Optimal Delay Time to Initiate Anticoagulation After Ischemic Stroke in Atrial Fibrillation

Long-term oral anticoagulation is standard for secondary stroke prevention in patients with atrial fibrillation (AFib). However, there is limited data and no consensus on the timing of when to initiate anticoagulation therapy, and concern that starting too soon risks symptomatic hemorrhagic transformation. These data are derived almost exclusively from heparins and Vitamin K antagonists (e.g.,warfarin). Now that NOACs have become the mainstay of stroke prophylaxis in AFib and have more rapid and consistent anticoagulation and fewer strokes (hemorrhagic especially), the question of optimal timing of NOAC initiation is of increasing importance. The primary aim is to determine the time-to-treatment interval with the lowest associated risk for adverse events in the context of anticoagulation therapy with NOACs for acute stroke patients with non-valvular AFib. The question will be investigated with a prospective, adaptive, randomized, controlled "dose-exploration" trial with the time to treatment with NOAC therapy treated as the incremental "dose". An adaptive, pragmatic trial will be performed that will not deviate from the treating physicians' usual practice except for randomizing the time to start the NOAC. Data collection will be limited to those fields necessary for the planned primary and secondary analyses. The composite primary outcome event will be any of the following within 30 days of the index stroke: Ischemic Events (symptomatic ischemic stroke or systemic embolism) or Hemorrhagic Events (symptomatic hemorrhagic transformation of index ischemic stroke, other symptomatic intracranial hemorrhage, or major extracranial hemorrhage). Four time-to-treatment intervals, i.e. study arms, between 2 and 14 days will be investigated: 72 (+/- 24) hours, 132 (+/- 12) hours, 228 (+/- 12) hours, and 324 (+/- 12) hours. An innovative adaptive design will be used which includes response adaptive randomization and modeling of ischemic and hemorrhagic outcome events. The ischemic and hemorrhagic events within the composite primary endpoint are modeled separately using their known monotonic property that the risk of an event increases (ischemic) or decreases (hemorrhage) as the time-to-treatment interval lengthens. Interim analysis will occur after 100 subjects are randomized and have the primary outcome, where the primary outcome will be analyzed and new randomization probabilities will be calculated to favor the arms that have a better risk-profile. Thereafter, interim analyses will occur after 50 subjects are randomized under the new randomization probabilities, and further new randomization probabilities will be calculated to favor the arms that have a better risk-profile.