Lim, Ghee Kheng

ABSTRACT Background Sodium‐glucose cotransporter 2 inhibitors (SGLT2‐Is) are primarily used to manage type 2 diabetes mellitus (T2DM) and heart failure (HF), but their impact on reducing stroke risk in patients with atrial fibrillation (AF) remains underexplored. We aimed to investigate the impact of SGLT2‐Is for stroke risk mitigation in patients with AF. Methods Using the TriNetX database, we performed a retrospective cohort study on patients ≥ 18 years with AF and on anticoagulation. These patients were then classified into two cohorts based on their use of SGLT2‐Is. To balance the baseline demographics, comorbidities, and medication use, propensity score matching (PSM) was used. The primary outcome was ischemic and hemorrhagic strokes, and the secondary outcomes were all‐cause mortality, all‐cause hospitalizations, need for AF cardioversion or antiarrhythmic drug initiation, and cardiac arrest. Adjusted odds ratio (aORs) and hazard ratios (HRs) were estimated for both the primary and secondary outcomes. Results A total of 152,778 patients with 76,389 patients were included in each cohort (SGLT2‐Is users, and non‐users) after PSM. The mean age of both SGLT2‐Is users and non‐users was 72.8 years. For the SGLT2‐Is users, the mean follow‐up period was 317 days, whereas the mean follow‐up period for the non‐users was 306 days. For the primary outcome, we found that SGLT2‐Is users had a lower risk of ischemic stroke (aOR: 0.889; 95% CI: 0.857–0.923) and hemorrhagic stroke (aOR: 0.682; 95% CI: 0.620–0.749) compared to non‐users. Regarding the secondary outcomes, SGLT2‐Is users also had lower risk for all‐cause mortality (aOR: 0.615; 95% CI: 0.595–0.636), all‐cause hospitalizations (aOR: 0.599; 95% CI: 0.587–0.611), need of AF cardioversion (aOR: 0.846; 95% CI: 0.813–0.881), antiarrhythmic drug initiation (aOR: 0.790; 95% CI: 0.766–0.815), and cardiac arrest (aOR: 0.643; 95% CI: 0.601–0.688). Conclusions SGLT2‐I use in patients with AF is associated with lower risks of stroke and other cardiovascular and non‐cardiovascular outcomes. Future prospective research is needed to validate these findings.