Effect of GLP-1 receptor agonists on upper gastrointestinal endoscopy outcomes: a systematic review and meta-analysis

Konstantinos Malandrisa, Savvas Papachristoua, Lito Psychou Derkab, Georgios Kalopitasc, Eleni Gkouraa, Aris Liakosa, Thomas Karagiannisa, Eleni Theocharidoua, Anastasios Manolakisb, Eleni Bekiaria, Apostolos Tsapasa,d

Aristotle University of Thessaloniki, Greece; University of Thessaly, Larissa, Greece; University of Oxford, UK

aSecond Medical Department, Aristotle University of Thessaloniki, Thessaloniki, Greece (Konstantinos Malandris, Savvas Papachristou, Eleni Gkoura, Aris Liakos, Thomas Karagiannis, Eleni Theocharidou, Eleni Bekiari, Apostolos Tsapas); bDepartment of Gastroenterology, University of Thessaly, Larissa, Greece (Lito Psychou Derka, Anastasios Manolakis); cFirst Medical Department, Aristotle University of Thessaloniki, Thessaloniki, Greece (Georgios Kalopitas); dHarris Manchester College, University of Oxford, Oxford, United Kingdom (Apostolos Tsapas)

Correspondence to: Konstantinos Malandris MD, MSc, Second Medical Department, Aristotle University of Thessaloniki, Konstantinoupoleos 49, 546 42, Thessaloniki, Greece, e-mail: kostas_malandris@yahoo.gr
Received 3 January 2026; accepted 1 March 2026; published online 24 April 2026
DOI: https://doi.org/10.20524/aog.2026.1067
© 2026 Hellenic Society of Gastroenterology

Abstract

Background Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) delay gastric emptying, raising concerns about potential aspiration risk during upper gastrointestinal (GI) endoscopy. We conducted a systematic review and meta-analysis to evaluate the effect of GLP-1 RA therapy on procedural outcomes in patients undergoing upper GI endoscopy.

Methods We searched Medline and Cochrane library up to July 2025 without restrictions. Eligible studies evaluated patients undergoing upper GI endoscopy, comparing those taking GLP-1 RAs with those who were not. Outcomes of interest were the incidence of retained gastric contents (RGC), bronchopulmonary aspiration, and procedure discontinuation. Pooled estimates are expressed as odds ratios (ORs) with 95% confidence intervals (CIs), using a random-effects meta-analysis with inverse variance weighting.

Results Twenty-four observational studies, predominantly retrospective, met the inclusion criteria: these comprised 184,707 participants, of whom 59,095 were taking GLP-1 RAs. Mean age was 58.7 years, 48.8% were women, and 51.2% had type 2 diabetes. Use of GLP-1 RAs was associated with higher rates of RGC (OR 4.82, 95%CI 3.66-6.35) and procedure discontinuation (OR 3.93, 95%CI 2.42-6.39) compared with control treatment. In contrast, the incidence of aspiration events was similar between groups (OR 1.1, 95%CI 0.84-1.48). Results remained consistent in a sensitivity analysis based on propensity score matching to control for confounders.

Conclusions GLP-1 RA therapy is associated with a greater incidence of RGC and higher rates of endoscopy termination, but not with a higher risk of aspiration. Adjusting the fasting duration, rather than routinely discontinuing GLP-1 RAs, may represent a reasonable management approach.

Keywords Upper gastrointestinal endoscopy, glucagon-like peptide-1 receptor agonists, systematic review, meta-analysis

Ann Gastroenterol 2026; 39 (4): 452-463

Introduction

Following their approval, glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have become one of the most widely used drug classes for the management of type 2 diabetes (T2D) [1]. Their widespread adoption has also been fueled by their proven efficacy in promoting weight loss among individuals with obesity, irrespective of diabetes status [2,3]. In addition, emerging evidence of potential therapeutic benefits in other conditions, including chronic kidney disease, neurodegenerative disorders and metabolic dysfunction-associated steatotic liver disease, is expected to expand their clinical indications [4,5].

GLP-1 receptor activation in the gastrointestinal (GI) tract reduces GI motility, thereby delaying gastric emptying. As a result, nausea, vomiting and abdominal discomfort are among the most commonly reported adverse effects in individuals treated with these agents [6]. The delayed gastric emptying associated with GLP-1 RAs has raised concerns for patients undergoing upper GI endoscopy, due to the potential risk of aspiration during sedation. In June 2023, the American Society of Anesthesiologists (ASA) issued a consensus-based guidance recommending that GLP-1 RAs be withheld prior to procedures [7]. In view of the limited evidence, the American Gastroenterological Association (AGA) released a clinical practice update advocating for an individualized, risk-based approach to the preprocedural management of these agents [8]. More recently, a multisociety clinical practice guidance document emphasized the importance of shared decision-making regarding GLP-1 RA management, while also highlighting the paucity of evidence on the optimal perioperative approach [9].

Likewise, recent observational studies could not establish a clear link between the preprocedural use of GLP-1 RAs and an increased risk of pulmonary aspiration during upper GI endoscopy [10,11]. To clarify this potential association and provide a comprehensive synthesis of the available evidence, we conducted a systematic review and meta-analysis assessing the effect of GLP-1 RA use on procedural outcomes in patients undergoing upper GI endoscopy.

Materials and methods

This systematic review and meta-analysis were conducted according to a prespecified protocol registered in PROSPERO (registration no. CRD420251167356). We report our methods and results in line with the PRISMA and MOOSE statements (Supplementary Tables 1 and 2) [12,13].

Eligibility criteria

We included comparative studies (randomized controlled trials or observational studies) that enrolled adult participants (aged >18 years) undergoing upper GI endoscopy, irrespective of indication. Eligible studies compared patients receiving GLP-1 RAs with those not on GLP-1 RA therapy. For semaglutide, both oral and injectable formulations were considered. We also included studies that recruited participants receiving tirzepatide. The primary outcome of interest was the incidence of retained gastric content (RGC), defined as the presence of any amount of either solid or liquid material in the stomach during endoscopy. Secondary outcomes of interest were the incidence of bronchopulmonary aspiration and rates of procedure discontinuation. We excluded non-comparative studies and studies not reporting at least 1 outcome of interest. In addition, we excluded studies comparing patients receiving GLP-1 RAs with those having discontinued GLP-1 RA therapy prior to endoscopy.

Information sources and study selection

We searched Medline via PubMed and the Cochrane Database from inception to July 7, 2025, without language or publication type restrictions. The search strategy combined free-text terms and controlled vocabulary, as presented in Supplementary Tables 3 and 4. To identify additional eligible studies, we manually screened the reference lists of relevant systematic reviews. We did not search conference proceedings from relevant scientific meetings, or contact study authors for additional information.

Search results were imported into reference manager software (Endnote, Clarivate Analytics), and duplicate records were removed. Two reviewers working independently assessed record eligibility initially at title and abstract level and then in full text. Any disagreements during the study selection process were resolved by a third reviewer. The study selection process was performed using the Covidence web application (Covidence systematic review software, Veritas Health Innovation).

Data extraction and risk of bias assessment

Two reviewers working independently extracted data from eligible studies using predesigned and pilot-tested forms. Extracted information included study characteristics, participant baseline characteristics and outcomes of interest. For studies that included participants undergoing combined upper GI endoscopy and colonoscopy, we preferentially extracted data specific to patients who underwent upper endoscopy alone, given the confounder of preprocedural colonoscopy preparation. In cases of multiple publications with potentially overlapping cohorts, we extracted data from the full text publication that provided the largest amount of information. Any disagreements during the data extraction process were resolved either through discussion between the original reviewers or by a senior reviewer.

Two reviewers independently assessed the risk of bias of the included studies, using the Risk of Bias in Non-randomized Studies – of Interventions, Version 2 (ROBINS-I v.2) tool [14]. Notably, all studies included in this systematic review and meta-analysis were observational. The assessment focused on the following domains: bias due to confounding, bias in classification of interventions, bias in the selection of participants included in the study (or in the analysis), bias due to deviations from intended interventions, bias due to missing data, bias arising from measurement of the outcome, and bias in selection of the reported result. Details on risk of bias judgments are presented in the electronic supplementary material. Any disagreements during the risk of bias assessment were resolved either through discussion between the original reviewers or by a senior reviewer.

Statistical analysis

All outcomes are reported as odds ratios (ORs) along with 95% confidence intervals (CIs). Given the expected between-study heterogeneity, we conducted a random-effects meta-analysis using inverse variance weighting. Between-study variance (Tau2) was estimated using the Restricted Maximum Likelihood (REML) method [15,16]. When at least 3 studies were available and Tau2 was greater than zero, the Hartung-Knapp-Sidik-Jonkman (HKSJ) method was applied to calculate confidence intervals for the pooled effect estimates [15,16]. In cases with only 2 studies, or where Tau2 was estimated as zero, the confidence intervals for the pooled effect estimates were calculated based on the Wald-type method [15]. Statistical heterogeneity was assessed using the chi-squared (χ2) test (with a significance threshold of P<0.10) and the I2 statistic, with I2 values exceeding 50% indicating substantial heterogeneity [16]. Studies with zero events in both arms were excluded from the meta-analysis. For the main analyses, we calculated prediction intervals (PIs) to quantify the expected range of effects in future studies. When at least 10 studies were included in the meta-analysis, we assessed publication bias through visual inspection of funnel plots [17]. All analyses were performed using RevMan Version 7.2.0.

Additional analysis

To assess the robustness of our findings and explore potential sources of heterogeneity we performed several sensitivity analyses. For all outcomes we conducted sensitivity analyses including only studies that: i) employed propensity score matching at baseline to control for confounders; ii) recruited solely participants with T2D; iii) were at low risk for bias; and iv) were published in full text. For the primary outcome, we conducted a post hoc analysis pooling adjusted ORs from primary studies, and a subgroup analysis based on the specific GLP-1 RA agent used. These additional analyses were not feasible for the incidence of aspiration, given the limited number of studies with available data.

One reviewer assessed the certainty in effect estimates from main analyses following the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach using the online GRADEpro Guideline Development Tool software [18]. We took into consideration the following domains: study design, risk of bias, inconsistency, indirectness, imprecision, and publication bias.

Results

Search results

Fig. 1 depicts the study selection process. In total, 255 records were screened at title and abstract level, of which 58 were reviewed in full text. Ultimately, 24 records describing 24 studies with 181,770 participants were included in the systematic review and meta-analysis [11,19-41]. During the full-text screening process, 3 studies were excluded because of concerns about overlapping cohorts [10,42,43].

thumblarge

Figure 1 Flow diagram of the study selection process

SRs, systematic reviews

Source: Page MJ, et al. BMJ 2021;372:n71. doi: 10.1136/bmj.n71.

This work is licensed under CC BY 4.0. To view a copy of this license, visit https://creativecommons.org/licenses/by/4.0/

Study and participant characteristics

The main characteristics of the included studies are presented in Table 1. Most studies were published between 2023 and 2025, and originated predominantly from the United States. Eight studies (33.3%) were identified solely as conference abstracts [20,21,25,29,30,32,33,39]. The majority of included studies were single-center, employing a retrospective design. Included studies mainly focused on participants undergoing elective procedures. Recruitment periods across studies spanned from 2005-2024. Fifteen studies (62.5%) applied propensity score matching for multiple variables (including age, T2D, body mass index [BMI] and sex) to ensure balanced baseline characteristics between groups [11,19-23,26,28,30,31,35,37-40]. Fasting protocols prior to procedures were largely unavailable across the included studies, and when reported, they were inconsistent. Supplementary Table 5 summarizes the fasting durations and dietary instructions as provided in the included studies. Among the 181,770 participants 59,098 (32.5%) were receiving GLP-1 RAs, 88,698 (48.8%) were females and 93,101 (51.2%) had T2D. Among studies with available data, the average mean age and BMI were 58.7 years and 33.0 kg/m2, respectively. Among participants treated with GLP-1 RAs, 10,387 (17.6%) received dulaglutide, 8352 semaglutide (14.1%), 7428 liraglutide (12.6%), and 518 (0.8%) tirzepatide. For the remaining 32,413 participants, the specific GLP-1 RA used was not reported.

Table 1 Baseline characteristics of included studies

thumblarge

RGC

Twenty studies, including 5619 participants in the GLP-1 RA group and 45,299 participants in the control group, reported rates of RGC [19-21,23,25-36,38-41]. The overall incidence of RGC was 11.0% among GLP-1 RA users and 2.7% in the control group. Treatment with GLP-1 RAs was associated with higher odds of RGC compared with controls (OR 4.82, 95%CI 3.66-6.35; I2=63%) (Fig. 2). Results remained consistent when prediction intervals were applied (95% PI 1.95-11.88) Similarly, findings were robust in a sensitivity analysis restricted to studies employing propensity score matching to adjust for baseline confounders (OR 3.79, 95%CI 2.96-4.84; I2=10%) (Supplementary Fig. 1). Comparable results were observed in an analysis limited to studies enrolling only participants with T2D (OR 4.09, 95%CI 3.06-5.45; I2=0%) (Supplementary Fig. 2). When synthesis was restricted to studies at low risk of bias or full-text publications, the effect estimates remained consistent with those of the primary analysis (Supplementary Figs. 3 and 4 respectively). The odds of RGC were similar across different GLP-1 RAs regimens, ranging from 3.48 for liraglutide to 4.35 for semaglutide (Supplementary Fig. 5). A meta-analysis of adjusted ORs yielded similar estimates to our main findings (OR 3.23, 95%CI 2.33-4.47; I2=70%) (Supplementary Fig. 6).

thumblarge

Figure 2 Meta analysis results for GLP-1 RAs vs. No GLP-1 RAs for incidence of retained gastric content

GLP-1 RAs, glucagon-like peptide 1 receptor agonists; CI, confidence interval

Aspiration

Thirteen studies, including 55,751 participants in the GLP-1 RA group and 113,240 participants in the control group, reported data on aspiration events [11,20-22,24,26,32,34-38,41]. The overall incidence of aspiration was 0.4% among GLP-1 RA users and 0.1% in the control group. Six studies were excluded from the meta-analysis as they had zero events in both arms [20,21,24,26,35,36]. The pooled analysis showed no significant difference in the odds of aspiration between the GLP-1 RA and control groups (OR 1.11, 95%CI 0.84-1.48; I2=9%) (Fig. 3). Results remained consistent when prediction intervals were applied (95%PI 0.76-1.63). Similar findings were observed in a sensitivity analysis restricted to studies employing propensity score matching to adjust for baseline confounders (OR 1.08, 95%CI 0.76-1.54; I2=17%) (Supplementary Fig. 7), studies enrolling only participants with T2D (OR 0.94, 95%CI 0.71-1.25; I2=0%) (Supplementary Fig. 8), studies at low risk for bias (OR 1.09, 95%CI 0.82-1.46; I2=12%) (Supplementary Fig. 9), and studies published as full text (OR 1.11, 95%CI 0.80-1.53; I2=10%) (Supplementary Fig. 10). Across individual agents, the rates of aspiration were 0.2% for semaglutide, 0.5% for liraglutide and 0.4% for dulaglutide.

thumblarge

Figure 3 Meta analysis results for GLP-1 RAs vs. No GLP-1 RAs for incidence of aspiration

GLP-1 RAs, glucagon-like peptide 1 receptor agonists; CI, confidence interval

Procedure discontinuation

Eleven studies, including 28,453 participants in the GLP-1 RA group and 87,879 participants in the control group, reported rates of procedure discontinuation [11,20,21,23,24,27,29,30,34,36,40]. The overall incidence of procedure discontinuation was 1.0% among GLP-1 RA users and 0.3% in the control group. One study was excluded from the meta-analysis as it had zero events in both arms [40]. Procedure discontinuation was more likely to occur in GLP-1 RA users compared to controls (OR 3.93, 95%CI 2.42-6.39; I2=53%) (Fig. 4). Results remained consistent when prediction intervals were applied (95%PI 1.16-13.38). Similar findings were observed in a sensitivity analysis based on propensity score matching, presence of T2D, low risk of bias and full-text publication (Supplementary Figs. 11-14, respectively).

thumblarge

Figure 4 Meta analysis results for GLP-1 RAs vs. No GLP-1 RAs for incidence of procedure discontinuation

GLP-1 RAs, glucagon-like peptide 1 receptor agonists; CI, confidence interval

Risk of bias and publication bias assessment

Most studies were at low risk for bias for the primary outcome (Supplementary Table 6). Four studies raised serious concerns over bias, mainly due to residual confounding [20,21,25,33]. Four additional studies published as conference abstracts were at moderate risk for bias, mainly due to concerns related to missing data [29,30,32,39]. Supplementary Tables 7 and 8 present the risk of bias assessment for the incidence of aspiration and procedure discontinuation, respectively. Visual inspection of funnel plots did not suggest the presence of small-study effect bias for RGC or procedure discontinuation rates (Supplementary Figs. 15-16). Owing to the limited number of studies reporting aspiration events, a respective analysis for this outcome could not be performed.

Grading of evidence

The confidence in the effect estimates for all outcomes was rated as very low (Supplementary Table 9). This was primarily due to imprecision in the effect estimates, arising from the limited number of events in both groups across all outcomes, particularly for the incidence of aspiration. In addition, the certainty of evidence was downgraded by 1 level because all available data were derived from observational studies.

Discussion

In this systematic review and meta-analysis, we assessed the effect of GLP-1 RA use on the risk of RGC, aspiration and procedure discontinuation in patients undergoing upper GI endoscopy. Our findings indicate that GLP-1 RA users are more likely to experience RGC (OR 4.82, 95%CI 3.66-6.35) or an aborted upper GI endoscopy (OR 3.93, 95%CI 2.42-6.39). This association remained consistent across multiple sensitivity analyses, including studies employing propensity score matching at baseline and those primarily involving participants with T2D. Moreover, the odds of RGC were comparable across different GLP-1 RAs. Notably, despite the increased risk of RGC, GLP-1 RA use was not associated with a higher incidence of aspiration. The absence of a statistically significant difference between groups in terms of aspiration, should be interpreted cautiously, given the small number of events in the groups (0.4% among GLP-1 RA users and 0.1% in the control group). Furthermore, these estimates derive from observational studies; therefore, the potential for residual confounding remains, despite the use of various methods in the primary studies to control for measured confounders. The confidence in our estimates was rated as very low, mainly because of the small number of events across all outcomes.

This systematic review and meta-analysis provide an updated summary of the available evidence on the effect of GLP-1 RA use on outcomes related to upper GI endoscopy. We included 24 studies, comprising more than 180,000 participants, while accounting for potential cohort overlap. The analysis adhered to rigorous methodological standards consistent with the latest Cochrane recommendations [15], and the robustness of the findings was confirmed through multiple sensitivity analyses. We evaluated the risk of bias using the recently revised ROBIN-I v.2 tool and assessed the certainty in our estimates using robust methodology.

Nevertheless, certain limitations should be acknowledged. Moderate to substantial heterogeneity was observed in most primary analyses, especially for the incidence of RGC (I2=63%). This heterogeneity was notably reduced in sensitivity analyses that included only propensity score-matched studies (I2=10%), suggesting that differences in methodological approaches were major contributors to the variability across effect estimates. Additional heterogeneity may have arisen from variations in fasting protocols, patient selection criteria and procedural sedation practices across studies. Moreover, the definition and grading of RGC were not uniform among the included studies. In our study, all types and grades of RGC were pooled together; this approach may have further contributed to the observed heterogeneity. In addition, 8 of the 24 studies (33%) were identified solely as conference abstracts; thus, a thorough evaluation of participant characteristics and applied methodology was not possible. However, sensitivity analyses excluding these studies yielded similar results to our main findings, supporting the robustness of our results across all main outcomes despite the inclusion of several studies published solely in abstract form.

Our findings are generally in line with previously published pertinent meta-analyses [44-48]. Facciorusso et al analyzed 13 studies including 84,065 participants and reported higher odds of RGC (OR 5.56, 95% CI 3.35-9.23) and procedure discontinuation (OR 5.13, 95%CI 3.01-8.75) among GLP-1 RA users compared with controls, while rates of aspiration did not differ significantly between groups (OR 1.75, 95%CI 0.64-4.77) [44]. Similarly, Abdulraheem et al synthesized 20 studies involving 164,222 participants and found comparable estimates, again showing no increase in aspiration events among GLP-1 RA users [45]. Consistent conclusions were also reached by Singh et al and Baig et al, who reported an increased risk of RGC and procedure discontinuation in GLP-1 RA users, without a corresponding increase in aspiration risk [47,48].

While our pooled estimates align with prior analyses, our study builds upon the existing literature through several methodological advantages. In contrast to all previously published meta-analyses, we provide prediction intervals to convey the expected range of effects in future studies, thereby enhancing the clinical interpretability of our results. Furthermore, whereas several previous systematic reviews and meta-analyses relied on the Newcastle–Ottawa Scale [44,46-48], we evaluated methodological quality using the updated ROBINS-I v.2 tool, allowing for a more comprehensive and methodologically rigorous assessment of bias in non-randomized studies of interventions. Moreover, in contrast to Abdulraheem et al, Singh et al and Tan et al, we formally appraised the certainty in our pooled estimates using the GRADE methodological framework, thereby strengthening the interpretability and clinical credibility of our results. Interestingly, 3 previous meta-analyses identified a protective effect of concurrent colonoscopy against RGC among GLP-1 RA users [45-47]. This observation may be attributed to the prolonged fasting period required for colonoscopy preparation, which typically involves a clear liquid diet the day before the procedure. Supporting this interpretation, quantitative evidence indicates that GLP-1 RA therapy is associated with an approximately 36-min delay on solid-phase scintigraphy and no significant differences in modalities reflective of liquid emptying [49].

In its position statement, the American Society for Gastrointestinal Endoscopy recommends withholding GLP-1 RAs on the day of the procedure for patients on daily formulations, and for 1 week in those receiving weekly dosing [50]. This approach is supported by a multicenter study involving 815 participants on GLP-1 RAs, which reported significantly lower rates of RGC among patients who withheld GLP-1 RAs before the procedure compared with those who did not (4.4% vs. 12.7%, P<0.001) [51]. However, in a cohort study including 629 bariatric patients, Jirapinyo et al observed no significant difference in the rates of early termination of upper endoscopy (1.4% vs. 0%, P=0.36) or RGC (6.4% vs. 2.7%, P=0.36) between patients who continued vs. those who discontinued GLP-1 RA therapy [52]. Taking this further, Santos et al evaluated the impact of varying preprocedural semaglutide interruption intervals on RGC during upper endoscopy [53]. Based on their findings, discontinuation of semaglutide for more than 21 days in patients with ongoing GI symptoms, and more than 14 days in asymptomatic individuals, was associated with comparable RGC rates to non-semaglutide users.

Withholding GLP-1 RAs in patients with obesity but without diabetes may pose fewer concerns, whereas interruption in patients with diabetes warrants more caution. Previous studies have identified poorly controlled diabetes and hyperglycemia as independent risk factors for gastric retention, suggesting that such patients may benefit from a prolonged (24 h) preprocedural liquid diet [9,51]. Phan et al, reported that, for every 1% increase in hemoglobin A1c, there was a 36% increase in the odds of retained food, after adjusting for medication type and GLP-1 RA withholding status [51]. Other factors that should be considered include the phase of GLP-1 RA therapy. Patients in the maintenance phase appear to be at lower risk of delayed gastric emptying compared to those who have recently undergone dose escalation [9]. Moreover, patients on weekly GLP-1 RA regimens are more likely to experience GI symptoms suggestive of gastric retention, in a dose-dependent manner, than those receiving daily formulations [9].

When clinical concern for RGC exists on the day of the procedure, such as in patients presenting with nausea, bloating, or abdominal discomfort despite prior GLP-1 RA discontinuation, point-of-care gastric ultrasound may be utilized to assess aspiration risk [54]. In such cases, a shared decision-making approach should be adopted, weighing the benefits and risks of proceeding with rapid sequence induction and tracheal intubation under general anesthesia to mitigate aspiration risk, vs. postponing or canceling the procedure [9]. It should also be noted that, in patients with persistent GI symptoms, even longer GLP-1 RA interruption intervals (exceeding 1 week) may be necessary to reduce the risk of RGC [53].

The current evidence regarding the impact of GLP-1 receptor agonists on upper GI endoscopy-related outcomes is largely based on observational, retrospective studies. Accordingly, prospective studies are urgently needed to generate higher-quality data, though their feasibility may be limited by the low incidence of key outcomes, necessitating large sample sizes. Most existing studies have excluded patients with neurological disorders, gastroparesis or surgically altered GI anatomy, populations in whom the true risk of RGC and aspiration may be greater. This underscores the need for further investigation in these higher-risk subgroups. Furthermore, the majority of available data derive from patients undergoing elective procedures, while hospitalized individuals were generally excluded. As a result, the generalizability of current findings to inpatient settings remains uncertain. Finally, most of the available evidence pertains to participants receiving semaglutide, liraglutide or dulaglutide, with only limited data currently available for tirzepatide.

In conclusion, data from observational studies suggest an increased risk of RGC and procedure discontinuation among patients receiving GLP-1 RAs who undergo upper GI endoscopy, although there is no clear evidence of an elevated risk for aspiration. Therefore, an individualized risk assessment with adjustment of fasting duration, rather than routine discontinuation of GLP-1 RAs prior to upper GI endoscopy, may represent a reasonable approach for these patients.

Summary Box

What is already known:

  • Glucagon-like peptide-1 receptor agonists (GLP-1RAs) are widely used for the management of diabetes

  • The delayed gastric emptying associated with GLP-1 RAs has raised concerns for patients undergoing upper gastrointestinal (GI) endoscopy, due to the potential risk of aspiration during sedation

  • Observational studies have yielded inconsistent findings regarding whether preprocedural use of GLP-1 RAs increases the risk of pulmonary aspiration during upper GI endoscopy

What the new findings are:

  • Treatment with GLP-1 RAs is associated with higher odds of retained gastric content during upper GI endoscopy compared with controls

  • Procedure discontinuation is more likely to occur in GLP-1 RA users compared to controls

  • GLP-1 RA use is not associated with higher rates of aspiration during upper GI endoscopy

References

1. Yao H, Zhang A, Li D, et al. Comparative effectiveness of GLP-1 receptor agonists on glycaemic control, body weight, and lipid profile for type 2 diabetes:systematic review and network meta-analysis. BMJ 2024;384:e076410.

2. Wilding JPH, Batterham RL, Calanna S, et al;STEP 1 Study Group. Once-weekly semaglutide in adults with overweight or obesity. N Engl J Med 2021;384:989-1002.

3. Davies M, Færch L, Jeppesen OK, et al;STEP 2 Study Group. Semaglutide 2·4 mg once a week in adults with overweight or obesity, and type 2 diabetes (STEP 2):a randomised, double-blind, double-dummy, placebo-controlled, phase 3 trial. Lancet 2021;397:971-984.

4. Sanyal AJ, Newsome PN, Kliers I, et al;ESSENCE Study Group. Phase 3 trial of semaglutide in metabolic dysfunction-associated steatohepatitis. N Engl J Med 2025;392:2089-2099.

5. Perkovic V, Tuttle KR, Rossing P, et al;FLOW Trial Committees and Investigators. Effects of semaglutide on chronic kidney disease in patients with type 2 diabetes. N Engl J Med 2024;391:109-121.

6. Karagiannis T, Malandris K, Avgerinos I, et al. Subcutaneously administered tirzepatide vs semaglutide for adults with type 2 diabetes:a systematic review and network meta-analysis of randomised controlled trials. Diabetologia 2024;67:1206-1222.

7. Joshi GP, Abdelmalak BB, Weigel WA, et al;American Society of Anesthesiologists (ASA) Task Force on Preoperative Fasting. American Society of Anesthesiologists consensus-based guidance on preoperative management of patients (adults and children) on glucagon-like peptide-1 (GLP-1) receptor agonists. 2023. Available from:https://www.asahq.org/about-asa/newsroom/news-releases/2023/06/american-society-of-anesthesiologists-consensus-based-guidance-on-preoperative [Accessed 26 March 2026].

8. Hashash JG, Thompson CC, Wang AY. AGA rapid clinical practice update on the management of patients taking GLP-1 receptor agonists prior to endoscopy:communication. Clin Gastroenterol Hepatol 2024;22:705-707.

9. Kindel TL, Wang AY, Wadhwa A, et al;Society of American Gastrointestinal and Endoscopic Surgeons. Multisociety clinical practice guidance for the safe use of glucagon-like peptide-1 receptor agonists in the perioperative period. Surg Obes Relat Dis 2024;20:1183-1186.

10. Yeo YH, Gaddam S, Ng WH, Huang PC, Ma KS, Rezaie A;Motility and Metabolic Pharmacoepidemiology Group. Increased risk of aspiration pneumonia associated with endoscopic procedures among patients with glucagon-like peptide 1 receptor agonist use. Gastroenterology 2024;167:402-404.

11. Alkabbani W, Suissa K, Gu KD, et al. Glucagon-like peptide-1 receptor agonists before upper gastrointestinal endoscopy and risk of pulmonary aspiration or discontinuation of procedure:cohort study. BMJ 2024;387:e080340.

12. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement:an updated guideline for reporting systematic reviews. BMJ 2021;372:n71.

13. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology:a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008-2012.

14. The risk of bias in non-randomised studies –of interventions, version 2 (ROBINS-I V2). Risk-of-bias.info website, 22 Nov 2024. Available from:https://www.riskofbias.info [Accessed 26 March 2026].

15. Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for systematic reviews of interventions version 6.5 (updated August 2024). Cochrane, 2024. Available from:https://www.cochrane.org/handbook [Accessed 26 March 2026].

16. Deeks JJ, Higgins JPT, Altman DG, McKenzie JE, Veroniki AA (editors). Chapter 10:Chapter 10:Analysing data and undertaking meta-analyses [last updated November 2024]. In:Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane handbook for systematic reviews of interventions version 6.5. Cochrane, 2024. Available from:https://cochrane.org/handbook [Accessed 26 March 2026].

17. Debray TPA, Moons KGM, Riley RD. Detecting small-study effects and funnel plot asymmetry in meta-analysis of survival data:a comparison of new and existing tests. Res Synth Methods 2018;9:41-50.

18. Guyatt G, Oxman AD, Akl EA, et al. GRADE guidelines:1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol 2011;64:383-394.

19. Abu-Freha N, Levi Z, Nevo-Shor A, et al. The impact of glucagon-like peptide-1 receptor agonist on the gastric residue in upper endoscopy. Diabetes Res Clin Pract 2024;217:111900.

20. Ayoub M, Chumbe JT, Varga SJ, Reddy Naravadi VV, Daglilar E, Faris C. The impact of glucagon-like peptide-1 receptor agonists use prior to esophagogastroduodenoscopy. Gastrointest Endosc 2024;99 Suppl:AB261.

21. Bahdi F, Bachu VS, Aldzhyan V, et al. S927 Risk of gastric food retention with glucagon-like peptide 1 receptor agonists versus diabetes mellitus:a case-control study from a large tertiary care health system. Am J Gastroenterol 2024;119(10S):S649.

22. Barlowe TS, Anderson C, Sandler RS, et al. Glucagon-like peptide-1 receptor agonists do not increase aspiration during upper endoscopy in patients with diabetes. Clin Gastroenterol Hepatol 2025;23:739-747.

23. Chapman MB, Norwood DA, Price C, et al. Effects of glucagon-like peptide-1 receptor agonists on gastric mucosal visibility and retained gastric contents during EGD. Gastrointest Endosc 2024;100:923-927.

24. Dev B, Hadi Y, Rizvi A, Cao C, Horwich B, Hoerter NA. Low rates of aborted endoscopy due to gastric food retention in patients on glucagon-like-peptide-1 receptor agonists. Dig Dis Sci 2025;70:1838-1843.

25. Elliott M, Travis B, King B, Hernandez MG, Khan S, Bay C. S896 GLP-1 receptor agonists and retained gastric content in type 2 diabetics:insights from a retrospective endoscopic analysis, a preliminary study. Am J Gastroenterol 2024;119(10S):S624-S625.

26. Garza K, Aminpour E, Shah J, et al. Glucagon-like peptide-1 receptor agonists increase solid gastric residue rates on upper endoscopy especially in patients with complicated diabetes:a case-control study. Am J Gastroenterol 2024;119:1081-1088.

27. Robalino Gonzaga E, Farooq A, Mohammed A, et al. Real-world impact of GLP-1 receptor agonists on endoscopic patient outcomes in an ambulatory setting:a retrospective study at a large tertiary center. J Clin Med 2024;13:5403.

28. Gu GH, Pauplis C, Seacor T, Devuni D, Krishnarao A. Association of semaglutide with retained gastric contents on endoscopy:retrospective analysis. Endosc Int Open 2025;13:a25501468.

29. Hernandez L, Ukwade D, Modi Z, et al. GLP-1 receptor agonist effect on retained gastric contents on EGD:a quality improvement study. Gastrointest Endosc 2024;99(Suppl):AB382.

30. Karlson R, Clukey J, Beck V, et al. Mo2015 Clinical impact of GLP-1RA on endoscopy:a retrospective cohort study in endoscopy patients. Gastroenterology 2024;166(Suppl):S-1197.

31. Kobori T, Onishi Y, Yoshida Y, et al. Association of glucagon-like peptide-1 receptor agonist treatment with gastric residue in an esophagogastroduodenoscopy. J Diabetes Investig 2023;14:767-773.

32. Markley A, Sandhu R, Seltzer E, et al. Glucagon-like peptide-1 receptor agonist (GLP-1 RA) use predicts retained gastric contents but not increased risk of major adverse anesthesia-related events during upper endoscopy. Gastrointest Endosc 2024;99(Suppl):AB1139.

33. Meluban L, Alter A, Spinnell M. Increased risk of residual gastric content in patients on glp-1 agonists undergoing esophagogastroduodenoscopy. Gastroenterology 2024;166(Suppl):S-1189-S-1190.

34. Nadeem D, Taye M, Still MD, et al. Effects of glucagon-like peptide-1 receptor agonists on upper endoscopy in diabetic and nondiabetic patients. Gastrointest Endosc 2024;100:745-749.

35. Nasser J, Hosseini A, Barlow G, et al. Food retention at endoscopy among adults using glucagon-like peptide-1 receptor agonists. JAMA Netw Open 2024;7:e2436783.

36. Panchal S, Mahmud N, Atkins JH, et al. Endoscopy and anesthesia outcomes associated with glucagon-like peptide-1 receptor agonist use in patients undergoing outpatient upper endoscopy. Gastrointest Endosc 2025;102:216-222.

37. Peng CY, Chang YC, Gong C, et al. Association between glucagon-like peptide-1 receptor agonists and aspiration pneumonia during endoscopic procedures. Anesthesiology 2024;141:1009-1012.

38. Quinn JA, Welch KM, Fujino E, et al. Perioperative glucagon-like peptide-1 receptor agonist use and retained gastric contents:A retrospective analysis of patients undergoing elective upper endoscopy. J Clin Anesth 2025;102:111776.

39. Rizvi A, Salgado S, Newberry C, et al. Endoscopic gastric food retention rates for GLP-1RA users can be potentially mitigated by pre-colonoscopy fasting protocols in a propensity-matched cohort study. Gastrointest Endosc 2024;99(Suppl):AB230-AB231.

40. Stark JE, Cole JL, Ghazarian RN, Klass MJ. Impact of glucagon-like peptide-1 receptor agonists (GLP-1RA) on food content during esophagogastroduodenoscopy (EGD). Ann Pharmacother 2022;56:922-926.

41. Wu F, Smith MR, Mueller AL, et al. Association of glucagon-like peptide receptor 1 agonist therapy with the presence of gastric contents in fasting patients undergoing endoscopy under anesthesia care:a historical cohort study. Can J Anaesth 2024;71:958-966.

42. Al Sakka Amini R, Ismail AS, Al-Aqrabawi M, et al. Risk of aspiration pneumonitis after elective esophagogastroduodenoscopy in patients on glucagon-like peptide-1 receptor agonists. Cureus 2024;16:e66311.

43. Suresh Kumar VC, Aloysius M, Nawaz A, Joyce J, Aswath G. Incidence of aspiration pneumonitis and emergent intubation in patients on glucagon-like peptide-1 inhibitors undergoing upper endoscopy and colonoscopy:a comparative study. Gastrointest Endosc 2024;99(Suppl):AB150-AB151.

44. Facciorusso A, Ramai D, Dhar J, et al. Effects of glucagon-like peptide-1 receptor agonists on upper gastrointestinal endoscopy:a meta-analysis. Clin Gastroenterol Hepatol 2025;23:715-725.

45. Abdulraheem A, Abujaber B, Ayers L, et al. Impact of GLP-1 receptor agonists on upper gastrointestinal endoscopy:an updated systematic review and meta-analysis. Surg Endosc 2025;39:5135-5151.

46. Tan Y, Zhang X, Lv XH, Sun YN, Yang JL, Xiao X. Glucagon-like peptide-1 receptor agonists increase the risk of residual gastric content and pulmonary aspiration on upper endoscopy:a meta-analysis. Dig Liver Dis 2025;57:1377-1385.

47. Baig MU, Piazza A, Lahooti A, et al. Glucagon-like peptide-1 receptor agonist use and the risk of residual gastric contents and aspiration in patients undergoing GI endoscopy:a systematic review and a meta-analysis. Gastrointest Endosc 2025;101:762-771.

48. Singh S, Rahman SH, Khan N, et al. Effects of glucagon-like peptide-1 receptor agonists on endoscopy outcomes:systematic review and meta-analysis. Gastrointest Endosc 2025;101:343-349.e5.

49. Hiramoto B, McCarty TR, Lodhia NA, et al. Quantified metrics of gastric emptying delay by glucagon-like peptide-1 agonists:a systematic review and meta-analysis with insights for periprocedural management. Am J Gastroenterol 2024;119:1126-1140.

50. Sharaiha RZ, Shukla AP, Sen S, et al. American Society for Gastrointestinal Endoscopy position statement on periendoscopic management of patients on glucagon-like peptide-1 receptor agonists and sodium-glucose cotransporter-2 inhibitors. Gastrointest Endosc 2025;101:285-294.

51. Phan J, Chang P, Issa D, et al. Glucagon-like peptide receptor agonists use before endoscopy is associated with low retained gastric contents:a multicenter cross-sectional analysis. Am J Gastroenterol 2025;120:554-561.

52. Jirapinyo P, Siranart N, Thompson CC. Glucagon-like peptide-1 receptor agonists and upper endoscopy:a real-world experience. Obesity (Silver Spring) 2025;33:250-256.

53. Santos LB, Mizubuti GB, da Silva LM, et al. Effect of various perioperative semaglutide interruption intervals on residual gastric content assessed by esophagogastroduodenoscopy:A retrospective single center observational study. J Clin Anesth 2024;99:111668.

54. Sen S, Potnuru PP, Hernandez N, et al. Glucagon-like peptide-1 receptor agonist use and residual gastric content before anesthesia. JAMA Surg 2024;159:660-667.

Notes

Conflict of Interest: None