The racehorses’ age and sex as well as racing career performance record immediately preceding the endoscopic examination on race day was extracted from race cards and included lifetime starts, lifetime wins, lifetime places (2nd, 3rd, or 4th) and lifetime stake earnings (South African Rand, ZAR). Further variables recorded on the day of endoscopic evaluation included horse name, microchip number, date of race, race venue, racing surface (turf or sand), trainer, jockey, locality (altitude or sea level), race reference number, number of starters, race time (day or night), race start time, race purse, time of endoscopic evaluation, elapsed time between race completion and endoscopic evaluation, weight carried, merit rating, draw number, starting bet, finishing position, horse length behind the winner, winner’s margin, finishing time of winner, finishing time of horse, race distance, track condition, falserail position and whether a horse was suspended due to epistaxis. Meteorological data were extracted from a commercial database (Pretoria Central Forecasting Office, South African Weather Service) and included penetrometer reading (an objective measure of track surface), ambient temperature, humidity, barometer, wind direction, wind speed, cumulative rainfall over the past 7 days, cumulative rainfall for the last 24 hours. Data was analyzed using Microsoft Excel (version 2003) and the statistical package NCSS (Hintze J. NCSS, PASS and GESS number cruncher statistical systems, Kaysville, Utah, 2004). Descriptive statistical analyses were initially carried out to summarize trends in the data. The relationship between EIPH and performance was evaluated by means of logistic regression. The presence of EIPH was defined as a dichotomous variable (yes vs. no) in 2 ways: severity grade 0 (no) vs. severity grade ≥ 1 (yes) and severity grade ≤ 1 (no) vs. severity grade ≥ 2 (yes).
Included in the regression model were variables that may have affected the racing performance or which may have acted as confounders. The variables were similar to those included by a previous study.11 Numeric independent variables included were race purse, elapsed time between race completion and endoscopic evaluation, weight carried, number of starters, race distance, penetrometer reading, age, lifetime starts. Dichotomized categorical independent variables included locality (sea level/altitude), sex (male/female) and EIPH (yes/no). The response variable for measuring performance was if a horse won (yes/no), or placed (1st, 2nd or 3rd) (yes/no) on the day of endoscopic examination. The Wald test was used to test the significance of the regression coefficient (β) and an estimated odds ratio (OR) with 95% confidence interval (CI) was calculated
for each associated regression coefficient. Multiple regression analysis was performed to assess the relationship between EIPH and the distance behind the winner. The same independent variables as used in the logistic regression were included in the model. However because the distance behind the winner was not normally distributed, logarithmic transformation was performed prior to analysis.
Least squares regression coefficient estimates were used to evaluate the relationship between the independent variables and distance behind the winner and a t-test was carried out to determine the significance of the coefficients. For all analyses, significance was set at P < 0.05. A total of 921 horses were included in the regression analyses as 84 horses had missing penetrometer readings and were excluded.
Similar to previous studies conducted abroad, this study found using tracheobronchoscopy a high overall prevalence of EIPH in Thoroughbred racehorses competing in South Africa.11,17,20,24 Moreover, a positive association was found between racing at sea level and the presence of EIPH severity grades ≥ 1 and ≥ 2. No relationship between the severity of EIPH and finishing in the first 3 positions nor distance behind the winner was found. In fact, a positive association between the presence of EIPH of severity grade ≥ 1 as well as ≥ 2 and higher odds of winning was identified. This study therefore concludes that in Thoroughbred racehorses competing in South Africa not medicated with furosemide nor using nasal dilator strips, the presence of EIPH is associated with superior performance and that racing at sea level is associated with an increased prevalence and severity of EIPH.
Historically, surveys of horse populations determining the prevalence and relationship with performance of EIPH have relied on the presence of epistaxis4,5,14,22,23,26,29 or tracheobronchoscopically detected blood.11,17,20,24 However, a diagnosis of EIPH based only on the presence or absence of epistaxis should be actively discouraged as its use as sole criteria for estimating the prevalence of EIPH is inaccurate. Epistaxis is an insensitive indicator of EIPH, occurring in only the most severely affected horse and may also be non-specific for pulmonary haemorrhage. As can be seen by the present study, although blood was detected in 54% of racehorses using tracheobronchoscopy, epistaxis was only present in 8/1005 (0.8%) of racehorses. Of these eight horses, 7 were affected by more severe grades of EIPH (6 horses had grade 4 EIPH and 1 horse had grade 3 EIPH), while one horse had no evidence of blood in the trachea despite having profuse epistaxis. Studies have used finishing first or in the first three positions as indicators of impaired racing performance and have found that epistaxis negatively impacts racing performance.17,26 However, these studies may have underestimated the true prevalence of EIPH by only reporting on the most severely affected horses, while those horses with less severe grades of EIPH were not included and so information regarding the relationship between less severe grades of EIPH and racing performance could not be made.17,26
In this study, we used tracheobronchoscopy to detect and quantify the presence and severity of EIPH. Therefore, in order to detect a horse affected with EIPH, blood needed to be present within the trachea and major bronchi at the time of tracheobronchoscopic evaluation. As has been eluded by another study, it is not certain whether horses that suffer minimal hemorrhage in the peripheral lung parenchyma may actually show blood within the airways.11 Moreover, as the movement of blood from the lung into the trachea may be time-dependant, and since this study endoscopically evaluated racehorses soon after racing, this study may have failed to identify such horses with the least severe grade of EIPH.
Chapter 1 Exercise-induced pulmonary haemorrhage: An introduction
1.4 CLINICAL SIGNS
1.6 THERAPEUTIC OPTIONS
1.7 RISK FACTORS
1.8 EFFECT ON PERFORMANCE
1.9 FIGURES AND TABLES
Chapter 2 Altitude affects the prevalence and severity of exercise-induced pulmonary heamorrhage in South African Thoroughbred racehorses
2.3 MATERIALS AND METHODS
2.7 FIGURES AND TABLES
Chapter 3 Pharyngeal, laryngeal and tracheal disorders in South African Thoroughbred racehorses: prevalence and relationship with performance
3.3 MATERIALS AND METHODS
3.7 FIGURES AND TABLES
Chapter 4 Reproducibility of endoscopic grading using tracheobronchoscopy in racehorses
4.3 MATERIALS AND METHODS
4.7 FIGURES AND TABLES
Chapter 5 Proinflammatory mRNA response in racehorses with exerciseinduced pulmonary haemorrhage
5.3 MATERIALS AND METHODS
5.7 FIGURES AND TABLES
Chapter 6 General discussion
6.1 EIPH: A SOUTH AFRICAN PERSPECTIVE
6.2 RESPIRATORY TRACT DISORDERS: PREVALENCE AND ASSOCIATION WITH RACING PERFORMANCE
6.3 DETECTION AND GRADING OF RESPIRATORY TRACT DISORDERS
6.4 SYSTEMIC INFLAMMATION AND THERAPUTIC INTERVENTIONS