The value of non-invasive myocardial work parameters combined with cardiac multi-dimensional parameters in predicting the risk of coronary heart disease

Abstract

: To investigate the application value of a comprehensive model integrating non-invasive myocardial work (MW) parameters and multi-dimensional cardiac parameters in risk prediction and early screening of coronary heart disease (CHD) , a prospective study was conducted involving 141 patients with CHD. Patients were divided based on coronary angiography (CAG) results: the CHD group (n=84, coronary stenosis ≥50%) and the non - CHD group (n=57, coronary stenosis <50%). Parameters showing intergroup differences were identified by univariate analysis. Independent risk factors for CHD were determined using multivariate binary logistic regression, which were then used to construct a multi-dimensional combined prediction model. Predictive performance was evaluated using receiver operating characteristic (ROC) curves, area under the curve (AUC), sensitivity, specificity, and optimal cut-off values. The DeLong test was used to compare AUC differences between the combined model and single parameters. Univariate analysis showed that the CHD group had significantly lower levels of global constructive work (GCW), global work efficiency (GWE), global work index (GWI), global longitudinal strain (GLS), left ventricular ejection fraction (LVEF), and high-density lipoprotein cholesterol (HDL-C), and significantly higher levels of global wasted work (GWW), epicardial adipose tissue (EAT), fasting blood glucose (FBG), triglycerides (TG), triglyceride-glucose (TyG) index, Gensini score, metabolic syndrome (Mets), and smoking history compared to the non-CHD group (all P<0.05). Multivariate logistic regression confirmed that GCW, GWE, EAT, and Mets were significantly associated with the occurrence of CHD. ROC curve analysis revealed that among individual parameters, EAT had the best predictive ability (AUC=0.760). The combined model incorporating the six above-mentioned parameters achieved an AUC of 0.940 (95% CI: 0.897 – 0.982), with a sensitivity of 89.3% and specificity of 89.5%, outperforming all single parameters, with an AUC significantly higher than that of EAT.