ON THE FRACTAL NATURE OF HEART-RATE-VARIABILITY IN HUMANS - EFFECTS OF RESPIRATORY SINUS ARRHYTHMIA

The purpose of the present study was to investigate the basic fractal nature of the variability in resting heart rate (HRV), relative to that in breathing frequency (BFV) and tidal volume (TW), and to test the hypothesis that fractal HRV is due to the fractal BFV and/or TW in humans. In addition, the possible fractal nature of respiratory volume curves (RVC) and HRV was observed. In the first study, eight subjects were tested while they sat quietly in a comfortable chair for 60 min. Beat-to-beat R-R intervals, i.e., HRV, and breath-by-breath BFV and TW were measured. In the second study, six subjects were tested while they were in the supine position for 20-30 min. The RVC was monitored continuously together with HRV. Coarse-graining spectral analysis (Yamamoto, Y., and R. L. Hughson, Physica D 68: 250-264, 1993) was applied to these signals to evaluate the percentage of random fractal components in the time series (%Fractal) and the spectral exponent (beta), which characterizes irregularity of the signals. The estimates of beta were determined for each variable only over the range normally used to evaluate HRV. Values for %Fractal and beta of both BFV and TW were significantly (P < 0.05) greater than those for HRV. In addition, there was no significant (P > 0.05) correlation between the beta values of HRV relative to either BFV (r = 0.14) or TW (r = 0.34). RVC showed a smooth oscillation as compared with HRV; %Fractal for RVC (42.3 +/- 21.7%, mean +/- SD) was significantly (P < 0.05) lower than that for HRV(78.5 +/- 4.2%). The %Fractal values in high-frequency ( > 0.15 Hz) bands were also significantly (P < 0.05) lower for RVC than for HRV. These results indicated that fractal components of HRV were probably not due to the respiratory modulation of heart rate.