The value of adding diffusion-weighted images for tumor detection and preoperative staging in renal pelvic carcinoma for the reader's experience.

Abstract

This retrospective study aimed to assess the value of adding diffusion-weighted magnetic resonance imaging (DWI) or gadolinium-enhanced fat-suppressed T1WI (CEI) to T2-weighted imaging (T2WI) for preoperative T categorization in renal pelvic carcinoma by the reader's experience using surgical specimens as the reference standard. Two radiologists (Reader 1; 3years, 2; 13years) reviewed 49 cases with urothelial carcinoma who underwent magnetic resonance imaging examination before surgery, independently, using three image sets: T2WI alone, T2WI plus DWI, and T2WI plus CEI for tumor detection and T categorization. The differences in the apparent diffusion coefficient values between tumors and renal parenchyma, histopathologic grade were analyzed. T2WI plus CEI or DWI had high detection rates (93.4%) compared to T2WI alone. When discriminating T3a/T3b, for Reader 1, the use of T2WI plus DWI (88.0%) and T2WI plus CEI (92.0%) was significantly more accurate than T2WI alone (73%), with AUCs of 0.86, 0.86 and 0.77, respectively. For Reader 2, the accuracies were high on all image sets, with AUCs of 0.87-0.95, and the mean ADC of the tumors was significantly lower than that of the normal renal parenchyma. In addition, the mean ADC values of high-grade tumors were significantly lower than that of low-grade tumors. DWI and CEI could be more helpful than T2WI alone for preoperative T categorization by less-experienced reader and DWI could be used for preoperative T categorization and for predicting the histopathologic grade of renal pelvic carcinoma.