Development and Validation of a Risk Score for Predicting lschemic Stroke After Transient lschemic Attack

Background: A risk stratification scale is es-sential to identify high-risk patients who had transient is-chemic attack (TIA) to prevent subsequent permanent dis-ability caused by ischemic stroke. Objective: This study aimed to develop and validate a scoring system to pre-dict acute ischemic stroke within 90 days after TIA in an emergency department (ED). Methods: We retrospectively analyzed the data of patients with TIA in a stroke registry between January 2011 and September 2018. Characteristics, medication history, electrocardiogram (ECG), and imaging findings were collected. Univariable and multivariable step-wise logistic regression analyses were performed to create an integer point system. The area under the receiver operat-ing characteristic curve (AUC) and the Hosmer-Lemeshow (HL) test were used to examine discrimination and calibra-tion. Youden's Index was also used to determine the best cutoff value. Results: A total of 557 patients were included, and the occurrence rate of acute ischemic stroke within 90 days after TIA was 5.03%. After multivariable analysis, a new integer point system was created-MESH (Medica-tion Electrocardiogram Stenosis Hypodense) score-which contained medication history (antiplatelet medication taken before admission, 1 point), right bundle branch block on electrocardiogram (1 point), intracranial stenosis > 50% (1 point), and size of the hypodense area on computed tomog-raphy (diameter > 4 cm, 2 points). The MESH score showed adequate discrimination (AUC = 0.78) and calibration (HL test = 0.78). The best cutoff value was 2 points, with a sensitivity of 60.71% and specificity of 81.66%. Conclusions: The MESH score indicated improved accuracy for TIA risk stratification in the ED setting. (c) 2022 Elsevier Inc. All rights reserved.