Abstract Number: Esoc2026a1282 Effect of Sodium–glucose Cotransporter 2 Inhibitors on Functional Outcomes After Endovascular Thrombectomy in Diabetic Stroke

Background and aims Although preclinical studies and limited clinical evidence suggest potential benefits of sodium-glucose cotransporter 2 inhibitors (SGLT2i) in acute ischemic stroke (AIS), their effects in patients with moderate-to-severe AIS requiring endovascular thrombectomy (EVT) remain unclear. This study aims to evaluate whether SGLT2i use is associated with improved 3-month functional outcomes in diabetic patients with AIS undergoing EVT. Methods We analyzed prospectively collected data from four comprehensive stroke centers in Korea, Italy, and Switzerland, including diabetic patients with AIS who underwent EVT. To address baseline imbalance between SGLT2i users and non-users, three propensity score-based approaches were applied: overlap weighting (OW), inverse probability of treatment weighting (IPTW), and 1:1 propensity score matching (PSM). Outcomes included 3-month modified Rankin Scale (mRS) shift, favorable (mRS 0-2), and excellent outcomes (mRS 0-1) at 3 months. Ordinal and binary logistic regression analyses were performed. Results Among 429 eligible patients (mean age, 73.5±11.1 years; 58.0% men; median [interquartile range] initial National Institutes of Health Stroke Scale score, 12 [6-18]), 51 (11.9%) received SGLT2i post-stroke. Successful recanalization was achieved in 85.3%. In the OW model, SGLT2i use was associated with improved 3-month mRS distribution (common OR, 2.31; 95% CI, 1.30-4.10) and higher odds of favorable outcome (OR, 2.26; 95% CI, 1.15-4.43). Findings were consistent across IPTW and PSM analyses, with the IPTW model additionally showing significance for excellent outcomes. No significant effect modification was observed across key clinical subgroups. Conclusions SGLT2i use was independently associated with improved 3-month functional outcomes after EVT in diabetic patients with AIS. Conflict of interest All authors: nothing to disclose Figure 1 - belongs to Results