Predictive Model Of MACE In Patients With Acute Myocardial Infarction Underwent PIC

OBJECTIVE: Acute myocardial infarction(AMI)is one of serious type coronary heart disease.The purpose of this study is to establish a predictive model of major adverse cardiac events(MACE)after patients in PCI that has acute myocardial infarction,to assist clinicians in judging the prognosis of patients with AMI,and to provide reference for secondary prevention AMI patients that take PCI.METHODS: Based on the AMI database that established in the Heart Center,Zhongshan Hospital affiliated to Dalian University from December 2015 to December2018.There are 632 patients were involved in this study.We record basic information(gender,height,weight,age,BMI,admission temperature,pulse,respiration and blood pressure),electrocardiogram(ECG)information(ventricular arrhythmia originate,pathological Q wave,T wave changes and ST elevation),Previous medical history(Hypertension,diabetes,coronary heart disease,previous myocardial infarction,PCI treatment,secondary prevention,criminal blood vessels,dyslipidemia,peripheral arteriosclerosis,deep vein thrombosis,cerebral infarction,atrial fibrillation,cardiomyopathy,chronic obstructive Lung disease.History of smoking and family history of coronary heart disease),Laboratory test results(Blood routine,urine routine,liver biochemistry,renal function,blood lipids,blood sugar,glycated hemoglobin,insulin,coagulation,and insulin enzymology.Troponin and other indicators are often tested multiple times during hospitalization,we were taking the highest level during hospitalization,and the remaining indicators are obtained by taking blood within 48 hours of admission),Echocardiography(IVSD,LVDD,LVPWD,AOD,LAD,RVOT,MPA,RVDD,LASD,RASD,LVEF,ventricular wall motion)and Coronary angiography results(Chronic occlusion,culprit vessel,whether there are triple-vessel disease,TIMI blood flow grade).Review the medical record information and follow up the patient for at least one year to investigate the occurrence of major adverse cardiovascular events(MACE).MACE events are defined as recurrent myocardial infarction,heart failure,malignant arrhythmias,cardiogenic shock,and cardiogenesis.Sudden death.The C4.5 decision tree algorithm was used to build a prediction model,and a decision tree was generated with the end point of the presence or absence of MACE events,and the model was adjusted with reference to previous research results related to the predictive index.Follow up the patient for at least one year and review the medical record information to investigate the occurrence of major adverse cardiovascular events(MACE).MACE events are defined as recurrent myocardial infarction,heart failure,malignant arrhythmias,cardiogenic shock,cardiogenesis and sudden death.The C4.5 decision tree algorithm was used to build a prediction model for this study.The decision tree was generated with the end point of the presence or absence of MACE events,and the model was adjusted with reference to previous research results related to the predictive index.RESULTS: 692 patients were enrolled in this study,including 31 with incomplete data,12 lost to follow-up,and 17 with unknown cause of death.632 patients completed follow-up,aged 32-103 years,mean age(69.4 ± 13.1)years,393 males and 239 females,followed up for 12-48 months,with an average follow-up of(14.4± 3.6)months.There were 467 cases in the MACE group,including 68 cases of recurrent myocardial infarction,321 cases of heart failure,19 cases of malignant arrhythmia,13 cases of cardiogenic shock,and 46 cases of sudden cardiac death.We have established a predictive model of major adverse cardiac events in patients with acute myocardial infarction after PCI,including systolic blood pressure,diastolic blood pressure,NT-pro BNP,white blood cell count,C-reactive protein,alanine aminotransferase,cholesterol,high-density lipoprotein,Uric acid,urea nitrogen,creatinine,blood glucose,decreased ST segment of ECG,Q wave,LVDD,LASD,LVEF,previous PCI surgery history,history of coronary heart disease,family history of coronary heart disease,history of diabetes,hyperlipidemia,peripheral arteriosclerosis,and deep vein embolism,and history of atrial fibrillation.Total 25 ofindicators with 38 branch.The prediction accuracy of the model in the validation data set is 84.35%.CONCLUSIONS: This study uses computer C4.5 decision tree algorithm to comprehensively analyze the basic information,laboratory indicators,cardiac ultrasound,and coronary angiography results of patients with acute myocardial infarction after PCI.Key indicators were screened out and a decision tree prediction model for predicting major cardiovascular adverse events(MACE)within 1 year after PCI for patients with acute myocardial infarction was generated.Compared with the current mainstream predictive indicators and scoring systems,this model incorporates more indicators as predictive factors and shows good predictive performance in the test data set.After further optimization and improvement,it can be used to assess the risk of major cardiovascular events in patients after PCI.