Abstract

Purpose: There is a sudden increase in the number of publications reporting on the use of the narrow band imaging (NBI) colonoscopy to differentiate neoplastic from hyperplastic colorectal polyps. These data would be valuable to allow us to understand its diagnostic operating characteristics, which, in turn, could provide insights of its potentials for clinical use. Methods: We extensively searched the literature using Medline, EMBASE and Cochrane Library databases and meeting abstracts; used an advanced bivariate meta-analysis to summarize the data, and a hierarchical model to synthesize the summary receiver-operating characteristic (ROC) curve. We calculated the summary sensitivity, specificity, and negative predictive value (NPV). Results: NBI diagnosis of colorectal polyps can be classified as an excellent test - the area under the ROC curve was 0.94 - based on 28 studies (18 manuscripts and 10 abstracts) involving 6,280 polyps in 3,036 patients. All studies reported the test to be worthwhile (Figure). The overall sensitivity was 91.0% (95% CI: 87.6-93.5%) and specificity was 82.6% (95% CI: 79.0-85.7%). In 6 studies (n=1567 polyps) that used high or low confidence diagnostic predictions, the overall sensitivity was 95.1% and specificity was 85% for a high confidence diagnosis, and the ROC area increased to 0.97. In the diagnosis of diminutive polyps (n=1850), including both high and low confidence predictions, NBI had a sensitivity of 89.4% (95%CI: 80.8-94.4%) and a specificity of 85.8% (95%CI=77.6-91.4%). The NPVs were 93.2, 90.1, and 86% for the proportion of neoplasms among polyps to be 40, 50 and 60%, respectively.Figure: [496] The Summary Receiver Operating Curve of the use of the Narrow Band Imaging to diagnose neoplastic and hyperplastic colorectal polyps. The size of the circles indicates the weight of each study.Conclusion: The accuracy of NBI diagnosis of neoplastic and hyperplastic colorectal polyps can be classified as excellent. It provides high (>90%) sensitivity and NPV for clinical implementation. Disclosure: Dr. McGill: nothing to disclose. Dr. Vangelis: nothing to disclose. Dr. Ioannides: nothing to disclose. Dr. Soetikno: consultant for Olympus Medical Systems Corporation, and has received research funding from Olympus America. Dr. Kaltenbach: consultant for Olympus Medical Systems Corporation, and has received research funding from Olympus America.Table: [496] Performance of real-time optical diagnosis with NBI colonoscopy

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call