| Objectives To create a single-center acute non-ST-segment elevation myocardial infarction,to explore its clinical characteristics,influencing factors of outcome and to construct a predictive scoring system to predict the prognosis of NSTEMI patients,so as to provide individualized and effective predictive tools for clinical prediction of prognosis in patients with NSTEMI.Methods A total of 497 patients who were definitely diagnosed NSTEMI in Hebei general hospital from January 2017 to September 2019 were enrolled.Collect baseline data,laboratory examination and treatment.Average follow-up time 26.86 ± 11.89 months,end-point event is MACE.Using R language,univariate analysis 、 Cox proportional risk regression model and other methods,the independent influencing factors of the occurrence of MACE were screened out,and build an nomogram.Area under the curve(AUC)was calculated to evaluate the model’s merits and weaknesses,and C-index was used to evaluate the consistency.Stratified the model for risk,compare the model with GRACE score,get the value of the prediction model in this study and GRACE scoreResults 1 NSTEMI clinical features: 497 patients enrolled in NSTEMI,average age is66.53±12.51,345 males;mean follow-up time was 26.86±11.89 months,a total MACE87 patients occurred during follow-up.2 The study population was divided into training set and external validation set.The training group was divided into two groups according to whether the end point event occurred during the follow-up period,and the differences between the two groups were compared.It was suggested that there were differences between the two groups in gender,age,smoking,Killip > grade 2,GRACE score,high risk and high risk,loop diuretics,euction fraction,albumin,uric acid,phosphorus,hemoglobin and in-hospital PCI.3 According to univariate and multivariate Cox analysis and other statistical methods,gender,age,hospital invasion,creatinine,ejection fraction,serum potassium,high-density lipoprotein,uric acid and Killip>grade 2 were the influencing factors for the occurrence of no MACE event in patients with NSTEMI at 1and 2 years,and the Nomogram was established,the C-index of the line diagram is 0.711.4 Compared to the GRACE score,the AUC of this model is higher in 1 year(0.719vs0.736)and 2 years(0.702vs0.743);C-index also outperformed the GRACE score(0.711vs0.582).5 Risk stratification of predictive model patient scores using k-means clustering,compared with the GRACE score risk stratification.The survival curve of the training set showed that the survival prognosis of the prediction model was different among the three levels.Analysis of Cox risk ratio model showed that HR=2.534(95%CI:1.445～4.441,P=0.001)in the high-risk versus low-risk groups;GRACE scores had no statistically significant difference in survival outcomes between the three levels;External validation set survival curves show that there are differences in survival and prognosis between the three layers.Analysis of Cox risk ratio model shows that: High-risk layer compared to low-risk layer,HR=9.141(95%CI: 2.655～31.472,P<0.001);GRACE scores had no statistically significant difference in survival outcomes between the three levels.6In the training set and the external validation set,whether to MACE events as the dependent variable,forecast model for risk score,GRACE score on the single factor and multi-factors COX regression as independent variables,the prediction model of HR=1.012(95%CI: 1.003～1.021,P=0.009),GRACE score HR=1.003(95%CI:0.996～1.009,P=0.425);External validation set results suggest prediction model HR =1.023(95%CI: 1.004～1.041,P=0.014),Grace score HR=0.998(95%CI: 0.984～1.012,P=0.796).Conclusions 1 Univariate and multivariate analysis of patients with NSTEMI revealed that gender,age,in-hospital invasion,creatinine,ejection fraction,serum potassium,high-density lipoprotein,uric acid,Killip > grade 2 were 9 independent influencing factors for the occurrence of MACE events outside the hospital;2 Based on the above 9influencing factors,the survival prediction models for NSTEMI patients without MACE events in 1 year and 2 years were obtained in this study,with good predictive efficiency and consistency(1-year AUC=0.736,2-year AUC=0.743,C-index =0.711);3 The nomogram prediction model has good differentiation and higher predictive value than GRACE score,which can provide an individualized and effective predictive tool for clinical decision-making and prognosis prediction of NSTEMI patients.Figure 5;Table 11;Reference 112... |
PDF Full Text Request