Diagnosis of narrow-band imaging in non-muscle-invasive bladder cancer: A systematic review and meta-analysis

Objectives The objective was to evaluate the diagnostic accuracy of cystoscopy assisted by narrow-band imaging compared with white-light imaging for non-muscle-invasive bladder cancer. Methods An electronic database search of PubMed, Embase, the Cochrane Library, Ovid and Web of Science was carried out for all articles comparing narrow-band imaging with white-light imaging cystoscopy in the detection of non-muscle-invasive bladder cancer. The review process followed the guidelines of the Cochrane Collaboration. Results Seven studies with prospectively collected data including a total of 1040 patients were identified, and 611 patients with 1476 tumors were detected by biopsy. In the patient- and tumor-level analysis, an additional 17% of patients (95% confidence interval, 10-25%) and an additional 24% of tumors (95% confidence interval, 17-31%) were detected by narrow-band imaging, respectively. In the patient- and tumor-level analysis, significantly higher detection rates using narrow-band imaging (rate difference 11%; 95% confidence interval 5-17%; P<0.001; and rate difference 19%; 95% confidence interval 12-26%; P<0.001, respectively) rather than white-light imaging were found. On the tumor level, an additional 28% of carcinoma insitu was detected (95% confidence interval 14-45%) by narrow-band imaging, and a significantly higher detection rate (rate difference 11%; 95% confidence interval 1-21%; P=0.03) was found. The false-positive detection rate of tumor level did not differ significantly between the two techniques. Conclusions Cystoscopy assisted by narrow-band imaging detects more patients and tumors of non-muscle-invasive bladder cancer than white-light imaging, and it might be an additional or alternative diagnostic technique for non-muscle-invasive bladder cancer.