Anaesthesia meta-analysis: more retained stomach contents, no measurable increase in aspiration

Researchers pooling data from 28 observational studies reported that people taking GLP-1 receptor agonists were much more likely to arrive for surgery with food still in their stomachs despite following fasting instructions — yet the pooled data showed no increase in pulmonary aspiration, the serious complication that fasting rules are designed to prevent. The systematic review and meta-analysis was published on 15 April 2025 in the journal Anaesthesia ^[1].

Aspiration happens when stomach contents pass into the lungs while a patient is sedated or under general anesthesia. Because GLP-1 drugs such as semaglutide (Ozempic, Wegovy, Rybelsus) and tirzepatide (Mounjaro, Zepbound) slow how fast the stomach empties after a meal, case reports have described patients who aspirated or had unexpected stomach contents even after fasting properly, prompting several professional organizations to flag a possible added risk ^[1].

What the analysis found

The team, led by researchers at the University of Melbourne and St Vincent's Hospital Melbourne, searched six databases and screened 9,010 studies, ultimately including 28 observational studies ^[1].

For the main safety question, nine studies covering 185,414 patients and 471 cases of pulmonary aspiration were pooled. GLP-1 exposure was not associated with aspiration, with an odds ratio of 1.04 and a 95% confidence interval of 0.87 to 1.25 ^[1]. An odds ratio of 1.00 means no difference between groups, and the confidence interval here sits close around that value — meaning the analysis did not detect a meaningful increase or decrease. The authors rated this finding low certainty of evidence using the GRADE framework ^[1].

The picture for stomach contents was very different. Across 18 studies involving 165,522 patients and 3,831 cases of residual gastric contents, GLP-1 use was linked to an odds ratio of 5.96 (95% CI 3.96 to 8.98) for having food or fluid left in the stomach despite appropriate fasting ^[1]. That is roughly a six-fold increase in the odds. This result was also rated low certainty ^[1].

A third analysis looked at whether skipping a dose helps. Five studies covering 1,706 patients and 208 cases of residual gastric contents found that withholding at least one dose of a GLP-1 drug before a procedure was associated with lower odds of residual gastric contents (odds ratio 0.51, 95% CI 0.33 to 0.81) ^[1]. The authors graded that evidence very low certainty — the weakest rating of the three ^[1].

The authors' conclusion: "Patients using glucagon-like peptide-1 receptor agonists are at increased risk of presenting for anaesthesia with residual gastric contents, though the available evidence does not indicate that this translates to an increased risk of pulmonary aspiration" ^[1].

Why it matters for patients

Anyone on a GLP-1 medication who is scheduled for surgery, a colonoscopy, a dental procedure under sedation or any other procedure requiring anesthesia has likely been asked about the drug, and may have been told to pause it. This analysis helps separate two things that often get merged in those conversations: the likelihood of stomach contents being present, which appears substantially higher, and the likelihood of actual aspiration, which the pooled data did not show to be higher ^[1].

That distinction matters because pausing a GLP-1 drug is not free of consequences — it can affect blood sugar control in people with type 2 diabetes and may disrupt a treatment schedule. The finding that withholding a dose was linked to fewer cases of residual gastric contents is the least certain of the three results in this analysis, so it does not settle how far in advance, or whether, a dose should be held ^[1].

Some important caveats are built into the study design. All 28 included studies were observational, not randomized trials, which is one reason the certainty ratings were low or very low ^[1]. Observational data can be affected by differences between the people who take these drugs and those who do not. Aspiration is also rare — 471 cases among 185,414 patients — so even a large pooled analysis may not detect a small increase in risk ^[1].

The analysis reported results for GLP-1 receptor agonists as a class. Whether risk differs between specific molecules, doses, or injectable versus oral formulations was not broken out in the published summary ^[1].

What happens next

The paper appears in Anaesthesia Volume 80, Issue 7, pages 846–858 ^[1]. Whether anesthesia societies revise their pre-procedure guidance in response has not been reported. Decisions about holding or continuing a GLP-1 medication before a procedure remain between patients and their surgical, anesthesia and prescribing teams.