Rhythms in the central nervous system and 1/f fluctuations of the heart rate.

We have developed a system to analyze heart rate variability (HRV) (power spectral array of the HRV) during 24 h ambulatory electrocardiographic monitoring. Several rhythms (circadian and several ultradian rhythms) were observed in the power spectral array of the heart rates and 1/f-like fluctuations in the log-log scaled heart rate power spectrum. The circadian change of the heart rate is closely related to the body temperature rhythm. The 90 min rhythm of HRV during sleep was suspected to be produced by the sleep cycle (REM/NREM) and the lower frequency peak of the HRV was coherent with oscillation in amplitude modulated respiration. These circadian and ultradian rhythm as assessed by heart rate variability exist both in normal subjects and in patients with autonomic failure. The power of the high frequency band decreases in subjects with autonomic failure. The power of low frequency components increases during periodic breathing or Cheyne-Stokes respiration. Log-log scaled analysis of the power spectrum of HRV disclosed that the slope of the HRV is markedly modulated by the range of the frequency applied for the least square regression line analysis. The increased power that might be produced by periodic breathing and decreased power in patients with autonomic failure might strongly modulate the slope of the log-log scaled HRV. It is concluded that the power spectral array of the HRV during 24 h period is useful in the detection of circadian and ultradian rhythm, and log-log scaled power spectra might be useful in the overall integration of the heart rate dynamics produced by the central nervous system. The several rhythm factors that might be produced by the central nervous system might modulate 1/f fluctuations of the HRV.