OUTCOMES ASSOCIATED WITH GLUCAGON-LIKE PEPTIDE-1 RECEPTOR AGONIST USE IN PATIENTS UNDERGOING OUTPATIENT UPPER ENDOSCOPY

Background: Glucagon-like peptide-1 receptor agonists (GLP1RAs) are associated with delayed gastric emptying, which has raised concerns regarding the risk of adverse anesthesia events (e.g., aspiration) among GLP1RA-treated patients. There is scarce literature evaluating endoscopic procedure-related complications in patients on GLP1RAs.
Methods: This retrospective analysis evaluates patients undergoing upper endoscopy at a large tertiary medical center from 10/1/2022 to 10/1/2023 and who were also prescribed a GLP1RA during this timeframe. The “GLP group” was defined by GLP1RA exposure at the time of endoscopy determined by review of pre-procedural medication reconciliation and pharmacy dispensation history. The control group was the “non-GLP group” and included those who did not fill the medication, were previously exposed but not in the six weeks prior to endoscopy, or who were exposed only after endoscopy. The primary outcomes were the presence of solid retained gastric contents (RGCs) noted during endoscopy or any anesthesia adverse events, such as the need for intubation, respiratory compromise, or suspected aspiration. Analysis was performed using the Chi-square test and multivariable logistic regression.
Results: A total of 599 patients (median [IQR] age, 60 [51-68] 61% female) were included in the study, of which 360 (60%) were on a GLP1RA, and 238 (40%) were not on a GLP1 RA at the time of endoscopy. There was no statistically significant difference between the GLP group and non-GLP group when assessing for concurrent opiate use (8.6% vs. 12%, p=0.19) or the presence of gastroparesis (3.6% vs. 5.4%, p=0.277). There was a greater proportion of patients with diabetes mellitus in the GLP group (68% vs. 57%, p=0.005). The likelihood of having solid RGCs was significantly higher in the GLP group on univariate analysis (Odds ratio [OR], 2.74; 95% CI, 1.29-5.81; p=0.008). The association between GLP1RAs and the presence of solid RGCs remained on multivariate analysis controlling for sex, age, concurrent opiate use, and the presence of diabetes or gastroparesis. There was no association between GLP1RA and solid RGC in patients also undergoing concurrent colonoscopy. One patient in the GLP group and two patients in the non-GLP group had complications, including intubation due to RGC and delayed extubation following planned general anesthesia, but there was no association with GLP1RA use (OR, 3.0; 95% CI, 0.27-33.7, p=0.365). There were no cases of aspiration pneumonitis in the study sample.
Conclusion: Patients taking GLP1RAs who undergo upper endoscopy have increased odds of having solid RGCs, but not if undergoing concomitant colonoscopy. This study does not demonstrate an increased risk of anesthesia adverse events in patients taking GLP1RAs undergoing upper endoscopy. Limitations of this study include the moderate sample size.