Glucagon-Like Peptide-1 Receptor Agonists and Peri-Procedural Aspiration Risk

Abstract

Context

Delayed gastric emptying caused by glucagon-like peptide-1 receptor agonists (GLP-1RAs) has raised concerns about increased aspiration risk during surgical and endoscopic procedures. In June 2023, the American Society of Anesthesiologists (ASA) recommended discontinuing GLP-1RAs one day (daily users) or one week (weekly users) before elective surgery or endoscopic esophagogastroduodenoscopy (EGD). In October 2024, the ASA reversed the initial recommendation and advised most patients to continue taking GLP-1RAs before elective surgery.

Objective

We conducted a systematic review of the evidence for or against the original recommendation.

Methods

We searched PubMed for retrospective cohort studies published between June 2023 and March 2025 investigating the association between GLP-1RA use and the risk of aspiration/pneumonia in patients undergoing elective surgery or endoscopic procedures. We calculated a summary risk ratio for studies that could be combined.

Results

We identified 3 studies of elective surgery and 4 of EGD using large databases to identify an increased risk of aspiration/pneumonia associated with GLP-1RA use. The 3 elective surgery studies had a combined risk ratio of 1.00 [0.76, 1.30]. The 4 EGD studies had a combined risk ratio of 1.10 [0.95, 1.27]. In one study, a parallel analysis of the aspiration/pneumonia risk associated with opioid medications found a risk ratio of 2.68 [1.89, 3.81], indicating that the methodology could detect an increased risk of aspiration/pneumonia from a motility inhibitor.

Conclusion

Although GLP-1RAs cause delayed gastric emptying, retrospective cohort studies using large real-world evidence databases have not consistently identified a GLP-1RA-associated risk of aspiration/pneumonia for elective surgical and endoscopic procedures.

Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have emerged as a widely prescribed therapeutic option for managing type 2 diabetes and obesity. Data show that 43% of adults who have been told by a doctor that they have diabetes and 22% who have been told by a doctor that they are overweight or obese have used a GLP-1RA [1, 2]. However, their mechanism of action, which includes delayed gastric emptying, has raised concerns about a potential increased risk of aspiration during surgical and endoscopic procedures [3, 4]. In response to these concerns, the American Society of Anesthesiologists (ASA) issued guidance on June 29, 2023, recommending the discontinuation of GLP-1RAs prior to elective procedures [5]. Specifically, the guidance suggested withholding daily GLP-1RAs for one day and weekly formulations for one week prior to the date of the elective procedure. On October 29, 2024, the ASA reversed the initial guidance and stated that most patients should continue taking their GLP-1RAs before elective surgery. This new guidance was released jointly by the ASA along with the American Gastroenterological Association, American Society for Metabolic and Bariatric Surgery, International Society of Perioperative Care of Patients with Obesity, and the Society of American Gastrointestinal and Endoscopic Surgeons [6, 7]. The ASA and the other organizations now recommend that use of GLP-1RAs in the perioperative period should be based on shared decision making by the patient along with procedural, anesthesia, and prescribing care teams balancing the metabolic need for the GLP-1RA with individual patient risk.

The initial ASA guidance did not cite studies showing an increased risk of aspiration/pneumonia in patients on GLP-1RAs undergoing elective procedures. Since the initial guidance, multiple retrospective cohort studies were reported comparing GLP-1RA users to nonusers for risks of postoperative respiratory complications, including aspiration and pneumonia, in various postoperative or postendoscopic settings [8-15]. To our knowledge, no review of this literature has been published evaluating relevant risks of GLP-1RAs administered prior to elective surgery or endoscopy. During the period of June 29, 2023, to October 29, 2024, there was pushback to the initial guidance [16] because of (i) the potential adverse impact of withholding GLP-1RAs on pre- and postoperative glycemic control in people with diabetes; and (ii) workflow interruptions for surgeons and hospital operating rooms when patients using GLP-1RAs unaware of a requirement to discontinue had to cancel procedures at the last moment.

We performed a systematic review and meta-analysis of retrospective cohort studies using data collected prior to the initial ASA guidance that were published between June 2023 and March 2025. We included studies that compared GLP-1RA users to nonusers for the risk of postoperative respiratory complications in patients undergoing elective surgery or endoscopic procedures. Our review aimed to assess the evidence supporting or refuting the initial ASA recommendations for preoperative GLP-1RA management.

Methods

Literature Search and Screening of Studies

We performed a PubMed search for all English language reports published from June 2023 to March 2025 that reported the association between GLP1-RAs and aspiration risk in esophagogastroduodenoscopy (EGD) or surgery. The search string was:

(((anesthesia OR sedation OR “aspiration pneumonia” OR aspiration OR “aspiration pneumonitis”) AND (surgery OR esophagogastroduodenoscopy OR EGD OR endoscop* OR upper endoscop* OR gastroscopy OR elective) AND (“GLP-1RA” OR “glucagon-like peptide-1 receptor” OR “glucagon-like peptide-1 receptor agonist” OR “GLP-1 receptor” OR “GLP-1 receptor agonist”) AND (complication OR risk OR adverse OR “aspiration pneumonia” OR “pneumonia”)) AND ((“2023/6/1”[Date - Publication] : “3000”[Date - Publication])) AND ((excludepreprints[Filter]) AND (english[Filter]))) AND ((y_1[Filter]) AND (excludepreprints[Filter]) AND (english[Filter]))

Two authors (C.N.H., M.A.K.) independently reviewed all potentially relevant abstracts and determined whether to advance to full-text review. The full-text review determined whether the study met the inclusion criteria for data extraction. Disagreements were resolved through consensus or consultation with a third reviewer (D.C.K.). We included retrospective cohort studies that examined the association between GLP-1RA use and aspiration risk in patients undergoing elective surgery or EGD. Studies were excluded if they focused on pediatric populations, only reported emergency procedures, lacked a control group of GLP-1RA nonusers, or used limited International Classification of Diseases, Tenth Revision, and Clinical Modification (ICD-10-CM) codes for aspiration outcomes. Table 1 presents the full inclusion and exclusion criteria.

Table 1.

Inclusion and exclusion criteria

	Inclusion	Exclusion
Population	• GLP-1RAs users	• Patients <18 years old
Intervention/exposure	• Elective surgery • Elective endoscopy • Anesthesia	• Emergency surgery • Electroconvulsive therapy
Outcome	• Aspiration • Pneumonia	• Residual gastric content as primary outcome • Noncomprehensive ICD-10-CM codes for aspiration
Study characteristics	• Retrospective cohort study • The time period of data collection included least 95% of the observation time to be before June 29, 2023	• Review articles • Case reports • No GLP-1RA nonusers control group

Abbreviations: GLP-1RA, glucagon-like peptide 1 receptor agonist; ICD-10-CM, International Classification of Diseases, Tenth Revision, Clinical Modification.

Data Extraction

Two reviewers (C.N.H., M.A.K.) independently assessed the quality of included studies. Using a standardized data extraction form, we extracted the following variables from each included study: study time frame, data source, type of analysis, post-procedure outcomes, type of procedure(s), anesthesia modality, and main conclusions. For each study group, we recorded the number of participants, the proportion of male participants, mean or median age with range, and the number of post-procedural adverse events.

Primary Data Analysis

To address potential confounding, we extracted adjusted risk estimates when available and noted the covariates included in each study's analysis model(s). All individual risk ratios were calculated using Microsoft Excel (Microsoft Corporation, Redmond, WA). All associated 95% CIs were computed using https://sample-size.net/risk-ratio/. Because of the heterogeneity in study designs, populations, and outcome measures, we grouped the studies into 2 categories for meta-analysis: the elective surgery group and the endoscopy group. Summary risk ratios (RRs) were computed using Stata version 18.5 (StataCorp, College Station, TX) employing a fixed-effects Mantel-Haenszel model.

Results

Literature Review

The initial PubMed search yielded 29 publications. One study was identified through a citation search, and 3 other studies were identified by an author (D.C.K.). After the title and abstract screening, 15 articles were selected for full-text review (17 were excluded). The full-text review resulted in 7 studies included for data extraction. A PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) diagram of the literature review process is shown in Fig. 1.

Figure 1.

Flow diagram for the systematic review of clinical studies presenting the risk of aspiration associated with GLP-1RAs users undergoing elective surgery/endoscopy. (Generated by Covidence and then modified). Abbreviations: GLP-1RAs, glucagon-like peptide-1 receptor agonists; ICD-10-CM, International Classification of Diseases, Tenth Revision, Clinical Modification.

Extracted Results

Our systematic review and meta-analysis included 7 studies examining aspiration risk in surgical and endoscopic procedures, with 541 408 patients (91 829 GLP-1RA users and 449 579 nonusers). Five studies in our review utilized data entirely collected prior to the ASA's initial recommendation [10, 12-14, 17]. Two studies in our review utilized data with at least 95% collected prior to the ASA's initial recommendation: (i) Alkabbani et al covered the period from January 2015 to August 2023; and (ii) Peng et al covered the period from April 2005 to February 2024. Regarding data collected after the ASA's initial guidance, Alkabbani et al included 3 out of 104 study months and Peng et al included 8 out of 226 study months. All other data in these 2 studies were collected prior to the publication of the ASA's initial guidance. We were interested in examining the clinical practice prior to the implementation of the new guidelines to provide a baseline for comparison in future studies that may examine the impact of the ASA's 2023 guidance.

Of the 7 included studies, 3 focused on elective surgical procedures, and 4 examined elective EGD (with 1 study also including colonoscopy performed in conjunction with EGD and another study also including lower endoscopy, as shown in Table 2). These 7 studies used large-scale medical and commercial databases, including Eversana Life Sciences, ICES, MarketScan, and TriNetX. With sample sizes ranging from 17 905 to 366 476 patients, these studies offer real-world evidence (RWE) on GLP-1RA use and associated risks of aspiration. Five studies (Klonoff et al, Welk et al, Alkabbani et al, Yeo et al, and Peng et al) used propensity matching. One study (Barlowe et al) used a Poisson regression model and another used multivariable logistic regression model (Chen et al) in their risk analysis to control for potential confounding factors. Table 2 presents the key characteristics and findings of these studies.

Table 2.

Seven studies reporting the risk of aspiration following elective surgery or elective EGD

Author (Reference number)	Month-year published	Time frame for data collection	Data source	Analysis method	Post-procedure outcome	N	Male	Age (mean)	Age range	Number of events	Events per 1000	N	Male	Age (mean)	Age range	Number of events	Events per 1000	Risk ratio [95% CI]	Type of procedure(s)	Type of anesthesia	Conclusion
Klonoff et al	Apr-24	Jan 2015-Apr 2023	Eversana Life Sciences	Propensity matched	Aspiration or pneumonitis within 72 hours	1296	477	58.7	18-89	7	5.40	1296	438	57.9	18-89	9	6.94	0.78 [0.29, 2.08]	Appendectomy, coronary artery bypass grafting, hernia repair, hip replacement, hysterectomy, prostatectomy, tonsillectomy, thyroidectomy, and spinal fusion	General endotracheal anesthesia	No increased risk of peri- and post-operative complications in GLP1-RA users undergoing surgery with general endotracheal anesthesia was identified.
Welk et al	May-24	Feb 2020-Mar 2023	Multiple linked administrative databases from Ontario, Canada (ICES)	Propensity matched	Pneumonia within 14 days	3833	2334	N/A	≥66	10	2.61	14 072	8524	N/A	≥66	49	3.48	0.75b [0.38, 1.48]	Aortic aneurysm repair, coronary artery bypass graft, aortic valve replacement, carotid endarterectomy, colorectal resection, lower limb revascularization, pancreaticoduodenectomy, partial liver resection, total hip replacement, Total knee replacement, radical prostatectomy, spinal decompression, craniotomy, total thyroidectomy, carpal tunnel release, minor urological/gynecologic procedure, urological endoscopic procedure, cholecystectomy, inguinal hernia repair, hysterectomy	General and spinal anesthetic	People with T2D who were using semaglutide at the time of elective surgery did not have an increased risk of pneumonia, of outcomes specific to aspiration, or of a need for bronchoscopy.
Chen et al	Mar-25	Apr 2020-Sep 2022	MarketScan	Multivariable Logistic Regression	Aspiration pneumonia within 30 days	5931	2429	N/A	≥18	42	7.08	360 545	157 257	N/A	≥18	2335	6.48	1.09b [0.81-1.48]	Bariatric surgical treatment, knee replacement, hip arthroplasty, colorectal resection, laminectomy, spinal fusion, coronary artery bypass graft, groin hernia repair, thyroidectomy, hysterectomy, lower extremity amputation, appendectomy, hip fracture surgery, tibial and ankle shaft fracture	General anesthesia and other types of anesthesia (not reported)	This cohort study found no significant association between the preoperative use of GLP-1 RAs and short-term postoperative aspiration pneumonia.
Barlowe et al	May-24	2005-2021	Truven Health Analytics MarketScan databases	Poisson regression	Aspiration, aspiration pneumonia, pneumonia, or respiratory failure within 14 days	15 119	6101	55a (49-60)	18-64	47	3.11	14 407	6959	57a (52-61)	18-64	54	3.75	0.83 [0.56, 1.23]	Upper endoscopy	Not reported	In patients with T2D prescribed GLP1-RA compared with patients prescribed DPP4i or opioids, no increased risk of pulmonary adverse events after EGD was observed. As an active comparator, the use of opioids is a class of medications that, similarly to GLP1-RAs, results in pharmacologic gastroparesis, yet has no peri-procedural management guidelines and provides important context.
Alkabbani et al	Oct-24	Jan 2016-Dec 2021 MarketScan Jan 2016-Aug 2023 Optum Clinformatics Data Mart	MarketScan Commercial Claims and Encounters, and Optum Clinformatics Data Mart	Propensity matched	Pulmonary aspiration within 48 hours	24 817	9023	59.9	≥18	103	4.15	18 537	6722	59.8	≥18	79	4.26	0.97 [0.73, 1.30]	Upper endoscopy	CPT codes 00731 and 00740 (upper gastrointestinal endoscopic procedures)	Compared with SGLT2 inhibitor use, GLP-1RA use was not associated with an increased risk of pulmonary aspiration, although it was associated with a higher risk for discontinuation of endoscopy.
Yeo et al	Jul-24	Jan 2018-Dec 2020	TriNetX	Propensity matched	Aspiration pneumonia within a month	15 144	5941	N/A	≤70	126	8.32	15 033	5904	N/A	≤70	94	6.25	1.33 [1.02, 1.74]	Upper and lower endoscopy	Not reported	The GLP-1RA user group compared to the nonuser group had a significantly higher incidence rate of aspiration pneumonia.
Peng et al	Nov-24	Apr 2005-Feb 2024	TriNetX	Propensity matched	Aspiration pneumonia within 7 days	25 689	10 005	58.5	≥18	123	4.79	25 689	9844	58.4	≥18	108	4.20	1.14 [0.88, 1.47]	Upper and lower endoscopy	General anesthesia	This population-based study found no increased risk of post-endoscopy aspiration pneumonia for users compared to nonusers of GLP-1RAs within 1 month prior to endoscopy.

Abbreviations: CPT, Current Procedural Terminology; DPP4i, dipeptidyl peptidase-4 inhibitors; EGD, esophagogastroduodenoscopy; GLP-1RA, glucagon-like peptide-1 receptor agonist; SGLT2, sodium-glucose cotransporter 2; T2D, type 2 diabetes.

^aMedian (interquartile range [IQR]).

^bCalculated risk ratio; authors reported the odds ratio.

Risk Ratios and Pooled Analysis

For elective surgery, 3 cohort studies (n = 386 973) found no increased risk of post-procedure aspiration or pneumonitis in GLP-1RA users. Klonoff et al reported a rate of aspiration or pneumonitis within 72 hours post-procedure of 5.40 per 1000 in GLP-1RA users vs 6.94 per 1000 in nonusers (RR 0.78, 95% CI [0.29-2.08]) [12]. Welk et al showed rates of pneumonia within 14 post-procedural days of 2.61 per 1000 in GLP-1RA users vs 3.48 per 1000 in nonusers (RR 0.75, 95% CI [0.38-1.48]) [13]. Chen et al reported a rate of aspiration or pneumonitis within 30 days post-procedure of 7.08 per 1000 in GLP-1RA users vs 6.48 per 1000 in nonusers (RR 1.09, 95% CI [0.81, 1.48]) [17]. The pooled analysis of these elective surgery studies yielded a combined risk ratio of 1.00 (95% CI [0.76, 1.30]) for aspiration events in GLP-1RA users compared to nonusers (Fig. 2). The combined risk ratio suggests no significant increase in aspiration risk associated with GLP-1RAs use in the surgical setting.

Figure 2.

(Upper panel) The combined risk ratios for 3 studies examining the risk of aspiration associated with GLP-1RA use compared to nonuse with elective surgery and (lower panel) 4 studies comparing the same 2 risks with EGD. Analyses used the fixed-effects Mantel-Haenszel model. Abbreviations: EGD, esophagogastroduodenoscopy; GLP1-RA, glucagon-like peptide-1 receptor agonist.

For EGD, 4 large cohort studies (n = 154 435) showed inconsistent results for the risk of post-procedure aspiration or pneumonitis in GLP-1RA users compared to nonusers. Barlowe et al found no significant increase in aspiration risk in the GLP-1RA user group vs the nonuser group within 14 days after elective EGD (3.11 vs 3.75 per 1000; RR 0.83, 95% CI [0.56-1.23]) [10]. Alkabbani et al reported no increased aspiration risk in the GLP-1RA user group vs the nonuser group within 48 hours post-EGD (4.15 vs 4.26 per 1000; RR 0.97, 95% CI [0.73-1.30]) [9]. Yeo et al observed a higher incidence of aspiration pneumonia in GLP-1RA users compared to nonusers within a month after an EGD procedure (8.32 vs 6.25 per 1000; RR 1.33, 95% CI [1.02-1.74]) [14]. Although the time frame specified to observe aspiration events after endoscopic procedures differs from Yeo et al, Peng et al observed a higher incidence of aspiration pneumonia in GLP-1RA users compared to nonusers within 7 days after the procedure (4.79 vs 4.20; RR 1.14, 95% CI [0.88, 1.47]). The pooled analysis of these 4 EGD studies resulted in a combined risk ratio of 1.10 (95% CI [0.95, 1.27]) for aspiration events in GLP-1RA users compared to nonusers (Fig. 2). The combined risk ratio indicates no statistically significant increase in aspiration risk for GLP-1RA users compared to nonusers undergoing endoscopic procedures.

Barlowe et al provided a sensitivity analysis by including opioids as an active comparator in their EGD study [10]. This inclusion is particularly relevant because opioids are known to delay gastric emptying and potentially increase aspiration risk [18]. Their analysis revealed that opioid users compared to nonusers had a significantly higher risk of aspiration (RR 2.68, 95% CI [1.89, 3.81]). The observed elevated risk ratio associated with opioid use demonstrates the Barlowe et al study's ability to detect increased aspiration risk associated with a medication that impairs gastric motility, lending credibility to its null findings for GLP-1RAs and risk of aspiration in patients undergoing EGD.

Discussion

Our systematic review and meta-analysis aimed to identify the effects of GLP-1RA use on the risk of aspiration and related pulmonary complications during elective surgery and EGD. Despite concerns about the delayed gastric emptying increasing the risk of post-procedure aspiration in GLP-1RA users vs nonusers [19], our analysis did not show that these medications significantly increase this risk in elective perioperative settings (RR 1.00, 95% CI [0.76, 1.30]) or peri-procedural settings (RR 1.08, 95% CI [0.91, 1.29]). Our results suggest that: (i) the initial guidance by the ASA intended to reduce pulmonary complications in GLP-1RA users during elective procedures may not have been warranted based on current evidence; and (ii) the revised ASA guidance is reasonable.

We did not include a study by Anazco et al because it did not compare GLP-1RA users to nonusers. This study reported an incidence of aspiration during elective EGD in GLP-1RA users of 0.67 per 1000 [20]. This was a lower incidence of complications on GLP-1RA than in the 4 EGD studies that we included (3.11, 4.15, 8.32, and 4.79 per 1000, respectively).

Do Nascimento et al conducted a systematic review and meta-analysis showing that GLP-1RA use was associated with an increased risk of pre-procedural gastrointestinal symptoms (odds ratio 7.66; 95% CI [3.42, 17.17]) and elevated residual gastric content (odds ratio 6.08; 95% CI [2.86, 12.94]) compared to controls [21]. They also found no significant difference in rates of perioperative hypoglycemia, hyperglycemia, or other complications. Additionally, several small cohort studies have reported an increased risk of residual gastric residue during EGD among adults using GLP-1RAs compared to those not using GLP-1RAs with risk ratios ranging from 2.92 to 15.14 [22-26]. These studies, however, either (i) did not report aspiration events as an outcome [24]; (ii) did not report any aspiration events in their cohort [22, 23]; or (iii) only reported a single aspiration event in their GLP-1RA group [25, 26].

The discrepancy between increased residual gastric content and lack of aspiration events warrants further investigation. While acute exposure to GLP-1 RAs during initiation or dose escalation is expected to delay gastric emptying, chronic use can lead to adaptive responses that attenuate this effect. This tachyphylaxis may partially explain why observational studies report minimal aspiration risk in patients on long-term GLP-1RA therapy. However, the heterogeneity in patient responses suggests that the risk of aspiration cannot be dismissed uniformly. It is possible that in the future, a heightened awareness of potential risks due to the ASA guidance will lead to more cautious perioperative management of patients on GLP-1RAs, potentially mitigating future aspiration risk through measures such as extended fasting periods, discontinuation in selected cases, use of gastric ultrasound to identify high-risk patients for pulmonary aspiration, or modified anesthetic techniques.

Limitations

Retrospective studies relying on administrative claims and electronic health record databases may not capture all instances of minor aspiration events or near-misses, and thus potentially underestimate the true risk. Future prospective randomized controlled trials (RCTs) could provide additional high-quality data, but these studies are expensive, time consuming, and possibly unethical. For identifying a rare side effect, RWE trials like the ones we analyzed are well established [27]. RWE trials can detect rare adverse events that may not occur frequently enough to be observed in smaller RCT populations. Furthermore, side effects might be identified in patient subgroups often excluded from RCTs, such as elderly patients or those with comorbidities [28]. Additionally, individual variations in gastric emptying, comorbidities, and concurrent medications may influence the risk of aspiration, but may not be fully captured in administrative claims and electronic health record databases. We are unaware of data about risks to glycemic control in the postoperative state and potential infections or other complications of hyperglycemia from halting GLP-1RA therapy one week prior to a surgical procedure.

Conclusion

Our study provides evidence that is inconsistent with the initial ASA guidance and which supports the revised ASA guidance regarding GLP-1RA use with surgical or endoscopic procedures. While our analysis suggests GLP-1RA use is not associated with a significantly increased risk of aspiration or related pulmonary complications in patients undergoing elective surgery or endoscopic procedures, clinicians should remain vigilant and consider individual patient factors when assessing aspiration risk. Further research is needed to definitively establish the safety profile of GLP-1RAs in the perioperative and peri-procedural settings and to refine future evidence-based guidance for their use.

Abbreviations

ASA

American Society of Anesthesiologists

EGD

esophagogastroduodenoscopy

GLP1-Ras

glucagon-like peptide-1 receptor agonists

ICD-10-CM

International Classification of Diseases

Tenth Revision

Clinical Modification

RCT

randomized controlled trial

RR

risk ratio

RWE

real-world evidence

Funding

None.

Disclosures

C.N.H. is a consultant for Liom. A.T.A. is a consultant for Liom. M.A.K. has nothing to disclose. G.E.U. is partly supported by research grants from the National Institutes of Health (NIH/NATS UL 3UL1TR002378-05S2) from the Clinical and Translational Science Award program, and from the National Institutes of Health and National Center for Research Resources (NIH/NIDDK 2P30DK111024-06). G.E.U. has received research support (to Emory University) from Bayer, Abbott, Corcept, and Dexcom, and has participated in Advisory Board for Dexcom and GlyCare. D.C.K. is a consultant for Afon, Embecta, Glucotrack, Lifecare, Novo, Samsung, SynchNeuro, and Thirdwayv.

Data Availability

Original data generated and analyzed during this study are included in this published article or in the data repositories listed in References.