A simple score to predict atherosclerotic or embolic intracranial large-vessel occlusion stroke before endovascular treatment

Abstract

OBJECTIVE The authors developed a method to predict the etiology of intracranial large-vessel occlusion stroke (ILVOS) before endovascular treatment. METHODS The authors retrospectively evaluated two etiologies of ILVOS-intracranial atherosclerotic stenosis-related occlusion (ICAS-O) and embolism-related occlusion (EMB-O)-in a cohort of patients from the National Comprehensive Stroke Center database of China. Patients were randomly divided into the derivation and validation cohorts at a ratio of 2:1. The authors derived the score in the derivation cohort and assessed the score in the validation cohort. RESULTS The authors identified 608 of 662 patients with ILVOS who received endovascular treatment during the study period. After adjustment for confounding factors, hypertension (OR 2.90, 95% CI 1.34-6.26), diabetes mellitus (OR 2.80, 95% CI 1.45-5.42), absence of atrial fibrillation (OR 27.29, 95% CI 13.27-56.09), National Institutes of Health Stroke Scale score < 7 (OR 2.92, 95% CI 1.22-6.99), and absence of the computed tomography hyperdense sign (OR 2.86, 95% CI 1.22-6.74) were significantly related to ICAS-O. A score was derived to help predict ICAS-O or EMB-O. The area under the curve values of the receiver operating characteristic curve for ICAS-O identification were 0.886 (95% CI 0.839-0.933) and 0.880 (95% CI 0.846-0.914) in the derivation and validation cohorts, respectively. CONCLUSIONS The atrial fibrillation-blood pressure-clinical neurological deficit-computed tomography hyperdense sign-diabetes mellitus (ABC2D) score can be used to identify atherosclerotic or embolic etiology of patients with ILVOS who require emergency endovascular treatment.