基于Lasso-Logistic回归分析构建中青年急性LHI静脉溶栓后出血转化预测模型  

Construction of a prediction model for hemorrhagic transformation after intravenous thrombolysis for acute LHI in young and middleaged patients based on Lasso-Logistic regression analysis

作　　者：张天姣 牛秋丽[1] 武娟 高文慧 ZHANG Tianjiao;NIU Qiuli;WU Juan;GAO Wenhui(Emergency Department,Xuanwu Hospital,Capital Medical University,Beijing 100032,China)

机构地区：[1]首都医科大学宣武医院急诊科,北京100032

出　　处：《中国急救复苏与灾害医学杂志》2024年第10期1316-1322,1365,共8页China Journal of Emergency Resuscitation and Disaster Medicine

基　　金：首都医科大学2022年教育教学改革研究课题立项项目(编号:2022JYY453)。

摘　　要：目的基于Lasso-Logistic回归分析中青年急性大面积脑梗死(LHI)患者静脉溶栓后出血转化(HT)的影响因素,并建立列线图预测模型。方法选取2022年5月—2023年10月首都医科大学宣武医院行阿替普酶静脉溶栓治疗的中青年急性LHI患者495例,溶栓后24 h是否发生HT分为HT组、非HT组。比较两组临床资料,通过Lasso-Logistic回归分析HT发生的影响因素,根据回归分析筛查出的指标构建列线图预测模型,并评价列线图预测模型的临床应用价值。结果495例中青年急性LHI患者中HT发生率为25.66%(127/495);HT组年龄、房颤占比、糖尿病占比、发病至溶栓时间(ONT)3.0~4.5 h占比、溶栓前美国国立卫生研究院卒中量表(NIHSS)评分、血清C反应蛋白(CRP)、降钙素原(PCT)、中性粒细胞/淋巴细胞比值(NLR)、基质金属蛋白酶-9(MMP-9)、脂蛋白相关磷脂酶A2(Lp-PLA2)、可溶性凝集素样氧化低密度脂蛋白受体-1(sLOX-1)水平高于非HT组,外周血血小板、尿酸水平低于非HT组(P<0.05);房颤、糖尿病、ONT、溶栓前NIHSS评分、外周血血小板、尿酸水平、血清NLR、MMP-9、Lp-PLA2、sLOX-1水平均是静脉溶栓治疗后HT发生的影响因素(P<0.05);列线图预测模型预测HT发生的AUC为0.893(95%CI:0.862~0.925),敏感度、特异度分别为83.65%、88.42%,该模型校准度良好,且具有明显的正向净收益。结论房颤、糖尿病、ONT、溶栓前NIHSS评分、外周血血小板、尿酸水平、血清NLR、MMP-9、Lp-PLA2、sLOX-1水平均是中青年急性LHI患者阿替普酶静脉溶栓治疗后发生HT的影响因素,依据上述影响因素建立列线图预测模型,该模型对静脉溶栓治疗后HT发生具有一定预测价值及临床实用性,临床可依据该模型筛选早期HT发生高危患者,并采取个体化治疗措施,以降低HT发生风险。Objective To analyze the influencing factors of hemorrhagic transformation(HT)after intravenous thrombolysis in young patients with acute large hemispheric infarcti(LHI)based on Lasso-Logistic regression analysis,and establish a nomogram prediction model.Methods A total of 495 young and middle-aged patients with acute LHI who received alteplase intravenous thrombolytic therapy in Xuanwu Hospital,Capital Medical University from May 2022 to October 2023 were selected.Whether HT occurred 24 hours after thrombolytic therapy was divided into HT group and non-HT group.The clinical data of the two groups were compared,the influencing factors of HT occurrence were analyzed by Lasso-Logistic regression,and a nomogram prediction model was constructed according to the indicators screened by regression analysis,and the clinical application value of the nomogram prediction model was evaluated.Results Among 495 young and middle-aged patients with acute LHI,the incidence of HT was 25.66%(127/495).The age,atrial fibrillation(AF)proportion,diabetes proportion,proportion of Time from onset to thrombolysis(ONT)of 3.0-4.5 hours,the National Institutes of Health Stroke Scale(NIHSS)score before thrombolysis,serum Creactive protein(CRP),procalcitonin(PCT),neutrophil-to-lymphocyte ratio(NLR),matrix metalloproteinase-9(MMP-9),Lipoprotein associated phospholipase A2(Lp-PLA2),soluble lectin like oxidized low-density lipoprotein receptor-1(sLOX-1)levels in the HT group were higher than those in the non-HT group,while peripheral blood platelet and uric acid levels were lower than those in the non-HT group(P<0.05).AF,diabetes,ONT,NIHSS score before thrombolysis,peripheral blood platelet,uric acid levels,serum NLR,MMP-9,Lp-PLA2,sLOX-1 levels were all factors affecting the occurrence of HT after intravenous thrombolysis treatment(P<0.05).The AUC of the nomogram prediction model for predicting the occurrence of HT was 0.893(95%CI:0.862-0.925),with a sensitivity and specificity of 83.65%and 88.42%,respectively.The model had good calibration and s

关 键 词：急性大面积脑梗死 出血转化 中青年 列线图 预测模型 Lasso-Logistic回归分析 

分 类 号：R743[医药卫生—神经病学与精神病学]