The agonists of the glucagon-like peptide-1 receptor (GLP-1R) have seen exponential growth since their commercialization in 2005 for the treatment of type 2 diabetes (T2DM) and obesity. GLP-1R agonists not only exert a potent incretin action, improving glycemic control, but also significantly reduce body weight and cardiovascular, renal, or hepatic metabolic disease risk.1–4 However, the actions of GLP-1R agonists on GI motility, with the frequent appearance of symptoms—nausea, vomiting, abdominal distention, diarrhea, or constipation—are a serious concern for endoscopists and anesthesiologists due to the potential risk of these patients having increased gastric residue retention and aspiration.
GLP-1R agonists, in fact, exert actions on gastric fundus relaxation, reduction of acid secretion, and inhibition of antral contractility, actions that are dependent on parasympathetic, vagal stimulation.5
However, the impact of delayed gastric emptying on sedation and aspiration risks during anesthesia is not clearly defined. Although there is abundant literature on the subject, it is highly heterogeneous due to the study design types, different methods of gastric content evaluation, various GLP-1R agonist molecules analyzed, and a variety of treatment regimens administered to populations that are not always homogeneous.
The question that arises for clinicians using GLP-1R agonists is whether these drugs delay gastric emptying in a transient or permanent manner, and what implications this may have when patients undergo procedures that require sedation and/or anesthesia.
The gold standard for correctly measuring gastric emptying is isotopic gastric emptying after the ingestion of a solid meal labeled with technetium-99m.6 A retention of ≥ 60% at 2 h or ≥ 10% at 4 h after ingestion of the preparation indicates delayed gastric emptying. Thus, in the study by Maselli et al.,7 in a controlled clinical trial with 136 patients with obesity, randomized to liraglutide 3 mg or placebo, they observed a longer gastric emptying time at 5 weeks after the start of treatment (191.6 vs. 105.9 min, p < 0.001), with this effect improving at 16 weeks (154.4 vs. 111.4 min, p < 0.001), evidencing a tachyphylaxis effect of the drug. Similar results of delayed gastric emptying have been found with the use of different GLP-1R agonists, both in monotherapy—liraglutide, lixisenatide, albiglutide, dulaglutide, semaglutide—and in dual combination—tirzepatide—or in the form of triagonists—retatrutide—,8–10 excluding, for now, oral semaglutide. A recent meta-analysis that analyzed 5 clinical trials (n = 247) confirmed that GLP-1R agonists have a mean gastric emptying time of 36 min (95%CI, 17.0–55.0; p < 0.01) longer than placebo.11 In contrast, in another 10 studies (n = 411) using the paracetamol absorption test, the area under the curve was comparable between GLP-1R agonists and placebo treatment. Of note, paracetamol surrogate test depends on both gastric emptying capacity and intestinal absorption, being a better reference for liquid emptying than for solids. However, it should not be routinely used to determine gastric emptying with GLP-1R agonists.
Liquid retention in the gastric cavity is usually not a problem for performing esophagogastroduodenoscopy, as the content is aspirated during the procedure. For example, in a study conducted with endoscopy and measurement of aspirate during the test, treatment with semaglutide was associated with a 5-fold higher risk of gastric retention (OR, 5.15 [95%CI, 1.92–12.92]), yet a protective effect was observed (OR, 0.25 [95%CI, 0.16−0.39]) when an esophagogastroduodenoscopy and colonoscopy were performed simultaneously. A preoperative diet restricted in residue and only liquid intake 24 h before the procedure certainly helps to avoid gastric retention of solid food. Other observational studies replicate similar data with different types of GLP-1R agonists, but gastric retention is also significant when GI symptoms—abdominal distention, nausea, vomiting, etc.—are present associated with the use of these drugs12–14 or in insulin-dependent T2DM patients on GLP-1R agonists and/or in those with macro- or microvascular complications.15
Some authors suggest performing an abdominal ultrasound in situ just before endoscopy or surgical anesthesia to check whether there is gastric content and make a decision at that moment. Although ultrasound is an accessible, non-invasive, and rapid method, it requires a trained operator, as it is more difficult to visualize gastric content in people with obesity. Two recent studies using abdominal ultrasound described greater gastric retention in people on GLP-1R agonists (between 40-56%) vs those not treated with such agent (3–19%),16,17 thus confirming previous results obtained with isotopic gastric emptying or endoscopic aspiration, showing gastric retention associated with GLP-1R agonist use. Additionally, it is worth noting that, in several of these studies, no notable differences were found across different types of agonists used, treatment duration—absence of tachyphylaxis—or even considering different periods of drug suspension (up to > 14 days) before anesthesia induction.
If we assume with this data that there is a higher probability of gastric retention with GLP-1R agonist treatment, is the risk of aspiration during sedation increased? In large retrospective studies with more than a million procedures performed, most studies have not found a significant association among GLP-1R agonist use, gastric content aspiration, and/or secondary pneumonia. In the study by Anazco et al.,18 the cumulative incidence rate of aspiration was 4.8/10,000 diagnostic endoscopic procedures among patients on GLP-1R agonists vs 4.6/10,000 in the untreated population. Similar data have been described in other studies of similar size,19–22 except for the work by Yeo et al.,23 who, analyzing the registry of 963,184 individuals undergoing endoscopy— 46,935 (4.9%) on GLP-1R agonist treatment—and applying a propensity score matching methodology, found that GLP-1R agonists were associated with an increased risk of aspiration (HR, 1.33 [95%CI, 1.02–1.74]; p = 0.036), especially when propofol was used in sedation induction. However, any of these observations, based on retrospective data analysis, should be interpreted with caution, as they may introduce validity biases by not reliably representing the analyzed population.
In everyday practice, patients on GLP-1R agonists arrive for endoscopy or surgery without precise instructions. But what do scientific societies say? In 2023, the American Society of Anesthesiologists established in its recommendations that: “In the case of patients receiving daily doses, it should be considered to suspend GLP-1R agonists on the day of the procedure or surgery. In the case of patients receiving weekly doses, it should be considered to suspend GLP-1R agonists 1 week prior to the procedure or surgery”.24 However, the American Gastroenterological Association,25 like other scientific societies,26 opposes the need to suspend the medication, as this could affect the patients’ metabolic control and weight management. They also remind that “for patients on GLP-1R agonists who have followed standard perioperative procedures—normally an 8-h fast for solid foods and a 2-h fast for liquids—and who do not have symptoms of nausea, vomiting, dyspepsia, or abdominal distention, it is recommended to proceed with upper and/or lower endoscopy.” Again, it is important to emphasize that individual variability in gastric emptying is high, particularly in people with T2DM. Therefore, we cannot predict which patients or which types of GLP-1R agonists, doses, or treatment durations will have a greater or lesser impact on gastric retention of solid food before a test. Even the absence of GI symptoms is not a guarantee of no gastric residue, as shown by data obtained with abdominal ultrasound.16,17
Recently, a clinical practice guideline from several scientific societies (American Gastroenterological Association, American Society for Metabolic and Bariatric Surgery, American Society of Anesthesiologists, International Society of Perioperative Care of Patients with Obesity, and Society of American Gastrointestinal and Endoscopic Surgeons) was published for the safe use of GLP-1R agonists in the perioperative period.27 Their recommendations are suggestions, as they are not supported by studies with sufficient scientific evidence. This document considers that those on GLP-1R agonists who are in the up titration phase, those with higher doses of the drug, weekly formulations, and those with GI symptoms or a history of gastroparesis are at greater risk of delayed gastric emptying and aspiration. Active treatment suspension is not recommended as a rule, but if doubts persist regarding the above recommendations, it is suggested to follow the advice of the American Society of Anesthesiologists, suspending GLP-1R agonists for 1 day or 1 week depending on the type of drug used.
None of the above consensuses or recommendations have been approved by any American or European scientific societies of endocrinology or diabetes. From our perspective, we agree with Jalleh et al.’s considerations for an individualized approach5: a) it is important to analyze clinical situations that may predispose to gastroparesis (advanced diabetes, chronic neurological diseases, opioid drugs, antidepressants, antipsychotics, proton pump inhibitors, etc.); b) avoid suspending GLP-1R agonists; c) proper preparation with a 24-h liquid diet and a 6−8 h fast prior to the procedure will ensure, in a high percentage of patients, the absence of gastric retention; d) optionally, and if available, the feasibility of performing an abdominal ultrasound to rule out residual gastric content, and e) also consider, in specific cases, the possibility of treatment with prokinetics, such as erythromycin (200 mg iv or 3 mg/kg). Some of these measures are also included in the multisociety consensus.27
To move out of the state of uncertainty surrounding the effects of GLP-1R agonists on gastric emptying and their implications for sedation or anesthesia induction, there are still some challenges ahead. Studies are needed to better investigate the physiology of gastric emptying in T2DM and obesity, before and after GLP-1R agonist treatment; additionally, prospective gastric emptying studies are needed to determine whether tachyphylaxis exists in different therapeutic situations, and more pharmacogenomics studies are needed to help differentiate patients with faster or slower gastric emptying and/or the presence of GI symptoms. Only with well-designed clinical trials will recommendations with a high level of evidence be made.
FundingNone declared.
