A large propensity-matched cohort study presented at the American College of Gastroenterology 2025 Annual Scientific Meeting in Phoenix revealed significant differences in esophageal complications between bariatric surgery and glucagon-like peptide-1 receptor agonist (GLP-1 RA) therapy for weight loss.
Bariatric surgery—particularly sleeve gastrectomy—nearly doubled the five-year risk of de novo GERD, esophagitis, and Barrett esophagus compared with GLP-1 RA therapy and was associated with greater need for endoscopic evaluation and acid-suppression therapy.
Himsikhar Khataniar, MD, and fellow researchers from Allegheny General Hospital in Pittsburgh, PA, and collaborating institutions, analyzed data from the TriNetX U.S. Collaborative Network. They identified adults with a body mass index greater than or equal to 30 kg/m² who either initiated FDA-approved weight-loss GLP-1 RA therapy (liraglutide, semaglutide, or tirzepatide) or underwent sleeve gastrectomy or Roux-en-Y gastric bypass from 2015 to 2020. While tirzepatide is a dual glucose-dependent insulinotropic polypeptide/GLP-1 agonist, researchers grouped it under GLP-1 RAs.
Exclusions included prior GLP-1 RA exposure, previous bariatric surgery, or any history of GERD, esophagitis, Barrett esophagus, esophageal cancer, scleroderma, pregnancy, or intensive-care admission. Propensity-score matching balanced 47 demographic, clinical, medication, and laboratory variables between groups.
Before matching, 89,422 bariatric surgery recipients (mean age 49 years; 73% female) and 84,689 GLP-1 RA users (mean age 53 years; 56% female) were identified. Post-matching cohorts were well-balanced, with a mean age of 51 years and 65% female representation in each group.
Dr. Khataniar and colleagues assessed de novo reflux esophagitis, GERD, Barrett esophagus, esophageal cancer, esophagogastroduodenoscopy (EGD) utilization, and initiation of acid-suppression therapy.
Reflux esophagitis occurred in 0.7% of surgery patients versus 0.4% of GLP-1 RA users, representing nearly double the risk with surgery. GERD developed in 16.9% of surgery patients compared with 9.8% of GLP-1 RA users, reflecting an 88% increased risk with surgical intervention.
Barrett esophagus developed in 0.5% of surgery patients versus 0.2% of GLP-1 RA users, showing double the risk with surgery. Esophageal cancer remained rare in both groups, with no statistically significant difference between treatments.
EGD utilization was substantially higher among surgery patients at 8.8% versus 3.7% in the GLP-1 RA group. Proton-pump inhibitor initiation occurred in 29.4% of surgery patients compared with 20.8% of GLP-1 RA users. H₂-receptor antagonist prescriptions were required in 15.6% of surgery patients versus 11% of GLP-1 RA users.
Among individual interventions, tirzepatide users showed the lowest GERD incidence at 6.4%, whereas sleeve gastrectomy had the highest rate at 19.1% relative to GLP-1 RAs.
All researchers reported no relevant financial relationships.
