NSAID initiation and risk for gastrointestinal bleeding in patients receiving GLP-1 receptor agonists: A retrospective cohort study

PURPOSE: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have been purported to increase risk for gastrointestinal (GI) bleeding. Our objective was to determine whether GLP-1 RAs are associated with increased risk for GI adverse events following the initiation of a nonsteroidal anti-inflammatory drug (NSAID). METHODS: This retrospective cohort study used national administrative data from the Veterans Health Administration to identify patients with type 2 diabetes who initiated an NSAID while concurrently receiving a GLP-1 RA or a comparator agent: a sodium-glucose cotransporter-2 inhibitor (SGLT-2i) or a dipeptidyl peptidase-4 inhibitor (DPP-4i). The primary outcome measure was the incidence of a GI bleed or ulcer within 12 months of NSAID initiation. Cox proportional hazards regression was used to compare risk for GI bleeding or ulcer between GLP-1 RA recipients and the comparator groups while adjusting for confounders. A sensitivity analysis was conducted using GI bleed as a secondary outcome. RESULTS: A total of 13,433 patients received a GLP-1 RA, 24,480 received an SGLT-2i, and 30,968 received a DPP-4i. The primary outcome occurred in 0.9% (n = 117) of patients receiving a concurrent GLP-1 RA, 1.0% (n = 243) of those receiving an SGLT-2i, and 1.0% (n = 308) of those receiving a DPP-4i. In adjusted regression analyses, no difference in risk for GI bleeding or ulcer was observed between GLP-1 RAs and either comparator (SGLT-2i: hazard ratio [HR], 0.85; 95% confidence interval [CI], 0.68-1.07; DPP-4i: HR, 0.82; 95% CI, 0.66- 1.01). Similar nonsignificant results were observed for the secondary outcome. CONCLUSION: We did not observe an increased risk for developing a GI bleed or ulcer within the year following NSAID initiation with concurrent GLP-1 RA therapy, relative to comparator agents.