Divergent Small Vessel Disease Burden in Warfarin-Associated and Direct Oral Anticoagulant-Associated Intracerebral Hemorrhage

Anticoagulant (AC)-associated intracerebral hemorrhage (ICH) is a significant complication of AC therapy, often leading to unfavorable outcomes.The association between increased AC intensity and a greater risk of ICH is well established.For vitamin K antagonist, the risk of ICH increases with an elevated international normalized ratio (INR). 1 Exposure-response analyses for direct oral anticoagulants (DOAC) similarly demonstrate an increased ICH risk with higher drug concentrations. 2In recent years, cerebral small vessel disease (SVD) has also emerged as an important factor contributing to ICH.Our previous research showed that DOAC users who experienced ICH during treatment exhibited a greater SVD burden. 3However, whether SVD burden differs between warfarin-associated and DOAC-associated ICH remains unclear.This study aimed to compare SVD burden among three cohorts of patients with spontaneous ICH, including DOAC users, warfarin users, and AC non-users.Additionally, AC users were analyzed according to different levels of AC intensity at presentation.This prospective cohort study enrolled consecutive patients aged ≥20 years admitted to the National Taiwan University Hospital for spontaneous ICH.According to the antithrombotic agents given before ICH, patients were classified as DOAC users or war-farin users.For each AC user, an AC non-user was matched by sex and age (±2 years).Patients with pure intraventricular hemorrhage, tumor bleeding, or cerebral venous sinus thrombosis were excluded.The study protocol received approval from the National Taiwan University Hospital International Ethics Committee (Rec.No. 202003059 RINA and Rec.No. 9561703046) and was registered on ClinicalTrials.gov(NCT06144866).Brain computed tomography (CT) was used to assess SVD burden based on white matter lesions (WML), lacunes and cerebral atrophy.The validated CT-evaluated SVD (CT-SVD) burden scoring system ranges from 0 to 3 (higher scores indicate greater burden). 4The specific definitions and scoring criteria are detailed in Supplementary Table 1.Hematoma volume was estimated using the ABC/2 formula. 5Hematoma expansion was defined as an increase of ≥6 mL or ≥33% in volume on follow-up imaging performed 24-48 hours after the initial scan. 6Neuroimaging was independently evaluated by a neuroradiologist (Y.F.C.) and a neurologist (C.H.C.).The inter-rater agreement, expressed as weighted Kappa values, was 0.72 (95% confidence interval [CI] 0.59-0.85)for the total WML score and 0.76 (95% CI 0.65-0.86)for the total atrophy score in our previous investigation. 4DOAC concentrations at hospital presentation were measured using an ultra-high-performance liquid chromatography-tandem mass