Glucagon-like Peptide-1 Receptor Agonists and Risk of Pneumonia and Postprocedure Hospitalization Among Patients Undergoing Gastrointestinal Endoscopy

Background and Aims: Safety concerns have been raised regarding the glucagon-like peptide-1 receptor agonist (GLP-1RA) use in patients undergoing procedural sedation and anesthesia. The objective of this study is to assess the association between GLP-1RA use and postprocedural risk of pneumonia and all causes of urgent care (UC), emergency department, and hospitalization. Methods: Using a propensity-matched retrospective cohort design, we matched patients with active GLP-1RA at the time of gastrointestinal endoscopic procedure between January 2008 and June 2023 to non-GLP-1RA-exposed controls (1:4) using propensity scores generated by logistic regression. The primary outcome was risk of postprocedural pneumonia and all-cause hospitalization further adjusted for type sedation (conscious sedation vs monitored anesthesia care) as well as study year. We further assessed association between pre-endoscopic GLP-1RA exposure and subsequent UC, emergency department visits, and 30-day mortality. Results: The propensity-matched cohort comprised 63% colonoscopy, 25% esophagogastroduodenoscopy and 12% bidirectional endoscopy among 3825 patients with active GLP-1RA matched to 14,920 controls. There was no significant association between active GLP-1RA use and postprocedure pneumonia (0.13% vs 0.25%, odds ratio 0.56 (95% confidence interval: 0.22, 1.45)) or 30-day postprocedural hospitalization (1.9% vs 2.6%, 0.76 (95% confidence interval: 0.59, 0.98)) compared to nonusers. No significant association was observed for UC, emergency department visits or mortality at 7- or 30-days postprocedure. Conclusion: Uninterrupted use of GLP-1RA was not associated with major adverse events among patients undergoing gastrointestinal endoscopy with sedation.