Benchmarking UK Lynch syndrome colonoscopy surveillance: quality metrics and diagnostic yield from the National Endoscopy Database

Background: Colonoscopy quality and diagnostic outcomes in Lynch syndrome have historically been heterogeneous. Following the introduction of the Lynch syndrome Bowel Cancer Screening Programme (BCSP), we assessed colonoscopy quality and diagnostic yield in Lynch syndrome using the National Endoscopy Database (NED), comparing outcomes between BCSP and non-BCSP providers. Methods: We performed a cross-sectional analysis of UK Lynch syndrome colonoscopies from NED between June 2024 and June 2025. Colonoscopy quality indicators and endoscopic detection rates for polyp (PDR) and advanced neoplasia (eANDR; polyps ≥10 mm or cancer; histopathology linkage was unavailable) were calculated and compared between BCSP and non-BCSP providers. Factors associated with neoplasia detection were identified using logistic regression. Results: Among 3196 colonoscopies, most were performed in England (90.2%), with 34.1% undertaken within BCSP. Overall colonoscopy quality was high: 97.6% were complete, 89.0% had adequate bowel preparation, and 96.3% had adequate tolerance. BCSP procedures demonstrated higher rates of completion (98.5% vs. 97.1%; P = 0.02) and adequate bowel preparation (92.7% vs. 86.7%; P < 0.001) than non-BCSP colonoscopies. PDR and eANDR were 46.8% and 4.7%, respectively, and were higher in BCSP than non-BCSP procedures (PDR 53.7% vs. 43.3%; eANDR 5.7% vs. 3.9%). Higher polyp detection was associated with increasing age (odds ratio [OR] 1.03, 95%CI 1.03-1.04), male gender (OR 1.48, 95%CI 1.26-1.73), BCSP status (OR 1.50, 95%CI 1.28-1.77), adequate bowel preparation (OR 1.81, 95%CI 1.40-2.34), and England nation (OR 1.99, 95%CI 1.22-3.36). Conclusions: Our NED evaluation demonstrated high colonoscopy performance in Lynch syndrome, with superior quality and endoscopic yield in BCSP procedures, supporting national integration of surveillance to establish Lynch syndrome-specific quality benchmarks.