Abstract
IMPORTANCE: Management after an ischemic stroke occurring despite direct oral anticoagulant (DOAC) therapy for atrial fibrillation (AF) varies widely. Switching anticoagulation is common in clinical practice, although evidence supporting this strategy is limited. OBJECTIVE: To evaluate whether continuation of treatment with the same DOAC was noninferior to switching oral anticoagulant therapy with respect to 90-day clinical outcomes. DESIGN, SETTING, AND PARTICIPANTS: This multicenter registry-based cohort study with an emulated target trial design included consecutive adult patients with AF who experienced a breakthrough ischemic stroke while receiving uninterrupted DOAC therapy and resumed anticoagulation therapy thereafter. Patients were enrolled between February 2020 and February 2025, across 35 stroke centers in 9 countries in Europe and North Africa, with a standardized 90-day follow-up. The dataset was locked on September 1, 2025. A noninferiority comparison of switching vs continuation strategies was performed. Baseline confounding was addressed using inverse probability of treatment weighting (IPTW). The primary noninferiority margin was an absolute risk difference of 3.0 percentage points in 90-day net clinical benefit. EXPOSURE: The intervention group switched to treatment with a different DOAC or vitamin K antagonist; the comparator group continued therapy with the prestroke DOAC. MAIN OUTCOMES AND MEASURES: The primary outcome was 90-day net clinical benefit, defined as recurrent ischemic stroke and moderate to severe bleeding. Secondary outcomes included recurrent ischemic events, symptomatic intracerebral hemorrhage, moderate to severe extracranial bleeding, all-cause mortality, and vascular death. RESULTS: Among 1006 patients included in the analysis (median age, 80.4 [IQR, 73.4-85.4] years; 503 female [50.0%] and 503 [50.0%] male), 463 (46.0%) continued the same DOAC therapy and 543 (54.0%) switched therapy. After IPTW adjustment, the 90-day net clinical benefit was 4.9% with switching and 5.1% with continuation, corresponding to a risk difference of -0.3 percentage points (90% CI, -2.7 to 2.1 percentage points), meeting the prespecified noninferiority criterion. For recurrent ischemic events and bleeding outcomes, the absolute differences were within the predefined noninferiority margins. Noninferiority was not demonstrated for all-cause or vascular mortality. CONCLUSIONS AND RELEVANCE: In patients with breakthrough ischemic stroke during DOAC therapy, switching anticoagulation treatment was not associated with clinically meaningful short-term benefit compared with continuation. These findings suggest that switching does not provide additional benefit compared with continuing treatment with the same DOAC. Randomized clinical trials are needed to identify strategies to improve secondary prevention after a breakthrough ischemic stroke.