Anaesthetic Adverse Events During Propofol-Based Sedation for ERCP: A Real-World Cohort Study

Background/Objectives: Endoscopic retrograde cholangiopancreatography (ERCP) requires deep sedation, which is increasingly provided by anaesthetists using propofol-based regimens. However, real-world data on the incidence and predictors of anaesthesia-related adverse events (AEs) in this setting remain limited. The objective of this study was to assess the frequency, predictors, and clinical significance of adverse events during anaesthetist-delivered sedation for ERCP, based on a propofol regimen. Methods: We conducted a retrospective single-centre cohort study including 388 consecutive adult patients who underwent ERCP with propofol-based sedation administered by an anaesthetist. Adverse events were classified into three tiers: Tier 1 (any adverse physiological events, including haemodynamic and respiratory threshold crossings), Tier 2 (clinically significant events requiring pharmacological intervention-the primary regression outcome), and Tier 3 (high-severity outcomes, reported descriptively). Independent predictors of Tier 2 events were identified using multivariable logistic regression. Results: Adverse physiological events occurred in 220 patients (56.7%), the majority of which were minor and self-limiting. Clinically significant events requiring active pharmacological intervention occurred in 108 patients (27.8%), with vasopressor-treated hypotension as the predominant component (88 patients, 22.7%). All bradycardia episodes required atropine administration (34 patients, 8.8%), while desaturation was largely self-limiting, with advanced airway management required in only three patients (0.8%). High-severity outcomes were rare (9 patients, 2.3%). In multivariable logistic regression predicting clinically significant adverse events, propofol dose (OR 1.20 per 10 mg, 95% CI 1.14-1.25, p < 0.001), ASA physical status (OR 1.63, 95% CI 1.07-2.49, p = 0.024), age (OR 1.04 per year, 95% CI 1.01-1.07, p = 0.007), and ketamine use, confounded by indication (OR 2.18, 95% CI 1.14-4.14, p = 0.018) were independent predictors. Model fit was good (Nagelkerke R² = 0.43). Conclusions: Adverse events are frequent when defined using inclusive criteria, but are predominantly minor in severity. Propofol dose is the principal modifiable risk factor, demonstrating a consistent dose-response relationship across multiple adverse outcomes. ASA physical status and age further identify patients at increased risk of clinically significant events requiring intervention. Ketamine use was associated with increased odds of adverse events; however, this association is likely confounded by indication and should not be interpreted as a direct causal effect. These findings support stepwise propofol titration guided by clinical sedation assessment, with avoidance of anticipatory dosing particularly in older patients and those with higher ASA scores, and highlight the safety of anaesthetist-led sedation in this setting.