Construction of a Prediction Model for Major Adverse Cardiovascular Events Complicating Coronary Non-Obstructive Myocardial Infarction Based on CysC and FAR

LI Di; JIANG Daxing; ZHU Jie

doi:10.3639/j.issn.2095-3097.2024.09.019

Abstract：Objective To construct a prediction model for major adverse cardiovascular events(MACE)myocar-dial infarction with non-obstructive coronary arteries(MINOCA)based on cystatin C(CysC)and fibrinogen albumin ratio(FAR).Methods 300 patients with MINOCA admitted to our hospital from January 2021 to December 2023 were select-ed,and were divided into the MACE group(82 patients)and the non-MACE group(218 patients)according to whether they had concurrent MACE at the 6-month follow-up.Clinical data of MINOCA patients were collected,and CysC and FAR were detected by fully automated biochemical analyzer.Single-factor and multifactor unconditional logistic regression were used to analyze the influencing factors affecting the concurrent MACE of MINOCA.A prediction model for MINOCA complicating MACE was constructed through the R language based on CysC,FAR.Consistency indices were used to assess the discrim-inative ability of the model.Calibration curves were used to assess the accuracy of the model.Decision curves were used to assess the clinical effectiveness of the model,While the Hosmer-Lemeshow test for model goodness of fit,and receiver operating characteristic(ROC)curves to ana-lyze model predictive energy efficiency.Results At 6 months of follow-up,the complication rate of MACE was 27.33％(82/300)in 300 patients with MINOCA.Univariate analysis showed that gender,age,left ven-tricular ejection fraction(LVEF),N-terminal pro B type natriuretic peptide,low-density lipoprotein cholesterol(LDL-C),blood creatinine,FIB,ALB,CysC,and FAR were associated with the complication of MACE in MINOCA(P＜0.05).Multifactorial unconditional Logistic regression showed that female,older age,high LDL-C,high CysC,and high FAR were independent risk factors for MINOCA complicating MACE,and high LVEF was an independent protective factor(P＜0.05).A regression equa-tion was established based on the independent influences of MINOCA complicating MACE[Logit(P)=-15.282+0.328 × sex-+0.074 × age-0.074 × LVEF+0.970 × LDL-C+0.435 × CysC+0.441 × FAR],the consistency index was 0.903(95％CI:0.898 to 0.907),the calibration curve was close to the ideal curve,the decision curve was higher than the extreme curve,and the Hos-mer-Lemeshow test yielded P＞0.05.The ROC curve showed that the area under the curve of the MINOCA concurrent MACE prediction model constructed on the basis of CysC and FAR predicted MINOCA concurrent MACE with an area under the curve of 0.903(95％CI:0.863 to 0.934),with a sensitivity and specificity of 0.7561 and 0.9174,respectively.Conclusion The pre-diction model for constructing MINOCA concurrent MACE based on CysC and FAR has high prediction energy efficiency for MINOCA concurrent MACE.