Peri-Procedural Safety of GLP-1 Receptor Agonists in Elective Endoscopy: A Multicenter Retrospective Cohort Study

Background and Aims: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) delay gastric emptying, raising concerns about periprocedural safety in elective endoscopy. We aimed to evaluate the association between pre-procedural GLP-1 RA use and post-procedural complications such as aspiration pneumonia. Methods: In this retrospective cohort study, adults (18-89 years) undergoing outpatient esophagogastroduodenoscopy or colonoscopy within the HCA Healthcare network from 1 July 2021 to 31 March 2024 were identified. Patients were classified as GLP-1 RA users (n = 953) or non-users (n = 3289) based on home medication records. Primary outcomes included aspiration, post-procedural oxygen requirement, hypotension, hospitalization, ICU admission, length of stay, and all-cause inpatient mortality. Multivariable logistic and negative-binomial regression models, incorporating an interaction term for anesthesia type, were adjusted for age, sex, body mass index, ASA class, and key comorbidities. Results: No aspiration events were reported in either group. GLP-1 RA use was associated with lower odds of post-procedural oxygen requirement (OR 0.43, 95% CI 0.25-0.76), hospitalization (OR 0.73, 95% CI 0.39-1.36), and mortality (0.1 vs. 0.9%, p = 0.014), and a shorter hospital stay (IRR 0.54, 95% CI 0.40-0.71). Rates of hypotension and ICU admission were similar between both groups. In anesthesia-stratified analysis among GLP-1 RA users, those receiving MAC/MS had higher odds of hospitalization compared with GA (OR 1.87, 95% CI 1.23-2.85, p = 0.003), whereas other outcomes were not significant. Conclusions: Pre-procedural GLP-1 RA therapy was not associated with increased peri-procedural complications. Although hospitalization was more frequent with MAC/MS, this difference did not extend to other clinically significant outcomes. Further prospective studies are needed to clarify the clinical implications of anesthesia choice.