Prospective Analysis of Confocal Laser Endomicroscopy for Assessment of the Resection Bed for Bladder Tumor

Abstract

Background and objective: Urothelial bladder cancer (UCB) care requires frequent follow-up cystoscopy and surgery. Confocal laser endomicroscopy (CLE), a probe-based optical technique for real-time microscopic evaluation, has shown promising accuracy for grading of UCB. We investigated the diagnostic accuracy of CLE-based assessment of the surgical radicality of the bladder resection bed (RB). Methods: We prospectively included 40 participants scheduled for transurethral resection of bladder tumors (TURBT) in two academic hospitals. Exclusion criteria were flat lesions, fluorescein allergy, and pregnancy. We performed CLE of the RB during TURBT. Histopathology of an RB biopsy was the reference test. Results at first cystoscopy 3 mo after TURBT are reported. A panel of two blinded observers evaluated the CLE images. The diagnostic accuracy of CLE for detection of detrusor muscle (DM) and residual tumor (rT) was calculated using 2 × 2 tables. Key findings and limitations: Histopathology for 22 CLE-matched RB biopsies revealed rT in four cases (18%) and DM in 13 (59%). The quality of CLE imaging was low in four (18%), moderate in 16 (73%), and good in two (9%) cases. CLE was able to correctly predict rT in two of the four cases (50%) identified on histopathology. The sensitivity, specificity, positive predictive value, and negative predictive value were 0.5 (95% confidence interval [CI] 0.07–0.93), 0.83 (95% CI 0.59–0.96), 0.4 (95% CI 0.05–0.85), and 0.88 (95% CI 0.64–0.99) for CLE prediction of rT, and 0.69 (95% CI 0.39–0.91), 0.33 (95% CI 0.07–0.7), 0.6 (95% CI 0.32–0.84), and 0.43 (95% CI 0.1–0.82) for prediction of DM, respectively. Five patients (23%) had rT at 3-mo follow-up; CLE had predicted rT in three, and histopathology had revealed rT in two cases at TURBT. Conclusions and clinical implications: CLE does not appear to be a reliable tool for detecting rT or DM in the RB after TURBT. Patient summary: We investigated a special imaging technique called confocal laser endomicroscopy (CLE) for checking the bladder after surgery for bladder cancer in a group of 40 patients. CLE results were compared to traditional biopsy results and the patients were checked after 3 months. CLE was not very reliable in detecting any remaining cancer (only 50% accurate) or important muscle tissue in the surgical area, and the quality of the images varied. While CLE shows some promise, it is not currently a dependable method for evaluating the bladder after bladder cancer surgery.