Evidence Report on the Safety of Gastrointestinal Endoscopy in Patients on Glucagon-like Peptide-1 Receptor Agonists: A Systematic Review and Meta-Analysis.

Abstract

Background/Objectives: Glucagon-like peptide-1 receptor agonists are increasingly used worldwide for weight and hyperglycemia management. There is an ongoing debate on the presence of increased gastric residue, leading to complications such as aspiration and overall safety in patients receiving upper gastrointestinal endoscopy. We aimed to study the effect of GLP-RAs on endoscopy outcomes. Methods: We conducted a detailed search of online databases to select the studies which provided details of the effects of GLP-RAs on patients undergoing endoscopy. The outcomes of interest were odds of retained gastric content (RGC), aspiration risk, and aborted and repeated procedures. A random effect model was used to calculate the pooled odds of outcomes with a 95% CI. We further calculated the pooled odds of predictive factors associated with an increased rate of retained gastric residues in the study population. Results: We included 12 studies with a total of 105,515 patients, of which 32,144 were on GLP-1 RAs and 73,273 were in the control group. A total of 234 (0.73%) aspiration events in GLP-RA users were noted compared to 257 (0.35%) events in the control group. No increased odds (1.26, 95% CI 0.86-1.87, I² 34%) of aspiration were found in GLP-1 users compared to the non-GLP-1 group. Patients on GLP-1 RA had increased RGC compared to the control group (OR 6.30, 95% CI 5.30-7.49, I² 0%). The pooled odds of aborted (OR 5.50, 95% CI 3.25-9.32, I² 0%) and repeated procedures (OR 2.19, 95% CI 1.42-3.38, I² 0%) were significantly higher in GLP-1 RA users. Patients taking Tirazepatide had the highest percentage of RGC (18.9%), while exenatide users had the lowest rate (6.2%) of food retention. Patients undergoing concomitant colonoscopy were found to have significantly low pooled odds of RGC (OR 0.26, 95% CI 0.04-0.48). GLP-1 RAs use was independently associated with increased odds of RGC (3.91, 95% CI 3.21-4.62, I² 0%). The results were homogenous and stayed consistent in the sensitivity analysis. Conclusions: Although the odds of RGC and aborted procedures are high in the GLP-1 RAs group compared to the control, no significant difference in the odds of aspiration was found between the two groups. Simple measures such as a clear liquid diet for 24 h, as routinely set for patients undergoing colonoscopy, may reduce the risk of retaining gastric residue in these patient populations.