GLP-1 Protocols and Bowel Obstruction: What the Ileus Warning Means and Which Symptoms Matter

Can a GLP-1 medicine cause a bowel obstruction? Bowel obstruction and ileus have been reported in people taking semaglutide and related drugs, and both now appear on the Ozempic and Wegovy labels. Whether the drugs actually raise the risk is less clear: the largest cohort studies found rates no higher than with another diabetes drug, and the absolute numbers are small either way.

GLP-1 receptor agonists slow the gut on purpose. Slower stomach emptying is part of how they reduce appetite, and constipation is one of the most common side effects in every trial. That makes the obstruction question a reasonable one to ask, especially for anyone with a history of abdominal surgery or who also takes other drugs that slow the bowel.

This article sets out what the label says, what the 2023 JAMA signal found, what the bigger studies since then show, and which symptoms separate ordinary GLP-1 constipation from something that needs a doctor the same day. None of the research below tested microdose regimens, and that gap matters for how the numbers should be read.

What is the difference between ileus and a bowel obstruction?

A blocked pipe versus a stalled one

A mechanical bowel obstruction means something physically blocks the intestine, while ileus means the intestine stops moving its contents along even though the passage is open. Common causes of mechanical obstruction are adhesions, the bands of scar tissue that can form after abdominal surgery, and hernias. Ileus is a motility problem, the bowel going quiet, and it often follows surgery, infection or drugs that slow the gut.

The distinction matters for GLP-1 medicines because their known action is on motility. A drug that slows transit has an obvious route to ileus and to severe constipation, but no direct route to creating scar tissue or a hernia. In people who already have adhesions, slower transit could plausibly make a partial narrowing more likely to cause symptoms, though no study has tested that idea directly. Our article on delayed gastric emptying explains the motility effect in more detail.

What does the FDA label say about ileus and GLP-1 drugs?

Postmarketing reports, not trial findings

The current Ozempic and Wegovy prescribing information list ileus, intestinal obstruction and severe constipation including fecal impaction under postmarketing experience. That section collects reports sent in after approval. The label itself notes that because these reports come from a population of uncertain size, it is not always possible to estimate their frequency or establish that the drug caused them.

The trial data in the same labels give a clearer picture of everyday bowel effects. In Ozempic's placebo-controlled trials in type 2 diabetes, constipation was reported by 1.5% on placebo, 5% on 0.5 mg and 3.1% on 1 mg. In the Wegovy weight-management trials, it was reported by 24% on 2.4 mg and 11% on placebo, and in the separate 7.2 mg trial by 20% on 7.2 mg, 19% on 2.4 mg and 8% on placebo.

Both labels also describe severe gastrointestinal reactions as a warning. Ozempic reports them in 0.4% on 0.5 mg and 0.8% on 1 mg versus 0% on placebo; Wegovy reports 4.1% on the injection versus 0.9% on placebo. These cover all serious stomach and bowel events, mostly nausea, vomiting and diarrhoea, not obstruction alone.

What did the 2023 JAMA study on GLP-1 side effects find?

The signal that made headlines

The 2023 JAMA research letter by Sodhi and colleagues linked GLP-1 drugs used for weight loss with a higher risk of bowel obstruction, but on a small number of cases. The team drew on a random sample of 16 million patients in the PharMetrics Plus claims database from 2006 to 2020 and compared 4144 liraglutide users and 613 semaglutide users with 654 people taking bupropion-naltrexone, a weight-loss drug that does not act on GLP-1 receptors.

The adjusted hazard ratio for bowel obstruction was 4.22, with a 95% confidence interval of 1.02 to 17.40. An interval that wide means the true effect could be anywhere from almost nothing to very large. The table behind that estimate shows 8.1 obstructions per 1000 person-years on liraglutide, 1.7 on bupropion-naltrexone and 0 on semaglutide, so the signal came from liraglutide.

The same study reported higher risks of pancreatitis and gastroparesis, which our articles on pancreatitis and lipase and gastric emptying cover. As a first look at weight-loss users rather than people with diabetes, the letter was useful. As a measure of how big the obstruction risk is, it was too small to say much.

Do larger studies confirm a higher risk of bowel obstruction?

UK, Scandinavia and Japan

Larger cohort studies give mixed answers, with the biggest ones finding no increase. Each compared new GLP-1 users with new users of SGLT2 inhibitors, another diabetes drug class, which helps make the groups comparable.

In the UK, Faillie and colleagues used the Clinical Practice Research Datalink for 2013 to 2019 and followed 25,617 GLP-1 users and 67,261 SGLT2 users, publishing in Clinical Pharmacology & Therapeutics in 2022. Intestinal obstruction needing hospital care occurred at 1.9 versus 1.1 per 1,000 person-years, a hazard ratio of 1.69 (95% CI 1.04 to 2.74). The ratio was highest, 3.48, after 1.6 years of use. The authors calculated that 1,223 people would need to take a GLP-1 drug for a year for one extra obstruction.

Ueda and colleagues then pooled national registers from Sweden, Denmark and Norway for 2013 to 2021, in Clinical Gastroenterology and Hepatology in 2024. Among 121,254 GLP-1 users and 185,027 SGLT2 users there were 557 obstruction events. The rate was 1.3 per 1000 person-years on GLP-1 drugs and 1.6 on SGLT2 inhibitors, a hazard ratio of 0.83 (95% CI 0.69 to 1.01). The authors concluded that earlier safety signals were not confirmed.

A 2026 Japanese claims study by Matsuo and colleagues in Diabetes, Obesity and Metabolism looked at 298,124 adults with type 2 diabetes: 25,779 starting a GLP-1 drug and 272,345 starting an SGLT2 inhibitor. The 1-year rate of obstruction or ileus severe enough to need tube decompression or surgery was 0.14% versus 0.17%, a hazard ratio of 0.80 (95% CI 0.52 to 1.24).

What do the randomized trials and meta-analyses show?

Pooling the evidence

Pooled randomized trials do not show a clear increase in bowel obstruction or ileus with GLP-1 drugs. Chiang and colleagues combined 55 placebo-controlled trials with 106,395 participants in Gastroenterology in 2025. GLP-1 drugs raised the risk of gallstones (risk ratio 1.46) and reflux (risk ratio 2.19), but the authors judged that they probably have little or no effect on other gut events, intestinal obstruction and paralytic ileus included.

A network meta-analysis by Chen and colleagues, published in 2026, gathered 50 randomized trials with 192,359 participants and compared individual drugs. No GLP-1 drug showed a significant increase in obstruction. Liraglutide was linked to a lower risk, with an odds ratio of 0.44, and the clearest signal across all drugs was for the SGLT2 inhibitor canagliflozin, at an odds ratio of 2.56.

A meta-analysis by Alfehaid and colleagues in Expert Opinion on Drug Safety mixed trials with observational studies. Across 6 studies and 550,426 participants, the odds ratio was 1.95 with a confidence interval of 0.43 to 8.79, which is not significant, and heterogeneity was very high (I2 = 94%). In its liraglutide subgroup the odds ratio was 3.0. Liraglutide therefore looks risky in some analyses and protective in others, which is a sign that the data cannot yet separate one GLP-1 drug from another on this outcome.

Is tirzepatide or semaglutide harder on the bowel?

A head-to-head cohort

Tirzepatide, semaglutide and dulaglutide had similar rates of severe gut events in the largest direct comparison so far. Crisafulli and colleagues, publishing in Annals of Internal Medicine, matched adults with type 2 diabetes starting each drug, with 65,238 matched pairs in the semaglutide versus dulaglutide comparison. The outcome combined acute pancreatitis, biliary disease, bowel obstruction, gastroparesis and severe constipation.

The hazard ratios were 0.96 for semaglutide versus dulaglutide, 0.96 for tirzepatide versus dulaglutide and 1.07 for tirzepatide versus semaglutide, none significant. Bowel obstruction was one part of that combined outcome rather than its own headline result. For readers weighing the two main drugs more broadly, our semaglutide vs tirzepatide comparison covers dosing and side effects.

Do opioid painkillers add to the risk on a GLP-1?

Two gut-slowing drugs at once

The type of opioid appears to matter for people already on a GLP-1. Bea and colleagues studied 411,188 adults with type 2 diabetes on a GLP-1 drug who started oxycodone, hydrocodone or tramadol, in a Diabetes Care paper published in 2026. They measured the 30-day risk of severe constipation, bowel obstruction or gastroparesis.

That risk was 0.51% with oxycodone, 0.35% with hydrocodone and 0.33% with tramadol. Oxycodone carried a higher risk than hydrocodone (risk ratio 1.48) and tramadol (risk ratio 1.55), driven by severe constipation and bowel obstruction rather than gastroparesis. The study did not include people without a GLP-1, so it shows how opioids compare with each other, not how much the GLP-1 adds. Post-surgery pain relief is the most common setting for this pairing; our guide to surgery and anesthesia on a GLP-1 covers the rest of that planning.

Which symptoms separate ordinary constipation from a possible blockage?

When to stop managing at home

Pain that keeps getting worse, a swollen belly, repeated vomiting and an inability to pass stool or gas together point toward a blockage rather than ordinary constipation. Constipation on a GLP-1 is usually uncomfortable but steady. An obstruction tends to escalate, and the vomiting is often what makes it hard to keep fluids down.

The everyday side of this, including why the standard fibre advice can backfire when meals shrink, is covered in our article on constipation on GLP-1 protocols.

Does a GLP-1 microdose lower the risk of bowel obstruction?

What the data can and cannot say

No study has measured bowel obstruction at microdose levels, so the honest answer is that the risk at low doses is unknown. Obstruction is rare enough that even studies of hundreds of thousands of people struggle to agree on whether standard doses raise it. A low-dose regimen would need far larger numbers to show anything.

Constipation, the more common bowel effect, offers a partial hint. In Ozempic's diabetes trials, constipation rates were 5% on 0.5 mg and 3.1% on 1 mg, far below the 24% seen with Wegovy 2.4 mg in weight-management trials. But these trials enrolled different people for different reasons, the 0.5 mg arm actually had more constipation than 1 mg, and the 7.2 mg Wegovy dose barely differed from 2.4 mg (20% versus 19%). The dose link for constipation is weaker than many people assume.

What can be said with more confidence is practical. Anyone with past abdominal surgery or a prior obstruction should mention it before starting a GLP-1 at any dose. New pain with vomiting deserves prompt attention, and other gut-slowing medicines, opioids especially, belong in the conversation with the prescriber. For how gradual dosing affects the more common side effects, see the gentle ramp.

Frequently Asked Questions