Dissociation between microneurographic and heart rate variability estimates of sympathetic tone in normal subjects and patients with heart failure

The concept that spectral analysis of heart rate variability (HRV) can estimate cardiac sympathetic nerve traffic in subjects with both normal and impaired left ventricular systolic function has not been validated against muscle sympathetic nerve activity (MSNA). We used coarse-graining spectral analysis to quantify the harmonic and non-harmonic, or fractal, components of HRV and to determine low-frequency (0.0–0.15 Hz; PL) and high-frequency (0.15–0.5 Hz; PH) harmonic power. To test the hypothesis that MSNA and HRV representations of sympathetic nerve activity (PL and PL/PH) increase in parallel in heart failure, we recorded heart rate and MSNA during supine rest in 35 patients (age 52.4±2 years; mean±S.E.M.), with a mean left ventricular ejection fraction of 22±2%, and in 34 age-matched normal subjects. Power density was log10 transformed. Mean MSNA was 52.9±2.6 bursts/min in heart failure patients and 34.9±1.9 bursts/min in normal subjects (P < 0.0001). In normal subjects, but not in heart failure patients, total power (PT) (r = -0.41; P = 0.02) and fractal power (PF) (r = -0.36; P = 0.04) were inversely related to age. In heart failure patients, total and fractal power were reduced (P < 0.009 for both), and were inversely related to MSNA burst frequency (r = -0.55, P = 0.001 and r = -0.60, P = 0.0003 respectively). In normal subjects, there was no relationship between MSNA and either PL or PH. In heart failure patients, as anticipated, PH was inversely related to MSNA (r = -0.41; P < 0.02). However, PL was also inversely rather than directly related to MSNA (r = 0.44 for 1/log10 PL; P < 0.01). There was no relationship between other sympathetic (PL/PH) or parasympathetic (PH/PT) indices and MSNA in either heart failure patients or normal subjects. The lack of concordance between these direct and indirect estimates of sympathetic nervous system activity indicates that this component of HRV cannot be used for between-subject comparisons of central sympathetic nervous outflow. It is the absence of low-frequency power that relates most closely to sympathetic activation in heart failure.