Colonel said:
His TT start time was 7:57am and with the half life being 4 hours he could of taken it that night before bed expecting it to be out his system by morning of the TT should he podium and be tested. (+-9am) and not expecting a pre-race test. Chances were it wasnt "in and out" as normal and hence the trace amount in that sample but once the race was over it was out resulting in no positive for the remaining tests.
If he was doping with it, he would have had a detectable level after as well as before the TT. And presumably of some other substance he was using it in conjuction with.
I finally found the
study I was looking for. This paper gives the renal clearance for probenecid as 0.6 – 0.8 ml/min. This refers to the volume of blood that contains the amount of drug appearing in the urine per unit of time. Knowing this, and the peak concentration of drug in the blood, one can calculate how much of a dose of probenecid appears in the urine per time.
For an oral dose of 250 mg, about 2 – 2.7 mg appears in the urine over a period of four hours, which corresponds to about the amount of urine that would be provided in one sample. This is about 1% of the dose. If the dose is increased, the % found in the urine in any given time period increases slightly, which means that if the dose were decreased below 250 mg, the % would be decreased. So 1% can be regarded as a maximum, in that for lower doses, such as contaminants, the % should be lower.
But if we use the 1% figure, we can apply it to Impey as follows: the minimum detectable dose seems to be about 0.05 ug/ml. In a typical 200 ml urine sample, this corresponds to a total of 10 ug. If this is 1% or less of the amount ingested, this means the latter must be at least 1 mg.
This is very high for a contaminant. In the article fmk linked to, the highest contaminant dose observed was about 200 ug/capsule, and even assuming this amount was on Impey’s capsules, some of it would be lost during handling. And I repeat, I’m using the most conservative estimates. The minimum sensitivity demanded by WADA is just 0.25 ug/ml, and for all we know, the amount Impey tested for may have been well above the minimum. And the % of the ingested amount in the urine of such a low dose would be expected to be less than 1%.
I really wonder, based on the similar Cielo case, if Impey’s experts may have simply demonstrated that his capsules were contaminated with probenecid, without showing that the amount present could have accounted for his positive. SAIDS might have felt that if any contamination were present, then one had to assume that was the cause of the positive. This is the way contaminated supplement cases generally work. I can understand the logic, except that one also wonders how easy it would have been to contaminate the capsules intentionally. One would have to know something about the drug to get the contamination level right—remember, this was the argument used by Ashenden against LA wrt the ’99 EPO samples—but knowing about the Cielo case, Impey could have just dusted the capsules and hoped for the best. In fact, if it turns out his level is too high to be accounted for by contamination, that would further suggest that he might have done that.
Hopefully, though, more details will eventually emerge. I hope they actually tested the idea with a known clean set of capsules. Beyond Impey's innocence or guilt, this should be done so that we can evaluate the seriousness of this problem for possible future cases.
Edit: I was able to access the first 3 pages, and one figure, in the above link. It confirmed, in a more quantitative way, what I had picked up from other studies--that very little unmetabolized probenecid appears in the urine. Most of the substance is excreted as probenecid glucuronide. If this is tested for, rather than probenecid itself, the sensitivity of the test will be greatly increased, and the amount needed to be ingested to get a level of 0.05 ug/ml can be reduced to about 40-50 ug. I've been uncertain about whether this is what is assayed in the test, because most studies I've seen refer to probenecid, not the glucuronide. But to maximize the sensitivity of the test, it would only make sense to assay the latter.