Five year analysis of Jockey Club horse-related injuries presenting to a trauma centre in Hong Kong☆
Article Outline
- Summary
- Introduction
- Methods
- Results
- Discussion
- Conclusion
- Conflict of interest statement
- References
- Copyright
Summary
Background
Professional horse riding-related injuries have not been studied before in Hong Kong, although horse racing takes place very regularly in the territory. In addition, the equestrian events of the 2008 Beijing Olympic Games will come to Hong Kong. This study analysed the pattern of horse-related injury among patients who presented to a trauma centre in a teaching hospital in Hong Kong.
Methods
Information from the trauma centre database was analysed retrospectively. The database includes trauma patients who had sustained potentially severe injuries that warranted initial assessment and resuscitation in a trauma resuscitation room (triage category 1 or 2). Data analysed included demographic variables, causes and mechanisms of injury, anatomical injuries, anatomical and physiological trauma scores, and patient outcome.
Results
Between January 2001 and June 2005, 2312 trauma patients were entered into the database. Thirty-six (1.6%) patients had sustained horse-related injuries (mean age 34 years, range 17–54; male to female ratio 32:4), all whilst at work in the Jockey Club. Twenty-two patients were injured between midnight and 09:00
h. This group stayed in the resuscitation room for longer prior to admission compared with patients presenting between 09:00h and midnight (median time 127min (interquartile range [IQR] 57–183) versus 58min (IQR 43–83), p=0.06). Twenty-five patients fell from horseback, whilst 11 were kicked by the horse.
Twenty patients had a single injury and 16 patients had multiple injuries. Eighteen patients had injuries to the thorax, abdomen, thoracolumbar spine or pelvis. Eleven patients had head, face and cervical spine injuries and 11 had limb injuries. Twenty-five patients were admitted, including four admitted to the intensive care unit (ICU). Ten patients required surgery. Median (standard deviation [S.D.]) probability of survival was 0.996 (0.052) and median revised trauma score (RTS) (S.D.) was 7.841 (0.624). There were no fatalities. Potentially serious horse-related injuries presented once every 6 weeks.
Conclusion
Most injuries are minor and affect the trunk but occur out of hours. Helmets, face shields and body protectors should be worn when riding or handling horses.
Keywords: Equestrian, Trauma, Hong Kong, Outcome, Prevention
Introduction
Hong Kong has a long history of horse racing dating from 1840. Hong Kong's first official race meeting was held at Happy Valley in December 1846.9 At that time, horse racing was considered a social event for the elite, but it became more popular among the general public in the 1960s. In response, a second racecourse for Hong Kong was constructed in Shatin in 1978. The equestrian events of the 2008 Olympic Games will be held in Shatin and other locations in Hong Kong. Due to its close proximity (5km) to the Prince of Wales Hospital (PWH), all trauma victims from the Shatin Racecourse are referred there for management.
Equestrians have been estimated to be at higher risk of serious injury than participants of many other sports; the incidence of serious injury per hour of riding is greater than during motorcycle and car racing.4, 7 Injuries from horse riding have been well documented in the literature, mainly from western countries.2, 3, 6, 7, 8, 10, 12, 13, 14 Data on recreational horse-related injuries in Hong Kong have been recently published.13
The aim of this study was to analyse the pattern of equestrian injuries in a Hong Kong trauma centre, and to propose measures on injury prevention and optimisation of trauma care.
Methods
PWH is an acute general hospital with 1200 beds and is the primary teaching hospital of the Chinese University of Hong Kong. Since March 2002, the PWH emergency department (ED) trauma centre has functioned as the regional level 1 trauma centre for the north east New Territories. It is equipped with two fully equipped resuscitation rooms dedicated for trauma care, with 24h computed tomography (CT) scanning available in an adjoining room. Around 520 trauma patients per annum are triaged to the trauma rooms according to high risk mechanism, of whom around 160 have an injury severity score (ISS) >15. The ED has an annual attendance of 180,000 patients per annum. A trauma team system is in effect for all severely injured patients according to predefined criteria, led by a specialist in emergency medicine 24h a day. All relevant surgical, intensive care and paediatric facilities are on site.
The trauma database of PWH was established in 2001. Information is collected on trauma patients who sustained injuries that warranted resuscitation or close monitoring in a trauma resuscitation room. Data from the trauma database was analysed retrospectively. Data analysed included demographic variables; mechanism of injury; anatomical and physiological trauma scores (including abbreviated injury scores for each body region (AIS), injury severity score and revised trauma score); calculated probabilities of survival (p's) and patient outcome. Patients were defined as professional equestrians if they were working or being employed to work on horses at the time of injury, which included jockeys, horse trainers, grooms, or work riders. Injured body parts were grouped into head and neck, face, trunk (including chest, abdomen and pelvis); and extremities.
Statistical analyses
Descriptive statistics were used to summarise the data. The difference between median times was analysed using the Mann–Whitney U-test. Statistical analysis was done using SPSS version 13.
Ethics
Institutional review board ethical approval requirement was waived for this retrospective observational study.
Results
Demographic data
Between January 2001 and June 2005, 2312 trauma patients were included in the trauma database of whom 36 (1.6%) had sustained horse-related injuries. The mean age was 34 years (age range 17–54) and the male to female ratio was 8:1. Review of case notes and ED records confirmed that all patients sustained their injuries while at work in the Jockey Club, i.e. they were all professional equestrians.
Among the 36 patients, 8 were jockeys, 5 were horse trainers, 9 were work riders, and 2 were grooms. Twelve others were all employed at the Jockey Club but their precise occupation was not recorded. Median length of stay was 1.5 days in hospital (range 0.1–280 days).
Table 1 has further details of injuries and mechanisms.
Table 1. Demographic and injury severity score (ISS), n=36
| ISS<5 | ISS 5–10 | ISS 11–14 | ISS>15 | Total | |
|---|---|---|---|---|---|
| Triage category 1 | 3 | 0 | 1 | 3 | 7 |
| Triage category 2 | 15 | 3 | 3 | 4 | 25 |
| Triage category 3 | 2 | 0 | 1 | 0 | 3 |
| Triage category 4 | 1 | 0 | 0 | 0 | 1 |
| Fell from horse | 15 | 2 | 4 | 4 | 25 |
| Kick by horse | 6 | 1 | 1 | 3 | 11 |
| Admit to specialty | 12 | 2 | 4 | 7 | 25 |
| Discharge from ED | 8 | 0 | 0 | 0 | 8 |
| Admit private hospital | 1 | 1 | 1 | 0 | 3 |
| ICU admission | 0 | 0 | 0 | 4 | 4 |
| Operation | 2 | 1 | 1 | 6 | 10 |
Pattern of injury
Twenty-two patients presented to the trauma centre between 00:00h and 09:00h (the period of the day when ED and hospital staffing is at a minimum), while 14 patients presented between 09:00h and 23:59h. Three patients in each of these groups required activation of the hospital trauma team. Trauma team activation criteria include multi-system blunt or penetrating trauma with unstable vital signs (hypotension; signs of shock; respiratory compromise; Glasgow Coma Score [GCS] ≤13), or significant anatomical injury (penetrating injury to head, neck, torso or groin; blunt or crush injury to chest or abdomen; flail chest; spinal injury with paralysis; two or more proximal long bone fractures; open or depressed skull fracture or suspected (unstable) pelvic fracture).
The group who arrived at the ED overnight stayed in the trauma resuscitation room for a longer period of time before being admitted (median time 127min (interquartile range [IQR] 57–183) versus 58min (IQR 43–83), p=0.06, Mann–Whitney U-test). There was no significant difference in median ISS between those arriving between 00:00h and 09:00h and those arriving after 09:00h (median 3 versus 5, p=0.4, Mann–Whitney U-test).
Twenty patients had an isolated injury (6 patients had been kicked [K], 14 patients had fallen [F]) and 16 (5 K, 11 F) had multiple injuries. Truncal injuries were the most common. Eighteen patients (6 K, 12 F) had injuries to the thorax, abdomen or pelvis. Ten patients (4 K, 6 F) had isolated contusions to the chest wall. Five patients (2 K, 3 F) had multiple rib fractures, with three pneumothoraces (1 K, 2 F).
Three patients had abdominal injuries. Two of them (1 K, 1 F) had lacerations of the liver which were managed conservatively, and one patient (K) had a ruptured spleen which required splenectomy. One patient sustained a pelvic fracture (K) and two patients (2 F) had lumbar spine fractures (no neurological sequelae in either case).
Eleven patients (4 K, 7 F) suffered from head and facial injuries, eight (4 K, 4 F) of which were minor head injuries. Two patients (2 F) had facial bone fractures. Three patients (2 K, 1 F) had a cervical spine injury, two of whom (1 K, 1 F) were tetraplegic.
Eleven patients (1 K, 10 F) had injuries to the extremities. Upper limb injuries were more common then lower limbs injuries (9 versus 2). Three patients had minor soft tissue injuries and three had fractured clavicles. There were three shoulder dislocations and two patients had fractures of tibia and fibula. Table 2 gives operative details.
Table 2. Operations performed
| Mode | Nature of injury | Operation performed | |
|---|---|---|---|
| 1 | Fall | Closed depressed fracture of posterior fossa with subarachnoid haemorrhage | Burr hole and elevation of fracture |
| 2 | Kick | Fracture dislocation C6 | Anterior spinal fusion |
| 3 | Fall | Burst fracture C5 | Anterior spinal fusion |
| 4 | Fall | Burst fracture L5 | Posterior spinal fusion |
| 5 | Fall | Ruptured spleen | Splenectomy |
| 6 | Kick | Closed fracture of pelvis (diastasis of pubic symphysis) | Open reduction and internal fixation of pelvis |
| 7 | Fall | Closed fracture of right tibia and fibula | Open reduction & internal fixation |
| 8 | Fall | Open fracture of left ankle, bimalleolar | Open reduction and plating |
| 9 | Fall | Recurrent dislocation of right shoulder | Arthroscopy of shoulder and repair of Bankart lesion |
| 10 | Fall | Closed fracture of nasal bone | Closed reduction |
Injury severity
Twenty-five patients were admitted to PWH while three patients were admitted to a private hospital. Ten patients required surgery (Table 2). Twenty-nine patients had an ISS<15. Among the seven patients with ISS>15, six required operative treatment and four were admitted to the intensive care unit with an average ICU length of stay of 8.8 days. Abbreviated injury score data are shown in Table 3. Median probability of survival and corresponding standard deviation (S.D.) was 0.996 (0.052). Median revised trauma score (S.D.) was 7.841 (0.624). There were no fatalities. Potentially serious horse-related injuries presented once every 6 weeks.
Table 3. Abbreviated injury severity score (AIS) in 36 patients
| Body region | AIS 0 | AIS 1 | AIS 2 | AIS 3 | AIS 4 | AIS 5 |
|---|---|---|---|---|---|---|
| Head and neck | 23 | 7 | 2 | 1 | 0 | 3 |
| Face | 33 | 1 | 2 | 0 | 0 | 0 |
| Chest | 26 | 5 | 0 | 3 | 1 | 1 |
| Abdomen | 31 | 0 | 1 | 2 | 2 | 0 |
| Extremities | 25 | 3 | 7 | 1 | 0 | 0 |
| External | 22 | 14 | 0 | 0 | 0 | 0 |
Discussion
Equestrian activities, both professional and recreational, are common throughout the world. The injury rate is understandably high due to the weight of the horses and the speed they can achieve. The pattern of equestrian injuries had been well studied in UK and Australia.1, 5, 10, 11, 12 However, horse riding is not a common sports activity in Hong Kong, borne out by our finding that all the horse-related injuries we encountered in our hospital were occupational.
All of the patients in this series work in the Jockey Club. The age and sex distribution is as expected as professional equestrians are predominantly young and male. This is in contrast to the findings of previously published literature, where female patients outnumbered males among amateur riders.10, 12, 14
The finding of the timing of injury (61% between 00:00h and 09:00h) is peculiar to professional equestrian injuries, as injuries occur more commonly around the time of the morning trot. Unfortunately, it is also the time when overall hospital medical and nursing manpower is at its lowest point in the day and there may be a lack of senior cover in the hospital.
There was a trend to a longer stay in the emergency department prior to admission for this group of patients (median time 127min versus 58min, p=0.06, Mann–Whitney U-test). This was despite the fact that the injuries sustained were on average, less severe in the early morning presentations.
Multiple factors could contribute to this delay. It can take longer to assess and resuscitate a patient during the early morning hours due to staff fatigue. Outwith office hours, patients are usually first assessed by junior medical officers as part of the trauma team, although the trauma team is directed by a specialist emergency physician 24h a day. The first call medical officer may not be experienced enough to make decisions on management plans and may need to wait for senior advice for major management decisions, such as the decision to go to the operating theatre. Delays in this decision making could have a clinically significant impact, for example, the decision to take a patient to the operating theatre for splenectomy after splenic rupture.
Previous studies on horse-related injuries had shown a reduction in severity and incidence of head injuries over the past 20 years. Barber in 1973 noted that 66% of admissions were attributable to head injury,1 while in 1996, Chitnavis noted the figure was reduced to 26%.5 The main reason for the reduction in head injuries is attributed to the improved protective helmet design. In our study, most head injuries were minor ones. Only one patient required operative neurosurgery for head injury. This may be attributed by the mandatory usage of standardised helmets during horse riding even though there was no clear documentation of the patient's helmet status during the accident.
Although a helmet can protect the cranium, it does not provide protection to the face or the cervical spine. In our study, two patients had serious facial injuries that warranted operative treatment. An Australian study 10 demonstrated a 12–35% rate of concomitant facial injuries among those patients who sustained head injuries. This suggests that helmet use does not protect against facial injuries. Helmets with full face shields may need to be considered in the future. The cervical spine remains a vulnerable body region; cervical spine injuries often lead to devastating consequences. However, primary prevention of cervical spine injury in equestrians is difficult to achieve. Healthcare providers must take all necessary precautions against secondary injury when dealing with potential cervical spine injuries.
Truncal injuries were the most common injuries encountered in our study population. This finding differs from those published previously in which injuries to the extremities were the most common among professional and amateur riders.10, 12, 13, 14 One possible explanation is the difference in the populations studied. In our study, we reviewed only patients who were managed in the trauma resuscitation room. These were potentially more seriously injured patients, possibly leading to a higher proportion of truncal injuries. Patients with very minor injuries were not included in our study. Many work related minor injuries are managed in the racecourse staff clinic, and some are referred to hospital and managed as non resuscitation room cases.
In this retrospective study, there was no consistent clear documentation of the protective devices status at the time of injury. It is assumed that jockeys and work riders wear full set of protective devices, i.e. helmet, body protector and boots, at the time of riding. The same may not hold true for grooms and horse trainers. In fact, a recently published study in Hong Kong shows that only around 70% of riders wear a body protector, while none of the grooms wear a body protector while working.13 The lack of body protection may account for the higher proportion of truncal injury in our study.
Previous studies suggest that admission rates vary between 15% and 27%.11, 13 The admission rate in our series was much higher because the trauma data base includes only resuscitation room patients, who were more seriously injured.
Approximately 75 jockeys and 66 horse trainers are employed by the Hong Kong Jockey Club,9 but the number of grooms and work riders are not known. As all major trauma cases are brought to PWH, it can be estimated that the incidence of major trauma among professional equestrians is very low. Assuming around 40 races take place per year at Shatin Racecourse, with 9 races per event, and 12 jockeys per race, there is an annual incidence of potentially serious trauma for professional jockeys of 0.042%.
Limitations
This is a retrospective study based on the data from a prospectively collected trauma database and case records. Information such as the patient's occupation, circumstances at the time of injury, etc., may not be complete. Helmet status was not consistently documented in the case record. The injury patterns described may not represent the complete profile of equestrian injuries in Hong Kong. The incidence calculations depend to some extent on the estimates made for the assumptions, although we believe them to be accurate at the time of writing. We acknowledge that the small numbers in the present study may limit the usefulness of the results to health care systems in other parts of the world.
Conclusion
Major trauma among professional riders is uncommon, with the trunk most commonly injured. Apart from a helmet, the use of full length body protector should be advocated when handling horses. Standardized helmets protect against head injuries but not facial injuries, helmets with full face shield should be considered for future use. Prompt decision making is crucial in the management of trauma patients, and we advocate specialist led trauma team provision on a 24h basis to optimise patient care.
Conflict of interest statement
Colin Graham is a member of Injury's Editorial Board. The other five authors have no conflict of interest to declare.
The Trauma and Emergency Centre at Prince of Wales Hospital is known as ‘The Jockey Club Trauma and Emergency Centre’ as the Hong Kong Jockey Club provided initial funding for the development of the trauma centre in 1997.
The Hong Kong Jockey Club has no input into the day-to-day running or management of the Trauma Centre, nor does the Trauma Centre receive any continuing money from the Hong Kong Jockey Club. The Hong Kong Jockey Club had no input into the design, execution or analysis of this study at any stage.
References
- . Horse play: survey of accidents with horses. BMJ. 1973;3:532–534
- . Accidents in equestrian sports. Am Fam Physician. 1987;36:209–214
- . Common injuries in horseback riding. A review. Sports Med. 1990;9:36–47
- Centre for Disease Control. Injuries associated with horseback riding: United States 1987–1988. MMWR Morb Mortal Wkly Rep 1990;39:329–32.
- Accidents with horses; what has changed in 20 years. Injury. 1996;27:103–105
- . Improving accident analysis and prevention by introducing an accident and emergency based relational database. Emerg Med J. 2002;19:573–575
- . Equestrian injuries. In: Schneider RC, Kennedy JC, Plant ML editor. Sports injuries: mechanism, prevention, and treatment. Baltimore, MD: Williams & Wilkins; 1985;p. 431–439
- . Equestrian injuries: a five-year review. J Emerg Med. 1994;12:143–145
- Jockey Club of Hong Kong. History [Hong Kong Jockey Club website]. Available at: http://www.hkjc.com/english/about_hkjc/history2.htm#1846. Accessed 19 October 2005.
- . Pattern of equestrian injuries presenting to a Sydney teaching hospital. ANZ J Surg. 2000;73:567–571
- . Riding and other equestrian injuries: considerable severity. Br J Sports Med. 1987;21:22–24
- . A changing pattern of injuries to horse riders. Emerg Med J. 2002;19:412–414
- . Horse related injuries: a local scene. Hong Kong J Emerg Med. 2004;11:133–141
- . Equestrian injuries: a five year review of hospital admissions in British Columbia, Canada. Inj Prev. 2000;6:59–61
☆ This paper was presented in poster format at the Scientific Symposium of the Hong Kong College of Emergency Medicine in Hong Kong on 28–29 October 2005.
PII: S0020-1383(06)00496-7
doi:10.1016/j.injury.2006.08.026
© 2006 Elsevier Ltd. All rights reserved.
