The use of swimming in thoroughbred racehorse training

by Dr Kathryn Nankervis SFHEA CVPM, The Equine Therapy Centre, Hartpury College

Swimming is common practice within the training and rehabilitation programmes of Thoroughbred racehorses yet there have been fewer than 20 scientific studies on swimming in the last 30 years. It is generally considered that swimming provides a good cardiovascular work-out for horses that are unable to gallop due to musculoskeletal injury. Heart rates during swimming are similar to those seen during canter or gallop work, i.e. from 160 to 220 beats/min (Jones and Hiraga 2008) with oxygen uptakes of approximately 70-80% of the typical maximal oxygen uptake of a trained horse (Thomas et al. 1980). Whilst the heart rates and oxygen uptakes during swimming and galloping may be similar, the biomechanical demands are entirely different. Unlike working overland, or on a water treadmill, the swimming horse is completely unable to generate locomotor force by pushing against the ground. In this respect, swimming exercise has certain advantages over work up the gallops as the horse exercises without any impact on the distal limbs. The disadvantage is that now the considerable mechanical advantage afforded by the ‘biological spring’ of the horse’s tendinous distal limb is absent and so all forward movement requires muscular effort.

During normal overland movement, the weight of the abdomen exerts a downward force on the spine, tending to draw the spine into extension. In water, the weight of the abdomen is supported, and the action of the hind limbs being brought forward under the body can generate considerable flexion (rounding) of the caudal thoracolumbar spine. Horses that are unable to flex the thoracolumbar region/lumbosacral joint due to back pain or stiffness are often less proficient swimmers. Indeed some authors report that swimming can exacerbate proximal limb lameness and/or back pain (Swanstrom and Lindy, 1974); Galloux et al. 1993). Therefore evaluation of an individual’s thoracolumbar comfort and range of movement should be undertaken before swimming to assess suitability.


Galloux, P., Goupil, X., Vial, C., Auvinet, B. and Lepage, O. (1994). Heart rate and blood lactic acid concentration of the horse during swimming training. The Equine athlete (USA).

Jones, J.H. and Hiraga, A. (2008). Metabolic, cardiovascular and respiratory responses to swimming in horses. In Proceedings of a Workshop on Exercise-induced pulmonary haemorrhage: State of current knowledge. Havemeyer Foundation Monograph Series no. 20 p. 34.

Thomas D. P., Fregin G. F., Gerber N. H., Ailes N. B. (1980). Cardiorespiratory adjustments to tethered-swimming in the horse. Pflügers Archiv: Europ. J. Physiol. 385, 65-70 Swanstrom, O.G. and Lindy, M. (1974). Therapeutic swimming.

Walmsley, E., Steel, C., Haines, G., Lumsden, J. and O'Sullivan, C. (2011). Colic after swimming exercise in racehorses: an investigation of incidence, management, surgical findings and outcome. Australian veterinary journal, 89(5), pp.180-183.

Consideration of the difference between galloping and swimming in terms of physiological and biomechanical demands raises certain questions; does swimming exercise improve hind limb muscular strength in a way which improves performance during galloping? Does swimming decrease the likelihood of musculoskeletal injury and should swimming exercise be additional to galloping or should it actually replace galloping within training programmes? Before these and other questions could be answered through scientific study, it would be useful to know a little more about how trainers currently use swimming within training programmes.

A pilot study carried out by Hartpury College and the Animal Health Trust investigated routine swimming practices amongst racehorse trainers. We were primarily interested in the frequency and duration of swimming sessions and to find out why trainers chose to swim their horses. We surveyed 13 Thoroughbred racehorse training yards where a total of 1,101 horses were swum, made up of a roughly 50:50 split of National Hunt and Flat racehorses. Nearly all yards surveyed had over 10 years’ experience of swimming horses. Of the 13 respondents, 9 yards had circular pools, 2 had straight pools and 2 yards had a pool which encompassed both. When a horse was first introduced to swimming, the swim session lasted on average just 5 mins (range 2 to 15 mins). For an experienced swimmer, the session lasted 16 mins on average (range 2 – 25 mins).

Post swimming colic is one of the risks associated with swimming horses. 50% of the respondents to our survey had horses which had experienced swimming associated colic (SAC) within the previous 12 months, varying in incidence from just ‘a couple’ to ‘quite regularly’. SAC incidence is reported to be less than 1 every 1000 swims (Walmsley et al. 2011), however for large yards swimming every day this could be up to one SAC per week. There was no indication from our study that any particular swimming practice was more or less likely to result in SAC. Some trainers simply opted not to swim horses that had experienced post swim colic, which would tend to reduce their incidence of SAC, whilst others felt that SAC was more likely if a horse became fatigued as a result of swimming.

We were surprised to find that 85% of yards surveyed (including all the National Hunt yards) swam all year round. Given that none of the pools were heated; horses are exposed to quite a wide range of water temperatures. None of the yards monitored the water temperature of their pools. Most yards employed rigorous warm up and cool down routines, with over half the yards putting horses on a horsewalker as part of their pre-swim and post swim routine.

We found that the most common reason for swimming was to exercise horses that were lame, and not in fact for training purposes. On average, racehorses swimming for training purposes alone were swum 6 sessions a week (i.e. once per day) alongside daily gallop work, but horses suffering from lameness and unable to be ridden were swum up to 14 sessions a week (twice each day). Eight of the 13 yards swam horses with back pain. Given that swimming is primarily used as a means of managing and rehabilitating horses recovering from lameness and back pain, further studies on the biomechanical and muscular demands of swimming would be useful not only to improve horse welfare but to enable trainers to minimise the number of days lost to injury whilst maximising the benefit of this very expensive, but enigmatic, training tool.

Animal Therapy Magazine