A cohort study of glucagon-like peptide-1 receptor agonists (GLP-1RAs) uncovered a strong association with chronic cough. A review of 427,555 individuals prescribed a GLP-1RA for diabetes had a 12% increased risk of a new chronic cough compared with 1.6 million individuals prescribed a different second-line diabetes agent. The GLP-1RA-associated risk jumped to 29% for individuals who do not have a diagnosis of gastroesophageal reflux disease (GERD).1
“The essential finding of an association between GLP-1RAs and chronic cough didn’t surprise me because we identified GERD as a likely mechanism, which is a very well-known side effect of GLP-1s,” said Tyler J. Gallagher, MD, MPH, resident physician in otolaryngology at the Keck School of Medicine at the University of Southern California. “But we were surprised that when we looked at a cohort where we entirely excluded a GERD diagnosis, there was still a strong association. This suggests that additional mechanisms beyond reflux may be contributing to chronic cough.”
GLP-1RAs have known associations with GERD and direct effects on the vagus nerve, Dr. Gallagher said. The adjusted hazard ratio (aHR) for chronic cough for a GLP-1RA compared with any second-line non-GLP-1 RA medication was 1.12 (95% CI; 1.08-1.16). Compared with dipeptidyl peptidase-4 (DPP-4) inhibitors, the aHR was 1.18 (95% CI; 1.11-1.26); and with any sulfonylurea, the aHR was 1.25 (95% CI; 1.18-1.32). However, compared with sodium-glucose co-transporter 2 (SGLT2) inhibitors, the aHR was 1.03 (95% CI; 0.98-1.09).
After excluding patients with a previous GERD diagnosis, the aHR for GLP-1RAs compared with any non-GLP-1RA medication jumped to 1.29 (95% CI; 1.17-1.42). In that cohort, the aHR for DPP-4 inhibitors was 1.36 (95% CI; 1.17-1.58); sulfonylureas was 1.25 (95% CI; 1.09-1.42); and SGLT2 inhibitors was 1.14 (95% CI; 1.02-1.28).
The mechanisms driving the association are unclear, Dr. Gallagher said. In addition to GERD-associated cough, individuals may have silent or laryngopharyngeal reflux (LPR), or airway reflux with no obvious symptoms of GERD. Vagus nerve stimulation may also cause neurogenic cough, and there may also be other mechanisms in play, he said.
Dr. Gallagher added that up to two-thirds of individuals with chronic cough show signs of gastrointestinal dysmotility. Proton pump inhibitors and other approaches focused on gastric acid suppression can provide relief from GERD but often have little effect on LPR. Because chronic cough associated with GLP-1RA use may be independent of stomach acid reflux with GERD, acid suppression alone may not provide relief in these individuals.
More investigation is clearly needed, Dr. Gallaher said. The research group is in the early stages of building a prospective study comparing individuals who are starting GLP-1RAs with those starting other medications. More granular patient-level data would allow more complete assessments of existing reflux disease, COPD, and other covariates, he said.
On the clinical side, the group may explore symptom severity and patient-reported outcomes, which could help determine what patients may need to consider alternatives to GLP-1RAs. There may also be dose-related effects and other modifiable factors that play into patient preferences in considering changing or stopping medication.
“For physicians, it is important to be consider that GLP-1RAs may be another potential contributor to chronic cough, similar to [angiotensin-converting enzyme] inhibitors,” Dr. Gallagher said. “This is an important consideration to be aware of for people who come in with chronic cough. Many patients derive substantial benefit from GLP-1RAs, but if they are bothered by cough symptoms, slowly weaning them off the GLP-1 is worth considering. Ultimately, this represents a shared cost-benefit decision that patients and clinicians should be comfortable exploring.”
References
1. Gallagher TJ, Razura DE, Li A, et al. Glucagon-like peptide-1 receptor agonists and chronic cough. JAMA Otolaryngol Head Neck Surg. doi:10.1001/jamaoto.2025.4181
