Tuesday, December 1, 2015

Thyroid Hormone Supplementation: Performance enabling or enhancing?

Supplementation with thyroxine (thyroid hormone) is commonplace among racehorse trainers, and recently this practice has been accompanied by a swirl of confusion.  Like so many other typical racetrack practices, the supplementation with thyroxine has been demonized in some corners, from being used to “cover up” cobalt abuse to being responsible for sudden deaths on the track.  Like so many other typical racetrack practices, thyroxine supplementation is defended by its proponents as making the horses “feel better and do better,” with no real explanation of how it might work. 

So, let’s start this discussion by reviewing what thyroxine actually does.  Thyroxine is normally produced in the thyroid gland, a paired gland on either side of the neck, just behind the head.  Thyroxine affects every cell in an animal’s body, affecting everything from normal growth to normal muscle development in response to exercise.  You can consider that thyroxine is the hormone that permits everything else in the body to function properly.  The basic metabolic rate, including heart rate and temperature are determined by thyroxine.  In fact, veterinarians recommend that horses which suffer from metabolic syndrome, a disease not dissimilar to type 2 diabetes in humans, be supplemented with thyroxine to increase their metabolism for the purpose of weight loss, which requires a dose 3 – 6 times the typical amount supplemented to the average racehorse.  Remarkably, these horses don’t seem to have a higher than average risk of sudden death and unquestionably have had no out of the ordinary exposure to cobalt, despite this large dose of thyroxine.  Most have minimal, transient signs or no signs at all of thyrotoxicosis (dangerously high levels of thyroxine in the blood).
What are the signs of thyrotoxicosis in horses?  There are no published reports…actually there are only a few published reports about thyroid hormone in horses at all.  This leaves us with a large hole in the scientific literature to assess whether our horses need extra thyroxine at all.  Thyrotoxicosis in other species is associated with increased heart rates, increased body temperatures, weight loss and diarrhea.  In fact, most veterinarians will tell you that nervousness and diarrhea accompany excessive thyroxine use, and can be seen occasionally in the first few days after starting the very high dose used in metabolic disease (Type 2 Diabetes-like) horses with high doses of thyroxine.  In humans, extremely high levels of thyroxine, such as those seen with thyroid gland tumors, are associated with cardiac arrhythmias and death, which is clearly the far-reaching source of the supposition made by racing officials when they posited a correlation between extra thyroxine and racetrack deaths.  Based on the fact that many racehorses are supplemented with thyroxine, and the spikes of sudden death were only observed in one jurisdiction, suggestions that there is a relationship between thyroxine supplementation and sudden death is irresponsible at best.

So, the next question is, why do so many trainers supplement racehorses with thyroxine, and is it a legitimate practice?  In actuality, while the scientific literature is sparse on the subject in horses, the only studies that exist clearly demonstrate that racehorses tend to be low in thyroxine1,2.  These old studies are likely the original source of the idea.  Luckily, there is a lot of good scientific evidence in other species.  Like the Thoroughbred racehorses in those two early studies, humans undergoing intense exercise can experience a low thyroxine3, which significantly impairs the individual’s ability to perform4.  Additionally, there is a lot of good scientific evidence in people5 and some limited data in horses6 that low thyroxine is associated with rhabdomyolysis (tying up).

Are there other causes besides intensive exercise that can cause low thyroxine in a race horse?  The test for thyroxine in horses measures both active (free) and inactive (protein bound) forms of thyroxine, and many exogenously administered substances, including estrogens and antibiotics can displace thyroxine from its protein binding sites, causing a falsely low thyroxine reading, without actually affecting the thyroid function in the horse.  Other exogenously administered substances can actually drop both the active and inactive forms of thyroxine in the blood.  For example, 5 days in a row of bute causes a profound drop in both forms of thyroxine7, which lasts 2 days for the active form and 10 days for the inactive form. 

Cobalt chloride also interferes with normal thyroid function.  While a critical and necessary mineral in trace amounts, cobalt has been used in higher than normal quantities for a variety of reasons in horses.  At levels exceeding daily requirements, especially many orders of magnitude above required amounts, cobalt activates and upregulates over 300 genes, of which a number might influence performance.  At modest amounts cobalt has been administered to racehorses to counteract low red blood cell counts, although there is no evidence that it works for this purpose, and has been used as a preventative for rhabdomyolysis8.  Very high persistent levels of cobalt in the blood interfere with the iodine uptake by the thyroid gland resulting in hypothyroidism, which is clearly the source of the rumored association between the two.  However, in addition to dropping the thyroxine level, injudicious use of cobalt causes heart and liver damage9.  Simple thyroxine supplementation does not counteract these adverse effects.  Additionally, the implementation of regulations limiting cobalt levels renders the exogenous administration of cobalt obsolete.

The veterinary community has not invested a lot of research on thyroxine in racehorses, but clearly the scientific data that do exist support limited use of this therapeutic substance in racehorses.  At the dosages typically used in racehorses, toxicity is unlikely, but the best method to determine if your horse needs supplementation is to have a baseline thyroxine test, and only supplement if your horse actually has a low thyroxine.  The best method to determine a reproducible level of thyroxine is to have the blood test drawn in the morning before training, when the horse has not received other medications within 10 days.

Conclusions?  The use of thyroxine supplementation is clearly an appropriate therapeutic treatment for horses, but it is critical to test for a low blood thyroid before starting on therapy, rather than indiscriminately treating all horses.  Has overuse of thyroxine been responsible for sudden deaths among racehorses?  Highly unlikely.  Is it good to investigate all possible variables when there is an unexpected number of deaths or other incidents involving animals?  Absolutely.  Is it irresponsible to suggest a cause without solid scientific evidence?  Yes.  Both regulators and horsemen are under pressure to have an answer whenever injuries or deaths occur, but kneejerk responses without science to back it up doesn’t get us any closer to solving the problems in our industry.  At the same time, it is critically important that we understand every therapy we institute in our athletes and apply each of them appropriately.

1.       1.      Bayly W, Andrea R, Smith B, Stensislie J, Bergsma G. Thyroid hormone concentrations in racing Thoroughbreds. Pferdeheilkunde; 12, 1996, pg 534-538.2.     Blackmore DJ, Greenwood RES, Johnson C. Observations on thyroid hormones in the blood of Thoroughbreds. Res Vet Sci; 25, 1978, pg 294-297.
3.     Pakarinen A, Häkkinen K, Alen M.  Serum thyroid hormones, thyrotropin and thyroxine binding globulin in elite athletes during very intense strength training of one week.  J Sports Med Phys Fitness. 1991 Jun;31(2):142-6.
4.     Werneck FZ1, Coelho EF, de Lima JR, Laterza MC, Barral MM, Teixeira PD, Vaisman M. Pulmonary Oxygen Uptake Kinetics During Exercise in Subclinical Hypothyroidism.  Thyroid. 2014 Mar 21. [Epub ahead of print]
5.     Lochmüller H, Reimers CD, Fischer P, Heuss D, Müller-Höcker J, Pongratz DE.  Exercise-induced myalgia in hypothyroidism.  Clin Investig. 1993 Dec;71(12):999-1001.
6.     Harris P, Marlin D, Gray J.  Equine thyroid function tests: a preliminary investigation.  Br Vet J. 1992 Jan-Feb;148(1), pg 71-80.
7.     Ramirez S, Wolfsheimer KJ, Moore RM, Mora F, Bueno AC, Mirza T. Duration of Effects of Phenylbutazone on Serum Total Thyroxine and Free Thyroxine Concentrations in Horses. J Vet Int Med 1997 11: 371–374.
8.     Fenger CK, Sacapolus, P.  2015.  What is Cobalt?  Horseman’s Journal Winter 2015.
9.     Ebert B, Jelkmann W. Intolerability of cobalt salt as erythropoietic agent. Drug Test Anal 2014 Mar;6(3):185-9.

Wednesday, October 28, 2015

Trace Environmental Substances Showing up as Post-Race “Positives” : What every horseman needs to know about being wrongly accused.

by Clara Fenger, DVM, PhD, DACVIM; Steven Barker, PhD; Keith Soring, DVM; Lee Shalgos, Esq.; Thomas Tobin, MRCVS, PhD, DABT.

The unveiling of the National Uniform Medication Program (NUMP) has come about with much fanfare and publicity; many states, such as the Mid-Atlantic States, signed on quickly while others have been more reticent to come on board.  The scientific basis for the thresholds which are supposed to permit appropriate use of the somewhat restricted list [27] of therapeutic medications while preventing unnecessary and excessive overmedication have hardly been forthcoming, and many hold-out states recognize these problems.  Despite the hurdles which NUMP faces from states unwilling to adopt such restrictive plans and from the internal deficiencies with the “Uniform Program”, such as insufficient scientific evidence, the over-regulation of therapeutic medications is not the only drug related obstacle which racing must face.

Substances used by humans, such as caffeine, cocaine and methamphetamine, commonly prescribed medications  for humans, such as tramadol and anti-anxiety drugs, and even over the counter medications like Aleve and ibuprofen have all come up as trace level “positives” in post-race samples in horse racing.  The biggest question surrounding these “positives” is:  Are these inadvertent environmental exposure, which horses picked up from eating contaminated hay, perhaps the result of a groom relieving himself innocently in a stall, or are they evidence of nefarious activity, forensic evidence of an attempt to steal a horse race?

In the early 1990’s, as the use of crack cocaine rose to a crescendo in the human population, there was a simultaneous rise in the general exposure of the population to trace levels of cocaine.  In fact, there are estimates that 90% of the paper currency in circulation is contaminated with measurable levels of the illegal substance.  People have even been wrongly accused of drug charges as a result of drug sniffing dog detection when they were only guilty of carrying large amounts of paper currency.  Our paper currency is so contaminated with drug residues that a confirmatory level of the major urinary cocaine metabolite, benzoylecgonine (BZE) at 100 ng/mL in urine is considered by the federal government to be a “negative” in that it is a forensically insignificant test.  Not everyone who handles money will test approaching this level:  this is one of the key points of environmental contamination related “positives”:  they strike randomly, and a number of important factors must align, including exposure to a source, individual metabolism and timing of the sample collection.   

The need for regulatory “cut-offs” for a number of environmental substances has been recognized by a number of states, which have put such cut-offs in place.  For example, in Ohio, there is a level for BZE (150 ng/mL in urine) and morphine (50 ng/mL in urine), the BZE “cut-off” being similar to the earlier  thresholds adopted by the federal government for BZE  as environmental contamination, as detailed in the 2012 National HBPA book on World Rules for Equine Drug Testing and Therapeutic Medication Regulation.

The potential for environmental residues of therapeutic substances to give rise to equine “positives” has been demonstrated repeatedly in the scientific literature.  When naproxen (Aleve ®) was administered to horses for 10 days, they continued to show measureable urine levels for over 45 days.  These researchers interpreted this prolonged elimination time  as being  due to stall contamination  with the medication and, consistent with this interpretation, when a horse which received no naproxen was placed in the stall of a naproxen treated horse, this horse  subsequently showed “positive” for urinary levels of naproxen for three days.  This occurred despite the fact that the stall and manger of the stall had been “thoroughly” cleaned.  Similar findings have been reported with flunixin (Banamine ®).  When untreated horses were placed in stalls of horses which had been treated with flunixin for 14 days, the untreated horses tested positive for 2 of the following 14 days.  Similar to what has been identified with BZE in humans, the appearance of an environmental contamination positive can clearly be associated with exposure to environmental residues.  In another study, isoxsuprine could be identified from the floor, walls manger, and even the cobwebs in the rafters of a stall of a horse which had received a therapeutic regimen of the substance, which was then discontinued for three weeks pre-race. A number of other studies have found similar results for a variety of drugs.

Key Factors Influencing a “Positive” from an environmental exposure

What are the key reasons that some substances can remain in the environment and therefore pose a risk for inadvertent environmental exposure to the competitive equine athlete?  There are three key criteria which must be met in order for a substance to represent a high risk.  The first criterion is that the substance, like isoxsuprine, must be chemically stable in the environment.  Once it is introduced into a stall it remains there, and hangs around waiting throughout the stall, including apparently in the cobwebs.  For an environmental contaminant be a significant problem, it first needs to be chemically stable and therefore persist in the environment.

The second point is that the contaminant must be easily absorbed, which usually means orally absorbed.  For an environmental substance to give rise to drug testing problems, it must be able to get into the horse from the environment.  Oral absorption is likely route number one, and we know that isoxsuprine is well absorbed orally.  This is more problematic than the aforementioned incidental exposure of humans to cocaine tainted currency:  horses, especially intact males, are well known to seek out urine in stalls, smelling and tasting the waste from any previous occupants to identify who may have been in that stall before them.  This is an animal behavior unparalleled by humans.

Next comes a very interesting point; although isoxsuprine is well absorbed orally, effective blood levels of isoxsuprine after oral administration are difficult to achieve.  Orally absorbed Isoxsuprine is so rapidly and effectively metabolized/glucuronidated on its first pass through the liver that it fails to rise to effective plasma concentrations after oral administration.  Orally absorbed isoxsuprine is essentially immediately glucuronidated, or changed into an inactive metabolite, while passing through the liver, and it is then shipped as the metabolite to the kidney for excretion.  This feature of isoxsuprine yields the highest known concentrations of any equine drug metabolite, peaking at 700,000 ng/mL.; given that about a 700 mg IV dose of isoxsuprine produced this urinary level of isoxsuprine in urine, a 1 mg dose could produce a 700ng/ml urinary concentration, readily explaining the ability of minute amounts of environmental isoxsuprine to produce low nanogram/ml urinary isoxsuprine glucuronide “positives” . 

To summarize, isoxsuprine is chemically stable in the environment, it is well absorbed orally and is, for all practical purposes, immediately converted into its glucuronide metabolite and excreted at very high concentrations in equine urine.  Take home message:  very small environmental traces of isoxsuprine passed in the urine of one horse, and lingering in its stall will result in sufficient isoxsuprine glucuronide in the urine to trigger a “positive” call for isoxsuprine in any untreated horse which subsequently inhabits the stall.

So the rules for a problem environmental substance are:  (1) it must be chemically stable in the environment; (2) it should be well absorbed orally; (3) it should be excreted in relatively high concentrations in the urine.  Any substance with these chemical/biological characteristics has the potential to be an environmental substance “positive” problem in equine drug testing, as we will see.


Substances  that become widely used in humans for foodstuff [caffeine] recreational [cocaine methamphetamine] or medical use [Ritalin], or as prescription pain medications, such as Tramadol, may well spill over into the equine world as trace environmental substances and  show up, usually at trace levels,  in racehorses.   In contrast to its therapeutic effectiveness as a painkiller in both humans and small animals, Tramadol in the horse has failed to show any significant analgesic effects in horses at similar doses.  At very high doses, exceeding 10 times the effective dose in humans, by weight, Tramadol can induce analgesic effects that may last for less than three hours.  This lack of efficacy in equines appears to stem from the rapid inactivation of the active primary metabolite, 4OH-tramadol, by glucuronidation, a unique feature in this species.  Nonetheless, this inactivated metabolite is detectable in the horse’s system for a long time after the last administration.

Tramadol is clearly stable in the environment; one environmental study looked into the fate and potential impacts of Tramadol after its introduction into water.  It was found that neither tramadol, nor its derivatives formed from exposure to light were significantly biodegradable according to standard test guidelines.  They simply remain in the water forever.  In Europe, Tramadol was #3 on the list of pharmaceuticals found in wastewater, ranked by median concentration, with the highest wastewater concentration reported being 1166 ng/L or 1.166 ng/ml.  Similar to flunixin and isoxsuprine, tramadol is highly persistent in the environment, and could readily serve as a substance of incidental environmental exposure in the horse.

On July 11th 2013 at Hoosier Park Indiana the Harness horse “Justice Jet” yielded a positive for “Tramadol”.  We assume that the actual chemical nature of the analyte was O-desmethyltramadol and the concentration identified was not available to us at the time of writing.  At the time of the identification Tramadol was classified by ARCI as a Class 2, Penalty Class A substance.  However, the trainer involved was apparently highly regarded, with a reportedly 30 year unblemished medication violation record.  Additionally, the trainer was unaware of any risk factors such as individuals prescribed Tramadol in the environment of or close to the horse in question.  Given the circumstances of this case and the apparently complete innocence of the trainer, a review of the ARCI penalty classification for Tramadol was requested by the Indiana Racing Authorities.  While Tramadol is still listed as an ARCI Class 2, penalty class A substance, the Indiana authorities chose to penalize the trainer involved with a relatively “modest” 15 day penalty and a $500.00 fine, a much less severe penalty than the in place ARCI penalty guidelines would have suggested.  It would also be fair to say that in absence of an identified prescription for Tramadol in the racing environment of the horse, this case becomes a simple generic environmental exposure case, with no specifically identified source for the detected Tramadol.  

In a December 4th 2013 case before the New Zealand Judicial Control Authority for Racing, human use of Tramadol was considered to account for “exposure” of a horse to Tramadol, i.e., a reported identification in a Standardbred racehorse.  The trainer/ person responsible for the horse had been prescribed Tramadol by her doctor for back pain, and the Tramadol metabolite O-desmethyltramadol was found in a post-race sample taken from the horse.  Under the New Zealand Rules of Harness Racing, the presence of Tramadol or its metabolites is prohibited. The person responsible admitted to taking 50 mg Tramadol capsules on race nights, and the most likely cause of exposure to the horse was considered to be through contamination of her hands after taking the medication and bridling the horse.  The concentration of tramadol identified in the sample was extremely low, 100 pg/mL (0.1 ng/ml), much lower than the LOD (limit of detection) in the recently published Knych et al. (2013) report.   It is also noted that this very low Tramadol concentration was not identified in the primary New Zealand testing laboratory, but was only identified upon additional and apparently Tramadol-unrelated high sensitivity retesting of the sample by the Hong Kong Jockey Club laboratory.  The trainer in question was fined $3,300 and to our knowledge was not suspended.

In this matter the New Zealand judicial authority appears to have accepted the likelihood of inadvertent exposure of the horse to very small and pharmacologically insignificant amounts of Tramadol associated with Tramadol being prescribed for medical purposes to the trainer in question. 

Similarly, in one California Tramadol positive in a racehorse, the trainer was prescribed Tramadol for chronic back pain associated with a training accident.  This trainer’s horse had tested “clean” only ten days earlier, which underscores the random nature of the environmental positive.


Cathinone, a component of the synthetic amphetamine “bath salts,” has been identified in trace amounts in racehorses in Iowa.  Bath salts hit the streets in the US in 2009 as a drug of human abuse, reaching epidemic proportions by 2011 and started to show up in racehorses in Iowa in 2011.  Further complicating the interpretation of the test, cathinone can be found in urine in some testing systems after the administration of the common antihistamine, Pyrilamine (TriHist®, Anihist®), ephedrine (Sudafed®), or propanolamine.  The AniAaaaAAThe first racehorse positive was in 2011 at a level of 3 ng/mL  in urine, too low to reflect any pharmacologic effect on the animal, and consistent with other drugs of human abuse:  environmental exposure at low levels.  The trainer was summarily suspended and even handcuffed and removed from the grounds in a squad car.  Shortly thereafter, other positive tests for cathinone started to trickle in, and over the course of 70 days, there were 16 trainers involved, encompassing 51% of the barns and 5 different veterinary practices.  The sheer number of “positives”, and the low level of the drug in the animals suggested that  it was likely that this was an environmental exposure (Kind et al., 2012).

A survey was conducted to identify any commonality among the positive identifications of cathinone, and none could be identified.  Hay and several species of broad leaf plants from local farms were analyzed without success and the source of the contamination could not be identified.   Human drug testing was inconclusive as well.   However, based on the levels found in the positive tests, an environmental contamination “cut-off” was determined to be 10 ng/mL, which was adopted by the Iowa Racing Commission as a new regulatory threshold. 

The cathinone positives have slowed down in Iowa after this initial spike, but regulators remain convinced that these positives resulted from some exposure in the environment, possibly of plant origin, although simple contamination from human recreational use cannot be ruled out. 


            Another emerging substance of inadvertent environmental exposure in horses is methamphetamine.  Like many of the other substances of concern, methamphetamine has become a widespread drug of human abuse.  A recent case involving a cluster of trace concentration methamphetamine positives in Canada provides a classic example.  These events started when a Michigan trainer purchased a large used horse trailer shortly before shipping her Quarter horses to Toronto to race at Ajax downs. The three horses that shipped in this trailer raced within two days of shipping, and all three tested positive for picogram urinary concentrations of methamphetamine, while another horse, which shipped in a different trailer, also raced but tested negative.  This circumstance presents a classic "cluster" of very low, picogram concentration methamphetamine positives, two occurring on day one and one the second day post shipping, with a horse not shipped in the suspect trailer testing negative.

The Ontario investigators interviewed the trainer and as part of their review tested a number of samples from the trainer's equipment, including a sample from the manger area of the trailer.  This trailer sample turned up “positive” for methamphetamine at 22 ng/milligram, providing a clear-cut environmental source for the methamphetamine identified in the horses shipped in this trailer. The most likely interpretation of these events is that this trailer, at some time in its previous life, had housed an illicit methamphetamine laboratory, and that traces of methamphetamine remained in the newly purchased trailer and transferred to the horses during shipping to Toronto. Similar environmental related methamphetamine contamination of humans working in or around methamphetamine synthesis facilities are well understood.  What is interesting, however, is that this is the first clear-cut link between an environmental source of methamphetamine and a classic "cluster" of low concentration environmentally driven equine methamphetamine identifications.

As with isoxsuprine, we can also estimate the dose of methamphetamine required to give rise to these reported identifications.  Work by Ohio State University researchers performed in the early 1970s shows that administration of a 150 mg dose of methamphetamine (a comparable dose to the effective dose in people) to a horse will yield a peak urinary methamphetamine concentration in the region of 7,000 ng per ml.  Simple arithmetic shows that 1 µg of methamphetamine, or 1/150,000th of the effective dose investigated by the researchers in Ohio, is all that is required to yield urinary concentration of 47 pg per ML, a concentration very close to the lowest urinary methamphetamine concentration reported in this Toronto cluster, which was 56 pg per ML, or parts per trillion in urine.  Methamphetamine adheres to thhe three key factors of an inadvertent environmental exposure:  (1) it is a chemically stable substance that is (2) orally absorbed from the environment of a horse and (3) can appear in urine at extremely low concentrations, but such findings are indicative of nothing more than totally minuscule exposure of the animal to the substance in question.  Further, the pervasiveness with which methamphetamine is used as a substance of human abuse greatly increases the likelihood that it will overlap with an unsuspecting population of horses.

Other Drugs of Human and Animal Use

There has also been a recent flurry of low-level “positives” for O-desmethylvenlafaxine, the major metabolite of venlafaxine, better known as the anti-depressant Effexor, in Canada, the USA and recently in India.  As with Tramadol, routes of human contamination of the horses are thought to be the same. In a small study conducted by Canadian authorities, a horse was dosed with venlafaxine and an aliquot of the horse urine used apparently “poured” on hay and the hay placed in a “clean” stable and housing a “clean” horse, rapidly leading to the “clean” horse becoming positive for O-desmethylvenlafaxine.  Indeed, low levels of this metabolite have also been found in supplements. This is similar to the sildenafil or Viagra positives seen several years ago, where a supplement source for the drug was identified in products produced by a compounding pharmacy.

Across the US and Canada there has also been a number of very low level levamisole positives. Levamisole, an “old time” wormer approved by the FDA for use in cattle and commonly used off label for horses was originally categorized as an Association of Racing Commissioners International, ARCI  Class 4 substance, but has been reclassified as a 2 with an A penalty.  It has experienced a resurgence in usage in horses in recent as a result of its immune stimulating properties, and has been widely recommended by researchers in the field of Equine Protozoal Myeloencephalitis.  Additionally, levamisole is very environmentally stable, is routinely used in food animals and can be found in fertilizers made from their excrement.  Experiments are currently underway to determine how much of this compound is taken up into food products such as corn and whether or not this is a potential source for some levamisole positives.  Nonetheless, levamisole can be found in runoff from organic fertilizer treated fields.

The list of such instances could continue and many more cases for low level positives in racing could be cited. This is, in many cases, the result of our ever increasing ability to detect pharmacologically insignificant amounts of drugs commonly used by humans and other animals in equine samples. More and more of the positives being called approach the low picogram (1,000th of a nanogram)/ml level and are being treated by many Commissions as if they were pharmacologically effective levels normally seen immediately following administration. There is, in such cases, no distinction being made between “trace” and perhaps even “ultra-trace” levels arising from environmental sources and what a true attempt to influence performance actually requires. “Zero Tolerance” or no tolerance for the finding of a substance, regardless of the level, has once again reared its ugly head but is now armed with more sensitive equipment than ever before.

As more therapeutic medications are prescribed to humans and used in food animals, as more enter the environment of the horse by a variety of mechanisms, as more medications and recreational substances  go into our wastewater, our foods, feeds, supplements and even the air, we will eventually reach a point where the sensitivity of our testing and the pervasiveness of these drugs will cause ever greater numbers of positives for the racing industry.  If we fail to address these facts the consequences are an ever increasing suspicion of corruption in racing, from the calling of drug “positives” at irrelevant levels which inevitably erode confidence in the integrity of the industry.  The attempts by several of the controlling associations to subsequently regulate common therapeutics by placing thresholds on useful veterinary products at the same level as may occur from contamination will only add to this problem. It is possible that their attempts will only result in the opposite of what they desired.


For reasons that are unclear the RMTC is uninterested in setting thresholds/cut-offs for inadvertent environmental exposure, despite the obvious need and at times specific requests for such guidance by regulators and horsemen’s organizations.   It is incumbent upon the regulators that police racing to distinguish between unavoidable trace level environmental exposure with no effect on the animal or the competitive event and no intentional malfeasance.  Innocent and unavoidable environmental sources should be considered in every instance in which low levels of substances are identified in post-race samples.  Identification of trace levels of substances of human abuse should be carefully investigated, and policies should be enacted to prevent the improper penalization of innocent horsemen.  In fact, in New York, the Court of Appeals has ruled that testing laboratories for humans could be held liable for calling positives on drug levels consistent with environmental contamination.

Guidance for Horsemen:

The primary source of environmental contamination positives in racehorses is exposure to humans who have handled or are taking prescription medications or illegal substances.  The best defense is careful oversight of your shed row.  If any grooms are suspected of illegal drug use, drug testing, with hair testing is by far the best option followed, where necessary, by removal from the care of the horses is the best option.  Any grooms taking prescription medications should be warned to wash their hands thoroughly between taking their medications and handling horses.   Tongue ties are of particular importance.  Use only new tongue ties on race-day, and do not allow grooms to carry them in their pockets, where they may be commingled with prescription pill fragments.   Carefully admonish grooms and riders NOT to urinate in the horse’s stalls.  Even over the counter pain medication, like Aleve® have been implicated in positive tests as a result of careless urination by a groom in a horse’s stall.

Horsemen and veterinarians should be cognizant of the source of their medications and supplements.  Positives may result from the use of supplements and compounded materials prepared and sold by less reputable compounding pharmacies, containing unidentified drugs.  Often, claims that sound too good to be true are exactly that. 

As the medication rules become more onerous, horsemen must be ever more proactive in controlling the environment of their charges.  Even when we have controlled everything in our shed rows, there will still be inadvertent environmental contamination positives, because we cannot control the ship-in barn.  Horses live in barns and not maximum security facilities.  We can only hope that the regulatory pendulum swings back towards reason, and our regulators can determine a way to avoid unfair penalizing of innocent horsemen, much like the justice system ultimately did the right thing by the unfairly convicted parolees of the 1990’s.

Suggested additional reading:

Kind AJ, Soring K, JD, Brewer K , Eisenberg, R., Hughes CG, Hartmann-Fishbach, P and Tobin T (2012) Cathinone and Related "Bath Salt" Substances - Detection in Equine Urine and Potential Sources, In press, The 19th International Conference of Racing Analysts and Veterinarians, University of Pennsylvania, Philadelphia, Pennsylvania, September 2012.

Wennerlund I, Ingvest-Larson C, Kallings P, Fredrickson E, Bondesson U.  2000.  Pharmacokinectics and urinary excretion of naproxen after repeated oral administration in the horse.  In RB Williams, E Houghton, JF Wade (Eds) Proceedings of the 13th Annual Conference of Racing Analysts and Veterinarians (pp. 195-200).  Cambridge, UK.

Norgren A, Ingvest-Larson C, Kallings P, Fredrickson E, Bondesson U.  2000.  Contamination and urinary excretion of flunixin after repeated administration in the horse.  .  In RB Williams, E Houghton, JF Wade (Eds) Proceedings of the 13th Annual Conference of Racing Analysts and Veterinarians (pp. 377-380).  Cambridge, UK.

Russell CS, Maynard S.  2000.  Environmental contamination with isoxsuprine.  In RB Williams, E Houghton, JF Wade (Eds) Proceedings of the 13th Annual Conference of Racing Analysts and Veterinarians (pp. 377-380).  Cambridge, UK.

Knych HK, Corado CR, McKemie DS, et al. 2013. Pharmacokinetics and pharmacodynamics of tramadol in horses following oral administration. J Vet Pharmacol Ther 2013;36:389-398.

Loos R, Carvalho R, Antonio DC, et al. 2013. EU-wide monitoring survey on emerging polar organic contaminants in wastewater treatment plant effluents. Water Res 47:6475-6487.

Thursday, April 16, 2015

What is Cobalt?

What is Cobalt?

By:  Clara Fenger, DVM, PhD, DACVIM and Pete Sacopulos, JD

Cobalt is a mineral which is present in animal systems predominantly as part of Vitamin B12, which is a key cofactor in DNA synthesis. It is a nutritionally required element with the lowest dietary requirement of all of the required trace minerals.   Recommended daily allowances for humans is about 0.1 µg/day in the form of Vitamin B12 (Unice et al., 2012).  The National Research Council, which determines daily requirements of nutrients for animals, has determined a dietary cobalt requirement of 0.5 to 1.1 mg/day in horses and 1.2 to 2.4 mg/day for a cow, depending upon weight and metabolic state (Lawrence et al., 2007).  Maximum tolerable amount is 25 mg/kg of dry matter of feed or forage (Lawrence et al., 2007).

Cobalt concentrations in drinking water typically range from 0.1–5 ng/mL, and daily cobalt intake ranges from 11–580 µg/day in humans (Unice et al, 2012; Schroeder, 1967). Legumes like alfalfa and clover are the principle sources of cobalt in the natural diet of horses, and there can be widely varied concentrations of cobalt in these forages across different geographical locations in the United States and internationally (Kubota et al, 1987, Herbert, 1996). Deficient soils can result in profound deficiencies in cattle and sheep, resulting in anemia, weight loss and general unthriftiness.  Deficiencies in these species are likely under identified, but readily corrected by supplementation (Merck Veterinary Manual, accessed 1/22/15).  In contrast, deficiency has not been identified in horses.

Oral absorption of cobalt salts in humans varies depending on the dose, compound, nutritional status of the individual and proximity in time to a meal (Tvermoes et al., 2013).  For example, less than 5% of an oral dose of cobalt oxide is absorbed, compared with 30% of an oral dose of cobalt chloride, as determined from rodent studies. Increasing the dose of cobalt does not tend to cause significant accumulation as increasing doses result in a smaller proportion of the dose being absorbed. In humans, approximately 80% of cobalt is excreted in the urine and about 15% in the feces (Lison, 2007).  No studies evaluating the oral absorption of cobalt have been done in horses.

Why is cobalt used in athletes?

Human recombinant erythropoietin (EPO) became available in the 1990’s, and rapidly became adopted into covert doping programs for human athletes.  EPO stimulates the bone marrow to produce red blood cells, increasing the oxygen carrying capacity of the blood.

The ability to test for EPO was developed in the early 2000’s, curbing its abuse, but it was pointed out in the medical literature (Lippi et al., 2005, 2006) that cobalt, the original treatment used to increase human hematocrits in the treatment of anemia, also had potential to be used/misused in athletics in much the same way as administration of EPO. The potential illicit use of cobalt to improve athletic performance is based on the cellular actions of cobalt, leading to its recent and reportedly widespread use in horse racing (Paulick, 2014; Merkeberg, 2013).

A high dose of cobalt triggers a series of events which increases endogenous blood concentrations of EPO. Cobalt increases a protein in the cell called Hypoxia Inducible Factor (HIF-1α).  High intracellular HIF-1α causes direct activation of the erythropoietin gene resulting in increased plasma concentrations of erythropoietin, which would then drive the increase in red cell formation and an equivalent increase in the animal’s hematocrit (Semenza, 2014).  In addition to its EPO effects, other effects include increased blood supply to the muscle, and increased efficiency of energy utilization.

The theory is interesting but does this really work in horses?  The hematological effects at high levels of cobalt can be impressive:  in laboratory animals (Gluhcheva et al., 2011), dogs (Fisher, 1959) and humans (Lippi et al., 2006) chronic daily administration of large quantities of cobalt chloride have resulted in significant increases in red blood cell counts and hemoglobin.  Lower doses appear to have no hematological effect.  In horses, single doses of 22 mg or 49 mg of Co2+  (Kynch et al., 2014), or up to 449 mg of Co2+  (pers. Comm. Mary Scollay) had no effect on red blood cell parameters.   The consensus from the literature including multiple studies in multiple species is that chronic blood concentrations of 300 ppb and less are not associated with hematological or toxicological effects, whereas chronic concentrations in excess of 300 ppb are associated with both hematological and toxicological effects (Finley et al, 2012).

In racehorses, cobalt is typically administered a day or two before racing.  It seems unlikely that cobalt directed changes in EPO gene transcription or capillarity of muscles could be in effect at the time of the race if cobalt is administered only a few days before competition.  However, a small subset of horses, specifically among Standardbreds, seem to have impressive racing performances pursuant to Co2+ administration two days pre-race.  One possible mechanism of action of this cation is its effect as a calcium channel blocker (Simonsen et al., 2012).  Standardbreds are unique among the racing breeds in that they are susceptible to “tying-up” during racing (Isgren et al., 2010).  Other racing disciplines do not suffer this condition during racing, but may “tie-up” before or after racing.  The underlying mechanism of tying-up is unclear, except that it is probably a heritable condition (Collinder et al., 1997), and likely mediated by calcium channels in the muscles (Lopez et al., 1995).  Certainly some of the preventative therapies for tying-up involve calcium channel blockers, such as Dantrolene (Lopez et al., 1995) or another divalent cation, magnesium (Fenger, unpublished observation).  Since tying-up is prevalent among Standardbreds, and other preventative treatments for this condition are banned on raceday, the purported performance enhancing effect of Co2+ administration two days before racing may simply be prevention of tying-up.   

Why should cobalt be regulated?

Other than a possible mechanism to prevent tying-up, or muscle cramping in some horses, there appears to be no effect of cobalt on horses in the doses most commonly used.  On the other hand, there are many reports of cobalt toxicity in people.  In a mining town in Peru, excessive exposure caused chronic excessive red blood cell production in miners to the point of sludging of blood (Jefferson, 2002) and the US Environmental Protection Agency has set safety levels for cobalt for the protection of workers at high risk, such as mining and some industrial jobs. 

Typical adverse reactions to chronic cobalt administration in humans include nausea, vomiting, heart failure, low thyroid hormone levels and goiter, with neurological symptoms being reported less frequently (Jelkmann, 2012).  Large circulating concentrations of cobalt interfere with the uptake of iodine into the thyroid gland, resulting in low thyroid hormone levels.  In the late 1960’s, cobalt was added to beer as an anti-foaming agent, and resulted in a number of cases of heart failure. Cobalt is taken up in high concentrations into the heart muscle, likely in its role as a calcium channel blocker, and has been linked to an epidemic of heart failure in a group of heavy beer drinkers from the cobalt-containing beer (Alexander, 1972).  This specific syndrome was clearly multi-factorial, as the specific disease syndrome did not match any others associated with excessive cobalt.

It is likely that these chronic high doses which are required to cause thyroid dysfunction or cardiac failure are not achievable in horses.  However, at the extremely high doses that are suspected of being in use, horses can exhibit impressive adverse effects, such as tremors, sweating and colic (pers. comm Mary Scollay).  All of these effects are transient, and the horses appeared to be completely normal within a few hours.  Nonetheless, there has been suspicion that some incidents of sudden death on the racetrack have been associated with high serum cobalt levels, suggesting that there may be a relationship.

What are current regs?

Hong Kong has had a cobalt threshold longer than any other jurisdiction, and the ‘in-house’ urine cobalt threshold is 100 ng/mL.  A recent paper out of Hong Kong suggests 2 ppb as an appropriate threshold in plasma (Ho et al., 2014).  Among the criticisms of this threshold by the Hong Kong authors is that one of the study populations was a group of 109 horses in the Emirates.  This population races under strict security and yet 6 horses were considered to be outliers and had to be eliminated from the population before the balance of the group would fit the 2 ppb threshold.  Thresholds for naturally occurring substances must be carefully considered, and elimination of data must be done only where considerable investigation is performed. 

In Australia, Harness racing officials have introduced a urinary threshold of 200 ng/mL after samples in certain harness horses there reportedly exceeded 3,500 ng/mL (Bartley, 2014).  Similar to the Hong Kong analysis, Hibbert (2014) examined a group of post-race samples and identified a “natural break” in the data, ultimately requiring the researchers to eliminate 17 horses from a population of 80 in order to make the proposed threshold fit.  Again, the elimination of 20% of the population without further investigation into alternative explanations, such as feed or hay sources is inappropriate for establishing a threshold for which the penalty is on par with erythropoietin.

Indiana is the first US jurisdiction to regulate and implement a cobalt rule.   Indiana’s rule regulating cobalt establishes a threshold of 25 ppb.   The regulation that was implemented, by way of emergency rule, became effective October 1, 2014, for horses in competition and January 1, 2015, for out-of-competition testing.   Indiana’s cobalt rule states:  “Cobalt – not to exceed twenty-five (25) parts per billion of cobalt in serum or plasma.  A sample from a horse tested and found by the Commission’s primary lab to have cobalt in excess of this threshold shall be placed and remain on the veterinarian’s list until the concentration of cobalt in serum or plasma has fallen below the designated threshold.”   71 IAC 8-1-9(a)(3)   Indiana’s rule regulating cobalt makes a positive test a category “A” penalty, as established by the Recommended Penalties and Model Rule, regardless of its presence in a post-race or out-of-competition sample.    71 IAC-8-1-7(b)

Interestingly, the Indiana Horse Racing Commission has issued a supplemental notice relative to the new emergency rule on cobalt that suggests leniency for those receiving a positive test for cobalt wherein the positive detected level is more than 25 ppb but less than 50 ppb. 

Indiana’s cobalt rule for in-competition testing became effective during Indiana’s 2014 season and, significantly, with only weeks left in that season.   Despite that being the case, there have been positive test results for cobalt all of which, as of this date, have been/are for in-competition testing.

Several other states have been closely observing Indiana’s lead to regulate cobalt.   The State of California is one of those states.   In fact, the State of California’s Horse Racing Board recently discussed a proposal by its Medications and Track Safety Committee to add Board rules to regulate the use of cobalt.   In October of last year, California’s Equine Medical Director, Dr. Rick Arthur, addressed the issue before the State of California Horse Racing Board.   In doing so, Dr. Arthur noted a study that the Board had performed at Maddy Laboratory that examined the results of Cobalt Chloride at low doses.   Dr. Arthur reported that: “. . .  what was found is that after the first elimination where probably 80% of it is eliminated in the first 48 hours . . . it takes weeks to get rid of the rest of it . . . that’s actually good news because it allows us to set a threshold that would eliminate its use . . . .”   (State of California Horse Racing Board Meeting Minutes for October, 2014)   Based on Dr. Arthur’s report and other considerations, the Executive Director of the State of California’s Horse Racing Board,  Rick Baedecher, stated that the Board would pursue a rule regulating cobalt at: “ . . . the 25 nanogram level.   And if a horse is above that level (the horse) will be placed on the vet’s list until . . . it tests lower than the threshold . . . .”  (State of California Horse Racing Board Meeting Minutes for October, 2014)    Dr. Arthur, at the October, 2014 meeting of the Board, also recommended the proposed rule regulating cobalt include out-of-competition testing.

Another state contemplating a rule regulating cobalt is Minnesota.   In the Fall of last year the Minnesota Racing Commission discussed the status of cobalt regulation in racing.    The chief veterinarian of the Minnesota Racing Commission, Dr. Lynn Hovda, provided a briefing to the Commission.  Dr. Hovda discussed both Indiana and California’s efforts in this regard and the reason for needed regulation.   The reasons given are those that are normally cited with regard to regulating cobalt, being the potential for toxicity to equine athletes if Cobalt Chloride is administered at high levels and a second reason being that it is believed that excessive amounts  of Cobalt Chloride may be performance enhancing.  The Minnesota Racing Commission has been and continues to move forward with adopting a rule for regulating cobalt.   It will likely closely follow the Indiana rule.   One exception may be the threshold level.    Discussions have included setting the threshold level for cobalt between 50 and 70 ppb, slightly higher than Indiana’s current threshold of 25 ppb.   (Minnesota Horse Racing Commission Meeting Minutes for September and October, 2014)

On March 4, 2014, the California Horse Racing Board (CHRB) issued a notification that they were about to commence monitoring cobalt concentrations, and they implemented a regulatory threshold of 25 ppb.  California, similar to the Hong Kong and Australian studies performed preliminary survey studies, which also excluded outliers in order to achieve their threshold.  The trend is clear:  in order to set the regulatory thresholds, horses with higher cobalt levels must be eliminated from the study data because they fall outside what looks “right.”  This is hardly appropriate science, nor the appropriate method of regulating a multi-million dollar industry.

On May 20, 2014, the Maryland Racing Commission decided that Maryland will begin testing horses for cobalt. The chairman of the board indicated that there was ‘no definitive threshold’ concentration for cobalt at that time. There are also no specific rules in place in Maryland relating to cobalt, and no comments were made on how a horse presenting with high concentrations of cobalt might be addressed (Vespe, 2014). 

In the absence of a state regulation, the Meadowlands Racetrack owner, Gural implemented a “house rule” regarding cobalt, using the current Hong Kong threshold of 10 ppb, as an out of competition threshold.

There is marked confusion over the regulation of cobalt, because experts disagree on an appropriate thresholds (Popot, 2014).  Surveys of horses have resulted in the wholesale elimination of “outlier” horses based on the assumption that these horses must have been treated with exogenous cobalt.  In a study where different laboratories across the globe tested the same samples, the variation between testing laboratories was as high as 82% for the testing in serum and 23% in urine.  Because of the greater agreement between lab testing methods using urine, the international community has settled on a urine threshold of 100 mcg/mL raceday threshold for uniformity.

No dose-response studies have been performed to determine at what level cobalt has any effect on horses.  However, a review of the literature in the animals which have been studied indicates that a sustained cobalt level above 300 ppb is required for cobalt to exert its hematopoietic and other effects (Finley, 2012).  Current regulatory focus on thresholds have sought levels which reflect likely exogenous administration of a naturally occurring substance which may or may not have been with intent, and not levels which actually reflect any performance enhancing effect.  This lack of valid scientific basis for the cobalt thresholds which have been adopted has led to headlines and career-threatening accusations across the turf media.


The regulation of cobalt in North America has come about with great fanfare and headlines, but the science has yet to catch up.  The thresholds adopted fail to hold up to scientific scrutiny, and, like so many other regulations in this day and age, are more likely to trap innocent horsemen than those actually cheating.

  1. Unice, K.M., Monnot, A.D., Gaffney, S.H., Yvermoes, B.E., Thuett, K.A., Paustenbach, D.J., Finley, B.L.., 2012.  Inorganic cobalt supplementation: prediction of cobalt levels in whole blood and urine using a biokinetic model. Food and Chemical Toxicology 50:2456-61. 
  2. Lawrence, L.M., Cymbaluk, N.F., Freeman, D. W. et al.  2007.  Nutrient Requirements of Horses.  Chapter 5.  Minerals.  The National Academies Press.  pp 87-88. 
  3. Schroeder, H.A., Nason, A.P., Tipton, I.H.  1967.  Essential Trace Minerals in Man:  Cobalt.  Journal of Chronic Disease.  20:869-890.
  4. Kubota, J., Welch, R.M., Van Campen, D.  1987.  Soil-related nutritional problem areas for grazing animals.  Advances in Soil Science.  Ed. B.A. Stewart.  P 193.
  5. Herbert, V., 1996. Vitamin B-12. Present knowledge in nutrition, 7th edition. Washington, D.C. ILSI p 191-205.
  6. http://www.merckmanuals.com/vet/management_and_nutrition/nutrition_cattle/nutritional_requirements_of_beef_cattle.html?qt=cobalt%20&alt=sh (accessed Jan 15, 2015)
  7. Tvermoes, B.E., Finley, B.L., Unice, K.M., Otani, J.M., Paustenbach, D.J., Gailbraith, D.A., 2013.  Cobalt whole blood concentrations in healthy adult male volunteers following two weeks of ingesting cobalt supplement. Food and Chemical Toxicology 53:432-39.
  8. Lison, D., 2007. Cobalt. In Handbook on the Toxicology of Metals, 3rd edition.  Nordberg, G.F., Fowler, B.A., Nordberg, M., and Friberg, L.T., Eds., Amsterdam: Elsevier Science Publishers. Pages 511-28.
  9. Lippi G, Franchini M, Guidi GC., 2005. Cobalt chloride administration in athletes: a new perspective in blood doping. British Journal of Sports Medicine 39(11):872
  10. Lippi, G., Franchini, M., Guidi, G.C., 2006.  Blood doping by cobalt. Should we measure cobalt in athletes? Journal Occupational Medicine and Toxicology. 1:18-20.
  11. Paulick, R., 2014.  Is cobalt a killer in horses? The Paulick Report. http://www.paulickreport.com/news/ray-s-paddock/is-cobalt-a-killer-in-horses/ (accessed June 1, 2014)
  12. Merkeberg J.  2013  Blood manipulation: current challenges from an anti-doping perspective. Sports Medicine in Hematology 2013:627-31.  doi: 10.1182/ asheducation-2013.1.627.
  13. Semenza, G.L.  2014., HIF-1 and human disease: one highly involved factor. Genes and Development 14:1983-1991.
  14. Gluhcheva, Y., Madzharova, M., Zhorovab, R., Atanasov, V., Ivanovac, J., Mitewa, M., 2011. Cobalt(II)-induced changes in hemoglobin content and iron concentration in mice from different age groups. Biotechnology & Biotechnological Equipment 26:126-128
  15. Fisher, J.W., 1959.  The effects of cobalt injections on total circulating red cell volume and bone marrow cytology in normal and adrenalectomized dogs. Endocrinology 64(4):522
  16. Knych, H.K., Arthur, R.M., Mitchell, M.M. et al.  2014.  Pharmacokinetics and selected pharmacodynamics of cobalt following a single intravenous administration to horses.  Drug Testing and Analysis.  DOI 10.1002/dta.1737.
  17. Finley, B.L., Monnot, A.D., Gaffney, S.H. et al.  2012.  Dose-response relationships for blood cobalt concentrations and health effects:  A review of the literature and application of a biokinetic model.  Journal of Toxicology and Environmental Health, Part B:  Critical Reviews 15(8):493-523.
  18. Simonsen, L.O., Harbak, H., Bennekou, P., 2012. Cobalt metabolism and toxicology—a brief update. Science of the Total Environment 432:210-15.
  19. Isgren, C. M., Upjohn, M. M., Fernandez-Fuente, M., Massey, C., Pollott, G., Verheyen, K. L. P., and Piercy, R. J. 2010. Epidemiology of Exertional Rhabdomyolysis Susceptibility in Standardbred Horses Reveals Associated Risk Factors and Underlying Enhanced Performance. PLoS ONE, 5(7), e11594. doi:10.1371/journal.pone.0011594.
  20. Collinder, E., Lindholm, A. and Rasmuson, M.  1997.  Genetic markers in standardbred trotters susceptible to the rhabdomyolysis syndrome.  Equine Veterinary Journal. 29:117-20.
  21. López, J.R., Linares, N., Cordovez, G. and Terzic, A. Elevated myoplasmic calcium in exercise-induced equine rhabdomyolysis.  Pflugers Arch. 1995 430:293-5.
  22. Jefferson, J.A.  2002.  Excessive erythrocytosis, chronic mountain sickness and its relation to serum cobalt levels. Lancet 359:407-8.
  23. Jelkmann, W., 2012. The disparate roles of cobalt in erythropoiesis, and doping relevance. Open Journal of Hematology 3-6.  DOI: http://dx.doi.org/10.13055/ojhmt_3_1_6.121211
  24. Alexander, C.S.  1972.  Cobalt-beer cardiomyopathy.  A clinical and pathologic study of 28 cases.  American Journal of Medicine 53(4):395-417.
  25. Ho, E.N.M., Chan, G.H.M., Wan, T.S.M. et al.  2014.  Controlling the misuse of cobalt in horses. Drug Testing and Analysis.  DOl 10.1 002/dta.1719.
  26. Bartley, P., 2014. Cobalt Chloride taking over from EPO?  Thoroughbred Village.  http://forum.thoroughbredvillage.com.au/cobalt-chloride-taking-over-from-epo_topic48586.html  (accessed June 1, 2014)
  27. Hibbert, D.B.  2014  Cobalt in Equine Urine.  Presented at The 20th International Conference of Racing Analysts and Veterinarians, September 20-27, Mauritius.
  28. Popot, M.A., Ho, E.M.N., Wan T.S.M. et al.  2014 An international collaboration on cobalt for setting up a threshold value.  Presented at The 20th International Conference of Racing Analysts and Veterinarians, September 20-27, Mauritius.