Detecting Heart Failure Relations: A Preliminary Study Integrating HRV, LVEF, and GLS in Patients with Ischemic Heart Disease and Dilated Cardiomyopathy

Cardiovascular diseases, such as Ischemic Heart Disease (IHD) and Dilated Cardiomyopathy (DCM), collectively represent the leading cause of mortality worldwide. In both pathological conditions, patients displaying heart failure symptoms emphasize the critical need for early detection, facilitating timely and appropriate care, enhancing patient outcomes, and optimizing healthcare resources. Heart rate variability (HRV), Left ventricular ejection fraction (LVEF) and Global longitudinal strain (GLS) are prominent parameters that could allow the identification of heart failure event. Therefore, the aim of our study was to develop an interpretable model that identify the relation between the occurrence of heart failure and HRV features, as well as LVEF, GLS, sex and age in patients with IHD and DCM. The study encompassed two groups: 126 patients with heart failure (HF group) and 126 patients without it (noHF group). GLS, LVEF, and linear and non-linear HRV features were extracted for each subject. Then, the interpretable model was produced by a logistic regression algorithm considering a set of features chosen with the univariate logistic regression method. The univariate logistic regression results indicate a significative correlation between the occurrence of heart failure events and the following parameters: LVEF, age, expBeta, HFn, and LF/HF. The obtained classification accuracy of produced model was 73% and the area under the ROC curve was 0.77. These preliminary findings showed that the identified parameters may be useful for stratification of IHD and DCM subjects with a risk of a heart failure event.