# Holding a GLP-1 Before Endoscopy or Surgery: What the Trial Found

By Nora Bissett, Pricing Editor (https://glpwatchdog.org/authors/nora-bissett) | Published: 2026-08-19 | Updated: 2026-10-30

Canonical page: https://glpwatchdog.org/research/glp1-before-endoscopy-and-surgery

> A 2026 randomized trial put retained stomach contents at 3.1% when one dose was held against 25.0% when it was not, and found something more useful about the day-before diet.

## Key points

- OCULUS randomized trial: 3.1% retained contents holding one dose vs 25.0% continuing.
- Stopped early at 60 patients, so trust the direction more than the exact size.
- Nobody on clear liquids the day before had a problem, in either arm.
- 262,018 patients: retained contents OR 4.54, but NO significant rise in aspiration pneumonia.

These drugs [slow the stomach deliberately](https://glpwatchdog.org/research/ozempic-stomach-paralysis-gastroparesis), which is a problem when someone is about to sedate you. In 2026 a randomized trial finally tested what to do about it: patients who held a single dose before an upper endoscopy had clinically significant retained stomach contents **3.1%** of the time, against **25.0%** for those who continued.[1] The trial was stopped early because the gap crossed its preset stopping boundary. But its most useful finding was about the prep, not the drug, and almost nobody quotes it.

## Why a slowed stomach matters here specifically

Sedation removes the reflexes that keep stomach contents out of your airway. Anesthetic fasting rules exist to make sure the stomach is empty by the time that happens, and they were written for stomachs emptying at a normal rate. A drug whose whole mechanism includes slowing gastric emptying can leave food behind after a fast that would ordinarily be sufficient.

The consequences run from inconvenient to serious: an examination the endoscopist cannot complete, a procedure abandoned partway, an unplanned breathing tube, or aspiration.

## What the randomized trial found

OCULUS ran at two US referral centers, randomizing adults on a stable GLP-1 or GLP-1/GIP dose to hold one dose before an upper endoscopy or to continue.[1] Its endpoint was a composite of the things that disrupt a procedure: contents preventing examination, premature termination or intubation, or an aspiration event needing extra monitoring or admission.

*OCULUS, preplanned interim analysis, 60 patients. Clinically significant residual gastric volume.*

| Group | Rate | Difference |
| --- | --- | --- |
| Held one dose | 3.1% | — |
| Continued | 25.0% | 21.9 points (90% CI 7.0–36.7), P = .003 |
| Upper endoscopy only — held | 5.0% | — |
| Upper endoscopy only — continued | **46.7%** | 41.7 points (90% CI 17.9–65.4), P = .001 |

> **Read those numbers with the trial's own caveat attached.** This is an *interim* analysis of **60 patients**, and the trial was terminated when the difference crossed a preestablished stopping boundary. Stopping early for benefit is the right thing to do for the participants, but it systematically *overstates* the size of the effect. The direction is well supported. The precise magnitude is not.

## The finding almost nobody quotes

The subgroup having an upper endoscopy plus a colonoscopy had 25 patients, who were therefore on **clear liquids the day before** as standard bowel prep. **Not one of them** had clinically significant retained contents, in either the continue arm or the hold arm.[1]

> Nobody on clear liquids the day before had a problem, whether they held the drug or not.

The trial's own conclusion draws that out: clear liquids the day prior may mitigate the risk regardless of GLP-1 or GIP use.[1] If that holds up in larger numbers, the practical answer to this whole question may turn out to be a diet instruction rather than a drug hold. That matters, because holding a dose has its own costs for someone whose blood sugar depends on it.

## Retained food is not the same as aspiration

This distinction is where most coverage goes wrong. A meta-analysis of 23 studies covering **262,018 patients** found GLP-1 users had substantially more retained gastric contents (an odds ratio of **4.54**, 95% CI 3.30–6.24) and more procedures terminated early.[2]

And it found **no significant difference in aspiration pneumonia**.[2] That is the outcome people are frightened of, and across a quarter of a million patients it did not separate.

Both facts are real and they support different actions. Four and a half times the odds of retained contents is a good reason to change how procedures are prepared for. It is not evidence that people on these drugs are being harmed at scale, and a page that reports the first number without the second is telling you half of it.

A second, independent meta-analysis published in *Anaesthesia* reached the same split from a separate set of studies, which is the strongest form this evidence can currently take.[4] Two teams, different literature searches, same dissociation.

*Anaesthesia meta-analysis, 28 observational studies. Both outcomes graded LOW certainty.[4]*

| Outcome | Studies / patients | Odds ratio (95% CI) |
| --- | --- | --- |
| Residual gastric contents despite fasting | 18 studies, 165,522 patients | **5.96 (3.96–8.98)** |
| **Pulmonary aspiration** | 9 studies, 185,414 patients, 471 cases | **1.04 (0.87–1.25)** |

An odds ratio of 1.04 with an interval from 0.87 to 1.25 is about as flat as a result gets. Six times the odds of retained food, and no detectable change in the harm that retained food is feared for.

## Does skipping a dose help?

This is the question patients ask, and the same review provides the only pooled answer available. Across five studies and 1,706 patients, **withholding at least one dose before a procedure was associated with lower odds of residual gastric contents, with an odds ratio of 0.51** (95% CI 0.33–0.81).[4]

> **That result is graded *very low* certainty, the weakest rung there is**, and it comes from observational data where the people who skipped a dose differ from those who did not. It is also the only pooled evidence anyone has, it points the way common sense would, and halving the odds of retained contents is a large effect. Whether to skip a dose is a decision for the team doing the procedure, and they need to know you take the drug either way.

## What the guidance says, and why it keeps changing

The American Gastroenterological Association issued a Rapid Clinical Practice Update on this, explicitly framed as timely guidance on a fast-moving question rather than a settled standard.[3] Anesthetic and gastroenterological societies in several countries have since published their own, and they do not agree in every detail.

Disagreement is normal for an evidence base whose first randomized trial is still being absorbed. The practical consequence for you is that the instruction you get depends on which society your proceduralist follows.

> **The one thing to do is tell them.** Before any procedure involving sedation (endoscopy, colonoscopy, surgery, dental work under sedation), say that you take a GLP-1, name it, and say when your last dose was. Do not assume it is in your chart, because [compounded GLP-1s often never reach the medical record](https://glpwatchdog.org/research/what-your-doctor-cannot-see). And do not stop on your own: what to hold and for how long depends on the procedure, the drug and why you are taking it. Our [pre-op stop date calendar](https://glpwatchdog.org/tools/glp1-pre-op-stop-date-calendar) prints the dates so you can take a specific question to that conversation. A blood thinner has its own stop time on its own label; see [blood thinners and a GLP-1](https://glpwatchdog.org/research/blood-thinners-and-a-glp1).

## Frequently asked questions

### Do I need to stop my GLP-1 before an endoscopy?

That is your proceduralist's decision, and the guidance differs by society and by procedure. The randomized evidence supports holding: in the OCULUS trial, 3.1% of patients who held one dose had clinically significant retained stomach contents against 25.0% of those who continued. Tell whoever is doing the procedure which drug you take and when your last dose was, and let them decide.

### Why does it matter for sedation?

Sedation removes the reflexes that keep stomach contents out of your airway, and fasting rules exist to ensure the stomach is empty by then. Those rules assume normal gastric emptying. These drugs slow it deliberately, so food can remain after a fast that would normally be enough.

### Does a clear liquid diet help?

It may matter more than holding the dose. In OCULUS, the subgroup also having a colonoscopy, and therefore on clear liquids the day before, had zero cases of clinically significant retained contents in either arm. The trial's authors suggest clear liquids the day prior may mitigate the risk regardless of GLP-1 use.

### Is aspiration pneumonia actually more common on these drugs?

Not by the largest analysis available. A meta-analysis of 23 studies covering 262,018 patients found 4.54 times the odds of retained gastric contents and more procedures terminated early, but no significant difference in aspiration pneumonia. Retained food and actual harm are different outcomes.

### How reliable is the OCULUS trial?

It is the best evidence available and it is small. The published result is a preplanned interim analysis of 60 patients, and the trial was stopped early after the gap crossed a preset stopping boundary. Stopping early for benefit is correct for participants and systematically overstates effect size, so trust the direction more than the exact percentages.

## References

1. Ahmad AI, Garg S, Jacobs J, et al. Holding vs Continuing GLP-1/GIP Agonists Before Upper Endoscopy: The OCULUS Randomized Clinical Trial. JAMA Internal Medicine. 2026. PMID: 41837981. https://pubmed.ncbi.nlm.nih.gov/41837981/
2. Baig MU, Piazza A, Lahooti A, et al. Glucagon-like peptide-1 receptor agonist use and the risk of residual gastric contents and aspiration in patients undergoing GI endoscopy: a systematic review and a meta-analysis. Gastrointestinal Endoscopy. 2025. PMID: 39694296. https://pubmed.ncbi.nlm.nih.gov/39694296/
3. Hashash JG, Thompson CC, Wang AY. AGA Rapid Clinical Practice Update on the Management of Patients Taking GLP-1 Receptor Agonists Prior to Endoscopy: Communication. Clinical Gastroenterology and Hepatology. 2024. PMID: 37944573. https://pubmed.ncbi.nlm.nih.gov/37944573/
4. Elkin J, Rele S, Sumithran P, et al. Association between glucagon-like peptide-1 receptor agonist use and peri-operative pulmonary aspiration: a systematic review and meta-analysis. Anaesthesia. 2025. PMID: 40230298. https://pubmed.ncbi.nlm.nih.gov/40230298/

---

Cite as: GLP Watchdog, Nora Bissett, "Holding a GLP-1 Before Endoscopy or Surgery: What the Trial Found", published 2026-08-19, updated 2026-10-30. https://glpwatchdog.org/research/glp1-before-endoscopy-and-surgery

Canonical HTML page: https://glpwatchdog.org/research/glp1-before-endoscopy-and-surgery

Medical disclaimer: Nothing here is medical advice. We report on this subject; we do not treat anyone. Whether to begin a medication, supplement, or treatment, whether to adjust it, and whether to stop are questions for a clinician licensed to answer them, who knows your history.
