Impact of low-dose alteplase bridging therapy on outcomes after mechanical thrombectomy in patients with atrial fibrillation-associated acute stroke: A retrospective cohort study

The optimal alteplase dose for bridging therapy in atrial fibrillation related large vessel occlusion stroke is uncertain. We compared low-dose alteplase plus mechanical thrombectomy with standard-dose bridging and direct thrombectomy. This single-center retrospective cohort included all consecutive eligible patients (n = 375) treated between January 2021 and December 2024; no a priori sample size calculation was performed. Patients received low-dose alteplase 0.6 mg/kg plus thrombectomy (group A, n = 120), standard-dose 0.9 mg/kg plus thrombectomy (group B, n = 125), or direct thrombectomy without thrombolysis (group C, n = 130). Primary outcomes were successful reperfusion (mTICI 2b-3), 90-day functional independence (mRS 0-2), and symptomatic intracerebral hemorrhage. Secondary outcomes included early neurologic change, 72 hour National Institutes of Health Stroke Scale, mortality, and key procedural metrics. Multivariable logistic regression adjusted for age, severity, time metrics, occlusion site, and comorbidities. Baseline features were similar. Versus direct thrombectomy, low-dose bridging improved reperfusion (91.7% vs 84.6) and increased 90-day independence (58.3% vs 45.4). Early improvement and 72 hour National Institutes of Health Stroke Scale favored low-dose bridging. Symptomatic intracerebral hemorrhage was lower with low-dose versus standard-dose bridging (3.3% vs 6.4) and comparable to direct thrombectomy. Low-dose bridging also shortened puncture-to-reperfusion time. In adjusted models, low-dose bridging predicted 90 day independence (odds ratio 1.68) and lower odds of symptomatic hemorrhage than standard-dose (odds ratio 0.48). Low-dose alteplase bridging before thrombectomy improved efficacy and safety in atrial fibrillation related large vessel occlusion stroke and may be preferable for patients at higher bleeding risk.