Abstract
Purpose
The standard procedure for diagnostics and follow-up for non-muscle invasive bladder cancer (NMIBC) is cystoscopy in the outpatient clinic. Suspicious lesions are biopsied for histopathological assessment. This pilot study aimed to evaluate the ability of Confocal Laser Endomicroscopy (CLE) to rule out High Grade Urothelial Carcinoma (HGUC) to select patients for in-office fulguration.
Materials and methods
We performed a prospective non-randomized, single surgeon study. Intraoperative CLE was performed independently by the surgeon and a blinded on-site uropathologist. Following the procedure, a CLE evaluation was performed by another blinded urologist. Lesions were classified as normal/inflammatory, Low Grade Urothelial Carcinoma (LGUC) or HGUC. With the histological evaluations as the gold standard we calculated sensitivity, specificity, PPV and NPV for HGUC and the accuracy for each CLE assessor. The primary outcome was the NPV for HGUC for the surgeon.
Results
Twelve patients with a total of 34 lesions were included. Six lesions were flat and 28 were exophytic. On histopathology, 25 lesions were classified as normal/inflammatory or LGUC, while nine were classified as HGUC. For the surgeon, the uropathologist and the second urologist, the sensitivity was 44%, 78% and 22%, respectively. Specificities for the three observers were 84%, 68% and 96%. This corresponded to PPVs for HGUC of 50%, 47% and 67% and NPV for HGUC of 81%, 89% and 77%.
Conclusions
In our hands the NPV of CLE is not high enough for it to be considered an alternative to histopathological assessment of bladder lesions.
Acknowledgements
The authors acknowledge Dr J. Liao, Department of Urology, Stanford University, USA, for sharing his ‘Teaching set for confocal laser endomicroscopy of the bladder’. We would like to thank Dr J. Baard, department of Urology, Amsterdam UMC, for her valuable assistance and her help with the CLE setup.
Disclosure statement
The Cellvizio® System was borrowed by Mauna Kea Technologies for a period of 30 days. No potential conflict of interest was reported by the author(s).
Research involving human participants
The study was approved by the Danish Capital Region ethics committee (H-15020548) and the Danish Data Protection Agency (2012-58-0004).
Informed consent
All patients provided verbal and written consent before inclusion.