ABSTRACT NUMBER: ESOC2026A1096 MECHANISMS OF RECURRENT ISCHEMIC STROKE AMONG PATIENTS WITH CRYPTOGENIC STROKE AND ATRIAL CARDIOPATHY: A SECONDARY ANALYSIS OF THE ARCADIA CLINICAL TRIAL

Abstract Background and aims Key characteristics of recurrent stroke after cryptogenic stroke are unclear. Methods ARCADIA was a multicenter, randomized trial comparing apixaban with aspirin in patients with cryptogenic stroke and biomarker evidence of atrial cardiopathy. Three vascular neurologists reviewed medical records to decide, by consensus, the most likely etiology of recurrent strokes using the TOAST classification. Among patients with recurrent ischemic stroke, we used logistic regression to evaluate associations between baseline atrial cardiopathy biomarker levels and embolic stroke of undetermined source (ESUS) versus other recurrent ischemic stroke mechanisms, before and after adjusting for demographics, comorbidities, treatment assignment, and baseline infarct topography. Results Among 1,015 randomized participants, 75 (7.4%) experienced recurrent ischemic stroke during a mean follow-up of 1.8 years. Recurrent ischemic strokes were most often cryptogenic (n = 56, 74.7%), of which half (n = 28) met criteria for ESUS. The remainder were due to large-artery (n = 8, 10.7%), cardioembolic (n = 7, 9.3%), small-artery (n = 2, 2.7%), and other mechanisms (n = 2, 2.7%). Without or with adjustment, the biomarkers were not associated with ESUS versus other ischemic stroke subtypes: NT-proBNP (OR per SD, 0.6; 95% CI, 0.2–2.1); P-wave terminal force in V1 (OR per SD, 1.40; 95% CI, 0.7–2.8); and left atrial diameter index (OR per SD, 0.6; 95% CI, 0.2–1.6). Conclusions Among patients with cryptogenic stroke and evidence of atrial cardiopathy, most recurrent ischemic strokes remained cryptogenic, and half of these met criteria for ESUS. Baseline atrial cardiopathy biomarker levels were not associated with ESUS recurrence compared with all non-ESUS recurrent ischemic stroke mechanisms. Conflict of interest Dr. Bianco, Prof. Navi, Dr Zang, Prof Elkind, Prof Tirschwell, Prof Kronmal, Prof. Elm and Prof. Broderick: nothing to disclose. Prof. Longstreth: funding The National Institutes of Health funded the ARCADIA trial, the BMS-Pfizer Alliance provided in-kind study drug to the StrokeNet Central Pharmacy for distribution, and Roche Diagnostics provided ancillary funding for laboratory supplies. Financial disclosures for Hooman Kamel: a Deputy Editor role for JAMA Neurology; clinical trial steering/executive committee roles for the LIBREXIA-AF (Janssen) trial, INTERCEPT (World Health Research Institute and Javelin) trial, and LAAOS-4 (Population Health Research Institute and Boston Scientific) trial; consulting or endpoint adjudication committee roles for AbbVie, Alnylam, Arthrosi, AstraZeneca, Bayer, Boehringer Ingelheim, Eli Lilly, GlaxoSmithKline, Medtronic, Novartis, and Novo Nordisk; and ownership interests in Ascential Technologies, Doug Labs, and TETmedical.