Multiparametric 3-T MRI for differentiating low-versus high-grade and category T1 versus T2 bladder urothelial...

OBJECTIVE. The purpose of this study was to determine an optimal multiparametric MRI protocol for characterizing tumors of low versus high grade and differentiating tumors as T1 versus T2 for preoperative staging of bladder urothelial carcinoma. SUBJECTS AND METHODS. Thirty-nine patients underwent MRI within 1 week before surgery. Three image sets-T2-weighted plus diffusion-weighted MRI (DWI), T2-weighted plus dynamic contrast-enhanced MRI (DCE-MRI), and T2-weighted plus DCEMRI plus DWI-were independently interpreted by two readers at 2-week intervals. ROC curves were plotted for both readers to compare the diagnostic efficacy of the three sets for detrusor muscle invasion for each reader, and the areas under the ROC curve were compared by use of the Bonferroni test. The apparent diffusion coefficient (ADC) values were correlated with histopathologic grade. RESULTS. A total of 49 category T1 and T2 lesions were analyzed. The average ADC of 11 low-grade tumors (1.141 ± 0.164 × 10(-3) mm(2)/s) was significantly (p < 0.05) higher than that of 20 high-grade malignant tumors (0.766 ± 0.091 × 10(-3) mm(2)/s). Neither reader considered T1 tumors as probably having muscle invasion (category T2) in the T2-weighted plus DWI image sets or the T2-weighted plus DWI plus DCE-MRI image sets. Using the T2-weighted plus DCE-MRI sets, the two readers overstaged 13 and 15 of 36 tumors by misdiagnosing category T1 as T2. With the cutoff ADC value of 0.899 × 10-3 mm(2)/s, the sensitivity and specificity for differentiating high- and low-grade bladder urothelial carcinoma were 100% and 95%. CONCLUSION. Multiparametric MRI with T2-weighted plus DWI plus DCE technique is the optimal protocol for preoperative staging of organ-confined bladder urothelial carcinoma. The ADC of low-grade tumors is significantly higher than that of high-grade tumors with 100% sensitivity and 95% specificity at a cutoff ADC value of 0.899 mm(2)/s.