GLP-1 medicines before surgery, anesthesia and colonoscopy: what each medical society says

Why anesthesia and endoscopy teams ask about GLP-1 medicines

GLP-1 medicines slow how fast the stomach empties. That is part of why people feel full for longer. It also means food or liquid can still be in the stomach after the usual fasting time before a procedure. Anesthesia and sedation relax the muscles that normally keep stomach contents down, so there is a small chance that this material travels into the windpipe and lungs. Doctors call that pulmonary aspiration. It is rare.

The same question comes up for upper endoscopy and colonoscopy, where sedation is common and a stomach that has not emptied can block the view or force the team to stop. The bodies below each tried to answer one question: what, if anything, should change around a procedure for someone who uses one of these medicines. They did not all land in the same place.

This page reports what each body says and what the studies under them found. It does not tell anyone to hold, continue or change a medicine. That decision belongs to the person and their own surgical, anesthesia and prescribing teams. For the same reason the page gives no hold schedule beyond what a named body has printed. The semaglutide and tirzepatide page explains how the medicines work, and questions to ask your prescriber has a list for the appointment.

What the FDA labels say

Ozempic, Wegovy, Zepbound and Mounjaro carry the same warning under the heading "Pulmonary aspiration during general anesthesia or deep sedation". In the parts of the labels the Atlas read, it makes four points.

• The medicine delays gastric emptying.
• There have been rare reports after the medicines reached the market of pulmonary aspiration in people having elective surgery or procedures that needed general anesthesia or deep sedation, including cases where stomach contents were still present despite reported adherence to fasting advice.
• The available data are not enough to say how to lower that risk, including whether changing fasting advice or temporarily stopping the medicine would reduce retained stomach contents.
• Prescribers are told to instruct patients to tell their healthcare providers about the medicine before any planned surgery or procedure.

The Ozempic and Mounjaro patient-counseling sections add that the medicine may make the stomach empty more slowly, which may lead to complications with anesthesia or deep sedation during planned procedures. The Atlas did not read that section of the Wegovy and Zepbound labels. One Wegovy label covers the injection and the tablets, and this warning names Wegovy without separating the two. In the parts the Atlas read, none of the labels tells patients to stop the medicine or to continue it. Labels change, so check the current text on DailyMed.

What each medical body says

The table runs from oldest to newest. Each row paraphrases the body's own document. Most of the bodies wrote for elective procedures, and the UK statement says its advice may fit less well in emergencies.

The multisociety guidance was issued by the ASA, AGA, American Society for Metabolic and Bariatric Surgery, International Society of Perioperative Care of Patients with Obesity and Society of American Gastrointestinal and Endoscopic Surgeons. The UK consensus came from the Association of Anaesthetists, Association of British Clinical Diabetologists, British Obesity and Metabolic Surgery Society, Centre for Perioperative Care, Joint British Diabetes Societies for Inpatient Care, Royal College of Anaesthetists, Society for Obesity and Bariatric Anaesthesia and UK Clinical Pharmacy Association. The Australian and New Zealand recommendations came from the Australian Diabetes Society, National Association of Clinical Obesity Services, Gastroenterological Society of Australia and Australian and New Zealand College of Anaesthetists. Dates are publication or posting dates as the Atlas found them. The Atlas read the multisociety guidance in its journal text. The ASA, AGA and Australian and New Zealand rows come from the bodies' web pages and summaries as the Atlas read them, and the AGA update's advice statements are behind a paywall, so that row follows the AGA's own web summary. The Korean row comes from the published abstract. The ASA page does not call its 2023 guidance withdrawn. It says the ASA later affirmed the value of the more recent multisociety document. The Atlas did not find a separate statement from the metabolic surgery societies or a European anesthesia society.

Where the statements split

A 2024 review read the statements of 14 anesthesia, endocrine and gastroenterology societies and found considerable variation on whether to withhold, for how long, on fasting and on rapid-sequence induction. Its authors tied the variation to a lack of high-quality data. The newer statements narrowed some of those gaps and left others open.

Cells paraphrase the documents and compress them, so read the source for the full wording. The labels use the same warning text for semaglutide and tirzepatide products.

The one randomized trial: OCULUS

OCULUS, published in JAMA Internal Medicine in 2026, randomized adults who took a GLP-1 or GLP-1/GIP medicine at a stable dose and were having elective upper endoscopy under moderate sedation or monitored anesthesia care. One group held a single dose before the procedure and the other continued as usual. People with planned general anesthesia, chronic opioid use or documented gastroparesis were left out.

The trial measured "clinically significant residual gastric volume": stomach contents that stopped the exam, forced an early stop or a breathing tube, or led to an aspiration event that needed extra monitoring or treatment. It was planned for 120 people and stopped at a preplanned halfway analysis of 60, when the difference crossed its stopping boundary.

Source: OCULUS, JAMA Internal Medicine 2026 (NCT06533527). Every case was stomach contents that prevented an adequate exam. No unplanned intubation, aspiration or hospital admission occurred, but the authors say the trial was not powered for rare events and that no observed events should not be read as proof of safety. The liquid-diet finding comes from a subgroup the trial was not powered to test. Upper GI symptoms did not predict who had retained contents, only one dose was held, and glucose control was not evaluated. The confidence intervals are 90%, as the paper reports them.

What the observational studies found

Most other evidence comes from comparing people who used GLP-1 medicines with people who did not, often from hospital or insurance records. These studies cannot show cause, and they disagree with one another.

These studies point in different directions. The UK cohort followed patients forward in time and found a large excess. Several large record-based studies found no excess or a lower risk of lung complications. Outcomes, procedures and designs differ, and none of the studies is a randomized comparison. The Atlas read the Garza, Barlowe, Chhabra, Chen and Potter studies as abstracts.

Colonoscopy and bowel preparation

For colonoscopy the practical question is whether the bowel is clean enough to see. Three meta-analyses of observational studies looked at it and did not agree.

BBPS is the Boston Bowel Preparation Scale, a standard score of how clean the bowel looks, where higher is cleaner. All three analyses agree that scores were slightly lower in users, and they disagree on whether more people had an inadequate prep. None pooled polyp or adenoma detection. The Chiu abstract says two studies found higher detection in users and one found no difference. The Atlas read the Ahmed and Chiu analyses as abstracts. A separate chart review at one hospital, Ghazanfar and colleagues in 2024, compared diets only among people on a GLP-1 medicine and found leftover food at upper endoscopy in 1.5% on a liquid or low-residue diet and 10% on a regular diet. It was retrospective and could not check whether patients followed the diet or took the medicine as prescribed.

Trials still running

ClinicalTrials.gov lists a randomized trial of a 24-hour clear liquid diet before upper endoscopy in people using GLP-1 medicines (NCT06654219, University of Texas Health Science Center at Houston, 136 people planned, recruiting when the Atlas looked). It also lists a randomized trial of a medicine that speeds stomach emptying before elective surgery in people using GLP-1 medicines for weight loss (NCT07100691, Mater Misericordiae University Hospital, 80 people planned). The registry still shows it as not yet recruiting, although its listed start and completion dates have passed. Neither has posted results.

Does the reason for taking it matter?

The labels and most statements treat the medicine the same whether it is used for diabetes or for weight. The multisociety guidance is one document for all uses. It says withholding the medicine only from people who take it for weight management could amount to obesity bias, and that for people with diabetes withholding can raise blood sugar, which has to be weighed against the aspiration risk. The UK and Australian and New Zealand statements raise the same blood sugar point. None of them gives a separate rule for weight management.

Terms you may hear on the day

Quick answers

Do the Ozempic, Wegovy, Zepbound and Mounjaro labels tell people to stop before surgery?

In the parts the Atlas read, no. The labels report rare cases of aspiration, say the data cannot show whether changing fasting advice or stopping the medicine would lower the risk, and tell prescribers to have patients mention the medicine before any planned procedure.

Do the medical societies agree on holding the medicine?

No. The ASA's 2023 guidance said to consider holding. The October 2024 multisociety guidance says most people can continue and describes holding as an option if the team finds the risk unacceptable. The UK, Australian and New Zealand statements advise against routine cessation before elective procedures, and the Korean statement leans toward continuing. A 2024 review of 14 societies found considerable variation.

Did the ASA change its advice?

The ASA issued consensus guidance in June 2023 that suggested considering a hold. In October 2024 it was one of five societies behind new guidance, and its 2023 page now says it affirmed the value of that more recent document. This page reports both and does not rank them.

Has holding the medicine been tested in a randomized trial?

For upper endoscopy, yes. OCULUS found less retained stomach content in people who held one dose, but it stopped early at 60 people and excluded planned general anesthesia. The Atlas did not find a randomized trial of holding against continuing before surgery under general anesthesia as of October 2026.

What about colonoscopy?

The AGA update and the multisociety guidance cover endoscopy as well as surgery. Studies of bowel preparation disagree on whether more users have an inadequate prep, and the evidence on whether polyp detection changes is thin.

Is tirzepatide treated differently from semaglutide?

The Mounjaro and Zepbound labels carry the same warning as Ozempic and Wegovy. The UK and Australian and New Zealand statements name tirzepatide alongside semaglutide, and the multisociety authors say their guidance may need changes for dual and triple agonists.

Does it matter whether the medicine is for diabetes or for weight?

The statements do not set separate rules by reason. They note that stopping can raise blood sugar in people with diabetes, and the multisociety guidance warns against singling out people who take the medicine for weight.

What about emergency surgery?

The 2023 ASA guidance says urgent and emergency cases go ahead and are treated as having a full stomach. The UK statement says its recommendations may be less applicable to emergency surgery and endoscopy.

Who decides what happens around a procedure?

The multisociety guidance and the UK consensus both describe it as a shared decision among the patient and the procedural, anesthesia and prescribing teams. The Atlas does not make that decision and cannot know the details of any one person's situation.

Questions to ask

• Who on my team is making the decision about my GLP-1 medicine around this procedure: the surgeon, the anesthesia team, the person who prescribes it, or all three?
• Does the team want to know which medicine I use, my dose, whether I am still increasing the dose and when I last took it?
• Is the plan going to include a diet change the day before, a stomach ultrasound or a change in how anesthesia is given?
• If I use the medicine for diabetes, who will manage my blood sugar around the procedure?
• Is my procedure planned or urgent, and does that change what the team does?
• Who do I contact, and when, if I have nausea, vomiting or belly pain close to the procedure date?

What is still unknown

• No randomized trial of holding against continuing before surgery under general anesthesia was found. OCULUS covers upper endoscopy and excluded planned general anesthesia.
• How many doses, if any, would need to be held to change stomach contents, and whether that differs by drug, dose or weekly versus daily forms, has not been tested.
• How often aspiration happens, and whether it differs by procedure, is unclear. The studies use different outcomes and point in different directions.
• Whether the 24-hour liquid diet works comes from one trial subgroup and from observational comparisons. Two registered trials may add data.
• Whether poorer bowel preparation changes polyp detection or the need for repeat colonoscopy is not settled.
• The Atlas did not read the patient-counseling sections of the Wegovy and Zepbound labels, the paywalled text of the AGA update or the full ASA, AGA and ANZCA documents.