SEARCH

SEARCH BY CITATION

Keywords:

  • horse;
  • abdominal muscles;
  • surface electromyography;
  • walk;
  • trot

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. Manufacturers' addresses
  9. References

Reason for performing study: The rectus abdominis (RA) and oblique external abdominal (OEA) muscles are both part of the construction of the equine trunk and thought to be essential for the function of the spine during locomotion. Although RA activity at trot has previously been investigated, the relationship between OEA and RA at walk and trot has not yet been described.

Objectives: To document abdominal muscle activities during walk and trot, and test the hypothesis that muscle activity at walk would be smaller than at trot.

Materials and methods: Six horses (8–20 years old, 450–700 kg) were used for surface electromyography (EMG) measurements, with EMG electrodes placed caudal to the sternum (RA) and at the level of the 16th rib (OEA). On all hooves, the withers and the sacrum reflective markers were placed to determine motion cycles. Normal distribution of data was tested using a Kolmogorov-Smirnov test and Student's t test was used to compare left-right and walk-trot differences (P<0.05).

Results: Minimum, maximum and mean EMG values recorded at walk were significantly higher at trot than at walk in all horses for OEA and in 5/6 horses for RA. At walk, EMG activity ranged from 8–44 mV (RA) and 7–54 mV (OEA). At trot, EMG activity ranged from 18–150 mV (RA) and 27–239 mV (OEA). There were statistically significant differences between maximum activities of left and right OEA and RA muscles at walk in all horses, and in 4/6 horses at trot.

Conclusions: Muscle activities of OEA and RA are smaller at walk than at trot. At walk, the OEA/RA ratio is lower than at trot. There are more significant correlations between muscle activities of both RA and OEA and limb movements at walk than at the trot.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. Manufacturers' addresses
  9. References

Surface electromyography (sEMG) provides a method to estimate the degree of muscular activation, by recording potentials from electrodes placed close to a selected muscle. Electrodes (typically of an Ag–AgCl design) are attached on the skin overlying the muscle to provide an indirect measure of muscle-generated potentials. The most relevant advantage over needle electrodes is the noninvasive nature of the surface electrodes. Surface electrodes can sample a large muscle volume (Basmajian and De Luca 1985) and in the horse most muscles are of a large volume. Needle EMG samples only the small volume of muscle in the direct vicinity of the needle and, therefore, needle EMG may be less representative of the overall function of an equine muscle than surface EMG.

In the horse, the oblique external abdominal muscle (OEA) has a fleshy origin on the lateral surface of the ribs and, more caudally, on the thoracolumbar fascia. It indirectly inserts on the prepubic tendon via the abdominal tunic (Nickel et al. 1984; Budras et al. 2009). The function of the OEA is the increase of intra-abdominal pressure necessary for urination, defecation and parturition. Additionally, it creates a firm yet pliable body wall supporting and protecting the abdominal contents.

The rectus abdominis muscle (RA) arises from the 4th to the 9th costal cartilages and adjacent parts of the sternum and ends on the prepubic tendon. It is marked by about 10 tendinous intersections, which attach the muscle to the external rectus sheath (Budras et al. 2009). This enclosure consists of external and internal sheaths, the former a combination of the 2 oblique muscle aponeuroses, the latter furnished by the transverse abdominal muscle (Budras et al. 2009). The equine spine has been likened to a ‘string and bow’ arrangement where the ‘bow’ is the rigid vertebral column and the ‘string’ is the musculature that keeps it under constant tension (Evans et al. 1995). This ‘string’ function is carried out by the epaxial muscles and by the RA, which creates dorsal flexion in the thoracolumbar spine.

Human abdominal muscles are known to be essential to achieve core stability, the ability of the trunk to support force production, and withstand forces acting upon it (Lederman 2010). Both RA and OEA activities have been evaluated in human subjects at stance and during walking and running, and marked differences in muscle recruitment pattern and relative muscle activities for the different gaits were detected (Capellini et al. 2006; Anders et al. 2007). In the horse, where back pain is a common but not fully understoodproblem (Jeffcott 1999), the activity of the abdominal muscles may also be influenced by inadequate or abnormal functioning of the thoracolumbar spine.

Although integrated EMG values of the equine RA have been reported at the trot (Robert et al. 2001a, 2002), no data on RA muscle activity at walk have been published. Also, equine OEA muscle activity values have not been reported at walk or trot. In the dog, OEA and RA activity were recorded (Tokuriki 1973a,b) at walk and at trot and graphical semiquantitative results were reported.

The goal of this study was therefore to simultaneously document equine RA and OEA muscle activity at walk and trot in order to evaluate the function of these muscles. We hypothesised that muscle activity in both muscles would be smaller at walk than at trot.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. Manufacturers' addresses
  9. References

Horses

From the teaching herd of the University of Veterinary Medicine, Vienna all available horses were examined for possible inclusion in this study. Prior to the measurements the horses were assessed clinically and horses with a history of colic, surgery or other major traumas in the ventral abdominal region were excluded. Also, horses were trotted in hand and only clinically sound horses were used for this experiment. Finally, 6 horses could be included in this study (2 Thoroughbreds, 2 Trotters and 2 Warmbloods; 2 geldings, 2 mares, 2 stallions; 8–20 years old; 450–700 kg). Selected horses were warmed-up and accustomed to the experimental set-up on the treadmill (Mustang 2200)1.

Data collection

Passive reflective makers were secured with adhesive tape on the 4 hooves, the withers and the sacrum. Three-dimensional kinematic data were collected using 10 infrared cameras (Eagle Digital RealTime System)2 recording at 120 Hz using kinematic software (Cortex 1.3)2.

For surface EMG measurements the skin over the muscles was shaved, pregelled bipolar AgCl electrodes (30 mm in diameter) were placed bilaterally over the RA muscle at the level of the end of the sternum and bilaterally over the OEA muscle at the level of lateral surface of the 16th rib. Electrodes were placed after palpation of bony landmarks, following prior determination of a suitable position of electrodes in 2 horses using ultrasound. The electrodes were held in position by use of nonirritant adhesive tape at a distance of 3 cm from each other and parallel to the direction of muscle fibres. A single reference electrode was placed over the left tuber coxae.

The sEMG measurements were taken and transmitted by use of a telemetric system (Telemyo Mini 16)3 (sample frequency 1.2 kHz). The sEMG data were synchronised with the kinematic software to obtain simultaneous recordings. Each horse walked and trotted at its own optimum speed (mean ± s.d. walk: 1.25 ± 0.06 m/s, range 1.2–1.35 m/s; trot: 3.23 ± 0.17 m/s, range 3.0–3.5 m/s). Data collection continued until 3 trials (each 10 s) for walk and trot had been recorded.

Data analysis and processing

The 3D coordinates of each marker during the time course of each experiment were calculated from the data using kinematic software. These time series were then smoothed by use of a Butterworth low-pass filter (cut-off frequency, 10 Hz). The data were split into motion cycles by MATLAB R2008b4 starting with the beginning of the right fore swing phase (breakover) and the duration of each motion cycle was calculated. A motion cycle is therefore defined as all the movements of the body occurring from the right fore breakover to the next right fore breakover.

The sEMG signal was rectified and sampling rate reduced to 120 Hz to make motion and sEMG comparable. A Butterworth low-pass filter was applied (fifth order; cut-off frequency, 10 Hz) to realise sampling, which was completed by calculating the mean of 10 samples. From all the motion cycles measured, the mean sEMG and mean motion for a motion cycle were calculated for each horse. At least 5 motion cycles for each horse at walk and trot were used and a mean walk and trot motion cycle was determined for each horse. The mean of a minimum of 15 motion cycles was considered for the overall mean motion cycle for each gait. For each of the 100 time points of the motion cycle this calculation was carried out. Maxima and Minima were defined as maximum and minimum sEMG values occurring within the mean motion cycle of each horse.

For each time point throughout the mean motion cycle the ratio of OEA:RA activity was calculated for each horse. These ratios were compared between walk and trot.

Statistical analysis

Statistical analyses were done using SPSS 17.05. Normal distribution of data was tested by use of a Kolmogorov-Smirnov test. Differences between left and right sides and between walk and trot were tested by use of a Student's t test for paired samples. Values of P<0.05 were considered significant. Pearson correlations were calculated for the muscle activity throughout the mean motion cycle of each horse of the left and right OEA and RA muscles and the vertical movement of each foot, the transversal movement of the withers and the sacrum over the mean motion cycle of each horse at walk and at trot. A total of 180 correlations were calculated, 30 per horse. Values of P<0.05 were considered significant, and values of P<0.01 were considered highly significant.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. Manufacturers' addresses
  9. References

In individual horses, minimum OEA muscle activity ranged from 7–16 mV at walk, and from 27–71 mV at trot. Maximum OEA muscle activity ranged from 16–54 mV at walk and from 60–239 mV at trot. For OEA for each motion cycle, one EMG maximum was found at walk and 2 at trot. Minimum RA muscle activity ranged from 8–16 mV at walk, and from 18–71 mV at trot. Maximum RA muscle activity ranged from 19–44 mV at walk and from 41–150 mV at trot. For RA for each motion cycle, 2 EMG maxima were found at walk and 2 at trot (see Figs 1, 2; Table 1).

image

Figure 1. Results of the smoothed surface EMG activity of the OEA and RA muscles and vertical movement of the 4 hooves at walk (OEA L: left obliquus externus, RA L: left rectus abdominis, OEA R: right obliquus externus, RA R: right rectus abdominis, LH: left hind, RH: right hind, LF: left fore, RF: right fore).

Download figure to PowerPoint

image

Figure 2. Results of the smoothed surface EMG activity of the OEA and RA muscles and vertical movement of the 4 hooves at trot (OEA L: left obliquus externus, RA L: left rectus abdominis, OEA R: right obliquus externus, RA R: right rectus abdominis, LH: left hind, RH: right hind, LF: left fore, RF: right fore).

Download figure to PowerPoint

Table 1. Mean ± s.d. values for mean, minimum and maximum surface EMG activity (mV) of the OEA and RA muscles at walk and trot
EMG activityOEA LOEA RRA LRA R
  1. OEA L: left obliquus externus, OEA R: right obliquus externus, RA L: left rectus abdominis, RA R: right rectus abdominis. *Statistically significant differences between left and right side muscle activity (P<0.05).

WalkMinimum (mV)9 ± 212 ± 312 ± 211 ± 2
Maximum (mV)25 ± 637 ± 1127 ± 923 ± 4
Mean (mV)15 ± 3*22 ± 7*19 ± 7*16 ± 2*
Difference between maximum and minimum (mV)15 ± 624 ± 915 ± 913 ± 4
TrotMinimum (mV)42 ± 1250 ± 1726 ± 734 ± 19
Maximum (mV)100 ± 24144 ± 7384 ± 24104 ± 36
Mean (mV)66 ± 14*83 ± 33*51 ± 11*62 ± 21*
Difference between maximum and minimum (mV)57 ± 2393 ± 5758 ± 2470 ± 28

There were significant differences between mean left and mean right muscle activities over the motion cycle in all horses at walk, and in 4/6 horses at trot. Between walk and trot muscle activity there were significant differences for OEA in all horses and for RA in 5/6 horses.

There were highly significant differences in OEA/RA ratios on both sides in all horses between walk and trot (P<0.00). At the walk the mean OEA/RA ratio was 1.2 (range 0.2–4.3, s.d. 0.6) and at the trot the mean OEA/RA ratio was 1.6 (range 0.3–5.0, s.d. 0.9).

At the walk, more correlations of muscle activity with kinematic parameters were significant or highly significant than at trot (walk 124/180, trot 57/180). A total of 39 correlations were significant both at walk and at trot, 15 were either positive or negative in both gaits, whereas 24 were positive in one and negative in the other gait (see Table 2).

Table 2. Pearson's correlation coefficients of the surface EMG at walk and trot with selected kinematic parameters and with the activity of other muscles throughout the mean motion cycle of each horse at each gait
  1. OEA L: left obliquus externus, OEA R: right obliquus externus, RA L: left rectus abdominis, RA R: right rectus abdominis) and the numbers 1–6 represent the individual horses. The kinematic parameters are the vertical movement of the hoof (LF: left fore, RF: right fore, LH: left hind, RH: right hind) and the transversal movement of the withers and sacrum. Only significant correlations are presented, significance level P<0.01 indicated by bold numbers, and significance level P<0.05 is indicated by italic numbers.

inline image

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. Manufacturers' addresses
  9. References

The application of sEMG in this study was used as an indicator of muscle activity. The signals provide the timing sequence of the 2 muscles during gait (De Luca 1997). Surface EMG analysis can be accomplished employing different techniques that allow the calculation of an envelope around the EMG signal. Low-pass filters have the disadvantage that afterwards some artefacts such as movement noise as well as cross-talk signals may still be present (De Luca et al. 2010). Besides the possibility of continuous muscle activation of the measured muscles, artefacts and cross-talks are possible additional reasons why in a living animal at stance sEMG signals do not reach zero values. All these influences are larger during locomotion. As the muscle activity pattern was important for the comparison of muscle activities in the present study, a low pass filter and reduction of the sampling rate were used to compare the EMG with the kinematic data (De Luca et al. 2010). A motion cycle adapted filtering technique has been shown to improve the signal to noise ratio in EMG measurements during locomotion (Peham et al. 2001a); however, in order not to bias the correlations calculated between the motion of the limbs and the EMG activities of the RA and OEA muscles, this filtering technique was not employed in the present study.

It should be remembered, that true force/power outputs can only be approximated by the EMG, even though a relationship between EMG signal and force has been investigated for sEMG in man (Milner-Brown and Stein 1975). In the present study, smoothed, filtered EMG signals (Peham et al. 2001a) were assessed, but we did not calculate integrated EMG, which might have offered additional information because of its linear relation to the force (Yoshida and Terao 2003).

The muscles used in this study were chosen because of their considerable size and we presumed that their activity patterns are correlated to the gait of the horse. Spinal energy transfer has been shown in man, and as the RA and OEA have an effect on trunk movement and mobility, they may be indirectly functional for the efficiency of the gait (Gracovetsky and Iacono 1987). Naturally, only superficially located muscles could be used for sEMG. Additional information on the activity of the obliquus internus abdominis muscle and especially on the activity of the transversusabdominis muscle would have been ideal to give a more comprehensive insight, especially as the transversus abdominis muscle has been assessed in a previous study at walk (Gutting et al. 1991). In order to achieve better core stability, e.g. for young race horses at risk of developing athletic performance related thoracolumbar pathologies, it would be of relevance to investigate when and how these trunk muscles are active, in order to improve training similar to human athletes where such strategies are now adopted (Borghuis et al. 2008). In several EMG studies the placement of the electrodes has been tested empirically (Wakeling et al. 2006; Zaneb et al. 2009) and in another study, an error analysis of electrode placement was done (Robert et al. 1999). Further investigation to optimise electrode placement in the OEA muscle should be the subject of a future methodological paper.

There have been several single (Tokuriki et al. 1991) and various breed studies (Robert et al. 2000, 2001b; Peham et al. 2001b; Licka et al. 2004) investigating equine EMG and both have interesting results. Using a wide variety of horses increases the variability of the results compared to single breed studies, but if a pattern is seen even within an nonhomogeneous study group then the results are more likely to be universally applicable to all equids.

In the walking dog, no activity of the OEA was documented, and continuous negligible or slight activity in the RA (Tokuriki 1973a), no clear on-off times were noted in the OEA and the RA muscles. Tokuriki (1973b) interpreted the OEA activity measured in the trotting dog as stabilisation of the axial skeleton by conforming the abdominal part in the midstance phase of hindlimbs. In the dog, this study found that the RA is always active, but its activity becomes weak when hindlimbs are in the swing phase. The activity of the canine OEA was graded as negligible during the mid-swing phase of the hindlimb and as slight and moderate during the stance phases of the hindlimbs at trot. The canine RA showed moderate and slight activity during the swing phase of hindlimbs and moderate activity during the stance phases of hindlimbs at trot. Comparing our results to these canine EMG activity patterns we found similar differences in magnitude between walk and trot. Also, no clear off times were determined for either muscle in the horse at walk and at trot.

The pattern of OEA activity is not as clearly defined as the RA pattern, or longissimus dorsi (LD) muscle patterns in previous studies (Robert et al. 2001a,b, 2002; Licka et al. 2004, 2009). Also, standard deviations are relatively high even within individual horses. A possible explanation for this is the recruitment of OEA for breathing, as it is described to be also an expiratory muscle (Nickel et al. 1984). In a study investigating regulation of respiratory muscles in adult horses at stance the muscle activity of the OEA during the expiratory phase was confirmed (Ainsworth et al. 1997). The fact that breathing is not coupled with locomotion at walk and trot (Ainsworth et al. 1997) and therefore the independently occurring necessity to activate OEA for expiration may create the large standard deviation seen. Additional studies at canter, where breathing and motion cycles are coupled, would help to further identify this effect. In analogy to the human OEA, where the muscle fascicle pattern was found to support the role of the OEA in the production of torque of the trunk (Urquhart et al. 2005), torque production may also be a function of the equine OEA. In our study we could not investigate torque production, as we did not measure synchronous detailed thoracolumbar spinal kinematics together with the EMG of the OEA, but this would be an interesting topic for future studies.

As hypothesised, maximum values at walk were smaller than at trot for both muscles investigated. Maximum values of activity were reported to allow a comparison of maximum activity of a muscle with the maximum activity of the other muscles measured. In a human study, the co-contraction (flexors/extensors) and the abdominal (EO/RA) synergist pattern was calculated as an indication of the concerted function of these muscles, which was also calculated in the present study. The muscle co-activation pattern can be interpreted as a support to provide trunk stability (Silfies et al. 2005). This calculation of ratios was also employed in an equine sEMG study looking at changes in muscle function in the presence of hindlimb lameness. In that study the reason given for using the ratio instead of the activity as such was that this reduces interindividual differences as the effect of factors such as skin conductivity and subcutaneous fat pads can be reduced (Zaneb et al. 2009).

Comparing the muscle activity patterns of OEA/RA ratios at walk and at trot, it is interesting that RA activity is relatively higher at the walk than at the trot, even though the vertical forces on the trunk (probably counteracted by RA among others) increase comparatively more at the trot than the transversal forces (probably counteracted by OEA among others).

The activity of OEA shows a clear phase shift between left and right, which indicates that the muscle is used more unilaterally. On the other hand, RA is active on both sides simultaneously, even though there is a difference between the magnitude of the left and right RA in the horses of our study, which may be related to the asymmetric distribution of the heavy intestines within the abdomen (Nickel et al. 1984), or with the fact that the horses were held from the left side. The simultaneous activity indicates that RA counteracts ventral spinal extension during the stance phases.

Horses used in this study were allowed to walk at their own optimum speed on the treadmill, and even at the slow walking speeds, RA activity was clearly visible. This is different from human individuals walking upright, where slow walk does not lead to muscle activity of the RA as at this speed and in this body position there is no need for toning the abdomen to protect the abdominal cavity (Saunders et al. 2004).

The tension of RA is thought to limit the passive thoracolumbar extension induced by the visceral mass acceleration during the stance phase (Denoix and Pailloux 1996). The speed and the relatively long suspension phase with the subsequent landing make this more relevant at the trot than at the walk.

Both at walk and at trot, the activity of RA is occurring at the times of low activity of the LD muscles, as measured in previous studies (Licka et al. 2004, 2009). A similar finding was reported in a study investigating EMG of the LD and RA as well as back angles at the trot (Robert et al. 1998). The RA was found to act during the stance phase and demonstrates antagonistic function and activity to the LD in one additional study at trot (Tokuriki et al. 1991, 1997). If RA were only used for vertebral stabilisation, then a simultaneous activity with the LD would have to be expected, but this is not the case. This shift in activity of RA may indicate the RA function of arching the back or at least reducing ventral spinal extension, which would possibly allow more cranial placement of the hind feet. At the end of the stance phase an ipsilateral small local maximum of RA activity was found. This could possibly represent the use of the RA muscle for cranial movement of the femur, where the RA attaches with the accessory ligament on the head of the femur (Budras et al. 2009).

The overall increased muscle activity at trot was interpreted as providing additional trunk stability and helping to create the rigid platform from which the limbs may swing faster (Rooney 1982).

In man, RA and OEA have been studied together (Saunders et al. 2004; Vera-Garcia et al. 2007). Hollowing the back, similar to ventral extension of the spine in the horse, had a less stabilising effect on the trunk than the bracing body position where a person sits bent forward with the hands behind their head. This position flexes the back and is therefore similar to dorsal flexion of the equine spine. This is interesting, as the dorsal angle of the equine back can be very well influenced by side reins, collection and the rider (Rhodin et al. 2005). A future perspective should therefore be the investigation of muscle activities in a variety of body postures during motion.

Correlations between motion characteristics and muscle activities were calculated throughout the motion cycle. In the 6 horses used for this study, a surprisingly large variation in correlations between vertical hoof movement, sacral bone and withers movement as well as muscle activities was found. In many cases, highly significant positive and negative correlations were documented for the same pair of parameters in different individuals. In fact only in 2 pairs of parameters at walk (OEA L and RH; RA L and LH) all 6 horses show similar and significant correlations. This may be due to the differences in conformation between horses, but is more likely to be due to the additional functions of the abdominal muscles, such as the need for breathing. It would be especially interesting for a future study to measure the abdominal muscles at canter, and to compare whether correlations in individuals are more similar in that gait, where breathing is coupled with locomotion.

In summary, clearly different abdominal muscle activities are present at walk and at trot. The biomechanical differences between the gaits are reflected in the correlations of muscle activities as well as in the maximum sEMG values reached. Stance phases of limbs require different stabilisation mechanisms to be employed in the 2 gaits to attenuate the ground contact forces, and probably concentric activity during breathing (OEA) is also present.

Manufacturers' addresses

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. Manufacturers' addresses
  9. References

1 Kagra AG, Fahrwangen, Switzerland.

2 Motion Analysis Corp., Santa Rosa, California, USA.

3 Noraxon Inc., Scottsdale, Arizona, USA.

4 The MathWorks Inc., Natick, Massachusetts, USA.

5 SPSS Inc., Chicago, Illinois, USA.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conflict of interest
  8. Manufacturers' addresses
  9. References
  • Ainsworth, D.M., Smith, C.A., Eicker, S.W., Ducharme, N.G., Henderson, K.S., Snedden, K. and Dempsey, J.A. (1997) Pulmonary-locomotory interactions in exercising dogs and horses. Respir. Physiol. 110, 287-294.
  • Anders, C., Wagner, H., Puta, C., Grassme, R., Petrovitch, A. and Scholle, H.C. (2007) Trunk muscle activation patterns during walking at different speeds. J. Electro. Kin. 17, 245-252.
  • Basmajian, J.V. and De Luca, C.J. (1985) Apparatus and technique. In: Muscles Alive. Their Functions Revealed by Electromyography, 5th edn., Williams and Wilkins, Baltimore. pp 22-50.
  • Borghuis, J., Hof, A.L. and Lemmink, K.A. (2008) The importance of sensory-motor control in providing core stability: implications for measurement and training. Sports Med. 38, 893-916.
  • Budras, K.-D., Sack, W.O. and Röck, S. (2009) Chapter 8: Abdominal wall and cavity. In: Anatomy of the Horse, 5th edn., Schlütersche Verlagsgesellschaft mbH & Co.KG, Hannover. pp 64-65.
  • Capellini, G., Ivanenko, Y.P., Poppele, R.E. and Lacquaniti, F. (2006) Motor patterns in human walking and running. J. Neurophysiol. 95, 3426-3437.
  • De Luca, C.J. (1997) The use of surface electromyography in biomechanics. J. appl. Biomech. 13, 135-163.
  • De Luca, C.J., Donald Gilmore, L., Kuznetsov, M. and Roy, S.H. (2010) Filtering the surface EMG signal: Movement artifact and baseline noise contamination. J. Biomech. 43, 1573-1579.
  • Denoix, J.M. and Pailloux, J.M. (1996) Muscles of the neck and trunk, Chapter 2: Anatomy and basic biomechanical concepts. In: Physical Therapy and Massage for the Horse, Manson Publishing Ltd., London. pp 30-36.
  • Evans, D.L., Jeffcott, L.B. and Knight, P.K. (1995) Effects of training on bone, Chapter: Performance-related problems and exercise physiology. In: The Equine Manual, 1st edn., Eds: A.J.Higgins and I.M.Wright, W.B. Saunders, London. p 957.
  • Gracovetsky, S.A. and Iacono, S. (1987) Energy transfers in the spinal engine. J. Biomed. Eng. 9, 99-114.
  • Gutting, S.M., Forster, H.V., Lowry, T.F., Brice, A.G. and Pan, L.G. (1991) Respiratory muscle recruitment in awake ponies during exercise and CO2 inhalation. Respir. Physiol. 86, 315-332.
  • Jeffcott, L.B. (1999) Back problems. Historical and clinical indications. Vet. Clin. N. Am.: Equine Pract. 15, 1-12.
  • Lederman, E. (2010) The myth of core stability. J. Bodyw. Mov. Ther. 14, 84-89.
  • Licka, T., Frey, A. and Peham, C. (2004) Electromyographic activity of the longissimus dorsi muscles in horses during trotting on a treadmill. Am. J. vet. Res. 65, 155-158.
  • Licka, T., Frey, A. and Peham, C. (2009) Electromyographic activity of the longissimus dorsi muscles in horses when walking on a treadmill. Vet. J. 180, 71-76.
  • Milner-Brown, H.S. and Stein, R.B. (1975) The relation between the surface electromyogram and muscular force. J. Physiol. 246, 549-569.
  • Nickel, R., Schummer, A. and Seiferle, E. (1984) Innervation der Bauchmuskeln. In: Lehrbuch Der Anatomie Der Haustiere I, 5th edn., Verlag Paul Parey, Berlin. p 336.
  • Peham, C., Licka, T. and Scheidl, M. (2001a) Evaluation of a signal-adapted filter for processing of periodic electromyography signals in horses walking on a treadmill. Am. J. vet. Res. 62, 1687-1689.
  • Peham, C., Frey, A., Licka, T. and Scheidl, M. (2001b) Evaluation of the EMG activity of the long back muscle during induced back movements at stance. Equine vet. J., Suppl. 33, 165-168.
  • Rhodin, M., Johnston, C., Roeathlisberger Holm, K., Wennerstrand, J. and Drevemo, S. (2005) The influence of head and neck position on kinematics of the back in riding horses at walk and trot. Equine vet. J. 37, 7-11.
  • Robert, C., Valette, J.P. and Denoix, J.M. (1998) Surface electromyographic analysis of normal horse locomotion: a preliminary report. In: Proceedings of the First Conference of the Equine Sports Medicine and Science. pp 80-85.
  • Robert, C., Valette, J.P. and Denoix, J.M. (2000) The effects of treadmill inclination and speed on the activity of two hindlimb muscles in the trotting horse. Equine vet. J. 32, 312-317.
  • Robert, C., Valette, J.P. and Denoix, J.M. (2001a) The effects of treadmill inclination and speed on the activity of three trunk muscles in the trotting horse. Equine vet. J. 33, 466-472.
  • Robert, C., Audigie, F., Valette, J.P., Pourcelot, P. and Denoix, J.M. (2001b) Effects of treadmill speed on the mechanics of the back in the trotting saddlehorse. Equine vet. J., Suppl. 33, 154-159.
  • Robert, C., Valette, J.P., Degueurce, C. and Denoix, J.M. (1999) Correlation between surface electromyography and kinematics of the hindlimb of horses at trot on a treadmill. Cells Tissues Organs 165, 113-122.
  • Robert, C., Valette, J.P., Pourcelot, P., Audigie, F. and Denoix, J.M. (2002) Effects of trotting speed on muscle activity and kinematics in saddlehorses. Equine vet. J., Suppl. 34, 295-301.
  • Rooney, J.R. (1982) The horse's back: Biomechanics of lameness. Equine Pract. 4, 17-27.
  • Saunders, S.W., Rath, D. and Hodges, P.W. (2004) Postural and respiratory activation of the trunk muscles changes with mode and speed of locomotion. Gait Posture 20, 280-290.
  • Silfies, S.P., Squillante, D., Maurer, P., Westcott, S. and Karduna, A.R. (2005) Trunk muscle recruitment patterns in specific chronic low back pain populations. Clin. Biomech. 20, 465-473.
  • Tokuriki, M. (1973a) Electromyographic and joint-mechanical studies in quadrupedal locomotion. I. Walk. Jap. J. vet. Sci. 35, 433-466.
  • Tokuriki, M. (1973b) Electromyographic and joint-mechanical studies in quadrupedal locomotion. II. Trot. Jap. J. vet. Sci. 35, 525-533.
  • Tokuriki, M., Nakada, A. and Aoki, O. (1991) Electromyographic activity of trunk muscles in horse during locomotion and synaptic connections of neurons of trunk muscles in the cat. In: Proceedings of the 1st International Workshop on Animal Locomotion, Eds: H.C.Schamhardt and A.J.Van Den Bogert, University of Utrecht, Utrecht. p 36.
  • Tokuriki, M., Otsuki, R., Kai, M., Hiraga, A. and Aoki, O. (1997) Electromyographic activity of trunk muscles in horse during locomotion on a treadmill. Proceedings of the 5th World Equine Veterinary Association Congress. p 26.
  • Urquhart, D.M., Barker, P.J., Hoges, P.W., Story, I.H. and Briggs, C.A. (2005) Regional morphology of the transversus abdominis and obliquus internus and externus abdominis muscle. Clin. Biomech. 20, 233-241.
  • Vera-Garcia, F.J., Elvira, J.L.L., Brown, S.H.M. and McGill, S.M. (2007) Effects of abdominal stabilization manoeuvres on the control of spine motion and stability against sudden trunk perturbations. J. Electro. Kin. 17, 556-567.
  • Wakeling, J.M., Barnett, K., Price, S. and Nankervis, K. (2006) Effects of manipulative therapy on the longissimus dorsi in the equine back. Equine Comp. Exercise Physiol. 3, 153-160.
  • Yoshida, M. and Terao, M. (2003) Suitable cutoff frequency of low-pass filter for estimating muscle by surface electromyogram. In: Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. 2, 1709-1711.
  • Zaneb, H., Kaufmann, V., Stanek, C., Peham, C. and Licka, T.F. (2009) Quantitative differences in activities of back and pelvic limb muscles during walking and trotting between chronically lame and nonlame horses. Am. J. vet. Res. 70, 1129-1134.