Sedative medication is commonly used to improve patient comfort and tolerability during colonoscopy. Standards for colonoscopy sedation are known to vary across countries, while different types of sedation are also associated with differences in clinical outcomes, as well as different economic and environmental burdens. In the Netherlands, insights into sedation practices across endoscopy units were lacking. This study aimed to evaluate national trends in colonoscopy sedation, inter-unit variability in sedation practices, and associations between type of sedation and various colonoscopy performance measures in the Dutch population.

All colonoscopies recorded in the Dutch Gastrointestinal Endoscopy Audit (DGEA) registry between January 1, 2020, and January 1, 2025, were eligible for inclusion. Sedation was classified as none, mild sedation (midazolam and/or opioids), or deep sedation (propofol). Trends and variability in colonoscopy sedation were analyzed across predefined indication categories: positive fecal immunochemical test for the colorectal cancer (CRC) screening program; post-polypectomy or CRC surveillance; familial CRC risk; inflammatory bowel disease (IBD); IBD surveillance; therapeutic procedures; and symptoms/other indications. The association between sedation type (none, mild, deep) and midazolam dosage (<2.5 mg, 2.6-5 mg, >5 mg) with the incidence of successful cecal intubation and polyp detection was assessed using generalized estimating equation Poisson regression.

A total of 260,316 colonoscopies were included. Mild sedation was used in the majority of colonoscopies (ranging from 88.9% to 92.6% over the years). Meanwhile, the use of deep sedation increased progressively from 5.8% in 2020 to 10.0% in 2024. The most notable increase in deep sedation was observed for IBD colonoscopies (11.6% to 23.2%; annual change 2.7% [95% CI 2.0-3.3]) and therapeutic colonoscopies (28.2% to 51.7%; annual change 5.3% [95% CI 4.3-6.3]). For procedures with mild sedation, the mean midazolam dosage varied by indication category, with the lowest dosage for colonoscopies for post-polypectomy or CRC surveillance (3.44 ± 1.29 mg) and the highest for IBD colonoscopies (4.26 ± 1.55 mg). Significant inter-unit variability in the overall mean midazolam dosage was observed (range: 2.13 ± 1.17 to 5.62 ± 1.70 mg), indicating considerable differences in midazolam dosing regimens across units. Compared to mild sedation with <2.5 mg midazolam (reference), a dosage of 2.6-5 mg was associated with a slightly higher incidence of successful cecal intubation (+0.6% [95% CI 0.4-0.8%]) and polyp detection (+1.7% [95% CI 0.4-3.0%]). Midazolam dosages >5 mg and propofol sedation did not improve cecal intubation, although they were associated with higher polyp detection rates (Table 1).

In the Netherlands, the use of deep sedation for colonoscopy is rising. Besides, sedation practices were illustrated to vary considerable across units, particularly in terms of midazolam dosing regimens. While a midazolam dosage of 2.6-5 mg was associated with slight improvements in cecal intubation and polyp detection compared to dosages <2.5 mg, the gain for daily practice seems very small. Therefore, the benefits of routinely administering doses >2.5 mg should be carefully weighed considering the associated greater risks, costs and environmental burdens of higher dosing. The findings of this study can be used to inform the development of guidelines for more standardized colonoscopy sedation.