ABSTRACT NUMBER: ESOC2026A507 DETERMINANTS AND CLINICAL IMPACT OF ANTICOAGULATION REVERSAL IN ORAL ANTICOAGULANT–RELATED INTRACEREBRAL HEMORRHAGE

Abstract Background and aims Oral anticoagulant (OAC)–related intracerebral hemorrhage (ICH) carries high mortality. Despite guideline recommendations, anticoagulation reversal is inconsistently used, particularly in severe presentations. We aimed to identify determinants of reversal administration and its association with mortality in real-world practice. Methods We analyzed data from the prospective, multicenter ARICH registry including consecutive OAC-related ICH patients admitted to 18 Spanish stroke centers (2019–2024). Outcomes included reversal administration (primary) and 24-hour and 90-day mortality. Hierarchical mixed-effects regression models with hospital-level random intercepts were used. Post-hoc analyses evaluated mortality in patients with ICH volume >60 mL and/or GCS ≤5, commonly excluded from RCTs. Results Among 891 patients (mean age 79.0±9.3 years, 541 [60.7%] male), 678 (76.1%) received reversal. Independent determinants included baseline mRS (aOR 0.84, 95% CI 0.71-0.99), GCS (aOR 1.24, 95% CI 1.13-1.36), ICH volume (per 10-mL increase, aOR 0.86, 95% CI 0.80-0.93), and time since last DOAC dose (per 60-minute increase, aOR 0.94, 95% CI 0.92-0.97). Overall 24-hour mortality was 159/889 (17.9%) and 90-day mortality was 429/887 (48.6%). Reversal was independently associated with lower 24-hour (aOR 0.32, 95% CI 0.18-0.56) and 90-day (aOR 0.59, 95% CI 0.35-0.99) mortality, and with improved outcomes across clinical severity strata: lower 24-hour mortality (Figure 1A) and higher 90-day survival (Figure 1B) (adjusted log-rank P<0.05 for all pairwise comparisons). Conclusions In real-world practice, anticoagulation reversal is influenced by perceived prognosis and is independently associated with lower 24-hour and 90-day mortality, including among patients with severe ICH presentations. Conflict of interest David Rodriguez-Luna: nothing to disclose. Olalla Pancorbo: nothing to disclose. Ana Núñez: nothing to disclose. Renato Simonetti: nothing to disclose. Laura Sánchez Cirera: nothing to disclose. Marta Serrano Ponz: nothing to disclose. Rocío Vera Lechuga: nothing to disclose. Cristina Pérez: nothing to disclose. Luis Prats-Sanchez: nothing to disclose. Carlos A. Molina: nothing to disclose. Figure 1 - belongs to Results