Abstract P605: Border-Zone Infarcts Predict Early Recurrence in Patients With Large Artery Atherosclerotic Subtype Despite Medical Treatment

Abstract

Objective: We aim to investigate infarct patterns and 90-day recurrence in patients with symptomatic intracranial atherosclerotic disease. Background: Large artery atherosclerosis subtype carries a high risk of early recurrent stroke despite medical management. Predictors of recurrence remain poorly understood. We hypothesized that border-zone infarcts are associated with a higher risk of recurrence. Design/Methods: We included consecutive patients admitted to NYU Langone Health (Manhattan and Brooklyn campuses) over 32-months with a diagnosis of acute ischemic stroke secondary to symptomatic intracranial or tandem atherosclerosis. Patients with purely extracranial stenosis were excluded. The primary predictor was infarct pattern (border-zone vs. non-border-zone infarction), defined in accordance to previous studies. We used univariate and multivariable cox-regression models to determine associations between infarct pattern and recurrent cerebrovascular events (RCVE) at 90-days. Results: Ninety-nine patients met the inclusion criteria; 95 were intracranial and 4 were tandem lesions. The mean age was 70 years, 58.6% were men, 15.1% Black and 16.1% Asian. The median symptom onset to arrival time was 1 day, nearly 74.7% of patients were treated with dual antiplatelet therapy and 99% were treated with high intensity statin. Within 90 days of follow up, 19.2% (19/99) had RCVE. In univariate analyses, the only factor associated with RCVE was border-zone infarct pattern when compared to non-border-zone infarct pattern (30.2% vs. 10.7%, p=0.02). In cox regression models, after adjusting for age and sex, border-zone infarct pattern was associated with increased risk of RCVE (adjusted HR 3.21 95% CI 1.21-8.51, p=0.019). Sensitivity analyses excluding patients with tandem lesions (n = 4) did not meaningfully change our findings (adjusted HR 3.04 95% CI 1.11-8.31, p=0.031). Conclusions: In real world post-SAMMPRIS medically treated patients with ICAD, infarct pattern was predictive of 90-day RCVE. Border-zone infarcts are likely a surrogate marker of impaired distal blood flow, highlighting the importance of targeting stroke mechanisms and developing alternative treatment strategies for this high-risk cohort.