Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have emerged as valuable therapeutic options for managing type 2 diabetes mellitus because they allow glycemic control without increasing the potential for hypoglycemia.  Also, they provide significant cardiovascular and renal benefits as well as lead to weight loss.  In addition to diabetes mellitus, they are being used extensively for managing obesity.  There is emerging evidence suggesting potential benefits of GLP-1RAs in the management of metabolic liver disease, obstructive sleep apnea, peripheral vascular disease, depression, addiction, and neurodegenerative disorders (e.g., Parkinson’s and Alzheimer’s disease).  With expanding indications, the clinical use of these drugs is expected to grow exponentially. Because GLP-1RAs can delay gastric emptying, they can potentially increase residual gastric volume despite recommended fasting duration.  This might increase the propensity for regurgitation and aspiration of gastric content during sedation and general anesthesia. 

In this issue of Anesthesiology, Oprea et al.  present a comprehensive review on preoperative care of patients receiving GLP-1 RAs. They endorse the concerns of delayed gastric emptying and the potential for regurgitation and aspiration of gastric content. The authors discuss the risks and benefits of preprocedure continuation or discontinuation of GLP-1RAs. They emphasize that medications should be withheld only when the perceived risks outweigh the benefits, and there are no alternatives to mitigate the risks. The authors appropriately argue that there are several benefits of continuing GLP-1RAs. Also, withholding GLP-1RA therapy for a longer duration (i.e., four to five half-lives) can impair glycemic control and lead to weight gain. Furthermore, if discontinued, a gradual restart may be necessary, which might lead to poor postoperative glycemic control and associated harm. Therefore, it is prudent to continue the GLP-1RAs and attempt to mitigate the aspiration risks through alternate approaches. This is in contrast to the American Society of Anesthesiologists (ASA; Schaumburg, Illinois) guidance recommending withholding the drugs prior to sedation or general anesthesia.  The ASA task force had an unenviable challenge of developing guidance to mitigate aspiration risk despite limited evidence. Of note, subsequent to the ASA guidance, several professional societies and reviews have recommended against withholding these drugs.

Oprea et al.  also present evidence suggesting that prolonging the fasting duration, particularly for solids, should reduce preprocedure residual gastric content. Although the available evidence is heterogenous and many studies assessing delayed gastric emptying are performed in nonsurgical settings, the recommendation for liquid diet for 24 h appears reasonable. However, the evidence for recommending high carbohydrate clear liquids for 8 h and no or low carbohydrate clear liquids for 4 h seems arbitrary as the evidence for this is questionable. Also, administration of preoperative carbohydrate load is not recommended in patients with diabetes mellitus, as it can cause hyperglycemia.  Management of perioperative hyperglycemia can be challenging, particularly in insulin-naïve patients which can lead to procedure delays and/or cancellation or unplanned hospital admission after ambulatory surgery. Furthermore, the authors’ recommendations for managing patients who are poorly tolerating GLP-1RAs scheduled for “time sensitive” procedures, are not pragmatic, as there may not be enough time for dietary modifications or fasting for 24 h or more.

Overall, the optimal approach to preprocedure management of patient receiving GLP-1 RA still remains controversial. So, what should a clinician do?

Given the complexities related to delayed gastric emptying and risks of regurgitation/aspiration associated with the use of GLP-1RAs, the strategies to ameliorate these risks should include a multipronged approach. It is critical to avoid scheduling elective surgery when the risks of delayed gastric emptying are high such as during the dose-escalation phase and in presence of significant gastrointestinal adverse effects. Although absence of gastrointestinal symptoms may not be a reliable indicator of abnormal gastric emptying, the presence of symptoms suggests the possibility of delayed gastric emptying.

Upon scheduling the procedure, it is advisable not to withhold the drugs prior to the procedure. However, it may be reasonable to suspend the treatment temporarily if surgery is contemplated during the early dose-escalation phase. Additional precautions would include prolonging the fasting duration for solids (e.g., liquid diet for 24 h) and dietary modifications particularly in patients at risk of high residual gastric content (e.g., drugs with longer duration of action, use of higher drug doses, presence of gastrointestinal symptoms, and comorbidities causing gastroparesis). On the day of the procedure, it is necessary to balance the available information with therapeutic and clinical options. It is prudent to perform point of care gastric ultrasound, if possible, to stratify patient risk. When in doubt, it is reasonable to consider the patient as having a “full stomach” and proceed accordingly.

There is an urgent need for well-conducted, adequately powered, prospective clinical trials to address the substantial gaps in knowledge surrounding the prevalence and duration of delayed gastric emptying in different populations (type 2 diabetes vs. obesity) and with different types and doses of GLP-1RAs. Also, it is necessary to assess the relation between the amount of gastric residual volume and content and the risk of clinically significant aspiration. Most studies do not consider the risk factors for delayed gastric emptying such as dose-escalation phase, the dose and dosing schedule, the timing of drug administration, the duration of preprocedure fasting, and other factors that may delay gastric emptying.

It is necessary to study the duration of fasting and type of diet to reduce the risk of retained gastric content. For example, a recent study in healthy volunteers reported that high-calorie fruit juice (e.g., mango juice) exhibited delayed emptying compared with a low-calorie fruit juice (e.g., blackcurrant juice).  Another research question includes evaluation of the effectiveness of prokinetics (e.g., erythromycin) in accelerating gastric emptying and the time required for this treatment to achieve gastric emptying. It is suggested that the variability in effectiveness and adverse effects of GLP-1RAs might be due to differences in pharmacogenomics thus understanding these factors should enable a more personalized approach to management.

Although there is significant interest in understanding the incidence of aspiration in patients on GLP-1RA, the approach used by most investigators is flawed and misleading. Most retrospective studies have assessed risk of aspiration using International Classification of Disease Tenth Revision, and Clinical Modification codes for aspiration as well as codes for respiratory or pulmonary complications.  Obviously, using pulmonary complications as a surrogate to assess the incidence of aspiration is inappropriate because there are numerous causes of postoperative pneumonia such as patient characteristics, surgery-related causes including postoperative care, and anesthesia-related causes (e.g., residual paralysis). Also, given the low incidence of aspiration, almost all studies are underpowered.

In summary, patients on GLP-1RA appear to have a higher risk of aspiration of gastric content. Although there are several published recommendations, they have significant limitations because the data to provide evidence-based clinical guidelines is sparse and of low quality. Nevertheless, elective procedures should be avoided during the dose-escalation phase and in patients with significant gastrointestinal symptoms. Also, there is consensus that preoperative withholding of GLP-1RAs is inadvisable, irrespective of the indication for their use. Furthermore, it seems appropriate to consume liquid diet for at least 24 h in patients at high risk of gastroparesis. There should be low threshold to perform gastric ultrasound and rapid sequence induction of general anesthesia, if appropriate. The available recommendations should be adapted to individual patients and facilities with the primary aim of maintaining patient safety. Finally, periprocedure management of patients receiving GLP-1RAs should be based on shared decision-making of the anesthesiologist, the proceduralist or surgeon, and the patient. Therefore, a multisociety consensus for the management of patients on GLP-1RA has recently been published15  and is described in detail in a letter published in this issue of Anesthesiology.