Sedation is essential in gastrointestinal (GI) endoscopy to optimize patient comfort, procedural quality, and endoscopist satisfaction [1,2]. While non-anesthesiologist-administered sedation (NAS) with propofol is well established for low-risk patients, i.e. American Society of Anesthesiologist (ASA) I–II [3], evidence regarding its use in higher-risk groups remains limited. ASA III patients present a greater likelihood of cardiopulmonary events, and current data are heterogeneous and underpowered [4,5]. This study aimed to evaluate the safety profile of NAS with propofol in ASA III patients undergoing routine diagnostic GI endoscopy.

This observational study included ASA III patients who underwent elective gastroscopy or colonoscopy under NAS with propofol, with or without adjunctive midazolam and/or fentanyl,

between 2017 and 2024. Sedation was administered by a member of the endoscopy team (endoscopist or nurse) who had completed the ESGE/ESGENA curriculum for sedation in GI endoscopy and was exclusively dedicated to sedation. The protocol dosages were as follows: propofol 0.3–0.6 mg/kg (with an additional dose of 0.3–0.4 mg/kg after 2 minutes), preceded or not by midazolam 0.02–0.03 mg/kg and/or fentanyl 0.6–1.3 μg/kg. Anesthesia-related adverse events (ARAEs) were categorized as minor, moderate, or severe based on the World Society of Intravenous Anesthesia definitions. Secondary outcomes included the identification of potential risk factors for ARAEs.

Among 1219 patients, 1423 procedures were analyzed (43.4% gastroscopies; 56.6% colonoscopies). A total of 79.2% of patients (965/1219) were aged 65 years or older. Four sedation regimens were used: propofol alone (19.3%), propofol+midazolam (18.7%), propofol+fentanyl (32.1%), and propofol+fentanyl+midazolam (29.9%). ARAEs occurred in 5.2% of procedures: 3.9% minor, 0.9% moderate, and 0.4% severe. No cases of tracheal intubation or death were observed. Hypotension was the most frequent event (3.6% requiring no intervention; 0.8% requiring fluids). Desaturation was rare (0.2% minor; one case requiring transient bag-valve ventilation). Severe ARAEs consisted only of bradycardia treated successfully with atropine. The incidence of ARAEs differed among sedation regimens (p=0.023): the highest rate was observed with propofol+midazolam (8.6%), while propofol+fentanyl had the lowest (3.7%). Post-hoc analysis showed a significant difference between these two regimens (p= 0.04, OR 2.46). Logistic regression analysis identified lower body mass index as the only independent predictor of ARAEs (p<0.001). Exploratory analyses to investigate these findings revealed a statistically significant negative correlation between BMI and the administered per-kilogram propofol dose (r= -0.196, p<0.001). An ANOVA test confirmed that patients in higher BMI classes (overweight and obese) received significantly lower mean per-kilogram propofol dosages compared to the normal-weight group (p<0.001). No differences were observed in the mean per-kilogram fentanyl and midazolam dosages. Age, procedure type, and duration were not associated with adverse events. Finally, most patients were elderly, with no increase in ARAEs compared with younger patients.

NAS with propofol appears safe in ASA III patients undergoing elective GI endoscopy when performed by trained personnel following standardized protocols. The overall incidence of ARAEs was low, and severe events were rare and manageable, even in elderly patients. Sedation regimen and BMI significantly influenced adverse event risk, suggesting the importance of tailored drug dosing. These findings support expanding NAS with propofol to higher-risk patients within structured, guideline-based pathways. This strategy may reduce the need for anesthesiologist involvement, improve resource utilization, and potentially lower procedural costs. The results further underline the need for prospective, multicenter studies to refine risk stratification and optimize sedation protocols.