Clinical Consequences of Delayed Gastric Emptying With GLP-1 Receptor Agonists and Tirzepatide

Abstract

Context: Glucagon-like peptide-1 (GLP-1) receptor agonists (RAs) are established therapeutics for type 2 diabetes and obesity. Among other mechanisms, they slow gastric emptying and motility of the small intestine. This helps to limit postprandial glycemic excursions and reduce chylomicron formation and triglyceride absorption. Conversely, motility effects may have detrimental consequences, eg, retained gastric contents at endoscopy or general anesthesia, potentially complicated by pulmonary aspiration or bowel obstruction.

Data acquisition: We searched the PubMed database for studies involving GLP-1RA therapy and adverse gastrointestinal/biliary events.

Data synthesis: Retained gastric contents at the time of upper gastrointestinal endoscopy are found more frequently with GLP-1 RAs but rarely are associated with pulmonary aspiration. Well-justified recommendations for the periprocedural management of GLP-1RAs (eg, whether to withhold these medications and for how long) are compromised by limited evidence. Important aspects to be considered are (1) their long half-lives, (2) the capacity of GLP-1 receptor agonism to slow gastric emptying even at physiological GLP-1 concentrations, (c) tachyphylaxis observed with prolonged treatment, and (d) the limited effect on gastric emptying in individuals with slow gastric emptying before initiating treatment. Little information is available on the influence of diabetes mellitus itself (ie, in the absence of GLP-1 RA treatment) on retained gastric contents and pulmonary aspiration.

Conclusion: Prolonged fasting periods regarding solid meal components, point-of-care ultrasound examination for retained gastric content, and the use of prokinetic medications like erythromycin may prove helpful and represent an important area needing further study to increase patient safety for those treated with GLP-1 RAs.

Keywords: GLP-1 receptor agonists; GLP-1/GIP dual receptor agonists; aspiration; gastric emptying; incretin mimetics; retained gastric content; tachyphylaxis.

Figure 1.

Effects of GLP-1RAs (lixisenatide and liraglutide) on gastric emptying in T2D. Gastric emptying (half-time) of a standardized meal (∼1175.7 kJ, egg, white bread, margarine, and water) in participants before and after treatment for 10 weeks with lixisenatide (20 µg per day; A) or liraglutide (1.8 mg per day; C). (B) and (D) show the change in gastric emptying half-times from baseline with lixisenatide (B) and liraglutide (D) treatment). (E) The x-fold change in the gastric emptying half-time (y-axis) is plotted vs the baseline gastric emptying half-time (solid components, measured by a 6-hour ¹³C-octanoate breath tests analyzed using Wagner-Nelson equations, x-axis), comparing the effects of the short-acting GLP-1RA lixisenatide (20 µg once daily over 10 weeks, n = 21) and the long-acting GLP-1RA liraglutide (1.8 mg once daily over 10 weeks, n = 26), with baseline gastric emptying (before GLP-1RA treatment). Linear regression analyses (regression equation, r², and related P-value) indicate a significant reduction in GLP-1RA effect to slow gastric emptying in subjects with relatively slower gastric emptying at baseline for both lixisenatide and liraglutide. Data are from Quast et al (40). Statistical analysis: linear regression analysis (regression line, r², and P-value). Please refer to Supplementary Table S1 for additional information on the effect of GLP-1RAs on gastric emptying in subjects with T2D and obesity. Abbreviations: GLP-1, glucagon-like peptide-1; RA, receptor agonist; T2D, type 2 diabetes.

Figure 2.

Duration of therapy with GLP-1RAs and their effects on gastric emptying in studies employing scintigraphy or ¹³C-octanoate breath tests (for the determination of gastric emptying half-times for solid meal components; A) or paracetamol absorption (expressed as the reduction in the area under the curve of paracetamol plasma concentrations during the first hour after a meal test; B), both plotted against the time after initiating GLP-1RA treatment. Pooled effect sizes for data displayed in panel (A) are shown separately for short-acting (exenatide twice a day, lixisenatide) and long-acting GLP-1RAs (C). Linear regression analysis (regression equations, r², and related P-values) revealed a significant reduction over time in the prolongation of gastric emptying half-times for long-acting GLP-1RAs supported by a similar trend for paracetamol absorption (A) and indicating tachyphylaxis but not for short-acting GLP-1RAs (B). The overall stronger effect on gastric emptying shown for short-acting GLP-1RAs in panel (C) may in part be the result of a lack of tachyphylaxis for these compounds. However, given the limitations of the paracetamol absorption test, interpretation must be considered in circumspect. Please refer to Supplementary Table S1 for more data on the effect of GLP-1 receptor agonists on gastric emptying in subjects with type 2 diabetes and obesity. Abbreviations: GLP-1, glucagon-like peptide-1; RA, receptor agonist.

Figure 3.

Effects of GLP-1RAs and the GIP/GLP-1 dual receptor agonist tirzepatide on ad libitum caloric intake of test meals in subjects with T2D (A) and obesity without T2D (B). The percent reduction vs baseline energy intake (± SEM) is displayed. Asterisks indicate a significant reduction (P < .05); n.s. indicates not significant (for GLP-1 receptor agonists studied head to head in some studies). Please refer to Supplementary Table S2 for additional information about the effect of GLP-1RAs on ad libitum energy intake and perceptions of hunger, satiation, fullness, and prospective food consumption in individuals with either T2D or obesity without T2D. Abbreviations: GLP-1, glucagon-like peptide-1; RA, receptor agonist; T2D, type 2 diabetes.

Figure 4.

Outcomes of retrospective studies summarizing monocentric experience with the incidence of residual gastric content (A) and evidence for aspiration (B) in the context of upper gastrointestinal endoscopy in participants (T2D and obesity) either treated with various GLP-1RAs (blue bars) or not treated with GLP-1RAs (grey bars). More subjects treated with GLP-1RAs had diabetes. In studies reporting both residual gastric content and aspiration, it can be concluded that 0.2% of those with residual gastric content also had evidence for aspiration (data not shown). For more information, please see Supplementary Table S3. Abbreviations: GLP-1RA, glucagon-like peptide-1 receptor agonist; T2D, type 2 diabetes.

Figure 5.

Indirect evidence to demonstrate that slow pot-prandial gastric emptying should not result in substantial gastric retention 12 hours after the last meal. (A) Six-hour gastric emptying tests (¹³C- octanoate breath for labeling solid meal components) analyzed using Wagner-Nelson equations were performed in 47 T2D subjects at baseline (black curves, with dotted lines indicating ± 95% confidence intervals). Results from subjects with substantially decelerated gastric emptying (at baseline or 10 weeks after initiating treatment with lixisenatide [20 µg/d] or liraglutide [1.8 mg/d]) are also shown. Gastric emptying curves between 6.5 and 12 hours after meal intake were interpolated assuming an exponential decay (from an initial plateau of 100%) based on the measurements during the initial 6 hours. Even those with slow gastric emptying achieved a gastric content near 0% by 12 hours, which would be a typical interval between a last meal (eg, dinner at 7 Pm) and an early procedure (endoscopy or general anesthesia) in the morning of the following day. Data used for display and modeling are from Quast et al (40). (B) Residual gastric content from food ingested on the previous day (or earlier) as visualized in the gastric antrum by upper gastrointestinal tract endoscopy (courtesy Diabetes Center Bad Lauterberg). Abbreviations: T2D, type 2 diabetes.

Figure 6.

Schematic diagram illustrating postprandial gastric emptying and interdigestive motility (especially migrating motor complex phase 3) that clear the upper gastrointestinal tract of residue, including indigestible, gastric content. While meal-related gastric emptying normally extends over a period of approximately 6 hours, after which the bulk of a meal has emptied, interdigestive gastrointestinal motility occurs at regular intervals between meals and is interrupted by meal ingestion. GLP-1 and GLP-1RAs decelerate meal-related gastric emptying (see Figs. 1 and 2 and Supplementary Table S1). Exogenous GLP-1 at both physiological and pharmacological doses has been shown to reduce the frequency of migrating motor complex phase 3 in individuals without diabetes (14). Studies with GLP-1RAs have not been reported. Abbreviations: GLP-1, glucagon-like peptide-1; RA, receptor agonist.

Figure 7.

Clinical recommendations for subjects treated with short- or long-acting GLP-1RAs who are scheduled for upper gastrointestinal endoscopy or general anesthesia. The recommendations reflect current knowledge about the effects of GLP-1 RAs on gastric emptying and represent the opinion of the authors that has not been endorsed by any learned society. Abbreviations: GLP-1RA, glucagon-like peptide-1 receptor agonist.