patients before urologic surgery AJR Am J Roentgenol 173:1273-7.
 Sheth S, Scatarige JC, Horton KM, et al. (2001) Current concepts in the diagnosis and management of renal cell carcinoma: role of multidetector CT and three-dimensional CT. Radiographics 21:S237-S254
 Coll DM, Smith RC. Update on radiological imaging of renal cell carcinoma. (2007) BJU International 99:1217-22
 Szolar DH, Kammerhuber F, Altziebler S. (1997) Multiphasic helical CT of the kidney: increased conspicuity for detection and characterization of small (< 3-cm) renal masses. Radiology 202(1):211-7.
 Kopka L, Fischer U, Zoeller G, et al. (1997) Dual-phase helical CT of the kidney: value of the corticomedullary and nephrographic phase for evaluation of renal lesions and preoperative staging of renal cell carcinoma. AJR Am J Roentgenol 169(6):1573-8
 Sacco E, Pinto F, Totaro A, D’Addessi A, Racioppi M, Gulino G, Volpe A, Marangi F, D’Agostino D, Bassi P (2011). Imaging of renal cell carcinoma: state of the art and recent advances. Urol Int 86:125-139. doi: 10.1159/000322724.
 Schimdt N, Hess V, Zumbrunn T, Rothermundt C, Bongartz G, Potthast S (2013) Choi response for prediction of survival in patients with metastatic renal cell carcinoma treated with anti-angiogenic therapies. Eur Radiol 23:632-9. doi: 10.1007/s00330-012-2640-x.
...(download the rest of the essay above)