论文部分内容阅读
目的探讨以改良弗明汉卒中风险评分(FSP)、脑血流动力学指标(CVHI)及脂蛋白磷脂酶A2(Lp-PLA2)构建的脑卒中预测模型对脑卒中的预测能力。方法收集101例首发缺血性脑卒中患者和156名社区非脑卒中常住人口的临床资料。利用多因素Logistic回归分析FSP、CVHI及Lp-PLA2预测脑卒中发生的价值,建立单纯FSP、FSP+CVHI、FSP+Lp-PLA2、FSP+CVHI+LpPLA2预测模型,通过受试者工作特征曲线(ROC)的曲线下面积(AUC)分析各模型的脑卒中预测能力。结果 FSP、CVHI、Lp-PLA2每增加1个等级发生脑卒中的风险分别增加2.85、3.25、7.75倍(P值分别为0.043、0.036、<0.001)。单纯FSP、FSP+CVHI、FSP+Lp-PLA2、FSP+CVHI+LpPLA2预测模型ROC的AUC分别为0.588、0.845、0.858和0.936。结论 FSP+CVHI+Lp-PLA2预测模型可有效预测脑卒中的发生。
Objective To investigate the predictive ability of stroke prediction model with modified stroke risk score (FSP), cerebral hemodynamic parameters (CVHI) and lipoprotein phospholipase A2 (Lp-PLA2). Methods The clinical data of 101 first-episode ischemic stroke patients and 156 non-stroke permanent residents were collected. The values of FSP, CVHI and Lp-PLA2 in prediction of stroke were analyzed by multivariate Logistic regression. The prediction models of simple FSP, FSP + CVHI, FSP + Lp-PLA2 and FSP + CVHI + LpPLA2 were established. ROC) of the area under the curve (AUC) analysis of the model of stroke prediction ability. Results The risk of stroke increased by 2.85, 3.25 and 7.75 times for each additional level of FSP, CVHI and Lp-PLA2 (P = 0.043, 0.036, <0.001 respectively). The AUC of ROC in simple FSP, FSP + CVHI, FSP + Lp-PLA2 and FSP + CVHI + LpPLA2 prediction models were 0.588, 0.845, 0.858 and 0.936, respectively. Conclusion FSP + CVHI + Lp-PLA2 prediction model can effectively predict the occurrence of stroke.