Symptom evaluation cannot exclude retained gastric content in patients taking glucagon‐like peptide‐1 receptor agonists: a reply

We thank Zavos for their comments [1] on our study of gastric ultrasound in patients receiving semaglutide [2], and we agree that little information is obtained from either the presence or the absence of gastrointestinal symptoms. We have concerns, however, about the false assumption of an empty stomach in patients receiving glucagon-like peptide-1 receptor agonist (GLP-1RA) therapy based on the absence of symptoms or of other risk factors, such as recent initiation or escalation of therapy. As Zavos notes, only four patients receiving GLP-1RA therapy in our study reported pre-operative upper gastrointestinal symptoms. Three of these had a positive gastric ultrasound, in each case based on solid residual gastric content, while one had a negative scan. The presence of gastrointestinal symptoms would therefore have identified only three out of the 21 patients receiving GLP-1RA therapy who had pre-operative residual gastric content. Of the three patients with solid residual gastric content, one had a low gastric volume (45 ml), one a high volume (277 ml) and one a volume that could not be measured because of the acoustic shadow of air bubbles (‘frosted glass’). Consequently, and contrary to what Zavos suggested, there does not appear to be a reliable correlation between gastric volume and upper gastrointestinal symptoms. Although we agree that evidence of pre-operative residual gastric content does not translate automatically into a risk of pulmonary aspiration, we would caution against over-interpreting the absence of an aspiration signal in retrospective studies, such as the meta-analysis by Elkin et al. [3]. Peri-operative measures including regional anaesthesia, gastric ultrasound and tailored airway management may have lowered the likelihood of aspiration, effectively ‘masking’ the underlying risk. Given the extensive reporting of pulmonary aspiration associated with GLP-1RA use, such an effect seems plausible. In the endoscopy literature there is also a signal of increased pulmonary aspiration in patients receiving GLP-1RA therapy (OR 2.29, 95%CI 1.36–3.87), as reported by Tan et al. [4]. In conclusion, until there is compelling evidence that residual gastric content in patients receiving GLP-1RA therapy carries no risk of peri-operative pulmonary aspiration, identifying and addressing residual gastric content before surgery wherever feasible appears to be a sensible approach. As we have shown, the absence of gastrointestinal symptoms is not a reliable indicator of gastric content, and we agree in principle with the alternative wording proposed by Zavos to emphasise this in the relevant guidelines [5].