Page Title

How credible is "The Black Box"?
6/3/08 By Kenneth W. Terpenning for KyHarnessRacing.com

In a previous article titled “The Black Box”, we covered its

purpose, its use, what the legal level of TCO2 measured by

the box is in a few jurisdictions and the penalties for those

involved in Kentucky and their horses that test positive for

pre-race testing.

The following information has come to this reporter’s

attention and I thought it is important enough to share

with all of you.

I have conducted many hours of research for this series of

articles including my attempts to contact racing officials in

Ohio, Kentucky, Pennsylvania, New York, New Jersey and

Indiana. In addition, I have attempted to contact multiple

trainers for comments on the issue.

Also

, I contacted Radiometer America, the company that

makes the blood gas analyzers

urrently being used in Ohio

and Kentu

cky. I have also read studies conducted by

Australia and s

everal colleges in the United States

regarding the influ

ence of baking soda on a horse’s

performance.

According to a letter I received from John A. Izzo, Deputy

Director and Legal Counsel for the Ohio State Racing

Commission dated May 22, 2008, Mr. Izzo states, “the

machine is commonly referred to as a radiometer and it is

my understanding that it is quite different from the “Black

Box” that was being used several years ago.”

So let’s call it a radiometer.

The “radiometer” was first utilized in harness racing to

catch those trainers using sodium bicarbonate to enhance

a horse’s performance. As history has shown, certain

trainers are deliberately giving their horses’ baking soda,

called milkshaking, for this very reason. Many trainers also

treat their horses with feed supplements containing baking

soda.

A controlled study completed in 2005 by the Department of

Large Animal Clinical Sciences (DLACS) studied 5,028 starts

made by 2,349 thoroughbreds racing in California trained

by 287 different trainers at 2 different racetracks. The

individuals involved in testing carefully monitored the

horses studied to insure they did not receive any baking

soda. The study showed that a majority of the horses that

finished in the top three positions of each of those races

observed maintained higher than normal pre-race TCO2

levels than those not finishing as well. Factors found to

significantly alter a horse’s TCO2 level pre-race ranged from

sex, class, distance, lasix administration, and cloudy

weather conditions. Hold on…something as simple as

clouds can alter your horses TCO2? That’s a new one on

me.

According to Mr. Izzo’s letter, it is never mentioned in the

DLACS’ study if the horses tested pre-race ever showed a

TCO2 level of 37 millimoles per liter or higher. It is also

referenced in his letter that “expert testimony at

administrative hearings in Ohio indicates that there is less

than a one in a million chance that a horse will have a

natural TCO2 level of greater than 37 millimoles per liter of

blood.”

Recently, the thoroughbred industry has undergone a

study to test the effects of baking soda on a horse’s

performance. Their mission is to prove baking soda truly

does not affect a horse’s performance as much as originally

thought and that the industry should stop testing for it. The

study information could not be found and it is not known if

there have been any results thus far.

When asked about the study, neither Marc Guilfoil, Deputy

Executive Director of the Kentucky Horse Racing Authority,

nor Mr. Izzo from Ohio was aware of its existence.

Lisa Underwood of the Kentucky Horse Racing Authority

failed to respond to any of my emails regarding this subject

about "The Black Box", including the question about this

study.

The Australian Harness Racing Council, an organization that

has well-researched this topic, only 38% of all horses that

carried a TCO2 level of over 36 millimoles per liter were

found to have had improved performance and many

horsemen back in the infancy of the use of baking soda

were well over-dosing their horses with baking soda trying

to get better performanc

e out of their horse.

A study by the AHRC proved that most trainers did not

know exactly what level of baking soda was correct

because of the horses either being over-trained or under-

trained and that can affect how much TCO2 they carry in

their blood. The AHRC also says a margin of error ranging

from plus

minus

1.2 to 1.4 millimoles per liter on any

given test exists meaning if your horse tests at 38.2

millimoles per liter and the level set by the racing

commission is 37 then you are handed a warning or

suspension and fine (depending on what number violation

this is for you as a trainer) when in fact your horse is within

the margin of error and should be exempt from penalty.

Questions have been brought up over the years about who

does the test, at what temperature the test is done, and

whether or not the sample or the equipment had been

tainted by other samples or products handled by the test

giver. Then there is also the matter of calibration of the

equipment. Like a police officer’s radar gun, the radiometer

must be calibrated on a regular basis to guarantee its

accuracy.

Mr. Izzo, from Ohio,

states, “Radiometers are at a secure

location at each track. To ensure that there was no

tampering with the machine since last used, as well as the

accuracy of the calculations, our operators run tests on the

radiometers prior to each day’s use to ensure that the

machine is running properly.” In Kentucky, Marc Guilfoil

responded that the radiometer’s are locked up in the test

barns at the tracks running and kept in the KHRA office

during non-race months. What does the company say the

margin of error is for their machines?

Radiometer America was asked about the margin of error

with their blood gas analyzers. In an emailed response

from a representative there, the company does not

calculate a margin of error on their machines because it is

incalculable due to pre-analytical errors that could occur like

mishandling of the samples taken from the horses. Errors

that could occur prior to the samples being tested are

sample mix-ups, dilution of the samples with other fluids or

water, the positioning of the needle when taking the

sample and whether or not a vein is accidentally punctured

as opposed to an artery, air bubbles, clotting, and even

variation in time passing between when the sample is

taken and when it is tested.

Mr. Izzo states, “I agree that there are reasons other than

baking soda use that may result in a horse having a high

TCO2 level. The Commssion’s goal is to prevent horses from

running in races with an artificially high TCO2 level.

Regardless of why the TCO2 level is high, a horse with an

artificially high TCO2 level has an unfair advantage over the

other horses in a race.”

zzo

continues,

“One of the experts the Commission uses

tells me that the measurement uncertainty (not the margin

of error) has recently been reassessed by the Australian

laboratories. Their new estimate of the measurement of

uncertainty is .6 millimoles per liter of blood. When the

Commission was testing TCO2 levels at the Analytical

Toxicology Laboratory on the Beckman ELISE, all results

were reported to the Commission with a ninety-five percent

confidence level. This means that scientific data showed

that a TCO2 level was at the level we reported, and that

there was a two and one half percent chance that the level

was even higher. We have continued with this confidence

level on the radiometers.”

Apparently there are many things that can go wrong with

the testing and the final results of TCO2 levels of pre-race

tests.

Unfortunately, there are too many factors and not enough

evidence one way or another to determine if the

radiometers are truly helpful to the sport.

Also

, let’s look at the different rules regarding the TCO2

legal levels from various jurisdictions. New Jersey and

Illinois allow a horse’s TCO2 to test at 37 millimoles per liter

when not on lasix and 39 millimoles per liter if on lasix.

Kentucky and Ohio take a stand on 37 millimoles per liter

regardless if on lasix or not. Pennsylvania, Indiana, and

New York failed to respond to my emails and phone calls

with regard to their rules. Does lasix cause a higher TCO2

level? The DLACS study says yes, New Jersey and Illinois

allow a leeway in the results for it, but Kentucky and Ohio

do not.

Many questions remain about baking soda, its effects,

radiometers and their accuracy and the accuracy of the

testing procedures and handling of the samples. It appears

there is room for errors, but that is like most things in life,

never a 100% guarantee. Why do we not have rule

uniformity from jurisdiction to jurisdiction, from state to

state with regard to testing, test results, penalties, and so

on?

With even the slightest of doubts cast on the testing and

its results and the possibility of errors, is the radiometer a

machine that would hold up in a court of law if challenged?

We let murderers off for reasonable doubt, what about

horsemen and women? Are they committing murder when a

horse tests high in TCO2? (We already know there are

other reasons a horse can test high for TCO2 aside from

baking soda.)

If the industry is going to test, why not find a machine that

not only tests for TCO2 levels but finds and determines the

cause for the high TCO2 before we penalize the trainer or

the horse. Ohio says it doesn’t matter why a horse tests

high…is that fair to everyone? Perhaps you feel the testing

is fair and nothing should change. That’s okay…but why?