Dose Insider / Analyses / GLP-1 bowel obstruction

Bowel Obstruction and Ileus on a GLP-1What the Labels and Studies Show

Updated November 20, 2026 路 21 min read 路 FDA labels 2026, three cohort studies, three meta-analyses

Ileus and intestinal obstruction are listed on the labels of Ozempic, Wegovy, Zepbound and Mounjaro among the reactions reported after approval, with no estimate of how often they happen. The studies disagree. A UK cohort found 1.9 hospital admissions for obstruction per 1,000 person-years on a GLP-1 medicine against 1.1 on a comparison drug, a hazard ratio of 1.69.

1.9 vs 1.1hospital admissions for intestinal obstruction per 1,000 person-years on a GLP-1 medicine and on a comparison drug in a UK cohort; a Scandinavian cohort found no increase
Obstructions per 1,000 a yearTwo cohort studies of people with type 2 diabetesUK COHORTGLP-1 medicine1.9SGLT-2 inhibitor1.1SCANDINAVIAN COHORTGLP-1 medicine1.3SGLT-2 inhibitor1.6Obstructions per 1,000 ayearTwo cohort studies of people with type 2 diabetesUK COHORTGLP-1 medicine1.9SGLT-2 inhibitor1.1SCANDINAVIAN COHORTGLP-1 medicine1.3SGLT-2 inhibitor1.6

The larger Scandinavian cohort found no increase on GLP-1 medicines.

On this page

A larger Scandinavian cohort found 1.3 against 1.6 and no increase. A U.S. claims study reported a hazard ratio of 4.22, driven by liraglutide, with no case among semaglutide users. Pooled randomized trials found probably little or no effect. The warning signs are the same whatever the cause: being unable to pass stool or gas, a swollen abdomen that does not go down, and vomiting that keeps on.1,5,6,7,9,11

Obstruction, ileus and impaction: what the words mean

MedlinePlus, the medical encyclopedia of the U.S. National Library of Medicine, defines intestinal obstruction as a partial or complete blockage of the bowel, so that the contents of the intestine cannot pass through it. It separates two kinds.11

The labels of the four medicines use three terms side by side: ileus, intestinal obstruction, and severe constipation including fecal impaction.1,2,3,4 They form a chain. Constipation is slow passage. Impaction is stool that has hardened and cannot be passed, which MedlinePlus counts among the mechanical causes of obstruction. Ileus is a bowel that has stopped moving with nothing in the way. A medicine that slows the gut could in principle contribute to the second and third. Whether these medicines do, and how often, is what the studies below tried to measure.

What the four labels say

In the section on reactions reported after approval, each of the four labels lists, under gastrointestinal reactions: ileus, intestinal obstruction, and severe constipation including fecal impaction. Each prefaces the list with the same caution. These reactions are reported voluntarily from a population of uncertain size, so it is not always possible to reliably estimate their frequency or establish a causal relationship to drug exposure.1,2,3,4

The clinical trials in the labels do not list obstruction or ileus among the reactions seen in the trials. What they give is the share of people with a severe stomach or gut reaction of any kind, under a warning headed "Severe Gastrointestinal Adverse Reactions".

Severe gut reactions of any kindFDA labels
Medicine

These figures cover every kind of severe stomach or gut reaction in the trials. None of the labels gives a separate number for obstruction or ileus.

The same warning ends with a restriction: the medicine is not recommended in patients with severe gastroparesis, a stomach that empties too slowly.1,2,3,4 Our page on Ozempic and gastroparesis covers that question.

Six studies, side by side

A list of reports cannot say whether a drug raises a risk. Comparative studies can, and six of them have looked at intestinal obstruction on GLP-1 medicines. Three are cohort studies of people in ordinary care and three are meta-analyses.

StudyWho was comparedResult for GLP-1 medicines
UK cohort, 202225,617 new users with type 2 diabetes against 67,261 users of SGLT-2 inhibitors, 2013 to 20191.9 against 1.1 admissions per 1,000 person-years; hazard ratio 1.69 (1.04 to 2.74)
Scandinavian cohort, 2024121,254 new users against 185,027 users of SGLT-2 inhibitors in Sweden, Denmark and Norway, 2013 to 20211.3 against 1.6 per 1,000 person-years; hazard ratio 0.83 (0.69 to 1.01)
U.S. claims study, 20234,757 users with obesity and no diabetes against 654 users of bupropion-naltrexone, 2006 to 2020Hazard ratio 4.22 (1.02 to 17.40); 73 cases on liraglutide, none on semaglutide
Meta-analysis of comparative studies, 20256 studies, 550,426 participantsOdds ratio 1.95 (0.43 to 8.79); for liraglutide 3.0 (2.03 to 4.45)
Meta-analysis of placebo-controlled trials, 202555 trials, 106,395 participantsProbably little or no effect on intestinal obstruction or paralytic ileus
Network meta-analysis of trials, 202650 trials, 192,359 participantsNo GLP-1 medicine with an increase; liraglutide odds ratio 0.44 (0.24 to 0.81)

Faillie 2022, Ueda 2024, Sodhi 2023, Alfehaid 2025, Chiang 2025, Chen 2026. Ranges in brackets are 95% confidence intervals.5,6,7,8,9,10

Each study's estimateSix studies
Show

A ratio of 1 means no difference from the comparison group. Coral bars have an interval entirely above 1, mint bars an interval that includes 1 or lies below it. Hazard ratios and odds ratios are not the same measure; they are shown together only to compare direction and size. The meta-analysis of 55 placebo-controlled trials is not on the chart because we read it as an abstract that gives no ratio for obstruction.

The UK cohort used primary care records linked to hospital data. It counted intestinal obstruction that required hospitalization and compared new users of GLP-1 medicines with new users of SGLT-2 inhibitors, another class of diabetes drugs. The hazard ratio was 1.69. The highest hazard ratio, 3.48, was seen after 1.6 years of use, and the authors put the number needed to harm after one year at 1,223: one additional obstruction for every 1,223 people treated for a year. The same study found a larger association for DPP-4 inhibitors, a different class that acts on the same hormone system: 2.7 against 1.0 per 1,000 person-years.6

The Scandinavian cohort repeated the design in the national registers of three countries, with almost five times as many GLP-1 users. It recorded 557 obstruction events and an adjusted rate of 1.3 per 1,000 person-years on GLP-1 medicines against 1.6 on SGLT-2 inhibitors. The authors concluded that the earlier safety signals were not confirmed.7

The U.S. claims study is the one behind most headlines. It looked at people with obesity and without diabetes, and compared semaglutide or liraglutide with bupropion-naltrexone. Bowel obstruction was recorded at 8.1 per 1,000 person-years on liraglutide (73 cases), 1.7 on the comparison drug (2 cases), and not at all among 613 semaglutide users. The adjusted hazard ratio of 4.22 has an interval from 1.02 to 17.40, and in the authors' sensitivity analysis that left hyperlipidemia out of the model it was 3.63 with an interval from 0.87 to 15.10, which includes 1.5

The meta-analysis of comparative studies screened 317 records, reviewed 14 studies and pooled 6. Its pooled odds ratio of 1.95 has an interval from 0.43 to 8.79, and the studies disagreed with each other so much that the authors reported 94% heterogeneity. Their conclusion was that GLP-1 receptor agonists as a class showed no clear increase in the risk of bowel obstruction or ileus, and that liraglutide is associated with a higher risk than semaglutide.8

The two meta-analyses of randomized trials point away from a drug effect. One pooled 55 placebo-controlled trials and concluded that these medicines probably have little or no effect on intestinal obstruction or paralytic ileus. The other pooled 50 trials by drug and dose. It found no GLP-1 medicine whose increase reached statistical significance, and for liraglutide a lower risk than in the control groups, an odds ratio of 0.44.9,10

How a GLP-1 could affect the bowel

The mechanism is a hypothesis, and the sources treat it as one. What is established is the effect higher up: all four labels state that the medicine delays the emptying of the stomach.1,2,3,4 For the intestine, the joint nutrition advisory of four U.S. medical societies says only that these medicines occasionally affect intestinal motility or secretions.13 The UK investigators opened their paper with the statement that GLP-1 medicines and DPP-4 inhibitors might increase the risk of intestinal obstruction and that real-world evidence was lacking, which is why they ran the study.6

Three indirect routes follow from the sources without any new assumption.

None of these has been measured in a study of people on a GLP-1. They are reasons the question is plausible, and no more than that.

Why the studies disagree

Four differences between the studies account for much of the spread.

The comparison drug. Both European cohorts compared GLP-1 medicines with SGLT-2 inhibitors as the reference group. The 2026 network meta-analysis of trials suggests that this reference is not neutral for one drug of that class: canagliflozin was associated with more intestinal obstruction than control therapies, an odds ratio of 2.56 with an interval from 1.01 to 6.49.10 If the comparison group carries a risk of its own, a hazard ratio against it moves down. The U.S. study used a different comparison, a weight-loss drug, in a different population.

The medicine. Liraglutide, a daily injection, dominates the older data. In the U.S. study all 73 cases among GLP-1 users were on liraglutide. The meta-analysis of comparative studies found the association for liraglutide and not for the class.5,8 Then the trials complicate it again: in randomized comparisons liraglutide had fewer obstructions than control groups.10 None of the six studies reports a result for tirzepatide alone.

What counted as a case. The UK cohort counted obstruction that required hospitalization. The U.S. study counted the first diagnosis code in claims.5,6 One possibility, which none of the papers tests, is that people on a medicine known for gut reactions are examined for obstruction more readily, so that more diagnoses are written down.

Size. The U.S. comparison group had 2 cases, which is why its interval runs from 1.02 to 17.40. The Scandinavian cohort had 557 events and an interval from 0.69 to 1.01.5,7 When a small study and a large one disagree, the large one carries more weight, though it does not settle a question about a different population.

Taken together: the best-powered cohort and both sets of pooled trials found no increase, two smaller observational studies found one, and all of them agree that the event is uncommon.

The fine print of each study

The headline ratio of each study hides details that change how much weight it deserves.

StudyDetail
UK cohortFor DPP-4 inhibitors the hazard ratio was 2.59 (1.52 to 4.42), rising to 9.53 after 1.8 years of use, with a number needed to harm of 603 after one year.
Scandinavian cohortFor DPP-4 inhibitors: 2.0 against 1.8 per 1,000 person-years, hazard ratio 1.13 (0.96 to 1.34), among 190,321 new users and 919 events.
U.S. claims studyThree versions of the analysis: 4.22 (1.02 to 17.40) in the main one, 3.63 (0.87 to 15.10) without hyperlipidemia in the model, and 2.44 (1.00 to 5.95) with a looser definition of obesity.
Network meta-analysis of trialsEight databases searched through January 21, 2025. Canagliflozin: number needed to harm 658. Liraglutide: an absolute risk reduction of 0.34%.

Faillie 2022, Ueda 2024, Sodhi 2023, Chen 2026.5,6,7,10

Two things stand out. In both European cohorts the signal for DPP-4 inhibitors, a class of diabetes tablets, was as large as or larger than the signal for GLP-1 medicines, and in Scandinavia neither was confirmed. And the U.S. estimate shrinks with each reasonable change to the analysis, down to an interval whose lower end sits at 1.00.

How large the risk is in absolute numbers

A ratio says nothing about how many people are affected. The rates do. In the two European cohorts intestinal obstruction occurred in 1 to 2 of every 1,000 people a year, on a GLP-1 medicine or on the comparison drug.6,7 The calculator applies each study's rates to a number of years.

Per 1,000 people, over my yearsThree cohorts
Study

The arithmetic multiplies each study's yearly rate by the years entered, which assumes the rate stays level. The UK study found the association strongest after 1.6 years, so that assumption is rough. The three studies followed different people with different comparison drugs, and their numbers should not be mixed.

Even the study with the highest hazard ratio describes a small absolute difference. In the UK cohort the gap was 0.8 admissions per 1,000 people a year. In the Scandinavian cohort the gap ran the other way, 0.3 fewer.6,7

Who is more likely to have an obstruction

The causes MedlinePlus lists are also the risk factors a prescriber would want to know about, with or without a GLP-1.11

A 2025 case report shows how these combine. A 59-year-old woman who had been on semaglutide for a year, with a recent dose increase, came to an emergency department with severe abdominal pain and about 15 episodes of vomiting within hours of an injection. Her history included a colectomy for perforated diverticulitis, a colostomy with reversal, eight abdominal hernia repairs, removal of the gallbladder and an appendectomy. A CT scan showed dilated loops of small bowel. She was managed without surgery, with bowel rest and a tube to decompress the stomach, and went home on the fifth day. The authors wrote that a direct causal relationship cannot be established, and that postoperative adhesions are a well-established cause of small bowel obstruction in patients like her. They recommend caution when prescribing GLP-1 medicines to people with a history of abdominal surgery and adhesions.12

For anyone with one of these in their history, the sources suggest three things to raise before starting a GLP-1 or at the next visit. The first is the history itself: operations on the abdomen or pelvis, a hernia, Crohn disease, an earlier obstruction. The second is constipation, because the labels list severe constipation with impaction beside obstruction and constipation is the stage that can be treated. The third is the plan for the four warning signs below: whom to call, and where to go out of hours.

One case proves nothing about frequency. It does describe the patient for whom the question matters most: someone whose bowel is already narrowed or tethered, in whom slower movement leaves less margin.

What applies to meMedlinePlus
I have or have had

The list follows the causes of obstruction and ileus in MedlinePlus. It does not score a risk; none of the studies here measured how much each factor adds on a GLP-1. A previous obstruction is on the list because the case report's authors asked about it, not because a study measured it.

From constipation to impaction

The third term on the labels, severe constipation including fecal impaction, is the one most people on these medicines can do something about. Constipation is common on them: 24% on Wegovy and 11% to 17% on Zepbound in the label trials.2,3 Impacted stool is on MedlinePlus's list of mechanical causes of obstruction.11

None of the studies above measured how often constipation on a GLP-1 progresses to impaction. The practical point is the order of events: impaction lies at the far end of constipation, so constipation is the stage at which there is still something to do. Our page on constipation on a GLP-1 covers fiber, liquids and what a gastroenterology guideline found for each laxative.

The warning signs

MedlinePlus lists these symptoms of intestinal obstruction:11

Most of that list is also the list of common reactions to a GLP-1 medicine, which is the difficulty. Nausea, vomiting, constipation, diarrhea, abdominal pain and distension all appear in the label tables. MedlinePlus's instruction on when to contact a provider picks out what is different. Contact your provider if you:11

What sets the four apart is persistence: pain or swelling that does not go away, vomiting that keeps on, and nothing passing. MedlinePlus lists dehydration, electrolyte imbalances, infection and a hole in the intestine among the complications of obstruction, and says that a blockage that cuts off the blood supply can cause tissue death.11 The National Institute of Diabetes and Digestive and Kidney Diseases is more explicit about timing. Its guidance on constipation says to see a doctor right away when constipation comes with an inability to pass gas, vomiting, or constant pain in the abdomen.14

Is this more than a side effect?MedlinePlus
Right now I
I also have

The four signs are MedlinePlus's reasons to contact a provider about a possible obstruction. A page cannot examine you. If something feels seriously wrong, do not wait for it to agree.

How an obstruction is found and treated

During an exam the provider may find bloating, tenderness or a hernia in the abdomen, MedlinePlus says. The tests that show an obstruction are an abdominal CT scan, an abdominal x-ray, a barium enema, and an upper GI and small bowel series.11

Treatment depends on the kind. A partial obstruction of the small bowel without complications can be managed by bowel rest, meaning nothing by mouth, with intravenous fluids and a tube passed through the nose into the stomach to suction out fluid and gas, which relieves the swelling and the vomiting. Surgery may be needed if the tube does not relieve the symptoms, if there are signs of tissue death, or if a tumor is the cause.11 The woman in the case report was treated in the first way and was passing bowel movements again on the fourth hospital day.12

On the outlook, MedlinePlus is brief: the outcome depends on the cause of the blockage, and most of the time the cause is successfully treated.11

The reason for acting early is in its list of complications: electrolyte imbalances, dehydration, a hole in the intestine, infection and jaundice. If the obstruction blocks the blood supply to the intestine, it may cause infection and tissue death, and MedlinePlus notes that the risk of tissue death depends on the cause of the blockage and on how long it has been present. Hernia, volvulus and intussusception carry the higher risk.11

What happens to the GLP-1 medicine afterwards is not covered by any of the sources here. The labels give no instruction specific to obstruction or ileus. The patient in the case report chose on her own to stop semaglutide after discharge.12 That decision belongs with the prescriber and the surgical team, who will know what caused the blockage.

Around an operation

Abdominal surgery is on MedlinePlus's list of causes of ileus: the bowel can stop moving for a time after an operation with no blockage present.11 Surgery also leaves the adhesions that cause mechanical obstruction later. Someone on a GLP-1 who is about to have an abdominal operation therefore has two reasons to make sure the surgical team knows about the medicine.

The labels add a third, which concerns the anesthesia and not the bowel. All four carry a warning about pulmonary aspiration during general anesthesia or deep sedation: there have been rare reports of people on these medicines who had residual stomach contents despite following the fasting instructions. The labels say the data are insufficient to recommend changing the fasting period or pausing the medicine, and they tell patients to inform their providers before any planned surgery or procedure.1,2,3,4

After the operation, the same warning signs apply as at any other time, and it is the surgical team's call whether and when the GLP-1 is restarted.

When it happens

The common stomach and gut reactions to these medicines cluster around the dose steps: the labels say that most reports of nausea, vomiting and diarrhea came during dose escalation.1,3,4 The one study that looked at timing for obstruction found a different pattern. In the UK cohort the hazard ratio was highest after 1.6 years of use, at 3.48 with an interval from 1.79 to 6.79.6 If that pattern is real, obstruction is not an early reaction that people grow out of, and a long uneventful stretch on the medicine is not a reason to dismiss the warning signs.

The case report describes the opposite timing, an obstruction within hours of an injection after a recent dose increase, in someone with heavy surgical scarring.12 The Scandinavian cohort, which found no excess, had a median follow-up of 0.9 years for GLP-1 users.7 No study here followed people on the weight-management doses for several years.

Diabetes or weight management

The evidence comes almost entirely from people with type 2 diabetes. Both European cohorts enrolled users of diabetes medicines, and the comparison drugs were diabetes drugs.6,7 The U.S. claims study is the exception: it selected people with a code for obesity and excluded anyone with a code for diabetes or a diabetes drug. Its data end in June 2020, before semaglutide was marketed for weight loss, and it did not include tirzepatide.5

The meta-analysis of 55 placebo-controlled trials covered people with type 2 diabetes, overweight or obesity, and fatty liver disease, and concluded for the whole set that the medicines probably have little or no effect on obstruction or ileus.9 For Wegovy at 2.4 mg and for Zepbound, the direct evidence on obstruction is the trial data in those pooled analyses and the reports listed on the labels.

Short answers to common questions

Can Ozempic cause a bowel obstruction?

Ileus and intestinal obstruction are on its label as reactions reported after approval, without a frequency. Comparative studies disagree: a UK cohort found a hazard ratio of 1.69, a larger Scandinavian cohort found no increase, and pooled randomized trials found probably little or no effect.

How common is it?

Uncommon. In the two European cohorts it occurred in 1 to 2 of every 1,000 people a year, with or without a GLP-1 medicine.

What is ileus?

A bowel that does not move its contents although nothing is blocking it. MedlinePlus also calls it paralytic ileus or pseudo-obstruction.

What are the warning signs?

Being unable to pass stool or gas, a swollen abdomen that does not go away, vomiting that keeps on, and abdominal pain that does not go away. MedlinePlus says to contact a provider for any of them, and NIDDK says to see a doctor right away when constipation comes with an inability to pass gas, vomiting or constant abdominal pain.

Is the risk higher with Wegovy or Zepbound than with Ozempic?

No study here compared them. The older observational signal is mainly for liraglutide; the U.S. claims study recorded no obstruction among 613 semaglutide users, and none of the studies reports tirzepatide separately.

Does previous abdominal surgery matter?

Adhesions from surgery head MedlinePlus's list of mechanical causes of obstruction. A 2025 case report of obstruction on semaglutide involved a woman with several abdominal operations, and its authors advise caution in such patients.

Can constipation turn into an obstruction?

Impacted stool is one of the mechanical causes MedlinePlus lists, and the labels report severe constipation including fecal impaction. How often that happens on a GLP-1 has not been measured in the studies here.

How is it treated?

A partial obstruction without complications is often managed with bowel rest, intravenous fluids and a tube to decompress the stomach. Surgery may be needed if that fails or if tissue is at risk.

How this page was built

Every number was read in its source on November 20, 2026. The label statements come from the current prescribing information of the four medicines. The U.S. claims study and the case report were read in full text. The two European cohorts and the three meta-analyses were read as abstracts, which give the main results and not the details of how cases were defined or adjusted. The definitions, symptoms, tests and treatment come from the MedlinePlus encyclopedia entry reviewed in April 2026.

The limits are these. The six studies used different populations, comparison drugs and definitions, and their ratios are shown together only to compare direction. Most of the data concern people with diabetes and older medicines, liraglutide above all; none reports tirzepatide on its own. The calculator multiplies a yearly rate by years, which assumes a level rate. The list of risk factors comes from general causes of obstruction and was not tested on people taking a GLP-1. This page does not suggest doses or advise stopping a medicine.

Sources

  1. OZEMPIC (semaglutide) injection prescribing information, NDA 209637/S-039, revised May 2026. U.S. Food and Drug Administration; sections 5, 6.2 and 7. accessdata.fda.gov
  2. WEGOVY (semaglutide) prescribing information, NDA 215256/S-025, revised June 2026. U.S. Food and Drug Administration; sections 5, 6.1, 6.2 and 7. accessdata.fda.gov
  3. ZEPBOUND (tirzepatide) prescribing information, NDA 217806/S-041, revised August 2026. U.S. Food and Drug Administration; sections 5, 6.1, 6.2 and 7. accessdata.fda.gov
  4. MOUNJARO (tirzepatide) prescribing information, NDA 215866/S-052, revised April 2026. U.S. Food and Drug Administration; sections 5, 6.2 and 7. accessdata.fda.gov
  5. Sodhi M, Rezaeianzadeh R, Kezouh A, Etminan M. Risk of gastrointestinal adverse events associated with glucagon-like peptide-1 receptor agonists for weight loss. JAMA 2023;330:1795-1797. doi:10.1001/jama.2023.19574
  6. Faillie JL, Yin H, Yu OHY, et al. Incretin-based drugs and risk of intestinal obstruction among patients with type 2 diabetes. Clin Pharmacol Ther 2022;111:272-282. doi:10.1002/cpt.2430
  7. Ueda P, Wintzell V, Melbye M, et al. Use of DPP4 inhibitors and GLP-1 receptor agonists and risk of intestinal obstruction: Scandinavian cohort study. Clin Gastroenterol Hepatol 2024;22:1226-1237. doi:10.1016/j.cgh.2023.08.034
  8. Alfehaid L, Alyami M, Almohareb S, et al. Evaluating bowel obstruction and ileus events in patients on GLP-1 receptor agonists: a systematic review and meta-analysis. Expert Opin Drug Saf 2026;25:1057-1065, published online February 2025. doi:10.1080/14740338.2025.2465870
  9. Chiang CH, Jaroenlapnopparat A, Colak SC, et al. Glucagon-like peptide-1 receptor agonists and gastrointestinal adverse events: a systematic review and meta-analysis. Gastroenterology 2025;169:1268-1281. doi:10.1053/j.gastro.2025.06.003
  10. Chen JJ, Hsu CW, Hung CM, et al. Agent- and dose-specific intestinal obstruction safety of GLP-1 receptor agonists and SGLT2 inhibitors: a network meta-analysis of randomized trials. Int J Mol Sci 2026;27:608. doi:10.3390/ijms27020608
  11. MedlinePlus Medical Encyclopedia. Intestinal obstruction and ileus. U.S. National Library of Medicine; reviewed April 27, 2026, read November 20, 2026. medlineplus.gov
  12. Nagib N, Avila D, Partyka J, Kamyab A. Weighing the risks: small bowel obstruction associated with semaglutide use in the postoperative abdomen. Cureus 2025;17(7):e88589. doi:10.7759/cureus.88589
  13. Mozaffarian D, Agarwal M, Aggarwal M, et al. Nutritional priorities to support GLP-1 therapy for obesity: a joint advisory from the American College of Lifestyle Medicine, the American Society for Nutrition, the Obesity Medicine Association, and The Obesity Society. Obesity (Silver Spring) 2025;33:1475-1503. doi:10.1002/oby.24336
  14. National Institute of Diabetes and Digestive and Kidney Diseases. Symptoms and Causes of Constipation. Last reviewed May 2018, read November 20, 2026. niddk.nih.gov

Related analyses

Log each dose and each symptom, and bring the dates to your prescriber.

Open Dose Insider