Oertzenweg 19 b
Tel.+49 30 838 62299 Fax.+49 30 838 62529
Heart rate variability is a noninvasive parameter to assess autonomic nervous system activity. In human medicine HRV became important in different sectors. In cardiology it is established to predict probability of survival for patients after myocardial infarction. It is also applied to evaluate diabetic neuropathy and it its established in psychoanalytic medicine and sports medicine to avoid chronic fatigue syndrome. In equine science HRV has been established basically to evaluate stress in horses in different situations that might be stressful such as transport, competitions, training methods and different husbandry conditions. Since it has been proven as a reliable parameter, HRV entered equine medicine to assess different stages of pain or to illustrate dysautonomia in nerve diseases like Equine Grass Sickness.
Objectives of this study were to figure out to what extent HRV in horses with colic is different and whether a conclusion can be drawn regarding predictability of survivability and the way the horse should be treated. In addition parameter of the humoral system were evaluated including well-tried cortisol and non-established β-endorphines.
For that purpose 43 horses admitted to equine hospital because of colic were distributed in 3 groups (group 1 = surgical, group 2 = conservative, group 3 = euthanized) and have been examined at 3 different measuring dates (day 1 = day of delivery, day 2 = day after delivery, day 3 = day of discharge). Blood samples were taken to be examined amongst others for β-endorphines and cortisol and HRV was determined out of 5-minutes-sequences of ECG records via time and frequency domain analysis. β-endorphin concentration was determined by ELISA and cortisol concentration was determined by high performance liquid chromatography. As time related parameters mean NN-interval (MeanNN), their standard deviation (SDNN) and absolute and relative number of pairs of adjacent NN intervals differing by more than 50 ms in the entire recording (NN50, pNN50) were calculated. As frequency related parameters high frequency power (HF, parasympathetic activity) and low frequency power (LF, mainly sympathetic activity) of spectral analysis were calculated and their ratio LF/HF.
On day 1 the patients out of the surgical and euthanized groups showed significant lower HF-values and significant higher LF-values and a significant higher LF/HF-ratio than the patients out of the conservative group. HF-power increased significantly during the course of the hospitalization in the surgical group while LF-power and LF/HF-ratio significantly decreased. On day 3 there was no difference between horses regarding HF-power, LF-power and LF/HF-ratio. Significant differences concerning NN50 and pNN50 were present only on day 2, when the horses out of group 3, that have been euthanized because of financial reasons for the main part, were no longer part of this study. At this juncture it was conspicuous that horses that had an infaust prognosis due to severe diseases like rupture of stomach or spleen NN50 and pNN50 were “0” already at the day of admission. Cortisol values of the conservative treated horses out of group 2 were significantly lower than the cortisol values of the horses out of the surgical and euthanized groups on day 1 and 2. During the course of the hospitalization cortisol decreased in all groups. There was a significant positive correlation between HRV parameters among each other and between cortisol, heart rate, packed cell volume, lactate and LF/HF ratio. Regarding β-endorphines there were no significant differences between the groups nor between the different measuring dates and there were no significant correlations between β-endorphines, cortisol and HRV parameters.
HRV as a sensitive non-invasive parameter and cortisol is, unlike β-endorphines, suitable to assess pain and stress in horses with colic, to serve as decision support concerning the therapy (surgical vs conservative) and to predict the outcome.
To pursue the question of whether the time domain parameters NN50 and pNN50, which is the number of pairs of adjacent NN intervals differing by more than 50 ms in the entire recording and NN50 count divided by the total number of all NN intervals, are consistently “0” in colicky horses with unfavorable prognosis further studies with higher caseloads are requested. In that case, horses could be saved further treatments and owners could spare costs.