Comparison of Narrowband Imaging and White-Light Endoscopy for Diagnosis and Screening of Nasopharyngeal Cancer.

Abstract

We compared the diagnostic accuracies of narrowband imaging and white-light endoscopy in the detection of nasopharyngeal cancer. Six databases (PubMed, Cochrane Database, Embase, Web of Science, SCOPUS, and Google Scholar). The 6 databases were thoroughly reviewed by 2 authors (working independently) from their dates of inception to December 2019. Nasopharyngeal mucosal or vascular changes detected by narrowband imaging were compared to those detected by white-light endoscopy. The authors extracted true-positive, true-negative, false-positive, and false-negative parameters for each study. Methodological quality was evaluated using the Quality Assessment of Diagnostic Accuracy Studies version 2 tool. The extent of interrater agreement was assessed. Eighteen prospective or retrospective studies were included. The diagnostic odds ratio of narrowband imaging was 77.560 (95% confidence interval [CI], 37.424-160.739). The area under the summary receiver operating characteristic curve was 0.926. The sensitivity, specificity, and negative predictive value were 0.871 (95% CI, 0.808-0.915), 0.905 (95% CI, 0.816-0.953), and 0.955 (95% CI, 0.906-0.979), respectively. The correlation between sensitivity and the false-positive rate was 0.284, indicating that heterogeneity was absent. Narrowband imaging exhibited moderate interrater reliability (0.7037; 95% CI, 0.6558-0.746). Subgroup analysis showed that vascular patterns revealed by endoscopy in a screened subgroup were significantly more diagnostically accurate than mucosal patterns used for surveillance of a recurrent cancer subgroup. Narrowband imaging exhibits high diagnostic accuracy and should be used in the diagnostic workup of nasopharyngeal cancer. However, further studies are necessary to confirm our results.