A Simple Score(ABC <sup>2</sup>D) to Predict Atherosclerotic Thrombosis or Embolic Intracranial Large Vessel Occlusion Stroke Before Endovascular Treatment

Abstract

Background: Identification of occlusion mechanism before endovascular treatment (EVT) is of great significance for intracranial large vessel occlusion stroke (ILVOs). Cardioembolism occlusion and large-artery atherosclerosis are the two common etiologies of ILVOs. However, there is few of valid methods to predict the two etiologies of ILVOs before EVT. Methods: We evaluated retrospectively two etiologies of ILVOs: intracranial atherosclerosis stenosis-related occlusion (ICAS-O) and embolism-related occlusion( EMB-O) in a cohort of patients at a national comprehensive stroke centre database of China over a 77-month period. Patients were divided at a ratio of 2:1 randomly into the training and the validation cohorts. We derived a score for ILVOs in the derivation cohort of patients , and assessed the score in the validation cohort. Results: We identified 608 from 662 patients with ILVOs who received EVT in the study period. Subsequent to adjustment for confounding factors, atrial fibrillation. (AF) (no vs. yes, OR=27·29, 95%CI: 13·27-56·09), hypertension (yes vs. no, OR=2·90, 95%CI: 1·34-6·26), diabetes mellitus (DM) (yes vs. no, OR=2·80, 95%CI: 1·45-5·42), National Institutes of Health Stroke Scale(NIHSS) (<7 vs. ³7, OR=2·92, 95%CI: 1·22-6·99), and computed tomography hyperdense sign (CTHS) (no vs. yes, OR=2·86, 95%CI: 1·22-6·74) appeared to be statistically significantly related to ICAS-O. We derived the AF-Blood pressure-Clinical neurological deficit-CTHS-DM (ABC2D) score to help to predict ICAS-O or EMB-O etiologies of ILVOs. The area under curve (AUC) of receiver operating characteristic(ROC) of ABC2D score for ICAS-O identification in the derivation and validation cohort were 0.886 (95%CI: 0·839-0·933) and 0·880 (95%CI: 0·846-0·914), respectively. The optimal cutoff value of an ABC2D score was 5 point with 74·5% sensitivity, 90·3% specificity, and 82·2% accuracy of predicting ICAS-O. Conclusions: Etiology of ILVOs before EVT seems to be highly predictable, although further validations and refinements are needed. In order to develop precise strategy of EVT, the ABC2D score may be used in routine clinical practice to identify etiology of ILVOs patients who need emergency EVT. Funding Information: This work was funded by grants from the Special Fund of Science and Technology of Guangdong Province, China (Reference No.2020S00050) and the High-level Hospital Construction Research Project of Maoming People’s Hospital (Reference No. Yueweihan 2018413). Declaration of Interests: We declare that we have no conflict of interest. Ethics Approval Statement: The Institutional Ethics Committee of our hospital has granted approval for our work (2019027).