Defining specific criteria for individualized treatment of clear cell RCC is difficult [6] because there are great differences in the biologic behaviors of these solid renal masses owing to anatomic, histologic, and genetic factors and clinical stage [7]. Compared with other histologic and clinical factors, including tumor size, microvascular invasion, necrosis, and invasion of the collecting system [8,9], high Fuhrman grade has been identified an independent risk factor for predicting metastasis after nephrectomy. Thus, preoperative characterization of the aggressiveness of clear cell RCC is vital for selecting the optimal therapy and thus helpful for improving active surveillance [10,11].DWI is gaining interest as a tool for determining localized renal mass, differentiating RCC subtypes, and predicting tumor aggressiveness [12][13][14][15][16][17]. Studies have shown differ-C lear cell renal cell carcinoma (RCC) is the most common type of primary renal malignancy in adults. It accounts for approximately 70% of all RCCs [1, 2] and has more aggressive behavior and worse prognosis than other subtypes [3,4]. Although surgery is the only curative therapy for localized clear cell RCC, the currently supported minimally invasive or noninvasive approachesthat is, percutaneous radiofrequency ablation, cryoablation, microwave ablation, laser ablation, and high-intensity focused ultrasound ablation-are alternatives to the surgical treatment of RCC for elderly patients with small incidentally found renal cortical lesions, patients with bilateral tumors, and patients with a solitary kidney who are at high risk of complete loss of renal function after surgery [5].OBJECTIVE. The purpose of this study was to compare histogram analysis of apparent diffusion coefficient (ADC) and R2* for differentiating low-grade from high-grade clear cell renal cell carcinoma (RCC).
MATERIALS AND METHODS.Forty-six patients with pathologically confirmed clear cell RCC underwent preoperative BOLD and DWI MRI of the kidneys. ADCs based on the entire tumor volume were calculated with b value combinations of 0 and 800 s/mm 2 . ROI-based R2* was calculated with eight TE combinations of 6.7-22.8 milliseconds. Histogram analysis of tumor ADCs and R2* values was performed to obtain mean; median; width; and fifth, 10th, 90th, and 95th percentiles and histogram inhomogeneity, kurtosis, and skewness for all lesions.RESULTS. Thirty-three low-grade and 13 high-grade clear cell RCCs were found at pathologic examination. The TNM classification and tumor volume of clear cell RCC significantly correlated with histogram ADC and R2* (ρ = -0.317 to 0.506; p < 0.05). High-grade clear cell RCC had significantly lower mean, median, and 10th percentile ADCs but higher inhomogeneity and median R2* than low-grade clear cell RCC (all p < 0.05). Compared with other histogram ADC and R2* indexes, 10th percentile ADC had the highest accuracy (91.3%) in discriminating low-from high-grade clear cell RCC. R2* in discriminating hemorrhage was achieved with a threshold of 68.95 Hz...