Analysis Of Risk Factors And Construction Of Risk Prediction Model For STEMI Patients Who Died In Hospital After PCI

ST-segment elevation myocardial infarction(STEMI)is usually caused by rupture,ulceration or erosion of coronary atherosclerotic plaque,causing thrombosis in one or more coronary arteries,resulting in decreased myocardial blood flow or distal thromboembolism,leading to epicardial coronary occlusion and transmural ischemia.A large number of cardiomyocytes undergo apoptosis and necrosis,serious loss of cardiomyocytes,activation of neuro-humoral regulatory mechanisms,and release of inflammatory mediators like tumor necrosis factor and so on,which ultimately lead to ventricular remodeling and cardiac insufficiency.Percutaneous coronary intervention(PCI)is an effective method to restore myocardial blood reperfusion,save ischemic myocardium,and protect cardiac function.According to current clinical guidelines,it is the primary approach to restoration of blood supply to infarct-related arteries and can significantly reduce mortality.The PEACE study showed that the frequency of primary PCI in STEMI patients has gradually increased in recent years,and the total ischemia time is shorter than in the past,however,the in-hospital and overall mortality and major complications have not been reduced accordingly.In-hospital mortality have been reported in approximately 6% of STEMI patients after PCI.Therefore,there is still room for improvement in the short-term prognosis of STEMI patients after PCI.In the past,clinicians mostly used single clinical indicators or myocardial injury markers to assess the risk and prognosis of patients.Although these indicators have certain value in predicting prognosis,they still have their limitations and are not comprehensive enough.Current guidelines encourage the use of clinical scoring systems,such as the Global Registry of Acute Coronary Events(GRACE)and the Thrombolysis in Myocardial Infarction(TIMI)risk score.However,data on STEMI patients with in-hospital deaths after PCI are still limited.The purpose of this study was to analyze the clinical characteristics of STEMI patients with in-hospital death after emergency PCI,to obtain independent risk factors for in-hospital death,to construct a prediction model of in-hospital death risk,and to compare with the currently commonly used scoring system to evaluate its prognosis value.In order to provide clinical guidance and improve the prognosis of STEMI patients.Part One Analysis of clinical characteristics of 297 STEMI patients who died in hospital after PCIObjective: In this study,we retrospectively analyzed the clinical characteristics of 297 STEMI patients who died in hospital after PCI,and analyzed their relationship with outcomes,so as to further study the prognosis of these patients.Methods: A total of 297 STEMI patients with in-hospital deaths who underwent PCI in 39 hospitals in Hebei Province from January 1,2018 to December 31,2019 were included in this study.The random number table method was used to select STEMI patients without in-hospital deaths.There were 779 patients in total.The differences in baseline data between the death group and the survival group were compared,and the clinical characteristics of the death group were analyzed.Results:1.Baseline data analysis: Compared with surviving patients,in-hospital death patients were older,had lower BMI,lower systolic and diastolic blood pressure on admission,and longer total ischemic time,lower proportion of males,higher proportion of Killip III-IV,higher proportion of cardiopulmonary resuscitation,higher proportion of prehospital use of ventilator and temporary pacemaker,and proportion of malignant arrhythmia high,short hospital stay;2.Analysis of past medical history: Compared with surviving patients,in-hospital mortality patients had higher proportions of history of hypertension, diabetes and PCI,higher history of coronary heart disease,higher history of atrial fibrillation,higher proportions of renal insufficiency history and family history of coronary heart disease,and lower proportion of smoking historically;3.Coronary angiography and PCI data analysis: Compared with the surviving patients,the death group had a higher proportion of long lesions and B2-C lesions,and a higher proportion of complications such as bleeding,coronary perforation,no-reflow,and dissection.The proportion of temporary pacemaker,ventilator,IABP and glycoprotein IIb/IIIa drugs during PCI is higher,and the culprit vessel is mostly anterior descending artery;4.Analysis of drug use in hospital: Compared with the patients in the survival group,the application ratios of ticagrelor/clopidogrel and ACEI/ARB in the death group were lower;5.Analysis of laboratory and ultrasound indicators: Compared with the survival group,the death group had lower hemoglobin,total cholesterol and triglyceride on admission,higher white blood cell,neutrophil-lymphocyte ratio,and systemic immune inflammatory index,and higher blood glucose levels at admission.The creatinine level was higher,the peak of CK-MB was higher,and the postoperative LVEF was lower.Summary:This study conducted a descriptive analysis of the clinical characteristics of STEMI patients who died in hospital after emergency PCI,in order to lay a foundation for screening independent risk factors for in-hospital deaths and for clinicians to identify high-risk patients.Part two Analysis of independent risk factors for in-hospital death in STEMI patients after PCIObjective: Univariate and multivariate Logistic regression analysis was performed on STEMI patients with in-hospital death after emergency PCI,to explore the independent risk factors of in-hospital death in STEMI patients after emergency PCI,and to deduce the risk score calculation formula for in-hospital death.Methods: This part of the study will include the 1076 patients in the first part.According to the ratio of 2:1,they were randomly divided into the training set and the testing set,with 717 cases in the training set and 359 cases in the testing set.Univariate Logistic regression analysis was used to screen out the potential risk factors for in-hospital death in STEMI patients after emergency PCI in the training group,and then multivariate Logistic regression analysis was used to determine the independent risk factors for in-hospital death in STEMI patients after emergency PCI.The proportion of regression coefficients in the regression model was used to assign scores to independent risk factors(where the presence and absence of dichotomous variables were assigned as 1 and 0 points,respectively),and the risk prediction formula for in-hospital death was obtained.Results:1.Univariate Logisitic regression analysis showed,older,lower BMI,lower admission systolic and diastolic blood pressure,longer total ischemic time,female,Killip grade III-IV,cardiopulmonary resuscitation,previous hypertension,diabetes,atrial fibrillation,coronary heart disease,history of renal insufficiency,smoking history,long lesions,use of temporary pacemaker,IABP,ventilator,residual stenosis after operation,no ACEI/ARB drugs before hospital,low admission hemoglobin,total cholesterol and triglyceride levels,high neutrophil-to-lymphocyte ratio and systemic immune inflammatory index,high random blood glucose,high creatinine level,high peak CK-MB,and low postoperative LVEF are potential risk factors for in-hospital death in STEMI patients after emergency PCI.2.Multivariate Logisitic regression analysis showed that age(OR=1.07,95%CI: 1.030-1.125,P<0.001),total ischemic time(OR=1.01,95%CI:1.005-1.012,P<0.001),HGB(OR=0.98,95%CI: 0.955-0.999,P<0.001),NLR(OR=1.14,95%CI: 1.060-1.229,P<0.001),admission random blood glucose(OR=1.23,95%CI: 1.085-1.401,P<0.001),postoperative LVEF(OR=0.88,95%CI: 0.834-0.913,P<0.001),CK-MB peak(OR=1.01,95%CI:1.004-1.012,P<0.001),serum total triglyceride(OR=0.52,95%CI:0.341-0.758,P<0.001),Killip grade III-IV(OR=11.29,95%CI:3.200-44.107,P<0.001),history of coronary heart disease(OR=11.83,95%CI:2.39-72.05,P=0.004),history of kidney functional insufficiency(OR=17.34,95%CI: 3.056-139.065,P<0.001),history of prehospital use of ACEI/ARB drugs(OR=0.01,95%CI: 0.002-0.035,P<0.001),intraoperative no-reflow(OR=26.90,95%CI: 4.214-210.198,P<0.001)were independent risk factors of in-hospital mortality in STEMI patients after emergency PCI.3.According to the β value of the regression model,the risk factors were assigned,and the risk prediction formula for in-hospital death after emergency PCI in STEMI patients was obtained =-1.7517 +(0.0718 * age)+(0.0083 *total ischemic time)+(0.0234 * admission hemoglobin)+(0.1329 * NLR)+(0.2062* admission random blood glucose)+(0.1333*postoperative LVEF)+(0.0079*CK-MB peak)+(0.6608*serum total triglyceride)+(2.4238*Killip III-IV grade)+(2.4705*history of coronary heart disease)+(2.8530*history of renal insufficiency)+(4.6396 *Pre-hospital history of ACEI/ARB drug application)+(3.2921* No reflow occurred during operation).Summary:1.Age,total ischemic time,HGB,NLR,admission random blood glucose,postoperative LVEF,peak CK-MB,serum total triglycerides,Killip Grade III-IV,history of coronary heart disease,history of renal insufficiency,prehospital ACEI/ARB drug application history,no reflow during operation was an independent risk factor for in-hospital death in STEMI patients after emergency PCI.2.Risk prediction formula for in-hospital death after emergency PCI in STEMI patients =-1.7517 +(0.0718 * age)+(0.0083 * total ischemic time)+(0.0234 * admission hemoglobin)+(0.1329 * NLR)+(0.2062 * random blood glucose at admission)+(0.1333 * postoperative LVEF)+(0.0079*peak CK-MB)+(0.6608*serum total triglycerides)+(2.4238*Killip grade III-IV)+(2.4705* history of coronary heart disease)+(2.8530*history of renal insufficiency)+(4.6396*prehospital history of ACEI/ARB drug use)+(3.2921 * No reflow occurred during the operation).Part Three Construction of a risk model for in-hospital mortality in STEMI patients after PCIObjective: To establish a simple and practical new model for predicting the risk of in-hospital death in patients with STEMI after emergency PCI,to provide a basis for clinicians to quickly stratify patients for risk and carry out personalized treatment,and to evaluate the application value of the predictive model to clinical prognosis.Methods: This part of the study included 1076 patients from the first part of the study.According to the ratio of 2:1,they were randomly divided into the training set and testing set,with 717 cases in the training set and 359 cases in the testing set.A risk prediction model was established based on the independent risk factor results of in-hospital death in STEMI patients after emergency PCI obtained in the second part of this study.Use the receiver operating characteristic curve(ROC)to evaluate the accuracy of the risk model in the testing set.The area under the curve(AUC)evaluates the predictive value of the model.The predictive accuracy of the risk model was measured using the C-index(Bootstrap method,1000 runs).Use the Hosmere-Lemeshow method to test the goodness of fit of the model,and P>0.05 indicated that the goodness of fit of the model was good.The clinical utility of predictive models was analyzed using decision curves.Results:1.According to the independent risk factors of in-hospital death after emergency PCI in STEMI patients in the second part of this study,a risk prediction model of in-hospital mortality was established.2.In the training set,the C-index was 0.947,indicating that predictive models are valuable in clinical practice.As shown by the ROC curve,AUC=0.947,95% CI: 0.927-0.967.The goodness-of-fit value is 0.683,indicating a good prediction accuracy.The DCA curves of the modeling group showed that the nomogram had a high overall net benefit in predicting in-hospital mortality after PCI.3.In the testing set,the C-index was 0.892.The ROC curve showedAUC=0.892,95% CI: 0.844-0.939.The goodness of fit value is 0.462.As the DCA curve shows,the nomogram has a good discriminative power and good prediction accuracy,and achieves a good net benefit.Summary: The risk model for in-hospital death of STEMI patients after emergency PCI constructed in this study includes thirteen simple and easy-to-obtained clinical indicators.They are age,total ischemic time,admission hemoglobin,NLR,admission random blood glucose,postoperative LVEF,peak CK-MB,serum total triglycerides,Killip III-IV,history of coronary heart disease,history of renal insufficiency,history of prehospital use of ACEI/ARB drugs,and intraoperative no-reflow.It has good predictive ability and good predictive value for in-hospital death after emergency PCI.Part Four Evaluation of prognostic score in STEMI patients after PCIObjective: To compare the predictive value of different scoring systems for in-hospital mortality in STEMI patients undergoing emergency PCI,as well as their combined application value with the risk prediction model derived in the second part of this study.Methods: A total of 359 patients in the validation set of the second part of this study were included.GRACE,TIMI-STEMI,CADILLAC,PAMI,SRI risk scores,and risk scores calculated in the second part of this study were performed for each patient according to the patient’s condition at admission.According to whether patients died in hospital,they were divided into death group(99 cases)and survival group(260 cases).The predictive power of each risk score and the score combined with the risk score obtained in the second part of the study was calculated for in-hospital mortality.Draw the ROC curve,calculate the AUC,and use the AUC to analyze and compare the predictive value of each score for in-hospital mortality.Results: ROC curve analysis of five scoring systems,GRACE,TIMI-STEMI,CADILLAC,PAMI,and SRI,combined with the risk prediction models obtained in the second part of this study,respectively,predicted the occurrence of in-hospital deaths in patients,and the AUCs were0.831,0.841,0.738,0.791,0.735 and 0.982,0.982,0.979,0.980,0.979 respectively,the combined prediction ability is higher than the ability of single application.Summary: Five scoring systems,GRACE,TIMI-STEMI,CADILLAC,PAMI,and SRI,all have good predictive value for in-hospital death in STEMI patients after PCI,but which scoring system has better predictive value remains unresolved.After combining with the risk prediction model obtained in the second part of this study,the predictive power of each was better than that of scoring alone.Conclusion:1.Age,total ischemic time,HGB,NLR,admission random blood glucose,postoperative LVEF,peak CK-MB,serum total triglycerides,Killip III-IV,history of coronary heart disease,history of renal insufficiency,prehospital ACEI/ARB drug application history,no reflow during operation was an independent risk factor for in-hospital death in STEMI patients after emergency PCI.2.The risk model for in-hospital death of STEMI patients after emergency PCI constructed in this study includes thirteen simple and easy-to-obtained clinical indicators.It has good predictive ability and good predictive value for in-hospital death after emergency PCI.3.Five scoring systems,GRACE,TIMI-STEMI,CADILLAC,PAMI,and SRI,all have good predictive value for in-hospital death in STEMI patients after PCI,but which scoring system has better predictive value remains unresolved.After combining with the risk prediction model obtained in the second part of this study,the predictive power of each was better than that of scoring alone.