GLP-1 Before an Endoscopy? Retained-Food Evidence (2026)

If you take Ozempic, Wegovy, Mounjaro, Zepbound, Rybelsus or any other GLP-1 drug and you have an upper endoscopy (EGD) scheduled, the question you're searching is straightforward: do I stop the drug first, and for how long? The honest answer is that the experts disagreed, and the safest move is to not decide this on your own. GLP-1 drugs slow gastric emptying, so the stomach can still hold food and fluid after a standard overnight fast — which raised concern about regurgitation and aspiration under sedation. In a pooled analysis of 23 studies and 262,018 patients, GLP-1 users were about 4.5× more likely to have retained gastric contents at endoscopy and more likely to have the procedure stopped early — but the rate of aspiration pneumonia was not significantly increased (Baig 2025 ^[1]). In June 2023 the American Society of Anesthesiologists (ASA) told patients to hold the drug; gastroenterology and endoscopy groups pushed back that blanket holding may be unnecessary and an individualized plan plus a clear-liquid diet is reasonable (Hashash 2024 ^[2]; Sharaiha 2025 ^[3]). This article walks through that disagreement, the retained-food evidence specific to EGD, and what it means for your prep. For the general surgery/anesthesia version see holding a GLP-1 before surgery, and for the lower-scope companion see GLP-1 before a colonoscopy.

The honest summary

• The concern is real but specific to retained food, not proven harm. At upper endoscopy, GLP-1 users had roughly 4.5× the odds of retained gastric contents and of needing the procedure aborted — but pooled data did not show a significant increase in aspiration pneumonia (Baig 2025^[1]).
• ASA said hold it (June 2023). The anesthesiologists' consensus advised stopping daily-dosed GLP-1s on the day of the procedure and weekly-dosed ones (semaglutide as Ozempic/Wegovy, tirzepatide as Mounjaro/Zepbound) about a week before (described in Hashash 2024^[2]).
• GI and endoscopy societies pushed back. The AGA's rapid update argued for an individualized approach rather than reflexive cancellation (Hashash 2024^[2]), and the ASGE position statement favored a 24-hour clear-liquid diet for GLP-1 users over routinely holding the drug (Sharaiha 2025^[3]).
• The 2024/2025 multisociety consensus split the difference. It favored shared decision-making plus a clear-liquid diet for at least 24 hours before — the kind of prep colonoscopy already uses — and reserved a strict hold or “full stomach” precautions for higher-risk situations (Kindel 2025^[4]).
• A one-week hold may not even empty the stomach. In one EGD study, holding semaglutide for fewer than 8 days still left a ~10-fold higher chance of retained contents; only stopping more than 14 days (no symptoms) reliably normalized it (Santos 2024^[5]).
• Longer use seems to matter. Smaller endoscopy series suggest the retained-food signal is strongest in people on the drug for many months (Wen 2025^[10]), though numbers are small.
• Bottom line: don't stop on your own. Protocols differ by center. Coordinate with the prescriber and the endoscopy unit, and follow your facility's written instructions.

Why an upper endoscopy raises the question

GLP-1 receptor agonists slow how fast the stomach empties — that delayed emptying is part of how they curb appetite. The downside is that your stomach can still contain solids or fluid even after the usual “nothing to eat after midnight” fast. An upper endoscopy (esophagogastroduodenoscopy, or EGD) is almost always done under sedation or anesthesia, and a stomach that isn't empty creates a theoretical risk of regurgitating and aspirating contents into the lungs. During an EGD the endoscopist can actually see the retained food, which is why this procedure produced the clearest early signal. The first hard data came from Silveira 2023^[6], a retrospective study of 404 patients having upper endoscopy: those who'd taken semaglutide within 30 days had retained gastric contents 24.2% of the time versus 5.1% of non-users — an adjusted odds ratio of about 5.2 — even though most had stopped the drug roughly 10 days before. A prospective gastric-ultrasound study in volunteers found the same direction of effect (Sherwin 2023^[8]).

Those single-center findings have since been replicated and pooled. Gu 2025^[7] retrospectively reviewed EGDs and again found semaglutide was a strong independent predictor of retained gastric contents. Baig 2025^[1] then combined 23 studies and 262,018 patients and found GLP-1 users had about 4.5× the odds of retained gastric contents (pooled OR ~4.5) and 4.5× the odds of having the endoscopy terminated early — but, importantly, no statistically significant increase in aspiration pneumonia. A second dedicated meta-analysis focused specifically on the aspiration endpoint reached a similar conclusion: retained contents are clearly more common, but documented pulmonary aspiration events remained rare across roughly 210,000 patients (Elmati 2025^[9]). In other words, the “food in the stomach” finding is robust; the leap from that to actual lung injury is not well supported by the outcome data so far. That gap is exactly why the specialties disagreed about how aggressively to hold the drug.

What the ASA said vs. what the endoscopy societies said

In June 2023 the American Society of Anesthesiologists issued consensus-based guidance recommending that patients hold their GLP-1 before elective procedures: skip the dose on the day of the procedure for daily-dosed agents, and stop weekly-dosed agents about a week beforehand. The ASA itself acknowledged the evidence base was sparse — limited largely to case reports — yet in practice the guidance led some anesthesia teams to cancel or postpone endoscopies when patients hadn't held the drug (described in Hashash 2024^[2]).

Gastroenterologists pushed back quickly. The American Gastroenterological Association's rapid clinical practice update (Hashash 2024^[2], Clinical Gastroenterology & Hepatology) argued for an individualized approach rather than blanket discontinuation — weighing why you take the drug, whether you have ongoing GI symptoms, how long you'll have fasted, and whether the procedure is urgent. Their reasoning: stopping a GLP-1 in someone with diabetes can worsen blood-sugar control (which itself harms outcomes), holding a single weekly dose may not reliably restore normal stomach emptying anyway, and cancelling needed cancer-screening or diagnostic procedures carries its own risk. In August 2023 five GI groups (AGA, ACG, ASGE, AASLD, NASPGHAN) issued a joint statement noting there was little or no data on actual aspiration risk to support routinely stopping GLP-1s before endoscopy.

The ASGE later formalized that view in a 2025 position statement (Sharaiha 2025^[3]) on peri-endoscopic management of patients on GLP-1s and SGLT2 inhibitors. For GLP-1 users it favored a clear-liquid diet for 24 hours before the endoscopy as the primary risk-reduction strategy, rather than reflexively holding the drug — reserving a hold, point-of-care gastric ultrasound, or “full stomach” anesthesia precautions for patients with ongoing GI symptoms or other risk factors. The 2024/2025 multisociety guidance — written jointly by anesthesiology, gastroenterology and bariatric-surgery groups (Kindel 2025^[4]) — tried to reconcile the camps: it recommended shared decision-making balancing the need for the drug against individual aspiration risk, offered the same 24-hour clear-liquid path “as performed in patients undergoing colonoscopy and bariatric surgery,” and for higher-risk cases suggested gastric ultrasound on the day, a rapid-sequence anesthesia technique, or rarely rescheduling. It still acknowledged the original ASA hold timing as a reasonable option when a hold is chosen.

How big is the retained-food signal really?

The honest answer: the relative increase is large and consistent, but the absolute rates vary a lot between centers, and the studies are mostly small, retrospective, and single-center. Reported rates of retained gastric contents in GLP-1 users range from roughly a quarter of patients (Silveira 2023^[6]) up to extreme figures in small series of long-term users — one 144-patient analysis reported retained contents in about 40% of the semaglutide/tirzepatide group versus ~3% of non-users (Gu 2025^[7]), and a small Cureus series found that nearly all patients on a GLP-1 for more than six months had retained contents (Wen 2025^[10]). Those eye-popping numbers come from tiny subgroups and shouldn't be read as a population-wide rate. What's consistent across the literature is the direction: GLP-1 use raises the chance of a non-empty stomach at EGD, the effect appears larger with longer treatment and with ongoing GI symptoms, and despite all of this, documented aspiration pneumonia remains rare (Baig 2025^[1]; Elmati 2025^[9]). Where the evidence is genuinely thin is the link from “more retained food” to “more harm” — so treat strong-sounding percentages with caution.

What this means for your prep — the practical upshot

• Don't stop your GLP-1 on your own. The right plan depends on the drug, your dose schedule, why you take it, and your facility's protocol — which vary between centers. Self-discontinuing can also worsen diabetes control.
• Tell BOTH your anesthesiologist/endoscopy unit AND your prescriber, well ahead. Mention the exact drug (Ozempic, Wegovy, Mounjaro, Zepbound, Rybelsus, etc.), your dose, and when you last took it. Ask specifically: should I hold a dose, and should I extend my clear-liquid diet?
• Expect a possible extended clear-liquid diet. The ASGE position statement and the multisociety consensus both favor at least 24 hours of clear liquids before the procedure for GLP-1 users (Sharaiha 2025^[3]; Kindel 2025^[4]).
• Flag any ongoing GI symptoms. Nausea, vomiting, bloating, or feeling full are the strongest predictors of retained stomach contents (Silveira 2023^[6]; Santos 2024^[5]). If you have them near your procedure date, say so — it may change the plan, including delaying the test.
• Mention how long you've been on the drug. Longer treatment may carry a higher chance of retained food (Wen 2025^[10]), so it's worth telling your team.
• Follow YOUR facility's written instructions. Protocols genuinely differ; the consensus documents leave room for local judgment. Your endoscopy unit's sheet is the one that counts.

Bottom line

GLP-1 drugs do leave more food in the stomach at upper endoscopy — about 4.5× the odds in pooled data — but that has not translated into a significant increase in aspiration pneumonia in the evidence so far (Baig 2025^[1]; Elmati 2025^[9]). The ASA initially said hold the drug; gastroenterology and endoscopy societies argued for individualized assessment and a 24-hour clear-liquid prep instead of reflexive cancellation (Hashash 2024^[2]; Sharaiha 2025^[3]; Kindel 2025^[4]), and a strict one-week hold may not even empty the stomach reliably (Santos 2024^[5]). Because the protocols genuinely differ by center, the single safest thing you can do is not decide alone: tell both your prescriber and the endoscopy unit which GLP-1 you take, how long you've been on it and when you last took it, ask about holding a dose and extending clear liquids, and follow your facility's written instructions.

This article is educational and is not medical advice. Every claim above is sourced to a peer-reviewed study, society position statement, or multisociety consensus indexed in PubMed, verified against the live PubMed database (and, for the consensus and position-statement wording, the open-access fulltext) before publication. The evidence is largely retrospective and single-center, and the link from retained gastric contents to actual aspiration harm remains uncertain. Coordinate your own pre-procedure plan with your prescriber and endoscopy team.