Frequency domain analysis of heart rate variability in horses at rest and during exercise

The pattern of variation in heart rate on a beat-to-beat basis contains information concerning sympathetic (SNS) and parasympathetic (PNS) contributions to autonomic nervous system (ANS) modulation of heart rate (HR). In the present study, heart period (RR interval) time series data were collected at rest and during 3 different treadmill exercise protocols from 6 Thoroughbred horses. Frequency and spectral power were determined in 3 frequency bands: very low (VLF) 0-less than or equal to 0.01, low (LO) >0.01-less than or equal to 0.07 and high (HI) >0.07-less than or equal to 0.5 cycles/beat. Indicators of sympathetic (SNSI = LO/HI) and parasympathetic (PNSI = HI/TOTAL) activity were calculated. Power in all bands fell progressively with increasing exercise intensity from rest to trot. At the gallop VLF and LO power continued to fall but III power rose. SNSI rose from rest to walk, then fell with increasing effort and was lowest at the gallop. PNSI fell from rest to walk, then rose and was highest at the gallop. Normalised HI power exceeded combined VLF and LO power at all gaits, with the ratio HI to LO power being lowest at the walk and highest at the gallop. ANS indicators showed considerable inter-horse variation, and varied less consistently than raw power with increasing physical effort. In the horses studied, the relationship between power and HR changed at exercise intensities associated with heart rates above approximately 120-130 beats/min. At this level, humoral and other non-neural mechanisms may become more important than autonomic modulation in influencing heart rate and heart rate variability (HRV). HRV at intense effort may be influenced by respiratory-gait entrainment, energetics of locomotion and work of breathing. HRV analysis in the frequency domain would appear to be of potential value as a noninvasive means of assessing autonomic modulation of heart rate at low exercise intensities, only. The technique may be a sensitive method for assessing exercise response to experimental manipulations and disease states.