| Objective:This study aimed to investigate the prediction of MACE during hospitalization in STEMI patients based on serum YKL-40,construct a nomogram prediction model,and evaluate its effectiveness.Methods:1.180 STEMI patients admitted to the Second People’s Hospital of Hefei from June 2020 to June 2022 were included in this study.General clinical baseline data,blood routine and biochemical data,and intervention-related data of hospitalized patients were collected in detail.To document whether in-hospital MACE occurred in STEMI patients.2.Machine learning random forest model was used to select important variables,and multivariate logistic regression was used to analyze the independent risk factors of in-hospital MACE in STEMI patients.3.A nomogram model was constructed,and the predictive efficacy was evaluated by C-index assessment,Bootstrap method repeated sampling 1000 times for internal validation,Hosmer-Lemeshow(H-L)fitted superiority test to assess the model fit,Brier score was calculated,and the calibration curve was drawn to evaluate the calibration of the model;receiver operating characteristic curve(ROC)was drawn to assess the efficacy of nomogram model and thrombolysis in myocardial infarction(TIMI)score in predicting in-hospital MACE after PCI in STEMI patients;decision curve analysis(DCA)and clinical impact curve(CIC)were used to assess the clinical application value of the model.Results:1.According to whether MACE occurred during hospitalization,the patients were divided into the MACE group(n=47)and the non-MACE group(n=133).The in-hospital MACE rate was 26.1%.The mean age of patients in the MACE group was65.64 ± 12.7 years older than 60.36±13.74 years in the non-MACE group(P<0.05).Patients in the MACE group had a median serum YKL-40 level of 1527.47(1206.36–1688.03)ng/d L greater than 1116.82(637.04,1535.74)ng/d L in the non-MACE group(P<0.05).Patients in the MACE group had higher high-density lipoprotein C(HDL-C),serum creatinine,uric acid,and fasting blood glucose(P<0.05)and lower hemoglobin,albumin,and left ventricular ejection fraction(LVEF)than patients in the non-MACE group(P<0.05).2.According to the random forest and multivariate logistic analysis results,we identified serum YKL-40,hemoglobin,fasting blood glucose,LVEF,and uric acid as independent risk factors for in-hospital MACE in STEMI patients.3.The nomogram was established by using the selected independent risk factors.The C-index of the nomogram model was 0.848,which was greater than the C-index of internal validation 0.832,and had better predictive power;the results of the H-L quasi-goodness test showed χ~2 = 3.124,P = 0.07715.After Bootstrap internal validation,the model calibration curve was close to the ideal model.The Brier score was 0.137,suggesting that the nomogram model predicted in-hospital MACE occurrence in acute ST-segment elevation myocardial infarction with good correlation and vital calibration with internal sampling validation.The results of DCA and CIC showed that the nomogram model had a high clinical application value.The area under the ROC curve(AUC)of the nomogram model in predicting in-hospital MACE after PCI in STEMI patients was 0.848(95% CI: 0.775-0.921)greater than that of TIMI score 0.727(95%CI: 0.643-0.812),P < 0.05.Conclusion:1.In this study,we analyzed the incidence of in-hospital MACE in STEMI patients after PCI to be 26.1% and identified serum YKL-40 as an independent risk factor for in-hospital MACE in STEMI patients after PCI.2.A nomogram model based on serum YKL-40 to predict the risk of in-hospital MACE after PCI in STEMI patients was constructed and validated according to random forest variable screening and multivariate logistic regression analysis results.This model can provide a scientific reference for predicting the occurrence of in-hospital MACE and improving the prognosis of STEMI patients after PCI. |
PDF Full Text Request