Diagnostic Performance of Narrow Band Imaging for Laryngeal Cancer: A Systematic Review and Meta-analysis.

Objective To evaluate the performance of narrow band imaging (NBI) for the diagnosis of laryngeal cancer and to compare the diagnostic value of NBI with that of white light endoscopy. Data Sources PubMed, Embase, Cochrane Library, and CNKI databases. Review Methods Data analyses were performed with Meta-DiSc. The updated Quality Assessment of Diagnostic Accuracy Studies-2 tool was used to assess study quality and potential bias. Publication bias was assessed with the Deeks's asymmetry test. The protocol used in this article has been published on PROSPERO and is in accordance with the PRISMA checklist. The registry number for this study is CRD42015025866. Results Six studies including 716 lesions were included in this meta-analysis. The pooled sensitivity, specificity, and diagnostic odds ratio for the NBI diagnosis of laryngeal cancer were 0.94 (95% confidence interval [95% CI]: 0.91-0.96), 0.89 (95% CI: 0.85-0.92), and 142.12 (95% CI: 46.42-435.15), respectively, and the area under receiver operating characteristics curve was 0.97. Among the 6 studies, 3 evaluated the diagnostic value of white light endoscopy, with a sensitivity of 0.81 (95% CI: 0.76-0.86), a specificity of 0.92 (95% CI: 0.88-0.95), and a diagnostic odds ratio of 33.82 (95% CI: 14.76-77.49). The evaluation of heterogeneity, calculated per the diagnostic odds ratio, gave an I2 of 66%. No marked publication bias ( P = .84) was detected in this meta-analysis. Conclusion The sensitivity of NBI is superior to white light endoscopy, and the potential value of NBI needs to be validated in future studies.