经皮冠状动脉介入治疗术后发生主要不良心血管事件的危险因素分析和Nomogram模型的构建

Risk factors of major adverse cardiovascular events and establishment of Nomogram model after percutaneous coronary intervention

目的研究急性ST段抬高型心肌梗死(STEMI)患者行经皮冠状动脉介入治疗(PCI)术后主要不良心血管事件(MACE)发生的独立危险因素并构建其Nomogram模型。方法应用病例对照研究设计,于2013年8月至2018年7月因急性ST段抬高型心肌梗死至苏州大学附属第一医院行PCI患者875例纳入本研究。收集患者的基本信息、实验室检查结果、心脏超声检查结果等指标。按照MACE的发生情况分为病例组与对照组。使用R程序中Logistics回归分析PCI术后出现MACE发生的独立危险因素,并建立Nomogram模型。应用Calibration曲线和受试者工作曲线(ROC)分析模型预测效能。结果与对照组相比,病例组患者脑卒中/短暂性脑缺血发作(TIA)和高血压的比例更高,心功能更差,病变血管数目更多,年龄更高;两组的肾小球滤过率(eGFR)、谷草转氨酶(AST)、白细胞(WBC)、血清肌酐(Cr)、超敏C反应蛋白(hs-CRP)、室上性心动过速和心室颤动(室颤)发生率、肌酸激酶同工酶(CK-MB)、左室收缩末期内径(LVESd)的差异均有统计学意义(P<0.05)。多因素Logistic回归显示,吸烟(OR=1.67,95%CI:1.13~2.47,P=0.027)、高血压(OR=2.04,95%CI:1.36~3.95,P=0.002)、Killip分级≥3级(OR=5.48,95%CI:2.41~12.46,P<0.001)、LVESd>40 mm(OR=1.6,95%CI:1.06~2.4,P=0.103)、hs-CRP>10 mg/L(OR=1.27,95%CI:0.87~1.87,P=0.947)为PCI术后MACE发生的独立危险因素。构建Nomogram预测模型,ROC曲线分析得出的曲线下面积(AUC)为0.666(95%CI:0.614~0.717,P<0.001),模型截断值为34分,灵敏度0.574,特异度0.685(P<0.05)。结论基于R-Logistic回归显示吸烟、高血压、Killip分级≥3级、LVESd>40 mm、hs-CRP>10 mg/L为PCI术后出现不良心血管事件结局的独立危险因素,建立Nomogram风险模型可以预测患者的个体化风险,对临床诊疗决策起到一定辅助作用。

Objective To study the independent risk factors of major adverse cardiovascular events(MACE)occurred in patients with acute ST-segment elevation myocardial infarction(STEMI)after percutaneous coronary intervention(PCI),and a Nomogram model was established.Methods STEMI patients(n=875)undergone PCI were chosen by using case-control study design from the First Affiliated Hospital of Soochow University from Aug.2013 to July 2018.The baseline data and results of laboratory examinations and echocardiography were collected from the patients.All patients were divided,according MACE occurrence state,into case group and control group.The independent risk factors of MACE after PCI were analyzed by using Logistics regression analysis in R program,and a Nomogram model was established.The predictive efficacy of Nomogram model was analyzed by using Calibration curve and ROC curve.Results The percentages of patients with stroke or transient ischemic attack(TIA)and hypertension were higher,heart function was worse,number of diseases vessels were more and age was elder in case group compared with control group.The differences in estimated glomerular filtration rate(eGFR),aspertate aminotransferase(AST),white blood cell(WBC),serum creatinine(SCr),high sensitivity C-reactive protein(hs-CRP),supraventricular tachycardia(SVT),ventricular fibrillation(VF),creatine kinase-MB isoenzyme(CK-MB)and left ventricular end-systolic diameter(LVESd)had statistical significance between 2 groups(P<0.05).The results of multi-factor Logistic regression analysis showed that smoking(OR=1.67,95%CI:1.13~2.47,P=0.027),hypertension(OR=2.04,95%CI:1.36~3.95,P=0.002),Killip grade≥3(OR=5.48,95%CI:2.41~12.46,P<0.001),LVESd>40 mm(OR=1.6,95%CI:1.06~2.4,P=0.103)and hs-CRP>10 mg/L(OR=1.27,95%CI:0.87-~1.87,P=0.947)were independent risk factors of MACE occurrence after PCI.A Nomogram predictive model was established and results of ROC curve analysis showed that AUC was 0.666(95%CI:0.614~0.717,P<0.001),cutoff value of the model was 34,sensitivity was 0.574 and specificity was 0.685(P<0.05).Conclusion Based on R-Logistic regression analysis,smoking,hypertension,Killip grade≥3,LVESd>40 mm and hs-CRP>10 mg/L were independent risk factors of MACE occurrence after PCI.The establishment of Nomogram risk model can predict individualized risk in patients,and has some assistant effects on making decisions of clinical diagnosis and treatment.