Equestrian surface biomechanics and the effects on the horse’s musculoskeletal system.

The protocol was applied to a harness racing trotter, and to a ridden sports horse (see figure 1), and was based on the simultaneous use of different measuring instruments:

• A 3D dynamometric shoe [Chateau et al., 2009b], and a 3D accelerometer attached to the horse’s hooves. Up to 2009, only one front foot was equipped ; since 2010 both front feet or one front and one hind foot are equipped simultaneously ;
• An ultrasound sensor, which enables to assess the force in the superficial digital flexor tendon (SDF) [Pourcelot et al., 2005 ; Crevier-Denoix et al., 2009a et b] ;
• One or two high frequency cameras (600 to 1000 Hz) ;
• A device to measure the horse’s speed (GPS on a horse being ridden).

All these measurements are synchronised. During the experiments, the horse’s speed was standardized to allow a comparison between the different track or arena surfaces.

The protocol was used to assess and compare around 20 different types of surface (e.g. waxed fibered sand track, crushed sand, dirt, clinker, beach sand…) and to assess the impact of maintenace procedures on the surfaces.

The results shown here were compiled from a same horse, tested at the same speed on 8 different training tracks and a hard pathway.

Measuring instruments used on the harnessed trotter

The instruments used and synchronised with one another are as follows :

A- a 3D dynamometric shoe, equipped with 4 triaxial sensors, measuring the reaction force in the ground in 3 directions, the moments and position of the pressure centre ;

B- a triaxial accelerometer, attached to the hoof wall, measuring the decelerations (shock on impact) and the vibrations generated by the shock ;

C- an ultrasound sensor to assess the force exerted on the superficial digital flexor tendon ;

D- inertia measuring devices (or accelerometers). Placed along the back (especially at the withers and the rump), these devices quantify dorsal mobility and locomotor dissymetries (lameness) ;

E- a high frequency kinematic system, which analyses joint movements in the limb, the orientation and sliding of the hoof. The high frequency camera, is positioned in a vehicle which follows the horse as it moves.

Impact shock (maximum vertical deceleration)

Of the 8 tracks tested, those made of waxed fibered sand gave the most damping, whereas the crushed sand were the firmest training surfaces.

The results on the clinker tracks, where the results were strongly influenced by the level of maintenance, appear to be intermediate sufaces. The pouzzalone track (had been well tilled on the day of the test) was noted as being highly damping (shock absorbing).

The hard pathway tested (access lane to the training tracks, not tilled) was dinstinctly different from the training tracks in that the intensity and variability of impact shocks was much higher (Maximum vertical deceleration of around 5000m/s2,  which is around 500 times the acceleration of gravity) (see figure 2)

Maximum longitudinal force

A deferred longitudinal force peak shows a delay in the immobilisation of the foot ; it implies a longer phase where the hoof sinks and slides into the ground at the beginnig of the support phase.

Speed at maximum load

The speed at maximum vertical load of a limb during the support phase (slope of the ratio of vertical-time force at the end of damping) is probably the criteria which is most relevant to qualify the idea of « trauma risk » linked to a surface (see figure 4)

Once again the pathway is very different from all of the training tracks (with an average value  for the ratio speed at maximum vertical load of around 16 tonnes per second). The other tracks are all fairly similar to one another, again with the waxed fibered sand and pouzzolane surfaces appearing to provide more shock absorbtion.

In 2009, the experimental protocol was adapted to a horse being ridden, 6 horses were thus tested at a gallop (30km/h, in a staight line) on the race track at Deauville-la Touques. The waxed fibered sand track was thus compared to a turf track.

In addition to the biomechanical parameters presented for trotter horses, new variables were calculated thanks to the improvements added to the protocols for kinematic analysis (see figure 5) and to the processing of the accelerometric and dynamometric data.

The additional variables are relative in particular to the speed of the hoof as it strikes the ground, its slide and penetration while braking, and to the orientation of the hoof and limb as a whole with relation to the track during the support phase.

Measuring equipment used for the tests on a horse ridden at a gallop

B- 2 triaxial accelerometers are also attached to the hoof wall of the same feet ;

C- an ultrasound sensor assesses the force exerted on the superficial digital flexor tendon (of the front right limb) ;

D- a GPS system, with the antenna placed on the horse’s rump, enables to measure the horse’s speed ;

E- recording devices are fitted on the horse’s and the rider’s backs ; they are controlled from a distance via a Wifi connection.

The track itself is marked out with kinematic markers which will serve to calibrate the measurements carried out with the two high frequency  fixed cameras, placed on the side of the track.

Interaction between the surface and the foot at the beginning of the support phase

Vertical and horizontal hoof speed on impact are significantly higher on turf, whether on a right leg lead or on a left leg lead gallop. The vertical shock of impact is more violent on turf (absolute peak value of the vertical deceleration significantly higher).

The time lapse before immobilisation of the hoof is substantially longer on the waxed fibered sand, and the vertical penetration of the hoof between impact and immobilisation is also significantly longer on the waxed surface. The horizontal deceleration peak (absolute value of the horizontal shock), the rebound which follows immobilisation and the maximum longitudinal force (braking) are substantially higher on turf.

Loading phase of a limb

Moreover, the maximum vertical loading speed of the limb (curve showing the vertical force- time ratio) and the vertical impulsion (area under the curve of the same diagram) ara also significantly lower on the waxed sand. It can also be noted that the vertical force peaks quite a bit later on the waxed sand surface.

Figure 6 ⇒ comparison between:

• Turf (green curves)
• Waxed sand surface (yellow curves).

Gallop left leg lead (30km/h), right leg fitted with measuring equipment.
On the waxed sand surface, the maximum vertical force (Fz) and maximum « braking force » (Fx) peaks have a lesser amplitude, and occur later than on turf.

Kinematic analysis

The kinematic measurements showed that the loading phase of the weight bearing limb is more gradual on the waxed fibered surface : the maximum angular speed peaks for the fetlock, carpus and elbow extension, and for the interphalanx and scapulohumeral flexion (shoulder, are significantly lower on the waxed surface during the first half of the support phase. For the distal joints, these observations are completely in tune with the angle of the hoof with regard to the track (larger variation of the angle on the waxed surface, and deper penetration of the heel in the first part of the support phase).

Inversely, push off is less effective on the waxed surface : when the limb leaves the ground at the end of the support phase, it is substantially more vertical (less oblique towards the rear) than on turf, the fetlock angle is less open, the elbow less extended.

This « relative delay » in the support phase on the waxed surface persists until the heels leave the ground (also with more delay time, expressed as a % of the support phase duration, on the waxed surface).

With the trotter in harness

The results, with the trotter in harness, regarding the biomechanical parameters which seem to be the most discriminating : vertical shock (maximum vertical deceleration) of the hoof on impact, the maximum longitudinal force (maximum horizontal force at the end of the slide), the corresponding time lapse (i.e. the lapse before immobilisation of the foot) and the speed of maximum vertical loading of the limb during the support phase. The waxed fibered sand surfaces were considerably more damping (shock absorbing) : they induce maximum deceleration of the foot on impact (shock), maximum longitudinal force (« braking » force, measured at the end of the slide when the foot becomes immobile) and maximum loading speed of the limb and of the superficial digital flexor tendon to be less than on traditional surfaces such as crushed sand. However immobilisation of the foot in the ground takes longer (sliding and ground penetration phase is longer) the length of the horse’s strides tends to be shorter (and the frequency of the strides is higher) and the relative duration of the support phase (ratio of the duration of support phase with regard to the duration of the stride) is also higher, which could be the basis for the sentiment that these tracks are « slower »/ «lessen performance » .

Tilling and sub layer

Tilling (dirt or sand tracks) considerably reduces the shock on impact (including the vibrations induced by the shock) as well as the maximum braking force. It was also demonstrated that the hardness of the under layer of a track (beneath the top 8 cms) has a significant impact on the maximum vertical force (in the middle of the support phase) as well as on the speed of maximum vertical load of the front limbs during  the support phase.

Water content

The influence of the water content in beach sand on the biomechanical variables was tested with 4 trotter horses on the beach at Varaville (Calvados). The comparison between a firm, wet sand (close to the water line, with  an average water content of around 20 %) with a deep wet sand (parallel to the first area but about 20 meters higher up, with an average water content of around 13,5 % ), had several significant effects : lessening of the length of the stride, increase in their frequency, and increase in the duration of the support phase (associated to a shortening of the swing phase) were observed in the deep wet sand, compared to the firm wet sand. Impact shock, more generally the forces exerted  (maximum braking force and maximum verical force) and the corresponding loading speeds, are lesser on the deep wet sand, which confirms the damping qualities of this type of surface.

However, during push off (around 70 to 85 % of the support phase), the tendancy switches round and longitudinal (push off) and vertical forces are higher, probably due to increased contraction of the propulsor muscles in the limb, needed to compensate the lack of responsivness in the deep going. However, this increased effort is in fact less effective since the fetlock joint remains more extended at the end of push off phase (during the last 25 % of the support phase) on deep sand compared to firm sand, a situation which could present a risk for the SDF tendon.

With the horse ridden at a gallop

The first results of tests on galloping horses confirmed the damping quality of the waxed-fibered sand surfaces (already demonstrated for the trotters).
Anatomical formations (bones, tendons, joints) in the distal part of the front limb are therefore less sollicited on this type of track than on the traditional turf tracks (which were tested on average in « good » conditions), particularly during the shock absorbtion phase. This effect is mainly due to greater deformation of the surface. However the push off effort supplied by the horse at breakover is probably greater, and less effective than on a turf track.

This is the first time that significant kinematic differences on the joint angles, and the angular speeds have been demonstrated with regard to the two types of track surface, those which are used the most frequently for flat racing worldwide.

Indeed, in the only other study to compare the kinematics of horses at a gallop on different tracks (turf vs dirt vs waxed fibered sand), our colleagues of the Veterinary school of Davis (California, USA ; Setterbo et al., 2008) only demonstrated differences in time scales (confirmed by our work). This statement of fact shows the performance level of our protocol, specifically regarding the superior resolution of our measurements.

When jumping

In 2010, the show jumping arena surfaces at Cabourg were compared in a first test during show-jumping (conditions tested : jumping and sharp turns).

The protocol was improved and then applied to the three arenas at the International equestrian facility in Deauville at the end of October 2010.

The analysis of these experiments is under way, but has already confirmed the differences between the leading and non leading front limbs during landing after jumping an obstacle (1m high). The maximum vertical force, and maximum load speed were higher for the non-leading leg compared to the leading leg ; the non-leading leg contributes more to push off than the leading leg, in which the braking phase is predominant. Moreover, the preliminary results of the study show that the duration of the support phase and the « impulsion » (area beneath the curve showing force) are greater on the leading leg, a factor which should be taken into account when examining show-jumpers presenting lameness or lack of performance.