Eighty per cent of adolescents taking a GLP-1 receptor agonist still had solid food in the stomach when they arrived for a procedure, despite having fasted for a median of 13 hours, well beyond the eight hours standard guidelines ask for. Among healthy adolescents fasting under the same instruction, the figure was 5%. The study, published on 8 September in the British Journal of Anaesthesia, used bedside ultrasound to look directly into the stomach instead of trusting the fasting rule to have worked.
What the study did
Anaesthetists at the Children’s Hospital of Philadelphia enrolled 67 patients aged 10 to 18 between June 2023 and November 2024 and sorted them into three groups: 20 taking a GLP-1 receptor agonist, 27 already considered at risk of delayed gastric emptying for other reasons, and 20 healthy controls. All followed the same instruction, no solids for eight hours and no clear liquids for one. In practice they fasted considerably longer than asked, a median of 13 hours for solids (interquartile range 12 to 14) and 12 hours for clear liquids.
Each patient then had a gastric ultrasound, scored on a three-point qualitative scale and measured quantitatively. The primary outcome was residual gastric contents, defined as visible solids or a gastric fluid volume of at least 1.5 ml per kilogram. Solids were present in 16 of 20 patients on a GLP-1 (80%), 17 of 27 in the at-risk group (63%) and 1 of 20 controls (5%). The difference survived propensity-score adjustment for age, body mass index, sex, race and fasting time (unadjusted p < 0.001; adjusted p < 0.02).
The comparison worth sitting with is not GLP-1 against healthy. It is GLP-1 at 80% against the group that already had a recognised reason for a slow stomach at 63%. Those percentages are 16 patients and 17 patients. With groups this small the gap between them is a handful of people wide, and it should not be read as showing that GLP-1 drugs empty the stomach more slowly than longstanding diabetes does. The contrast that separates cleanly is the 5% in healthy controls, a sixteenfold difference in proportion, and that is what the adjusted comparison is holding up.
For anyone using one of these GLP-1 medicines and facing surgery, an endoscopy or any sedation, the practical read is unchanged in direction and firmer in degree: say that you are on one, and say it well before the day rather than in the holding bay. Adult anaesthesia guidance has been moving this way for two years. What is new here is that it is a prospective look at adolescents, a group whose prescriptions have grown since GLP-1 approvals reached age 10 for type 2 diabetes and age 12 for obesity, and that fasting longer on your own initiative is plainly not a fix.
What it does not tell you
This is a single-centre cohort of 67 people with only 20 in the GLP-1 arm. Residual stomach contents on ultrasound is a surrogate rather than a harm: no aspiration events are reported, so the study measures a risk factor and not an outcome. The abstract does not say which GLP-1 drugs those 20 patients were taking, at what dose, or how long before the procedure the last dose was given, and that last variable is precisely what current pre-procedure advice turns on. Because patients were not randomised, the groups differ in ways that a propensity adjustment can only partly absorb.
Sources: O’Brien EM, Gallop R, Washburn M, Dumas A, Como N, Kilungya F, Willi SM, Lin EE. “Delayed gastric emptying in adolescent patients on GLP-1 receptor agonists with standard preoperative fasting guidelines: a prospective cohort study.” British Journal of Anaesthesia, online ahead of print 8 September 2026, doi:10.1016/j.bja.2026.07.016. The authors declare no conflicts of interest.
