The predictive value of surgeon's perception during transurethral resection of bladder tumours

Abstract

Introduction: Little is known about the urologists' ability to predict grade, stage and the presence of muscular layer during transurethral resection of bladder tumours (TURBT). This study aimed to assess the clinical prediction of bladder tumour stage and grade based solely upon cystoscopic appearances at the first TURBT. Methods: We conducted a prospective study on 100 consecutive patients undergoing TURBT for newly diagnosed bladder tumours. During the initial TURBT, Urology residents and seniors predicted the histological grade, invasion and the presence of detrusor muscle in the specimen, compared with the final histopathology result. Results: From 100 patients, 91 were suitable for analysis. The residents correctly predicted tumour invasiveness in 75.8% of cases, while the seniors correctly predicted in 86.8%. High-grade tumours were reported in 77.7% of the specimens and were correctly predicted in 74.7% and 76.9%, by residents and seniors, respectively. Muscle representativeness was found in 80.2% of resections. In 73.6% and 75.8% of the cases, the residents and the seniors, respectively, correctly predicted the presence or the absence of the detrusor muscle layer. The positive predictive value (PPV) for this parameter was 79.2% for the resident, and 80.6% for the senior. Conclusions: The surgeon's naked eye analysis showed a good predictive ability to detect non-muscle invasive and high-grade bladder tumours in TURBT specimens. Results demonstrate that the senior has a higher predictive ability than the resident, though, the differences were not statistically significant. The PPV for muscle representativeness reinforces the idea of performing an extensive and careful TURBT.