• View in gallery
    Figure 1—

    High-detail radiograph depicting the location of fractures in proximal sesamoid bones obtained from cadavers of racing Thoroughbreds. The bone on the left has no fracture; notice that the width (horizontal lines) and height (vertical lines) are indicated. Fractures in the other bones (from left to right) are categorized as apical, midbody, basilar, axial, or abaxial.

  • View in gallery
    Figure 2—

    High-detail radiograph depicting fracture orientation in proximal sesamoid bones. Fracture orientation (from left to right) was categorized as complete transverse, split transverse, partial transverse, oblique, or longitudinal.

  • View in gallery
    Figure 3—

    High-detail radiograph depicting fracture configuration in proximal sesamoid bones. Fracture configure was categorized as simple (left), comminuted (middle), or avulsion (right).

  • View in gallery
    Figure 4—

    High-detail radiograph depicting characterization of fracture margins in proximal sesamoid bones. Fracture margins were categorized as distinct (left) or indistinct (right).

  • View in gallery
    Figure 5—

    High-detail radiograph depicting osteophyte location on proximal sesamoid bones. Osteophyte location (white circles) was categorized (from left to right) as apical, axial, abaxial wing, abaxial insertion, basilar abaxial, basilar middle, or basilar axial.

  • View in gallery
    Figure 6—

    High-detail radiograph of a proximal sesamoid bone. Notice that an abaxial vascular channel (within white circle) progresses distal and axial through the bone.

  • View in gallery
    Figure 7—

    Age distribution of horses with fractures of the proximal sesamoid bones that resulted in death or euthanasia of the affected horse. Values reported represent the proportion of affected horses relative to the corresponding population of actively training or racing Thoroughbreds in California and were categorized on the basis of age of horse and fracture location.

  • 1.

    Johnson BJ, Stover SM & Daft BM, et al. Causes of death in racehorses over a 2 year period. Equine Vet J 1994;26:327330.

  • 2.

    Estberg L, Stover SM & Gardner IA, et al. Fatal musculoskeletal injuries incurred during racing and training in Thoroughbreds. J Am Vet Med Assoc 1996;208:9296.

    • Search Google Scholar
    • Export Citation
  • 3.

    Peloso JG, Mundy GD, Cohen ND. Prevalence of, and factors associated with, musculoskeletal racing injuries of Thoroughbreds. J Am Vet Med Assoc 1994;204:620626.

    • Search Google Scholar
    • Export Citation
  • 4.

    Cohen ND, Peloso JG & Mundy GD, et al. Racing-related factors and results of prerace physical inspection and their association with musculoskeletal injuries incurred in Thoroughbreds during races. J Am Vet Med Assoc 1997;211:454463.

    • Search Google Scholar
    • Export Citation
  • 5.

    Hernandez J, Hawkins DL, Scollay MC. Race-start characteristics and risk of catastrophic musculoskeletal injury in Thoroughbred racehorses. J Am Vet Med Assoc 2001;218:8386.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6.

    Krook L, Maylin GA. Fractures in Thoroughbred race horses. Cornell Vet 1988;78 (suppl):1133.

  • 7.

    Stover SM, Johnson BJ & Daft BM, et al. An association between complete and incomplete stress fractures of the humerus in racehorses. Equine Vet J 1992;24:260263.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8.

    Riggs CM. Aetiopathogenesis of parasagittal fractures of the distal condyles of the third metacarpal and third metatarsal bones— review of the literature. Equine Vet J 1999;31:116120.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9.

    Poulos PW. Radiographic and histologic assessment of proximal sesamoid bone changes in young and working horses, in Proceedings. 35th Annu Conv Am Assoc Equine Pract 1989;34:347358.

    • Search Google Scholar
    • Export Citation
  • 10.

    O'Brien TR, Morgan JP & Wheat JD, et al. Sesamoiditis in the Thoroughbred: a radiographic study. J Am Vet Radiol Soc 1971;12:7587.

  • 11.

    Kane AJ, McIlwraith CW & Park RD, et al. Radiographic changes in Thoroughbred yearlings. Part 2: associations with racing performance. Equine Vet J 2003;35:366374.

    • Search Google Scholar
    • Export Citation
  • 12.

    Kane AJ, Park RD & McIlwraith CW, et al. Radiographic changes in Thoroughbred yearlings. Part 1: prevalence at the time of the yearling sales. Equine Vet J 2003;35:354365.

    • Search Google Scholar
    • Export Citation
  • 13.

    Spike-Pierce DL, Bramlage LR. Correlation of racing performance with radiographic changes in the proximal sesamoid bones of 487 Thoroughbred yearlings. Equine Vet J 2003;35:350353.

    • Search Google Scholar
    • Export Citation
  • 14.

    Trumble TN, Arnoczky SP & Stick JA, et al. Clinical relevance of the microvasculature of the equine proximal sesamoid bone. Am J Vet Res 1995;56:720724.

    • Search Google Scholar
    • Export Citation
  • 15.

    Hardy J, Marcoux M, Breton L. Clinical relevance of radiographic findings in proximal sesamoid bones of two-year-old Standardbreds in their first year of race training. J Am Vet Med Assoc 1991;198:20892094.

    • Search Google Scholar
    • Export Citation
  • 16.

    Grondahl AM, Gaustad G, Engeland A. Progression and association with lameness and racing performance of radiographic changes in the proximal sesamoid bones of young standardbred trotters. Equine Vet J 1994;26:152155.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17.

    Hill AE, Gardner IA & Carpenter TE, et al. Effects of injury to the suspensory apparatus, exercise, and horseshoe characteristics on the risk of lateral condylar fracture and suspensory apparatus failure in forelimbs of Thoroughbred racehorses. Am J Vet Res 2004;65:15081517.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18.

    Schneider RK. Incidence and location of fractures of the proximal sesamoids and proximal extremity of the first phalanx, in Proceedings. 26th Annu Conv Am Assoc Equine Pract 1980;25:157158.

    • Search Google Scholar
    • Export Citation
  • 19.

    Fretz PB, Barber SM & Bailey JV, et al. Management of proximal sesamoid bone fractures in the horse. J Am Vet Med Assoc 1984;185:282284.

  • 20.

    Watanabe T. Localized lesions of bone in the fetlock joint and their diagnosis by xeroradiography in racehorses. Centaur 1985;2:7984.

    • Search Google Scholar
    • Export Citation
  • 21.

    Bassage LH II, Richardson DW. Longitudinal fractures of the condyles of the third metacarpal and metatarsal bones in racehorses: 224 cases (1986–1995). J Am Vet Med Assoc 1998;212:17571764.

    • Search Google Scholar
    • Export Citation
  • 22.

    Greet TRC. Condylar fracture of the cannon bone with axial sesamoid fracture in three horses. Vet Rec 1987;120:223225.

  • 23.

    Barclay WP, Foerner JJ, Phillips TN. Axial sesamoid injuries associated with lateral condylar fractures in horses. J Am Vet Med Assoc 1985;186:278279.

    • Search Google Scholar
    • Export Citation
  • 24.

    Weaver JC, Stover SM, O'Brien TR. Radiographic anatomy of soft tissue attachments in the equine metacarpophalangeal and proximal phalangeal region. Equine Vet J 1992;24:310315.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 25.

    Wisner ER, O'Brien TR & Pool RR, et al. Osteomyelitis of the axial border of the proximal sesamoid bones in seven horses. Equine Vet J 1991;23:383389.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 26.

    Cornelissen BPM, Rijkenhuizen ABM, Barneveld A. The arterial shift features in the equine proximal sesamoid bone. Vet Q 1996;18:S110S116.

Advertisement

Association between findings on palmarodorsal radiographic images and detection of a fracture in the proximal sesamoid bones of forelimbs obtained from cadavers of racing Thoroughbreds

Lucy A. AnthenillJD Wheat Veterinary Orthopedic Research Laboratory, Department of Anatomy, Physiology, and Cell Biology, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Lucy A. Anthenill in
Current site
Google Scholar
PubMed
Close
 DVM
,
Susan M. StoverJD Wheat Veterinary Orthopedic Research Laboratory, Department of Anatomy, Physiology, and Cell Biology, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Susan M. Stover in
Current site
Google Scholar
PubMed
Close
 DVM, PhD
,
Ian A. GardnerDepartment of Medicine and Epidemiology, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Ian A. Gardner in
Current site
Google Scholar
PubMed
Close
 BVSc, MPVM, PhD
,
Ashley E. HillJD Wheat Veterinary Orthopedic Research Laboratory, Department of Anatomy, Physiology, and Cell Biology, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Ashley E. Hill in
Current site
Google Scholar
PubMed
Close
 DVM, MPVM, PhD
,
Christina M. LeeJD Wheat Veterinary Orthopedic Research Laboratory, Department of Anatomy, Physiology, and Cell Biology, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Christina M. Lee in
Current site
Google Scholar
PubMed
Close
 BS
,
Mark L. AndersonCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Mark L. Anderson in
Current site
Google Scholar
PubMed
Close
 DVM
,
Bradd C. BarrCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Bradd C. Barr in
Current site
Google Scholar
PubMed
Close
 DVM
,
Deryck H. ReadCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Deryck H. Read in
Current site
Google Scholar
PubMed
Close
 DVM
,
Bill J. JohnsonCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Bill J. Johnson in
Current site
Google Scholar
PubMed
Close
 DVM
,
Leslie W. WoodsCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Leslie W. Woods in
Current site
Google Scholar
PubMed
Close
 DVM
,
Barbara M. DaftCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Barbara M. Daft in
Current site
Google Scholar
PubMed
Close
 DVM
,
Hailu KindeCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Hailu Kinde in
Current site
Google Scholar
PubMed
Close
 DVM, MPVM
,
Janet D. MooreCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Janet D. Moore in
Current site
Google Scholar
PubMed
Close
 DVM
,
Cynthia A. FarmanCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Cynthia A. Farman in
Current site
Google Scholar
PubMed
Close
 DVM
,
Jenee S. OdaniCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Jenee S. Odani in
Current site
Google Scholar
PubMed
Close
 DVM
,
Patricia A. PesaventoCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Patricia A. Pesavento in
Current site
Google Scholar
PubMed
Close
 DVM
,
Francisco A. UzalCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Francisco A. Uzal in
Current site
Google Scholar
PubMed
Close
 DVM
,
James T. CaseCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by James T. Case in
Current site
Google Scholar
PubMed
Close
 DVM, PhD
, and
Alex A. ArdansCalifornia Animal Health and Food Safety Laboratory System, School of Veterinary Medicine, University of California, Davis, CA 95616.

Search for other papers by Alex A. Ardans in
Current site
Google Scholar
PubMed
Close
 DVM

Abstract

Objective—To determine the distribution for limbs and bones in horses with fractures of the proximal sesamoid bones and relationships with findings on palmarodorsal radiographic images.

Sample Population—Proximal sesamoid bones obtained from both forelimbs of cadavers of 328 racing Thoroughbreds.

Procedure—Osteophytes; large vascular channels; and fracture location, orientation, configuration, and margin distinctness were categorized by use of high-detail contact palmarodorsal radiographs. Distributions of findings were determined. Relationships between radiographic findings and fracture characteristics were examined by use of χ2 and logistic regression techniques.

Results—Fractures were detected in 136 (41.5%) horses. Biaxial fractures were evident in 109 (80%) horses with a fracture. Osteophytes and large vascular channels were evident in 266 (81%) and 325 (99%) horses, respectively. Medial bones typically had complete transverse or split transverse simple fractures, indistinct fracture margins, > 1 vascular channel that was > 1 mm in width, and osteophytes in abaxial wing and basilar middle or basilar abaxial locations. Lateral bones typically had an oblique fracture and distinct fracture margins. Odds of proximal sesamoid bone fracture were approximately 2 to 5 times higher in bones without radiographic evidence of osteophytes or large vascular channels, respectively.

Conclusions and Clinical Relevance—Biaxial fractures of proximal sesamoid bones were common in cadavers of racing Thoroughbreds. Differences between medial and lateral bones for characteristics associated with fracture may relate to differences in fracture pathogeneses for these bones. Osteophytes and vascular channels were common findings; however, fractures were less likely to occur in bones with these features.

Abstract

Objective—To determine the distribution for limbs and bones in horses with fractures of the proximal sesamoid bones and relationships with findings on palmarodorsal radiographic images.

Sample Population—Proximal sesamoid bones obtained from both forelimbs of cadavers of 328 racing Thoroughbreds.

Procedure—Osteophytes; large vascular channels; and fracture location, orientation, configuration, and margin distinctness were categorized by use of high-detail contact palmarodorsal radiographs. Distributions of findings were determined. Relationships between radiographic findings and fracture characteristics were examined by use of χ2 and logistic regression techniques.

Results—Fractures were detected in 136 (41.5%) horses. Biaxial fractures were evident in 109 (80%) horses with a fracture. Osteophytes and large vascular channels were evident in 266 (81%) and 325 (99%) horses, respectively. Medial bones typically had complete transverse or split transverse simple fractures, indistinct fracture margins, > 1 vascular channel that was > 1 mm in width, and osteophytes in abaxial wing and basilar middle or basilar abaxial locations. Lateral bones typically had an oblique fracture and distinct fracture margins. Odds of proximal sesamoid bone fracture were approximately 2 to 5 times higher in bones without radiographic evidence of osteophytes or large vascular channels, respectively.

Conclusions and Clinical Relevance—Biaxial fractures of proximal sesamoid bones were common in cadavers of racing Thoroughbreds. Differences between medial and lateral bones for characteristics associated with fracture may relate to differences in fracture pathogeneses for these bones. Osteophytes and vascular channels were common findings; however, fractures were less likely to occur in bones with these features.

Racing and training Thoroughbreds often die or are euthanized because of injuries to the suspensory apparatus, including fractures of the proximal sesamoid bones.1–3 Fractures of the proximal sesamoid bones in the forelimbs were detected in 30% to 50% of Thoroughbreds in California that died as a result of injuries during racing or training.1,2 Fractures of the proximal sesamoid bones of the forelimbs were also common fatal musculoskeletal injuries of Thoroughbreds at several racetracks in Kentucky,4 Florida,5 and New York.6 Injuries to the proximal sesamoid bones accompanied by injury to the suspensory apparatus or fracture of the third metacarpal condyle also account for a large number of noncatastrophic career-ending injuries.4 Because fractures of the proximal sesamoid bones are common and reduce the likelihood for future performance, prevention of such fractures should be a high priority.

Analysis of evidence indicates that some common bone fractures in equine athletes are the acute manifestation of a chronic process. Stress fractures predispose long bones to complete fractures. Stress fractures may be evident as periosteal or endosteal callus or an incomplete fracture line.7 Fractures of the lateral condyle of the third metacarpal bone have been associated with subchondral bone changes.8 Early detection of bony changes allows modification of training regimens to allow resolution of existing damage and prevent catastrophic injury. On the basis of findings in long bones, bony changes may also precede fractures of proximal sesamoid bones, and detection of changes may be useful for preventing fractures in proximal sesamoid bones.

In addition to fractures, other radiographic findings for proximal sesamoid bones include enthesophytes and osteophytes9,10 as well as vascular channels.11–15 These visible changes may be indicative of pending fracture of proximal sesamoid bones.

Proximal sesamoid bones develop 2 types of osteophytes that can be detected on palmarodorsal radiographs (periarticular osteophytes that form along the articular surface margins and enthesophytes that arise within the fibrous attachments of the suspensory, intersesamoidean, and distal sesamoidean ligaments).9,10 Enthesophytes and osteophytes are most often seen in the periarticular apical region associated with the nonarticular portion of the abaxial surface and at the basilar margin; they can coincide in multiple locations. Enthesophytes and osteophytes have been associated with clinical lameness and sesamoiditis.10 Yearling Thoroughbreds with enthesophytes on the proximal sesamoid bones of the forelimbs, which were detected on survey radiographs routinely conducted on all yearling Thoroughbreds, had fewer race starts, compared with the number of race starts for yearlings without these changes.11

Vascular channels appear as linear lucencies within the body of proximal sesamoid bones. In addition, funnel-shaped lucencies at the abaxial margin of the bone may continue as linear lucencies.12–15 These channels most commonly course from the abaxial surface toward the axial surface. Abnormal vascular channels have been defined as channels within the body of a proximal sesamoid bone that are ≥ 2 mm in width and have nonparallel margins on oblique radiographic views of the forelimb.13 Other investigators have considered funnel-shaped lucencies located abaxial to the body of the bone to be abnormal.12,15

Abnormal vascular channels have been associated with poor racing performance. In a radiographic study13 of 487 racing 2to 3-year-old Thoroughbreds, a lower mean number of race starts and lower total career earnings were associated with abnormal vascular channels observed on survey radiographs taken of these horses when they were yearlings, compared with values for horses that did not have these changes when they were yearlings. In contrast, no relationship was evident between race performance and abnormal vascular channels detected on prepurchase radiographs of yearlings for 946 Thoroughbreds.11 Similarly, in a radiographic study16 of 753 young Standardbred trotters for which radiographs were taken when the horses were 6 to 21 months of age, abnormal vascular channels were not associated with total earnings for the racing season when the horses were 3 to 4 years old.

Identification of patterns of distribution for fractures of the proximal sesamoid bones, osteophytes, and vascular channels may be useful to trainers, owners, and veterinarians in understanding the etiopathogenesis of changes in the proximal sesamoid bones that lead to fractures in racing Thoroughbreds. We hypothesized that similar to radiographic findings in long bones,7,8 radiographic findings (vascular channels and osteophytes) in proximal sesamoid bones may be associated with fractures. Therefore, the objectives of the study reported here were to document by use of palmarodorsal radiographs the prevalence and distribution of fractures of proximal sesamoid bones, osteophytes, and vascular channels within a group of cadavers of racing and training Thoroughbreds in California and examine any relationships between radiographic findings and fractures. These results may provide insight into the pathogenesis of injuries to the proximal sesamoid bones and be valuable for use in preventing fractures of the proximal sesamoid bones.

Materials and Methods

Study population—Proximal sesamoid bones from both forelimbs of 328 Thoroughbred cadavers obtained between November 11, 1999, and December 19, 2002, through the CAHFSLS for the CHRB Postmortem Program were examined. Pathologists at the CAHFSLS examine all horses that die or are euthanized on sanctioned racetrack premises in California. The 328 horses were a subset of 660 Thoroughbreds examined through the CHRB Postmortem Program during that time period and were reported as part of another study.17 Age and sex for all horses that raced or had official timed workouts at racetracks in California during the same time period were obtained as a customized report from a commercial database.a

Data collection—Necropsy data were obtained from CAHFSLS records. Postmortem information included age and sex of each racehorse and cause, date, and location of death. Contact palmarodorsal radiographs were taken of the paired medial and lateral proximal sesamoid bones attached to the intersesamoidean ligament for both forelimbs by use of a cabinet radiographic unitb on high-detail mammography film.c Radiographs were scannedd and saved as digital image files.e All observations were made by 1 investigator (LAA) who used high-detail radiographs or digital images of the radiographs.

Bone height and width were determined only for intact proximal sesamoid bones. Height and width were determined parallel and perpendicular, respectively, to the axial margin of each bone by rotating the figure for alignment on a rectangular grid (Figure 1).

Figure 1—
Figure 1—

High-detail radiograph depicting the location of fractures in proximal sesamoid bones obtained from cadavers of racing Thoroughbreds. The bone on the left has no fracture; notice that the width (horizontal lines) and height (vertical lines) are indicated. Fractures in the other bones (from left to right) are categorized as apical, midbody, basilar, axial, or abaxial.

Citation: American Journal of Veterinary Research 67, 5; 10.2460/ajvr.67.5.858

Evidence, or lack thereof, of fractures, osteophytes, and vascular channels was documented for each proximal sesamoid bone. Fractures within bones were categorized on the basis of location, orientation, configuration, and margin quality. Osteophytes were categorized on the basis of location. Width and number of vascular channels were recorded.

Fracture location—Fracture location was categorized as apical (proximal third), midbody (middle third), or basilar (distal third) for fractures with a major transverse component or as abaxial or axial for longitudinally oriented fractures adjacent to the abaxial and axial surfaces, respectively (Figure 1). For all bones with fractures, division into thirds for assessment of fracture location was made on the basis of measurements obtained from the corresponding intact medial or lateral contralateral bone.

Fracture orientation—Fracture orientation was categorized as complete transverse, split transverse, partial transverse, oblique, or longitudinal (Figure 2). Complete transverse fractures extended from axial to abaxial surfaces, deviated from a transverse plane perpendicular to the axial margin of the bone by < 30°, and did not have evidence of other major fracture components. Split transverse fractures had a complete transverse fracture with an additional longitudinal component that divided the distal fragment into 2 pieces. Partial transverse fractures started on the axial or abaxial surface but deviated before reaching the other surface and entered the basilar portion of the bone. Oblique fractures deviated from the transverse plane by > 30° but < 60°. Longitudinal fractures deviated from the transverse plane by > 60°.

Figure 2—
Figure 2—

High-detail radiograph depicting fracture orientation in proximal sesamoid bones. Fracture orientation (from left to right) was categorized as complete transverse, split transverse, partial transverse, oblique, or longitudinal.

Citation: American Journal of Veterinary Research 67, 5; 10.2460/ajvr.67.5.858

Fracture configuration—Fracture configuration was categorized as simple, comminuted, or avulsion (Figure 3). A simple fracture divided the bone into 2 major fragments. Comminuted fractures had ≥ 3 fracture fragments that were larger than 1 cm in any dimension, which excluded small bone fragments (ie, chips). Avulsion fractures were near a bone margin within the origin or insertion of a ligamentous attachment to the proximal sesamoid bones.

Figure 3—
Figure 3—

High-detail radiograph depicting fracture configuration in proximal sesamoid bones. Fracture configure was categorized as simple (left), comminuted (middle), or avulsion (right).

Citation: American Journal of Veterinary Research 67, 5; 10.2460/ajvr.67.5.858

Fracture margins—Fracture margins were subjectively categorized as distinct when any portion of the edge of the fracture surface was sharp and smooth (Figure 4). In contrast, fracture margins were categorized as indistinct when the margin was blurry or fuzzy.

Figure 4—
Figure 4—

High-detail radiograph depicting characterization of fracture margins in proximal sesamoid bones. Fracture margins were categorized as distinct (left) or indistinct (right).

Citation: American Journal of Veterinary Research 67, 5; 10.2460/ajvr.67.5.858

Osteophytes—All bony outgrowths were classified in the general category of osteophytes because it was difficult to distinguish enthesophytes from periarticular osteophytes on a single palmarodorsal radiographic projection. Osteophyte location was categorized as apical, axial, or abaxial in the proximal region; basilar axial; basilar middle; or basilar abaxial in the basilar region (Figure 5). Apical osteophytes were at the proximal periarticular margin of the bone, adjacent to the insertion of the intersesamoidean ligament and the medial or lateral branches of the suspensory ligament. Axial osteophytes were along the axial margin between the bone pair within the attachment of the intersesamoidean ligament. Abaxial osteophytes were subdivided into those at the abaxial insertion of the medial and lateral branches of the suspensory ligament (hereafter referred to as the abaxial fossa) and those at the portion of the abaxial surface that projects proximal to the insertion sites of the suspensory ligament and forms a nonarticular convex ridge of bone covered by the fibrous tissue of the intersesamoidean ligament (hereafter referred to as the abaxial wing). Basilar osteophytes were subclassified into basilar axial, basilar middle, or basilar abaxial osteophytes on the basis of location within the respective one third of the basilar margin of each proximal sesamoid bone.

Figure 5—
Figure 5—

High-detail radiograph depicting osteophyte location on proximal sesamoid bones. Osteophyte location (white circles) was categorized (from left to right) as apical, axial, abaxial wing, abaxial insertion, basilar abaxial, basilar middle, or basilar axial.

Citation: American Journal of Veterinary Research 67, 5; 10.2460/ajvr.67.5.858

Vascular channels—Evidence of vascular channels was assessed for abaxial and basilar locations (Figure 6). Abaxial vascular channels were detected when they were visible at the abaxial margin and coursed within the body of the bone toward the axial margin. Basilar channels extended proximally from the base of the proximal sesamoid bone into the body. For abaxial channels, the number of visible channels and number of channels with a width > 1 mm at any point along the channel length were recorded.

Figure 6—
Figure 6—

High-detail radiograph of a proximal sesamoid bone. Notice that an abaxial vascular channel (within white circle) progresses distal and axial through the bone.

Citation: American Journal of Veterinary Research 67, 5; 10.2460/ajvr.67.5.858

Statistical analysis—Distributions for age and sex of horses for the sample population and the population of all horses that died and entered the CHRB Postmortem Program and between the populations of all horses that raced or trained in California during the same time period as the study period were compared by use of contingency tables and χ2 analyses, except for categories with sparse data. For comparisons in contingency tables, horses that were ≥ 6 years old were combined into a category to provide values > 5 for minimum expected cell frequencies.

Differences in bone height and width between limbs (left and right forelimbs) and bones (lateral and medial) were examined by use of paired t tests. Contingency tables, χ2 analyses, or the McNemar test for paired data were used to examine limb and bone distributions among categoric variables for all bones. Logistic regression models were used similarly, but they were also used to account for multiple observations from left and right limbs and medial and lateral bones from each horse. Logistic regression models typically included limb, bone, horse, sex (female, gelding, and sexually intact male), and the limb-by-bone interaction. Comparisons between the distributions of levels of dependent variables on the basis of age of horse were considered in χ2 models because of the categoric nature of the age variable. The relationships between fractures and osteophytes and fractures and vascular channels were explored by use of similar contingency tables and logistic regression models to obtain odds ratios. Significance was set at values of P ≤ 0.05. Results for χ2 tests were reported, except when logistic regression models differed in significance. Only significant interactions were reported.

Results

Study population—During the 37-month period, radiographic features of 1,312 proximal sesamoid bones from the forelimbs of 328 Thoroughbred cadavers were examined (Table 1). This convenience sample represented 49% of all Thoroughbreds that died or were euthanized and that were subsequently necropsied as part of the CHRB Postmortem Program during this time period. Females (125 [38%]) and geldings (131 [40%]) were more common than sexually intact males (72 [22%]). Most commonly, horses were 3 years old (114 [35%]). The proportion of 3-year-old sexually intact male horses (46%) was significantly (P = 0.015) higher, compared with the proportion of 3-yearold females (30%) or geldings (33%).

Table 1—

Age and sex distributions of the sample population of 328 Thoroughbred cadavers, all CHRB Postmortem Program submissions, and all horses that raced or had official timed workouts California during the 37-month study period.

SexAge (y)Total
12345≥ 6
Sample population
 Females0243834199124 (38)
 Geldings01944252222132 (40)
 Sexually intact males01333183572 (22)
Total056 (17)115 (35)77 (23)44 (14)36 (11)328
CHRB Postmortem Program submissions
 Females15280582819238 (36)
 Geldings03077474155250 (38)
 Sexually intact males0239673133172 (26)
Total184 (13)196 (30)172 (26)100 (15)107 (16)660
California racetrack population
 Females15,1645,8383,4511,6101,11717,181 (44)
 Geldings13,2713,6702,6181,6512,15113,362 (34)
 Sexually intact males22,3852,7121,7639177718,550 (22)
Total410,820 (28)12,220 (31)7,832 (20)4,178 (11)4,039 (10)39,093

Numbers in parentheses are percentages.

Sex distribution—Sex distribution for the sample population did not differ significantly (P = 0.368) from that of all CHRB submissions (Table 1). Sex distribution of horses that had a race or official timed workout in California during the same time period as the study differed significantly from the sex distribution for horses in the study population (P = 0.043) and CHRB Postmortem Program (P < 0.001). Geldings were overrepresented and females were underrepresented in the study population and CHRB submissions.

Age distribution—Age distribution for the study population did not differ significantly (P = 0.094) from that of the CHRB submissions (Table 1). Age distribution of horses that had a race or official timed workout in California during the same time period as the study differed significantly from the age distribution for horses in the study population (P = 0.001) and CHRB submissions (P < 0.001). Two-year-old horses were underrepresented and older horses (> 3 years old) overrepresented in the study population and CHRB submissions.

Bone dimensions—Bone height was significantly (P = 0.005) less for medial bones (mean ± SD of 3.13 ± 0.17 cm and 3.13 ± 0.16 cm for the left and right forelimbs, respectively) than for lateral bones (3.17 ± 0.17cm and 3.17 ± 0.16 cm for the left and right forelimbs, respectively) but did not differ significantly (P = 0.650) between forelimbs. Mean bone width was significantly (P < 0.001) greater for medial bones of the left forelimb (2.47 ± 0.12 cm), compared with mean width of lateral bones of the left forelimb (2.43 ± 0.12 cm), but did not differ significantly (P = 0.500) between medial bones of the right forelimb (2.50 ± 0.12 cm), compared with width for lateral bones of the right forelimb (2.50 ± 0.12 cm). Bones of the left forelimb had significantly (P < 0.001) less width, compared with width of bones of the right forelimb.

Fractures—Fractures were detected in 251 proximal sesamoid bones of 136 (41.5%) horses of the study population. Biaxial fractures in which both the medial and lateral bones within 1 limb were fractured were detected in 109 (80%) horses with a fracture of the proximal sesamoid bones, whereas a single bone within a limb was fractured in 27 (20%) horses. Six horses had at least 1 fracture in a proximal sesamoid bone in both forelimbs, and 1 horse had a fracture in all 4 proximal sesamoid bones in the forelimbs. We did not detect significant differences for fracture distributions between left (129) and right (122) forelimbs (P = 0.535) or between lateral (121) and medial (130) bones (P = 0.426).

Fracture location—Fractures were most commonly detected in the midbody and basilar locations (Table 2). Eighty-four horses had 109 midbody fractures, 70 horses had 81 basilar fractures, and 29 horses had 30 axial fractures. Apical fractures and abaxial fractures were less common. We did not detect significant (P = 0.904) differences for the distributions of midbody fractures between the left and right forelimbs or between medial and lateral bones. Basilar fractures were significantly (P < 0.001) more frequent in medial bones (56 medial basilar fractures of 81 basilar fractures [69%]), compared with the frequency in lateral bones (25/81 [31%]), but we did not detect a significant (P = 0.264) difference for the distribution of basilar fractures between the left and right forelimbs. Axial fractures were significantly (P < 0.001) more frequent in lateral bones (27/30 axial fractures [90%]), compared with the frequency in medial bones (3/30 [10%]), but we did not detect a significant (P = 0.656) difference for the distribution of axial fractures between the left and right forelimbs. Similarly, we did not detect significant differences in the distribution between forelimbs for apical (P = 0.950) or abaxial (P = 0.226) fractures or between medial or lateral bones for apical (P = 0.594) or abaxial (P = 0.062) fractures.

Table 2—

Location of fractures in proximal sesamoid bones of 328 Thoroughbred cadavers on the basis of forelimb and proximal sesamoid bone.

LocationLeft forelimbRight forelimbTotal
LateralMedialMedialLateral
Apical774624
Midbody27272728109
Basilar1529271081
Abaxial02417
Axial14121330

Fracture orientation—Bones most commonly had a complete transverse fracture (100/251 [40%] fractured bones), followed by bones with an oblique (54 [22%]), split transverse (39 [16%]), longitudinal (37 [14%]), or partial transverse (21 [8%]) fracture (Table 3). Complete transverse fractures were detected significantly (P < 0.001) more often in medial bones (70%), compared with the frequency in lateral bones (30%), but we did not detect a significant (P = 0.381) difference in frequency between the left and right forelimbs. Split transverse fractures also were detected significantly (P < 0.001) more often in the medial bones (77%), compared with the frequency in the lateral bones (23%), but we did not detect a significant (P = 0.766) difference in frequency between the left and right forelimbs.

Table 3—

Orientation of fractures in proximal sesamoid bones of 328 Thoroughbred cadavers on the basis of forelimb, proximal sesamoid bone, and fracture location.

LocationLeft forelimbRight forelimb
LateralMedialMedialLateral
Apex
 Complete transverse3431
 Oblique4315
Midbody
 Complete transverse9221713
 Oblique155713
 Partial transverse0001
 Split transverse3031
Base
 Complete transverse313120
Oblique1000
 Partial transverse9137
 Split transverse215123
Axial longitudinal141213
Abaxial longitudinal0241

Oblique fractures of the proximal sesamoid bones were detected significantly (P < 0.001) more often in lateral bones (38/54 [70%] oblique fractures), compared with the frequency in medial bones (16/54 [30%]). Similarly, partial transverse fractures were detected significantly (P < 0.001) more often in lateral bones (17/21 [81%] partial transverse fractures), compared with the frequency in medial bones (4/21 [19%]). We did not detect significant differences between the left and right forelimbs for frequency of oblique (P = 0.804) or partial transverse (P = 0.446) fractures. Longitudinal fractures were detected significantly (P < 0.001) more often in lateral bones (28/37 [75%] longitudinal fractures), compared with the frequency in medial bones (9/37 [25%]), but there was not a significant (P = 0.366) difference in distributions between the left and right forelimbs. Complete transverse and oblique fractures were detected significantly (P < 0.001) more often in the midbody and base of the proximal sesamoid bones, whereas split and partial transverse fractures were detected significantly (P < 0.001) more often in the base of the proximal sesamoid bones.

Fracture configuration—Simple fractures were detected most commonly (151/251 [60%] fractured proximal sesamoid bones), followed by comminuted (61 [24%]) and avulsion (39 [16%]) fractures (Table 4). Simple fractures were detected significantly (P = 0.004) more often on medial bones (89 [59%]), compared with the frequency for lateral bones (62 [41%]), largely as a result of fractures in the basilar region; we did not detect significant (P = 0.575) differences in distributions between the left and right forelimbs. Simple and comminuted fractures were detected significantly (P < 0.001) more often within the middle of the body of the proximal sesamoid bone. Comminuted fractures were detected significantly (P = 0.011) more often on the left forelimb (38/61 [62%] comminuted fractures), compared with the frequency for the right forelimb (23/61 [38%]); there was not a significant (P = 0.707) difference in distributions between the lateral and medial proximal sesamoid bones. Axial avulsion fractures were detected significantly (P < 0.001) more often on lateral bones (28/31 [90%] axial avulsion fractures), compared with the frequency for medial bones (9/31 [10%]), but distributions did not differ significantly (P = 0.250) between the left and right forelimbs. Abaxial avulsion fractures were most common on medial bones (6/7 fractures).

Table 4—

Configuration of fractures of the proximal sesamoid bones in 328 Thoroughbred cadavers on the basis of forelimb, proximal sesamoid bone, and fracture location.

LocationLeft forelimbRight forelimb
LateralMedialMedialLateral
Apical
 Simple6436
 Comminuted1310
Avulsion0000
Midbody
 Simple16161820
 Comminuted111198
 Avulsion0000
Basilar
 Simple724247
 Comminuted7532
Avulsion1001
Axial avulsion141213
Abaxial avulsion0241

Fracture margins—Fracture margins were distinct on the radiographic image of 164 of 251 (65%) fractured proximal sesamoid bones (Table 5). There was not a significant (P = 0.550) difference in appearance of the fracture margin between the left and right forelimbs, whereas fracture margins were significantly (P < 0.001) more distinct for lateral proximal sesamoid bones (78% of lateral bones), compared with results for the medial proximal sesamoid bones (54% of medial bones).

Table 5—

Margin quality of fractured proximal sesamoid bones of 328 Thoroughbred cadavers on the basis of forelimb and proximal sesamoid bone as determined by examination of palmarodorsal radiographic images.

LocationLeft forelimbRight forelimbTotal
LateralMedialMedialLateral
Distinct47353547164
Indistinct (ie, fuzzy)1631291187

Osteophytes—Osteophytes were detected at the apex, abaxial fossa, abaxial wing, or base of the proximal sesamoid bone in 267 horses (81% of the sample population; Table 6). Fifty-eight (22%) horses had osteophytes of all 4 proximal sesamoid bones. Osteophytes were detected significantly (P < 0.001) more often in the medial bones (395/666 [59%] bones with osteophytes), compared with the frequency for the lateral bones (271 [41%]). There was not a significant (P = 0.144) difference in distribution of osteophytes between the left and right forelimbs.

Table 6—

Distribution of osteophytes and vascular channels in proximal sesamoid bones of 328 Thoroughbred cadavers on the basis of forelimb and proximal sesamoid bone.

Radiographic changeLeft forelimbRight forelimb
LateralMedialMedialLateral
Proximal osteophytes
 Apex3523
 Axial20181111
 Abaxial fossa16284119
 Abaxial wing30947238
Basilar osteophytes
 Basilar abaxial7713612082
 Basilar middle9514413687
 Basilar axial25241426
Vascular channels
 All271291290269
 > 1 mm in width8311610390

Osteophytes were detected significantly (P = 0.001) more often on the base (996/1,377 [70%] distinct osteophytes) than on the apex, axial, or abaxial surfaces (411/1,377 [30%]) of the proximal sesamoid bones. Two hundred fifteen horses had a basilar osteophyte in at least 1 of the 4 proximal sesamoid bones, whereas 153 had an osteophyte within the body or apex of at least 1 of the 4 proximal sesamoid bones.

Significantly (P < 0.001) more bones had basilar osteophytes in the basilar middle (462 basilar middle osteophytes of 996 total basilar osteophytes [48%]) and basilar abaxial (416/996 [43%]) regions, compared with the frequency for the basilar axial region (89/996 [9%]). Basilar osteophytes were detected significantly (P < 0.001) more often in the medial bones (574/996 [59%]), compared with the frequency for the lateral bones (392/996 [41%]). Distribution of basilar osteophytes did not differ significantly (P = 0.059) between the left (501/996 [52%]) and right (465/996 [48%]) forelimbs.

Abaxial wing osteophytes were the most common osteophyte in the proximal region. These osteophytes were detected significantly (P < 0.001) more often on the medial bones (166/234 [71%] abaxial wing osteophytes), compared with the frequency for the lateral bones (68/234 [29%]); the distribution did not differ significantly (P = 0.467) between the left and right forelimbs. Abaxial fossa osteophytes were detected significantly (P < 0.001) more often on the medial bones (69/104 [68%] abaxial fossa osteophytes), compared with the frequency for the lateral bones (35/104 [34%]), and significantly (P = 0.037) more often on the right forelimb (60/104 [58%]), compared with the frequency for the left forelimb (44/104 [42%]). Axial margin osteophytes were detected significantly (P = 0.002) more often on the left forelimb (38/60 [63%] axial margin osteophytes), compared with the frequency for the right forelimb (22/60 [37%]); there was not a significant (P = 0.712) difference in distribution between the medial and lateral bones.

Vascular channels—Three hundred twenty-four (99%) horses in the study population had abaxial vascular channels visible on high-detail radiographs, whereas 11 horses (13 proximal sesamoid bones) had basilar vascular channels (Table 6). Abaxial vascular channels were detected significantly (P < 0.001) more often in the medial proximal sesamoid bones, compared with the frequency for the lateral proximal sesamoid bones, when examined at a horse-incident level (logistic regression); however, we did not detect a significant (P = 0.106) difference in distribution between medial and lateral bones by use of a χ2 analysis. There was not a significant (P = 0.773) difference in distribution of abaxial vascular channels between the left and right forelimbs.

Proximal sesamoid bones most often had 1 or 2 visible channels, and rarely were > 3 channels visible. Channels measuring > 1 mm in width were seen in 221 (67%) horses. There was a significant (P = 0.012) difference in distribution of these larger channels between medial (219/392 [56%] bones with channels > 1 mm in width) and lateral (173/392 [44%]) bones; however, there was not a significant (P = 0.761) difference in distribution between the left and right forelimbs.

Basilar vascular channels were detected extremely infrequently (13 bones with visible basilar vascular channels); therefore, statistical analysis of pattern distribution was not performed. Basilar vascular channels were visible in 3 lateral and 4 medial proximal sesamoid bones in the left forelimb and 3 medial and 3 lateral proximal sesamoid bones in the right forelimb.

Association of sex or age with fractures—We did not detect significant differences between distributions of fractures of the proximal sesamoid bones in the study population among sex groups (females, geldings, and sexually intact males; P = 0.116) or among age groups (2, 3, 4, 5, and ≥ 6 years old; P = 0.659). Fiveyear-old horses had the highest prevalence of fatal fractures when the population of horses that had raced or had an official timed workout during the study period was used as the denominator (Figure 7).

Figure 7—
Figure 7—

Age distribution of horses with fractures of the proximal sesamoid bones that resulted in death or euthanasia of the affected horse. Values reported represent the proportion of affected horses relative to the corresponding population of actively training or racing Thoroughbreds in California and were categorized on the basis of age of horse and fracture location.

Citation: American Journal of Veterinary Research 67, 5; 10.2460/ajvr.67.5.858

Association of sex or age with fracture location— Associations between sex or age groups and fracture location (midbody vs apical and basilar) per affected horse were examined. There was a significant (P = 0.005) difference in sex distribution for bones with a midbody fracture; 20 of 124 (16%) females, 38 of 132 (29%) geldings, and 26 of 72 (36%) sexually intact males had bones with a midbody fracture. There was not a significant (P = 0.840) difference in distribution of bones with a midbody fracture among age groups. Distributions of basilar fractures did not differ significantly among sex (P = 0.734) or age (P = 0.569) groups. Five-year-old horses had the highest incidence of midbody, basilar, and axial fractures within the racetrack population, whereas incidence of apical fractures tended to peak in horses at 6 years of age (Figure 7).

Association between sex or age and osteophytes—Osteophytes were observed in 203 (81%) horses in the sample population. We did not detect significant associations for the incidence of osteophytes among sex (P = 0.160) or age (P = 0.480) groups.

Association between sex or age and vascular channels—Distribution of large vascular channels differed significantly (P < 0.001) among sex groups (56% of females, 79% of geldings, and 67% of sexually intact males had at least 1 visible channel > 1 mm in width). Distribution of large vascular channels did not differ significantly (P = 0.130) among age groups.

Association between fractures of proximal sesamoid bones and osteophytes or vascular channels—Detection of osteophytes and vascular channels was compared to detection of fractures within proximal sesamoid bones (Table 7). Osteophytes were observed in 99 of 251 (39%) fractured proximal sesamoid bones and 567 of 1,061 (53%) nonfractured bones. Fractures were observed in 99 of 666 (15%) bones with osteophytes and 152 of 646 (23%) proximal sesamoid bones without osteophytes. The likelihood of fracture in bones with osteophytes was approximately half (odds ratio, 0.57; 95% confidence interval, 0.43 to 0.75) that for bones without osteophytes.

Table 7—

Association between detection of a fracture and detection of osteophytes or vascular channels in proximal sesamoid bones of 328 Thoroughbred cadavers.

Radiographic findingFractureNo fracture
Osteophytes
 Detected99567
 Not detected152494
Vascular channels
 Detected160961
 Not detected91100

Vascular channels were seen in 160 of 251 (64%) fractured bones and 961 of 1,061 (91%) nonfractured bones. Fractures were seen in 160 of 1,121 (14%) bones with vascular channels and 91 of 191 (48%) bones without vascular channels. The likelihood of fracture in bones with vascular channels was approximately one fifth (odds ratio, 0.18; 95% confidence interval, 0.13 to 0.25) that for bones without vascular channels. Osteophytes and vascular channels were less likely to be detected in fractured proximal sesamoid bones than in nonfractured proximal sesamoid bones.

Discussion

Findings on palmarodorsal radiographs and fracture characteristics were categorized for proximal sesamoid bones from the forelimbs of Thoroughbred cadavers. Of the 328 equine cadavers, 136 (41.5%) had at least 1 fractured proximal sesamoid bone. Few differences in fracture characteristics were apparent between left and right forelimbs. Medial bones appeared preferentially affected with complete transverse or split transverse simple fractures. Fractured medial bones tended to have indistinct fracture margins; > 1 vascular channel > 1 mm in width; and osteophytes in abaxial wing, basilar middle, or basilar abaxial locations. Oblique, partial transverse, and axial fractures were detected predominantly in lateral bones. Fracture prevalence relative to horses actively training and racing increased with age up to 5 years. Osteophytes and vascular channels were common in fractured and nonfractured proximal sesamoid bones. In contrast to our hypothesis, odds for incurring a fracture in a proximal sesamoid bone were lower in bones with osteophytes and vascular channels than in bones without these features.

The sample population for the study reported here reflected all Thoroughbreds that died or were euthanized for any reason at a California racetrack, but it differed from the population of Thoroughbreds actively racing and training at CHRB-sanctioned racetracks during the study period. Compared with the population of Thoroughbreds that were actively racing or training, geldings were overrepresented and females underrepresented in the sample population. Differences in sex distribution may have reflected a greater likelihood for rehabilitation of injured females and sexually intact males for breeding potential, in contrast to the outcome for geldings. Two-year-old horses were underrepresented and older horses overrepresented in the sample population. The higher prevalence of horses > 2 years old in the sample population probably reflected a need for exposure, or repeated exposure, to factors (including fractures of the proximal sesamoid bones) that result in death of racehorses. Age prevalence could also have been attributable to efforts to salvage younger horses for future racing or sales in preference to older horses with less potential for future earnings.

Findings of the study reported here are most reflective of severe fractures of the proximal sesamoid bones. Only the more severe types of fractured proximal sesamoid bones warrant euthanasia of affected horses because of a grave prognosis for recovery. Midbody and basilar fractures were most common in the horses of our study. Midbody fractures were also the most common type of fracture in 19 Thoroughbred cadavers with fractures of the proximal sesamoid bones at New York racetracks.6 In contrast, other racehorses incur fractures of the proximal sesamoid bones more commonly in the apex than in the midbody or basilar areas.18,19 In the study reported here, avulsion fractures were limited to longitudinal axial and abaxial fractures. Inferences should not be made to more manageable types of fractures of the proximal sesamoid bones, such as isolated apical chip fractures or avulsion fractures of the suspensory branch. However, because complete midbody and basilar fractures are difficult to manage clinically, the findings of our study are most likely relevant to these types of fractures in equine patients. The relationships between detection of a fracture and other radiographic findings have relevance to the fractures observed in this sample population.

The high prevalence of fractures in the proximal sesamoid bones of the forelimbs in cadavers of racing Thoroughbreds has been documented.1–3,6 In the convenience sample of equine cadavers in the study reported here, 135 (41%) horses had a fracture of the proximal sesamoid bones, and 109 (80%) of these horses had biaxial (medial and lateral bones within a limb) fractures. This biaxial pattern of fractures in the proximal sesamoid bones was similar to the pattern of fractures in proximal sesamoid bones observed in Thoroughbreds with musculoskeletal injuries in Florida5 and New York.6 Female horses tended to have fewer midbody fractures than was evident for sexually intact males and geldings. Five-year-old horses had the highest incidence of fractures, as well as midbody, basilar, and axial fractures, for the racetrack population (horses that raced or had official timed workouts during the same time period).

The distribution of fractures did not differ between forelimbs and bones in the study reported here. This is in contrast to findings from other studies, although comminuted fractures were typically detected in the left forelimb. A similar preference in the left forelimb has also been observed in several other studies.2,3,20 In a study2 of horses in California, fractures of the proximal sesamoid bones that resulted in death or euthanasia of the affected horse were observed more commonly in the left (29 fractures) than right (9 fractures) forelimb for racing Thoroughbred horses; however, differences were not found between forelimbs for fractures incurred during training. Similarly, the proportion of racing Thoroughbreds in Kentucky with injuries to the proximal sesamoid bones was significantly higher for the left forelimb than the right forelimb.3 In a small number of racing Thoroughbreds with fractures of the proximal sesamoid bones in Japan, where horses run in a clockwise direction on the racecourse, the medial bone was fractured in 70% of left forelimbs and the lateral bone was fractured in 86% of right forelimbs.20 Limb and bone predilections for fractures of the proximal sesamoid bones may have diminished or been related to increased accuracy associated with the large sample size of our study population (328 horses). Thus, the data from our study do not support a cause for severe fractures of proximal sesamoid bones that is related to racetrack geometry or direction of racing.

Differences in limb and bone distributions were apparent for some specific fractures of the proximal sesamoid bones in our sample population of Thoroughbred cadavers. Complete transverse fractures in the midbody, complete and split transverse fractures of the basilar area, and simple fracture configurations were detected most commonly within the medial bone of both forelimbs. Oblique fractures of the midbody, partial transverse fractures of the basilar area, and axial fractures were seen most often in the lateral proximal sesamoid bones. The transverse orientation of fractures of medial bones in the study reported here could indicate that fractures in medial proximal sesamoid bones were attributable to longitudinally oriented tensile forces. The oblique orientation of fractures of the lateral proximal sesamoid bones could indicate that the direction of tensile load deviated from longitudinal or that an axially directed compressive component was superimposed on tension. Conceivably, fractures could occur first in the medial bone and result in valgus angulation of the digit centered at the metacarpophalangeal joint. Angulation of the limb could alter the direction of tension on the lateral proximal sesamoid bone and promote oblique propagation of the subsequent fracture of the lateral proximal sesamoid bone.

Axial fractures in the study reported here were detected most often in the lateral proximal sesamoid bones, concurrent with a lateral condylar fracture of the contralateral third metacarpal bone (20/30 horses). This is in agreement with several studies21–23 in which investigators found that axial fractures of the proximal sesamoid bones in Thoroughbreds are associated with a lateral condylar fracture of the corresponding third metacarpal bone. Displacement of lateral condylar fractures could result in abaxially directed forces on the proximal sesamoid bone through the palmar annular ligament and lateral collateral attachments between the condylar fragment and adjacent proximal sesamoid bone. Consequently, tension between the proximal sesamoid bone and its axial attachment (ie, the intersesamoidean ligament) could result in avulsion of the axial margin of the proximal sesamoid bone. Abaxial fractures were rare, and no limb or bone predilections could be inferred from the low numbers in our study.

Osteophytes were common (266 [81%] horses) and most prevalent on the base of the medial proximal sesamoid bone. The most common sites (middle and abaxial regions of the base) correspond to origins of the straight, oblique, and short distal ligaments of the proximal sesamoid bones,24 although periarticular osteophytes along the basilar articular margin would be difficult to differentiate from nonarticular enthesophytes on the palmarodorsal contact radiographic view used in the study reported here. Proximal osteophytes (apical, axial, and abaxial) were less prevalent than basilar osteophytes but were similarly most common on medial bones. Axial and abaxial fossa osteophytes were most prevalent in the left and right forelimbs, respectively, and abaxial fossa and wing osteophytes were most prevalent in medial bones.

Proximally, apical osteophytes are likely to have an articular component, abaxial fossa osteophytes are most likely enthesophytes within the insertions of the medial and lateral branches of the suspensory ligament, and abaxial wing osteophytes are most likely enthesophytes within the attachments of the intersesamoidean ligament and palmar annular ligament of the metacarpophalangeal joint. Abaxial wing osteophytes have been associated with the fibrous tissue collars that develop around the entrance of the nutrient arteries of the proximal sesamoid bones.9 Axial osteophytes may represent foci of bone production or mineralization of soft tissue in response to pathologic processes or thrombosis of the vessels that cross the intersesamoidean ligament.25 Intuitively, osteophytes were expected to increase with age of the horses; however, this relationship was not detected. Perhaps osteophyte formation is initiated early in the racing career of horses. Osteophyte size was not quantitated; thus, we could not determine whether osteophytes enlarged over time.

Vascular channels were detected in abaxial locations in 325 (99%) horses, but basilar channels were observed in only 10 (3%) horses. Abaxial vascular channels appear as linear lucencies near the abaxial surface and within the body of a proximal sesamoid bone that proceed toward the axial margin.12–15 Medial proximal sesamoid bones tended to have more abaxial vascular channels than were seen in the lateral proximal sesamoid bones. Vascular channels > 1 mm in width appeared more frequently on the medial bones. Irregular funnel-shaped lucencies in the abaxial wing palmar to the insertion of the suspensory ligament branches have been considered to be evidence of irregular enlargement of the sesamoidean vessels.10 However, microscopic evaluation9 has revealed that these cones are composed of nonmineralized fibrocartilage that most likely protects and supports the vessels entering the proximal sesamoid bones. Abnormal vascular channels have been defined as a channel with a width ≥ 2 mm with nonparallel radiographic margins on oblique views of the intact metacarpophalangeal joint of a live horse, including or excluding funnelshaped abaxial lucencies.12,13,15 It has been suggested26 that basilar vascular channels develop as a compensatory means of restoring compromised arterial supply to bones with sesamoiditis or sesamoidosis. In the study reported here, evaluation of vascular channels was conducted on the basis of criteria similar to those reported in another study13 in which width was measured on vascular channels within the body of the bone (not the abaxial wing); however, we used a channel width of 1 mm because magnification was expected to be less for contact radiographs than for clinical radiographs in live horses. Vascular channels were not assigned a rating of normal or abnormal; rather, observations were recorded to document the prevalence and distribution of these findings.

Features that are detected in similar locations may provide clues to the pathogenesis of fractures. Medial bones appeared preferentially affected with complete transverse or split transverse simple fractures and tended to have indistinct fracture margins, > 1 vascular channel > 1 mm in width, and osteophytes in abaxial wing and basilar middle or basilar abaxial locations. Indistinct margins may represent overlapping fracture fragments or possibly focal osteoporosis attributable to bone remodeling. In contrast, the orientation of oblique fractures and distinct fracture margins of lateral bones may reflect acute fracture secondary to overload after fracture of a medial proximal sesamoid bone.

Fractures were most likely to be found in proximal sesamoid bones without osteophytes or vascular channels. Although osteophytes and vascular channels were extremely common features in fractured and nonfractured bones, they were most frequently seen in nonfractured bones. Consequently, the odds of fracture of a proximal sesamoid bone were approximately 2 to 5 times higher in bones without evidence of these features. Thus, osteophytes or vascular channels may represent physiologic adaptation to biomechanical stresses associated with race training and racing. Some findings may be associated with growth and skeletal development. Large vascular channels have commonly been observed in young prospective racing horses.12,13,15 Alternatively, within the study reported here, some radiographic lesions (osteophytes and vascular channels) may have been obscured by fracture planes and fragments in fractured proximal sesamoid bones and thus not observed or associated with fractures. Small osteophytes may not have been observed on the single palmarodorsal radiographic projection. Furthermore, osteophytes were not subcategorized on the basis of size, which may have been related to likelihood or propensity for fracture.

Radiographic findings were determined from examination of 1 palmarodorsal radiographic projection. Additional radiographic views (eg, lateromedial and oblique) would have been useful for detecting osteophytes and enthesophytes obscured by overlying bone tissues and contours on the palmarodorsal projection. Furthermore, additional views would have enhanced the ability to differentiate osteophytes from enthesophytes. Oblique views may have enhanced our ability to detect vascular channels. Radiographic findings are also affected by radiographic technique, image contrast, and image brightness. For example, the likelihood of detecting osteophytes can be enhanced or diminished by changes in radiographic exposure, image contrast, or image brightness.

In the study reported here, radiographic positioning, exposure technique, image contrast, and image brightness were optimized for image quality and standardized for all bones. Thus, comparisons of palmarodorsal radiographic findings between fractured and nonfractured proximal sesamoid bones should be valid, although additional features may have been evident or further enhanced by use of alternative radiographic views. Radiographs taken for standard clinical prepurchase examinations can be compromised by horse movement, scatter of the x-ray beam, contours of overlying soft tissue, film quality, and exposure variation.11–13 The high-detail mammography film and contact radiographic technique used in the study reported here optimized enhancement of fine details and features, especially small vascular channels.

In the study reported here, biaxial fractures of the proximal sesamoid bones were commonly associated with death or euthanasia of racing Thoroughbreds in California. The patterns of palmarodorsal radiographic features appeared to differ between medial and lateral proximal sesamoid bones. Osteophytes and vascular channels were common findings; however, fractures were less likely to be found in bones with these features. Histologic and epidemiologic studies of fractured proximal sesamoid bones may elucidate tissuelevel changes and risk factors that will enhance our understanding of fracture etiopathogenesis and opportunities for prevention.

ABBREVIATIONS

CAHFSLS

California Animal Health and Food Safety Laboratory System

CHRB

California Horse Racing Board

a.

Jockey Club Information Systems, Lexington, Ky.

b.

Faxitron 805, Field Emission Corp, McMinnville, Ore.

c.

Kodak XOmat TL, high-speed mammography film, Eastman Kodak Co, Rochester, NY.

d.

Lumiscan 75 interface, Lumisys Inc, Sunnyvale, Calif.

e.

Adobe photoshop, Adobe Systems Inc, San Jose, Calif.

References

  • 1.

    Johnson BJ, Stover SM & Daft BM, et al. Causes of death in racehorses over a 2 year period. Equine Vet J 1994;26:327330.

  • 2.

    Estberg L, Stover SM & Gardner IA, et al. Fatal musculoskeletal injuries incurred during racing and training in Thoroughbreds. J Am Vet Med Assoc 1996;208:9296.

    • Search Google Scholar
    • Export Citation
  • 3.

    Peloso JG, Mundy GD, Cohen ND. Prevalence of, and factors associated with, musculoskeletal racing injuries of Thoroughbreds. J Am Vet Med Assoc 1994;204:620626.

    • Search Google Scholar
    • Export Citation
  • 4.

    Cohen ND, Peloso JG & Mundy GD, et al. Racing-related factors and results of prerace physical inspection and their association with musculoskeletal injuries incurred in Thoroughbreds during races. J Am Vet Med Assoc 1997;211:454463.

    • Search Google Scholar
    • Export Citation
  • 5.

    Hernandez J, Hawkins DL, Scollay MC. Race-start characteristics and risk of catastrophic musculoskeletal injury in Thoroughbred racehorses. J Am Vet Med Assoc 2001;218:8386.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6.

    Krook L, Maylin GA. Fractures in Thoroughbred race horses. Cornell Vet 1988;78 (suppl):1133.

  • 7.

    Stover SM, Johnson BJ & Daft BM, et al. An association between complete and incomplete stress fractures of the humerus in racehorses. Equine Vet J 1992;24:260263.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8.

    Riggs CM. Aetiopathogenesis of parasagittal fractures of the distal condyles of the third metacarpal and third metatarsal bones— review of the literature. Equine Vet J 1999;31:116120.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9.

    Poulos PW. Radiographic and histologic assessment of proximal sesamoid bone changes in young and working horses, in Proceedings. 35th Annu Conv Am Assoc Equine Pract 1989;34:347358.

    • Search Google Scholar
    • Export Citation
  • 10.

    O'Brien TR, Morgan JP & Wheat JD, et al. Sesamoiditis in the Thoroughbred: a radiographic study. J Am Vet Radiol Soc 1971;12:7587.

  • 11.

    Kane AJ, McIlwraith CW & Park RD, et al. Radiographic changes in Thoroughbred yearlings. Part 2: associations with racing performance. Equine Vet J 2003;35:366374.

    • Search Google Scholar
    • Export Citation
  • 12.

    Kane AJ, Park RD & McIlwraith CW, et al. Radiographic changes in Thoroughbred yearlings. Part 1: prevalence at the time of the yearling sales. Equine Vet J 2003;35:354365.

    • Search Google Scholar
    • Export Citation
  • 13.

    Spike-Pierce DL, Bramlage LR. Correlation of racing performance with radiographic changes in the proximal sesamoid bones of 487 Thoroughbred yearlings. Equine Vet J 2003;35:350353.

    • Search Google Scholar
    • Export Citation
  • 14.

    Trumble TN, Arnoczky SP & Stick JA, et al. Clinical relevance of the microvasculature of the equine proximal sesamoid bone. Am J Vet Res 1995;56:720724.

    • Search Google Scholar
    • Export Citation
  • 15.

    Hardy J, Marcoux M, Breton L. Clinical relevance of radiographic findings in proximal sesamoid bones of two-year-old Standardbreds in their first year of race training. J Am Vet Med Assoc 1991;198:20892094.

    • Search Google Scholar
    • Export Citation
  • 16.

    Grondahl AM, Gaustad G, Engeland A. Progression and association with lameness and racing performance of radiographic changes in the proximal sesamoid bones of young standardbred trotters. Equine Vet J 1994;26:152155.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17.

    Hill AE, Gardner IA & Carpenter TE, et al. Effects of injury to the suspensory apparatus, exercise, and horseshoe characteristics on the risk of lateral condylar fracture and suspensory apparatus failure in forelimbs of Thoroughbred racehorses. Am J Vet Res 2004;65:15081517.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18.

    Schneider RK. Incidence and location of fractures of the proximal sesamoids and proximal extremity of the first phalanx, in Proceedings. 26th Annu Conv Am Assoc Equine Pract 1980;25:157158.

    • Search Google Scholar
    • Export Citation
  • 19.

    Fretz PB, Barber SM & Bailey JV, et al. Management of proximal sesamoid bone fractures in the horse. J Am Vet Med Assoc 1984;185:282284.

  • 20.

    Watanabe T. Localized lesions of bone in the fetlock joint and their diagnosis by xeroradiography in racehorses. Centaur 1985;2:7984.

    • Search Google Scholar
    • Export Citation
  • 21.

    Bassage LH II, Richardson DW. Longitudinal fractures of the condyles of the third metacarpal and metatarsal bones in racehorses: 224 cases (1986–1995). J Am Vet Med Assoc 1998;212:17571764.

    • Search Google Scholar
    • Export Citation
  • 22.

    Greet TRC. Condylar fracture of the cannon bone with axial sesamoid fracture in three horses. Vet Rec 1987;120:223225.

  • 23.

    Barclay WP, Foerner JJ, Phillips TN. Axial sesamoid injuries associated with lateral condylar fractures in horses. J Am Vet Med Assoc 1985;186:278279.

    • Search Google Scholar
    • Export Citation
  • 24.

    Weaver JC, Stover SM, O'Brien TR. Radiographic anatomy of soft tissue attachments in the equine metacarpophalangeal and proximal phalangeal region. Equine Vet J 1992;24:310315.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 25.

    Wisner ER, O'Brien TR & Pool RR, et al. Osteomyelitis of the axial border of the proximal sesamoid bones in seven horses. Equine Vet J 1991;23:383389.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 26.

    Cornelissen BPM, Rijkenhuizen ABM, Barneveld A. The arterial shift features in the equine proximal sesamoid bone. Vet Q 1996;18:S110S116.

Contributor Notes

Dr. Johnson's present address is Oklahoma Animal Disease Diagnostic Laboratory, Center for Veterinary Health Sciences, Oklahoma State University, Stillwater, OK 74078.

Supported by the Grayson Jockey Club Research Foundation Incorporated and the Center for Equine Health with funds provided by Oak Tree Racing Association, the State of California pari-mutuel fund, and contributions by private donors.

Presented in part at the Symposium of the American College of Veterinary Surgeons, San Diego, October 2005.

The authors thank Dr. Lisa Heath for technical assistance.

Address correspondence to Dr. Stover.