Effect of fellow involvement on colonoscopy outcomes: A systematic review and meta-analysis

Background and aims: The effect of fellow involvement on colonoscopy outcomes is controversial. Thus, we evaluated this effect on adenoma detection rate (ADR) and on other colonoscopy quality indicators. Methods: MEDLINE and Cochrane central register of controlled trials were searched up to September 2018 for studies evaluating fellow-involved colonoscopies vs. attending physicians-only examinations in terms of colonoscopy outcomes. Primary outcome was ADR, while advanced ADR (AADR), mean number of adenomas per colonoscopy (MAC), cecum intubation rate (CIR) and adverse events rate comprised the secondary outcomes. The effect size on study outcomes was calculated using random-effects model and it is presented as Odds Ratio (OR) or Mean Difference (MD) with 95% confidence interval (CI). Results: Nineteen observational studies involving 34,059 patients (fellow-involved 16,875, attending physician-only 17,184) were included. Compared to the attending physician-only group, fellow involvement marginally increased ADR [OR (95% CI) = 1.12 (1.00-1.26); p = 0.06, I-2 = 76%]. Attending physicians with low-to-moderate ADR (< 35%) benefited most from fellow's participation [OR (95% CI): 1.26 (1.13-1.40) vs. 1.12 (1.00-1.26); p = 0.03 when ADR < 35% and OR (95% CI): 1.29 (1.13-1.46) vs. 0.95 (0.78-1.16); p = 0.01 when ADR < 30%, respectively]. Moreover, fellow-involved group had higher MAC compared to attending-only group [MD (95% CI) = 0.12 (0.04-0.20); p = 0.002, I-2 = 53%]. No benefit from fellow involvement was detected either for AADR, CIR or adverse events rate. Conclusions: Fellow involvement during colonoscopy is associated with more adenomas detected per procedure and with higher ADR when the attending physician-only group ADR is less than 35%. (C) 2019 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.