Objective To investigate the independent risk factors of premature coronary heart disease(PCHD)and constructed and verified the PCHD risk prediction model to improve the poor applicability of the existing prediction models in young people.Methods A total of 1305 young patients suspected of coronary heart disease from the First Hospital of Lanzhou University from January to December 2023 were continuously collected.According to the coronary angiography results,the patients were divided into the PCHD group(745 cases)and non coronary heart disease(NCHD)group(560 cases),and the training set and validation set were randomly included in the ratio of 8:2.The baseline data of the two groups were compared.LASSO regression was used for preliminary screening of risk factors,and logistic regression analysis was performed in the training set to screen the independent risk factors for PCHD;and then the PCHD risk prediction model was constructed.Receiver operating characteristic curve(ROC)was drawn and area under the curve(AUC)was calculated.Calibration curve and decision curve analysis(DCA)were drawn for model verification and evaluation.Results A total of 22 non-zero coefficients were selected by LASSO regression.Multivariate logistic regression analysis showed that male,age,history of hypertension,diabetes,family history of hypertension,low level apolipoprotein AI(ApoAI),lipoprotein a[Lp(a)],random glucose(RG)and aspartate aminotransferase(AST)were the independent risk factors for PCHD(P＜0.05).Based on the constructed PCHD nomogram model,the ROC curve showed that the training set AUC=0.786(95％CI 0.758-0.813)and the verification set AUC=0.796(95％CI 0.742-0.849).The Hosmer-Lemeshow shows that the calibration curves of the training set(P=0.795)and the validation set(P=0.558)fitted the ideal curve well.DCA showed that when using this model for clinical decision making,the net benefit for patients was＞0 when the training set and validation set threshold probabilities were 8％-72％and 8％-60％.Conclusion In the group of young people with coronary heart disease related symptoms,the PCHD risk prediction model constructed in this study based on male,age,history of hypertension,history of diabetes,family history of hypertension,ApoAI,Lp(a),RG and AST performs well in terms of differentiation,accuracy and clinical applicability.
关键词
早发冠心病/心血管疾病/年轻人群/风险预测模型
Key words
Premature coronary heart disease/Cardiovascular diseases/Young people/Risk prediction model
维普
万方数据