Establishment of risk prediction model for acute myocardial infarction in elderly patients with hyperuricemia / 中华老年医学杂志

ABSTRACT

Objective:To establish a risk prediction model of acute myocardial infarction(AMI)in elderly patients with hyperuricemia(HUA).Methods:A total of 148 elderly patients with HUA in Shanghai Fourth People's Hospital Affiliated to Tongji University for prospective study from March 2020 to March 2021 were divided into AMI group and non-AMI group according to whether they were complicated with AMI.The baseline data of patients were recorded at admission and compared between the two groups.Multivariate Logistic regression analysis was used to examine the risk factors for AMI in elderly patients with HUA.And AMI risk prediction model was established.The ROC curves were plotted.Using the ROC curve, the prediction efficiency of AMI risk prediction model was evaluated.Results:Among 148 elderly patients with HUA, 22 cases(14.86%)had AMI.The proportion of smoking history in AMI group was higher than non-group( χ2=6.192, P=0.013). The diastolic blood pressure, systolic blood pressure, fasting blood glucose(FPG), low density lipoprotein cholesterol(LDL-C), serum uric acid(SUA)and homocysteine(Hcy)were statistically significantly higher in AMI group than in non-AMI group( U=1.973, 2.652, 2.016, 1.364, 2.151, 2.973, 5.972, P=0.049, 0.008, 0.044, 0.032, 0.003, <0.001). The multiple Logistic regression analysis showed that smoking history, high DBP, high SBP, high FPG, high LDL-C, high SUA and high Hcy were risk factors for AMI in elderly HUA patients( OR=3.120, 1.072, 1.887, 2.358, 3.365, 1.031, 3.773, P=0.016, 0.001, <0.001, 0.008, 0.028, 0.001, <0.001). By substituting the risk factor index and the regression coefficient into the Logistic regression equation formula, the obtained final fitting equation was AMI=1.138×smoking history(no=0, yes=1)+ 0.070×DBP+ 0.635×SBP+ 0.858×FPG+ 1.213×low density lipoprotein+ 0.030×SUA+ 1.328×Hcy.The ROC curve was drawn and showed that the AUC of forecast value of AMI risk prediction model in predicting the risk of AMI in elderly HUA patients was 0.847, >0.80, which had certain predictive value.Hosmer-Leme show goodness of fit test χ2=5.743, P=0.676, P>0.05, the model had good correction ability and satisfactory fitting. Conclusions:AMI in elderly hyperuricemia patients are affected by multiple risk factors.The AMI risk prediction model is established via combining with multiple risk factors.The model has certain efficiency in predicting the occurrence of AMI in elderly patients with hyperuricemia.