How GLP-1 Gastric-Emptying Delay Is Measured (2026)

Almost every GLP-1 explainer says the same thing — these drugs “slow gastric emptying” — but few say how that is actually measured, how large the delay really is, or the part that surprises people most: the slowing largely plateaus, and partly fades, over weeks of treatment. That last fact (called tachyphylaxis) explains a lot — why a single first dose can hit you hard while month three feels calmer, why holding a weekly drug for a few days may not fully empty the stomach, and why the “food still in the stomach” finding at endoscopy is real even though aspiration pneumonia has not gone up in pooled data. This article walks through the four ways gastric emptying is measured (scintigraphy, the acetaminophen-absorption test, breath tests, and MRI), what the numbers show, and the biology of why the effect plateaus. For the practical safety angle, see stopping GLP-1s before surgery and GLP-1s and colonoscopy prep.

The honest summary

• The delay is real and measurable. By gastric scintigraphy — the reference standard — pooled data put the gastric half-emptying time at about 138 minutes on a GLP-1 versus 95 minutes on placebo, a roughly 36-minute delay (Hiramoto 2024^[1]).
• How you measure it changes the answer. Scintigraphy and breath tests detect the slowing; the acetaminophen-absorption test often does not, because over 4–5 hours the stomach catches up and total drug absorption looks normal even though the early emptying is delayed (Hiramoto 2024^[1]).
• The effect is biggest after the first dose. A single early dose can roughly double half-emptying time, but with continued weekly dosing the slowing shrinks substantially — the plateau, or tachyphylaxis (Urva 2020^[2]; Jalleh 2024^[3]).
• It was first shown in hours, not weeks. During continuous GLP-1 infusion, the braking effect on a second meal just 4 hours later was already much weaker than on the first — rapid tachyphylaxis, probably acting through the vagus nerve (Nauck 2011^[4]).
• Some residual slowing usually remains. In a liraglutide study, 57% had a marked delay at 5 weeks, but only about half of those still had it at 16 weeks; roughly 30% of the whole group retained a delay long-term (Camilleri 2024^[5]).
• Why it matters clinically: slowed emptying drives satiety and blunts post-meal glucose spikes, and it is why stomachs can hold contents at endoscopy — about 6× the odds of residual gastric contents — yet pooled data show no increase in aspiration (Elkin 2025^[6]).

Why gastric emptying matters in the first place

Slowing how fast the stomach empties is not a side effect of GLP-1 drugs — it is one of the core ways they work. GLP-1 is a gut hormone that acts as a physiologic “brake” on the stomach: when nutrients reach the small intestine, GLP-1 release tells the stomach to hold food back so the gut is not overwhelmed. The foundational human work showed this is dose-dependent — the more GLP-1 in the blood, the slower the stomach empties — and that this braking effect on emptying does more to lower post-meal glucose than the hormone's insulin-boosting effect does (Willms 1996^[7]). Two consequences follow. First, food stays in the stomach longer, so you feel full sooner and longer — the satiety that drives weight loss. Second, glucose and oral drugs enter the bloodstream more gradually, which smooths post-meal sugar spikes but can also alter how some oral medications are absorbed.

The four ways gastric emptying is measured

1. Gastric scintigraphy — the reference standard

Scintigraphy is the test motility specialists treat as the gold standard. You eat a standardized meal — in the consensus protocol, an egg-white sandwich labeled with a trace amount of a radioisotope (technetium-99m sulfur colloid) — and a gamma camera takes images at 0, 1, 2, and 4 hours so radiologists can watch the radioactive meal physically leave the stomach (Abell 2008^[8]). The standardized read-outs are the half-emptying time (how long until half the meal has left) and the percentage retained at set time points; in the consensus protocol, more than 10% of the meal still in the stomach at 4 hours counts as delayed emptying. Because it images the actual meal, scintigraphy reliably detects the GLP-1 slowing. In Hiramoto's pooled analysis, scintigraphy gave a mean half-emptying time of about 138 minutes on a GLP-1 versus 95 minutes on placebo — the headline 36-minute delay (Hiramoto 2024^[1]).

2. The acetaminophen (paracetamol) absorption test

This is the clinical-pharmacology workhorse, and it explains a lot of apparent contradictions. Acetaminophen is barely absorbed by the stomach but rapidly absorbed by the small intestine — so how fast it shows up in the blood is a proxy for how fast the stomach handed it off. You swallow a dose (often in a meal or yogurt) and blood levels are tracked. A GLP-1 that delays emptying flattens and delays the early acetaminophen curve. Semaglutide, for example, cut first-hour acetaminophen absorption by about 27% versus placebo (Hjerpsted 2018^[9]). But here is the catch: when researchers measured the total amount absorbed over 4–5 hours (AUC), there was often no significant difference — across ten such studies the test showed no meaningful delay, because the stomach catches up later (Hiramoto 2024^[1]). So the acetaminophen test is excellent for the question it was designed for — does the drug change overall absorption of oral medicines? — but it can understate the early-phase slowing that scintigraphy and breath tests pick up.

3. Stable-isotope breath tests

Here the meal is labeled with a non-radioactive carbon isotope (carbon-13) bound to a fatty acid or other substrate. As the meal leaves the stomach and is metabolized, labeled carbon dioxide appears in the breath, and the rate of its appearance tracks gastric emptying. Breath tests are radiation-free and can be done outside a nuclear-medicine suite, which makes them attractive for research. They detect the GLP-1 delay in the same direction as scintigraphy, though Hiramoto's team found the breath-test studies too heterogeneous in their read-outs to pool into a single number (Hiramoto 2024^[1]).

4. MRI and gastric ultrasound

Imaging the stomach directly is the newest frontier. Serial MRI can measure gastric volume and content over time without any radiation, and bedside gastric ultrasound — increasingly used by anesthesiologists on the day of a procedure — can show whether a fasted stomach still holds solid food or a large fluid volume. Ultrasound does not give a clean half-emptying number, but it answers the practical question that matters before sedation: is this stomach empty right now? That is exactly why some perioperative protocols now favor point-of-care gastric ultrasound over a fixed hold rule (Jalleh 2024^[3]).

How big is the delay, really?

The most defensible single number comes from Hiramoto's 2024 systematic review and meta-analysis (36 studies, 1,574 patients across all methods): by scintigraphy, a roughly 36-minute longer half-emptying time on a GLP-1 (about 138 vs 95 minutes) (Hiramoto 2024^[1]). After a single first dose the effect can be far larger — early-dose studies have shown half-emptying time roughly doubling — but that peak is transient (see the plateau section below). The magnitude also varies a lot between people: in healthy volunteers some show a striking delay and others almost none, which is part of why no fixed pre-procedure rule fits everyone (Camilleri 2024^[5]). And the dual GIP/GLP-1 agonist tirzepatide (Mounjaro, Zepbound) delays emptying after a first dose to a degree similar to long-acting GLP-1 agonists — not dramatically more — and shows the same fade with continued dosing (Urva 2020^[2]).

Why it plateaus: tachyphylaxis

The single most under-appreciated fact about GLP-1 and the stomach is that the slowing does not keep getting stronger — it largely plateaus, and partly reverses, with continued exposure. This is called tachyphylaxis. It was first shown not over months but over hours: in a continuous GLP-1 infusion study, the braking effect on a second meal eaten just four hours after the first was already much weaker than on the first meal. The authors concluded the effect undergoes rapid tachyphylaxis, most likely through adaptation in the vagus nerve pathway that GLP-1 uses to signal the stomach, rather than the receptors disappearing (Nauck 2011^[4]).

The same pattern shows up over weeks with the actual drugs. Tirzepatide and dulaglutide delayed emptying sharply after the first dose, but the effect diminished markedly after repeated dosing (Urva 2020^[2]). In a liraglutide study tracking gastric emptying over 16 weeks, 57% of patients had a very significant delay at 5 weeks — but among those, only about half still had it at 16 weeks, and roughly 30% of the whole group retained a measurable delay long-term (Camilleri 2024^[5]). The leading explanation is that long-acting agents — semaglutide, tirzepatide, dulaglutide — keep the GLP-1 receptor continuously occupied, and continuous stimulation drives faster neural accommodation, so the gastric-emptying brake fades even as the appetite and glucose benefits persist. Short-acting agents that come and go (the older twice-daily exenatide, once-daily lixisenatide) show less of this fade and keep slowing emptying more consistently (Jalleh 2024^[3]).

How this connects to satiety, glucose, and aspiration

The same delayed emptying does three jobs. It prolongs fullness, because a stomach that stays distended longer keeps sending satiety signals — a major contributor (alongside central appetite suppression) to eating less. It blunts post-meal glucose spikes, because nutrients trickle into the small intestine instead of arriving in a flood. And it is the reason stomachs can still hold contents after the usual fast before sedation. That last point is where the measurement story meets real-world safety: in the largest pooled analysis, GLP-1 users had roughly 6× the odds of residual gastric contents at procedures (OR 5.96) — yet no significant increase in actual pulmonary aspiration (OR 1.04, across 185,414 patients) (Elkin 2025^[6]). The “food in the stomach” finding is robust and directly traceable to the mechanism this article describes; the leap to lung injury is not well supported by outcome data so far — which is exactly why the perioperative guidance has shifted toward individualized assessment rather than blanket cancellation (Jalleh 2024^[3]).

Bottom line

GLP-1 drugs measurably slow gastric emptying — about a 36-minute longer half-emptying time by scintigraphy in pooled data (Hiramoto 2024^[1]) — but how you measure it matters: scintigraphy and breath tests catch the early slowing, while the acetaminophen-absorption test often normalizes because the stomach catches up over 4–5 hours. The effect is largest after the first dose and then plateaus and partly fades with continued long-acting dosing, a vagally mediated tachyphylaxis first shown within hours of infusion (Nauck 2011^[4]) and confirmed over weeks with modern drugs (Urva 2020^[2]; Camilleri 2024^[5]). That plateau explains the early-heavy nausea, the imperfect logic of short pre-procedure holds, and why the appetite and glucose benefits outlast the gastric brake. It also frames the safety data: more residual stomach contents at endoscopy, but no measured rise in aspiration (Elkin 2025^[6]).

This article is educational and is not medical advice. Every claim above is sourced to a peer-reviewed study, society consensus document, or systematic review indexed in PubMed, verified against the live PubMed database (and, where noted, the open-access fulltext) before publication. Discuss any decision to start, hold, or stop a GLP-1 with your prescriber.