The Gallbladder Does Not Care How the Weight Came Off

A single-centre cohort finds GLP-1 receptor agonist users reach acute pancreatitis with higher predicted severity, pointing back at a biliary complication bariatric surgery has spent two decades learning to prevent.

bariatric surgeryGLP-1complications

The gallbladder does not record how the weight came off. That is the uncomfortable premise behind a single-centre cohort published on 17 November in Surgical Endoscopy, which reviewed 149 patients admitted with acute pancreatitis at the Hospital of the University of Pennsylvania between March 2017 and August 2024 and split them by whether they had been taking a GLP-1 receptor agonist for at least three months before the diagnosis.

Forty-nine of them had: 32.9% of the cohort. After adjustment, agonist exposure was independently associated with a Bedside Index for Severity in Acute Pancreatitis score of 2 or higher (odds ratio 5.25, 95% CI 1.13–24.34, p = 0.034). Age was the only other predictor that reached significance, at 1.07 per year. Pancreatitis-associated ascites was more frequent in the exposed group; new-onset diabetes and exocrine pancreatic insufficiency were less frequent; length of stay and post-pancreatitis sequelae did not separate the two groups.

Read the arithmetic before the headline. Forty-nine exposed patients, a score dichotomised at a bedside threshold, a confidence interval running from a barely detectable effect to a twenty-five-fold one. This is a retrospective chart review, not a trial, and its senior author discloses industry consulting and equity holdings.

What the paper does have is a mechanism it is willing to name. Nearly half of the agonist users developed pancreatitis within a year of starting therapy, most had lost weight beforehand, and biliary pathology — stones, choledocholithiasis or sludge — appeared in both groups at rates that did not differ significantly, including in patients whose gallbladder had already been removed. The authors read that as a biliary contribution rather than an injury specific to the drug.

The oldest complication in the field

Rapid weight loss causing gallstones is not new to bariatric surgery; it is the complication the specialty has the longest record on. A meta-analysis in BMC Surgery in June pooled ten randomised trials and 3,575 participants on prophylactic ursodeoxycholic acid: stone formation fell by 67% at 12 to 24 months and 73% at six months, symptomatic stones by 70%, while cholecystectomy rates did not differ significantly. Doses of 500–600 mg daily for six to twelve months carried the effect, whichever operation had been performed.

A scoping review in Obesity Surgery in April mapped how unsettled the protocol still is: ten studies from 2014 to 2023, control-arm stone incidence of 25% to 40%, treated incidence of 1.4% to 10.8%, relative risks from 0.13 to 0.36, six months of 500–600 mg daily as the commonest regimen, early initiation described as crucial, and no standardisation across centres.

The mechanism review in Frontiers in Surgery puts the range behind those numbers at 10.4% to 52.8% in the first postoperative year, attributing it to injury to the hepatic branch of the vagus, altered bile composition, sharply reduced food intake, gut-hormone shifts and dysbiosis — with sleeve gastrectomy generally less risky than bypass. A 150-patient sleeve series in Langenbeck’s Archives, with ultrasound scheduled at three, six and twelve months, found thirteen stones at a median of six months and significantly greater early weight loss among the patients who formed them.

When the stones end up in the duct rather than the gallbladder, the anatomy a bypass created becomes the problem, which is why single-session versus two-stage EDGE has its own literature and still no randomised answer.

A different instrument, measuring something else

Against that body of work sits the incretin evidence, and it is a different kind of evidence. A scoping review of tirzepatide’s pancreatic and gallbladder safety, published in October in Diabetes, Metabolic Syndrome and Obesity, gathered 11 randomised trials, five observational studies, four case reports, a EudraVigilance analysis and 575 FAERS reports. In the trials, pancreatitis appeared in 11 participants across five studies (0.2% to 3.7%) and cholelithiasis in 73 across eleven (0.3% to 3.7%).

Those percentages sit an order of magnitude below the surgical control arms, and the reason is largely ascertainment. The bariatric trials looked for stones on a schedule, in patients losing weight at a measured rate; the drug trials counted events that were reported. The review’s own verdict is that missing exposure denominators make the sources incomparable.

What is missing is a denominator

No trial has done for a patient losing a fifth of body weight on tirzepatide what ten randomised trials did for a patient losing it after a sleeve, and ursodeoxycholic acid has not been tested in agonist-treated patients at all. The Penn authors stop short of recommending prophylaxis; their conclusion asks for attention to the biliary route and notes that biliary disease appeared after cholecystectomy too, which points at bile composition rather than a gallbladder simply sitting still.

NIDDK’s patient information states the mechanism without preference for the method: fast weight loss, “like from weight-loss surgery,” is listed among the factors that make gallstones more likely, and an untreated blockage of the bile or pancreatic duct is described as potentially fatal. That is physiology, not a ranking of treatments.

What the field has, in the end, is one version of this complication with a protocol resting on ten randomised trials and years of scheduled follow-up — the kind of long-horizon record the LABS cohort, funded by NIDDK, was designed to produce — and a second route to the same weight loss with a pharmacovigilance file and no protocol at all. Whether that asymmetry is physiology or simply elapsed time is the question the next set of denominators will answer.

Sources

  1. Urias E, et al. Glucagon-like peptide-1 receptor agonist use is associated with higher severity of acute pancreatitis. Surgical Endoscopy, 17 November 2026.
  2. Alrubaiaan A, et al. Efficacy of prophylactic ursodeoxycholic acid in preventing gallstone formation after metabolic bariatric surgery: an updated systematic review and meta-analysis of randomized controlled trials. BMC Surgery, 30 June 2026.
  3. da Silva MQ, et al. UDCA dosage, duration, and initiation timing for post-metabolic bariatric surgery cholelithiasis prevention: a scoping review. Obesity Surgery, April 2026.
  4. Chen S, et al. Gallstones after bariatric surgery: mechanisms and prophylaxis. Frontiers in Surgery, 13 March 2025.
  5. Dogan S, Donder Y. The effect of body mass index change on cholelithiasis after laparoscopic sleeve gastrectomy: a single center experience. Langenbeck's Archives of Surgery, 13 August 2025.
  6. Kutsyba T, et al. Pancreatic and gallbladder adverse events associated with tirzepatide in obesity management: a scoping review. Diabetes, Metabolic Syndrome and Obesity, 23 October 2026.
  7. National Institute of Diabetes and Digestive and Kidney Diseases. Gallstones: definition and facts.
From the blog

New research notes