Session S82.3

Linear and Nonlinear Measures of Heart Rate Variability as Risk Stratifiers in Heart Failure Patients

A Voss*, R Schroeder, M Vallverdú, I Cygankiewicz,
R Vázquez, A Bayes de Luna, P Caminal

University of Applied Sciences
Jena, Germany

Heart failure is a major and growing public health concern affecting about 23 million people worldwide, particularly in the industrialized countries with ageing populations. Results from the Framingham study revealed five-year survival rates between 62% and 75% in men and between 38% and 42% in women. In various studies several non-invasive approaches for risk stratification of patients with heart failure were developed to identify patients at high risk of cardiac death who would benefit from a preventive therapy. However, most of these applied methods could not contribute considerably to such risk stratification.
The aim of this study was to investigate whether measures from linear (according to the Task Force recommendations) and from nonlinear (detrended fluctuation analysis and symbolic dynamics) heart rate variability (HRV) analysis enhance risk prediction in patients with heart failure.
In the scope of the Spanish multicenter study MUSIC2 509 heart failure (HeF) patients were enrolled. These patients could be separated in two subgroups: low risk (LR: stable condition, N=459) and high risk group (HR: cardiac death, N=50). From every patient a 24h long-term ECG was recorded. From the extracted heart rate time series HRV measures were calculated applying standard HRV analysis, DFA and SD.
Two parameters with a significance value p<0.001 from DFA (alpha1) and SD (averaged 5 symbols probability) and five measures from standard HRV (frequency domain: LFn, HFn, LF/HF) and from SD (high variability index, single word distribution) with a significance level p<0.05 seem to be suitable for an enhanced risk classification in HeF patients. Comparable results with only slight decreased significant values were achieved considering only subgroups with ischemic heart failure (LR: N=221, HR: N=35).
In conclusion, these results show that HRV measures from frequency domain and nonlinear dynamics contribute to an enhanced risk stratification in heart failure patients probably independently from the origin of heart failure.

(Abstract Control Number: 178)