Reasons for performing study: Massage is widely used in physiotherapy, but there has been little previous research examining its effectiveness in increasing equine soft tissue length.
Objectives: To determine the effect of massage on equine hindlimb protraction. We hypothesised that massage to the caudal muscles of the equine hindlimb, specifically the superficial gluteal, semitendinosus, biceps femoris and semimembranosus muscles, can increase passive and active hindlimb protraction.
Materials and methods: The study used a crossover design in which 8 horses were randomly assigned to 2 groups of 4, A and B. Group A received massage for 30 min, group B received sham treatment for 30 min. The procedure was repeated following a 7 day ‘washout’ period, when group A received the 30 min sham treatment and group B, 30 min of massage. Passive hindlimb protraction was measured before and after each intervention, using a modified version of the human ‘Sit and Reach test’ for general muscle flexibility. Active protraction was measured using 2 dimensional kinematic analysis of stride length. The data analysis used the Wilcoxon signed rank test at a significance of P<0.05.
Results: Massage to the caudal limb muscles significantly increased passive (P = 0.01) and active limb (P = 0.01) protraction.
Conclusions: This study indicates that massage can increase protraction of the equine hindlimb. Massage may, therefore, play a valuable role in the development of strategies used to improve a horse's locomotor function, e.g. during rehabilitation or optimum performance for competition.
Massage is defined as the systematic therapeutic stroking or kneading of the body or one of its parts (Blood and Studdert 1999). It is one of the oldest treatment modalities, dating back 300–400 years BC. Massage is currently widely used in human and veterinary physiotherapy for both its physiological and psychological effects (Valberg 1996; Roetting 1999).
Massage increases the pliability of the intramuscular (i.m.) connective tissue (Holey and Cooke 1999) and maintains muscle flexibility following trauma (Olmstead 1985). Massage is used therapeutically to reduce muscle tone by decreasing the excitability of alpha-motorneurones through its effects on the cutaneous mechanoreceptors and pressure receptors (Goldberg et al. 1992). As such, massage is a valuable physiotherapy technique that can be used to improve muscle flexibility with the intention to improve joint range of motion and has a variety of clinical applications.
Psychologically, massage therapy reduces feelings of anxiety and improves self-esteem in human patients (Hernandez-Reif et al. 1998). It is likely emotions also play an important role in the horse, particularly with respect to its sporting performance. The horse may not function optimally unless it is psychologically comfortable and confident (Denoix and Pailloux 1996) and massage may be able to promote this feeling of well being prior to competition.
One of the primary goals of training young horses is to achieve ‘engagement’ of the hindquarters (Crossley 1993), which is associated with an earlier maximal protraction of the hindlimb with respect to the ipsilateral forelimb (Back and Clayton 2001). A greater synchronicity of the ipsilateral retraction of the forelimb and protraction of the hindlimb at the trot is linked to an increased suppleness in the lateral bending of the body and is observed in horses with a naturally good gait (Back et al. 1994b). If massage to the caudal muscles of the hindlimb can increase protraction in ‘normal subjects’, it may well be beneficial in enhancing equine performance pre- and post trauma.
In this study we explore the effects of massage to the caudal muscles (retractors) of the equine hindlimb on active and passive limb protraction and stride length. More specifically we hypothesise that:
Massage significantly increases the passive movement range of protraction (passive is defined as being performed by an operator rather than the horse).
Massage significantly increases the active protraction of the equine hindlimb, which results in an increase in stride length (active is defined as being performed by the horse).
Materials and methods
The study was conducted under the auspices of the RVC and approved by the RVC ethics committee. The subjects comprised 8 mixed breed geldings, of a similar height (16–17.3 hh), aged between 8 and 13 years. All horses were in regular work and identically managed at the same riding school. A clinical examination was undertaken and animals were excluded if they were known to have a skin disease, systemic or local infection, soft tissue injury, or any orthopaedic condition.
The study used a crossover design with a ‘washout period’ of 7 days. Horses were randomly divided into 2 groups of 4 (A and B) using the table of random numbers and the restricted randomisation method (Petrie and Watson 1999). Horses in Group A received the massage treatment, followed by the ‘washout’ period of 7 days and then the sham procedure. Horses in Group B received the sham procedure, followed by the washout period and then the massage treatment.
Each horse received 30 min of effleurage massage interspersed with 3, 30 s bursts of circular kneading by the same veterinary physiotherapist to the superficial gluteal, semitendinosus, biceps femoris and semimembranosus muscles, i.e. the caudal muscles of the equine hindlimb. Effleurage is defined as a technique in massage in which long, light or firm strokes are used; whereas kneading, a form of Petrissage, utilises a grasping, rolling and pressing movement (The C.V. Mosby Company 1990). The therapist manually applied effleurage with an alternating force of between 2.3–9.5 kg, which had previously been determined by simulating the technique using vertical weigh scales1.
During the 30 min sham procedure, horses stood in the same environment with the therapist standing adjacent to their hindquarters. The therapist was positioned as they would be to carry out the massage.
The determination of the effect of massage on passive limb protraction
A static modified human ‘Sit and Reach test’ (Jones and Barker 1996) was used to measure passive hindlimb protraction before and after intervention. Prior to the measurement procedure, all horses stood squarely with the fore- and hindlimbs positioned in mid-stance with the metatarsals vertical (Leach et al. 1984). The hind leg was then protracted forward passively to its ‘end of range’ by the clinician, in the direction of the ipsilateral front limb.
End of range was taken as the point where ‘tissue resistance’ was felt by the clinician and where any further forward movement would instigate active foot placement with associated lateral flexion of the hindquarters. At the end of the passive limb protraction, an independent assessor recorded the distance between a mark made on the concrete yard surface using a spirit level2 and the toe of the ipsilateral front foot. The clinician performing the passive range of motion was blinded to the measurements. Five successive passive protraction movements were performed and their measurements recorded both pre- and post massage/sham procedure.
The determination of the effect of massage on active limb protraction
Stride length was used as a measure of active hindlimb protraction (Barrey 1999) and assessed using 2-dimensional kinematic data analysis recorded on a video camera3. Yellow insulation tape mounted on ‘Colplast’4 was positioned midline around the toe of the right hind hoof wall to improve the visibility of the foot against the riding surface. Stride length was repeatedly measured 5 times pre- and post massage/sham procedure whilst the horses trotted over a predetermined riding track (16 × 2 m), which had white marker posts placed at 2 m intervals to act as a reference system for calculating stride length (Fig 1).
Figure 1. Start stride for horse on track. Yellow marker indicates toe of hind hoof. Distance between poles is set at 2 m.
Motion capture was recorded at 25 frames/s at a shutter speed of 1/8000 second using the levelled video camera placed at a distance of 17 m perpendicular to the plane of the track. A zoom lens increased the image size and minimised the effect of parallax error (Clayton 1990).
Still images (800,000 pixels) were transferred to a computer and the position of the first and last footfall of the right hind foot was calculated relative to the white marker poles using the Cool Ruler5. Parallax errors (using the principle of similar triangles) and the average stride length were then determined.
The validity of the measurement technique using the computer still images had previously been determined and confirmed that the measurement of stride length from the still images on the computer was reliable. The repeatability analysis of measurement of still images showed the mean of the differences as 0.015 and the standard deviation (s.d.) of the differences as 0.023.
Data were checked for normality of distribution. Mean, median and standard deviations were calculated for stride length. Comparisons between conditions were made using the Wilcoxon signed rank test at a significance level of P<0.05.
The effect of the sham procedure on the passive movement of the equine hindlimb
There was no significant effect of sham procedure on passive hindlimb protraction (P = 0.12) (Table 1).
Table 1. Comparison of pre- and post measurements for sham and massage for passive range of protraction
HorseSham results (m)Pre-post sham (b)–(c)Massage results (m)Pre-post massage (e)–(f)Difference between col (d) and col (g) (cm)
A positive value in column (d) and column (g) indicates an increased range of protraction.
Mean value 0.00 0.09−0.09
The effect of massage on the passive movement of the equine hindlimb
All horses showed a significant increase in the range of passive movement following massage (P = 0.01) (Table 1). The distance between the toe of the hind foot and the toe of the front foot decreased by 0.09 ± 0.07 m (mean ± s.d.) and a median of 0.08 m, following massage.
Statistical comparison of passive movement following massage when compared to sham treatment showed that there was a trend for massage to increase limb protraction (P = 0.07) by 0.09 ± 0.10 m (mean ± s.d.) and a median of 0.07 m when compared to the sham procedure (Table 1).
The effect of the sham procedure on the active movement of the equine hindlimb
There was no significant increase in active limb protraction following the sham procedure (P = 0.23) (Table 2).
Table 2. Comparison of pre- and post measurements for sham and massage for active stride length
HorseSham results (m)Pre-post sham (b)–(c)Massage results (m)Pre-post massage (e)–(f)Difference between col (d) and col (g) (m)
The differences are expressed as median values. On average the post is greater than the pre- and represented by a negative value in columns (d) and (g).
Mean value −0.03 −0.180.15
The effect of massage on the active movement of the equine hindlimb
All 8 horses significantly increased in their active hindlimb protraction measurements following massage (P = 0.01), reflected by 0.18 ± 0.17 m (mean ± s.d.) and a median of 0.12 m increase in stride length (Table 2) and an associated increase in trotting speed (Table 3).
Table 3. The mean trotting speeds of the individual horses for active limb protraction
HorseMean speed of trotting horses (m/s)
Pre massagePost massagePre shamPost sham
Statistical comparison of stride length following massage when compared to sham treatment, showed that massage significantly increased stride length (P = 0.03) a mean of 0.15 ± 0.2 m (mean ± s.d.) and a median of 0.14 m, when compared to the sham procedure (Table 2).
This study demonstrates that massage to the caudal muscles of the equine hindlimb, that is the superficial gluteal, semitendinosus, biceps femoris and semimembranosus muscles, significantly increases both passive (P = 0.01) and active (P = 0.01) hindlimb protraction.
Our findings are consistent with those of Crosman et al. (1984), who reported that massage of the human hamstring muscles significantly improved the range of motion at the hip joint. Massage improved range of motion in children suffering from burns (Morien et al. 2008).
It is known that skeletal muscle tends to remain in a mild state of contraction maintained by the interaction between the excitatory impulses from the muscle spindle and inhibitory impulses from the golgi tendon organs. Both deep and superficial massage in man have been shown to reduce H-reflex activity (Goldberg et al. 1992; Sullivan et al. 1993) and therefore muscle tone. This reduction in muscle tone increases muscle compliance and may be the reason why the horses increased both their passive and active hindlimb protraction range. Massage may also increase the pliability of i.m. connective tissue and mobilise interfaces within the muscle allowing greater soft tissue flexibility (Holey and Cooke 1999) and this effect may also have contributed to the increase in hindlimb protraction.
The psychological effects of massage in man are well documented (Hernandez-Reif et al. 1998; Hemmings et al. 2000). Massage increases endorphin release, which may explain the reported sensation of well-being post massage (Kaada and Torsteimbo 1987). Massage may also have had a beneficial psychological effect on the horses used in this study, aiding relaxation and improving their sense of well-being. Relaxation is associated with a reduction in muscle tone (Beider and Moyer 2007), which enabled a greater freedom of limb protraction and the relaxed animal may also have been more compliant allowing the therapist to move the limb through a greater range of motion.
We also found that the horses increased their active stride length post massage and that this was associated with an increase in trotting speed (Table 3). A positive relationship between speed and stride length has been demonstrated previously by Deuel and Park in 1990, where the speed of the extended trot of high scoring dressage horses at the Seoul Olympics was found to be closely influenced by stride length. If massage to the caudal muscles of the hindlimb can increase protraction, measured as stride length, in ‘normal subjects’ it may well enhance equine performance and be of use post trauma. Some of the horses in the study also showed an increase in speed post sham (Table 3) but we found that this was not associated with such large changes in stride length, and may have resulted from an increase in stride frequency. It is possible that we were unable to adequately control for fluctuations of individual trotting speed in the field environment, which may have influenced our findings regarding stride length, and it is recommended that speed is controlled more rigorously in future research.
Further study may also include the effect of increasing temperature on soft tissue pliability and hence the range of limb protraction. For example, ‘Horse 6’ demonstrated an increase in passive protraction after the sham treatment and this may have resulted from increasing levels of sunshine during the procedure, giving rise to heating of the soft tissues (Strickler et al. 1990). In future, the use of a thermostatically-controlled box is indicated to avoid fluctuations in ambient temperature.
We believe this study is one of the first to attempt to utilise objective outcome measures to determine the effectiveness of one physiotherapeutic intervention, namely massage, on equine hindlimb protraction and stride length. Previously studies have tended to combine the effect of massage with another modality, which often means that it is difficult to separate the specific effects of massage alone.
We attempted to ensure that the experimental conditions were the same for all the animals in both the massage and sham groups. The intention was to minimise any physical or emotional stress in the subjects, which may have resulted in an increase in muscle tone that may have compromised muscle flexibility measurements of range of motion. As such we obtained measurements immediately post massage to assess its immediate effects (Cawley 1997). In an effort to blind the operators, the person collecting the data was unaware of whether the horse had been sham treated or massaged and the clinician who did the ‘Sit and Reach test’ for passive protraction was unable to view the outcome measurement. Although the sit and reach method is well established for assessing flexibility in humans it is recommended that an in-depth reliability pilot study should be carried out regarding its further use in horses.
The small sample size limited our statistical analysis to nonparametric tests, which are conservative by nature, and as such, our findings should be viewed as a preliminary insight into the effectiveness of massage on equine hindlimb protraction. It is recommended that further experiments be undertaken using a larger cohort of animals to confirm the findings and observations of this work. In addition, it would be interesting to determine if this massage technique has a similar effect on hindlimb protraction in different breeds and ages of horse. It would also be useful to compare the effectiveness of different forms of massage and their subsequent effects on limb protraction during different gaits whilst the horse exercised on different surfaces. Future study may also include measuring hindlimb joint angles on the treated limb, the comparative leg length between individuals and determining vertical velocity at push-off, as an increase in vertical velocity on push-off is influential on stride length (Clayton 1994a).
In summary, we found effleurage and circular kneading increased both active and passive limb protraction in our cohort of horses, but further work is necessitated to determine its true effectiveness and enable a reliable evaluation of its efficacy (Ernst 1999) in both the human and the animal subject.
This study indicates that massage to the caudal muscles of the equine hindlimb increases its active protractive movement measured as stride length. Massage may therefore play a valuable role in the development of strategies used to improve a horse's locomotor function, e.g. during a period of rehabilitation or when optimum performance is required for competition.
The authors would like to thank Dr Jo Price and the staff at the Structure and Motion Laboratory at the RVC, Dr Aviva Petrie of UCL, and Harris Croft Riding School of Wootton Bassett for their kind permission to use their horses and facilities.
Conflicts of interest
The authors have declared no potential conflicts.
1 Hanson weigh scales, France.
2 Fisco Tools Ltd, UK.
3 Panasonic NV-DS15B, Japan.
4 Naturin medical, UK.
5 FABSoft. Inc, USA.
Back, W. and Clayton, H. (2001) Intra-limb Coordination; the forelimb and the hind limb. In: Equine Locomotion, Ed: W.Back, H.Clayton. W.B. Saunders, London. p 120.
Back, W., Barneveld, A., Schamhardt, H.C., Bruin, G. and Hartman, W. (1994b) Kinematic detection of superior gait quality in young trotting warmbloods. Vet. Q. 16, 91-96.
Barrey, E. (1999) Methods, applications and limitations of gait analysis in horses. Vet. J. 157, 7-22.
Web of Science® Times Cited: 15
Beider, S. and Moyer, A. (2007) Randomised controlled trials of pediatric massage: a review. Evid. Based Complement. Alternat. Med. 4,23-24.
Web of Science® Times Cited: 14
Blood, D.C. and Studdert, V.P. (1999) Saunders Comprehensive Veterinary Dictionary, 2nd edn., Ed: C.Byres, W.B. Saunders, London. p 699.
Cawley, N. (1997) A critique of the methodology of research studies evaluating massage. Eur. J. Cancer Care 6, 23-31.
Clayton, H.M. (1990) Gait evaluation: making the most of your home video system. Proc. Am. Ass. equine Practners. 36, 447-455.
Clayton, H.M. (1994a) Comparison of the stride kinematics of the collected, working, medium and extended trot in horses. Equine vet. J.26, 230-234.
Web of Science® Times Cited: 36
Crosman, L., Chateauvert, S. and Weisberg, J. (1984) The effects of massage to the hamstring muscle group on range of motion. J. Orthop. Sports Phys. Ther. 6, 168-172.
Crossley, A. (1993) Riding to the bit. In: Training the Young Horse: The First Two Years, 1st edn., Stanley Paul and Co. Ltd, London. p95.
Denoix, J. and Pailloux, J. (1996) Neurophysiology. In: Physical Therapy and Massage For The Horse, Manson Publishing Ltd, London. p 10.
Deuel, N.R. and Park, J. (1990) The gait patterns of Olympic dressage horses. Int. J. Sport Biomech. 6, 198-226.
Web of Science® Times Cited: 18
Ernst, E. (1999) Massage therapy for low back pain: a systemic review. J. Pain Symptom Manage. 17, 65-69.
Web of Science® Times Cited: 53
Goldberg, J., Sullivan, S.J. and Seaborne, D.E. (1992) The effect of two intensities of massage on the H-reflex amplitude. Phys. Ther.72, 449-457.
Web of Science® Times Cited: 20
Hemmings, B., Smith, N., Graydon, J. and Dyson, R. (2000) Effects of massage on physiological restoration, perceived recovery, and repeated sports performance. Br. J. Sports Med. 34, 109-115.
Web of Science® Times Cited: 37
Hernandez–Reif, M., Field, T., Field, T. and Theakston, H. (1998) Multiple sclerosis patients benefit from massage therapy. J. Bodywork Mov. Ther. 2, 168-174.
Holey, E. and Cooke, E. (1999) Therapeutic Massage, 3rd edn., W.B. Saunders Company Ltd, London.
Jones, K. and Barker, K. (1996) Modified sit and reach test. In: Human Movement Explained, Butterworth-Heinmann Ltd, Oxford. p 253.
Kaada, B. and Torsteimbo, O. (1987) Vasoactive intestinal polypeptides in connective tissue massage. Gen. Pharmacol. 18, 379-384.
Web of Science® Times Cited: 7
Leach, D.H., Omrod, K. and Clayton, H.M. (1984) Standardized terminology for the discription and analysis of equine locomotion. Equine vet. J. 16, 522-528.
Web of Science® Times Cited: 28
Morien, A., Garrison, D. and Smith, N.K. (2008) Range of motion improves after massage in children with burns: a pilot study. J. Bodywork Mov. Ther. 12, 67-71.
Olmstead, M.L. (1985) Tendon, muscle and ligament injuries and surgery. In: Small Animal Orthopaedics, Mosby, Missouri. pp 473-499.
Petrie, A. and Watson, P. (1999) Statistics for Veterinary and Animal Science, Blackwell Science Ltd, Oxford.
Roetting, A.K. (1999) Manual lymphatic drainage. Treatment of the horse's leg. Available at: http://www.diss.fu-berlin.de/diss/receive/FUDISS_thesis_000000000152 (accessed 4 August 2010).
Strickler, T., Malone, T. and Garrett, W.E. (1990) The effects of passive warming on muscle injury. Am. J. Sports Med. 18, 141-145.
Web of Science® Times Cited: 35
Sullivan, S.J., Seguin, S. and Seaborne, D. (1993) Reduction of H-reflex amplitude during the application of effleurage to the triceps surae in neurologically healthy subjects. Physiother. Theory Pract. 9, 25-31.
The C.V. Mosby Company (1990) Mosby's Dictionary, 3rd edn., The C.V. Mosby Company, St. Louis, Missouri. p 406.
Valberg, S.J. (1996) Muscular causes of exercise intolerance in horses. Vet. Clin. N. Am.: Equine Pract. 12, 495-515.