GLP-1 GI safety, what the trial data and pharmacovigilance actually say about gastroparesis, gallbladder, and pancreatitis

Key takeaways

• GI side effects are the most common adverse events of the GLP-1 class. STEP-1 reported nausea in 44 percent, diarrhea in 32 percent, vomiting in 25 percent, and constipation in 24 percent of the semaglutide arm. Most are mild to moderate and concentrated in the titration phase.
• The FDA added gastroparesis-related labeling guidance to the GLP-1 product family in 2023 and 2024; current Wegovy, Ozempic, and Mounjaro labels include warnings on severe GI adverse reactions and delayed gastric emptying.
• The Sodhi et al. 2023 JAMA pharmacovigilance analysis using real-world data flagged elevated risk of gastroparesis (HR 3.67), pancreatitis (HR 9.09), and bowel obstruction (HR 4.22) on GLP-1 vs. bupropion-naltrexone; the trial data show smaller absolute risks but the directional signal is consistent.
• Gallbladder events (cholelithiasis, cholecystitis) are a recognized class effect. The He et al. 2022 meta-analysis of 76 randomized trials reported a 37 percent relative increase in gallbladder or biliary disease.
• The American Society of Anesthesiologists issued perioperative guidance in 2023 (updated 2024) recommending withholding GLP-1 prior to elective surgery to reduce aspiration risk from delayed gastric emptying.

Trial rates of GI adverse events

The GLP-1 class slows gastric emptying as part of its mechanism for appetite suppression and postprandial glycemic control. Slower gastric emptying is the proximate cause of the most common adverse events. STEP-1 (semaglutide 2.4 mg, 68 weeks) reported nausea in 44.2 percent, diarrhea in 31.5 percent, vomiting in 24.8 percent, and constipation in 23.4 percent of the active arm versus 16.8, 15.9, 6.6, and 9.5 percent on placebo. Most events were mild to moderate and concentrated in the dose-titration phase .

SURMOUNT-1 (tirzepatide 5/10/15 mg, 72 weeks) reported a similar pattern with a slight increase in GI event frequency at higher doses: nausea 24 to 33 percent, diarrhea 19 to 23 percent, vomiting 8 to 12 percent across the three dose arms . The trials report study-drug discontinuation due to GI events of approximately 4 to 7 percent across the class.

Gastroparesis: signal, evidence, labeling

Delayed gastric emptying is the intended pharmacology, not an adverse effect on its own. Gastroparesis as a clinical syndrome (chronic delayed gastric emptying with severe symptoms in the absence of mechanical obstruction) is a different matter. Case reports of severe gastroparesis on GLP-1, sometimes persistent after discontinuation, prompted the FDA labeling action in 2023 .

The Sodhi et al. 2023 JAMA pharmacovigilance analysis used a large US claims database to compare GLP-1 users with bupropion-naltrexone users for weight loss. After adjustment, GLP-1 use was associated with hazard ratios of 3.67 for gastroparesis, 9.09 for pancreatitis, and 4.22 for bowel obstruction . The analysis is hypothesis-generating because it is observational and the gastroparesis ICD coding can be heterogeneous, but it materially shaped the conversation around GLP-1 GI risk.

Current Wegovy, Ozempic, Mounjaro, and Zepbound prescribing information include warnings or precautions on severe gastrointestinal adverse reactions. Some labels reference acute pancreatitis as a separate warning .

Gallbladder and biliary events

Cholelithiasis and cholecystitis are recognized class effects. The He et al. 2022 meta-analysis of 76 randomized trials reported a 37 percent relative increase in gallbladder or biliary disease (RR 1.37, 95 percent CI 1.23 to 1.52) on GLP-1 vs. comparator . The mechanism is multifactorial: rapid weight loss is itself a risk factor for cholelithiasis, and GLP-1 may directly slow gallbladder motility.

Approved GLP-1 product labels include warnings or precautions on acute gallbladder disease. The absolute rate is low but clinically relevant; users with new right-upper-quadrant pain after starting a GLP-1 should be evaluated .

Pancreatitis: trial signal vs. observational signal

Acute pancreatitis was an early concern across the GLP-1 class and is on every approved label as a precaution. In the cardiovascular outcomes trials and weight-loss trials, the acute pancreatitis rate has been low and not consistently elevated relative to placebo. SUSTAIN-6 reported pancreatitis in less than 0.5 percent of subjects . SELECT did not report excess pancreatitis .

Observational pharmacovigilance signals (Sodhi et al. 2023) report higher hazard ratios. The reconciliation is partly population (claims-database analyses include populations not enrolled in the trials), partly diagnostic ascertainment (claims-database pancreatitis codes can include non-acute presentations), and partly statistical power (trials are powered for cardiovascular outcomes, not adverse-event detection at low absolute rates) .

Perioperative considerations and aspiration risk

Delayed gastric emptying raises aspiration risk during anesthesia induction, particularly when the patient has been NPO for the conventional 6 to 8 hours that assume normal gastric transit. Case reports of food residue on endoscopy and intraoperative aspiration prompted the American Society of Anesthesiologists to issue perioperative guidance in 2023, with an update in 2024.

The current ASA guidance recommends a multidisciplinary approach: consider holding once-weekly GLP-1 for one week and once-daily GLP-1 for 24 hours prior to elective procedures, with extended preoperative fasting for those who have not held the drug, and point-of-care gastric ultrasound where available. The PubMed-indexed guidance documentation summarizes the rationale .

What this means for users

• Most GI events are titration-phase and dose-related; slow titration and not skipping the lower doses reduces severity.
• Persistent severe vomiting or upper-abdominal pain on a GLP-1 should not be assumed to be ordinary side-effect tolerance and warrants clinical evaluation.
• New right-upper-quadrant pain after rapid weight loss on a GLP-1 should prompt evaluation for cholelithiasis or cholecystitis.
• Tell every prescriber and the anesthesiology team about GLP-1 use before elective procedures or endoscopy.
• Compounded GLP-1 products do not necessarily reproduce the labeled dose, fill, or excipient profile of approved products. The FDA has issued repeated cautions on dosing errors with compounded GLP-1 .