Influence of shoes with different weights on the motion of the limbs in Icelandic horses during toelt at different speeds



Reason for performing study: Weight boots are commonly used for Icelandic horses to increase the height of the flight arc of the forelimbs in toelt.

Objective: To show the influence of weights and toelting speed on the height of the swing phase.

Materials and methods: Eight Icelandic horses (mean ± s.d. 12 ± 3 years old, 369 ± 46 kg) were used. Reflecting makers were placed on the dorsal side of each hoof. The motion was collected with a kinematic system (10 cameras, 120 Hz sample rate, 1.3 Mpixels resolution). The horses were ridden in toelt by 2 experienced riders on a treadmill at 2 different speeds (2.96 m/s ± 0.30 and 4.10 m/s ± 0.32). At each speed the horses were measured wearing no boots, light boots (170 g) and heavy boots (280 g) on both fore hooves. The measurement sequence was varied between horses. A Kolmogorov-Smirnov test was carried out to test for normal distribution of data and ANOVA for repeated measurements were used to compare differences (P<0.05).

Results: The weight as well as the speed of toelt had a significant influence on the height of the flight arc. At the lower speed, the mean ± s.d. height was 163 ± 55 mm, whereas at the higher speed the mean height was 228 ± 60 mm. The heavy weights increased the mean height at the lower speed from 152 ± 38 to 169 ± 48 mm and at the higher speed from 214 ± 60 to 245 ± 60 mm.

Conclusions: This investigation shows that Icelandic horses can be expected to show a better toelt in competitions with weights, and ridden at a higher speed. For muscle adaptation to occur, weights should therefore be used during competitions and training.


Icelandic horses are strongly influenced of Nordic and middle European breeds (Adalsteinsson and Hampel 1998) and they have developed (or maintained) a fourth gait, the so-called toelt. This is a symmetric 4 beat gait in which the pattern of stance phases is right hind, right front, left hind, left front. This gait was first studied by Hildebrand (1965). Over the last decades, the Icelandic horse has become very popular in Europe, but scientific literature on the toelt has until recently remained scarce.

Since 2000 several scientific papers have been published (Zips et al. 2001; Biknevicius et al. 2004, 2006; Robilliard et al. 2007; Starke et al. 2009). The precise sequence of footfalls in toelt has been documented (Robilliard et al. 2007). The intervals between the landing of limbs should be of similar duration, and single support phases and double support phases should alternate. Double support phases of fore- and hindlimbs are either diagonal or ipsilateral and these should also be of approximately similar duration. Toelt is considered to be a gait without a suspension phase, even though a very short suspension phase was measurable in most Icelandic horses at toelt (Zips et al. 2001). In toelt competitions speed and the height of the flight arc of the hoof are crucial to achieve high marks. The regulations state that, in the speed chosen by the rider, the horse should be supple with high and long strides and have full engagement of the hindquarters and harmonious carriage (Anon 2009).

Icelandic horses are ridden using a special set of tack, with saddles fitted to their body shape and simple bridles. Many Icelandic horses are born with the ability to toelt, even though they may not show this gait unsupported. The development and perfection of the horse's own natural ability to toelt is the largest part of the training, and the success of this toelt training plays an important role in the horse's riding and breeding value. For the accentuation of the swing phase, weighted boots are popular as training aids, as experience has shown that these boots raise the flight arc of the forelimbs. Although there is a lot of equestrian information regarding the training of horses with these boots, no scientific investigation into their effect has been carried out.

There are a number of studies investigating the effect of boots in other horse breeds. In showjumpers, weighted PE boots induce hyperflexion of the hindlimbs and incline the horse to jump fences cleanly (Murphy 2008, 2009). In forelimbs, protective or supportive boots reaching from the mid-pastern region to the mid cannon were shown to reduce maximum extension of the fetlock at the trot (Kicker et al. 2004). Clayton et al. (2008) showed that a mild tactile stimulation of the coronet in itself leads to a higher flight arc of the hooves in horses. This indicates that, besides the mechanic effects of boots, an independent neural response (feedback mechanism) also exists, which leads to hyperflexion.

The aim of the present study was to investigate the influence of weights and speed on the height of the flight arc of the forelimb of Icelandic horses at toelt. The hypothesis was that there was a positive correlation between the height of the flight arc and both the weight added to the hoof and the toelting speed.

Materials and methods


Eight adult Icelandic horses (mean ± s.d. 12 ± 3 years, 369 ± 46 kg) without any sign of lameness were used for this investigation. The horseshoes were in accordance with the regulations of the Austrian Icelandic Horse Association (Anon 2009b). The horses were well-trained and ridden several times a week. The horses selected had a placid temperament suitable for this kinematic analysis. During 2 training sessions, horses were accustomed to the treadmill (Mustang 2200)1. Horses that were not able to toelt under a rider on the treadmill were excluded from the study.

Data collection

Riders: Two experienced competition riders rode the horses on the treadmill1 at toelt with and without weighted boots. Each rider rode 4 of the 8 horses.

Boots: Two different weighted boots2 were used. Each horse was measured without boots (0 g) and with weighted boots of 170 g and of 280 g placed on the fore hooves (Fig 1a).

Figure 1a.

The weight boots.

Markers: On each hoof, a passive reflective marker was placed on the dorsal hoof wall and secured in place using adhesive tape. A second marker on the distal edge of the dorsal hoof wall was placed for an additional study, but was not considered for this study (Fig 1b). Additional markers were placed on the lateral aspect of the fetlock joint, the cannon bone and on the radius, to allow determination of the angles of the fetlock and carpus joints.

Figure 1b.

Marker set-up on the dorsal hoof wall.

Training procedure

After the clinical investigation of the horses, horse and rider were allowed a short period of warm-up in the indoor riding school. Then the horses were trained to toelt on the treadmill, initially without and subsequently with a rider. The range of possible toelting speeds (lowest possible toelt speed [slow] and highest possible toelt speed [fast]) was determined for each horse. Only horses displaying a stable toelt pattern at 2 clearly different speeds were selected for the study, and the markers were attached.

Three-dimensional kinematic data were collected using 10 infrared cameras (Eagle Digital RealTime System)3 recording at 120 Hz.

The horses were analysed in the 2 previously determined speeds, without (0 g) and with weighted boots (170 and 280 g). Therefore, a total of 6 measurements were carried out per horse. Data were captured in each setting over a period of 40 s (2 times 20 s). At least 20 motion cycles of each horse and each set up were analysed. A random order of measurement settings was used.

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 (Cortex 1.3)3. These time series were then smoothed by use of a Butterworth low-pass filtered (cut-off frequency, 20 Hz). The data were split into motion cycles by MATLAB R2008b4 starting with the beginning of the right fore swing phase and the duration of the motion cycle was calculated. The maximum height of the flight arc of the hoof was determined. Angles of the fetlock and carpus joints were calculated in the sagittal plane.

Also the maximum potential energy (Nm) was determined using the following equation: Epot= hmax×g× m with hmax the maximum height of the flight arc of the hoof, g the gravitation (9.81 m/s2), and m the mass of the distal limb (from the carpus) set at 2.5 kg (Buchner et al. 1997) (with added weights 0 g/170 g/280 g).

Statistical analysis

Statistical analyses were done with the SPSS 17.05 software. The data were tested for normal distribution applying a Kolmogorov- Smirnov test. The height and speed were evaluated using the ANOVA for repeated measures with a Bonferroni post hoc test (P<0.05).


In the 8 horses, the slow speed of toelt without boots was mean ± s.d. 2.96 ± 0.30 m/s. There were no significant differences between the toelting speeds of the trials without and with both types of weighted boots (0 g: 3.07 ± 0.47 m/s; 170 g: 3.00 ± 0.31 m/s; 280 g: 2.80 ± 0.44 m/s).

The fast toelting speed without boots was 4.10 ± 0.32 m/s. Again there were no significant differences between the toelting speeds of the trials without and with weighted boots (170 g: 4.20 ± 0.64 m/s; 280 g: 3.99 ± 0.34 m/s). The height of the arc of the fore hoof flight was significantly increased with speed and with weighted boots, while no significant changes of the hindlimbs were found (Table 1, Fig 2). The maximum potential energy was significantly increased with speed and with weighted boots (Table 2). There was no significant difference between minimum angles in forelimb fetlock and carpal joints in the 6 settings of the study. The difference between minimum and maximum angles of the forelimb fetlock joints and of the carpal joints was positively correlated in the 6 settings used in this study, with higher weights and faster speeds correlating with increased flexion of the fetlock and carpal joints. No significant difference in the range of fetlock and carpal joint angles was found (Table 3).

Table 1. The results of 8 horses at slow and fast toelting speeds ridden on a treadmill, without (0 g) and with weighted boots (170 and 280 g). The mean ± s.d. maximum height of the flight arc of the fore- and hindlimbs is shown. Significant differences were present in the forelimbs between each of the 6 different settings. No significant differences were determined in the hindlimbs
SpeedWeight (g)Height forelimb (mm)Height hindlimb (mm)
Slow0152 ± 3891 ± 28
170167 ± 57100 ± 27
280169 ± 4696 ± 24
Fast0214 ± 6096 ± 32
170225 ± 5897 ± 28
280245 ± 59100 ± 33
Figure 2.

The results of 8 horses at slow and fast toelting speeds ridden on a treadmill, without (0 g) and with weighted boots (170 and 280 g). The increase of the height of the flight arc of the fore limbs is shown. Significant differences were present in the forelimbs between each of the 6 different settings.

Table 2. Calculated mean ± s.d. maximum potential energy of 8 horses at slow and fast toelting speeds ridden on a treadmill, without (0 g) and with weighted boots (170 and 280 g). The mass of the distal limb was set at 2.5 kg. Significant differences were present in the forelimbs between each of the 6 different settings. No significant differences were determined in the hindlimbs
SpeedWeight (g)Potential energy
Slow03.73 ± 0.93
1704.38 ± 1.45
2804.62 ± 1.28
Fast05.24 ± 1.47
1705.90 ± 1.57
2806.67 ± 1.62
Table 3. Mean ± s.d. difference between the minimum and maximum angles (°) of the forelimb fetlock and carpal joints of 8 horses at slow and fast toelting speeds ridden on a treadmill, without and with weighted boots (170 and 280 g). A positive correlation was present between carpal and fetlock joint angles in the 6 settings
SpeedWeight (g)Carpal jointFetlock joint
Slow093.91 ± 19.1498.07 ± 12.81
170101.95 ± 14.65105.81 ± 15.95
280102.95 ± 12.25105.66 ± 28.76
Fast0101.92 ± 27.90101.81 ± 16.12
170107.94 ± 7.67109.22 ± 12.75
280113.24 ± 9.48105.05 ± 19.82


The weight most commonly attached to the distal limb of horses is the shoe, usually a steel shoe. Willemen et al. (1999) observed that the weight of the shoe primarily affects variables in the swing phase, at high speed. Our study showed that even at the slower toelting speed, the swing phase significantly changed using weights.

In human walkers, the attachment of a 2 kg weight to one distal limb (Noble and Prentice 2006) led to a change in both left and right limb kinematics when walking on a treadmill, a situation they named ‘calibration’ and after removal of the weight ‘re-calibration process’. In the horse, a similar change in the use of the hindlimbs can be expected, even if the kinematics used in our study did not show this effect. Further studies, investigating the centre of body mass and the precise placement of the hindlimbs during exercise with weighted forelimbs will be interesting.

Ankle weights are commonly used in human resistance training (Morrissey et al. 1995) and rehabilitation (Jung et al. 2010), where they have been shown to change the muscle recruitment (Fiebert et al. 2001) of the limb. In the aforementioned study of Noble and Prentice (2006) the mass of 2 kg led to a reduction in knee angle at the walk on the treadmill, which is different from our results with a maximum of 280 g. This difference is probably as much due to the differences in the weight used, as it could be the consequence of the difference between the species and the joints that are involved.

In the present study, the carpal joint and the fetlock joints were correlated with each other when comparing the 6 different settings, but there were no significant differences between the weight groups. Therefore, the main activity for the increased flexion of the whole limb appears to have been done by the elbow and shoulder joints. However, it is also possible, that the small number of horses with moderate interindividual variation between limb angles investigated in our study did not allow us to identify a smaller difference between weighted and nonweighted scenarios.

The weight of the boots changes the swing phase and the landing of the weighted limbs compared to the nonweighted limbs. Setterbo et al. (2009) showed the influence of the surface for the impact, with a mean acceleration on turf surface of 42.9 g. Clearly, additional weight on the hoof increases the impact. Using the Newton's Second Law (Force = acceleration × mass = 42.9 × 9.81 × 0.28 = 118 N), the impact while using boots (0.28 kg) will be increased by 118 N. Additionally the potential energy increased with speed and weights from 3.7–6.7 Nm in our study. Furthermore the kinetic energy will increase because of the increasing mass and the faster footfall. This increased energy will lead to a higher load on the contra-lateral limb in the stance phase and on the upper limb joints in the swing phase.

Due to the additional weight the initial lift-off of the limb is higher, as the pull of the body on the limb has to overcome a larger gravitational pull. This pull consists of the muscular contraction of the upper limb as well as the passive pull of the trunk moving away from the limb at stance. Muscular adaptation is necessary, because an increased force must be produced from the biceps brachii muscle, leading to potential overload injuries of the bicipital bursa and tendon.

A variety of methods has been employed to improve the quality of the Icelandic horse's toelt. The tactile stimulation of a lightweight device (55 g) with 7 chains attached to the pastern has been shown to increase the height of the flight arc of hooves (Clayton et al. 2008) until a certain habituation occurred. Our findings show that the forelimb in toelting Icelandic horses reacts similar to the hindlimb in showjumpers (Murphy 2009), as hooves were lifted higher in showjumpers with weighted boots than without them.

Enhancement in performance may also increase the risk of tissue damage and/or overexertion of the competition horse's lumbar musculature where weighted boots are used in the hindlimbs of showjumpers (Murphy 2009), and similar risks apply to the thoracic and forelimb musculature of the Icelandic horse toelting with weighted boots. In the absence of scientific appraisal, it is unclear whether such boots are acceptable and innovative training aids within equitation (Murphy 2008). Weighted boots influence the motion and the loads of the limbs; therefore long-term investigations will be necessary to ascertain that this does not increase the risk for soft tissue and joint diseases.


We would like to thank Sebastian Peham his support during the data processing.

Conflicts of interest

None declared.

Manufacturers' addresses

1 Kagra AG, Fahrwangen, Switzerland.

2 Topreiter Corp., Austria.

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

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

5 SPSS Inc., Chicago, Illinois, USA.