VI-RADS score and tumor contact length in MRI: A potential method for the detection of muscle invasion in bladder cancer

Abstract

Background: Vesical Imaging Reporting and Data System (VI-RADS) is a useful tool for evaluating muscle layer invasion of bladder cancer (BCa) on magnetic resonance imaging (MRI). Purpose: To evaluate the diagnostic performance of bladder MRI to detect the muscle layer invasion of BCa using VI-RADS score and quantitative MRI parameters. Methods: Preoperative bladder MRI was performed in 73 BCa patients. Two observers independently evaluated the MR blinded to histopathological data and classified the tumors according to VI-RADS criteria. Moreover, the quantitative parameters (maximum tumor diameter; Dmax, tumor contact length; TCL, and tumor apparent diffusion coefficient; ADC values) were independently measured by observers. The diagnostic performance of the VI-RADS score and quantitative values were evaluated by using receiver operating characteristic (ROC) analysis. Interobserver agreement was evaluated using the weighted-kappa coefficient (Kappa). Results: For the VI-RADS score, the AUC (area under the curve) was 0.968 and accuracy was 90.4% for Observer 1, and AUC was 0.953, accuracy was 89% for Observer 2. The AUC of TCL, TCL/DMax, and ADC values was 0.918, 0.675, and 0.832. In patients with a VI-RADS score >= 3, when a threshold value of TCL > 19.5 mm is used as complementary to the VI-RADS score, the accuracy of MRI for Observer-1 increases 100% and 97.26% for Observer-2. There was a good-excellent agreement between the observers in assessing the VI-RADS scores and quantitative parameters. Conclusion: Evaluation of bladder MRI using both VI-RADS criteria and TCL is successful and highly reproducible for detecting muscle layer invasion in patients with BCa.