Post-Mortem Study of Thoroughbred Fatalities in Victoria, Australia Between 2001 and 2004

Post-mortem examinations are an essential part of any study of racehorse fatalities. First, they ensure accurate assessment of musculoskeletal injuries warranting euthanasia. Second, they accurately identify the causes of sudden death, particularly cardiovascular or respiratory failure.

1. Introduction

Death or catastrophic injury of horses during training or racing is an important cause of economic loss within the Australian racing industry. Previous Australian studies of racecourse mortalities have relied on the unverified observations of racecourse veterinarians to categorize the injuries responsible for death or euthanasia. The present study is the first in Australia to employ post-mortem examinations to thoroughly document the range of injuries sustained and to accurately identify the causes of sudden death on racecourses.

2. Materials and Methods

A prospective study was undertaken over a 3-yr period from February 1, 2001 to October 31, 2004. Post-mortem examinations were performed on all Thoroughbred horses that died suddenly or were euthanized because of injuries sustained while racing or training on any of the four metropolitan racecourses in Victoria during this period. In addition, owners, trainers, and/or racetrack veterinarians elected to submit for post-mortem examination 16 horses that died while racing or training on country tracks during the study period.

For each horse, a fatality report form was completed at the time of death by the attending veterinarian and submitted to Racing Victoria Ltd. for inclusion in a national database (i-RIS). A copy of the fatality report accompanied each horse submitted for necropsy. The report included details on the identification of the horse, track conditions, race type and length, a description of the incident, the time of death, and a provisional diagnosis. If euthanasia was performed, details of the method of euthanasia were also provided. Horses were transported to the Veterinary Clinical Center of The University of Melbourne where necropsies were performed by one of four veterinary pathologists within 4 - 6 h of death using a standardized protocol. Samples of all viscera, including cardiac conduction tissue dissected in toto and brain, were preserved in 10% neutral buffered formalin. Tissues were embedded in paraffin, sectioned at 2 μm, and stained with hematoxylin and eosin for light microscopic examination.

Total numbers of horses that died suddenly or were euthanized on Victorian metropolitan and country racetracks were obtained from Racing Victoria Ltd. records. Proportional mortality rates were calculated by dividing the number of fatalities caused by sudden death by the total number of fatalities [1]. Proportions were compared with two-tailed p-values calculated using Fisher's Exact Test and WINPEPI [2]. WINPEPI was also used to calculate 95% CI (Wilson's method) and to compare each proportion with a hypothetical value of 0.5.

3. Results

Over 3 yr, 77 post-mortem examinations were performed, with 61 (79%) of the horses submitted from metropolitan racecourses and 16 horses submitted from country racecourses. Of the 77 fatalities, 40% (31 of 77) were sustained in flat races, 25% (19 of 77) in jumps races (10 hurdle and 9 steeple races), 3% (2 of 77) in barrier trials, and 32% (25 of 77) in track or non-raceday events. Of the 77 horses, 52 were euthanized (68%), and the remaining 25 horses died suddenly without veterinary assistance (Table 1). Females accounted for 21% (16 of 77; 13 - 31% CI) of the horses. Most of the affected males were geldings (56 of 61; 2%). The age of affected horses ranged from 2 to 10 yr with a mean age of 5.2 yr. The mean age of females was 4 yr (range = 2 - 8 yr), and the mean age of males was 5.5 yr (range = 2 - 10 yr). The mean age of horses dying in flat races was 4.8 yr, whereas those dying in jumps races had a mean age of 7.1 yr. In track events, the mean age was 4.4 yr, and in barrier trials, the mean age was 4 yr.

Catastrophic musculoskeletal injury was the cause of death or euthanasia in 57 of 77 horses (74%). Most of these horses (52 of 57; 91%) were euthanized. Of the 57 horses with musculoskeletal injuries, 45 (79%) sustained catastrophic injuries to the appendicular skeleton (limbs including pelvic bones) or associated soft tissues (Table 2), 9 (16%) sustained cranial or vertebral fractures, 1 (2%) sustained multiple rib and proximal forelimb fractures and a brain laceration during a steeplechase, 1 (2%) succumbed during a flat race to massive rupture of the paravertebral musculature resulting in sudden death, and 1 (2%) died suddenly after sustaining severe intramuscular hemorrhage during a flat race. Of the 45 limb injuries, 20 occurred in flat races, 11 in jumps races, 12 in track/training sessions, and 2 in hurdles barrier trials. Cranial or vertebral fractures accounted for a greater proportion of fatalities in jumps races (6 of 19; 32%) than in flat races (2 of 31; 6%; p = 0.04) and track/training sessions (1 of 25; 4%; p = 0.03). Of the 45 horses with appendicular skeletal or soft-tissue injuries, 40 (89%) sustained injuries to a single limb and 5 (11%) had concurrent injuries to two limbs. One horse sustained injuries to both the right forelimb and right hindlimb. Two horses sustained catastrophic injuries to both forelimbs, and two horses sustained injuries to both hindlimbs. In the 40 horses with catastrophic injury to a single limb, the most common injury involved the distal limb (30 of 40; 75%); and a forelimb (32 of 40; 80%) was more frequently affected than a hindlimb. In the same subgroup of horses, 23 animals sustained an injury to a left limb (23 of 40; 58%; 42 - 72% CI); and 17 horses (17 of 40; 43%; 29 - 58% CI) sustained an injury to a right limb. The proportion of horses with a single catastrophic left limb injury was not significantly different from the proportion of horses with a single catastrophic right limb injury (p = 0.43). The most common limb injury involved fracture of the metacarpus III (MC III) or metatarsus III (MT III; 12 of 45; 27%).

Twenty horses in the study (26%) died suddenly for reasons other than musculoskeletal injury. Seventeen horses died (6 during flat races and 11 during track/training sessions) with acute pulmonary edema, congestion, and/or hemorrhage. Two horses died (one in a flat race and one in a track session) exsanguinated into the peritoneal cavity, and one horse died (in a steeplechase) with no significant lesions detectable. In the 17 horses with acute lung lesions, 15 horses had severe pulmonary hemorrhage and 15 had moderate to severe pulmonary edema. Of the six horses with severe pulmonary edema, four also had severe pulmonary hemorrhage. In two horses, severe pulmonary edema and congestion were the predominant findings. Significant microscopic lesions were detectable in the hearts of only two horses with pulmonary lesions.

4. Discussion

Most fatalities of Thoroughbred racehorses in racing and training in Victoria between 2001 and 2004 were caused by euthanasia after a catastrophic injury to a forelimb. This is consistent with previous studies in other countries [3-9]. There is no difference in the proportion of left and right limbs affected, despite the unidirectional, counterclockwise mode of racing in Victoria. This contrasts with studies in the United Kingdom [7,9] that reported a pre-disposition of the right limb to injury and a study in the United States that found a pre-disposition of the left limb to injury during racing and of the right limb during training [3]. In racing, catastrophic injury to the MC III or MT III was the most common injury, followed by injuries to MC III/MT III accompanied by fractures of the proximal sesamoids and phalanx I. These findings are similar to a previous report from Victoria [10]. However, in North American studies, fractures of the proximal sesamoids were the most common injuries, followed by injuries to MC III and carpal bones [6,8]. In the United Kingdom, fractures of the MC III accompanied by fractures of the medial proximal sesamoid bone and phalanx I were found to be the most common racing injuries [5,7,9].

This study also suggests that sudden death is a more important contributor to racing fatalities than has been reported previously in Victoria [10] and in other countries. Sudden death accounted for 26% (46 of 180) of all fatalities on Victorian racetracks during the study period compared with 9% (58 of 659) of Thoroughbred racing fatalities reported in a Californian study [11] and 12% (15 of 127) of racing fatalities in a British study [9]. Future investigation of causes of sudden death may be facilitated by analysis of blood samples obtained immediately after death. Assays of markers of peracute myocardial injury (such as cardiac troponin I) may help to confirm the presence of myocardial degeneration or necrosis. Additionally, measurement of electrolytes (especially potassium, calcium, and magnesium) may identify potential triggers of such injury. Coagulation studies on point of death blood samples may also be beneficial in investigating cases of sudden death from massive hemorrhage.

This study was made possible by funding from Racing Victoria Ltd. , Rural Industries Research and Development Corporation, and the Equine Center of The University of Melbourne. The authors gratefully acknowledge Garry Anderson (The University of Melbourne) for statistical assistance.