预测STEMI患者梗死相关血管内红色血栓形成风险的列线图模型的构建  

Construction of nomogram model to predict risk of generating red thrombus in IRA of patients with STEMI

作　　者：赵文博 楚英杰 董淑娟 李静超 余海佳 韩永梅 张路遥 ZHAO Wen-bo;CHU YING-jie;DONG Shu-juan;LI JING-chao;YU HAI-jia;HAN YONG-mei;ZHANG LU-yao(Department of Cardiology,People’s Hospital of Zhengzhou University,Henan Provincial People’s Hospital,Zhengzhou 450003,China)

机构地区：[1]郑州大学人民医院,河南省人民医院心血管内科,郑州450003

出　　处：《医药论坛杂志》2022年第23期1-6,共6页Journal of Medical Forum

基　　金：河南省重点科技攻关计划项目(122102310068)。

摘　　要：目的探究ST段抬高型心肌组织梗死(ST-segment elevation myocardial infarction,STEMI)患者梗死相关血管(infarct-related artery,IRA)内形成红色血栓的预测因素,并构建预测STEMI患者IRA内红色血栓形成风险的列线图模型。方法收集60例2021年5月—2021年12月在河南省人民医院接受急诊经皮冠状动脉介入治疗(percutaneous transluminal coronary intervention,PCI)治疗及光学相干断层成像(optical coherence tomography,OCT)技术观察IRA内血栓类型的STEMI患者,根据梗死相关血管内血栓类型,分为红色血栓组29例和白色血栓组31例,比较两组临床资料。经由logistic逐步回归筛选出非零系数的预测因素。以预测因素构建logistic模型并检验拟合优度。以预测因素建立STEMI患者IRA内红色血栓形成风险的列线图模型,采用C-指数评估预测模型的预测效能。内部模型验证采用引导方法进行验证,绘制校准曲线,评估模型的校准程度;绘制ROC曲线分析列线图模型预测STEMI患者IRA血管内红色血栓形成风险的预测效能。结果以logistic逐步回归筛选出门囊时间(OR=1.226,95%CI:1.005~1.535)、男性(OR=10.031,95%CI:1.457~104.009)、血清脑利钠肽(OR=0.986,95%CI:0.968~0.999)、血清谷草转氨酶(OR=0.991,95%CI:0.985~0.996)、血清总胆固醇(OR=1.145,95%CI:1.015~1.380)、血清高密度脂蛋白胆固醇(OR=0.001,95%CI:0.001~0.221)、白细胞计数(OR=1.289,95%CI:0.983~1.747)、淋巴细胞计数(OR=2.271,95%CI:1.005~6.086)、单核细胞计数(OR=0,95%CI:0~0.019)、室壁运动异常(OR=9.913,95%CI:1.935~75.594)为预测因素,构建logistic模型并进行拟合优度检验,良好(χ2=6.774,df=8,P=0.566)。构建列线图模型,C-指数为0.852(95%CI:0.774~0.930)。列线图模型预测STEMI患者IRA内形成红色血栓的风险的AUC(0.841)。结论基于门囊时间、男性、血清脑利钠肽、血清谷草转氨酶、血清总胆固醇、血清高密度脂蛋白胆固醇、白细胞计数、淋巴细胞计数、单核Objective To explore the predictors of generating the red thrombus in IRA of patients with ST-segment elevation myocardial infarction(STEMI),and to construct a nomogram model for predicting the risk of generating the red thrombus in IRA of patients with STEMI.Methods We consecutively reviewed 60 patients of STEMI who underwent the PCI and OCT in emergency,we observed the thrombosis which was attached to the vascular wall in IRA through the OCT.They were divided into red thrombus group(29 cases)and white thrombus group(31cases).Moreover,through comparing it by analysis of the clinical date,Stepwise logistics regression was used to screen out predictive factors with non-zero coefficients,and using the predictive factors to construct a logistics model and testing the goodness of fit.Using the predictive factors to construct a nomogram model to predict the risk of generating the red thrombus in IRA of patients with STEMI,and the C-index was used to assess the predictive performance of the nomogram model.we used bootstrap method to verify the model internally and drawn the calibration curve to evaluate the calibration.ROC curve was drawn to analyze the predictive performance of nomogram model on the risk of generating the red thrombus in IRA of patients with STEMI.Results Door to balloon time(OR=1.226,95%CI:1.005-1.535),male(OR=10.031,95%CI:1.457-104.009),BNP(OR=0.986,95%CI:0.968-0.999),AST(OR=0.991,95%CI:0.985-0.996),TC(OR=1.145,95%CI:1.015-1.380),HDL_C(OR=0.001,95%CI:0.001-0.221),WBC(OR=1.289,95%CI:0.983-1.747),lymphocyte(OR=2.271,95%CI:1.005-6.086),monocyte(OR=0,95%CI:0-0.019),the wall motion abnormality(OR=9.913,95%CI:1.935-75.594)were the predictors of generating the red thrombus in IRA of patients with STEMI.Through constructing the logistic model and testing the goodness of fit,the model is fine(χ2=6.774,df=8,P=0.566).Through the predictors,we constructed the nomogram.The model displayed good predictive performance with a C-index of 0.852(95%CI:0.774-0.930).The model of calibration curve and the ideal model ar

关 键 词：经皮冠状动脉介入治疗 血栓形成 光学相干断层成像 列线图模型 

分 类 号：R54[医药卫生—心血管疾病]