Gastric ultrasound in patients receiving semaglutide: a prospective, multicentre, matched control study

Citation:

Vlaeminck N, Van de Putte P, Dekeyser M, Baert N, Wallyn A, Vernieuwe L, et al. Gastric ultrasound in patients receiving semaglutide: a prospective, multicentre, matched control study. Anaesthesia. 2026. doi:10.1111/anae.70129

This episode examines a prospective matched case control study of semaglutide treated adults having elective surgery with pre operative gastric ultrasound. The study suggests semaglutide users may be more likely to have ultrasound defined full stomachs and solid gastric contents than matched non users (low certainty). Because the evidence is observational from a single small cohort, these results are hypothesis generating rather than practice changing.

Study at a glance

- Design and setting: Prospective, multicentre, matched case control study using individual matching, conducted at one academic and three large community hospitals in Belgium (Antwerp University Hospital, Imelda Hospital, AZ Delta Hospital, AZ Klina Hospital). Pre operative point of care gastric ultrasound was performed in the holding area approximately 30 minutes before elective surgery under general anaesthesia.

- Population: Eighty eight adults scheduled for elective surgery under general anaesthesia were included: 44 receiving semaglutide and 44 matched controls. Inclusion required adherence to fasting guidelines (>2 h liquids, >6 h solids) and guideline recommended glucagon like peptide receptor agonist withholding, plus matching on age (within 5 years), body mass index category (<25, 25–30, 30–35, >35 kg.m-2) and diabetes status. Mean age was 59 (14.0) vs 58 (13.8) years and body mass index 32.6 (4.7) vs 32.5 (5.6) kg.m-2 in semaglutide and control groups, respectively.

- Exposure and comparator: Exposure was current semaglutide treatment (any dose, administration route, frequency or indication) with pre operative withholding per American Society of Anesthesiologists and European guidance (1 week for weekly dosing, 1 day for daily dosing). The comparator group comprised matched control patients not receiving semaglutide, undergoing similar elective surgery under general anaesthesia.

- Primary outcome: Primary outcome was a full stomach on pre operative gastric ultrasound, defined as visible solid content in any position or calculated gastric volume > 1.5 ml.kg-1 in right lateral decubitus. Full stomach was observed in 21/43 semaglutide patients (49%) vs 8/44 controls (18%), adjusted odds ratio 4.29 (95% confidence interval 1.63 to 11.29), p = 0.003 (low certainty). This suggests semaglutide use may increase the odds of an ultrasound defined full stomach compared with matched non users.

- Key secondary outcome: A key secondary outcome was solid gastric contents on ultrasound. Solids were seen in 18/43 semaglutide patients (42%) vs 3/44 controls (7%), adjusted odds ratio 9.85 (95% confidence interval 2.57 to 37.76), p < 0.001 (low certainty), indicating semaglutide treatment may greatly increase the odds of visible solid gastric contents despite standard fasting and drug withholding.

- Confounding: Confounding was addressed by prospective individual matching on age, body mass index category and diabetes status, and by excluding conditions known to delay gastric emptying; analyses used logistic regression with generalised estimating equations for matched pairs. However, no propensity score or broader multivariable adjustment was reported, and baseline differences such as American Society of Anesthesiologists physical status distribution and prevalence of gastrointestinal symptoms remained, so residual confounding is likely and its direction is unpredictable.

- Risk of bias and certainty: Overall risk of bias was judged Moderate by ROBINS I, mainly due to residual confounding and potential unblinded, expectation influenced assessment of ultrasound outcomes, which could exaggerate harms in semaglutide users; selection, exposure classification, deviations from intended interventions, missing data, and selective reporting were at Low risk. For the primary outcome (full stomach) and solid contents, certainty of evidence was rated Low; for change in anaesthetic plan (24/44 vs 8/43, adjusted odds ratio 5.26, 95% confidence interval 1.96 to 14.11, p < 0.001) and for total and weight adjusted gastric volume (medians 61 vs 52 ml, p = 0.568; 0.58 vs 0.57 ml.kg-1, p = 0.520, with 13 vs 1 inconclusive scans), certainty was Very Low due to risk of bias and imprecision.