38 Heart rate variability as a predictor of racing performances in standardbred horses

• Published in: Journal of equine veterinary science 2023-05, Vol.124, p.104340, Article 104340
• Main Authors: Vasseur, M., Lepers, R., Julliand, S., Langevin, N., Martin, A.
• Format: Article
• Language: English
• Subjects: Cardiac Monitoring; Fitness; Trotter
• ISSN: 0737-0806; 1542-7412
• DOI: 10.1016/j.jevs.2023.104340

Heart Rate Variability (HRV) is a non-invasive tool for assessing the autonomic regulation of heart rate through variations in RR-intervals. It reflects the control of parasympathetic and sympathetic autonomic functions in the cardiovascular system. In humans, HRV is used to adjust exercise training for improving performance. Also, some HRV parameters have been correlated to running performance and fitness status. In horses, although HRV has been used to assess behavioral response to stress, little is known regarding its correlation with performance. This experiment aimed to investigate if resting values of HRV correlate to racing performance in Standardbred horses. Forty-eight horses (27 stallions, 4 geldings and 17 mares; 2 to 7 years old; 2 to 95 races run) from Hunter Valley stables were followed over one year. Electrocardiograms were recorded twice (6 mo apart) using heart rate monitors Equimetre (Arioneo, France) for 10 min in the early morning when horses were in individual stalls. Horses did not undergo intense exercise the day before data collection. Analysis of HRV parameters was performed on Kubios HRV software to determine Standard Deviation (SDNN), Root Mean Square of Successive Differences (RMSSD), number of successive intervals differing more than 50ms (pNN50) and Standard Deviation from Poincaré plot (SD1 and SD2). Racing performances (total earnings, earnings perrace, racing time) were collected from LeTrot database for each horse over the whole career, and close to the day of measurement (one month before to 2 mo after). Performance and HRV parameters were correlated using Pearson coefficients, and linear regressions were performed to predict racing performances. Best time in career and close to HRV measurement were negatively correlated to SDNN, RMSSD, pNN50, SD1 and SD2 (P < 0.01; −0.46P < 0.03; 0.27 > r > 0.35). Total earnings and earnings per race during the 3 mo close to HRV measurement were correlated with RMSSD, pNN50 and SD1 (P < 0.05; 0.29 > r > 0.52). Prediction of total earnings and earnings per race from the HRV parameters within 3 mo close to the measure was modest (adjusted R2 = 0.31 and 0.24; P = 0.0004 and 0.0016 respectively). In human athletes, higher HRV parameters indicate greater parasympathetic tone, better training adaptation and increased recovery. In our study, horses with higher HRV parameters possibly also recovered better, as the trainer decided to make them race more (P < 0.01; R > 0.36). Further research focu