Anesthetic Considerations in Adult Patients on Glucagon-Like Peptide-1 Receptor Agonists: Gastrointestinal Focus

New classes of antihyperglycemic agents continue to be introduced for the management of diabetes mellitus (DM).1 Endogenous glucagon-like peptide-1 (GLP-1) is an incretin hormone that stimulates insulin production and secretion from pancreatic β cells and reduces glucagon secretion from α cells.2 In addition, GLP-1 also reduces appetite and slows gastric emptying. The GLP-1 receptor agonist (GLP-1RA) has been shown to improve glycemic control in patients with DM and to reduce weight.1,2 Furthermore, GLP-1RAs have cardiac and renal protective properties, which are often beneficial in this patient population.3 Therefore, GLP-1RAs are increasingly being used for the treatment of type 2 DM and obesity. It is incumbent on anesthesiologists to be well-versed in the pharmacology and perioperative considerations of GLP-1RAs (Table).4 Table. - Summary of Glucagon-Like Peptide-1 Receptor Agonists Generic drug Brand name Administration/frequency Elimination half-life (t 1/2) Dulaglutide Trulicity SQ injection × once weekly ≈5 d Exenatide (ER) Bydureon, BCise SQ injection × once weekly ≈2.4 h Exenatide (IR) Byetta SQ injection × twice daily ≈2.4 h Liraglutidea Saxenda, Victoza SQ injection × once daily ≈13 h Lixisenatideb Adlyxin SQ injection × once daily 2–4 h Semaglutide Ozempic, Wegovy SQ injection × once weekly ≈7 d Semaglutide Rybelsus Oral, once daily ≈7 d Tirzepatide Mounjaro SQ injection × once weekly ≈5 d Abbreviations: ER, extended release; IR, immediate release; SQ, subcutaneous.aXultophy is a combination product containing liraglutide and insulin degludec (ultralong-acting basal insulin analog).bSoliqua is a combination product containing lixisenatide and glargine. GLP-1RAs are associated with adverse gastrointestinal (GI) effects such as nausea, vomiting, diarrhea, constipation, dyspepsia, abdominal distension, and abdominal discomfort/pain.1,2 GLP-1RAs can also delay gastric emptying, which can be further exacerbated by the gastroparesis associated with advanced DM.5 In September 2023, the US Food and Drug Administration updated the warning label for Ozempic (semaglutide) to include ileus along with the current list of side effects (ie, nausea, vomiting, diarrhea, and constipation).6 Delayed gastric emptying as well as ileus can increase the residual gastric volume (RGV),7–11 even if current preoperative fasting recommendations are followed.12,13 This can increase the risk of regurgitation and pulmonary aspiration of gastric contents during sedation or general anesthesia. In fact, several anecdotal reports of regurgitation and aspiration have been recently published.14–16 Given this patient safety concern the American Society of Anesthesiologists (ASA) recently provided consensus-based guidance on preoperative management of patients on a GLP-1RA.17 The aims of this commentary are (1) to review the available literature on GLP-1RAs, including the studies evaluating RGV in fasting patients, (2) to address the several associated clinical controversies, and (3) to identify research needed to improve patient care and safety of this population. GLP-1RA-INDUCED DELAYED GASTRIC EMPTYING A review of the literature suggests that gastric emptying is delayed with the initiation of GLP-1RA therapy; however, there is variability in the methodology used to measure gastric emptying.18–23 This may explain the conflicting reports on delayed gastric emptying for these medications. Most studies have used the acetaminophen absorption test, which measures gastric emptying of liquids and is not reliable. Scintigraphy is considered the “gold standard,” as it measures gastric emptying for both solids and liquids.2 Moreover, all gastric emptying tests have considerable interindividual variation, although this can be counteracted by intrasubject measurement comparisons. Another confounding factor is that glycemic control can influence gastric emptying. While it is usually normal or modestly accelerated with well-controlled DM, inadequate glycemic control may delay gastric emptying.2,5 Furthermore, although long-acting GLP-1RAs slow gastric emptying, it is less pronounced than short-acting GLP-1RAs.20,21 In addition, the effect on gastric emptying depends on the dose and duration of medication use. Delay in gastric emptying is attenuated with dose escalation and with duration, which suggests possible tachyphylaxis. Delayed emptying occurs within 12 weeks of semaglutide use and then tends to subside or resolve after 20 weeks.22 Accordingly, adverse GI effects attributed to delayed gastric emptying tend to peak at around 12 weeks, and subsequently subside. Overall, several factors influence GLP-1RA-related delayed gastric emptying. These include the type of drug (ie, short-acting versus long-acting), the drug dose and duration, and glycemic control (good versus poor control). Despite these variables, there remains a concern for delayed gastric emptying and possibility of increased preoperative RGV. After 12 weeks of therapy, patients receiving semaglutide retained 37% of solid meal after 4 hours, as compared with no gastric retention in the placebo group.23 GLP-1RA-RELATED FASTING RESIDUAL GASTRIC VOLUME Several studies have assessed RGV in fasting patients undergoing esophagogastroduodenoscopy (EGD). A retrospective cohort study of patients taking GLP-1RA (n = 59) with matched controls (n = 118) reported RGV in 6.8% (4 of 59) patients taking GLP-1RA versus 1.7% (2 of 118) in the control group.7 Another case-control study observed an overall 1.5% (17 of 1128) incidence of RGV in unmatched patients undergoing EGD.8 With propensity score matching (n = 205 pairs) these authors reported that even after ≥12 hours of fasting, RGV was 5.4% in patients taking a GLP-1RA compared with 0.5% controls. A retrospective study of patients undergoing EGD observed that 6.7% (27 of 886) had RGV.9 For patients taking semaglutide for weight loss, 24.2% (8 of 33) had RGV, while 5.1% (19 of 371) in the control group had RGV. The mean duration of withholding semaglutide was 10 days, which was similar with or without RGV. There was also no relationship between semaglutide use and the amount/volume of RGV. However, adverse GI symptoms were associated with increased RGV. In contrast, EGD combined with colonoscopy had a lower incidence of RGV.9 In this series, a patient with previous gastric bypass receiving semaglutide experienced pulmonary aspiration even after fasting for 12 hours. Of note, a large retrospective study (n = 85,116) observed that only 3% of EGD patients had RGV.7 In this study, the use of a GLP-1RA (6/85,116 patients) was not associated with increased residual gastric content.10 These authors concluded that RGV encountered during EGD should only be used as a surrogate for delayed gastric emptying. Sherwin et al11 performed gastric ultrasound to evaluate the presence of solid residual content after an 8-hour fast in volunteers (n = 10) on short-term course of semaglutide compared with controls (n = 10). Participants underwent a second ultrasound evaluation 2 hours after drinking 12 ounces (approximately 350 mL) of water. In the supine position, 70% of semaglutide participants and 10% of control participants had solids present on gastric ultrasound. In the lateral position, 90% of semaglutide participants and 20% of control participants had solids. The authors concluded that GLP-1RAs increase RGV after an overnight fast and 2 hours after clear liquid ingestion.11 Overall, studies evaluating RGV during EGD have several limitations including combining patients on short-acting and long-acting GLP-1RA, variable indication for use of these medication, variable dose and duration of therapy, and variable risk factors for delayed gastric emptying. Importantly, RGV was defined as presence of any solids in the stomach or >0.8 mL/kg of fluid content as measured from the aspiration/suction canister, but its clinical relevance regarding regurgitation and pulmonary aspiration is unknown. Despite these limitations, the study findings corroborate those reporting delayed gastric emptying with GLP-1RA therapy. PREPROCEDURE WITHHOLDING of GLP-1RAs Because GLP-1 only functions during hyperglycemia, the risk for hypoglycemia while on a GLP-1RA during fasting is extremely low. Therefore, guidelines on perioperative management of patients with DM have recommended to either continue the daily dose of these drugs or to withhold it the day before or day of the procedure.1 For patients on weekly dosing, it is recommended to hold the dose for 7 days. However, these recommendations are aimed at maintaining perioperative glycemic control and do not address the concerns about delayed gastric emptying and the potential for increased regurgitation and aspiration. The ASA task force recommends that patients hold their daily dose of GLP-1RA on the day of the procedure and their weekly dose 7 days before the procedure.17 However, Jones et al24 suggest holding GLP-1RA administered for weight loss for at least 3 half-lives, which represents ≈88% drug clearance. With an elimination half-life of a long-acting GLP-1RA (eg, semaglutide) being 7 days, this would be approximately 3 weeks. For patients with DM, these authors recommend consulting an endocrinologist to assess the risk/benefits of holding these drugs for at least 3 half-lives. Given that it takes 4–5 half-lives for a drug to be eliminated from the body, the suggestion to hold the drugs for 3 half-lives appears arbitrary. In fact, there are no good data to suggest a correlation between half-life of a GLP-1RA and gastric emptying. Half-life is a complex and complicated concept; therefore, it is often challenging to incorporate it into clinical practice and to use it as a tool for clinical decision-making. Predicting drug elimination based on half-life can go awry. Of note, if GLP-1RAs prescribed for DM are held for longer than the dosing schedule, consider consulting an endocrinologist for bridging the antidiabetic therapy. However, it has been argued that preoperatively withholding GLP-1RA in patients with DM may be detrimental as it can result in hyperglycemia and associated adverse events, or it can necessitate bridging therapy with supplemental insulin, which can increase the likelihood of hypoglycemia.25 The ASA task force recommendations do not differentiate between patients taking a GLP-1RA for DM versus for weight loss.17 In contrast, others have suggested that the approach to preoperative care should vary based on the indication of GLP-1RA use (DM versus weight loss), dose (low versus high) and dosing schedule (daily versus weekly), or the type of procedure/surgery (GI procedure versus others). However, evidence to guide drug withholding based on these variables is lacking. PREPROCEDURE FASTING INTERVAL FOR PATIENTS ON A GLP-1RA The concern about pulmonary aspiration has led to the notion that the longer the fast the safer it is for the patient. Given that GLP-1RAs delay gastric emptying and increase RGV even after conventional fasting times, some have suggested prolonged fasting for solids and administration of an interim liquid diet, as well as prolonged immediate preoperative fasting for clear liquids (ie, longer than 2 hours). However, prolonged fasting may not reliably result in an empty stomach. Instead, intake of clear liquids may paradoxically reduce RGV. Furthermore, even if regurgitation and aspiration of clear fluid occurs, it is unlikely to result in significant morbidity.12,13 Importantly, longer fasting times are associated with significant adverse physiological derangement and adverse patient-reported outcomes such as discomfort, anxiety, thirst, hunger, nausea, and omission of medications.26 Therefore, further prolongation of fasting duration appears unreasonable, particularly given the lack of evidence to suggest a safe fasting duration. DAY OF PROCEDURE MANAGEMENT OF PATIENTS ON A GLP-1RA Patients arriving for elective procedure with their GLP-1RA withheld as recommended, and with no GI symptoms (eg, severe nausea/vomiting/retching, abdominal bloating, or abdominal pain) may proceed with their procedure (Figure). However, if the GLP-1RAs are not held as advised and/or if the patient has significant GI symptoms, consider evaluating RGV using point-of-care gastric ultrasound. Gastric ultrasound has been validated to quantify the volume of solids and liquid contents in the stomach. It could help risk-stratify patients and assist decision-making. However, it can be technically challenging, and may be inconclusive in some cases. In addition, there is the potential for both false positive and negative results. In a retrospective cohort study (n = 512), 6% of patients had solid content or >1.5 ml/kg of clear fluid despite no risk factors for delayed gastric emptying, but the clinical implications of these findings are uncertain.27Figure.: Management of a patient on GLP-1 RA on the day of the procedure. GI indicates gastrointestinal; GLP-1, glucagon-like peptide-1; RA, receptor agonist; RSI, rapid sequence induction.It is recommended that if the stomach is empty, one can proceed as usual. However, if the stomach is full or if gastric ultrasound is inconclusive or not possible, consider delaying the procedure or treat the patient as “full stomach” and manage accordingly. The ASA difficult airway guidelines recommend that patients at high risk of aspiration and with a documented or suspected difficult airway receive awake tracheal intubation.28 Some argue that withholding a GLP-1RA and/or the absence of GI symptoms may not be a reliable indication of the absence of risk, and therefore it may be safer treating patients on a GLP-1RA as “full stomach.” OTHER MEASURES TO MITIGATE THE RISKS OF REGURGITATION AND ASPIRATION Medications that promote gastric emptying and increase lower esophageal sphincter tone (eg, metoclopramide), reduce gastric volume (eg, proton pump inhibitors29), or increase gastric pH (eg, proton pump inhibitors and H2 receptor antagonists) have been used to prevent adverse outcomes with pulmonary aspiration of gastric content. However, their role in influencing clinical outcomes remains controversial. There also are concerns of associated adverse effects such as extrapyramidal symptoms, dizziness, and sedation. Another intervention that has been advocated but remains controversial is the placement of nasogastric tube for aspiration of gastric contents.24,30 Its placement in an awake state can be markedly uncomfortable and may not reliably empty the stomach. However, this could be considered during the procedure before tracheal extubation. One of the factors commonly overlooked is the possibility of regurgitation and aspiration after tracheal extubation. Therefore, the chosen general anesthetic technique should allow for rapid recovery of baseline mental status.31,32 An extubation plan should be formulated before anesthesia and consider patient-related, anesthetic, surgical, and human factors.33 The patient’s clinical condition should be optimized for extubation. Awake extubation is the standard technique. CONCLUSIONS The evidence to provide guidance for preoperative management of GLP-1RAs to prevent regurgitation and aspiration of gastric contents is sparse and of limited quality (case reports and observational trials). Nevertheless, it is necessary to recognize the potential risk of aspiration. Strategies to ameliorate the risk of regurgitation/aspiration in patients taking a GLP-1RA include when feasible, holding the medication before scheduled procedures. Caution is advised, particularly during the initial 12 to 20 weeks of GLP-1RA therapy, which may be a critical interval for delayed gastric emptying. In addition, if possible, consider preoperative gastric ultrasound to screen patients for RGV. There should be a low threshold for considering rapid sequence induction to secure the airway, although its role in reducing the risk of pulmonary aspiration during induction of general anesthesia remains controversial. Shared decision-making principles dictate that the potential risks and benefits of each option should be openly and transparently discussed with the patient and the proceduralist/surgeon. High-quality, adequately powered studies are urgently needed to investigate the safety of GLP-1RAs as it relates to peri-anesthetic management. This can be challenging given the extremely low risk of pulmonary aspiration in the elective setting, significant variabilities in the effects of GLP-1RA on gastric emptying, lack of relationship between gastric emptying and duration of preoperative fasting on RGV, and lack of relationship between RGV and regurgitation subsequent pulmonary aspiration. DISCLOSURES Name: Girish P. Joshi, MBBS, MD, FFARCSI. Contribution: G. Joshi wrote and revised the manuscript, and approved the final version. This manuscript was handled by: Thomas R. Vetter, MD, MPH, MFA.