Computer-aided quality assurance versus standard colonoscopy: A systematic review and meta-analysis of randomized-controlled trials

BACKGROUND: Computer-aided quality assurance (CAQ) can provide feedback to improve mucosal inspection during colonoscopy. This systematic review and meta-analysis of randomized controlled trials (RCTs) evaluated the efficacy of CAQ versus standard colonoscopy in enhancing colonoscopy quality indicators.

METHODS: Adhering to PRISMA guidelines, we searched MEDLINE, Embase, and CENTRAL for RCTs comparing CAQ with standard colonoscopy. The primary outcome was adenoma detection rate (ADR). Secondary outcomes included detection rates for advanced adenoma (AADR) and adenocarcinoma (ACDR), adenomas per colonoscopy (APC), and withdrawal time. Pooled risk ratios (RRs) and weighted mean differences (WMDs) were calculated using random-effects models.

RESULTS: Six RCTs involving 5687 colonoscopies (CAQ 2895; standard colonoscopy 2792) were included. Compared with standard colonoscopy, CAQ increased ADR by 43% (RR [95%CI]: 1.43 [1.17-1.73]; P < 0.001, I2 = 73.5%), with heterogeneity decreasing substantially after omitting one study (1.52 [1.32-1.75]; P < 0.001, I2 = 11.9%); AADR by 54% (1.54 [1.26-1.87]; P < 0.001, I2 = 0.0%); ACDR by 30% (1.30 [1.00-1.68]; P = 0.047, I2 = 0.0%); APC by 0.14 adenomas (WMD [95%CI]: 0.14 [0.08-0.20]; P < 0.001, I2 = 53.2%); adenomas ≥10 mm by 0.03 adenomas (0.03 [0.02-0.05]; P < 0.001, I2 = 45.1%); and withdrawal time increased by 0.64 minutes (0.64 [0.04-1.24]; P = 0.04, I2 = 96.1%).

CONCLUSIONS: CAQ significantly improved ADR and detection of clinically significant lesions. While findings for AADR and ACDR should be interpreted with caution due to low event rates, the improvements in detection of adenomas ≥10 mm and APC suggest that CAQ may lower the risk of post-colonoscopy colorectal cancer, a hypothesis that warrants evaluation in future long-term analyses of patient outcomes.