| Background: The conventional white light colonoscopy(WLC)examination has adenoma miss rate of approximately 22%,therefore it is significant to find a new and optimized colonoscopy technique improving the adenoma detection rate(ADR)and reducing this miss rate.There are multiple randomized controlled trials comparing the narrow band imaging(NBI)with WLC to evaluate the detection effect of colorectal adenomas and polyps,but the conclusions are not consistent,hence a meta-analysis is required to resolve this problem.Objective: To ascertain whether NBI improves ADR.Methods: “MEDLINE,Embase,Pub Med,Cochrane Library,CNKI,VIP Citation Databases,Ten thousand the Databases” were searched to November 2016,by using a combination of the following terms: “colorectal neoplasm”,“colonoscop* or endoscop*”,”NBI” and “electronic chromo endoscopy”.Results: Finally,16 prospective randomized controlled trials(RCT)comparing NBI with WLC for detection of adenoma or polyps were identified to have this meta-analysis.There was no statistically significant difference in the overall ADR and polyp detection rate(PDR)comparing the use of NBI with WLC(NBI vs.WLC ADR: 38.3% vs.36.6%,P =0.08,[relative risk,RR 1.06;95% confidence interval,CI 0.99-1.12],I~2=5%;PDR: 43.6% vs.41.3%,P=0.14,[relative risk,RR 1.07;95%CI 0.98-1.18],I~2=59%).When the subgroup analysis basing on the differences of colonoscopy definition was taken,we found that PDR was significantly higher in the high definition NBI group compared to the overall WLC(including standard definition and high definition pooled)group in fixed-effects meta-analysis(68.8% vs.60.4%,P=0.008,RR 1.14;95%CI 1.03-1.25,I~2=8%).Furthermore,the flat adenoma detection rate was significantly higher in the NBI group compared to WLC group in random-effects meta-analysis(23.4% vs.15.8%,P=0.03,RR 1.53;95%CI 1.03-2.27,I~2=56%).When the mean number of adenomas and polyps per participants was analyzed,no significant difference was found between NBI and WLC(1.183 vs.1.098,P=0.47,[mean difference,MD 0.03;95%CI-0.05 to 0.12,I~2=0 %];1.152 vs.1.043,P=0.41,[mean difference,MD 0.14;95%CI-0.19 to 0.47],I~2=96%).However,the mean flat adenoma number per participants was significantly higher in the NBI group compared to WLC group(0.668 vs.0.588,P = 0.03,MD 0.12;95% CI 0.01-0.24,I~2=0%).When the overall adenoma miss rate and polyp miss rate were analyzed comparing NBI with WLC,disappointedly there was no statistically significant difference(22.7% vs.20.7%,P=0.17,RR 1.12;95%CI 0.96-1.30,I~2=38%;22.2% vs.26.0%,P=0.23,RR 0.85;95%CI 0.66-1.10,I~2=63%).Conclusions: Overall,NBI do not enhance the ADR and PDR,not reduce the adenoma and polyp miss rate.However,the high definition NBI can improve the PDR.NBI can enhance the detection of flat adenomas. |
PDF Full Text Request