Since the May Special General Meeting on this topic Ken Dyer has researched from the available AHRC papers and documentation in summary form the key questions and answers forming Council’s earlier considerations on establishing the relevant threshold level of 35mmol/l.

As you are aware Ken, as former AHRC Chief Executive, was also the Secretary to Council’s Working Party in this area. Click here to read the results of his research together with Council’s current Rules on Prohibited Substances. This paper was presented to the Annual General Meeting on 25^th and 26^th August 2000.

It is provided for information.

Rod Pollock
Chief Executive

30^th August 2000

TCO₂ Questions and Answers

INTRODUCTION

• In a natural habitat a horse forages for grasses, hay etc which contain elements of sodium, bicarbonate and many other substances. One combination gives a horse an endogenous level of sodium bicarbonate and as a final simplified result thereof a horse has a natural level of TCO2 when tested. This natural level can be anywhere from 26/27mmol/l to 31/32mmol/l.

• When taken in hand to an environment where the horse is fed with commercially collected grasses (hay, lucerne, oats etc) plus additives such as electrolytes and prepared pellets which may contain additional elements etc, the level of TC02 consolidates from about 27/28mmol/l to 32/33mmol/l. Most horses have a set range between which their TC02 levels float, however, the extreme range of 33mmol/l is seldom exceeded.

• These TC02 levels will not change from the normal range unless the diet of the horse is changed either deliberately or accidentally.

• Certain proprietory brands of products containing sodium bicarbonate, citrates, other substances will have the effect of increasing the pH level in the blood.

• In an exercise, equitage, endurance or racing environment the level of pH in the blood is most crucial in counteracting the build up of lactic acid in the body of a horse as a result of exercise. If the build up of lactic acid can be controlled by what is called buffering or alkinizing substances then the horse will go longer at the peak rate of endurance or speed. The same applies to human athletes in middle distance running, rowing etc. In sprint racing the build up of lactic acid is not so obvious and has little effect.

• Added to this equation is the fact that a horse can be undertrained, scientifically trained, or overtrained. A horse that is scientifically trained or well trained cannot be improved very much by the administration of alkinizing agents, whereas over or undertrained horses seem to be the horses which are the greatest effected by an increase of pH in their blood.

• In a racing environment a horseman would have to be a great judge of his horse and training methods to know whether an administration of a buffering substance or alkinizing agent would beneficially effect the horse.

• In North American tests in 1996 it was found that some horses performed worse when administered with a buffering substance or alkinizing agent.  These buffering or alkinizing agents are known under general terminology as "MILKSHAKES".

• In Australia when "milkshakes" were first used some horsemen erroneously thought that the greater the quantity the greater effect. Some horses tested were shown to have levels in excess of 50mmol/l.

The following questions and answers give reasons why authorities have determined arbitrary levels and why they arrived at the level of 35mmol/l as the best means of controlling the use of buffering substances or alkinizing agents whether sodium bicarbonate, citrates or other buffering-alkinizing substances. They are written in general terms so that specifics may require further elaboration if necessary. Further questions and answers will be added in due course.

A "milkshake" is a colloquial term used to describe a concoction comprising sodium bicarbonate (usually up to about 500 gm), and glucose often in combination with other additives. Citrates and other non-bicarbonate alkanizising agents are interchangeable with bicarbonates as buffering substances. The eventual results of non-bicarbonate compounds are similar to bicarbonate since they all result in elevated blood pH, bicarbonate and TC02. When mixed together the ingredients develop a white frothy appearance. Milkshakes are usually administered by tube as a stomach drench.

Milkshakes affect the performance of a horse because they can cause profound physiologic changes. The premise of milkshaking is based on the lowering of the deleterious effects of lactic acid by raising the pH of the body. The intent of the user is to alter/manipulate the performance of a horse. The basic purpose for the use of this product is to enhance performance through the effect of bicarbonate on the general body metabolism. Bicarbonate has an effect of countering the lactic acid naturally produced by horses during high speed performance or exercise. It has been found in a number of studies that a high concentration of lactic acid in the blood and muscle is correlated with fatigue. Administration of a buffer (ie bicarbonate, citrates etc) helps to negate the effect of lactic acid produced and stops the decrease in pH.

Many trainers believe the administration of sodium bicarbonate enhances an animal’s performance, conversely the reverse applies thereby the ability to manipulate form.

The bottom line for racing authorities using blood gas testing is that they are really testing for all alkalizing substances with similar effects, not just bicarbonate based agents, and blood gas machines can detect the effects of mutiple alkalizing substances without differentiation. Likewise, a horseman that gives a non-bicarbonate based substance in adequate quantities to cause significant alkalosis may not legitimately claim to have not given a "milk shake" .

Administration of alkalizing agents is detected by measuring the total carbon dioxide(TC02) concentration in venous blood samples, rather than directly measuring the bicarbonate concentration. The TC02 concentration is the usual measure of bicarbonate concentration in samples analized by an autoanalyzer.

The bicarbonate concentration is a measure of the bicarbonate ion (HC03), while TC02 concentration includes both bicarbonate and free carbon dioxide in the plasma.

Bicarbonate of soda and citrates are listed in the main pharmacopoeias which describe their preparation, properties, usage, and side affects. Tris buffers and citrates are also listed. The current Australian Rules of Harness Racing prescribe the total carbon dioxide (TC02) at a level of 35.0 millimoles per litre in plasma and alkalising substances found in a horse that has tested higher than a level of 35mmol/l are defined as drugs.

There is ample anecdotal evidence that milkshaking was occurring prior to 1987, but detection and governance procedures did not occur in harness racing until around 1989 despite many complaints from the industry at large over a period of a few years that some trainers were using illicit preparations of unknown combination of substances.

Milkshaking in harness racing was detected by autoanalizer in1989 in Ontario, Illinois, New Jersey, and Florida and about the same time in New Zealand and Australia.

There is anecdotal evidence also that bicarbonate of soda has been surreptitiously used over many years in human athletics particularly in medium and long distance running events and in cycling and rowing.

In 1990 it was reported there was convincing anecdotal evidence that bicarbonate administration would improve 400 metre human running times by about 7 to 10%, however, under controlled conditions official tests were carried out on rowers at Launceston, Tasmania by the State Institute of Technology and the results were reported in the Australian Journal of Science and Medicine in Sport. The results indicated that in 6 minutes of rowing, the rowers that had been given an administration of sodium bicarbonate were able to row 48 metres further than those rowers on placebos.

Prof.  Irvine in a 1991 paper presented to the World Trotting Conference in Auckland reported "Bicarbonate dosing has been used successfully to counteract the effects of strenuous exercise in animals and man according to papers published in scientific journals for the last 60 years..." he goes on to say "several of the older papers gave inconsistent results between horses and I wasn’t convinced of its usefulness in postponing fatigue. In most papers about 30% performed better, 50% showed no change and 20% were upset by treatment and performed worse...." and later "Some research that I have done may provide the explanation for this inconclusive result.  Firstly the amount of bicarbonate has to be sufficient to counteract the excess acid produced by exercise which is around 600 grams. The average dose of sodium bicarbonate in the British Veterinary Codex is 60 grams. In the research papers the highest dose given was 135 grams per horse. The amount of sodium bicarbonate used in milkshakes is 300 grams which is a very large quantity compared with usual medical doses. The doses of around 100 grams used in most experiments would show little elevation of blood bicarbonate and neutralises only a fraction of the 600 grams of lactic acid produced so it isn’t surprising that those experiments showed inconsistent effects on performance although there was still a marked improvement in some horses,".

In another report published in the Australian Standardbred (Nov 91) and headed "Benefits of sodium bicarbonate on racing standardbreds" it is stated "the February edition of the AEVA’s Veterinarian...describes the effect of sodium bicarbonate given just prior to racing of standardbreds. Twenty two standardbreds were paired and participated in a crossover trial when competing in two races at least a week apart. They were treated with 300mg/kg NaHco3 orally 2 1/2 hours before racing or with a placebo of salt and dextrose. The treatments were switched for the second race. Blood tests and lactate were measured before and after racing.

Results: Racing times were a mean 1.1 seconds faster in those treated with NaHC03 and blood pH was significantly elevated .Notably, post exercise lactate clearance rates were significantly enhanced in treated horses."

In a study by Lawrence (cited by Dr Phillip Swann in his book "Performance Drugs In Sport") on thoroughbreds racing one mile the results showed that milkshakes improved the performance of racehorses by 1.6 to 5.2 seconds with an average of 2.8 seconds. However, the control group of horse in the same experiment who were dosed with a placebo of sugar,salt, corn syrup and water also showed an improvement between 0.6 to 2.4 seconds an average of 1.1 seconds. It was concluded that that milkshakes elevated blood pH and tended to improve the performance of 8 of the horses but to reduce performance of 6 of the horses. Further on Dr Swann indicates that it appears that bicarbonate drenches will have minor effects on scientifically trained horses and they can effect (either beneficially or adversely) traditionally trained horses. In the same chapter Dr Swann canvasses the effect (if any) of bicarbonate on overtrained horses.

Initially the authorities in Australia did not envisage the extent to which trainers were milkshaking their horses. Given free rein trainers had capitalised on the inaction of authorities until at last the complaints of those trainers not milkshaking their horses became so deafening and officials became so concerned that hastily arranged surprise testings were commissioned at harness racing meetings and revealed the enormous headway milkshaking had made into the harness racing industry.

The last straw so far as officials of Controlling Bodies were concerned was the surprise testing of all horses racing at the Harold Park meetings on 26th July, 1991 and again 9^th August 1991. These tests conclusively showed that trainers were ignoring the then current rules in the absence of testing which in reality was caused by the lack of suitable equipment and stronger rules to conduct and governance the tests and results. At this time the equipment necessary to test was only available at hospitals and medical clinics and the equipment thereof was not necessarily calibrated uniformly.

The testing at Harold Park was commissioned by the Harness Racing Authority of NSW and was carried out by the University of Sydney under Prof. Reuben Rose and David Lloyd on the basis that horses with a venous blood bicarbonate concentration below the level of 35mmol/l were considered to be within the normal range. Horses between 35.1mmol/l and 36mmol/l were considered to be suspicious but were included in the normal range in this particular survey. Horses above 36 were thought to have been dosed with sodium bicarbonate. In hindsight Prof.Rose and David Lloyd got the levels nearly right at their first attempt! The results were startling as 50.6% of the 79 horses tested on the first night proved to be positive (ie above 36mmol/l.). The mean blood bicarbonate concentration for all horses tested was 37.6mmol/l (+- 4.9).After the experience of the first night it was decided to establish whether the first night’s testing was an aberration- as it turned out there was only a marginal difference.

Overall, the testing of all horses on the two nights of racing gave authorities the unique experience of being able to test performance under race conditions of one group of horses which were racing under their normal conditions against another group which quite obviously had been milkshaked by their trainers.

Given the fact that strict governance is now in place the 1991 window of opportunity is not likely to be repeated.

The following chart clearly indicates the effect of milkshakes in a racing situation and the "edge" that a milkshaked group of horses has over horses racing in their normal condition.

Not all milkshaked horses performed above the form of non-milkshaked horses nevertheless the milkshaked horses performed better even in the lower ranked placings (ie there were more milkshaked horses in the better placings and less in the lower placings). The chart shows the placings of horses under the current governed level of 35mmol/l and the placings of horses under the 1991 Rose/Lloyd test level of 36mmol/l.

Re-examining the 1991 statistics under current conditions where the maximum level is 35mmol/l, the testing over two nights reveals that 47% of the total 162 horses tested were negative (ie under 35mmol/l); 32% were 35mmol/l and over whereas another 21% were above the level of 40mmol/l. The highest reading was 50.7mmol/l. Thirteen winners of the eighteen races conducted were above the current level of 35mmol/l. One horse won with a level of 46.2mmol/l.

Clearly, the great majority of trainers thought that milkshakes helped their horses overcome the fatigue problem. Equally clearly, trainers were still experimenting with the dosage levels but had no idea whether their horses were overtrained or undertrained which has a plus or minus effect.

As result of the above testing and somewhat similar results in other States, the AHRC and its Members in November, 1991, banned stomach tubing on the day of the race. However, this was not enough and much more stringent rules were later introduced.

A later series of experiments were carried out at the Equine Performance and Drug Evaluation Laboratory, Dept of Veterinary Clinical Sciences, University of Sydney, later in 1991 to determine the feasibility of pre-race blood sampling and post race analysis of the blood, to detect sodium bicarbonate, the principal component of the so-called "milkshakes".

The Main Findings were:

• Use of autoanalyzers to measure the total carbon dioxide concentration in the blood is an accurate method for detecting sodium bicarbonate administration.

- A range of 26.3mmol/l to 35.2mmol/l assumes that 1 normal horse in 1000 will fall outside this range.

- A range of 25.4mmol/l to 36.1mmol/l assumes that 1 normal horse in 10,000 will fall outside this range. This is obviously a more conservative range.

- A range of from 24.6mmol/l to 36.9mmol/l assumes that 1 normal horse in 100,000 will fall outside this range. This is an exceptionally conservative range.

There were other studies carried out around this time as follows:-

These data were combined to give a mean TC02 concentration of 30.8mmol/l for horses in full training in their stables.

There was no doubt whatever that the use of milkshakes in 1990-1991 discontented the large proportion of those in the harness racing industry. Many industry participants at the time felt they had to use milkshakes just to be competitive with those who had to win at any price- while those who chose to remain within the drug free rules raced at a severe disadvantage. All saw an opportunity to "move up" their horses without fear of being caught.

The main effects briefly are:

• MANIPULATION OF FORM

Should a trainer know that his horse can be beneficially or adversely affected by milkshakes administration then obviously the trainer can pick and choose when to use milkshakes to best effect. Obviously the public is not a party to this information.

• THE WELFARE OF THE HORSE

Usage of milkshakes has detrimental effects on the horse. Scouring seems to be the worst feature of overuse from which some horses take a long time to recover. Some horses never recover and are useless forevermore as racehorses. In addition it is a matter of record that horses have died from drowning through incorrect insertion of the tubing. One of the earliest officially recorded deaths by drowning-where the tube instead of resting in the stomach was poked through the lung wall unknowingly, occurred at the Macau Raceway in 1981. The RSPCA, the media, the public, and the vast majority of those engaged in the harness racing industry are totally opposed to the abuse of horses through the usage of drugs. Many horseman, however, are not so supportive of the duty of care.

• HIDES THE USE OF OTHER DRUGS

For a period of time many believed -on anecdotal grass roots evidence and some academic thought- that milkshakes would hide/mask the use/detection of other drugs but it could not be proven. Anecdotal evidence eventually became reality in the Meadow Mirage case when after many months of testing, analysts found the drug etorphine("a positive but light reaction") hidden behind an administration of bicarbonate.(See Appeal Judge JG Garvey’s summary of this case for further information.).

Palfium was also found in other cases in conjunction with milkshakes-this substance, however, was easily found by analysts.

"The administration of alkaline substances such as sodium bicarbonate can effect the excretion of some drugs. For instance some basic drugs may be excreted in lower concentrations and for a longer duration whilst some acidic drugs may be excreted more rapidly".

• DISTORTION OF PERFORMANCES FOR STUD BOOK PURPOSES

The widespread use of milkshakes and the subsequent improvement in time performances distorted Stud Book comparisons between horses which had raced in a drug free environment and those that had been surreptitiously "moved up".

Initially, The Controlling Bodies used Hospital and Medical Clinic equipment that was nearby, however, not all towns had this service available and consequently no blood samples were taken for pre-race testing. Later the Controlling Bodies purchased autoanalisers of various brand names for mobile testing around all tracks. Since 1992 autoanalisers have undergone many modifications each upgrade making the equipment more sensitive.

The equipment now used is still similar to that used in Hospitals and Medical Clinics for human patients and are calibrated to a certain uniform standard.

The Beckman instrument has been the favoured test equipment in Australia and overseas but like all other equipment (ie radar speed guns) requires constant certification that the equipment is calibrated properly and is working accurately otherwise without regular checking "it leaves the door open" to technical disputes.

The blood sample tests analysed by Rose/Lloyd in 1991 showed that 95 standardbred horses in training but at rest had a reading of between 26.8mmol/l to the highest at 34.3mmol/l. From this they were able to deduce "that 1 normal horse in 100,000 will fall outside this range" but this was qualified by the remark that "this was a exceptionally conservative range"-see Para 8 for full quote.

On 16/8/93 the WATA tabled at an AHRC/DLC meeting statistics it had collated from tests conducted in WA for the period 27/11/92 to 30/6/93. Of 4200 samples taken on Raceways the corrected average was 29.28mmol/l. Furthermore only 7 samples were over the level of 35mmol/l for the period 27/11/92 to 27/7/93 for 4829 samples.

Other State Controlling Bodies at this period of time had similar test sample levels on a greater number of test samples.

Currently, the TC02 test levels of every winner are progressively published in WATA’s monthly magazine. These listings show levels of between 28mmol/l to 33mmol/l. Very rarely does a horse exceed 33mmol/l and it should be remembered that these tests were done on the racetrack and not at rest in the stables or at pasture.

It should also be remembered that the laboratories allow +- 1.2 to 1.4 so that the proclaimed level is a net level of 35mmol/l whereas the horse in reality would test at 36.2mmol/l or 36.4mmol/l according to the State.

The following graph as published in NZ Harness Racing Weekly on 5/7/2000 is a summary of HRNZ’s blood bicarbonate testing since February 1991, when testing was introduced.

The simple answer is No. This question was first discussed by AHRC in 1990-1991 when it decided that its policy and that of its Members would be completely in line with the policy expressed by the Australian Conference of Principal Clubs.

That policy was expressed to the Australian Trainers Association in a letter dated 6/3/89 as follows:

There is no rule or policy that indicates that the industry must be advised of any new procedures. The rules provide for drug free racing and participants use drugs on their horses at their own risk in a racing situation.

• In November the authorities in Canada required that horsemen bring their engaged horses to a public detention barn 4 ½ hours before the race in which the horses were engaged. This procedure was in response to the fact that milkshakes to be effective must be given around 1-2 hours prior to the race.

• In NZ authorities began testing from 1/2/91 but did not set a level.

• On the 9/11/91 the AHRC following representations from its constituent bodies introduced a rule banning stomach drenching on the day of the race in which the horse was engaged.

• On 1/12/91 Harness Racing Victoria began testing.

• During 1991 the WA Trotting Association adopted rules banning the use of milkshakes but did not establish a level.

• On 1/1/92 the NSW Harness Racing Authority promulgated rules to apply a level of 37mmol/l as from that date.

• In November 1992 following a loss in the Chambers case the WATA introduced a level of 35mmol/l based on testing levels of all horses racing at Gloucester Park and other tracks in Western Australia.

• On 30/4/93 the AHRC adopted a level of 35mmol/l which it recommended to its Members to commence from 1/7/93. Tasmania and South Australia were the last two States to adopt the 35 level in 1994.

The level of 35mmol/l was established on the basis that all Controlling Bodies had many thousands of test results which conclusively showed that levels for individual horses on racetracks were between 29mmol/l to 32mmol/l. Higher levels were indications that such horses had suspiciously higher intakes of bicarbonate in their feed either given accidentally or deliberately, still higher levels indicated that horses had been milkshaked with sodium bicarbonate or other buffering agents.

It was also based on the fact that a horse having a normal level of from 29mmol/l to 32mmol/l would have to be administered a substantial amount (ie around 300/600 g)of sodium bicarbonate or buffering agent to be effective in a race. The substantial amount would be enough to send the TC02 level over 35mmol/l and perhaps 36mmol/l or 37mmol/l. However, a level of 36mmol/l or 37mmol/l would ensure that a high percentage of milkshaked horses would be missed. This is because the levels of 35mmol/l, 36mmol/l or 37mmol/l are net levels as all tests are given a +- rating of 1.2 or 1.4 for the uncertainty factor. Thus the gross levels would be 36.2mmol/l for 35mmol/l, 37.2mmol/l for 36mmol/l and 38.2mmol/l for 37mmol/l.

There appears to be no doubt that feed high in alkalising agents can effect the TC02 levels in a horse.

Feed incorporating supplements, additives, electrolytes etc can increase to a small extent the normal level of TC02 in a horse, whilst lucerne and other hay, grasses etc may also increase the normal level. Pelletised feed may include some or all of these ingredients.

Whilst commercially collected or manufactured feed may increase the TC02 level there are many studies of TC02 levels of racing horses at rest in stables which show that the range still does alter significantly from 29-33mmol/l except in some isolated cases.

It is fair to say that the TC02 level of the average horse has increased overall over the past few years but the range is still within acceptable boundaries and there is still a considerable buffer for horsemen to adequately feed their horses with modern feedstuffs.

Horsemen with horses on a natural low level are probably in a safer position than a horse with a level in the higher range of normal.

According to Dr. Kevin Kline (US) no matter whether a small or large dose of an alkalising agent is administered the rate of increase in TC02 is the same for the first two horses.

Sodium is found in sodium bicarbonate, sodium acetate, sodium citrate,sodium phosphate etc-all these substances are generally included in the term "milkshakes".

Sodium does not have any relationship to pH or bicarbonate or to exercise.

Sodium levels, unlike TC02 do not change and continue to rise for a period of three hours after administration regardless of whether the horse was exercised or not. However, there may be other considerations for the rise in sodium levels and therefore a high sodium level is not necessarily proof of bicarbonate administration or other alkalising administration. Nevertheless a high sodium level is an alert or factor to be taken into account.

Further references - CLICK HERE to go to:

AUSTRALIAN HARNESS RACING RULES, PART 12
PROHIBITED SUBSTANCES