关键词: 卒中; 量表; 复发; 验证; 预测

Abstract: Objective To retrospectively validate the prediction of the essen stroke risk score(ESRS) in non-atrial fibrillation patients with acute ischemic stroke hospitalized in a stroke unit of China, to modify it, and to investigate the prediction of the streamlined ESRS.Methods Five hundred and ten hospitalized patients with acute cerebral infarction were collected within 24 hours, and clinical follow-up were available in 468 cases after 12 months. The ESRS were analyzed in accordance with the follow-up results and were divided into recurrence group and the non-recurrence group. Basic demographic information and various risk factors were compared via statistical analysis in both groups. We modified the ESRS and validated the stroke recurrence rate of the ESRS with the streamlined ESRS.Results In recurrence group(N=201) and the non-recurrence group(N=267), four risk factors of ESRS, including hypertension(P=0.0010), other heart disease(P=0.0162), smoking(P=0.0220),and previous transient ischemic attack or ischemic stroke, (P=0.0000) had significant difference (P<0.05). Three other risk factors, age(P=0.2565), previous myocardial infarction(P=0.5072), peripheral arterial disease(P=0.1578), showed no significant difference between the two groups(P>0.05). The differences of recurrence rate in diabetic patient group and non-diabetic patient group closely reached to statistical significance(P=0.07). The area under receiver operating characteristic(ROC) curve of the streamlined ESRS predicting the recurrence rate of stroke was 0.645(95%CI:0.596~0.695) and the area under ROC curve of the complete ESRS was 0.650(95%CI:0.601~0.699). The difference was not statistically significant(P=0.8875>0.05).Conclusion In this study, the ESRS and the streamlined ESRS could predict the recurrence of stroke, which achieved similar prediction accuracies.

Key words: Stroke; Scale; Recurrence; Validation; Forecasting