Outcomes of Patients With Embolic Stroke of Undetermined Source Treated With Antiplatelet Agents or Anticoagulation: A Multicenter Cohort Study

Abstract

Background and objectives: Embolic stroke of undetermined source (ESUS) can be attributed to a variety of potential embolic sources, with differential response to anticoagulation.

Methods: A multicenter, retrospective observational cohort study (27 sites) of consecutive adult patients with acute ischemic stroke due to ESUS (admitted 2015-2024) was conducted. The aim was to compare outcomes after antiplatelet(s) vs anticoagulant (±antiplatelet) treatment in patients with ESUS across potential embolic sources. The time from admission to the primary composite outcome of recurrent stroke, major bleeding, or death was assessed using adjusted Cox proportional hazard regression (clustered by site) and propensity score (PS) matching with (1) inverse probability of treatment weighting (IPTW) and (2) 10:1 nearest-neighbor matching with replacement, adjusting for age, stroke severity, and potential embolic sources (e.g., left ventricular injury and patent foramen ovale). Recurrent stroke, major bleeding, and death were also assessed as secondary outcomes, with stratification by potential embolic sources.

Results: Of the 2,328 included patients (n = 230 treated with anticoagulation), the median age was 65 years (interquartile range [IQR] 54-75), 50% were female, and the median NIH Stroke Scale score was 4 (IQR 2-11). Compared with patients treated using antiplatelet(s) therapies, those treated with anticoagulants were not at a lower risk of the primary outcome in the adjusted Cox model (adjusted hazard ratio [aHR] 1.00, 95% CI 0.69-1.45), adjusted IPTW regression model (aHR 1.15, 95% CI 0.79-1.66), or 10:1 PS-matched regression model (aHR 1.00, 95% CI 0.70-1.44). In patients with left ventricular injury, anticoagulation was associated with a lower rate of the primary outcome (aHR 0.35, 95% CI 0.16-0.77; p-interaction <0.01) and trended toward a lower rate of recurrent ischemic stroke (aHR 0.22, 95% CI 0.05-1.08; p-interaction = 0.04) when compared with patients treated with antiplatelet(s).

Discussion: These real-world data validate randomized trial results in ESUS, which reported no net benefit of anticoagulation over antiplatelet therapy. These data suggest possible benefit of anticoagulation in patients with left ventricular injury, as in previous cohort studies, although the findings are limited by the small number of patients treated with anticoagulation. Future trials should evaluate treatment differences in this subgroup.

Trial registration information: Cardiac Abnormalities in Stroke Prevention and Risk of Recurrence; registration ID: NCT06398366. Registered on May 3, 2024.

Classification of evidence: This study provides Class III evidence that in patients with ESUS, anticoagulation was not superior to antiplatelet therapy in reducing the risk of recurrent stroke, bleeding, or death.