2595-P: Tirzepatide vs. Semaglutide and the Risk of Incident Gastroparesis: A Propensity-Matched Analysis of 828,426 Patients

Introduction and Objective: GLP-1 receptor agonists are widely used for type 2 diabetes and obesity, and delayed gastric emptying has become an important safety consideration. This study evaluated whether adults initiating Tirzepatide (TZ) experienced a higher risk of incident gastroparesis than those on Semaglutide (SM). Methods: A retrospective cohort study conducted using the TriNetX Network. Adults starting TZ or SM were included, while those with prior GLP-1 agonist exposure, pre-existing gastroparesis, diabetic neuropathy, gastrectomy, or bariatric surgery were excluded. Propensity score matching (1:1) on age, sex, and type 2 diabetes produced 828,426 matched individuals. Outcomes included incident gastroparesis diagnoses and gastric emptying studies over 365 days, with a 180-day sensitivity analysis. Risk ratios (RR) and hazard ratios (HR) calculated. Results: At 365 days, TZ users had a markedly higher incidence of gastroparesis compared with SM (0.22% vs 0.002%; RR 88.0, 95% CI 45.7-169.2; HR 88.0, 95% CI 50.0-154.8; p<0.001), representing an almost 100-fold relative increase despite low absolute event rates. Gastric emptying study rates were similar (0.054% vs 0.059%; RR 0.91, p=0.322). Sensitivity analysis (180-day follow-up) confirmed higher gastroparesis risk with TZ (0.164% vs 0.003%; RR 50.7; HR 52.6; p<0.001) with comparable gastric emptying studies (0.040% vs 0.043%; RR 0.92, p=0.431). The number needed to harm at 365 days was 461. Conclusion: TZ use was associated with a significantly higher risk of incidence of gastroparesis compared with SM, while rates of gastric emptying studies remained similar. Potential explanations for this discrepancy include greater gastric motility suppression related to GLP-1 activity, dose escalation differences, or increased clinical recognition of GI symptoms. Although absolute rates were low and causality cannot be inferred, magnitude and consistency of the observed associations suggest differences in gastrointestinal tolerability between GLP-1-based therapies. Disclosure A. Vanaparti: None. T. Ikpeze: None. H.M. Baloch: None.