Background
Frusemide is a high ceiling diuretic commonly used in the management of exercise induced pulmonary haemorrhage in racehorses. Despite its long-standing use, there is little reliable data on its urinary excretion which is directly applicable to drug testing in Australia. Betamethasone and triamcinolone acetonide are corticosteroid drugs with profound anti-inflammatory properties and are widely used in the treatment of racing injuries. In the last few years, they have been increasingly administered intraarticularly (IA) rather than by intravenous (IV) or intramuscular (IM) routes. There is little or no information on the excretion and detection of these drugs after IA administration and hence there is no data upon which veterinarians can base informed decisions on the use of these drugs.

Methods
Robust and properly validated detection methodology has been applied to the measurement of these three drugs in blood and urine. Methods based on gas chromatography mass spectrometry and liquid chromatography mass spectrometry have been developed to ensure that the data generated are as definitive and reliable as possible and that experimental protocols are aligned with those being used to carry out related studies in other countries. This ensures that data will be transferable and enables the range of drugs for which reliable excretion data are available to be greatly extended.

Results
It has been shown that after IA administration, betamethasone may be detected for up to three days in urine, albeit at very low concentrations. There is no evidence of any significant retention of the drug in synovial fluid and hence it is unlikely that its systemic persistence and excretion in urine is prolonged by this route of administration. This is also supported by the noted suppression of endogenous hydrocortisone production which is of only two-three days duration.

The study on frusemide has shown that this drug may be reliably detected up to 48 hours in blood and urine. Data collected on the volume of urine voided and urine specific gravity may be interpreted to suggest that the diuretic effect would not extend beyond that period. Therefore, a 48 hour post administration detection time can be recommended with a high level of confidence, given that intersubject variability was relatively small.

The study of triamcinolone acetonide administered by IA injection has demonstrated that this drug and its metabolite can be detected for up to 130 (5.5 days) and 170 (7 days) hours respectively, provided that methods of appropriate sensitivity are applied. The detection times are roughly equivalent to the time for which endogenous hydrocortisone is suppressed suggesting that they may also correlate with the duration of action of the drug. Inter-subject variability was relatively small with respect to the excretion of the drug and its metabolite after 60 hours. However, there was considerable variation in the rate at which endogenous hydrocortisone levels came back to pre-administration values.

Implications
This data can now be used by veterinarians to make more informed decisions about the treatment of horses, particularly in the weeks or days leading up to a race. The data will also assist the racing authorities to draw conclusion about purported treatment regimens in the event that one of these drugs is detected in a pre- or post-race sample.

Recommendations
We suggest that studies of this type continue so that similar relevant data can be generated for a realistically wide range of drugs commonly used for therapeutic purposes in race and competitive horses.