摘要：
Objective This study aims to explore the risk factors of hemorrhagic transformation (HT),and compare the predictive ability of different predictive models.Methods Retrospective analysis of clinical data of patients with intravenous thrombolysis in our hospital from June 2015 to June 2018.The risk factors were statistically analyzed.The results were fitted into a new predictive model named MODEL.The area under receiver operating characteristic (ROC) curve was used to compare the predictive ability of the MODEL and some traditional methods including the Multicenter Stroke Survey (MSS),and the Totaled Health Risks In Vascular Events Score (THRIVE),and the Glucose at presentation,Race,Age,Sex,systolic blood Pressure at presentation,Severity of stroke at presentation (GRASPS) to HT.Results The results of single factor analysis showed that there were significant differences in the history of hypertension,the history of atrial fibrillation,the systolic pressure before thrombolysis,the National Institute of Health Stroke Scale (NIHSS) score before thrombolysis,the level of plasma total cholesterol (TG),low density lipoprotein cholesterol (LDL-C),apolipoprotein B (Apo B),D dimer (DD),and the platelet distribution width and the sodium ion.Multivariate logistic regression analysis confirmed that the NIHSS score before thrombolysis,the level of plasma LDL-C and DD were independent risk factors for HT.The formation of a new predictive model was logi(p) =0.21 +0.99 × NIHSSscore-0.834 × LDL-C +0.002 × DD,which was defined as predictive mode of MODEL.The four predictive models all had predictive effects on HT.Z test results showed that,compared with the GRASPS,the MODEL and MSS showed a higher predictive power.Conclusions The NIHSS score before thrombolysis,and the level of plasma LDL-C,and DD were independent risk factors for HT after intravenous thrombolysis.MODEL and MSS had better predictive performance.%目的 探讨急性脑梗死静脉溶栓后继发出血转化(HT)的危险因素,并比较不同预测模型的预测能力.方法 回顾性分析2015年6月至2018年6月于本院静脉溶栓的急性脑梗死患者的临床资料,对其危险因素进行统计学分析,结果拟合为新的预测模型MODEL,应用ROC曲线下面积比较MODEL、多中心卒中调查预测模型(MSS)、血管事件健康风险汇总评分(THRIVE)、GRASPS评分对继发HT的预测能力.结果 单因素分析显示,高血压病史、房颤病史、溶栓前收缩压、溶栓前美国国立卫生研究院卒中量表(NIHSS)评分、总胆固醇(TG)、低密度脂蛋白胆固醇(LDL-C)、载脂蛋白B(Apo B)、D二聚体(DD)、血小板分布宽度、钠离子与静脉溶栓后继发HT有关(P＜0.05).多因素logistic回归分析证实,溶栓前NIHSS评分、LDL-C、DD是HT的独立危险因素,形成新的预测模型:logit(p) =0.21 +0.99×溶栓前NIHSS评分-0.834×LDL-C+0.002×DD,定义为MODEL预测模型.4种预测模型对继发HT均有预测作用,预测能力从强至弱分别为MODEL＞ THRIVE＞ MSS＞GRASPS.结论 溶栓前NIHSS评分、LDL-C、DD是影响静脉溶栓后继发HT的独立危险因素,MOD-EL有较好的预测效能.
