Impact of GLP-1 receptor agonists on risk of stroke in patients with atrial fibrillation: a Danish nationwide cohort study

Introduction Atrial fibrillation (AF) is associated with markedly increased risk of stroke and remains a major contributor to cardiovascular morbidity and mortality. Glucagon-like-peptide-1 receptor agonists (GLP1) known for managing diabetes (DM) and reducing cardiovascular risk, have recently shown promising results in promoting weight loss, improving cardiometabolic profiles, and improving heart failure symptoms. However, their potential to address AF-related complications, such as stroke remains largely unexplored. Purpose The objective of this study was to examine the association between GLP1-use and the risk stroke in patients with AF in a nationwide population. Methods We conducted a nationwide retrospective cohort study including all patients ≥18 years diagnosed with AF (ICD-10 I48) between January 1, 2010 and January 1, 2023, utilizing Danish nationwide registries. Patients were required to be alive 90 days after AF discharge. Diabetes was defined by redeemed prescription for glucose-lowering drugs (ATC A10). GLP1 users were defined as a redeemed prescription within 1 year before to 90 days after AF discharge. Follow-up started 91 days after AF discharge. The primary outcome was ischemic stroke, and secondary outcome was all- cause mortality. Hazard ratio (HR) was estimated using Cox proportional hazard models adjusted for age, sex and comorbidities. Propensity score matching (PSM) was performed using nearest-neighbor matching. Results A total of 41,469 patients with AF and diabetes were included; 4,849 (11.7%) were GLP-1 users and 36,620 (88.3%) were non GLP1-users. Mean age was 62.2±10.2 years for GLP-1 users and 68.8±9.7 years for non GLP1-users. GLP-1 users were associated with a lower risk of stroke in the unadjusted analysis (HR 0.65, 95% CI 0.56–0.75) and remained significantly associated with reduced stroke risk after adjustment for age, sex and CHA2DS2-VA-Score (HR 0.69, 95% CI 0.57–0.81). Cumulative incidence curves demonstrated early and sustained separation, with consistently lower absolute stroke risk among GLP1 users (Figure 1). 1-year mortality was 7.9% (95% CI 7.1-8.7) vs. 13.2% (95% CI 12.8-13.5), and 3-year mortality 18.9% (95% CI 17.6-20.3) vs. 30.6 (95% 30.1-31.1) in GLP1-users and non GLP1-user, respectively. GLP1 users were independently associated with lower all-cause mortality (HR 0.86, 95% CI 0.81-0.91). In the PSM cohort (n=9,698), GLP-1 users were associated with a lower risk of stroke (HR 0.72, 95% CI 0.60–0.86) and lower all-cause mortality (HR 0.86 95% CI 0.80-0.93). Conclusion In this nationwide cohort of patients with AF and DM, GLP1 users were associated with a significantly lower risk of stroke and all-cause mortality compared to non GLP1 users. The association remained consistent after multivariable adjustment and in PSM-analyses.