**5.2 Analysis of HRV**

Since HRV reflects cardiac autonomic outflow, attempts have been made to assess this outflow by analyzing HRV. Time domain analysis with the use of standard deviation of R-R interval has been proposed as measures of parasympathetic activity. But this is a nonspecific quantifier of HRV and we cannot analyze the factors which produce this variability. To solve this problem, frequency domain analysis with the use of power spectrum has proven useful to sort out the variability into components which the whole variability is consisted of. In this method, the variability is mathematically transformed into frequency components, and the power of each frequency is calculated. In this way, we can understand which frequency components make up the variability and how much influence they have on the whole.

Example of a power spectrum of HRV in human is shown in Fig. 9. In human beings, three major components can be observed. One in the low frequency (LF) area of 0.04-0.15 Hz, one in the high frequency (HF) area of around 0.20 Hz and one below the LF. The LF power which is the components between 0.04-0.15 Hz in human, reflect the heart rate fluctuating at a cycle of about 10 seconds. This component is said to be the result of the Mayer wave of arterial pressure reflecting on the burst rate of the sino atrial node through baroreflex (Scher, 1977). Both the sympathetic and parasympathetic outflow are considered to regulate the LF components (Akselrod, et al., 1981; Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). The HF power which is the components between 0.15-0.40 Hz in human, derives from respiratory sinus arrhythmia (Hirsch & Bishop, 1981). The frequency of the component is this area coincides with the frequency of respiration. This component is said to be the respiratory system ad afferent signals from receptors in the lung influencing the cardiovascular system. Only the parasympathetic outflow is considered to regulate the HF components (Akselrod, et al., 1981; Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996).

Recent Advances in Telemetry Monitoring and Analysis for Laboratory Animals 153

HF power. These powers have units of msec2. The ratio of LF power and HF power (LF/HF)

All animals shared a characteristic pattern in their power spectrum analysis. Representative

power spectra of HRV in each animal species are shown in Fig. 10.

Fig. 10. Representative power spectra obtained from many animal species

were decided in each animal species as shown in Table 1.

summarized in Table 2.

There were two major spectral components of LF and HF spectra for HRV. Since the HF power is represented by the component corresponding to respiration, the range of HF was set so that the respiration rate would be included in it. As for the LF, the upper limit was set at the same frequency as the lower limit of HF. The lower limit of LF was set according to the resampling time of the R-R interval time series. In the method of fast Fourier transform, the components at very low frequencies include noise from the data analyzed and makes that part unreliable. The frequency range which includes this noise is in relation to the resampling time. With this in mind, we have set the lower limit of LF according to the limit we observed to be a reliable one. On the basis of these data, two frequency bands of interest

The values of HRV in each animal species obtained from our experiments are also

was also calculated and this is unitless.

Fig. 9. Power spectrum obtained from human
