Trends and Variations in Colonoscopy Sedation: Insights based on the Dutch Gastrointestinal Endoscopy Audit.

Abstract

Background Sedative medication is routinely used during colonoscopy. Sedation practices may vary in terms of the type and dose of administered medication, possibly impacting clinical outcomes, associated risks, and financial and environmental burdens. We aimed to assess trends and variations in colonoscopy sedation in the Netherlands. Methods Data prospectively recorded in the Dutch Gastrointestinal Endoscopy Audit registry between January 2020 and January 2025 were analyzed. Annual trends in the application of different types of sedation were assessed. Variations across endoscopy units were evaluated by comparing midazolam dosing regimens and the proportional application of propofol-based sedation. Regression analyses assessing the association between sedation and cecal intubation and polyp detection were performed. Results Data from 285,241 colonoscopies were included. Application of propofol-based sedation increased from 5.8% (2020) to 10.0% (2024). The largest increases were observed for IBD-related (11.6-23.2%) and therapeutic indications (28.2-51.7%). For midazolam-based sedation, the mean dose administered varied considerably across units (2.1-5.6mg), as did the proportional application of propofol-based sedation (median 5.0% [IQR 2.3-15.7]; range 0-38%). Compared to midazolam-based sedation using doses equal to or lower than 2.5 mg, propofol-based sedation was not associated with improved cecal intubation (-0.7%; 95% confidence interval [CI] -1.0 to -0.3%), although it was associated with polyp detection (+5.5%; 95% CI +2.8 to +8.3%). Conclusion Colonoscopy sedation practices vary across Dutch endoscopy units in terms of midazolam dosing regimens and the proportional application of propofol-based sedation. In general, propofol-based colonoscopy sedation is increasing. This study's findings may inform guidelines for more standardized, evidence-based, and resource-conscious sedation practices.