Information-Theoretic Characterization of Short-Term Heart Rate Variability in Ischemic and Non-Ischemic Heart Failure Patients During Exercise

Abstract

Heart failure with reduced ejection fraction (HFrEF) remains a major cause of morbidity and mortality worldwide. The underlying etiology, ischemic (IHF) or nonischemic (NIHF), influences disease course, autonomic control and therapy response. This study investigates whether nonlinear information-theoretic heart rate variability (HRV) metrics can reveal etiology-specific differences in cardiac autonomic dynamics under exercise stress overlooked by traditional methods. Fifty-two transplant-listed HFrEF patients underwent electrocardiogram recordings at rest and during graded exercise. RR interval time series were derived and HRV features computed across three phases: rest, early, and late exercise. Metrics included conventional time- and frequency-domain indices and entropy-based measures of complexity, Conditional Entropy (CE) and Self-Entropy (SE). Traditional HRV indices showed expected exercise-related trends but no group differences. In contrast, information-theoretic metrics revealed distinct autonomic signatures: NIHF patients displayed higher CE and lower SE during late exercise, indicating greater HRV complexity and reduced predictability. In conclusion, information-theoretic HRV analysis revealed differences missed by conventional methods, possibly enhancing risk assessment and understanding of autonomic dysfunction in HFrEF.