Equine Drugs, Medications, and Performance Altering Substances: Their Performance Effects, Detection, and Regulation

Dr. Thomas Tobin, Dr. Julio Gutierrez, Emily Schwartz, 
Dr. Fernanda Camargo, and Charlie Hughes 
Equine Pharmacology, Therapeutics and Toxicology Laboratory
The Maxwell H. Gluck Equine Research Center
University of Kentucky
Lexington, KY 40546-0099

Dr. Rodney Eisenberg
Frontier Biopharm
6013 Atwood Drive, Suite 300
Richmond, KY 40475
e-mail: rod@frontierbiopharm.com

Dr. Andreas Lehner
Diagnostic Center for Population and Animal Health
College of Veterinary Medicine
Michigan State University
4125 Beaumont Road
Lansing, MI  48910

Mr. Kent Stirling
Florida Horsemen’s Benevolent and Protective Association 
P.O Box 1808
Opa Loca, FL 33055-0808 

Based on a presentation to the Equine Law section
of the Kentucky Bar Association at Keeneland,
Lexington, Kentucky, Oct 21, 2005 
(webpage updated Dec 2010)

Table of Contents

  1   Summary
  2   Background and Definitions
  3   History
  4   Can Drugs or Medications Influence the Outcome of a Race?
  5   The Introduction of ELISA Testing (1988) 
  6   Mass Spectral Confirmation
  7   Liquid Chromatography/Mass Spectrometry/Mass Spectrometry 
  8   "Zero Tolerance" Testing
  9   Numbers of Medication Molecules: Medication Dosing and Elimination
10   Thresholds, Including "No Effect Thresholds" (NETs)
11   Withdrawal Time Guidelines
12   Reference Standards
13   Medication Rules
14   The Current Racing Medication Testing Consortium (RMTC) &
       Association of Racing Commissioners International (ARCI) Rule
15   Further Reading
16   Appendices 

1. Summary

Thoroughbred Racing has been testing for drugs and medications since about 1903.  Today, racehorse testing is by far the longest established, broadest in scope and most sensitive drug testing performed on earth. Racehorse testing is also performed within an extremely stringent regulatory context, and my understanding is that many of our constitutional protections as US citizens are inoperative in the racing environment. Racehorse testing is also remarkably “clean,” as the incidence of deliberate use of performance affecting substances seems to be very small. 

There are good reasons for all of the above. It is empirically clear that medications are highly likely to influence the performance of racing horses, although the scientific evidence for actual improved performance is much less than overwhelming. 

In the mid-nineteen eighties, however, the use of high potency drugs with clear potential to affect performance was not particularly well controlled. Following a directive from the Kentucky State Racing Commission, an interdisciplinary team at the University of Kentucky worked on adapting ELISA testing to racing chemistry; this proprietary technology was at that time a major step towards solving the problem of the abuse of high potency drugs in racing horses, and these tests are now marketed worldwide out of Lexington (www.neogen.com/forensickits.htm)

One of the lessons that came out of ELISA testing is that advances in drug detection/testing are research driven. Once a medication is “called positive”, that is the first “positive” is called and prosecuted, the rate of use of the substance  drops dramatically, to close to zero, but not quite zero; it appears that there are always people ready to try a medication that worked for them, or for a colleague, or  a rival, in the past.

Overall, the rate at which performance altering medication violations are reported in racing is extremely small. For example, from 1995–1999 there were about 3 positives for every 100,000 samples for Association of Racing Commissioners International [ARCI] Class 1 violations after trace level identifications of dietary and environmental substances are eliminated. By far, the most common identifications reported in racing are residual “traces” of well recognized and widely used therapeutic medications, so called “tail-ends” of therapeutic medications , and traces of dietary and environmental substances that also happen to be ARCI substances, for example trace level identifications of caffeine and other substances widely used by humans. 

The ease with which such “traces” of therapeutic medications, dietary and environmental substances can be detected using current testing technology has now clearly led scientists and regulators away from the old “zero tolerance” approach, which many authorities now see as outdated, to defined regulatory limits or “thresholds” for therapeutic medications, endogenous, dietary and environmental substances. 

This situation was driven in large part by ELISA testing, which allows highly sensitive detection of trace amounts (tail ends) of therapeutic medications, environmental and dietary substances. In the nineteen nineties, following another Kentucky Racing Commission directive, the   University of Kentucky program at The Maxwell H. Gluck Equine Research Center pioneered the basic research that underpins the evolving and now in principle very well established concept concerning the use of regulatory “thresholds” in racing regulation.

More recent challenges include developing effective regulatory methods for the newer recombinant hormonal products such as the various human recombinant erythropoietin products and variants thereof and growth hormones. More recently, a high quality ELISA test has been made available for human recombinant erythropoietin and racing chemistry has scored a major scientific breakthrough by developing the first mass spectral confirmation method to detect use of recombinant human erythropoietin (rhEPO) in horses or, indeed, in any species. 

2. Background and Definitions

There are at least 30 million known chemical substances and 4,000 or more prescription medications. Racing regulators in the United States , therefore, divide drugs and medications into two major groups:
The largest group of concern to regulators is  the "performance-enhancing substances", whose identification in a horse is viewed with great regulatory concern. Testing for these substances usually proceeds at the highest level of sensitivity possible; so-called "zero-tolerance" testing. About 900 or so substances are classified by the Association of Racing Commissioners International (ARCI) Uniform Classification System for Foreign Substances as potentially performance enhancing in a five class system, the most complete listing of such substances available anywhere in the world (http://www.arci.com/druglisting.pdf). 
 The second and smaller group comprises the "therapeutic medications", recognized by the American Association of Equine Practitioners [AAEP] and the Racing Medication and Testing Consortium [RMTC]. There are approximately 50 plus of these medications used therapeutically in horses in training (Table 1). Since about the year 2000, it has come to be much more generally accepted that we must set “limitations” on the sensitivity of testing for therapeutic medications. These limitations are variously called thresholds or reporting levels, or decision levels (
California ) apparently depending on the semantic preference of the individual jurisdiction. 

Table 1.  Therapeutic Medications Routinely Used and Identified as Necessary by the Veterinary Advisory Committee — (Racing Medication and Testing Consortium [RMTC] draft list of therapeutic medications, 2005) 

1. Acepromazine 17. Dipyrone  33. Omeprazole 
2. Albuterol 18. Flunixin  34. Pentoxifylline
3. Aminocaproic Acid 19. Fluprednisolone 35. Phenylbutazone
4. Atropine 20. Fluphenazine 36. Phenytoin
5. Beclomethasone 21. Furosemide 37. Prednisolone
6. Betamethasone 22. Glycopyrrolate  38. Prednisone
7. Boldenone 23. Guaifenesin 39. Procaine Penicillin
8. Butorphanol  24. Hydroxyzine 40. Pyrilamine
9. Cimetidine 25. Isoflupredone 41. Ranitidine
10. Clenbuterol 26. Isoxsuprine 42. Reserpine
11. Cromolyn 27. Ketoprofen 43. Stanozolol
12. Dantrolene 28. Lidocaine  44. Testosterone
13. Detomidine  29. Mepivacaine  45. Triamcinolone
14. Dexamethasone 30. Methocarbamol  46. Trichlomethiazide
15. Diazepam 31. Methylprednisolone
16. DMSO 32. Nandrolone 

3. History

Up to about 100 years ago there was little concern about the use of medication in racing horses, and particularly so in North America . The 1800s had seen the purification of cocaine and morphine and availability of these substances in pure form made the acute stimulant medication of racing horses a reality. Around the turn-of-the-century (1890-1910), a number of American trainers went to Europe , taking with them these new “American” medications.  As a group, these trainers were so successful that they became known in European racing circles  as the "Yankee Alchemists."

Figure 1.    Carl Vernet depicts apparently routine 
                      pre-race medication of horses, ca. 1810

In the early 1900s the Honorable Mr. George Lambton (Fig. 2), the sartorially correct and socially prominent leading English trainer of his time, grew tired of losing races to the "Yankee Alchemists," as he also soon grew tired of politely requesting the English Jockey Club to do “something” about the problem. He therefore purchased some of the American "medications," and publicly announced that certain horses in certain races were going to be, well, shall we say "medicated." These activities rapidly gained the Jockey Club's attention, and in 1903 the Jockey Club made the medication of a racing horse an offense against the rules of racing in England.  While the record is silent as to how these medications were to be detected, the prescribed punishment was to be "ruled off the turf," a punishment still in place in parts of the English speaking world.  Hon George Lambton
Figure 2.  
The Honorable George Lambton


    Figure 3.
Somewhat farther from home, an American trainer by the name of Jack Keene was also having a very good run in Russia.  Mr. Keene’s run, however, came to an abrupt halt one day when he was met in the paddock by a Russian racing official, followed by Russian chemist, complete with a basket of frogs. Some saliva was taken from Mr. Keene's horse, and presumably force-fed to the frog, which then reportedly behaved in a most un-frog-like way.  Mr. Keene's horse was duly declared "positive," and  shortly thereafter Mr. Keene left Russia and returned to Kentucky and to his family farm, Keeneland.   

Classic analytical chemistry based race testing as we know it started in France , apparently before 1910.  In 1935, Mr. William Woodward of Woodward Stakes fame, sent a Dr. Catlett, a veterinarian and Dr. Charles Morgan, a chemist, from Florida to France to learn the French drug testing techniques. They returned to Florida and set up the first US drug testing lab; later the New York Racing Commission opened a racing chemistry laboratory on the10th floor of a building on Chambers Street in Manhattan . In 1947 the professional association of racing chemists, the Association of Official Racing Chemists [AORC] was formed. 

Figure 4.   
The late Mr. Robert Vessiny, Truesdail Labs, Tustin, CA, circa 2000.  
Robert Vessiny's professional career began in 1941 at the NY Racing Commission Laboratory on Chambers Street in Manhattan and continued until 2005, covering virtually the entire history of US racehorse testing which started about 1935 in Florida and New York under Dr. Charles Morgan.

4. Can Drugs or Medications Influence 
     the Outcome of a Race?

Drugs and medications can be used to influence the outcome of races in a number of ways. Acute stimulant medication is the administration of a stimulant substance to a horse shortly before post. Among the especially useful agents in this area are the opiates, which have long been used in racing horses, and also the amphetamine-like stimulants, and most especially methylphenidate (Ritalin). All of these substances have been widely used, the opiates likely for hundreds of years, and presumably particularly so when testing for these agents was not available.
        Figure 5.

        Figure 6.     Narcotic drug family dose response curves

As drug testing improved in sensitivity those individuals seeking an “opiate edge” began to use the more potent and thus more difficult to detect opiates. The unnamed but highly potent opiate at the far left of the above family of opiate dose response curves (Fig. 6) is etorphine, or “elephant juice.” Etorphine is one of the most potent opiates known and at the time that this figure was published in “Drugs and the Performance Horse” in 1981, there was no test available that could detect it, which is the reason that this medication is unidentified in this original paper.  This figure also shows, for one family of substances, the 10,000 fold range in dose/potency from the least potent opiate tested on the right, meperidine, at about a one gram/horse dose, to the very highly potent etorphine on the far left, with 50 micrograms (50 millionths of a gram) producing an equivalent pharmacological effect to one gram or more of meperidine. And, of course, etorphine was also, in round figures, about 10,000 time more difficult for the chemist to detect than the old standbys of morphine and heroin, one of whose street names was “horse.” This great increase in the potency of medications being used in horses set the stage for the development of ELISA Testing, as we will discuss later.

As well as being administered stimulants to win races, horses can also be medicated to win by relaxing them and allowing the horse to run its best possible race. The widely used tranquilizer acepromazine, and any number of related or equivalent agents, have reportedly been used in this way. 

Improving a horse’s “wind” by opening its airways through the use of bronchodilators may also improve the performance of horse, and especially a horse that is sub-clinically broncho-constricted. In this regard, the best selling ELISA test at one time was a particularly sensitive and broad scope bronchodilator test, the availability of which test abruptly stopped the less than therapeutic use in racing horses of a bronchodilator called terbutaline. 

 Figure 7.   
 Only triple dead heat on record

 Figure 8.  
 Grindstone wins the 1996 
 Kentucky Derby by a nose 

The difficulty with trying to scientifically demonstrate performance effects of drugs in small numbers of horses is that the drug needs to produce a positive performance effect of about the same magnitude as Secretariat’s win at Belmont (Fig. 9) to meet the lowest level of statistical significance acceptable in science. This is a considerable experimental challenge; another way of looking at this is that successful horse trainers make far more subtle and discriminating judgments than most scientists, of which I think there is no doubt whatsoever. 

Figure 9.  Secretariat wins at Belmont 

Veterinarians certify horses as being sound in "wind and limb." Obviously, medications that can affect these parameters and also the “attitude” or “behavior” of a horse have the potential to affect both the presentation of a horse and also, presumably, the results of the ultimate performance analysis, the outcome of a race.  By the mid-1980s the use of highly potent drugs, and most especially highly potent opiate type medications such as fentanyl (Sublimase) and etorphine, and high potency bronchodilators to improve the racing performance of horses had created  considerable problems for race horse testing and also for racing.

5. The Introduction of ELISA Testing, 1988
In the mid-1980s, race testing was for all practical purposes dependent on a primary screening technique called Thin Layer Chromatography (TLC). This technology has many useful qualities, being inexpensive and fast, but it is not particularly sensitive, and in the mid-1980s some horsemen were reportedly using high potency narcotics, stimulants, bronchodilators and tranquilizers with impunity. In 1985 we were requested (directed?) by the then Kentucky State Racing Commission to "fix this problem." The solution that we developed, ELISA testing for high potency drugs and medications, is in place and widely used around the world today, and is evidenced here in Lexington by a thriving commercial concern, Neogen Corp, on Nandino Boulevard, employing 100 people and bringing in about US $50 million a year into Lexington (not all through ELISA tests – www.neogen.com/forensickits.htm).
The term ELISA is an acronym that stands for Enzyme Linked ImmunoSorbent Assay. Simply put, an ELISA test is a variant on the home pregnancy test technology. It requires a drop of urine; it can be performed relatively rapidly, it is/can be highly sensitive and can be read by eye. When ELISA testing was first introduced, the problem was to keep the technology from "putting down” too many trainers, especially in those jurisdictions that had frozen “back samples.” Let me simply say that this was a turbulent time for me professionally, but matters eventually settled down and, as I indicated, ELISA testing is in many areas the backbone of drug screening worldwide today.

The term ELISA is an acronym that stands for Enyme Linked ImmunoSorbent Assay. Simply put, an ELISA test is a variant on the home pregnancy test technology. It requires a drop of urine; it can be performed relatively rapidly, it is/can be highly sensitive and can be read by eye. When ELISA testing was first introduced, the problem was to keep the technology from "putting down” too many trainers, especially in those jurisdictions that had frozen “back samples”. Let me simply say that this was a turbulent time for me professionally, but matters eventually settled down and, as I indicated, ELISA testing is the backbone of drug screening worldwide today.
Figure 10.    ELISA Testing
This is a 96 well ELISA plate in which the full blue color of an ELISA negative has been developed in most wells. The clear wells on the left hand side are the “positive controls” containing calibration standards. All of the other wells represent ELISA “negative” urine samples. A “track” ELISA positive would show up as a clear well in the middle of the blue samples, in laboratory jargon, a so called “whiteout,” or an ELISA positive.
    Figure 11.    ELISA Test Results

An ELISA test will usually detect about 5 ng/ml (or 5 parts per billion) or less of drug or drug metabolite in the sample. Some tests are 10 fold more sensitive, detecting down to the high parts per trillion. To put these figures in perspective, one part per billion is one second in your life if you are 32 years old. 
To put the matter of testing sensitivity into regulatory perspective, a sure prescription for regulatory friction/problems is a therapeutic medication (or a dietary or environmental substance) given at higher (gram) doses to horses, excreted efficiently in urine, and being tested for by an analyst with a highly sensitive ELISA test with no thresholds/decision levels in place. In the absence of “thresholds” or detection/decision limits in place, a sensitive ELISA test can become basically a hunting tool/license for forensic chemists.  Isoxsuprine, an ARCI class 4 therapeutic medication, administered orally in gram amounts per day, and excreted very efficiently and at very high concentrations in equine urine,  is a classic example. 
Finally, we must always remember that an ELISA test simply binds to and “sees” one side/surface of the medication molecule. Therefore, while an ELISA “negative” is almost certainly a true negative, an ELISA test will, by definition, interact to some extent with  substances other than the drug in question.  As such, the rule with an ELISA “positive” is that it can always be, by definition, a so-called “false positive.” Which is, of course,  why chemists invariably follow ELISA screening with the much more chemically specific technique of Mass Spectral “confirmation."
And again, on the other hand, within the performance limits of the assay, an ELISA "negative" is virtually certainly a true negative. 

6. Mass Spectral Confirmation

While ELISA screening/testing is fast and highly sensitive, it is, as set forth above, far from specific. The second and absolutely critical and essential step in the testing process is confirmatory testing, usually by Mass Spectroscopy. In this step, the molecule is isolated and its precise mass measured, and the molecule is also broken into a series of fragments.  Both the mass and relative proportions of these fragments (the fragmentation pattern) are specific for the given drug, and are then matched with known certified reference standards run through the Mass Spectrometer  in parallel with the test samples.  As such, a full scan mass spectrum, with appropriate matching controls, is the "gold standard" in drug testing, and is considered definitive evidence for the presence of the substance in the sample in question. Additionally racing also almost always allows collection of an independent or "split" or "referee" sample, which the trainer can have analyzed independently [although sometimes more independently than others!] to check for the presence of the substance in question. Independent replication of the primary findings in the "split" or "referee" analysis usually neutralizes any substantial challenges in the area of the substance claimed identified and for some classes of substances, namely therapeutic medications, dietary and environmental substances, may well yield important mitigating or exculpatory information.
  Figure 12.  
  Dr. Lehner and the LC/MS/MS

Figure 13.  

Figure 14.

Comparison of Mass Spectra of a post-race etorphine and an authentic standard. The lower figure shows the mass spectrum of an authentic etorphine laboratory standard. Note the molecular ion at mass 483, the base peak at mass 272 and the various other ions of the standard or control spectrum. Note the very close correspondence of the standard or control mass spectrum with the mass spectrum of the derivatized material recovered from the post-race sample, indicating that the material recovered from the post race sample is indistinguishable from authentic derivatized reference standard etorphine.

7. Liquid Chromatography/Mass Spectrometry/
    Mass Spectrometry  (LC/MS/MS)

More recently an even higher sensitivity and specificity technology has become widely available, namely, Liquid Chromatography/Mass-Spectrometry/Mass-Spectrometry (LC/MS/MS). 

Figure 15. 
atic of MS/MS showing the relationship between the Mass Spectrometers and the parent/precursor and product/daughter ions.
LC/MS/MS technology allows the unequivocal identification and quantification of substances down to low picogram per milliliter, that is, low part per trillion concentrations, or at times even lower concentrations. The technology is called LC because the separation technique, the liquid chromatographic (LC) “feed” into the mass spectrometer, which expands the range of substances that can be chromatographed and therefore analyzed.  The actual analysis takes place in a dual stage Mass Spectrometer (MS/MS), linked by a reaction/fragmentation chamber. The first Mass Spectrometer stage is calibrated/set to identify a specific drug/molecular mass of interest, let us say clenbuterol, which has a molecular mass of 277.19.  This drug molecule/mass then feeds into the reaction/fragmentation chamber, where it is fragmented, and the molecular fragments are then fed into the second Mass Spectrometer stage, where the product/daughter ion fragments are identified.  Figure 16 below shows the fragmentation of clenbuterol in our hands, showing how the 277.19 mass clenbuterol molecule fragments to yield an intermediate of 259 and a final mass of 203,  one of several possible fragmentation sequences as set forth in (Fig. 16).   

Figure 16. 
The entire family of clenbuterol fragments, going from 277->203, 279->205, 277->259, 279->261, 277->57 and 277->57.  As we indicated, since each drug molecule has a specific starting mass, and a specific fragmentation pattern and the relationship between parent and daughter ions can be clearly identified, this is a highly sensitive and highly specific analytical technique.









      Figure 17.

Finally, given the very high sensitivity and specificity of this testing procedure, it can be used to detect parts per trillion or picograms per ML of drug in relatively small sample volumes, such as a one milliliter plasma sample, as set forth in Reference 9 below.
This very high and continually increasing sensitivity of drug testing techniques and instrumentation brings with it the problem of detecting more and more minute traces of drugs, including therapeutic medications used legitimately and appropriately to protect the health and welfare of the horse, traces that are clearly pharmacologically and forensically irrelevant.  As such, the extremely high sensitivity of current analytical techniques has created significant problems, and perhaps has made the so called “zero-tolerance” concept or approach to equine drug testing
irrational and irrelevant when applied to therapeutic medications and dietary and environmental substances.  

8. Zero Tolerance Testing

“Zero Tolerance” testing is not testing down to "Zero" molecules, which no analytical chemist can yet accomplish, but rather testing to the Limit of Detection (LOD) of the best available technology. While this may be an entirely appropriate analytical approach to the regulation of performance altering substances which have no place in racing, it is absolutely not considered appropriate for therapeutic medications. Therapeutic medications are substances used to maintain the health and welfare of horses, and to arbitrarily change the sensitivity of testing for these substances depending on either the whim of the chemist or today's availability of improved testing technologies, is entirely inappropriate, as we will see from review of the following basic mathematics of drug/medication dosing and drug elimination.

9. Numbers of Medication Molecules:
    Medication Dosing and Elimination

When you administer a dose of phenylbutazone to a horse, you administer about the same number of phenylbutazone molecules as there are stars in the known universe, that is about 6 followed by 21 zeros molecules. This is a very large number of molecules indeed. 
    Figure 18.

The horse will eliminate the bulk of this dose of phenylbutazone quite rapidly. If phenylbutazone in the horse has a 7.22 hour half-life, 50% of the drug will be eliminated by 7.22 hours after dosing, 75% by 14.44 hours post dosing, 87.5 by about 21 hours post dosing, and 90% by 24 hours after dosing. At the end of day 1, when 90% of the drug is eliminated, the pharmacological effect of the drug is, for all practical purposes, gone, but there is still present in the horse the not inconsiderable number of 6 followed by 20 zeros worth of phenylbutazone molecules.  Every day another 90% of the drug in the body of the horse will be eliminated, and other zero drops off. 

However,  if the chemist really wants to look, with current technology he or she can easily find traces of the phenylbutazone or its metabolites for 14 days or more after administration, a time post-administration that even the most conservative chemists and regulators generally do not wish to pursue a medication identification. However, the question now arises of when, precisely, should the chemist stop pursuing these traces? Or at what trace concentration should racing regulators cease being concerned?
10. Thresholds, Including "No Effect Thresholds" (NETs)
The answer to this question is simple; the chemist should stop pursuing these traces precisely when he/she is instructed to stop. It is, however, slightly more complicated to determine the exact point at which chemists should be instructed to cease and desist their analytical endeavors.
We approached this question experimentally in the
Maxwell H. Gluck Equine Research Center during the second half of the 1990s. Simply put, we administered decreasing doses of local anesthetics to horses until we saw no detectable local anesthetic effect, which gave us the No Effect Dose for the drug, in this case a local anesthetic. Then we measured the concentrations of the drug, actually a recovered drug metabolite fragment, in the urine, and the concentrations we came up with are, by experimental demonstration, not associated with any pharmacological effect. These concentrations then become “No Effect Thresholds” in urine for the specific therapeutic medication.  These No Effect Threshold [NET] concentrations can then be written into the medication rules, and the chemist can then be advised not to test below these now scientifically defined "no effect" regulatory threshold concentrations. 


  Figures 19-21.      Equine response to heat lamp stimuli

We presented this scenario hypothetically in 1994-95, and then started the actual research on regulatory thresholds. We were immediately vigorously attacked from conservative quarters, at first anonymously and libelously. In 1996 one of these libelous letters surfaced signed by Mrs. Donna Ewing of the Illinois Hooved Animal Humane Society. The University of Kentucky "encouraged me,” shall we say, to sue, which I did.  While I eventually dropped the suit,  its filing  had the desired effect of silencing the complaining parties, who have not been heard from since with regard to racing. More to the scientific point, we completed our ongoing thresholds research and published it in the refereed scientific literature. By the year 2000, the intellectual concept and more importantly, the actual word "thresholds" became more or less "safe" for a courageous racing administrator to allow past his (or her) lips.  Indeedby today  December 2010, the concept of regulation by the use of specified concentration “thresholds” in plasma or in urine is extremely well established, at least in North America

In this regard, is of interest to note that the concept/approach of "zero tolerance" was, to some extent, officially voted out of favor and “off the regulatory island” in one of the opening papers of the 13th International Conference of Racing Analysts and Veterinarians (ICRAV 2000) in  Cambridge, England. In this paper Professor Robert L. Smith addressed the concept of zero tolerance, which he considered a "fading illusion," and reviewed the events "which are increasingly undermining the suitability of this approach."  In his words, "The zero tolerance approach . . . is in essence an illusion in which the magician is the racing chemist." He continued, "The zero tolerance approach is both philosophically and pragmatically unsound. . . . The goal for the future integrity of racing is to develop "reporting values" for therapeutic substances based upon rigorous analysis of their pharmacological and pharmacokinetic properties and using an appropriate model.”

11. Withdrawal Time Guidelines
Let us now move from the theoretical and illusionary concept of "zero tolerance" to practical horsemen’s concerns. A “regulatory threshold” or a “reporting level” is a concentration value (such as, for example 10 parts per billion in urine) that has, in the larger scheme of things, little actual reality for horsemen.  This is because a horseman or, for that matter, a chemist locked out of his/her laboratory, cannot “see” 10 parts per billion of anything in horse urine. What the horseman needs are clear, transparent “withdrawal time guidelines,” i.e., guidelines as to when he/she should stop administering the medication prior to post so that the blood or urine "reading" comes in below the stipulated regulatory threshold, whatever that particular threshold may be. 
Establishing withdrawal time guidelines is considerably more difficult than determining threshold concentrations. The only way to answer the guidelines question is, again, by actual experimental determination, followed by field application. The specific medication product/formulation in question must be defined/specified, and the formulation, dose, route, and duration and number of administrations specified. The medication must be administered to a significant number, hopefully at least 20 or more, Thoroughbred horses in training, and the blood or urinary concentrations of the parent medication or its principal urinary metabolite/analyte followed over a period of time. The laboratory performing the analyses should be appropriately accredited (American Association of Laboratory Accreditation, A2LA), and have in place a validated quantitative analytical method for the threshold substance at concentrations down to concentrations significantly below the lowest concentration of interest  in the experimental model/horses. (http://hbpa.org/resources/MedicationPolicy.pdf). 
The data obtained must then be analyzed statistically, and fitted to a defined mathematical distribution. One can then use this mathematical distribution to advise horsemen that if they administer the drug following X stipulation doses/days, and stop administration at Y hours prior to post, there will be a Z probability of exceeding the regulatory threshold. One of the things that everybody must understand is that if you administer a medication to a horse at any time close to post, there is always a finite mathematical probability of exceeding the threshold; all anybody can do is estimate as accurately as possible the statistical probability of exceeding the regulatory threshold, i.e., of incurring an “overage”, and make sure that the risk of an inadvertent or statistical “overage” is a risk that the horseman can live with. 
This finite probability of a therapeutic medication overage is most likely the reason that regulatory authorities are almost invariably reluctant to be associated with “withdrawal time” guidelines. While a 1/1000 risk of a “positive” may be an entirely acceptable risk for an individual horseman running  a small number of horses, if the authority approves a given “withdrawal time,” it immediately assumes responsibility for all 10-20,000 or more samples tested in the jurisdiction, which increases the probability of a statistical overage 10-20,000 fold, or more if the authority tests more than 20,000 samples per year. 
The first formal scientific approach to this question of determining a regulatory threshold linked to a specific withdrawal time was undertaken at the request of the Kentucky Horsemen's Benevolent and Protective Association [KY HBPA] in the early 1980s.  Responding to a request from President Ed Flint of the Kentucky HBPA, we defined the population distribution characteristics of furosemide in equine plasma after its administration to 49 horses [Chay et al, Ref 11].  This work was soon published, and then picked up by a number of racing states, starting with Oklahoma
in about 1986.  Since then, the regulatory threshold for furosemide, adjusted upwards to 100 ng/mL to allow for “field” variability and linked to a 1.010 urinary specific gravity value, has become the scientific basis for the widely applied four hour furosemide rule in the United States, as set forth in the ARCI medication thresholds/rules.   
At the personal level, given the current state of knowledge, it is at best extremely challenging to provide useful "withdrawal time information" advice. The number of factors which affect the withdrawal time is very large indeed, and in the absence of a defined threshold ("Zero Tolerance" testing) it may be little more that a guessing game.  Whenever I get a “withdrawal time” estimate request, I try to make these uncertainties clear, and to also clearly communicate that any estimates offered are nothing more that my best professional opinion, and my opinion is always qualified with the statement that "there are no guarantees in life, and this caveat most certainly includes “withdrawal time” estimates."
The various factors that can affect "withdrawal times" are set forth in some detail in Appendix #1, Thomas Tobin and Kent H. Stirling (2009) Equine Drug Testing & Therapeutic Medication Regulation; 2009 Policy of the National Horsemen's Benevolent and Protective Association, Inc. Wind Publications, 600 Overbrook Dr, Nicholasville, KY 40356, USA pp 170

12.  Reference Standards
When we administer a medication to a horse [or a human] what the chemist finds in the urine is generally not the drug itself, but a chemically modified [metabolized] form of the drug linked to a sugar molecule, glucuronic acid.  Figure 23 below sets forth this process for lidocaine which is first hydroxylated by the horse to give rise to 3-hydroxylidocaine. The hydroxylated lidocaine is then linked to glucuronic acid to give the final metabolite, 3-hydroxylidocaine glucuronide.  When this resultant highly water-soluble glucuronide metabolite of lidocaine enters the urine, it cannot be re-absorbed by the horse, and the end result is a high urinary concentration of this glucuronide conjugated metabolite of lidocaine, which is removed from the body of the horse the next time the horse urinates.
And as an aside, where the metabolized drug goes after urination can be a matter of some regulatory significance.  If the dose of drug administration to horses large, and the drug/metabolite is excreted at high concentrations in the horse's urine, the horse will contaminate his stall environment. It has been shown that a "clean" horse put into the environmentally contaminated stall can immediately go "positive" for the medication in question, creating an interesting regulatory circumstance, and demonstrating another compelling argument in favor of regulatory thresholds for therapeutic medications.

Figure 23.
Lidocaine is first hydroxylated by the horse to give rise to 3-hydroxylidocaine; this hydroxylated lidocaine is then linked to glucuronic acid to give the final highly water soluble metabolite, 3-hydroxylidocaine glucuronide.  Analyzing this sample, the chemist first hydrolyses off the glucuronic acid, and then recovers the 3-hydroxylidocaine metabolite fragment from the urine sample and quantifies it. To correctly perform this analysis requires a certified reference standard for 3-hydroxylidocaine and a deuterated internal standard in which 10 of the hydrogen atoms in 3-hydroxylidocaine have been replaced with 10 deuterium atoms, as set forth in Appendix II   

To quantify the amount of this 3-hydroxylidocaine glucuronide metabolite in a urine sample, the chemist needs a certified reference standard for 3-hydroxylidocaine, and also what is called a deuterated version of 3-hydroxylidocaine, which serves as an internal or "loss check" reference standard throughout the analytical procedure.  At the time that we started this work there was no source for these reference standards and internal standards, so we began creating a line of certified reference standards and internal standards for equine therapeutic medications.  Creation of these standards is not a trivial process, and the synthetic scheme set forth below is that generated by our colleagues Dr. Rodney Eisenberg and Dr. Julio Gutierrez for the synthesis of the required 3-hydroxylidocaine certified reference standard and internal standard, as set forth in APPENDIX III.

13. Medication Rules

More than forty years ago, when the Kentucky medication rule was first being formulated (even before I came to Lexington)  there were no thresholds or regulatory limits anywhere.  Indeed, there were very few, if any, quantitative analytical methods applied to racing.  Under these circumstances the Kentucky medication rule was clear, simple, effective and highly practical. You could not run your horse on stimulants, depressants, local anesthetics, tranquilizers or narcotic analgesics, the classic performance altering substances. However, the use of substances that were perceived as therapeutic was permitted, with the goal of protecting the health and welfare of the horse. This Kentucky rule was well fitted to the regulatory technology then available, and indeed is, I understand, close to the rule currently applied in human athletics. At that time, this very practical rule had been in place in Kentucky for at least 30 years and, to the best of my knowledge, served the horses and horsemen of Kentucky well.   

14. The Current Racing Medication Testing 
      Consortium (RMTC) & Association of Racing
      Commissioners International (ARCI) Rule

As of December 2010 the following is a summary of the RMTC/ARCI model rules referring to thresholds for therapeutic medications. Additionally, one must keep in mind that at any given time individual states may have in place thresholds that at times differ from and/or extend the RMTC/ARCI model rules.  Many of these individual state thresholds are presented in the National HBPA booklet, Appendix #1, but given the rate of change of medication rules, for definitive information with respect to any individual state, the most up to date version of the individual state’ medication rules should always be consulted.  

The current Association of Racing Commissioners International model rule includes the thresholds presented below (Table 2.).  Again, for definitive information the Association of Racing Commissioners International should be consulted.  See the ARCI website  http://


Phenylbutazone - 2 micrograms per milliliter; 
Flunixin -  20 nanograms per milliliter; 
Ketoprofen -  10 nanograms per milliliter. 
Phenylbutazone [subthreshold] - 0.5 micrograms per milliliter;


Furosemide  - 100  nanograms per milliliter, urinary specific gravity < 1,010


Cimetidine  - up to 24 hours before post 
Omeprazole - up to 24 hours before post 
Ranitidine up to 24 hours before post 


Caffeine  -  100 ng of caffeine per ml in plasma or serum      


a) 16β-hydroxystanozolol (metabolite of stanozolol (Winstrol)) - 1 ng/ml in urine for all horses regardless of sex; 

(b) Boldenone (Equipoise® is the undecylenate ester of boldenone) in male horses other than geldings - 15 ng/ml in urine. No boldenone shall be permitted in geldings or female horses. 

(c) Nandrolone (Durabolin® is the phenylpropionate ester and Deca-Durabolin® is the decanoate ester) 

(A) In geldings - 1 ng/ml in urine 

(B) In fillies and mares - 1 ng/ml in urine 

(d) Testosterone 

(A) In geldings - 20 ng/ml in urine 

(B) In fillies and mares - 55 ng/ml in urine 


The threshold for TCO2 is 37.0 millimoles per liter of plasma/serum or a base excess level of 10.0 millimoles, and; 

The decision level to be used for the regulation of TCO2 is 37.0 millimoles per liter of plasma/serum plus the measurement uncertainty of the laboratory analyzing the sample, or a base excess level of 10.4 millimoles per liter of plasma/serum. 

 Table 2.      RMTC/ARCI model threshold rules 

Finally, as of this writing, the RMTC appears committed to developing regulatory thresholds in plasma for all of the therapeutic medications listed in Table 1.   This is a significant regulatory advance based on recent technological and regulatory developments set
forth in this medication regulation overview.  
13. Further Reading

1/ www.thomastobin.com 

2/ Thomas Tobin.   Drugs and the Performance Horse by Thomas Tobin, 463 pages, Charles C. Thomas, Springfield, Illinois, 1981.

Thomas Tobin and Kent H. Stirling.  Equine Drug Testing & Therapeutic Medication Regulation: 
2009 Policy of the National Horsemen's Benevolent and Protective Association, Inc.
,  Wind Publications, Nicholasville
, Kentucky, 2009,  170 pages.

4/ Tobin T, Mundy GD,
Stanley SD, Sams RA, Crone D (eds): Testing for Therapeutic Medications, Environmental and Dietary Substances in Racing Horses, Proceedings of Workshop, Lexington, KY, 220 pages, 1995. [KY Ag Exp Sta #95-14-058]

5/ The Association of Racing Commissioners International (ARCI) Uniform Classification System for Foreign Substances:   http://www.arci.com/druglisting.pdf

6/ Tobin T, Harkins JD, Sams RA: Testing for therapeutic medications: Analytical/ pharmacological relationships and the need for “limitations” on the sensitivity of testing for certain agents.  J Vet Pharm Therap, 22:220-233. 1999. [KY Ag Exp Sta #98-14-134].

7/ Smith R.L “The zero tolerance approach to doping control in horse racing: a fading illusion.” Proceedings of the 13th International Conference of Racing Analysts and Veterinarians (ICRAV)
Cambridge , United Kingdom , p 9-14, 2000. 

8/ Tobin T, Watt DS, Kwiatkowski S, Tai HH, Blake JW, McDonald J, Prange CA, Wie S. Non-isotopic immunoassay drug tests in racing horses: a review of their application to pre- and post-race testing, drug quantitation, and human drug testing. Res Commun Chem Pathol Pharmacol. 1988 Dec;62(3):371-95.

9/ Neogen ELISA tests: www.neogen.com/forensickits.htm

10/ Lehner A.F.; Harkins J.D.; Karpiesiuk W.; Woods W.E.; Robinson N.E.; Dirikolu L.; Fisher M.; Tobin T; .Clenbuterol in the Horse: Confirmation and Quantitation of Serum Clenbuterol by LC/MS/MS after Oral and Intratracheal Administration; Journal of Analytical Toxicology, Volume 25, Number 4, May/June 2001 , pp. 280-287(8)

Chay S, Woods WE, Rowse K, Nugent TE, Blake JW, Tobin T. 1983The pharmacology of furosemide in the horse. V. Pharmacokinetics and blood levels of furosemide after intravenous administration Drug Metab Dispos. May-Jun;11(3):226-31.

12/ Thomas Tobin, et al.  Furosemide in the Horse.  Wind Publications, Nicholasville, Kentucky, 2000.

Appendix #1

Table of Contents from Equine Drug Testing & Therapeutic Medication Regulation: 
2009 Policy of the National Horsemen's Benevolent and Protective Association, Inc.

by Thomas Tobin and Kent H. Stirling
Wind Publications, Nicholasville
, KY 40356 , USA  pp 170


Appendix #2


The synthesis of 3-hydroxylidocaine started with commercially available 2,6-dimethylaniline 1, which required 4 steps to produce 3-hydroxylidocaine.

Scheme 1:     Synthesis of 3-hydroxylidocaine.

Synthesis of 3-hydroxylidocaine-d10 starts with the same starting material 1, above, and  then follows steps similar to those used to produce 3-hydroxylidocaine..

Scheme 2:     Synthesis of 3-hydroxylidocaine-d10.


Appendix #3


Julio Gutierrez, Wojtek Karpiesiuk, Gabrielle Herrensmith, Elizabeth Armstrong, 
Charlie Hughes, Job Tharappel, and Thomas Tobin.
Maxwell H. Gluck Equine Research Center
Department of Veterinary Science, 
University of Kentucky , Lexington , KY 40546

Rodney Eisenberg. 
Frontier Biopharm, 
P.O. Box 614 , Richmond , KY 40476

Brent Mayer and Emilie Stanley.  
Neogen Corp., 
Nandino Boulevard , Lexington , KY 40511  

The equine medication metabolite standards and stable isotope reference standards listed here are being synthesized by a team of scientists, starting with Dr. Wojciech Karpiesiuk in the Gluck Equine Research Center, more recently in cooperation with Dr. Rodney Eisenberg and more recently again with Dr. Julio Gutierrez and Ms. Gabrielle Herrensmith, whose synthetic methods have been guided by the analytical skills of Dr. Andreas Lehner, Dr. Job Tharappel and Mr. Charlie Hughes. These syntheses have been an ongoing project, starting in the mid-nineties with the need to synthesize reference standards for unique equine drug metabolites and metabolite fragments. More recently, there has been an increasing need for high-quality certified reference standards and stable isotope internal standards for use in quantitative equine medication regulation, or more simply, to allow the regulatory application of thresholds for therapeutic medications.
University of Kentucky and Frontier Biopharm, in cooperation with Neogen Corp., are working to provide the horse race testing community with certified quantitative analytical reference standards for equine therapeutic medications that will satisfy the ISO-17025 and ISO-34 requirements for quantitative measurements, and related reference materials which are increasingly required for equine forensic testing. These reference standards will meet industry requirements for chemical identity, spectroscopic and chemical purity, and the levels of residual volatile solvents, water and inorganic, in accordance with accepted ISO analytical standards.
Synthesis of these standards has been made possible by ongoing research support from a large number of Horsemen’s Benevolent and Protective Association’s, starting with the Florida HBPA in the mid-90s, and since then including the National Horsemen’s Benevolent and Protective Association and the Alabama, Arizona, Arkansas, Canada, Charles Town West Virginia, Florida, Iowa, Indiana, Kentucky, Louisiana, Michigan, Minnesota, Nebraska, Ohio, Oklahoma, Ontario Canada, Oregon, Pennsylvania, Tampa Bay Downs Florida, Texas, Washington State, and West Virginia Horsemen’s Benevolent and Protective Associations, the Kentucky Equine Drug Research Council, and the Kentucky Racing Commission, the Kentucky Horse Racing Authority and the Kentucky Science and Engineering Foundation, Grant agreement KSEF 148-502-05-160 with the Kentucky Science and Technology Corporation, and this support is gratefully acknowledged, along with the ongoing support of the Director and Faculty of the Maxwell H. Gluck Equine Research Center and the University of Kentucky College of Agriculture. 


Standards Available or in Synthesis


Parent Medication

Regulatory Analyte or 
Deuterated Internal Standard



2-(1-hydroxyethyl) promazine sulfoxide (D4)
(deuterated metabolite)



2-(1-hydroxyethyl) promazine sulfoxide






Clenbuterol (D9 )
(deuterated standard)






Carboxydetomidine (D4)
(deuterated metabolite)






Hydroxydetomidine (D4)
(deuterated metabolite)



Furosemide (D5)
(deuterated standard)



Flunixin (D3)
(deuterated standard)



Guafenesin (D4)
(deuterated standard )



Ketoprofen (D3)
(deuterated standard )












Methocarbamol (D4)
(deuterated standard )


Modafinil acid

Modafinil acid



Phenylbutazone (D9 )
(deuterated standard)



Procaine (D10)
(deuterated  standard )






Promethazine sulfoxide




Click here to see Certificates of Analysis as a PDF document.    
Metabolites are filed under the name of the parent medication.        zA thru M    z N thru Z  
Wait a moment for PDF to load after clicking.
If you do not have the Acrobat PDF Reader you may download Acrobat here.


Also Synthesized or Synthesis in Progress


Parent Medication

Regulatory Analyte or
Deuterated Internal Standard



Acepromazine Sulfoxide 



N-2,4-Dimethylphenyl-N’-methyl-formamidine (D6)
(deuterated standard)






(metabolite )



2-(2-)4-Amino-3,5- dicholorophenyl)-2-hydroxyethylamino)-2-methyl-propan-1-ol


Colterol and Bitolterol










3-hydroxylidocaine (D10)
(deuterated standard )






3-Hydroxymepivacaine (D3)
(deuterated standard )






2-(1-Hydroxypropyl)promazine sulfoxide



Pyrilamine (D4)
(deuterated standard)














Appendix #4

Association of Racing Commissioners International “Drug Positives”

ARCI Drug Postive Rulings From 8/1/2004 Thru 8/1/2005




National Horsemen’s Benevolent and Protective Association, Inc. Proposed National Policy on Drug Testing and Therapeutic Medication. J Eq Vet Sci 23(1): 4-5, 18-40, 2003. http://hbpa.org/resources/MedicationPolicy.pdf 

Published as Kentucky Agricultural Experiment Station Article #_____ with the approval of the Dean and Director, College of Agriculture and the Kentucky Agriculture Experiment Station.

Publication #359 From the Equine Pharmacology, Therapeutics and Toxicology Program of the Maxwell H. Gluck Equine Research Center, University of Kentucky, Lexington, KY 40546-0099.

Supported by the National, Alabama, Arizona, Arkansas, Charles Town WV, Florida, Iowa, Kentucky, Louisiana, Michigan, Minnesota, Nebraska, Ohio, Oklahoma, Oregon, Pennsylvania, Tampa Bay Downs, Texas, Washington State, West Virginia, Ontario, and Canada Horsemen's Benevolent and Protective Associations.