A training program of a new simplified classification of magnified narrow band imaging for superficial esophageal squamous cell carcinoma.

ABSTRACT

BACKGROUND AND AIM: Optimal staging of the invasion depth of superficial esophageal squamous cell carcinoma is vital before endoscopic treatment. A new simplified magnified narrow-band imaging (M-NBI) classification system based on vascular architecture has recently been developed by the Japan Esophageal Society; however, its validity remains uncertain. METHODS: A total of 11 experienced and 11 inexperienced endoscopists were invited to join an endoscopic training program, which was composed of pretest, educational section, and post-test. The pretest and post-test sections included a set of endoscopic photos from 40 subjects with superficial esophageal squamous cell carcinoma with various invasion depths. Each subject appeared twice in the test, one with white-light imaging (WLI) only and the other with both WLI and M-NBI. The educational section included lectures and video demonstrations. RESULTS: The accuracy of WLI alone and combined with M-NBI at baseline were 0.53, 0.57 and 0.43, 0.41 for the experienced and inexperienced endoscopists, respectively, which then improved to 0.57, 0.63 and 0.49, 0.52 after training. Inter-observer agreement (k-value) of WLI alone and combined WLI and M-NBI for the experienced and inexperienced endoscopists also improved from 0.61, 0.61, and 0.61, 0.53 to 0.68, 0.71, and 0.71, 0.59, respectively. Multivariate analysis revealed that the educational course but not experience in endoscopy, NBI, or magnification significantly improved the diagnostic accuracy. M-NBI had a significant additional benefit to WLI, with an improvement in accuracy from 36% to 56% for the cases with m3/sm1 cancers (P < 0.05). CONCLUSIONS: A well-designed training program can improve the diagnostic accuracy in evaluating cancer invasion depth, with substantial agreement.