Background: Whether hemorrhagic transformation (HT) modifies the treatment effect of early versus late initiation of direct oral anticoagulation (DOAC) in people with ischemic stroke and atrial fibrillation is unknown.
Methods: This is a post-hoc analysis of the ELAN trial. The primary outcome was a composite of recurrent ischemic stroke, symptomatic intracranial hemorrhage (sICH), major extracranial bleeding, systemic embolism, or vascular death within 30 days. Secondary outcomes were the individual components, 30- and 90-day functional outcome. We estimated outcomes based on HT, subclassified as hemorrhagic infarction (HI) or parenchymal hemorrhage (PH) on pre-randomization imaging (core-lab rating) using adjusted risk differences (aRD) between treatment arms.
Results: Overall, 247/1970 (12.5%) participants had HT (114 HI 1, 77 HI 2, 34 PH 1, 22 PH 2). For the primary outcome, the estimated aRD (early versus late) was -2.2% (-7.8 to 3.5%) in people with HT (HI: -4.7%, -10.8 to 1.4%; PH: 6.1%, -8.4 to 20.7%), and -0.9% (-2.6 to 0.8%) in people without. Numbers of sICH were identical in people with and without HT. With early treatment, the estimated aRD for poor 90-day functional outcome (mRS 3-6) was 11.4% (-0.9 to 23.7%) in participants with HT (HI: 7.2%, -6.6 to 21.0%; PH: 25.1%, 0.2 to 50.0%), and -2.6% (-7.1 to 1.8%) in people without HT.
Conclusions: We found no evidence of major treatment effect heterogeneity or safety concerns with early versus late DOAC initiation in people with and without HT. However, early DOAC initiation may worsen functional outcomes in people with PH.