GLP-1 receptor agonists in obstructive sleep apnea: A propensity score-matched real-world analysis

Glucagon-like peptide-1 receptor agonists (GLP-1As) confer cerebrovascular benefits in diabetes and obesity, but their long-term effects in obstructive sleep apnea (OSA), which is frequently associated with obesity and linked to poor cerebrovascular outcomes, remain unclear.

We conducted a retrospective cohort study using the TriNetX US Collaborative Network between January 1st 2016 to December 31st 2025. Adults with OSA were identified; exposure was initiation of a GLP-1A within 6 months before up to 1 month after OSA diagnosis. Propensity score matching was performed 1:1. Outcomes were assessed at 1, 3, and 5 years and included ischemic stroke, intracranial hemorrhage, emergency department visits, inpatient hospitalizations, and all-cause mortality. Kaplan-Meier analyses and Cox proportional hazards models were used. CPAP-restricted and tirzepatide-specific subgroup analyses were also performed.

After matching, 438,844 patients were included in each cohort. Across 1-, 3-, and 5-year follow-up, GLP-1A exposure was associated with lower hazards of ischemic stroke (HRs, 0.75, 0.83, and 0.87), intracranial hemorrhage (HRs, 0.44, 0.56, and 0.61), emergency department visits (HRs, 0.77, 0.86, and 0.87), inpatient hospitalizations (HRs, 0.59, 0.67, and 0.69), and all-cause mortality (HRs, 0.38, 0.49, and 0.54; all p < 0.001). Associations were directionally consistent in CPAP-restricted and tirzepatide-specific subgroup analyses.

In this large real-world cohort of patients with OSA, GLP-1A exposure was associated with lower hazards of cerebrovascular events, healthcare utilization, and all-cause mortality across 1-, 3-, and 5-year follow-up. These findings are hypothesis-generating and require validation in prospective studies evaluating incretin-based therapies as potential adjuncts to standard OSA management.