TDK conference 2019

Horse keeping has undergone radical changes over the last few decades. As a result, many new stressors have been introduced into the life of horses, which also affect animal welfare, performance and the development of certain adaptive diseases. Stress can be defined as the physiological, homeostatic, and behavioural responses of a horse that result from interactions with environmental stimuli. Heart rate variability (HRV) describes the inequality between RR intervals. This variability is determined by the coordinated function of the efferent and afferent ANS branches. HRV is a commonly used tool when trying to assess the functioning of cardiac autonomic regulation.

According to our basic hypothesis, there is a correlation between the temperament and the HRV parameters. We hypothesized higher sympathetic and lower vagal activity in temperamental horses than in calm, more cooperative ones. According to our hypothesis calm horses with higher HRV are, less exposed to environmental stimuli and more able to cope with stress. A total of 51 animals, 29 horses and 22 ponies were involved in the study. We recorded the breed, age, gender and exercise level of the horse examined. The required temperament data were obtained from a modified questionnaire survey. ECG recordings were collected with KRUUSE Televet 100 - Telemetric ECG. We recorded 30 minutes of resting ECG while horses were freely moving in their individual boxes. From the 30 minutes recordings 3 minutes of data free of any artefact was selected for further HRV analysis. The analysis was performed by the Kubios software without using any filters. Correlations between HRV parameters and temperament scores were evaluated by Spearman's rank correlation test.

When evaluating all horses in a single data set we could not demonstrate any significant correlation between HRV parameters and temperaments scores. When evaluating data sets separately according to fitness levels certain correlations could be identified. In horses, with low-moderate exercise levels, temperament scores were correlated to HF values. In horses with more intense exercise levels temperament scores correlated with some HRV values (Max HR p=0.02223, Mean HR p=0.003516, Mean RR p=0.003516, Min HR p=0.01551, PNS index p=0.01214, SNS index p=0.02872, HF HZ_FFT p=0.01861. Interestingly the correlation was always inverse than expected. Horses subjectively described with calm, and cooperative temperament had higher HR values, and lower parasympathetic influence while recording the resting ECG. Our findings are consistent with previous studies where HRV data were not useful in reflecting visible behavioural types or rather the other way around subjective temperament questionnaires do not reflect the horse neurohormonal stress situation and stress coping strategies. It can be also stated that the HRV parameters of horses at different training levels are not comparable, because activity has a much greater effect on HRV than temperament.