Reasons for performing study: Strongyle worm burdens are assumed to subclinically affect equine performance. This assumption appears to be particularly pronounced in the equine racing industry.
Hypothesis: Race results of Standardbred trotters are negatively affected by high strongyle faecal egg count levels.
Methods: Faecal samples were obtained from 213 racing Standardbred trotters, aged ≥2 years, and stabled at training facilities of 21 professional trainers with license at racecourses in Denmark. Strongyle egg counts were generated using a McMaster technique. Race results were recorded as the finishing position of the horse (position 1–3 vs. finishing lower) and winning purse. The effect of strongyle egg counts on performance was assessed using regression analyses.
Results: Strongyle egg counts ranged from 0–3500 with a mean of 319 and a median of 150 eggs/g. Finishing in positions 1–3 was significantly associated with higher egg counts.
Conclusions: Race performance of the population of professionally trained Danish Standardbred trotters was not negatively affected by higher strongyle faecal egg count levels.
Potential relevance: The traditional frequent anthelmintic treatments of racehorses may be inordinate.
Strongyle infections are ubiquitous in equine populations worldwide. Although horses can harbour hundreds of thousands of cyathostomins without clinical signs of parasitic disease, there are defined disease complexes caused by helminth infections. Of most concern is the larval cyathostominosis caused by massive synchronous re-emergence of encysted larvae from the walls of the large intestine. This condition is reported to have a case-fatality rate of about 50% (Love et al. 1999). In addition, Parascaris equorum infections may lead to small intestinal impactions in foals and yearlings (Cribb et al. 2006), and Anoplocephala perfoliata infection can cause spasmodic colic and ileal impaction (Proudman and Edwards 1993; Proudman et al. 1998). Large strongyle parasites have become very rare in managed horses but Strongylus vulgaris is notorious for causing thromboembolic colic, which can be fatal in horses (Enigk 1951; Duncan and Pirie 1975). However, given the general ubiquity of gastrointestinal helminths, clinical disease remains a relatively rare condition.
Subclinical parasitic infections are often assumed, by horse owners as well as veterinarians, to affect horses in various ways. Retarded growth and ill-thrift can be observed in young horses with heavy parasite burdens, but no observational studies have systematically investigated this in the field setting. Similarly, it appears to be widely assumed that parasites can cause poor performance in competition horses, but this is not supported by any published study. In the equine racing industry in particular, there is a tradition for frequent prophylactic anthelmintic treatments as a part of a general health care plan (Anon 1998; Earle et al. 2002; Comer et al. 2006), presumably based on the belief that parasites affect race performance.
The aim of the present study was to evaluate whether strongyle faecal egg counts had a negative influence on race performance of Danish Standardbred trotters.
Materials and methods
The study was carried out between 27th July and 12th October 2009. Twenty-one licensed trotting trainers from various regions of Denmark collaborated for the study. Each trainer was represented with 6–14 horses in the final data set. The horses had been in race training for at least 3 months, were raced at 6 different race tracks and all had at least one valid start at an official totalisator race during the study period.
Faecal samples were collected from a total of 213 Standardbred trotters. The age of the horses ranged from 2–12 years (mean = 4.8, median = 4) and 52 (24%) were stallions, 79 (37%) geldings and 82 (38%) mares.
The participating stables all treated all horses with anthelmintics 3–4 times a year, but the horses enrolled in the study had not received anthelmintic treatment for 12 weeks prior to faecal sampling. The latest anthelmintic treatment had been given 3–11 months prior to the study, with a majority of trainers using ivermectin, 3 using moxidectin and one trainer using pyrantel embonate.
According to the homepage of Danish Trotting Association, a total of 590 horses were registered for Standardbred racing in Denmark during the study period. Since not all registered horses were necessarily raced, it can be concluded that our sample covered 36% of all Standardbred trotters in racing in Denmark, but probably represented a larger share of those raced.
Fresh faecal samples were collected from the stall floors according to published recommendations (Nielsen et al. 2010a). Samples were kept in airtight plastic bags and refrigerated at 5°C until analysed the following day.
Faecal egg counts (FEC) were performed using a McMaster technique with a detection limit of 50 eggs/g (Roepstorff and Nansen 1998). A modified McMaster method (Proudman and Edwards 1992) was used to detect presence of anoplocephalid eggs in the samples. Individual larval cultures were performed from all horses shedding 200 eggs/g or more. Ten grams of faeces were moistened with tap water, mixed with 4 g of vermiculite and set up for culture in a humidity chamber at room temperature. After 14 days, cultures were sedimented with the Baermann method for 24 h, and the whole sediment was examined under the microscope at 100 and 400× magnification. Third stage larvae of Strongylus vulgaris were identified according to the criteria described by Russell (1948) and horses were recorded as being either S. vulgaris positive or negative.
Race results, including finishing position and winning purse for each horse, were obtained from the official homepage of Danish Trotting Association. Range of the winning purse was 0–245,000 DKR (0–33,000 Euros) with a mean of 5298 and a median of 750 DKR.
Data were organised, edited and analysed using the SAS software, version 9.2.
The effect of strongyle egg counts on each horse's race performance was assessed using the finishing position as well as the winning purse. The effect was evaluated in 2 ways. In the first analysis the horses were categorised into 4 groups based on the egg count (FEC = 0; 0<FEC≤200; 200<FEC≤1000; and FEC>1000) while in the second analysis the FEC was kept as a continuous variable.
The effect of strongyle egg counts on finishing position (finishing 1–3 or 1–5, respectively) was analysed using logistic regression (the procedure NLMIXED in SAS). The effect of strongyle egg counts on the winning purse of the horse was analysed using linear regression (the procedure MIXED in SAS). For the latter, the horses were divided into 2 groups based on the winning purse being above or below 5000 DKR (about 700 Euros).
The analysis of the effect of strongyle egg counts on finishing position and winning purse was controlled for the effect of gender (stallion, mare, gelding), age (2, 3, 4, 5 and >5 years) and trainer (random effect). The statistical significances of the effects were assessed using likelihood ratio test and a significance level of 5%.
Strongyle FECs ranged from 0–3500 with a mean of 319 and a median of 150 eggs/g. Eighteen (8.5%) of horses had Anoplocehala perfoliata eggs and 17 (8.0%) were positive for Strongylus vulgaris on the larval cultures.
Statistically significant associations were found between finishing position in races and FEC. Results from the analyses using FEC as a continuous variable are presented in Table 1. When categorising FECs into 4 groups and using horses with FEC = 0 as reference, the odds ratios (OR) for horses finishing in position 1–3 were as follows for the 3 other groups; 0<FEC<200: 1.1, 200<FEC<1000: 0.9 and FEC>1000: 5.6. This all indicates that horses finishing in position 1–3 had higher egg counts. There were no significant effects of gender, age or trainer (Table 1). Figure 1 shows FECs of horses finishing first to third vs. horses finishing fourth or below. The analysis comparing horses finishing in position 1–5 with the remainder of the horses yielded similar results (not shown).
Table 1. Estimated odds ratios (OR) and 95% confidence intervals obtained in logistic regression for factors influencing the probability of 213 Danish Standardbred trotters obtaining finishing position 1–3 in races
OR (95% Wald confidence interval)
The effect of the strongyle faecal egg count is given per 100 eggs/g increase. *Variable significant at 0.05 level.
Figure 2 illustrates egg counts of horses winning above and below 5000 DKR. The linear regression suggested an increase of 450 DKR per 100 eggs/g, but the 95% confidence interval included 0 (-74–1024 DKR), and the association was not significant. Gender had a significant effect on winning purse (stallions had a significantly higher winning purse), whereas age and trainer did not.
The present study is the first published attempt to relate faecal egg count levels to equine performance. A large proportion of Danish Standardbred trotters were included in the study, and their race results were not negatively affected by strongyle faecal egg count levels. This raises the question of whether the intensive anthelmintic treatment regimens often practised in the racing industry are justified.
A prerequisite for discussing the outcomes of this study is a good understanding of the interpretation of the strongyle FEC. In a recently published study, no direct linear associations were found between strongyle faecal egg counts and total luminal strongyle worm burdens in horses (Nielsen et al. 2010b). However, when egg counts were categorised into defined FEC levels, horses with egg counts below thresholds in the range of 100–500 eggs/g had significantly smaller worm burdens than horses above these thresholds (Nielsen et al. 2010b). This suggests that the level of FEC gives us some broad indication of the parasite burdens, but above 500 eggs/g no quantitative information can be concluded from the egg count number. In the present study, we also divided egg counts into relevant categories that could be expected to correspond to differences in worm burden. In addition, several studies have shown that horses tend to maintain the same level of FEC over time even in the absence of anthelmintic treatment (Gomez and Georgi 1991; Döpfer et al. 2004; Nielsen et al. 2006; Becher et al. 2010). This suggests that some horses are capable of maintaining lower worm burdens than others. Given that important confounding variables, such as age and gender, were accounted for in our analyses, it appears reasonably justified to use FECs as a broad indicator of strongyle worm burden, and it is therefore interesting that no associations with race results were observed in the present study. However, it should be borne in mind that the study population had undergone a preselection for apparently healthy individuals. Horses showing clinical signs of any kind of disease would probably not be raced, so it can be argued that horses affected by their parasite burdens never became part of the data set. On the other hand, our specific aim was to study the possible subclinical influences of strongyle parasites, which justifies a possible exclusion of clinically affected animals.
It has been shown that strongyle FECs tend to decline with age, while luminal worm burdens remain relatively constant (Chapman et al. 2003). This suggests that FECs are partially influenced by acquired or age-dependent immunity. Theoretically, higher stress levels related to training and racing, and possibly affecting the immune status, could be speculated to explain the observed association between higher FECs and better race results. The analysis using the 4 FEC categories indicated that horses with a high FEC (>1000 eggs/g) had a tendency towards a better finishing position, whereas there were no significant differences between horses having lower FEC levels. Although this finding was statistically significant, it should be borne in mind that only 15 horses had egg counts >1000 eggs/g and this result should therefore be interpreted with caution.
While it can be concluded that the strongyle egg count yields some information about parasite burden, another relevant discussion is how to best measure equine performance. Racing results are influenced by many factors besides FEC and it is difficult in a study design to account for them all. In the present study we accounted for the influence of the trainer (training and management) by including trainer as random effect. However, there are horse specific factors not accounted for, including other physical conditions affecting performance such as acquired, undiscovered injuries, and possibly even tactical decisions made by the driver during the races. We chose to create 2 different outcome variables describing the race results to add strength to our measurements. The difficulty of a given race was accounted for by using winning purse as outcome variable, because races with higher winning prizes are more prestigious and attract the better horses. For a more accurate measure of equine performance in future studies, horses could be evaluated on a treadmill, which is often used in exercise physiology studies.
In conclusion, the present study suggested that Standardbred trotters in Denmark were not affected negatively by strongyle FEC levels during racing. Further studies are needed to elucidate the role of strongyle worm burdens in performance horses to identify suitable anthelmintic treatment regimens for these high-performance athletes.
Conflicts of interest
No conflicts of interest have been declared.
Source of funding
The study was financially supported by The Anniversary Foundation of Kongeriget Danmark Horse Insurance, The Hans Nielsen Foundation and Merial Norden A/S.
The authors are very grateful to veterinary practitioners Karsten Holm and Sune Hansen as well as veterinary student Trine Månsson for assisting with sample collection. Finally, colleagues at Ansager Dyrehospital are acknowledged for great patience and help with the laboratory analyses. Dr Pikka Jokelainen, University of Helsinki, is warmly acknowledged for critically reviewing the manuscript.