Abstract
Objective—To quantify changes in hoof wall strain distribution associated with exercise and time in Standardbreds.
Animals—18 young adult Standardbreds.
Procedures—9 horses were exercised 4 d/wk for 30 to 45 minutes at a medium trot for 4 months; 9 nonexercised horses served as the control group. Rosette strain gauges were used to measure the principal surface strains at the toe, lateral quarter of the hoof wall (LQ), and medial quarter of the hoof wall (MQ) of the right forefoot at the beginning and end of the experiment. Midstance maximal (msϵ1) and minimal (msϵ2) principal and peak minimal principal (pkϵ2) surface strains were measured; SDs of each of those variables were also calculated. Results were compared through ANOVA of time and exercise effects between and within the groups.
Results—Both the exercised and nonexercised groups had changes in strain distribution in their hooves over time. The msϵ1 did not change significantly with exercise; however, it changed significantly in both groups at both hoof quarters over time. At the beginning of the study, mean msϵ2 and pkϵ2 values were significantly higher in the exercised group than in the control group at the MQ and LQ but not at the toe. At the end of the study, these values were significantly higher in the control group than in the exercised group at the toe but not at the MQ or LQ.
Conclusions and Clinical Relevance—Detected changes in hoof wall surface strain may indicate the ability of hoof capsule material to respond to exercise. A better understanding of hoof adaptation to applied forces may allow implementation of proper trimming and shoeing techniques to promote adaptation to exercise loads in horses.
In horses, the mechanical properties of the hoof wall are, to a large extent, a function of its structural arrangement. Grossly defined parts of the hoof protect underlying structures of the foot and initiate dissipation of concussive forces on impact when the hoof strikes the ground.1 In response to the imposed stresses, the hoof undergoes elastic deformation. During weight bearing, the dorsal hoof wall rotates caudodistally while the quarters and heels flare abaxially.2 These deformations of the whole hoof capsule cause regional variations in the stress and strain of the material of the wall, which can be recorded as principal strains.
The combination of the horse's weight, high-speed gait patterns, and contact with uneven surfaces augment the chance of developing high localized stress concentrations within the hoof.3 However, the equine hoof wall has a unique architectural design that enables it to withstand the forces generated during weight bearing and through the impact of the hoof hitting the ground. The hoof is a dynamic structure able to respond to changes in its environment, such as changes in loading.4–9 Various factors influence its biomechanics, including but not limited to shape,10–16 shoeing,17–22 substrate,23–26 breed,27 speed,28 and rider.29 Although it has been shown that the equine limb as a whole can adapt to the stresses imposed, much of this attention has been paid to the third metacarpal bone and associated tendons and ligaments,30–33 whereas some of the hoof's responses to exercise have not been documented. There is empirical evidence that the hoof may respond to changes in its environment,6 but to the authors' knowledge, this has not been explored experimentally.
Biological responses of the hoof to loading under conditions of controlled exercise have not been quantified in experimental conditions. The purpose of the study reported here was to quantify changes in hoof wall strain distribution associated with exercise and time in Standardbreds.
Materials and Methods
Animals—Eighteen young adult Standardbreds (6 females and 12 geldings) with a mean ± SD body weight of 461 ± 36 kg were randomly assigned to control (n = 9) and exercised (9) groups. Age, body weight (expressed as body mass), and hoof conformation were not considered in the randomization process.
Study design—The control group received no forced exercise, whereas each horse in the exercised group was exercised 4 times/wk. To properly condition the horses in the exercised group, the duration of exercise began at 10 min/d and progressively increased within a period of 10 days to 45 min/d, which was considered routine exercise. The horses were trotted at an approximate speed of 7 ± 1 m/s (medium trot) on a 700-m straight track surfaced with crushed limestone on a gravel base to mimic a North American harness track. At the end of the track, the horses were randomly turned to the left or right before accelerating to trot again. This process was repeated for the allotted exercise time. Two comparable exercise jog carts were used by drivers of similar body weight (approx 55 ± 8 kg). The study protocol was approved by the University of Guelph Animal Care Committee and conformed to guidelines of the Canadian Council of Animal Care.
Hoof maintenance—All horses' feet were trimmed and shod by the same farrier at the beginning of the study and were maintained on a routine schedule every 6 weeks. The same hoof-trimming protocol was applied to all feet of all horses, and no correctional trimming technique was used. Briefly, the toe was trimmed down to a point at which, from a lateral view of the hoof, the straight alignment through the phalanges and the hoof wall (foot-pastern axis) was preserved. The lateral and medial aspects of the hoof walls and heels were trimmed to conserve mediolateral symmetry. The same type of shoe (flat steel, full swedge [lengthwise indentation in the web of the shoe] Standardbred training shoe) was used for all feet in all horses.a The swedge in the shoe was used for better traction on a stone track. No gait-altering devices such as caulks, trailers, or wedges were used. On the basis of hoof sizes, 3 shoe sizes (00, 1, and 2 weighing approx 166, 178, and 196 g, respectively) were used.
Strain measurement—With the horse standing, 3 rosette strain gauges were attached to the hoof wall of the right forefoot at the toe and at the widest point of the LQ and MQ, approximately 1.3 cm below the coronary band.b The strain gauges were glued to the hoof wall with a thin layer of cyanoacrylate adhesive.c Gauges were placed in such a way that the primary axis was aligned with the hoof wall tubules as they could be grossly seen. The wiring connections originating from the strain gauges were secured to each horse's body and the shaft of a sulky (a light cart with 2 wheels and a seat for the driver) by use of bandaging taped and connected to the data acquisition and logging system computers,e which were placed on the sulky (Figure 1).
Figure 1—Photograph showing the placement of rosette strain gauges (3; arrows) at the toe, MQ, and LQ of a horse for the purpose of measuring changes in hoof surface strain distribution in response to exercise. The wires were secured to the limb and body and connected to the data logger, which was mounted on the sulky.
Citation: American Journal of Veterinary Research 72, 4; 10.2460/ajvr.72.4.484
A wheel-mounted speedometer recorded the rate of wheel revolutions, which was later converted to velocity of the cart. The data were processed by use of a custom-written program in data analysis softwaref that identified the zero line when calibrating. Midstance maximal principal strain, msϵ2, and pkϵ2 were calculated for each stance during the trial (Figure 2). Data were obtained at the beginning and at the end of the study.
Figure 2—Sample traces of principal strains for 1 stance in a horse, illustrating the measurements made on every such trace (magnitudes of msϵ1, msϵ2, and pkϵ2). ϵ1 = Maximal principal strain. ϵ2 = Minimal principal strain.
Citation: American Journal of Veterinary Research 72, 4; 10.2460/ajvr.72.4.484
Statistical analysis—Data were examined in a quadratic model, with comparisons made between the groups and within each group over 4 months (June to October), at the beginning and at the end of the study. A generalized linear mixed model was developed with the aid of statistical software.g The design was a completely randomized split plot with exercise effect as the whole-plot factor and time as the split-plot factor; times were split within horse because each horse had 2 measurement times. Position of the strain gauges (toe, MQ, and LQ) on the hoof wall and speed were also included in the model. The assumptions of ANOVA were assessed via comprehensive residual analyses. Normality was examined by performance of the Shapiro-Wilk, Kolmogorov-Smirnov, Cramer–von Mises, and Anderson-Darling tests and measurements of skewness and kurtosis. Residuals were plotted against the predicted values and explanatory variables in the model. Because multiple strides were measured, the presence of repeated measures was included as another data component. To handle this, the summary statistics mean and SD were used as the units of observation in the analysis as described elsewhere.34 Means and variances of the 3 measured variables (msϵ1, msϵ2, and pkϵ2) were analyzed in both groups at the beginning and the end of the study.
To evaluate the strain pattern at a similar range of speed, data were ignored when they were recorded at < 4 m/s, presumably derived from the beginning or the end part of the 700-m run, or at > 8 m/s, presumably from some horses that occasionally ran faster than average. Data were pooled into 2 speed categories: 4 to 6 m/s and 6 to 8 m/s. For each horse, approximately 180 strides were recorded at variable speeds in the test run, and a minimum of 10 and maximum of 80 strides were randomly picked for each speed category. To examine the repeatability of the randomized selection procedure, the selection was repeated 2 more times, and the msϵ2 means were tested each time.
Model building was iterative, starting with models containing all primary factors and all possible interactions among those factors. The P values of each of the main effects and the interactions between those factors were examined. Group and time were kept in the model as main effects. However, when an interaction was significant (P < 0.05), the main effect of any factor in that interaction was retained in the model to preserve hierarchy. The same approach was used for each of the 3 strain variables. For analyses of means, data were normally distributed and not transformed. For ANOVAs, data were not normally distributed; therefore, they were transformed (by calculation of the square root), leading to analyses yielding SDs as summary statistics.
Results
Animals—The mean age of the study horses was 6.3 ± 2.7 years in the control group and 5.4 ± 2.6 years in the exercised group. All horses were kept in 1 large paddock for the duration of the experiment and had free access to water and pasture. Through the exercise period, horses in the exercised group had a mean of 47,000 strides (range, 21,000 to 81,000 strides). Mean and variability of msϵ1, msϵ2, and mean pkϵ2 were compared between the exercised and control groups and also between measurement times. The effect of speed (4 to 6 m/s vs 6 to 8 m/s) was nonsignificant in all comparisons and consequently was removed from the models.
Mean msϵ1—With the data pooled for all gauge positions and test times, the magnitude of msϵ1 ranged from −4,741 to 2,304 μϵ for the control group and from −2,041 to 2,968 μϵ for the exercised group. Mean msϵ1 did not differ significantly (P = 0.98) between control and exercised horses. However, when data for both groups were pooled, the mean msϵ1 did vary significantly according to the position of the gauge on the hoof (toe, MQ, and LQ) and time, with some significant interactions. When the effect of gauge position was evaluated, the MQ mean msϵ1 differed from both the toe (P < 0.001) and LQ (P = 0.005) values, but the mean msϵ1 for the toe and LQ did not differ significantly (P = 0.226) from each other. In a comparison of changes at each position by time, values differed significantly for the MQ and LQ but not for the toe (Figure 3). Mean msϵ1 at the toe, MQ, and LQ were 235, −13, and 359 μϵ at the start and 319, −345, and −76 μϵ at the end of the study, respectively.
Figure 3—Mean msϵ1 at the MQ, toe, and LQ before (white bars) and after (black bars) a 4-month period in which Standardbreds were exercised (n = 9) or received no forced exercise (9). The control and exercised groups did not differ significantly (P ≥ 0.05), and therefore, their values were pooled. *Within a location, values for the start and end of the study differed significantly (P < 0.05).
Citation: American Journal of Veterinary Research 72, 4; 10.2460/ajvr.72.4.484
Mean msϵ2—Most individual means for msϵ2 were compressive (negative), including when data were pooled for all gauge positions and test times. Values ranged from −9,781 to 102 μϵ in the control group and from −6,605 to −5 μϵ in the exercised group. Among the primary factors, only gauge position at the hoof was significant (P < 0.001); significant interactions between gauge position and time and between group and time necessitated the retention of the 3-way interaction among those factors in the final analytic model. At the beginning of the study, mean msϵ2 was significantly higher in the exercised group than in the control group at the MQ (P = 0.047) and LQ (P = 0.007) but not at the toe (P = 0.934). At the end of the study, the opposite pattern was found: mean msϵ2 was significantly higher in the control group than in the exercised group at the toe (P = 0.043) but not at the MQ (P = 0.837) or LQ (P = 0.060). Between the beginning and the end of the study, the absolute value of mean msϵ2 increased, changing from −1,354 to −1,856 μϵ at the MQ (nonsignificantly; P = 0.098) and from −2,031 to −3,605 μϵ at the LQ in the control group (significantly; P < 0.001). However, it did not vary significantly (P = 0.767) at the toe over time. In the exercised group, mean msϵ2 changed significantly (P = 0.009) from −3,777 to −2,970 at the toe, but the changes at the MQ (from −2,046 to −1,781 μe) and LQ (from 2,982 to −2,910 μϵ) were not significant (P = 0.384 and P = 0.812, respectively).
Mean pkϵ2—All individual magnitudes of mean pkϵ2 were compressive and ranged from −10,613 to −228 μϵ in the control group and from −6,898 to −633 μϵ in the exercised group. The patterns of changes in the mean pkϵ2 in the control and exercised groups were similar to those in the mean msϵ2. Among the primary factors, only gauge position at the hoof had a significant (P < 0.001) effect, whereas time and group did not. All pairwise interactions among position, time, and groups were significant, so the 3-way interaction among those factors was included in the final analytic model. At the beginning of the study, mean pkϵ2 was significantly higher in the exercised group than in the control group at the MQ (P = 0.033) and LQ (P = 0.003) but not at the toe (P = 0.683). At the end, the groups were not significantly different at any position. In the control group, the absolute value of mean pkϵ2 significantly increased at the MQ (P = 0.005) and LQ (P < 0.001), changing from 1,687 to −2,539 μϵ at the MQ and from −2,265 to −3,594 μϵ at the LQ. Those changes were subtle at the toe (from −4,298 to −4,012 μϵ) and nonsignificant (P = 0.351). In the exercised group, mean pkϵ2 decreased with time at all 3 positions: from −2,435 to −2,179 μϵ at the MQ, from −3,313 to −3,104 μϵ at the LQ, and from −4,147 to −3,303 μϵ at the toe. Those changes were significant (P = 0.006) only at the toe (Figure 4).
Figure 4—Mean msϵ2 (A) and pkϵ2 (B) at the MQ (white bars), toe (gray bars), and LQ (black bars) before (start) and after (end) a 4-month period in which Standardbreds were exercised (Ex; n = 9) or received no forced exercise (Ctrl; 9). a–eValues with the same letters differ significantly (P < 0.05).
Citation: American Journal of Veterinary Research 72, 4; 10.2460/ajvr.72.4.484
Analysis of variability in the data—Analysis of the SD for msϵ1, msϵ2, and pkϵ2 revealed similar patterns of significance to those of the means themselves. In all comparisons, gauge position had a significant (P < 0.001) effect. Among the other primary factors, time and group had significant effects by virtue of significant interactions with each other and with position. The SD of all 3 measured variables followed a similar pattern: it increased in the control group and decreased in the exercised group at the quarters, and it decreased at the toe in both groups.
SD of msϵ1—For msϵ1, there were fewer variations (smaller SDs) at the toe than at the MQ or LQ. The control and exercised groups did not differ significantly from each other at the beginning and at the end of the study. In the control group, the SDs differed significantly between measurement times within both groups at the toe (P = 0.001) and at the LQ (P = 0.016) but not at the MQ (P = 0.315; Figure 5). In the exercised group, values differed significantly at the toe (P < 0.001) but not at the MQ (P = 0.764) or LQ (P = 0.191).
Figure 5—Standard deviations of the msϵ1 (A), msϵ2 (B), and pkϵ2 (C) at the MQ (white bars), toe (gray bars), and LQ (black bars) before (start) and after (end) a 4-month period in which Standardbreds were regularly exercised (Ex; n = 9) or received no forced exercise (Ctrl; 9). a–fValues with the same letters differ significantly (P < 0.05).
Citation: American Journal of Veterinary Research 72, 4; 10.2460/ajvr.72.4.484
SD of mse2—For msϵ2, the SD differed significantly between the groups at the LQ but not at the other 2 positions at both measurement times. The control and exercised groups were significantly different at the LQ at the beginning (P = 0.011) and at the toe (P = 0.008) and LQ (P = 0.039) at the end of the study. Values differed significantly between measurement times at the LQ (P = 0.003) in the control group and at the toe (P < 0.001) in the exercised group (Figure 5). In the control group, values for the MQ and LQ did not differ significantly (P = 0.053 and P = 0.051, respectively). In the exercised group, the same was true for the MQ and toe (P = 0.327 and P = 0.053, respectively).
SD of pkϵ2—Similar to the SD for msϵ2, the SD for pkϵ2 was consistently larger at the LQ than at the 2 other positions. The control and exercised groups differed significantly at the LQ at the beginning (P = 0.028) and at the end (P = 0.035) of the study and at the toe at the end of the study (P = 0.021). Values differed significantly between measurement times at the toe (P = 0.027) and LQ (P = 0.006) in the control group but only at the toe in the exercised group (P < 0.001).
Discussion
In the study reported here, midstance strains were measured because forces acting on the equine hoof are known to peak at midstance,35 whereas pkϵ2 represents the greatest strain in hoof wall material. We detected significant changes in the magnitude of msϵ2 and pkϵ2 between exercised and nonexercised horses at certain hoof positions (ie, the toe and LQ) and a reduction in the SDs of msϵ1, msϵ2, and pkϵ2 in the exercised group between the beginning and end of the study. Mean msϵ1 did not differ in response to exercise, but it significantly changed between measurement times, so data for the control and exercised group were combined and evaluated together. The fact that msϵ1 became more compressive at the MQ and LQ with time indicated that these aspects of the hoof wall were likely to allow changes in tensile strain; such changes were even more compressive at the MQ than at the LQ (Figure 3).
Mean msϵ2 and pkϵ2 changed in response to exercise; their absolute value significantly decreased at the toe. At the MQ and LQ, the changes in these values assumed an interesting pattern: the absolute value of mean msϵ2 and pkϵ2 decreased with time in the exercised group and increased in the control group. This would support the hypothesis that the MQ and LQ are more likely to respond to changes in exercise regimen than is the toe. The observed pattern of increasing the SDs (ie, variations) of msϵ1, msϵ2, and pkϵ2 at the MQ and LQ in the control group and their reduction in the exercised group suggested a possible role of the exercise in reducing that variation.
The observed differences in the hoof wall surface strain distribution could have been related to a number of factors, of which 2 may have had greater influence: shape and material properties of the hoof capsule. As part of a larger study,36 we examined the morphometric changes in the hoof shape and found the differences between the control and exercised groups were not significant. Therefore, it is likely that the changes detected in the present study were linked with changes in the hoof's material properties. Whether these changes were associated with adaption to exercise, changes in hoof wall quality, or another reason is not clear. Although it would have been valuable information, the biochemical and structural properties of the hoof wall were not examined because our study did not include euthanasia of horses to obtain such data.
The effect of the season (eg, cold and dry weather vs humid and warm weather) on hoof shape has been reported.37,38 However, there are few reports39–41 on the effect of exercise on hoof shape. Thoroughbred racehorses reportedly have a reduction in hoof angle with exercise, whereas after a period of rest, the hoof angle increases again.40 Among the other factors that may influence the response to exercise are exercise speed and intensity. The reported peak strain at a gallop and trot would be 6 and 2 times that at a walk, respectively.42 Higher speeds impose more strain on the skeletal system, which results in more adaptive responses and possibly increased strength. In the study reported here, the exercised group received only mild to moderate exercise (trotting), and it is unknown whether exercising the horses at higher speeds could have induced different changes in an adaptive response.
Interestingly, strain distribution was different between exercised and nonexercised horses at the beginning of the study. Whether this difference was attributable to chance or other factors was not apparent. Mean age and body weight were similar between the groups. The hoof shape was examined in another study,36 and it did not appear to differ significantly between the groups. All horses were acquired a minimum of 4 weeks prior to commencing the present study and were kept in a small barn with no opportunity to exercise because of quarantine protocols. The effect of the previous exercise history (if any) on the observed differences could not be determined because such historical data were not available. To minimize the chance of environmental variability, all horses were housed in the same paddock and on the same substrate (pasture). The process of randomization should also have reduced the effect of other confounding factors.
The changes reported here developed in response to 4 months of mild to moderate exercise on a hard surface in trotting Standardbreds. Whether changes to stimuli or environment such as exercise intensity, exercise duration, faster speed, or exercise surface could induce different responses requires additional investigation. Our findings indicated that responses of the highly keratinized hoof to this moderate exercise regimen were subtle but present.
ABBREVIATIONS
| LQ | Lateral quarter of the hoof wall |
| MQ | Medial quarter of the hoof wall |
| μϵ | Microstrain |
| msϵ1 | Midstance maximal principal strain |
| msϵ2 | Midstance minimal principal strain |
| pkϵ2 | Peak minimal principal strain |
Professional Farrier Supply Inc, Orangeville, ON, Canada.
N32-FA-2–120–11, Showa Measuring Instruments Co Ltd, Tokyo, Japan.
Loctite 411, Loctite Canada Inc, Mississauga, ON, Canada.
Vetrap, 3M Animal Care Product, Saint Paul, Minn.
Logbook/300, Iotech Inc, Cleveland, Ohio.
Gauss Data Analysis Software, Aptech Industries, Seattle, Wash.
PROC MIXED, SAS, version 9.1.3, SAS Institute Inc, Cary, NC.
References
- 1.↑
Barrey E. Investigation of the vertical hoof force distribution in the equine forelimb with an instrumented horseboot. Equine Vet J Suppl1990;(9): 35–38.
- 2.↑
Thomason JJ, Biewener AA, Bertram JEA. Surface strain on the equine hoof wall in vivo: implications for the material design and functional morphology of the wall. J Exp Biol 1992; 166:145–165.
- 3.↑
Kasapi MQ, Gosline JM. Design complexity and fracture control in the equine hoof wall. J Exp Biol 1997; 200:1639–1659.
- 4.
Pollitt CC. The anatomy and physiology of the hoof wall. Equine Vet Educ 1998; 4:3–10.
- 5.
Douglas JE, Thomason JJ. Shape, orientation and spacing of the primary epidermal laminae in the hooves of neonatal and adult horses (Equus caballus). Cells Tissues Organs 2000; 166:304–318.
- 6.↑
Bowker RM. The growth and adaptive capabilities of the hoof wall and sole: functional changes in response to stress, in Proceedings. 49th Annu Conv Am Assoc Equine Pract 2003; 146–148.
- 7.
Bowker RM. Contrasting structural morphologies of “good” and “bad” footed horses, in Proceedings. 49th Annu Conv Am Assoc Equine Pract 2003; 186–209.
- 8.
Thomason JJ, Bignell WB, Batiste D, et al. Effects of hoof shape, body mass and velocity on surface strain in the wall of the unshod forefeet of Standardbreds trotting on a treadmill. Equine Comp Exercise Physiol 2004; 1:87–97.
- 9.
Sarratt SM, Hood DM. Evaluation of architectural changes along the proximal to distal regions of the dorsal laminar interface in the equine hoof. Am J Vet Res 2005; 66:277–283.
- 10.
Balch OK, Ratzalff MH, Hyde ML, et al. Locomotor effects of hoof angle and mediolateral balance of horses exercising on a high-speed treadmill: preliminary results, in Proceedings. Annu Conv Am Assoc Equine Pract 1991; 37:687–705.
- 11.
Roland E, Stover SM, Hull ML, et al. Geometric symmetry of the solar surface of hooves of thoroughbred racehorses. Am J Vet Res 2003; 64:1030–1039.
- 12.
Thomason JJ, Douglas JE, Sears W. Morphology of the laminar junction in relation to the shape of the hoof capsule and distal phalanx in adult horses (Equus caballus). Cells Tissues Organs 2001; 168:295–311.
- 13.
Thomason JJ, MacClinchey HL, Faramarzi B, et al. Mechanical behaviour and quantitative morphology of the equine laminar junction. Anat Rec A Discov Mol Cell Evol Biol 2005; 283:366–379.
- 14.
Eliashar E, McGuigan MP, Wilson AM. Relationship of foot conformation and force applied to the navicular bone of sound horses at the trot. Equine Vet J 2004; 36:431–435.
- 15.
van Heel MC, Barneveld A, van Weeren PR, et al. Dynamic pressure measurements for the detailed study of hoof balance: the effect of trimming. Equine Vet J 2004; 36:778–782.
- 16.
van Heel MC, van Weeren PR, Back W. Shoeing sound warm-blood horses with a rolled toe optimizes hoof-unrollment and lowers peak loading during breakover. Equine Vet J 2006; 38:258–262.
- 17.
Hinterhofer C, Stanek C, Haider H. The effect of flat horseshoes, raised heels and lowered heels on the biomechanics of the equine hoof assessed by finite element analysis. Zentralbl Veterinarmed A 2000; 47:73–82.
- 18.
Back WB, Remmen JL, Knaap J, et al. Effect of lateral heel wedges on sagittal and transverse plane kinematics of trotting Shetland ponies and the influence of feeding and training regimes. Equine Vet J 2003; 35:606–612.
- 19.
Eliashar E, McGuigan MP, Rogers KA, et al. A comparison of three horse shoeing styles on the kinetics of breakover in sound horses. Equine Vet J 2002; 34:184–190.
- 20.
Chateau H, Degueurce C, Denoix JM. Effects of a 6 degree elevation of the heels on 3D kinematics of the distal portion of the forelimb in the walking horse. Equine Vet J 2004; 36:649–654.
- 21.
Chateau H, Degueurce C, Denoix JM. Effect of egg bar shoes on the three-dimensional kinematics of the distal forelimb in horses walking on a sand track. Equine Vet J2006;377–382.
- 22.
Lawson SE, Chateau H, Pourcelot P, et al. Effect of toe and heel elevation on calculated tendon strains in the horse and the influence of the proximal interphalangeal joint. J Anat 2007; 210:583–591.
- 23.
Thomason JJ. Variation in surface strain on the equine hoof wall at the midstep with shoeing, gait, substrate, direction of travel, and hoof shape. Equine Vet J Suppl1998;(26): 86–95.
- 24.
Crevier-Denoix N, Pourcelot P, Ravary B, et al. Influence of track surface on the equine superficial digital flexor tendon loading in two horses at high speed trot. Equine Vet J 2009; 41:257–261.
- 25.
Setterbo JJ, Garcia TC, Campbell IP, et al. Hoof accelerations and ground reaction forces of Thoroughbred racehorses measured on dirt, synthetic, and turf track surfaces. Am J Vet Res 2009; 70:1220–1229.
- 26.
Gustås P, Johnston C, Drevemo S. Ground reaction force and hoof deceleration patterns on two different surfaces at the trot. Equine Comp Exercise Physiol 2006; 3:209–216.
- 27.↑
Back W, MacAllister C, van Heel M, et al. Vertical frontlimb ground reaction forces of sound and lame Warmbloods differ from those in Quarter Horses. J Equine Vet Sci 2007; 27:123–129.
- 28.↑
Gustås P, Johnston C, Henderstorm U, et al. A field study on hoof deceleration at impact in Standardbred trotters at various speeds. Equine Comp Exercise Physiol 2006; 3:161–168.
- 29.↑
Morales JL, Manchado M, Vivo J, et al. Angular kinematic patterns of limbs in elite and riding horses at trot. Equine Vet J 1998; 30:528–533.
- 30.
Davies HM. Strain in the yearling equine metacarpus during locomotion. Equine Vet J 2001; 33:16–20.
- 31.
Hoekstra NE, Nielson BD, Rosenstein DS, et al. Comparison of bone mineral content and bone metabolism in stall versus pressure reared horses. Equine Vet J Suppl1999;(30): 601–604.
- 32.
Nielson BD, O'Conner CI, Rosenstein DS, et al. Influence of trotting and supplemental weight on metacarpal bone development. Equine Vet J Suppl2002;(34): 236–240.
- 33.
Davies HM. Estimating peak strains associated with fast exercise in Thoroughbred racehorses. Equine Vet J Suppl2006;(36): 440–444.
- 34.↑
Everett BS. The analysis of repeated measures: a practical review with examples. Statistician 1995; 44:113–135.
- 35.↑
Merkens HW, Schamhardt HC, van Osch GJV, et al. Ground reaction force patterns of Dutch Warmblood horses at a normal trot. Equine Vet J 1993; 25:134–137.
- 36.↑
Faramarzi B, Thomason JJ, Sears WC. Changes in growth of the hoof wall and hoof morphology in response to regular periods of trotting exercise in Standardbreds. Am J Vet Res 2009; 70:1354–1364.
- 37.
Davies HM. No foot, no horse. The clinical implications of modeling the hoof capsule. Equine Vet J 2002; 34:646–647.
- 38.
Kainer R. Functional anatomy of equine locomotor organs. In: Adam's lameness in horses. Philadelphia: Lippincott Williams & Wilkins, 2002;6–8.
- 39.
Florence L, McDonnell SM. Hoof growth and wear of semi-feral ponies during an annual summer ‘self-trimming’ period. Equine Vet J 2006; 38:642–645.
- 40.↑
Peel JA, Peel MB, Davies HM. The effect of gallop training on hoof angle in Thoroughbred racehorses. Equine Vet J Suppl2006;(36): 431–434.
- 41.
Decurnex V, Anderson GA, Davies HM. Influence of different exercise regimes on the proximal hoof circumference in young Thoroughbred horses. Equine Vet J 2009; 41:233–236.
- 42.↑
Davies HM, McCarthy RN, Jeffcott LB. Surface strain on the dorsal metacarpus of Thoroughbreds at different speeds and gaits. Acta Anat (Basel) 1993; 146:148–153.
