S1258 Randomized Controlled Trial of Holding Versus Continuing GLP-1 or GIP Agonists Before Upper Endoscopy (OCULUS Trial)

Introduction: Guidelines recommend that patients on daily dosing hold glucagon-like peptide-1 (GLP-1) and gastric inhibitory polypeptide (GIP) agonists on the day of endoscopy, and patients on weekly dosing hold the drug a week prior. These recommendations were rapidly incorporated into clinical practice; however, the data supporting them are conflicting and based on low-quality evidence. We hypothesized that holding a single dose of a GLP-1 or GIP agonist prior to endoscopy would be non-inferior to continuing these medications for reducing the risk of clinically relevant residual gastric volume (CR-RGV). Methods: Single-blinded, randomized controlled trial at 2 tertiary centers, approved by the Institutional Review Board, and registered on clinicaltrials.gov (NCT06533527). Eligible outpatients on stable maintenance dosing of GLP-1/GIP medications planned for an upper endoscopy (esophagogastroduodenoscopy, endoscopic ultrasound, or endoscopic retrograde cholangiopancreatography) were invited to participate. Patients were randomly assigned to either continue GLP-1/GIP agonist therapy (“continue” group) or to hold therapy as outlined in guideline recommendations (“hold” group). Both endoscopy and anesthesiology teams were blinded to study group assignment. Using Farrington-Manning’s test, we estimated that 120 patients were needed (60 in each group) to be 80% certain the upper limit of a 1-sided 95% confidence interval would exclude a difference in favor of the standard group of more than 3% (non-inferiority limit). Interim analyses were planned at 20% and 50% of recruitment to assess if the study should be terminated prematurely for safety, or to recalculate the sample size. Results: Sixty patients completed the study protocol, and an interim analysis was conducted at 50% enrollment as planned. There were no significant differences in baseline characteristics. In the intention-to-treat analysis, CR-RGV was significantly higher in the “continue” group. The study was terminated because the Z-value was 2.75, which exceeded the O’Brien-Fleming stopping boundary of 2.5491. There were no cases of CR-RGV in the subgroup who also underwent colonoscopy and were on 1 day of clear liquid diet (0/13 continued, 0/12 held). There were no study-related adverse events. Conclusion: Continuing GLP-1 and GIP-1 agonists prior to upper endoscopy significantly increases the risk of CR-RGV that precludes adequate examination. Holding a single dose of the medication significantly reduces the risk of CR-RGV. Clear liquids the day prior to endoscopy may eliminate the risk of CR-RGV, regardless of whether the medication is held or continued.