Racing surfaces could affect the risk for musculoskeletal injuries of racehorses through the transfer of forces from the surface to structures of the limbs. Surface forces are transferred through horseshoes at the interface of the surface and hoof. Forces have been modified to prevent injury in horses through the use of therapeutic shoes.1 Wear patterns on a horseshoe may reflect forces transferred between the surface and limb during the interaction of the hoof with the surface.2,3 Understanding of the hoof-shoe interaction and effects of racetrack surfaces on this interaction could lead to optimizing the transfer of loads from racing surfaces to a hoof and injury prevention.2,4
A hoof typically expands at the heels during loading.5,6 General shoeing practice is to place nails within the dorsal half of a hoof to allow for expansion of the heels during the stance phase and limb loading.7 Because a horseshoe is stiffer than the hoof wall, heel expansion during loading would result in sliding of the hoof wall relative to the solar surface of the horseshoe when shear forces exceed frictional resistance at the wall-shoe interface. To our knowledge, there are no data on shoe wear at the horseshoe-hoof wall interface.
The objective of the study reported here was to measure horseshoe wear, as indicated by groove dimensions on the solar surface of aluminum racehorse shoes, and compare groove dimensions between dirt and synthetic racing surfaces to provide insights on differences of hoof impacts on various surface types. Characteristics of each horseshoe were also considered to determine their contribution to groove dimensions for various surface types. We hypothesized that length and width of horseshoe wear grooves would differ on the basis of surface type and shoeing characteristics.
Materials and Methods
Samples
Horseshoes (n = 1,121) that had been removed from 242 Thoroughbred racehorses by their respective farriers during routine horseshoeing procedures from December 3, 2009, through March 3, 2011, were used in the study. Racehorses trained or raced (or both) at 4 racetracks that had a dirt surface (Pleasanton, San Luis Rey Downs, and Santa Anita) or synthetic surface (Hollywood Park and Santa Anita). Santa Anita initially had a synthetic surface, but this was changed to a dirt surface (December 7, 2010); thus, the analysis included data for 5 racetrack-surface combinations.
Experimental procedures
Horseshoes were marked to indicate the horse identification number and the limb from which each horseshoe was removed. Horseshoes were categorized by side (left vs right) and body region of limb (forelimb vs hind limb). A set of horseshoes from a specific horse was placed into a plastic bag and labeled with corresponding identifying information. Horseshoes were excluded from the study when a second set of horseshoes for a horse were collected during a second shoeing procedure while that horse still trained on the same surface, a horse trained or raced on both dirt and synthetic surfaces, they consisted of material other than aluminum, they had modifications other than reshaping (eg, heel shortened or bar shoes), or they comprised a small sample size (n < 5 horses) for a particular style of horseshoe. Amount of time the horseshoes were on a horse and the amount of work each horse completed while wearing the horseshoes were unknown. Farriers differed among the racetracks and were also unknown. Horses were shod with 6 types of horseshoe.a–f
Data collection
The solar surface of each shoe was digitally photographed.g Photographs were obtained with a calibration rule under standardized conditions. Resolution was 118.5 pixels/mm; thus, measurements could be resolved to within 0.008 mm (eg, dimension of a pixel).
Lines were drawn on horseshoe photographs to mark reference points for measurements (Figure 1). Lines intersected at right angles. A line (line AB) was drawn through the palmar (or plantar) edges of the medial and lateral heels of the horseshoe. A second line (line CD) was perpendicular to line AB and was drawn through a point equidistant between the most dorsal nail holes. A third line (line EG) was parallel to line AB and was drawn through the widest portion of the horseshoe. Point F was the intersection of lines CD and EG. A fourth line (line HI) was perpendicular to line AB and was drawn from the last nail hole that contained a nail on both the medial and lateral aspects of the horseshoe.
Coordinate system used for measurements on the solar surface of a horseshoe. Line AB indicates heel edges of the horseshoe, line CD is horseshoe length, line EG is horseshoe width, line FD is length of the heel of the horseshoe, and line HI is the distance from the last nail hole that contained a nail to the heel of the horseshoe. The boundary of 1 of the 2 grooves on the heel of the horseshoe is indicated (dashed line); length (double-headed black arrow) and width (double-headed white arrow) of the groove are indicated.
Citation: American Journal of Veterinary Research 77, 9; 10.2460/ajvr.77.9.983
Dimensions of wear grooves at the medial and lateral heels of the solar surface of each horseshoe were measured. Wear grooves were defined as any pitting, smoothing, or change from the pattern stamped on a horseshoe or as part of a horseshoe as originally cast.
Variables were measured on digital images by use of an image-analysis program.h Shoe length was the distance on line CD from the dorsal margin of the shoe to the intersection with line AB. Shoe width was the distance on line EG between the medial and lateral margins of the horseshoe. Web width was the width (medial to lateral) of the left side of the horseshoe. Heel length was the length of line FD. Length of a wear groove was the distance between the dorsal and palmar (or plantar) edges of the groove. Maximum width of a wear groove was the distance of a line perpendicular to the edges of the groove at the widest point of the groove. Distance from the palmar (or plantar) edge of the groove to the heel of the horseshoe was also measured.
Because limb, horseshoe, and hoof characteristics could have confounded results, several characteristics were recorded. These included side (left vs right), body region of limb (forelimb vs hind limb), toe grab8 (none, low [> 2 and ≤ 4 mm], medium [> 4 and ≤ 6 mm], and high [> 6 and ≤ 8 mm]), and horseshoe clip (none, toe, and side). In addition, variables included horseshoe size (4,5,6,7, and 8) and, for both medial and lateral heels of horseshoes, nail distance (distance from the last nail hole that contained a nail to the heel of the horseshoe [eg, length of line HI]). Horseshoes on which the size was illegible were treated as missing data.
Statistical analysis
Effects of racetrack surface, racetrack, and horseshoe variables on dimensions (length and width) of wear grooves were assessed by use of a mixed-model ANOVA.i The model was stratified on the basis of body region of limb (ie, separate models for the forelimbs and hind limbs) because statistical results differed for forelimb and hind limb horseshoes. All potential main effects, including those significant in a univariable analysis or that were biologically plausible or potential confounders, were included in the model. Two-way interactions between racetrack surface and all other variables significant in a univariable analysis were also included. A backward-stepwise elimination procedure was used, and variables with P < 0.05 were excluded from each successive iteration until there was a final model with only those variables that remained significant (P ≤ 0.05). This included variables that were part of significant interactions, even if the main effects of the interaction variables were not significant. Pairwise comparisons were examined with least squares mean differences and P values adjusted by use of the Tukey method. A Shapiro-Wilk test was performed to evaluate normal distribution of ANOVA residuals (W > 0.96).
Results
Horseshoes
Data for 1,014 shoes obtained from 233 horses were analyzed. Horseshoes were approximately equally distributed between dirt (462/1,014 [46%] shoes) and synthetic (552/1,014 [54%] shoes) surfaces (Table 1). There were 475/1,014 (47%) horseshoes obtained from forelimbs and 539/1,014 (53%) horseshoes obtained from hind limbs.
Number (%) of horseshoes collected from the forelimbs and hind limbs of Thoroughbreds racing or training (or both), on the basis of racetrack and surface type.
| Horseshoes (No. [%]) | |||
|---|---|---|---|
| Variable | Forelimb | Hind limb | Total |
| Surface and racetrack Dirt | |||
| Pleasanton | 14 (44) | 18 (56) | 32 (3) |
| Santa Anita | 168 (46) | 195 (54) | 363 (36) |
| San Luis Rey Downs | 22 (33) | 45 (67) | 67 (7) |
| Total dirt | 204 (44) | 258 (56) | 462 (46) |
| Synthetic | |||
| Hollywood Park | 152 (50) | 155 (50) | 307 (30) |
| Santa Anita | 119 (49) | 126 (51) | 245 (24) |
| Total synthetic | 271 (49) | 281 (51) | 552 (54) |
| Limb | |||
| Left | 237 (47) | 268 (53) | 505 (50) |
| Right | 238 (47) | 271 (53) | 509 (50) |
| Type of horseshoe* | |||
| a | 7 (47) | 8 (53) | 15 (1) |
| b | 36 (54) | 31 (46) | 67 (7) |
| c | 25 (100) | 0 (0) | 25 (2) |
| d | 321 (52) | 298 (48) | 619 (61) |
| e | 86 (55) | 69 (45) | 155 (15) |
| f | 0 (0) | 133 (100) | 133 (13) |
| Toe grab† | |||
| Low | 475 (54) | 406 (46) | 881 (87) |
| Medium | 0 (0) | 133 (100) | 133 (13) |
| Horseshoe clip | |||
| None | 475 (56) | 372 (44) | 847 (84) |
| Side clip | 0 (0) | 32 (100) | 32 (3) |
| Toe clip | 0 (0) | 135 (100) | 135 (13) |
A total of 1,014 horseshoes were analyzed from all limbs of all horses at all racetracks.
Horseshoes were shod with 6 types of horseshoe.a–f
Toe grab was characterized as none, low (> 2 and ≤ 4 mm), medium (> 4 and ≤ 6 mm), and high (> 6 and ≤ 8 mm).
Type d horseshoes were the ones most frequently worn (321/475 [68%] horseshoes on the forelimbs and 298/539 [55%] horseshoes on the hind limbs). Type e horseshoes were the second most frequently worn horseshoe on the forelimbs (86/475 [18%]), whereas the next most frequently worn horseshoes on the hind limbs were type f (133/539 [25%] and type e (69/539 [13%]. Other shoe types were represented in smaller numbers (Table 1).
All of the toe grabs on horseshoes on the forelimbs were low (475/475 [100%]). For horseshoes on the hind limb, 406 of 539 (75%) toe grabs were low and 133 of 539 (25%) were medium.
No clips were found on horseshoes on the forelimbs. However, for horseshoes on the hind limbs, 135 of 539 (25%) had toe clips and 32 of 539 (6%) had side clips.
Racetrack surface
Differences between dirt and synthetic surfaces were identified for length and width of the lateral groove for horseshoes on the forelimbs (Table 2). Mean length of the lateral groove was 13% greater for synthetic surfaces than for dirt surfaces. Mean width of the lateral groove was 17% greater for synthetic surfaces than for dirt surfaces. No other significant differences between surfaces were found for groove length or width for horseshoes on the forelimbs or hind limbs.
Mean length and width of grooves in the medial or lateral heels of horseshoes collected from the forelimbs and hind limbs of Thoroughbreds racing or training (or both) at various racetracks with dirt or synthetic surfaces.
| Medial groove | Lateral groove | |||
|---|---|---|---|---|
| Variable | Length (mm) | Width (mm) | Length (mm) | Width (mm) |
| Forelimb | ||||
| Dirt | ||||
| Pleasanton | 3.89 ± 0.43 | 0.81 ± 0.07 | 3.53 ± 0.39 | ND |
| Santa Anita | 4.58 ± 0.13 | 0.66 ± 0.02 | 4.66 ± 0.12 | ND |
| San Luis Rey Downs | 4.62 ± 0.34 | 0.71 ± 0.05 | 4.61 ± 0.31 | ND |
| Total dirt | NR | NR | 4.27 ± 0.18 | 0.66 ± 0.02 |
| Synthetic | ||||
| Hollywood Park | 5.55 ± 0.19 | 0.65 ± 0.03 | 4.91 ± 0.19 | NR |
| Santa Anita | 4.79 ± 0.16 | 0.77 ± 0.25 | 4.79 ± 0.15 | NR |
| Total synthetic | NR | NR | 4.85 ± 0.13 | 0.78 ± 0.02 |
| Hind limb Dirt | ||||
| Pleasanton | 3.67 ± 0.41 | 0.91 ± 0.05 | 3.40 ± 0.28 | 0.77 ± 0.05 |
| Santa Anita | 4.89 ± 0.18 | 0.67 ± 0.02 | 4.45 ± 0.09 | 0.63 ± 0.15 |
| San Luis Rey Downs | 4.15 ± 0.26 | 0.75 ± 0.03 | 4.23 ± 0.18 | 0.70 ± 0.03 |
| Total dirt | NR | NR | NR | NR |
| Synthetic | ||||
| Hollywood Park | 4.27 ± 0.24 | 0.63 ± 0.02 | 4.31 ± 0.14 | 0.66 ± 0.02 |
| Santa Anita | 4.59 ± 0.20 | 0.66 ± 0.02 | 4.50 ± 0.12 | 0.65 0.02 |
| Total synthetic | NR | NR | NR | NR |
| Horseshoe clip | ||||
| None | 4.60 ± 0.12 | NR | NR | NR |
| Side clip | 4.58 ± 0.37 | NR | NR | NR |
| Toe clip | 3.82 ± 0.20 | NR | NR | NR |
| Limb | ||||
| Left forelimb | NR | NR | 4.80 ± 0.13 | 0.76 ± 0.02 |
| Right forelimb | NR | NR | 4.32 ± 0.13 | 0.68 ± 0.02 |
| Left hind limb | 3.96 ± 0.14 | 0.69 ± 0.02 | NR | NR |
| Right hind limb | 4.70 ± 0.14 | 0.74 ± 0.02 | NR | NR |
Values reported are least squares mean ± SEM; only values included in the final multivariable models are reported.
NR = Not reported.
Racetrack
Effects of racetrack were apparent for length of the medial and lateral grooves. Mean lengths of the groove for horseshoes on the forelimbs and hind limbs were shorter for the Pleasanton (dirt) racetrack than for all other racetracks (Table 2). Medial forelimb groove length was longest for the Hollywood Park (synthetic) racetrack. Length of the medial groove for horseshoes on the hind limbs for the Pleasanton racetrack was significantly (P = 0.017) smaller than for the Santa Anita (dirt), whereas the length for horseshoes on the hind limbs for the Santa Anita (dirt) was significantly longer than for San Luis Rey Downs (dirt; P = 0.005) and Hollywood Park (synthetic; P = 0.009). Lengths of the medial and lateral grooves for horseshoes on the forelimbs and hind limbs did not differ significantly between Santa Anita (dirt) and Santa Anita (synthetic).
Effects of racetrack were also apparent for width of medial and lateral grooves (Figure 2). Pleasanton (dirt) had the largest width for lateral grooves on horseshoes on the hind limbs. Santa Anita (dirt) had the smallest width for medial groves on horseshoes on the hind limbs for all dirt racetracks. Widths of medial and lateral grooves were larger for horseshoes on the forelimbs and hind limbs for Pleasanton (dirt) than for Santa Anita (dirt). Width of lateral and medial grooves were both significantly (P < 0.001) larger on horseshoes for Santa Anita (synthetic) than for Santa Anita (dirt). Width of medial grooves on horseshoes on the forelimbs were significantly (P = 0.002) different between Hollywood Park (synthetic) and Santa Anita (synthetic). Width of lateral grooves on horseshoes on the forelimbs did not differ among racetracks; however, the grooves were wider for racetracks with a synthetic surface than for racetracks with a dirt surface.
Mean length (A–D) and width (E–H) of the groove on the medial (B,C,F, and G) and lateral (A,D,E, and H) heels of horseshoes on the forelimbs (A, B, E, and F) and hind limbs (C, D, G, and H) of Thoroughbreds training or racing (or both) on racetracks with a dirt (gray bars) or synthetic (black bars) surface. Values reported are least squares mean ± SEM. The racetracks were Hollywood Park (HP), Pleasanton (PL), Santa Anita–dirt (SAD), Santa Anita–synthetic (SAS), and San Luis Rey Downs (SLRD). a–c Values with different superscript letters differed significantly (P < 0.05).
Citation: American Journal of Veterinary Research 77, 9; 10.2460/ajvr.77.9.983
Nail distance
Distance from the last nail hole in which a nail was inserted to the heel of the horseshoe was associated with length of the groove (Table 3). However, magnitudes of the relationships were low (all values of R2 ≤ 0.081) when this distance was compared for length and width of grooves (Figure 3). When nail distance values were adjusted for effects in the model, magnitudes of the relationship increased (all values of R2 ≤ 0.697). Overall, as nail distance increased, length of the groove also increased, although length of lateral grooves for horseshoes on the forelimb on dirt surfaces did not differ significantly as nail distance increased. Width of grooves for horseshoes on both the forelimbs and hind limbs on dirt surfaces increased as nail distance increased. There was no significant change or a decrease in width of grooves for horseshoes on the forelimbs and hind limbs on synthetic surfaces.
Effects of distance from the last nail hole that contained a nail to the heel of the horseshoe (nail distance) on length (A) and length adjusted on the basis of effects of the statistical analytic model (B) of the medial groove on horseshoes on the forelimbs of Thoroughbreds racing or training (or both) at racetracks with a dirt (black circles) or synthetic (gray circles) surface.
Citation: American Journal of Veterinary Research 77, 9; 10.2460/ajvr.77.9.983
Results of statistical analysis of ANOVA models for medial and lateral grooves on horseshoes collected from the forelimbs and hind limbs of Thoroughbreds racing or training (or both) at various racetracks with dirt or synthetic surfaces.
| Forelimb | Hind limb | |||||||
|---|---|---|---|---|---|---|---|---|
| Medial groove | Lateral groove | Medial groove | Lateral groove | |||||
| Variable | Length | Width | Length | Width | Length | Width | Length | Width |
| Surface | NS | NS | 0.005 | 0.001 | NS | NS | NS | NS |
| Racetrack (surface) | 0.001 | 0.001 | 0.047 | NS | 0.006 | 0.001 | 0.007 | 0.021 |
| Nail distance* | 0.028 | 0.004 | 0.012 | 0.319 | 0.001 | 0.002 | 0.001 | NS |
| Nail distance (surface*) | NS | NS | 0.001 | 0.001 | NS | NS | NS | NS |
| Horseshoe length | 0.046 | NS | 0.003 | NS | NS | 0.056 | NS | NS |
| Horseshoe length (surface) | NS | NS | NS | NS | NS | 0.002 | NS | NS |
| Horseshoe width | NS | 0.123 | NS | 0.001 | NS | NS | 0.007 | NS |
| Horseshoe width (surface) | NS | 0.025 | NS | NS | NS | NS | NS | NS |
| Horseshoe clip† | NA | NA | NA | NA | 0.001 | NS | NS | NS |
| Limb‡ | NS | NS | 0.001 | 0.001 | 0.001 | 0.017 | NS | NS |
Values were considered significant at P ≤ 0.05.
Nail distance was the distance from the last nail hole that contained a nail to the heel of the horseshoe.
Categories were toe and side clip.
Categories were right and left.
NA = Not applicable because no clips were found on horseshoes on the forelimbs. NS = Not significant (P > 0.05).
Horseshoe clips
Presence of a toe clip on horseshoes on the hind limbs was positively associated with length of the medial groove (Table 3). Horseshoes without toe clips had grooves that were 20% longer than the grooves for horseshoes with toe clips. There was no apparent effect for side clips.
Horseshoe length
Length of a horseshoe on the forelimbs was positively associated with lengths of medial and lateral grooves (Table 3). However, magnitudes of the relationships were biologically unimportant (R2 < 0.03).
Horseshoe width
Width of a horseshoe was associated with width of the lateral groove on horseshoes on the forelimbs (R2 = 0.213; P < 0.001) and with length of the lateral groove on horseshoes on the hind limbs (R2 = 0.014; P = 0.007). As horseshoe width increased, width of lateral grooves on horseshoes on the forelimbs and length of lateral grooves on horseshoes on the hind limbs increased, although the magnitude of these increases was small.
Limb side
Mean length and width of lateral grooves were significantly (P < 0.001) larger for horseshoes on the left forelimbs, compared with values for horseshoes on the right forelimbs (Table 2). Length and width of medial grooves were smaller (17% [P < 0.001] and 7% [P = 0.017], respectively) for horseshoes on the left hind limbs than for horseshoes on the right hind limbs.
Discussion
Length and width of grooves worn in the solar surface of the medial and lateral heels of horseshoes on the forelimbs and hind limbs were compared among racetracks and between surfaces. Groove lengths and widths differed among racetracks, but overall differences of the effects between dirt and synthetic surfaces did not have consistent patterns. Additional factors that were associated with groove length and width included the distance from the last horseshoe nail to the heel of the horseshoe, horseshoe dimensions, and presence or absence of horseshoe clips.
Effects of racetrack surface (dirt vs synthetic) only had one significant influence on the length or width of the grooves measured, but differences in surfaces among racetracks were apparent. Differences in racetrack surfaces result from differences in material composition, temperature, moisture content, base layers, banking, and general maintenance practices. Variability of surface characteristics within the same surface type can differ by as much as 300%.9 In the study reported here, differences among the racetracks were more apparent than were differences between surface types. These differences were even apparent between racetracks that had the same surface type. Dirt surfaces were found on racetracks located in both northern and southern California, which possibly reflected different types of soil composition as well as differences in environmental conditions. Track maintenance at each location could also have contributed to these differences. Harrowing depth, moisture application, and general upkeep are often the responsibility of different individuals and may differ with surface composition and use. One track (Santa Anita) had both synthetic and dirt surface types during the study, and horseshoes were collected during both time periods. However, there were no significant differences for length of grooves between the 2 surface types at this racetrack. Differences between surface types represented ≤ 6.3% of the groove length. Although a larger sample size may have allowed detection of significant differences, such differences may not be biologically relevant.
Track maintenance and surface content were not evaluated in this study, and quantitative measures of the mechanical behavior of the surfaces were not known. Consistency between measurement outcomes for a track surface could be related to uniform management for each specific racetrack. Greater variation might be expected among racetracks. Because of the retrospective nature of this study, material properties of the surface were not known, but they would be needed in future studies to better understand the relationships between surface mechanical properties and hoof biomechanics.
Various factors, including horseshoe and track surface materials, can contribute to hoof vibrations, ground reaction forces, and hoof accelerations for a racehorse. Fewer and less impactful vibrations are expected to cause less damage to hooves, tissues, and joints. Slower hoof accelerations as a result of a more compliant surface could lessen the impact on a horse's limbs during heel strike and stance. Lower hoof vibrations, slower hoof accelerations, and lower ground reaction forces were found for synthetic surfaces than for dirt surfaces in other studies.10,11 In another study,12 investigators compared dampening effects of shoeing and found that aluminum horseshoes resulted in lower hoof vibrations than for steel horseshoes. All of the horseshoes in the present study were made of aluminum. Aluminum is a lighter and less rigid metal than steel, which could allow a horseshoe to absorb some of the vibrational energy exerted by the hoof. It is possible that wear patterns in horseshoes are a reflection of this energy as it is absorbed by the horseshoes through movement and friction.
Horseshoe, hoof, and distance from the last nail to the end of the horseshoe have the potential to affect movement between the horseshoe and hoof. Lengths of grooves in heels of horseshoes were longer for horseshoes that had a greater nail distance. As this distance increased, length and width of grooves increased. It was likely that this increase in wear was attributable to the opportunity for a greater length of hoof to expand with nails placed farther from the heel of the horseshoe. General shoeing practice is to place nails in the region from the toe to the widest part of the hoof.7 For the horseshoes in the study reported here, many had nails that were placed more palmar (or plantar) to the widest part of the hoof. Thus, nails were actually placed closer to the heel of the hoof. It is possible that placement of these nails closer to the heel interfered with the natural ability of a hoof to expand during the stance phase; thus, there were shorter grooves in horseshoes in which nails were placed in a more palmar (or plantar) location.
Whether nail placement closer to the heel is an indicator of possible interference of force dissipation and possible cause of injury remains to be determined. Force is absorbed by structures of the limb and hoof during the stance phase. This includes expansion of heel and frog structures of the hoof. Prevention of heel expansion may contribute to injury. To our knowledge, there is no published information regarding the amount of heel expansion that would be considered normal. We also are not aware of any publications on the extent of interference that nail placement has on heel expansion. Obtaining an understanding of this behavior in the hoof would aid in understanding the nature of the hoof as an absorptive structure as well as the manner in which interference with this role could possibly lead to injury. Because of the interesting findings regarding nail placement and the biological relevance, future studies on the amount of heel expansion for hooves with horseshoes versus without horseshoes are required.
Groove length was shorter on horseshoes with toe clips. It is likely that there was less friction because clips are a rigid structure that prevent the hoof from sliding forward or sideways, depending on the site of clip placement.13 Clips are used for horses that have a tendency to dislodge horseshoes because of sideways or forward slippage of the horseshoe. Clips also provide an opportunity to relieve some of the stress placed on the nails as a result of movement of the hoof. The lack of significant differences for groove length between horseshoes with no clip and horseshoes with side clips may have been attributable to the ability of a hoof to still slide forward in a horseshoe with side clips, whereas toe clips prevented this type of movement.
Length and width were greater for lateral grooves of horseshoes on the left forelimbs and medial grooves of horseshoes on the right hind limbs. This could have been a reflection of increased forces on these aspects of the hoof. Hoof wall strains during turning are greater on the inside quarter of a hoof than on the outside quarter of a hoof.14 In another study,15 investigators detected increases in subchondral bone density for the medial aspect of the third metacarpal bone of the right forelimb of horses performing training in a counterclockwise direction, which is consistent with findings for the present study. In the United States, horses run in a counterclockwise direction when racing and training and land with the right hind limb first and the left forelimb last on rounded aspects of the track. This diagonal transition through a left turn could be associated with greater loads on the medial aspect of the hind limb during hoof landing and the lateral aspect of the left forelimb as weight is moved over the hoof.16 The greater values for length and width of grooves of horseshoes on these limbs provided insight to hoof behavior and loading of the limb during the stance phase and could provide information that will aid in understanding the occurrence of injuries. Duration of hoof slide on synthetic surfaces is reduced, compared with the duration of hoof slide on dirt surfaces,10 which could have accounted for the larger width of the lateral groove on horseshoes on the forelimbs.
The study had several limitations. There were relatively small sample sizes for 2 racetracks (Pleasanton and San Luis Rey Downs). Information about the amount of time (duration) that the horseshoes were worn by the horses and the exercise activity of the horses was not known. For purposes of the study, it was assumed that all horses were in active race training and had the opportunity to create grooves in a relatively consistent manner. Depth of grooves would likely increase with an increase in the amount of exercise, but length and width of grooves are a reflection of the range of hoof deformation during each phase of the gait. The range of deformation would be dependent on the magnitude of loading, which is related to horse speed and gait, but would be independent of the amount of exercise and duration of time the shoes were on the hooves. Therefore, length and width, but not depth, of grooves were quantified in the present study.
Hoof conformation was also unknown and could have been a confounding factor because movement of the hoof wall could have differed among hooves with differences in hoof wall angles (eg, upright hoof or splayed heels). It is possible that conformation of hooves may have played a role in groove development in the present study.
For the study reported here, differences in wear grooves on horseshoes were more apparent among racetracks than between surface types. This highlights the need to quantitatively characterize surface behaviors in future studies. Horseshoe characteristics and size were contributing factors to the dimensions of the wear grooves. The manner in which a horseshoe was attached to a hoof was also a factor in the length of the grooves. Overall, this study revealed that racetrack and shoeing practices were factors that likely affected hoof behavior during racing and race training activities.
Acknowledgments
This manuscript represents a portion of a thesis submitted by Ms. Dahl to the University of California-Davis Department of Animal Biology as partial fulfillment of the requirements for a Master of Science degree.
Supported in part by the California Horse Racing Board and the Center for Equine Health with funds provided by the State of California satellite wagering fund and contributions by private donors.
The authors thank Kirk Breed, Tom McCarthy, Victor Tovar, and Tanya Garcia-Nolen for technical assistance and Dr. Neil Willits for statistical assistance.
Footnotes
Kings Plate XT, Royal Kerchkeart, Vogelwaarde, The Netherlands.
Tradition XT, Royal Kerchkeart, Vogelwaarde, The Netherlands.
Silver Queen, Thoro'bred Inc, Anaheim, Calif.
Queen XT, Thoro'bred Inc, Anaheim, Calif.
Synergy, Royal Kerchkeart, Vogelwaarde, The Netherlands.
Regular, Royal Kerchkeart, Vogelwaarde, The Netherlands.
Nikon D1X camera with 60-mm Nikkor lens, Nikon, Melville, NY.
Image J, US National Institutes of Health. Available at: rsbweb.nih.gov/ij/. Accessed Jul 3, 2013.
SAS, version 9.4, SAS Institute Inc, Cary, NC.
References
1. Riemersma DJ, van den Bogert AJ, Jansen MO, et al. Influence of shoeing on ground reaction forces and tendon strains in the forelimbs of ponies. Equine Vet J 1996; 28: 126–132.
2. Oomen AM, Oosterlinck M, Pille F, et al. Use of a pressure plate to analyse the toe-heel load redistribution underneath a normal shoe and a shoe with a wide toe in sound warmblood horses at the walk and trot. Res Vet Sci 2012; 93: 1026–1031.
3. Hampson BA, Connelley AD, de Laat MA, et al. Sole depth and weight-bearing characteristics of the palmar surface of the feet of feral horses and domestic Thoroughbreds. Am J Vet Res 2011; 72: 727–735.
4. Schaer BL, Ryan CT, Boston RC, et al. The horse-racetrack interface: a preliminary study on the effect of shoeing on impact trauma using a novel wireless data acquisition system. Equine Vet J 2006; 38: 664–670.
5. Roepstorff L, Johnston C, Drevemo S. In vivo and in vitro heel expansion in relation to shoeing and frog pressure. Equine Vet J Suppl 2001; 33: 54–57.
6. Yoshihara E, Takahashi T, Otsuka N, et al. Heel movement in horses: comparison between glued and nailed horse shoes at different speeds. Equine Vet J Suppl 2010; 38: 431–435.
7. Balch OK, Butler D, Collier MA. Balancing the normal foot: hoof preparation, shoe fit and shoe modification in the performance horse. In: Martinelli MJ, Mair TS, Green RE, eds. Equine veterinary education. Ashford, Kent, England: Geerings of Ashford Ltd, 1998;30–41.
8. Kane AJ, Stover SM, Gardner IA, et al. Hoof size, shape, and balance as possible risk factors for catastrophic musculoskeletal injury of Thoroughbred racehorses. Am J Vet Res 1998; 59: 1545–1552.
9. Peterson ML, McIlwraith CW, Reiser RF. Development of a system for the in-situ characterisation of thoroughbred horse racing track surfaces. Biosystems Eng 2008; 101: 260–269.
10. Symons JE, Garcia TC, Stover SM. Distal hindlimb kinematics of galloping Thoroughbred racehorses on dirt and synthetic racetrack surfaces. Equine Vet J 2014; 46: 227–232.
11. 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.
12. Benoit P, Barrey E, Regnault JC, et al. Comparison of the damping effect of different shoeing by the measurement of hoof acceleration. Acta Anat (Basel) 1993; 146: 109–113.
13. Hinterhofer C, Stanek C, Haider H. Finite element analysis (FEA) as a model to predict effects of farriery on the equine hoof. Equine Vet J Suppl 2001; 33: 58–62.
14. 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 Suppl 1998; 26: 86–95.
15. Firth EC, Rogers CW, Doube M, et al. Musculoskeletal responses of 2-year-old Thoroughbred horses to early training. 6. Bone parameters in the third metacarpal and third metatarsal bones. N Z Vet J 2005; 53: 101–112.
16. Van Heel MCV, 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.
