Skeletal injuries are common in athletic horses. Brian D. Nielsen, Department of Animal Science at Michigan State University, looked at over 30 years of research, primarily in racehorses who are subjected to the greatest strain on leg bones from a young age. Besides demonstrating how research evolves and how one project can lead to another, the research provides recommendations, supported by science, on how to decrease injuries in athletic horses, with implications for humans as well.

Studies included the role dietary silicon can play in racehorse injuries, an observation of mineral loss from the cannon bone after the commencement of training, loss associated with horses being removed from pasture and placed into stalls, the use of short sprints to maintain or increase bone strength, and how endurance exercise without high-speed exercise fails to cause bone to become stronger.

Following is a brief synopsis of  A Review of Three Decades of Research Dedicated to Making Equine Bones Stronger: Implications for Horses and Humans.

Dietary Silicon

The initial project began with the examination as to whether supplementing a bioavailable source of silicon (Si) could decrease injury rates in equine athletes. With 53 Quarter Horses in race training in the blinded and placebo-controlled study, benefits from supplementing sodium aluminosilicate (SZA), the Si source, were documented.

All three supplemented groups (low, medium, and high dosages) had more horses complete the required race program (nine races scheduled two weeks apart) than were injured compared to the control group which had more horses injured than were able to complete the study without injury. (It should be noted, injuries were not of catastrophic nature but simply were injuries that required the horses to miss days of training.) Further, the medium and high dosage groups completed substantially greater distances in training before experiencing an injury or completing the project if no injury occurred compared to the control group.

Finally, the medium treatment group had a faster average race time than the control and low Si groups at the race distance of 320 m. While supplementation was not believed to make horses faster, it was concluded that faster individuals within the medium treatment group were able to better withstand the rigors of race training without experiencing injury.

Several future studies into possible reasons for the decrease in injury rates looked at supplemented vs unsupplemented yearlings and post-partum broodmare. These two studies suggest that an altered rate of bone turnover may have influenced injury rates by allowing for a more rapid rate of bone repair.

There were concerns about the amount of aluminum (Al) that was present in the SZA supplement, which prompted the evaluation of another Si source without Al ‒ oligomeric orthosilicic acid (OSA), which appeared to be a viable option to provide dietary bioavailable Si without adding substantial amounts of Al to the equine diet. Another option may be a mineral supplement from a marine source. In that study, yearling horses given the marine mineral supplement had enhanced bone turnover.

In another study, retired Standardbred racehorses were supplemented with a source of bioavailable Si to examine whether it could affect lameness, particularly through its potential role in collagen synthesis. In this 84-day study, 10 horses were assigned to a Si-supplemented group or control. No differences in lameness examination scores, radiographic scores of joints, or other indications of collagen degradation or synthesis were observed. With the mean age of horses being over 10 years, this study questioned whether supplementation provides little benefit in aged animals. It also questioned whether the dosage provided in a commercially available form was insufficient to elicit a response.

Combined, these studies suggest the importance of careful scrutiny of commercial products that make claims based on studies that were not performed on their products, or that recommend dosages not shown to produce benefits in published research.

Bone Loss Associated with Stall Housing

The initial study mentioned above utilizing 53 horses in race training revealed a surprising result. When horses commenced race training, the RBAE (an estimate of bone mineral content) had decreased by day 62 of training. The RBAE remained low through day 104 of the study, but had begun to increase by the conclusion of the study at day 244.

It was not surprising that the greatest injury rates were occurring at the time when bone was the weakest and horses were beginning to race.

These findings were surprising, as most physiological systems are typically believed to increase in strength when athletic training commences (cardiovascular, muscular, respiratory). Thus, to find the skeletal system losing bone mass was confusing and alarming. Further, it should be noted that most bone-related injuries happened between days 60 and 120 of training when the bone mass was at its lowest. Accompanying this, horses began racing during week 9 of the study. While it was surprising that horses were losing bone mass during the initial stages of training, it was not surprising that the greatest injury rates were occurring at the time when bone was the weakest and horses were beginning to race. The combination of fast speeds and weak bones logically results in injuries.

It was noted that prior to entering race training, the young horses had been moved from pasture housing and placed into box stalls. By eliminating access to free exercise and having no fast exercise during the early part of training, it was hypothesized that the loss of bone may have been caused by the lack of loading on the skeleton.

At the time, it was not commonly believed among horsemen that confinement had anything to do with bone loss. Luckily, one of the researchers in the latter two studies had no previous horse experience that could bias her, although she had extensive experience with bone loss in human bed-rest patients. Lacking horse experience kept her from having any preconceived notions based upon how things have traditionally been done with horses.

To test the hypothesis that stalling of horses with no access to high-speed exercise was responsible for bone loss, 16 Arabian yearlings were randomly divided into two groups. Half of the horses remained on pasture while the other half were moved into box stall housing with one hour of walking daily on a mechanical walker. By day 28 of the study, the RBAE of the third metacarpus of box-stalled horses had decreased and remained low throughout the 140-day study.

It was concerning to note that horses kept in stalls had lower bone mass at the end of the nearly five-month study than they did when they started it.

Even when horses were started under saddle and began race training after 12 weeks in the study, no increase in bone mass was seen during 8 weeks of slow racetrack training. Markers of bone formation were lower and markers of bone resorption were higher in the confined horses at days 14 and 28 compared with the pastured horses, and both markers subsequently returned to baseline in the confined horses. Those alterations suggest bone formation decreased and bone resorption increased in the confined horses, thus explaining the loss of bone.

With this study closely mirroring how yearling racehorses are managed while being prepared for sales and during the early training, it was concerning to note that horses kept in stalls had lower bone mass at the end of the nearly five-month study than they did when they started it. The results suggested a probable cause for the high injury rates observed in young horses managed in that fashion.

Another study examined bone mineral content in 11 mature Arabian horses (ages four to seven years) that had been previously conditioned but were then placed into box stalls for 12 weeks. Despite being walked on a mechanical walker daily and being fed dietary Ca at twice the 1997 Nutrient Requirements of Horses-recommended amounts, horses lost bone mass.

To investigate the question as to whether having horses housed on pasture completely (as opposed to partially) was necessary to avoid bone loss, 17 weanling Arabian horses were randomly assigned to three treatment groups: (1) housed on pasture, (2) housed in stalls, and (3) housed in stalls for 12 hours per day and housed on pasture for 12 hours per day. After 56 days, greater increases in bone mass of the third metacarpal were observed in both groups allowed access to pasture, as opposed to the group confined completely to stalls, suggesting that even partial turnout could prevent bone loss associated with disuse.

Speed and Bone Health

A study utilized 18 weanlings divided into three groups: (1) group-housed, (2) box stall-confined with no exercise, and (3) box stall-confined with a daily sprint of 82 m five days a week. The eight-week study saw increased bone mineral content and altered bone geometry in the confined horses that were allowed to sprint short distances compared to the confined horses that experienced no sprinting.

It should be noted that not all exercise produces similar benefits. For instance, circular exercise such as lunging is commonly done with horses. While believed to be detrimental to joint health, various bone parameters have been shown to be altered between inside and outside legs of juvenile bull calves exercised in only one direction five days per week in a study lasting seven weeks. Further, bone responds more to mechanical strain (the amount of bending) than it does the number of times it is bent.

With this in mind, it was not surprising that, during a conditioning study lasting 78 days, progressively increasing the amount of weight carried up to 45 kg resulted in increased RBAE in young horses exercised in a walker compared to control horses that were exercised similarly, but without carrying supplemental weight. Prior to entering the conditioning period, all horses were stall-confined for 108 days, during which time they had lost bone mineral content of the third metacarpal.

Housing horses on pasture does not guarantee they will perform exercise necessary to enhance bone strength, but it does increase the likelihood of it.

Likewise, while it has long been believed that months of slow training will increase bone strength and prevent injuries, science does not support this. Using 11 two-year-old Arabians that were split into two groups, the influence endurance exercise has on bone mineral content was examined. One group was trained on a high-speed treadmill for 90 days, then placed on a regular exercise schedule including a 60 km endurance test every three weeks. The other group served as controls and lived on pasture without any forced exercise. At day 162, no differences were seen between treatments in RBAE, suggesting endurance exercise does little to alter bone density compared with free-choice exercise on pasture.

Housing horses on pasture does not guarantee they will perform exercise necessary to enhance bone strength, but it does increase the likelihood of it. By contrast, if confined to a stall and never afforded the opportunity to run, it can be assured that skeletal strength will be compromised.

Pharmaceutical Factors

Research also suggests caution be taken when providing pharmaceuticals that may inhibit mineral absorption or enhance mineral loss, specifically bisphosphonates or corticosteroids. Masking pain when an injury is still present increases the chance for greater injury, potentially even catastrophic in nature, to occur.

In recent years, there has been growing concern regarding the use of bisphosphonates, approved in 2014 to treat navicular disease in horses over the age of four years. Bisphosphonates inhibit osteoclasts, whose function is to resorb bone, raising concern regarding whether this will impair bone healing, leaving horses, particularly young horses in training, more susceptible to injury.

Furosemide is commonly given to racehorses in North America to decrease the incidence of exercise-induced pulmonary hemorrhage. As the usage of such has been shown to negatively impact Ca balance for several days after administration, concern existed as to whether it could have detrimental effects on bone. A study involving ten horses, serving as both controls and furosemide-treated, during two eight-day periods of urine and feces collection showed that Ca balance returned to baseline in three days after furosemide administration and thus it does not present much risk to bone.

Likewise, omeprazole has been commonly provided to aid in healing or preventing gastric ulcers in horses, but concerns have existed whether the suppression of gastric acid may inhibit absorption of Ca and thus impair skeletal health. Using a preventative dose (1 mg/kg BW daily) for up to two months provided no indication of such, although usage for longer periods or at the treatment dose (4 mg/kg BW daily) could still pose a risk.


  • An initial study investigating the role of bioavailable silicon in the diets of horses in race training produced the unexpected finding of decreased bone mineral content of the third metacarpus subsequent to the onset of training.
  • Further studies revealed this decrease to be associated with stall housing eliminating high-speed exercise, leading to disuse osteopenia.
  • Only relatively short sprints were necessary to maintain bone strength and as few as one sprint per week provided the needed stimuli.
  • Endurance exercise without speed fails to elicit the same benefits to bone.
  • Proper nutrition is also required for optimal bone health, but without the right exercise, strong bone cannot be maintained.
  • Several pharmaceuticals may have unintended consequences capable of impairing bone health.
  • Many of the factors influencing bone health in horses also exist in humans including a sedentary lifestyle, improper nutrition, and pharmaceutical side-effects.

Read the entire paper here.