• View in gallery

    Representative EMG activity of a Thoroughbred's left LDm, TFLm, GMm, and BFm over 4 consecutive strides during galloping on a treadmill (grade, 3%) at a constant speed (13.5 m/s) to achieve fatigue after approximately 360 seconds. For this horse, the left hind limb was the lead limb after 30 seconds of galloping and EMG activity recordings were obtained (start of gallop exercise [left column]). Recordings were obtained at 30-second intervals thereafter until the horse could not maintain its position on the treadmill despite humane encouragement. The data for the horse in the fatigued state were those obtained at the last point at which the left hind limb was again the lead limb prior to exhaustion (right column). In this horse, all electrodes remained at the surfaces of the recorded muscles, and the baseline EMG activity of all muscles is stable throughout the experiment. When electrodes were detached or EMG data were excessively noisy for any horse in the study, those data were excluded from further analysis.

  • View in gallery

    Diagram to illustrate the calculation of the iEMG value and MF for the left and right LDm, TFLm, GMm, and BFm of Thoroughbreds used in a study to quantify fatigue-induced EMG changes in their hind limb muscles. For each left or right muscle, the iEMG value (the total of the red area under the traces) for a stride (red dashed arrow) was measured at the predetermined time points during galloping and trotting and the mean of the values for 5 consecutive strides was used for statistical analysis. Fast Fourier transform was applied to EMG signals for a period of 100 milliseconds (horizontal thick arrow), and the maximum amplitude of the EMG wave (designated by the vertical arrow) was located at the midpoint of the signal region. Then, MF was calculated mathematically as the point where the Fourier profile was bisected in area. Mean MF was derived from 5 consecutive muscle discharges and used for statistical analysis.

  • View in gallery

    Typical changes in the iEMG value of a Thoroughbred's left GMm during galloping at a constant speed (14.7 m/s) on a treadmill (3% grade). For this horse, the left hind limb was the lead limb at the 30-second time point; thus, the left muscle was analyzed as the lead limb muscle (circles) and the right muscle was analyzed as the trailing limb muscle (triangles). Recordings were obtained at 30-second intervals thereafter until the horse could not maintain its position on the treadmill despite humane encouragement. The duration of galloping until exhaustion was 365 seconds for this horse. The last data collection point at which the left hind limb was the lead limb prior to exhaustion was 330 seconds; the iEMG value at that time was considered the value at fatigue.

  • View in gallery

    Changes in SF for each of 8 Thoroughbreds (represented by a different color) that were instrumented with electrodes to assess EMG activity of the left and right LDm, TFLm, GMm, and BFm during galloping on a treadmill (grade, 3%) at a constant speed (12.6 to 14.7 m/s) to achieve fatigue after approximately 360 seconds. During the gallop exercise, SF was determined every 30 seconds. The SF was calculated as the inverse of stride duration of the lead limb. Stride frequency of all horses decreased over time.

  • View in gallery

    Changes in iEMG values for the LDm, TFLm, GMm, and BFm of the lead hind limb (A) and the trailing hind limb (B) of the 8 Thoroughbreds in Figure 4 during galloping to exhaustion. A—For the lead limb, iEMG values for the GMm and BFm decreased with fatigue, whereas those of the LDm and TFLm did not change with fatigue. B—For the trailing limb, the iEMG values of the GMm and BFm decreased with fatigue, whereas those of the LDm and TFLm did not change with fatigue. Data were not available for each muscle from all horses because there was excessive signal noise or the electrodes were detached during a measurement in some horses. *For this muscle, the value at fatigue for each horse is significantly (all P ≤ 0.01) less than that at the start of galloping. †For this muscle, the value at fatigue for each horse is significantly (P ≤ 0.05) less than that at the start of galloping. See Figure 4 for key.

  • View in gallery

    Changes in iEMG of the left LDm, TFLm, GMm, and BFm muscles of the 8 Thoroughbreds in Figure 4 during trotting prior to the galloping exercise and when fatigued after the galloping exercise. For each horse, the trotting exercise was performed on a treadmill inclined to a 3% grade and involved trotting at 3.5 m/s for 3 minutes prior to the gallop exercise; after galloping to exhaustion, each horse was trotted at a speed of 3.5 m/s for 3 minutes. The iEMG values were assessed during trotting at the 60-second time point before and after galloping, and these values were compared. For all muscles, the iEMG values did not change with fatigue. Data were not available for each muscle from all horses because there was excessive signal noise or the electrodes were detached during a measurement in some horses. See Figure 4 for remainder of key.

  • 1. Yoshikawa T, Mori S, Santiesteban AJ, et al. The effects of muscle fatigue on bone strain. J Exp Biol 1994;188:217233.

  • 2. Butcher MT, Hermanson JW, Ducharme NG, et al. Superficial digital flexor tendon lesions in racehorses as a sequela to muscle fatigue: a preliminary study. Equine Vet J 2007;39:540545.

    • Search Google Scholar
    • Export Citation
  • 3. Takahashi T, Kasashima Y, Ueno Y. Association between race history and risk of superficial digital flexor tendon injury in Thoroughbred racehorses. J Am Vet Med Assoc 2004;225:9093.

    • Search Google Scholar
    • Export Citation
  • 4. Hamlin MJ, Shearman J, Hopkins W. Changes in physiological parameters in overtrained Standardbred racehorses. Equine Vet J 2002;34:383388.

    • Search Google Scholar
    • Export Citation
  • 5. Valentin S, Zsoldos RR. Surface electromyography in animal biomechanics: A systematic review. J Electromyogr Kinesiol 2016;28:167183.

    • Search Google Scholar
    • Export Citation
  • 6. Williams JM. Electromyography in the horse: a useful technology? J Equine Vet Sci 2017;60:4358.

  • 7. Clayton H, van Weeren PR. Measurement techniques for gait analysis. In: Back W, Clayton H, eds. Equine locomotion. 2nd ed. London: Saunders, 2013;3160.

    • Search Google Scholar
    • Export Citation
  • 8. Kamen G. Electromyographic kinesiology. In: Robertson DGE, Caldwell GE, Hamill J, et al. eds. Research methods in biomechanics. 2nd ed. Champaign, Ill: Human Kinetics, 2014;179201.

    • Search Google Scholar
    • Export Citation
  • 9. Vanhatalo A, Poole DC, DiMenna FJ, et al. Muscle fiber recruitment and the slow component of O2 uptake: constant work rate vs. all-out sprint exercise. Am J Physiol Regul Integr Comp Physiol 2011;300:R700R707.

    • Search Google Scholar
    • Export Citation
  • 10. Ament W, Bonga GJ, Hof AL, et al. EMG median power frequency in an exhausting exercise. J Electromyogr Kinesiol 1993;3:214220.

  • 11. Komi PV, Tesch P. EMG frequency spectrum, muscle structure, and fatigue during dynamic contractions in man. Eur J Appl Physiol Occup Physiol 1979;42:4150.

    • Search Google Scholar
    • Export Citation
  • 12. Colborne GR, Birtles D, Cacchione I. Electromyographic and kinematic indicators of fatigue in horses: a pilot study. Equine Vet J Suppl 2001;33:8993.

    • Search Google Scholar
    • Export Citation
  • 13. Cheung TK, Warren L, Lawrence L, et al. Electromyographic activity of the long digital extensor muscle in the exercising Thoroughbred horse. Equine Vet J 1998;30:251255.

    • Search Google Scholar
    • Export Citation
  • 14. Takahashi T. The effect of age on the racing speed of Thoroughbred racehorses. J Equine Sci 2015;26:4348.

  • 15. Clayton HM, Hodson E, Lanovaz J, et al. The hindlimb in walking horses: 2. Net joint moments and joint powers. Equine Vet J 2001;33:4448.

    • Search Google Scholar
    • Export Citation
  • 16. Dutto DJ, Hoyt DF, Clayton HM, et al. Joint work and power for both the forelimb and hindlimb during trotting in the horse. J Exp Biol 2006;209:39903999.

    • Search Google Scholar
    • Export Citation
  • 17. Audigié F, Pourcelot P, Degueurce C, et al. Kinematics of the equine back: flexion-extension movements in sound trotting horses. Equine Vet J Suppl 1999;30:210213.

    • Search Google Scholar
    • Export Citation
  • 18. Robert C, Audigie F, Valette JP, et al. Effects of treadmill speed on the mechanics of the back in the trotting saddlehorse. Equine Vet J Suppl 2001;33:154159.

    • Search Google Scholar
    • Export Citation
  • 19. Kawai M, Minami Y, Sayama Y, et al. Muscle fiber population and biochemical properties of whole body muscles in Thoroughbred horses. Anat Rec (Hoboken) 2009;292:16631669.

    • Search Google Scholar
    • Export Citation
  • 20. Eto D, Yamano S, Kasashima Y, et al. Effect of controlled exercise on middle gluteal muscle fibre composition in Thoroughbred foals. Equine Vet J 2003;35:676680.

    • Search Google Scholar
    • Export Citation
  • 21. Takahashi T, Matsui A, Mukai K, et al. The effects of inclination (up and down) of the treadmill on the electromyogram activities of the forelimb and hind limb muscles at a walk and a trot in Thoroughbred horses. J Equine Sci 2014;25:7377.

    • Search Google Scholar
    • Export Citation
  • 22. Nukaga H, Takeda T, Nakajima K, et al. Masseter muscle activity in track and field athletes: A pilot study. Open Dent J 2016;10:474485.

    • Search Google Scholar
    • Export Citation
  • 23. Payne RC, Hutchinson JR, Robilliard JJ, et al. Functional specialisation of pelvic limb anatomy in horses (Equus caballus). J Anat 2005;206:557574.

    • Search Google Scholar
    • Export Citation
  • 24. Yamano S, Eto D, Hiraga A, et al. Recruitment pattern of muscle fibre type during high intensity exercise (60–100% VO2 max) in Thoroughbred horses. Res Vet Sci 2006;80:109115.

    • Search Google Scholar
    • Export Citation
  • 25. Votion D-M, Navet R, Lacombe VA, et al. Muscle energetics in exercising horses. Equine Comp Exerc Physiol 2007;4:105118.

  • 26. Parsons KJ, Pfau T, Wilson AM. High-speed gallop locomotion in the thoroughbred racehorse. I. The effect of incline on stride parameters. J Exp Biol 2008;211:935944.

    • Search Google Scholar
    • Export Citation
  • 27. López-Rivero J, Serrano A, Diz A, et al. Variability of muscle fibre composition and fibre size in the horse gluteus medius: an enzyme-histochemical and morphometric study. J Anat 1992;181:110.

    • Search Google Scholar
    • Export Citation
  • 28. Wickler SJ, Greene H, Egan K, et al. Stride parameters and hindlimb length in horses fatigued on a treadmill and at an endurance ride. Equine Vet J Suppl 2006;38:6064.

    • Search Google Scholar
    • Export Citation
  • 29. Johnston C, Gottlieb-Vedi M, Drevemo S, et al. The kinematics of loading and fatigue in the Standardbred trotter. Equine Vet J Suppl 1999;31:249253.

    • Search Google Scholar
    • Export Citation
  • 30. Ronéus N, Essén-Gustavsson B, Johnston C, et al. Lactate response to maximal exercise on the track: relation to muscle characteristics and kinematic variables. Equine Vet J 1995;27:191194.

    • Search Google Scholar
    • Export Citation
  • 31. Biewener AA. Muscle-tendon stresses and elastic energy storage during locomotion in the horse. Comp Biochem Physiol B 1998;120:7387.

    • Search Google Scholar
    • Export Citation
  • 32. Minetti AE, Ardigo L, Reinach E, et al. The relationship between mechanical work and energy expenditure of locomotion in horses. J Exp Biol 1999;202:23292338.

    • Search Google Scholar
    • Export Citation
  • 33. Tokuriki M, Aoki O. Electromyographic activity of the hindlimb muscles during the walk, trot and canter. Equine Vet J 1995;27:152155.

    • Search Google Scholar
    • Export Citation
  • 34. Licka TF, Peham C, Frey A. Electromyographic activity of the longissimus dorsi muscles in horses during trotting on a treadmill. Am J Vet Res 2004;65:155158.

    • Search Google Scholar
    • Export Citation
  • 35. de Cocq P, Weeren Pv, Back W. Effects of girth, saddle and weight on movements of the horse. Equine Vet J 2004;36:758763.

  • 36. Fruehwirth B, Peham C, Scheidl M, et al. Evaluation of pressure distribution under an English saddle at walk, trot and canter. Equine Vet J 2004;36:754757.

    • Search Google Scholar
    • Export Citation
  • 37. Doud JR, Walsh J. Muscle fatigue and muscle length interaction: effect on the EMG frequency components. Electromyogr Clin Neurophysiol 1995;35:331339.

    • Search Google Scholar
    • Export Citation
  • 38. Masuda K, Masuda T, Sadoyama T, et al. Changes in surface EMG parameters during static and dynamic fatiguing contractions. J Electromyogr Kinesiol 1999;9:3946.

    • Search Google Scholar
    • Export Citation
  • 39. De Luca CJ, Gilmore LD, Kuznetsov M, et al. Filtering the surface EMG signal: Movement artifact and baseline noise contamination. J Biomech 2010;43:15731579.

    • Search Google Scholar
    • Export Citation
  • 40. Stewart D, Macaluso A, De Vito GD. The effect of an active warm-up on surface EMG and muscle performance in healthy humans. Eur J Appl Physiol 2003;89:509513.

    • Search Google Scholar
    • Export Citation
  • 41. Harrison AP. A more precise, repeatable and diagnostic alternative to surface electromyography – an appraisal of clinical utility of acoustic myography. Clin Physiol Funct Imaging 2018;38:312325.

    • Search Google Scholar
    • Export Citation
  • 42. Riis KH, Harrison AP, Riis-Olesen K. Non-invasive assessment of equine muscular function: a case study. Open Vet J 2013;3:8084.

Advertisement

Electromyographic changes in hind limbs of Thoroughbreds with fatigue induced by treadmill exercise

Yuji Takahashi DVM1, Kazutaka Mukai DVM, PhD2, Akira Matsui PhD3, Hajime Ohmura DVM, PhD4, and Toshiyuki Takahashi DVM, PhD5
View More View Less
  • 1 Sports Science Division, Equine Research Institute, Japan Racing Association, 1400-4 Shiba, Shimotsuke, Tochigi, Japan.
  • | 2 Sports Science Division, Equine Research Institute, Japan Racing Association, 1400-4 Shiba, Shimotsuke, Tochigi, Japan.
  • | 3 Sports Science Division, Equine Research Institute, Japan Racing Association, 1400-4 Shiba, Shimotsuke, Tochigi, Japan.
  • | 4 Sports Science Division, Equine Research Institute, Japan Racing Association, 1400-4 Shiba, Shimotsuke, Tochigi, Japan.
  • | 5 Sports Science Division, Equine Research Institute, Japan Racing Association, 1400-4 Shiba, Shimotsuke, Tochigi, Japan.

Abstract

OBJECTIVE To quantify fatigue-induced electromyographic changes in hind limb muscles in horses.

ANIMALS 8 Thoroughbreds.

PROCEDURES The left and right hind limb longissimus dorsi, tensor fasciae latae, gluteus medius, and biceps femoris muscles were instrumented for surface electromyography. Hoof strain gauges were attached to confirm stride cycle. Each horse was galloped on a treadmill (grade, 3%) at a constant speed (12.6 to 14.7 m/s) to achieve fatigue after approximately 360 seconds. Before and after this exercise, the horses were trotted at 3.5 m/s. At 30-second intervals during galloping an integrated electromyography (iEMG) value for a stride and the median frequency of muscle discharge (MF) in each limb were measured. The mean of stride frequency (SF), iEMG value, and MF of 5 consecutive strides at the start and end of galloping for the lead and trailing limbs were compared. For trotting, these variables were compared at 60 seconds before and after galloping.

RESULTS The mean ± SD value for SF decreased over time (2.14 ± 0.06 to 2.05 ± 0.07 stride/s). In both the lead and trailing limbs, fatigue decreased the iEMG values of the gluteus medius and biceps femoris muscles but not those of the longissimus dorsi and tensor fasciae latae muscles. The MF did not change for any muscle during galloping with fatigue. The SF, iEMG value, and MF did not change during trotting with fatigue.

CONCLUSIONS AND CLINICAL RELEVANCE Fatigue induced by high-speed galloping decreased the gluteus medius and biceps femoris muscles' iEMG values in Thoroughbreds. Fatigue of these less fatigue-resistant hind limb muscles would affect a horse's speed.

Abstract

OBJECTIVE To quantify fatigue-induced electromyographic changes in hind limb muscles in horses.

ANIMALS 8 Thoroughbreds.

PROCEDURES The left and right hind limb longissimus dorsi, tensor fasciae latae, gluteus medius, and biceps femoris muscles were instrumented for surface electromyography. Hoof strain gauges were attached to confirm stride cycle. Each horse was galloped on a treadmill (grade, 3%) at a constant speed (12.6 to 14.7 m/s) to achieve fatigue after approximately 360 seconds. Before and after this exercise, the horses were trotted at 3.5 m/s. At 30-second intervals during galloping an integrated electromyography (iEMG) value for a stride and the median frequency of muscle discharge (MF) in each limb were measured. The mean of stride frequency (SF), iEMG value, and MF of 5 consecutive strides at the start and end of galloping for the lead and trailing limbs were compared. For trotting, these variables were compared at 60 seconds before and after galloping.

RESULTS The mean ± SD value for SF decreased over time (2.14 ± 0.06 to 2.05 ± 0.07 stride/s). In both the lead and trailing limbs, fatigue decreased the iEMG values of the gluteus medius and biceps femoris muscles but not those of the longissimus dorsi and tensor fasciae latae muscles. The MF did not change for any muscle during galloping with fatigue. The SF, iEMG value, and MF did not change during trotting with fatigue.

CONCLUSIONS AND CLINICAL RELEVANCE Fatigue induced by high-speed galloping decreased the gluteus medius and biceps femoris muscles' iEMG values in Thoroughbreds. Fatigue of these less fatigue-resistant hind limb muscles would affect a horse's speed.

Muscular fatigue can cause detrimental effects to the musculoskeletal system that may lead to injury.1,2 Thoroughbred racehorses participating in longdistance races have a higher risk of injury to the superficial digital flexor tendon,3 suggesting that fatigue is associated with injury. Furthermore, fatigue associated with overtraining causes poor performance in horses.4 Therefore, understanding the mechanisms of fatigue may lead to the prevention of injuries and better training regimens.

Surface EMG is a noninvasive and useful technique for investigating muscle activity in animals, including horses,5,6 because it records the sum of the motor unit potentials of subcutaneous muscles.6–8 There are several variables that are appropriate for analysis of the EMG signal. Integrated EMG determines the area under a fully rectified EMG trace and total muscular activity over time.8 Another variable that reflects motor unit behavior is the RMS,8 which is calculated as follows:

article image

where EMG is the value of the EMG signal at each moment of time (t), and T represents the duration of the analyzed signal.8 Furthermore, the MF of the EMG signal calculated by spectral analysis, such as FFT, is often used to describe the EMG frequency characteristics.8 These variables are reliable for assessment of muscular fatigue in humans; for example, it is known that fatigue causes an increase of iEMG9 and a decrease in MF.10,11 In horses, a negative correlation between the MF of deltoid muscles and time during galloping was found without lead limb changes, although the correlation was not significant, likely owing to the small sample size.12 In addition, the RMS of the long digital extensor muscles during trotting increased when horses were fatigued.13 However, little is known about fatigue-induced EMG changes in other muscles during galloping.

During races, Thoroughbred racehorses gallop at speeds > 17 m/s while carrying a jockey's weight.14 It is well known that the hip joints are a main source for generation of propulsive power,15,16 and that trunk muscles are important for stabilization of vertebrae.17,18 Therefore, understanding fatigue of these muscles is important for the horseracing industry. The muscles of Thoroughbreds have a much higher percentage of type II muscle fiber than the percentage of type I muscle fiber, especially in hind limb and trunk muscles.19 Therefore, those muscles should be less fatigue resistant.

The objective of the present study was to quantify changes in iEMG value and MF in hind limb muscles of horses with fatigue during galloping. Targeted muscles were the LDm, TFLm, GMm, and BFm. With the exception of the TFLm, the percentage of type II muscle fiber is much higher than the percentage of type I muscle fiber in these muscles.19 On the basis of previous reports,9,10,12,13 we hypothesized that the iEMG values of the LDm, GMm, and BFm in Thoroughbreds would increase, whereas the MF of those muscles would decrease, with fatigue during galloping.

Materials and Methods

The study was approved by the Animal Welfare and Ethics Committee of the Japan Racing Association Equine Research Institute.

Animals

Eight Thoroughbreds from the Japan Racing Association's Equine Research Institute herd were studied. They were clinically nonlame (ie, no subjectively evident lameness) as determined on the basis of results of a physical examination and brief lameness examination performed by 2 experienced veterinarians. Among the horses, there were 5 geldings and 3 mares (age range, 5 to 9 years; weight range, 474 to 564 kg). All of the horses were familiar with running on a treadmill.a

EMG data acquisition

Electrode positions—The EMG signals for each hind limb were recorded from the LDm at the level of the 16th thoracic vertebra and from the TFLm at just under the tuber coxae, which is the muscular part of the TFLm and less likely to be affected by skin movement. The EMG signals were recorded from the GMm on an imaginary line drawn from the tuber coxae to the root of the tail at a third of the distance from the tuber coxae20 and from the BFm at its upper third portion.21 We had verified the accurate location of each muscle not only by ultrasonography but also by dissections of some horses euthanized for other research projects; therefore, we were assured that targeted muscles in these locations were not covered by other muscles. The skin over each muscle was shaved and cleaned with alcohol around the muscle belly.

Measuring equipment—The sEMG data were collected from both left and right sides of muscles with a multichannel telemetry system that is used to analyze human dynamic movements.22 The compact electrode telemetry system had active, reference, and ground electrodes and a wireless transmitterb that sends sEMG data to a host computer for real-time display and storage.c We connected the transmitters via snap-type lead cablesd to surface Ag-AgCl electrodese (diameter, 20 mm), which were attached to each muscle parallel to the muscle fibers. The distance between each surface electrode was 30 mm, with active and reference electrodes set side by side. To strengthen the attachment of the surface electrodes, fast-acting gluef and foam padsg that had strong adhesion were used. Cables were also attached to each horse's body with foam padsg to reduce motion artifact as much as possible. We verified that they remained at the surfaces of the recorded muscles until the experiment was finished and the sEMG data were not noisy (Figure 1). The sEMG data were omitted from the analysis if there was excessive noise or if the electrodes were detached during a measurement.

Figure 1—
Figure 1—

Representative EMG activity of a Thoroughbred's left LDm, TFLm, GMm, and BFm over 4 consecutive strides during galloping on a treadmill (grade, 3%) at a constant speed (13.5 m/s) to achieve fatigue after approximately 360 seconds. For this horse, the left hind limb was the lead limb after 30 seconds of galloping and EMG activity recordings were obtained (start of gallop exercise [left column]). Recordings were obtained at 30-second intervals thereafter until the horse could not maintain its position on the treadmill despite humane encouragement. The data for the horse in the fatigued state were those obtained at the last point at which the left hind limb was again the lead limb prior to exhaustion (right column). In this horse, all electrodes remained at the surfaces of the recorded muscles, and the baseline EMG activity of all muscles is stable throughout the experiment. When electrodes were detached or EMG data were excessively noisy for any horse in the study, those data were excluded from further analysis.

Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.828

Foot-on and foot-off events were determined with strain gaugesh that were attached firmly with gluef to the dorsal midline of the hind limb hooves. They were connected to dynamic strain-measuring instrumentsi that were connected to the input terminal boxj of the multichannel telemetry system via coaxial cable with BNC (Bayonet Neill-Concelman) connectors. The EMG signals were amplified (200X) and hoof strain-gauge signals were recorded at 1 kHz with a 10-bit resolution (5 mV; full scale) and filtered (band-pass filter, 15 to 500 Hz for EMG; low-pass filter, 250 Hz for strain gauges).

Experimental protocol

Experiments were performed on a treadmilla inclined to a 3% grade. For each horse, the exercise protocol involved walking at 1.7 m/s for 1 minute, trotting at 3.5 m/s for 3 minutes, and then galloping at a constant speed until the horse was exhausted (ie, it could not maintain its position on the treadmill despite minimal humane encouragement). The galloping speed was set to make each horse run for approximately 360 seconds, which had been determined for each horse during the preceding week. After this gallop exercise, each horse was trotted at a speed of 3.5 m/s for 3 minutes followed by walking at 1.7 m/s for 10 minutes.

Data analysis

During the gallop exercise, SF, iEMG value, and MF were measured every 30 seconds. During trotting before and after the gallop exercise, those variables were measured at the 60-second time point. The SF and iEMG value for a stride were measured (Figure 2); for purposes of statistical analysis, the mean from 5 consecutive strides was calculated for each variable starting at each time point. The SF (strides/s) was calculated as the inverse of stride duration of the lead limb during gallop and left limb during trotting. The MF, which was determined by FFT and defined as the point where the Fourier profile was bisected in area,12 was measured for a muscle discharge, and the mean of 5 consecutive MFs was calculated for statistical analysis. To obtain MF during a muscle discharge, an FFT (256 points, which were processed with a Hanning window) was applied for a 100-millisecond period during both galloping and trotting (before and after galloping). Muscle discharge was defined as the EMG signal for a 100-millisecond period when the maximal amplitude of the EMG signal was located at the midpoint of the wave, to which FFT was applied (Figure 2). The total number of points sampled by the FFT was only 100, and points extending beyond this limit for the FFT were set to zero.

Figure 2—
Figure 2—

Diagram to illustrate the calculation of the iEMG value and MF for the left and right LDm, TFLm, GMm, and BFm of Thoroughbreds used in a study to quantify fatigue-induced EMG changes in their hind limb muscles. For each left or right muscle, the iEMG value (the total of the red area under the traces) for a stride (red dashed arrow) was measured at the predetermined time points during galloping and trotting and the mean of the values for 5 consecutive strides was used for statistical analysis. Fast Fourier transform was applied to EMG signals for a period of 100 milliseconds (horizontal thick arrow), and the maximum amplitude of the EMG wave (designated by the vertical arrow) was located at the midpoint of the signal region. Then, MF was calculated mathematically as the point where the Fourier profile was bisected in area. Mean MF was derived from 5 consecutive muscle discharges and used for statistical analysis.

Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.828

During the gallop exercise, each hind limb was defined as the lead or trailing limb during the first 30-second interval prior to data collection. For example, if the left hind limb was the lead limb at 30 seconds into the run, then the left muscles were defined as the lead limb and the right muscles were defined as the trailing limb. Because the frequency of switching the lead limb or the timing of the lead-limb changes depended on the horse, the SF, iEMG value, and MF of the lead or trailing limb were measured every 30 seconds and were compared with data from the same lead or trailing limb just before exhaustion (Figure 3). In other words, all 3 variables at fatigue were those obtained at the last time when the lead limb was on the same side as it was at 30 seconds prior to exhaustion.

Figure 3—
Figure 3—

Typical changes in the iEMG value of a Thoroughbred's left GMm during galloping at a constant speed (14.7 m/s) on a treadmill (3% grade). For this horse, the left hind limb was the lead limb at the 30-second time point; thus, the left muscle was analyzed as the lead limb muscle (circles) and the right muscle was analyzed as the trailing limb muscle (triangles). Recordings were obtained at 30-second intervals thereafter until the horse could not maintain its position on the treadmill despite humane encouragement. The duration of galloping until exhaustion was 365 seconds for this horse. The last data collection point at which the left hind limb was the lead limb prior to exhaustion was 330 seconds; the iEMG value at that time was considered the value at fatigue.

Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.828

For trotting analysis, comparisons of all variables were made between the data obtained at the 60-second time point before and after galloping. Because lead and trailing limb were not identified during trotting, the comparison was made between the start and fatigued values for the same-side muscles.

The Shapiro-Wilk test and F-test were applied to the data to test the normality distribution and equal variances, respectively. A paired t test was used to compare differences between data obtained at the start of the gallop exercise (ie, at 30 seconds into the run) and when horses were fatigued at the end of the gallop exercise in both the lead and trailing limbs, as well as to compare the values of both sides (left and right hind limbs) during trotting both before and after the gallop exercise. Statistical analyses were performed with commercial softwarek with significance set at a value of P ≤ 0.05.

Results

All data were confirmed to be distributed normally and of equal variance. Electrodes were attached to 64 muscles in total; however, data from 28 muscles were excluded from analysis because of sEMG noise or electrode detachment during the experiment.

During the gallop exercise, the horses' running speeds ranged from 12.6 m/s to 14.7 m/s. Six horses became exhausted after 360 to 390 seconds; 1 horse was exhausted between 330 and 360 seconds and the other horse was exhausted between 240 and 270 seconds (Figure 4). The number of times that horses changed their lead hind limb during the experiments ranged from 2 to 7.

Figure 4—
Figure 4—

Changes in SF for each of 8 Thoroughbreds (represented by a different color) that were instrumented with electrodes to assess EMG activity of the left and right LDm, TFLm, GMm, and BFm during galloping on a treadmill (grade, 3%) at a constant speed (12.6 to 14.7 m/s) to achieve fatigue after approximately 360 seconds. During the gallop exercise, SF was determined every 30 seconds. The SF was calculated as the inverse of stride duration of the lead limb. Stride frequency of all horses decreased over time.

Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.828

The SF of all horses decreased with time (Figure 4). The mean ± SD value for SF was 2.14 ± 0.06 strides/s at the start of the gallop exercise and 2.05 ± 0.07 strides/s when fatigued at the end of the gallop exercise, which represented a significant (P = 0.001) decrease. During the gallop exercise, fatigue decreased the iEMG values of the GMm and BFm both in the lead limb and the trailing limb, whereas the iEMG values of the LDm and TFLm of the lead limb and the trailing limb did not change (Figure 5). The MF did not change with fatigue in any muscle in either the lead limb or trailing limb (Table 1).

Figure 5—
Figure 5—

Changes in iEMG values for the LDm, TFLm, GMm, and BFm of the lead hind limb (A) and the trailing hind limb (B) of the 8 Thoroughbreds in Figure 4 during galloping to exhaustion. A—For the lead limb, iEMG values for the GMm and BFm decreased with fatigue, whereas those of the LDm and TFLm did not change with fatigue. B—For the trailing limb, the iEMG values of the GMm and BFm decreased with fatigue, whereas those of the LDm and TFLm did not change with fatigue. Data were not available for each muscle from all horses because there was excessive signal noise or the electrodes were detached during a measurement in some horses. *For this muscle, the value at fatigue for each horse is significantly (all P ≤ 0.01) less than that at the start of galloping. †For this muscle, the value at fatigue for each horse is significantly (P ≤ 0.05) less than that at the start of galloping. See Figure 4 for key.

Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.828

Table 1—

Values of MF for the LDm, TFLm, GMm, and BFm of the lead and trailing hind limbs of 8 Thoroughbreds determined during galloping on a treadmill (grade, 3%) at a constant speed (12.6 to 14.7 m/s) to achieve fatigue after approximately 360 seconds.

 Lead hind limbTrailing hind limb
Time pointLDm (n = 3)TFLm (n = 3)GMm (n = 6)BFm (n = 6)LDm (n = 4)TFLm (n = 5)GMm (n = 5)BFm (n = 4)
Start45.8 ± 5.059.3 ± 3.534.9 ± 12.437.0 ± 10.851.7 ± 1.656.5 ± 13.646.3 ± 14.244.1 ± 11.2
At fatigue40.9 ± 5.258.3 ± 9.334.3 ± 12.438.1 ± 12.450.3 ± 4.745.4 ± 7.246.2 ± 5.741.7 ± 9.8

Data are reported as the mean ± SD. During galloping, the lead limb at the 30-second time point was designated as the lead limb and the other hind limb was designated as the trailing limb for purposes of analysis. The MF value at the 30-second time point was considered the value at the start of the gallop exercise. Recordings were obtained at 30-second intervals thereafter until the horse could not maintain its position on the treadmill despite humane encouragement. For each horse, the last data collection point was that at which the designated lead limb was again the lead limb prior to exhaustion; the MF value at that time was considered the value at fatigue. To obtain the MF, FFT was applied to EMG signals for a period of 100 milliseconds, and the maximum amplitude of the EMG wave was located at the midpoint of the signal region. Then, MF was calculated mathematically as the point where the Fourier profile was bisected in area. At each time point, a mean MF value was derived from 5 consecutive muscle discharges for each horse, and an overall mean MF value calculated. Data were not available for each muscle from all horses because there was excessive signal noise or the electrodes were detached during a measurement in some horses. For all muscles, the MF value did not change with fatigue.

During trotting before and after the gallop exercise, mean SF was 1.45 ± 0.06 strides/s and 1.44 ± 0.07 strides/s, respectively. Therefore, SF did not change with fatigue (P = 0.20). The iEMG values of the left hind limb muscles during trotting when the horses were fatigued did not change from values obtained when the horses were trotting but not fatigued (Figure 6). Similarly, the iEMG values of the right hind limb muscles during trotting when the horses were fatigued did not change from values obtained when the horses were trotting but not fatigued. During trotting, the MF in any muscle of either the left or right hind limb did not change as a result of fatigue (Table 2).

Figure 6—
Figure 6—

Changes in iEMG of the left LDm, TFLm, GMm, and BFm muscles of the 8 Thoroughbreds in Figure 4 during trotting prior to the galloping exercise and when fatigued after the galloping exercise. For each horse, the trotting exercise was performed on a treadmill inclined to a 3% grade and involved trotting at 3.5 m/s for 3 minutes prior to the gallop exercise; after galloping to exhaustion, each horse was trotted at a speed of 3.5 m/s for 3 minutes. The iEMG values were assessed during trotting at the 60-second time point before and after galloping, and these values were compared. For all muscles, the iEMG values did not change with fatigue. Data were not available for each muscle from all horses because there was excessive signal noise or the electrodes were detached during a measurement in some horses. See Figure 4 for remainder of key.

Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.828

Table 2—

Values of MF for the LDm, TFLm, GMm, and BFm of the left and right hind limbs of the 8 Thoroughbreds in Table 1 determined during trotting prior to the gallop exercise and when fatigued after the gallop exercise.

 Left hind limbRight hind limb
Time pointLDm (n = 3)TFLm (n = 3)GMm (n = 7)BFm (n = 6)LDm (n = 4)TFLm (n = 5)GMm (n = 4)BFm (n = 4)
Start46.9 ± 11.463.9 ± 18.241.8 ± 8.940.4 ± 9.056.9 ± 13.155.6 ± 11.347.8 ± 4.840.3 ± 6.7
At fatigue44.9 ± 14.939.1 ± 14.948.2 ± 9.643.7 ± 7.539.1 ± 3.150.6 ± 9.444.1 ± 10.640.9 ± 8.1

Data are reported as the mean ± SD. For each horse, the trotting exercise was performed on a treadmill inclined to a 3% grade and involved trotting at 3.5 m/s for 3 minutes prior to the gallop exercise; after galloping to exhaustion, each horse was trotted at a speed of 3.5 m/s for 3 minutes. The MF values during trotting were calculated at the 60-second time point before and after galloping. For all muscles, the MF value did not change with fatigue.

See Table 1 for remainder of key.

Discussion

Results of the present study in Thoroughbreds indicated that the iEMG values of the GMm and BFm, which are the largest contributors to propulsion and impulsion,7,23 decreased when fatigued during galloping, whereas the iEMG values of the LDm and TFLm did not change with fatigue. In contrast, none of the assessed hind limb muscles had a fatigue-associated decrease in MF. Furthermore, neither the iEMG value nor MF during trotting were affected by fatigue.

To our knowledge, the present study was the first to quantify fatigue-induced EMG changes in major hind limb muscles of Thoroughbreds. Electromyographic activity (characterized by the iEMG value) reflects motor unit activity, including the recruitment pattern.6,11 It is likely that most available motor units, including type IIx muscle fibers, would be recruited at the onset of galloping because it is high-intensity exercise.24–26 Furthermore, the GMm and BFm have high proportions of type II muscle fibers and lower proportions of type I muscle fibers,19,27 suggesting that these muscles are less resistant to fatigue.11 Decreases in iEMG values for the GMm and BFm associated with fatigue may suggest that recruited motor units, mainly type II muscle fibers, contribute less when fatigued. In addition, the GMm and BFm in horses have large cross-sectional areas to generate force and power,23 indicating that the activities of these muscles are critical for high-speed galloping. Therefore, the inability to maintain speed with fatigue may be attributable to decreased power exerted by the GMm and BFm.

It is well known that the SF decreases and that stride length increases for fatigued horses on a treadmill on which speed is constrained.12,28,29 Results of another study30 indicated that a low percentage of type IIB fibers can cause low SF. During galloping on inclined surfaces, SF is important for providing net vertical work.26 In the present study, SF of all horses decreased over time; eventually, they could not keep their position on the treadmill despite humane encouragement. It is possible that a decrease in type IIx fiber recruitment attributable to fatigue might be associated with the observed decrease in SF, leading to the inability of the horses to maintain speed.

Muscle force is determined not only by spatial factors, namely how accurately electrodes can detect active motor units, but also by temporal factors, namely the firing rate of motor units.8 Our hypothesis was based on results of human studies,8,9 which indicated that maintaining a constant work rate until exhaustion requires additional muscle fiber recruitment and an increase in the iEMG value to exert the same mechanical power output. In addition, fatigue causes synchronization or a decrease of firing rates, leading to a decrease of MF.8 However, contrary to our hypothesis, the iEMG values of the GMm and BFm of horses running at constant speed decreased with fatigue, presumably for the aforementioned reasons. It is possible that, unlike in humans, fatigue in horses during galloping might result in greater activities of other muscles in the forelimbs or the distal portion of the hind limbs13 or in elastic structures,31,32 compared with the situation in a nonfatigued condition. This could subsequently lead a horse to compensate for the decrease in GMm or BFm activities. Indeed, Standardbred trotters adapt to fatigue with kinematic changes (eg, excursions of their metacarpophalangeal or carpal joints),29 suggesting the possibility that high loads are being applied to tendons or ligaments. Further EMG research involving other muscles will be required to determine how locomotion changes with fatigue.

In horses, the TFLm works mainly as an extensor or stabilizer muscle of the stife joint in the stance phase by tightening the fascia lata around that joint during locomotion.33 Furthermore, a high percentage of type I fibers is present in the TFLm,19 suggesting that this muscle is resistant to fatigue. In humans, the iEMG value of the vastus lateralis muscle with a high percentage of slow-twitch fibers did not decrease with fatigue, whereas the iEMG values of those muscles with higher proportions of fast-twitch muscles decreased with fatigue during isokinetic contractions.11 These findings were consistent with the results of the present study, in that fatigue-induced decreases in the iEMG value of the TFLm in horses did not occur.

The iEMG value of the LDm did not change with fatigue although the composition of muscle fiber type is similar to that of the BFm.19 We speculated that this is because the LDm has a quite different function from that of the BFm. Although the mechanical work required to replace energy lost to the environment or to power tasks, such as acceleration, is most likely done by the hind limb muscles (eg, the GMm or BFm),23 the LDm is mainly responsible for stabilization of the vertebral column against dynamic forces.34 In addition, it is known that the storage and release of elastic energy could occur along the vertebral column during galloping.32 The iEMG value of the LDm in the present study might not have changed because of the elastic energy use along the vertebral column. The calculation of elastic energy with fatigue would be required to reveal the reason for the different results between GMm or BFm and LDm. Although the iEMG value of the horses LDm did not change in the present study, it is possible that the iEMG value of the LDm might change if horses carried riders or additional weight because of increased loading of the LDm.35,36 In that case, according to the size principle,25 more type II muscle fibers might be recruited from the onset of galloping to counter a rider's weight, and the iEMG value might decrease with fatigue. Further research is warranted to reveal the relation between fatigue of the LDm in horses carrying riders or added weights.

During trotting, the fluctuations of kinetic and potential energies are in phase and elastic energy storage helps to compensate for the decrease in mechanical energy,31,32 whereas walking gaits have an out-of-phase relationship between these 2 energy forms.32 Moreover, slow trotting uses more elastic energy, compared with that used during high-speed trotting, suggesting that speed might be associated with elastic recovery.31 Cheung et al13 reported that fatigue increased the RMS of the long digital extensor muscle during trotting; however, during trotting in the present study, the iEMG value of the muscles investigated did not change with fatigue. It is possible that these contradictory results might be attributable to the difference in protocols. In the present study, the horses might have trotted with more elastic energy and less muscle activity because of the lighter exercise-intensity protocol (galloping at 3.5 m/s on a 3% incline), compared with the effect of the exercise-intensity protocol used in the previous study (galloping at 4 m/s on a 10% incline).13 An additional factor could be that different muscles have different responses to fatigue. For example, muscles in the distal portion of the limb, such as long digital extensor muscle, might be more activated during trotting when fatigued.13 Further studies will be needed to elucidate the response to fatigue of different muscles that have different muscle fiber types, lengths, volumes, and functions.

Contrary to our hypothesis, no muscles had a fatigue-induced change in MF regardless of gait. A decrease in MF in deltoid muscles in Thoroughbreds has been detected but only when horses ran without lead-limb changes.12 When lead-limb changes occur, the rate of fatigue in a given limb can change. For example, it is possible that a shift from right to left and back again might relieve the strain off 1 limb. However, it is difficult to control lead-limb changes during high-speed galloping. In addition, FFT is ideally performed for isometric contractions8 because muscle length affects the MF of the EMG signal.37 However, in animal experiments, establishment of an isometric protocol is difficult; to compensate, we applied FFT for only a short interval. Even during this short interval, concentric and eccentric contractions could have occurred, and this would have affected the EMG signals. Although a decrease in MF is observed with dynamic contractions in humans,10,38 it can be more difficult to determine a decrease in MF with fatigue in equids because changes in equine muscle fiber length may be more dynamic than those in humans.

A limitation of the present study derived from the nature of sEMG. Although we analyzed the data that seemed not to be affected by motion artifacts, we could not omit all motion artifacts. In humans, muscular fatigue is often measured with an isometric protocol.8 With dynamic contractions, greater motion artifacts contaminate the sEMG signal because of the motion of the electrode relative to the muscle during the contraction.39 These motion artifacts are unavoidable in animal experiments because true isometric maximal voluntary contraction measurement as part of a static muscle contraction protocol similar to that used in humans8,39 is impossible.5 Although most motion artifacts are < 20 Hz in human experiments,39 there has been no study to investigate the relation between equine motion artifacts and EMG signals. Moreover, there are several factors that can affect the EMG signal such as muscle temperature, blood flow, and dehydration.8,40 The sEMG signal is a combined neuromuscular signal that includes those factors; however, recent acoustic myography that measures vibration generated by muscles can detect pure muscle contraction with less noise from motion artifacts.41 Although there are few published reports of the use of acoustic myography in veterinary species,42 investigating acoustic myographic changes with fatigue might provide new insights into understanding muscular fatigue in animals.

In the present study, the iEMG values of the GMm and BFm of Thoroughbreds decreased with gallop-induced fatigue, leading to the horses' inability to maintain high speed, but the iEMG values of the LDm and TFLm did not change with fatigue. In our opinion, iEMG appears to be superior to the assessment of MF for detection of muscular fatigue in horses with gallop-induced fatigue.

Acknowledgments

Supported by the Japan Racing Association.

Presented in abstract form at the 8th International Conference on Canine and Equine Locomotion, London, England, August 2016.

The authors thank Dr. James H. Jones of the School of Veterinary Medicine, University of California–Davis for comments and English language assistance on the manuscript.

ABREVIATIONS

EMG

Electromyography

FFT

Fast Fourier transform

GMm

Gluteus medius muscle

iEMG

Integrated EMG

LDm

Longissimus dorsi muscle

MF

Median frequency of muscle discharge

RMS

Root-mean square

sEMG

Surface EMG

SF

Stride frequency

TFLm

Tensor fasciae latae muscle

Footnotes

a.

SÄTO AB, Knivsta, Sweden.

b.

ZB-150H, Nihon Kohden, Tokyo, Japan.

c.

WEB-7000, Nihon Kohden, Tokyo, Japan.

d.

Lead wire C K251, Nihon Kohden, Tokyo, Japan

e.

Vitrode J, Nihon Kohden, Tokyo, Japan.

f.

Gachi, Kokuyo, Osaka, Japan.

g.

Foam pad 75A, Nihon Kohden, Tokyo, Japan.

h.

N22-FA-10-120-11-VS3, Showa Measuring Instruments Inc, Tokyo, Japan.

i.

DPM-612B, Kyowa Electronic Instruments, Tokyo, Japan.

j.

JC-130H, Nihon Kohden, Tokyo, Japan

k.

JMP 13.1.0, SAS Institute Inc, Cary, NC.

References

  • 1. Yoshikawa T, Mori S, Santiesteban AJ, et al. The effects of muscle fatigue on bone strain. J Exp Biol 1994;188:217233.

  • 2. Butcher MT, Hermanson JW, Ducharme NG, et al. Superficial digital flexor tendon lesions in racehorses as a sequela to muscle fatigue: a preliminary study. Equine Vet J 2007;39:540545.

    • Search Google Scholar
    • Export Citation
  • 3. Takahashi T, Kasashima Y, Ueno Y. Association between race history and risk of superficial digital flexor tendon injury in Thoroughbred racehorses. J Am Vet Med Assoc 2004;225:9093.

    • Search Google Scholar
    • Export Citation
  • 4. Hamlin MJ, Shearman J, Hopkins W. Changes in physiological parameters in overtrained Standardbred racehorses. Equine Vet J 2002;34:383388.

    • Search Google Scholar
    • Export Citation
  • 5. Valentin S, Zsoldos RR. Surface electromyography in animal biomechanics: A systematic review. J Electromyogr Kinesiol 2016;28:167183.

    • Search Google Scholar
    • Export Citation
  • 6. Williams JM. Electromyography in the horse: a useful technology? J Equine Vet Sci 2017;60:4358.

  • 7. Clayton H, van Weeren PR. Measurement techniques for gait analysis. In: Back W, Clayton H, eds. Equine locomotion. 2nd ed. London: Saunders, 2013;3160.

    • Search Google Scholar
    • Export Citation
  • 8. Kamen G. Electromyographic kinesiology. In: Robertson DGE, Caldwell GE, Hamill J, et al. eds. Research methods in biomechanics. 2nd ed. Champaign, Ill: Human Kinetics, 2014;179201.

    • Search Google Scholar
    • Export Citation
  • 9. Vanhatalo A, Poole DC, DiMenna FJ, et al. Muscle fiber recruitment and the slow component of O2 uptake: constant work rate vs. all-out sprint exercise. Am J Physiol Regul Integr Comp Physiol 2011;300:R700R707.

    • Search Google Scholar
    • Export Citation
  • 10. Ament W, Bonga GJ, Hof AL, et al. EMG median power frequency in an exhausting exercise. J Electromyogr Kinesiol 1993;3:214220.

  • 11. Komi PV, Tesch P. EMG frequency spectrum, muscle structure, and fatigue during dynamic contractions in man. Eur J Appl Physiol Occup Physiol 1979;42:4150.

    • Search Google Scholar
    • Export Citation
  • 12. Colborne GR, Birtles D, Cacchione I. Electromyographic and kinematic indicators of fatigue in horses: a pilot study. Equine Vet J Suppl 2001;33:8993.

    • Search Google Scholar
    • Export Citation
  • 13. Cheung TK, Warren L, Lawrence L, et al. Electromyographic activity of the long digital extensor muscle in the exercising Thoroughbred horse. Equine Vet J 1998;30:251255.

    • Search Google Scholar
    • Export Citation
  • 14. Takahashi T. The effect of age on the racing speed of Thoroughbred racehorses. J Equine Sci 2015;26:4348.

  • 15. Clayton HM, Hodson E, Lanovaz J, et al. The hindlimb in walking horses: 2. Net joint moments and joint powers. Equine Vet J 2001;33:4448.

    • Search Google Scholar
    • Export Citation
  • 16. Dutto DJ, Hoyt DF, Clayton HM, et al. Joint work and power for both the forelimb and hindlimb during trotting in the horse. J Exp Biol 2006;209:39903999.

    • Search Google Scholar
    • Export Citation
  • 17. Audigié F, Pourcelot P, Degueurce C, et al. Kinematics of the equine back: flexion-extension movements in sound trotting horses. Equine Vet J Suppl 1999;30:210213.

    • Search Google Scholar
    • Export Citation
  • 18. Robert C, Audigie F, Valette JP, et al. Effects of treadmill speed on the mechanics of the back in the trotting saddlehorse. Equine Vet J Suppl 2001;33:154159.

    • Search Google Scholar
    • Export Citation
  • 19. Kawai M, Minami Y, Sayama Y, et al. Muscle fiber population and biochemical properties of whole body muscles in Thoroughbred horses. Anat Rec (Hoboken) 2009;292:16631669.

    • Search Google Scholar
    • Export Citation
  • 20. Eto D, Yamano S, Kasashima Y, et al. Effect of controlled exercise on middle gluteal muscle fibre composition in Thoroughbred foals. Equine Vet J 2003;35:676680.

    • Search Google Scholar
    • Export Citation
  • 21. Takahashi T, Matsui A, Mukai K, et al. The effects of inclination (up and down) of the treadmill on the electromyogram activities of the forelimb and hind limb muscles at a walk and a trot in Thoroughbred horses. J Equine Sci 2014;25:7377.

    • Search Google Scholar
    • Export Citation
  • 22. Nukaga H, Takeda T, Nakajima K, et al. Masseter muscle activity in track and field athletes: A pilot study. Open Dent J 2016;10:474485.

    • Search Google Scholar
    • Export Citation
  • 23. Payne RC, Hutchinson JR, Robilliard JJ, et al. Functional specialisation of pelvic limb anatomy in horses (Equus caballus). J Anat 2005;206:557574.

    • Search Google Scholar
    • Export Citation
  • 24. Yamano S, Eto D, Hiraga A, et al. Recruitment pattern of muscle fibre type during high intensity exercise (60–100% VO2 max) in Thoroughbred horses. Res Vet Sci 2006;80:109115.

    • Search Google Scholar
    • Export Citation
  • 25. Votion D-M, Navet R, Lacombe VA, et al. Muscle energetics in exercising horses. Equine Comp Exerc Physiol 2007;4:105118.

  • 26. Parsons KJ, Pfau T, Wilson AM. High-speed gallop locomotion in the thoroughbred racehorse. I. The effect of incline on stride parameters. J Exp Biol 2008;211:935944.

    • Search Google Scholar
    • Export Citation
  • 27. López-Rivero J, Serrano A, Diz A, et al. Variability of muscle fibre composition and fibre size in the horse gluteus medius: an enzyme-histochemical and morphometric study. J Anat 1992;181:110.

    • Search Google Scholar
    • Export Citation
  • 28. Wickler SJ, Greene H, Egan K, et al. Stride parameters and hindlimb length in horses fatigued on a treadmill and at an endurance ride. Equine Vet J Suppl 2006;38:6064.

    • Search Google Scholar
    • Export Citation
  • 29. Johnston C, Gottlieb-Vedi M, Drevemo S, et al. The kinematics of loading and fatigue in the Standardbred trotter. Equine Vet J Suppl 1999;31:249253.

    • Search Google Scholar
    • Export Citation
  • 30. Ronéus N, Essén-Gustavsson B, Johnston C, et al. Lactate response to maximal exercise on the track: relation to muscle characteristics and kinematic variables. Equine Vet J 1995;27:191194.

    • Search Google Scholar
    • Export Citation
  • 31. Biewener AA. Muscle-tendon stresses and elastic energy storage during locomotion in the horse. Comp Biochem Physiol B 1998;120:7387.

    • Search Google Scholar
    • Export Citation
  • 32. Minetti AE, Ardigo L, Reinach E, et al. The relationship between mechanical work and energy expenditure of locomotion in horses. J Exp Biol 1999;202:23292338.

    • Search Google Scholar
    • Export Citation
  • 33. Tokuriki M, Aoki O. Electromyographic activity of the hindlimb muscles during the walk, trot and canter. Equine Vet J 1995;27:152155.

    • Search Google Scholar
    • Export Citation
  • 34. Licka TF, Peham C, Frey A. Electromyographic activity of the longissimus dorsi muscles in horses during trotting on a treadmill. Am J Vet Res 2004;65:155158.

    • Search Google Scholar
    • Export Citation
  • 35. de Cocq P, Weeren Pv, Back W. Effects of girth, saddle and weight on movements of the horse. Equine Vet J 2004;36:758763.

  • 36. Fruehwirth B, Peham C, Scheidl M, et al. Evaluation of pressure distribution under an English saddle at walk, trot and canter. Equine Vet J 2004;36:754757.

    • Search Google Scholar
    • Export Citation
  • 37. Doud JR, Walsh J. Muscle fatigue and muscle length interaction: effect on the EMG frequency components. Electromyogr Clin Neurophysiol 1995;35:331339.

    • Search Google Scholar
    • Export Citation
  • 38. Masuda K, Masuda T, Sadoyama T, et al. Changes in surface EMG parameters during static and dynamic fatiguing contractions. J Electromyogr Kinesiol 1999;9:3946.

    • Search Google Scholar
    • Export Citation
  • 39. De Luca CJ, Gilmore LD, Kuznetsov M, et al. Filtering the surface EMG signal: Movement artifact and baseline noise contamination. J Biomech 2010;43:15731579.

    • Search Google Scholar
    • Export Citation
  • 40. Stewart D, Macaluso A, De Vito GD. The effect of an active warm-up on surface EMG and muscle performance in healthy humans. Eur J Appl Physiol 2003;89:509513.

    • Search Google Scholar
    • Export Citation
  • 41. Harrison AP. A more precise, repeatable and diagnostic alternative to surface electromyography – an appraisal of clinical utility of acoustic myography. Clin Physiol Funct Imaging 2018;38:312325.

    • Search Google Scholar
    • Export Citation
  • 42. Riis KH, Harrison AP, Riis-Olesen K. Non-invasive assessment of equine muscular function: a case study. Open Vet J 2013;3:8084.

Contributor Notes

Address correspondence to Dr. Takahashi (yuji_takahashi@equinst.go.jp).