Clinical Guidance · GLP-1 Therapy
In 2023 a wave of coverage attached the phrase "stomach paralysis" to incretin medications, and the phrase has stayed attached to them since. It is a poor description of what these medications do, although it was not invented out of nothing. Tirzepatide genuinely does slow the rate at which the stomach empties, that slowing is genuinely part of why people eat less, and in a small number of people the effect becomes pronounced enough to resemble a motility disorder.
The difficulty for anyone trying to make sense of this is that the intended effect and the unwanted one sit on the same continuum, so there is no clean threshold where useful fullness becomes a clinical problem. What exists instead is a spectrum of severity, a set of imperfect measurements, and a body of observational research showing an elevated relative risk against a low absolute one. What follows is what delayed gastric emptying actually is, what the evidence does and does not establish about gastroparesis on tirzepatide, why the findings from endoscopy suites turned into formal guidance, and how ordinary adjustment can be distinguished from something worth raising with a clinician.
Delayed gastric emptying means that food leaves the stomach and enters the small intestine more slowly than it otherwise would, and on tirzepatide this happens by design because both incretin receptors the molecule activates influence gastric motility. The resulting delay accounts for a meaningful share of why appetite falls during treatment.
Tirzepatide activates the GLP-1 receptor and the GIP receptor, and GLP-1 receptor activation is the better characterized of the two in this context. Signalling through vagal pathways and the brainstem, it reduces the strength of the antral contractions that push stomach contents forward while increasing tone at the pylorus, the valve controlling outflow into the duodenum. Food therefore remains in the stomach longer, gastric distension persists after a smaller volume of food, and the satiety signals that normally follow a large meal arrive after a modest one instead.
This is not paralysis, and it is not damage. Gastric muscle and its nerve supply remain intact, and the effect is pharmacological, meaning that it tracks the presence of the medication and reverses as the medication clears. That distinction matters for anyone worried that tirzepatide is causing permanent injury, because the available evidence does not support that interpretation for the overwhelming majority of people taking it.
What the research consistently emphasises is the considerable variation between individuals, since studies measuring emptying directly have found that some participants show a modest change while others show a pronounced one at the same dose. Nobody can currently predict in advance which group a given person will fall into.
Figure 1
Reported Gastroparesis Diagnoses per 1,000 Person Years in an Obesity Population
Source: Published observational cohort data in obesity populations, as summarized in reviews of gastrointestinal outcomes with incretin therapies. Observational data cannot establish causation, and people taking these medications interact with the healthcare system more frequently, which raises the likelihood that a motility problem is investigated and coded. Individual results vary.
Incretin medications have been associated with an increased relative risk of a gastroparesis diagnosis in observational studies, although the absolute risk in those studies remained low and generally stayed under one percent of users. The relationship is further complicated by the fact that these medications produce symptoms resembling gastroparesis in many people who do not have the disorder.
Gastroparesis as a clinical diagnosis means significantly delayed gastric emptying, confirmed by objective testing, in the absence of mechanical obstruction, together with symptoms such as persistent nausea, vomiting, early fullness, and upper abdominal discomfort. The operative word is objective, since symptoms on their own do not establish the diagnosis.
Published cohort analyses have reported relative risks in the range of roughly three to four times that of comparator therapies, and analyses in type 2 diabetes populations have reported figures of similar magnitude. Conference data comparing tirzepatide with semaglutide directly in people without diabetes has suggested a lower rate of new gastroparesis diagnoses with tirzepatide, although findings presented in abstract form should be treated as preliminary until they appear in full peer reviewed publication.
Several caveats deserve weight here. Observational cohorts cannot establish causation, and obesity and type 2 diabetes are themselves associated with delayed gastric emptying independent of any medication, while a threefold increase applied to an uncommon event still produces an uncommon event. The clinically useful framing is that a meaningful number of people taking tirzepatide will experience symptoms overlapping with gastroparesis, a considerably smaller number will meet diagnostic criteria on objective testing, and reported cases have generally improved after the medication was reduced or stopped.
Figure 2
Relative Risk Estimates Reported Across Published Cohorts
Sources: Published observational cohorts of gastroparesis diagnoses in obesity and type 2 diabetes populations, and a retrospective endoscopy series of 35,183 patients. Relative risk describes the ratio between groups and says nothing about how common an event is in absolute terms, which remained below one percent per year for gastroparesis diagnoses in the cohorts summarized above.
They fade partially. Research on long acting incretin medications has documented tachyphylaxis, meaning that the magnitude of the emptying delay diminishes over weeks to months of continued treatment, although a residual slowing remains rather than resolving completely.
This adaptation is one reason that people frequently describe the first several weeks on a given dose as the most difficult period, with fullness and nausea easing as the body adjusts. Regression analyses across studies of long acting agents have shown a significant reduction over time in the prolongation of gastric emptying half times, while short acting agents have not shown the same pattern of adaptation.
There is an important asymmetry here, because each dose increase reintroduces some of the effect, which explains why titration schedules move gradually and why a step up in dose often brings back symptoms that had previously settled. Tolerance developed at one dose does not fully carry over to the next, so early symptoms are a poor guide to long term tolerability, and someone who struggles during the first month of a dose is not necessarily someone who will struggle at that dose in the fourth month.
Figure 3
Tachyphylaxis: How the Emptying Delay Changes With Continued Use (Schematic)
Note: This figure is a schematic representation of the direction and shape of the effect described in regression analyses of gastric emptying studies, and the plotted values are illustrative rather than measured trial data. Each dose increase can temporarily reintroduce part of the initial effect. Individual results vary.
Symptoms diverge from measurements because nausea, vomiting, and fullness on tirzepatide arise from several mechanisms at once, and gastric emptying is only one of them. Studies have repeatedly found weak correlation between how delayed a person's measured emptying is and how severe their gastrointestinal symptoms are.
Incretin receptors are present in the area postrema and the nucleus tractus solitarius in the brainstem, regions involved in nausea and vomiting that sit outside the blood brain barrier, and activating them produces nausea directly without any involvement of the stomach. Motility changes further along the gastrointestinal tract contribute their own symptoms, and some research has examined altered small bowel transit and bacterial overgrowth as additional contributors.
This decoupling cuts in both directions in practice, since substantial nausea does not confirm that the stomach is emptying dangerously slowly, while comfortable digestion does not confirm that emptying is normal. It is one of the main reasons that clinicians rely on objective testing rather than symptom reports when gastroparesis is genuinely suspected.
Figure 4
Gastrointestinal Effects Reported in Tirzepatide Clinical Trial Programs
Sources: Approximate frequencies drawn from the tirzepatide phase 3 trial programs, where reported rates varied by dose, indication, and study population. Values are rounded and shown for orientation rather than as precise figures for any single trial. These common effects are most pronounced during dose escalation and are distinct from a gastroparesis diagnosis, which requires objective testing. Individual results vary.
Those findings showed that a meaningful minority of people taking incretin medications still have food in the stomach after a standard fasting period, which matters for procedures requiring sedation. In the largest series, covering 35,183 patients, retained gastric contents were found in 13.6% of incretin users compared with 2.3% of non users, and the rate of aborted procedures rose by a comparable factor.
That finding drove genuine concern about pulmonary aspiration, in which stomach contents enter the airway under sedation, although the data on whether aspiration itself increases are less consistent than the retained contents data. Among patients found to have retained contents, roughly 0.2% showed evident aspiration. One analysis reported a higher risk of aspiration pneumonia among incretin users with a hazard ratio of 2.28, while another found no increased rate of pulmonary aspiration under general anesthesia, at 0.54% among users compared with 0.69% among non users. For context, aspiration during elective surgery occurs in roughly one per 3,000 cases overall.
Early guidance from the American Society of Anesthesiologists suggested withholding long acting agents for a week before procedures, and that approach drew criticism on pharmacological grounds, since semaglutide has a half life of around 7.6 days and even physiological concentrations of GLP-1 are sufficient to slow emptying, so a single skipped dose does not clear the effect. It also carried its own costs in glycemic control and treatment continuity.
Figure 5
Retained Gastric Contents at Endoscopy After Standard Fasting
Source: Retrospective endoscopy series of 35,183 patients reporting retained gastric contents in 13.6% of incretin medication users compared with 2.3% of non users, with a comparable increase in aborted procedures, observed irrespective of diabetes status. Retained contents describe what was seen at endoscopy rather than a gastroparesis diagnosis. Individual results vary.
| Consideration | Earlier single society guidance | 2024 multisociety guidance |
|---|---|---|
| Holding the medication | Withhold one week for long acting agents before a procedure | Continue therapy for most patients undergoing elective procedures |
| Diet before the procedure | Standard fasting protocol | Liquid diet for 24 hours beforehand for patients assessed as higher risk |
| Assessing stomach contents | Not routinely specified | Point of care gastric ultrasound immediately before the procedure in higher risk patients |
| Anesthetic approach | Not specifically addressed | Technique adjusted to minimise aspiration risk where indicated |
| Document status | Society statement | Described as guidance rather than an evidence based guideline, emphasising shared decision making |
Sources: American Society of Anesthesiologists guidance and the 2024 multisociety clinical practice guidance developed jointly by anesthesiology, gastroenterology, bariatric surgery, and endoscopic surgery societies. Decisions about any individual procedure rest with the treating team.
The 2024 multisociety guidance, developed jointly by the anesthesiology, gastroenterology, bariatric surgery, and endoscopic surgery societies, took a different position and concluded that most patients should continue their medication before elective procedures, with risk mitigation reserved for those assessed as highest risk. The actionable point for anyone taking tirzepatide is straightforward, since telling the anesthesiologist and the proceduralist about the medication well before the day of the procedure allows the plan to be made deliberately rather than in a preoperative bay.
Symptoms that persist between doses rather than clustering after them, that fail to settle while a dose is held steady, or that involve vomiting food eaten many hours earlier all warrant a conversation with the prescribing clinician. The pattern matters considerably more than any single symptom in isolation.
Nothing described here substitutes for individual medical judgement, and decisions about dose reduction, pausing treatment, or investigating symptoms belong with the clinician who knows the full clinical picture. Stopping a prescribed medication without that conversation carries risks of its own.
Most of what helps involves reducing the volume and the workload the stomach has to handle at any one time, which means working with the slowed emptying rather than against it. Smaller and more frequent meals ask less of a stomach that is emptying slowly than three conventional meals do.
Meal composition. High fat meals empty most slowly of all and often provoke the greatest discomfort, so changing the composition of difficult meals tends to help more than simply shrinking them. Liquid and semi liquid foods pass through more readily than solids, which is the same principle behind the liquid diet in the multisociety procedural guidance, and it can be applied on a difficult day as well.
Timing and posture. Separating fluids from solid food by 20 to 30 minutes reduces gastric volume at any single moment, while remaining upright for a period after eating rather than lying down uses gravity in the right direction and also reduces reflux, which frequently accompanies slowed emptying. Careful chewing matters more than it does normally, since mechanical breakdown in the mouth substitutes for gastric work that is happening less efficiently.
Titration. Slower titration is often the most effective adjustment available, and it is a clinical decision rather than a self directed one. Spending longer at a tolerated dose before stepping up, or moving back down a step, resolves a large proportion of persistent symptoms, because the tolerability of the dose someone is actually taking matters more than the theoretical benefit of a dose they cannot maintain.
Slowed gastric emptying on tirzepatide is not a warning sign in itself, since it is the mechanism working and accounts for a real share of the appetite reduction people experience. The risk of a formal gastroparesis diagnosis appears elevated relative to comparator therapies in observational research, although the absolute incidence in those cohorts stayed low and reported cases have generally improved after treatment was reduced or stopped.
The points most worth carrying forward are that symptoms are an unreliable measure of how slowly the stomach is actually emptying, that early difficulty on a dose usually eases as the body adapts, and that procedures involving sedation require advance disclosure so the team can plan, with current guidance favouring continuation of the medication alongside targeted precautions. The symptoms that genuinely warrant attention are those that persist across the full weekly cycle, involve vomiting food eaten hours earlier, prevent fluid intake, or bring severe pain.
A 3 minute intake is all it takes. A physician reviews your information and identifies the protocol matched to your specific symptom profile.
See Our Protocols