Vesical Imaging-Reporting and Data System (VI-RADS®): Initial experience in the classification of muscle-invasive...

to analyze the diagnostic accuracy of the VI-RADS® system in the differentiation of non-muscle-invasive bladder tumors (NMIBT from muscle-invasive bladder tumors (MIBT in suspicious cystoscopic findings without prior transurethral resection (TUR evaluated by radiologists with no prior experience in its use.

retrospective study carried out with 18 patients with suspicious lesions in cystoscopy. All of them underwent MRI of the bladder. Two radiologists with no prior experience in the use of the VI-RADS® system evaluated the results. All patients underwent TUR of the suspicious lesions after MRI. The sensitivity and specificity of the system were analyzed for VI-RADS® values ≥ 3 or VI-RADS® ≥ 4, as well as the Cohen's kappa coefficient between both radiologists.

the mean values of sensitivity and specificity of both radiologists considering both the VI-RADS® ≥ 3 or VI-RADS® ≥ 4 values were 91.7% and 87.5%, respectively. The kappa coefficient considering the VI-RADS® ≥ 3 as positive, was 0.551 (P<.05), while considering the VI-RADS® ≥ 4 as positive, it was 0.571 (P<.05).

The VI--RADS® system presents excellent sensitivity (91.7% and specificity (87.5% values in the classification of MIBT performed by radiologists with no prior experience in its use, with a moderate interobserver agreement.