36 The use of NBI is not routine in clinical practice, and many gastroenterologists remain unfamiliar with its use. For NBI to be widely applied to polyp differentiation in the community, several criteria must be fulfilled including: (i) good interobserver agreement and specified endoscopic criteria for histology; (ii) development of teaching tools for learning NBI; and (iii)

demonstration that practicing endoscopists can acquire skills using NBI. Two prospective observational single-centre studies have shown that short NBI training sessions were effective for physicians with varying levels of endoscopic experience in distinguishing between hyperplastic and adenomatous polyps on NBI.37,38 Most studies that have reported on interobserver agreement selleckchem in polyp differentiation have provided insufficient details on the methodology.29,32 East et al. showed moderate-to-good agreement for Kudo pit pattern (k-value 0.48) and vascular pattern intensity (k-value 0.64) in the FK228 mw assessment of 32 polyps by two observers.12 Rastogi et al. recently reported no significant difference in the kappa value for interobserver

prediction for polyp type on NBI between experienced and less experienced gastroenterologists.39 In a prospective study involving less-experienced endoscopists (colonoscopy > 5 years but no experience with NBI) and highly experienced endoscopists (routinely PI-1840 used NBI for > 5 years), the diagnostic accuracy of polyps based on Sano and Kudo classification systems using NBI with high magnification improved in the less experienced endoscopist group to levels equivalent to that of the highly experienced endoscopists group after expanded training.40 These results suggested that NBI can be effectively learnt with dedicated training. Further studies should assess the impact of training on in vivo histological prediction during live colonoscopy and whether improvement can be sustained over time. Unlike chromoendoscopy, the NBI system is convenient because it features a simple one-touch

button for changing from white light to NBI and does not require indigocarmine dye spraying. The procedure entails minimal time implications and little additional cost to the procedure. It is currently too early to conclude whether chromoendoscopy will be replaced by NBI. Randomized controlled studies have shown that chromoendoscopy improved the detection of flat and small adenomatous polyps3,41,42 and neoplasia in ulcerative colitis. However due to the increased procedure time, higher cost and labor-intensive procedure, chromoendoscopy has not been implemented in routine practice.43 Although NBI can potentially provide accurate definition of vascular structures in the colon and represents an attractive substitute for chromoendoscopy, several questions remain unanswered.