Phenylbutazone blood and urine concentrations, pharmacokinetics, and effects on biomarkers of inflammation in horses following intravenous and oral administration of clinical doses.

Phenylbutazone (PBZ) is a potent mon-steroidal anti-inflammatory drug used commonly in performance horses. The objectives of the current study were to describe blood and urine concentrations and the pharmacokinetics of PBZ and its metabolites following intravenous (IV) and oral administration and to describe the duration of pharmacodynamic effect. To that end, 17 horses received an IV administration and 18 horses an oral administration of 2 g of PBZ. Blood and urine samples were collected prior to and for up to 96 hours post drug administration. Whole blood samples were collected at various time points and challenged with lipopolysaccharide or calcium ionophore to induce ex vivo synthesis of eicosanoids. Concentrations of PBZ and eicosanoids were measured using liquid chromatography-tandem mass spectrometry (LC-MS/MS) and non-compartmental pharmacokinetic analysis performed on concentration data from IV and oral administration. Serum concentrations of PBZ and its metabolites were below the limit of quantitation at 96 hours post administration. The volume of distribution at steady state, systemic clearance, and terminal half-life was 0.194 ± 0.019 L/kg, 23.9 ± 4.48 mL/h/kg, and 10.9 ± 5.32 hours, respectively. The terminal half-life following oral administration was 13.4 ± 3.01 (paste) and 15.1 ± 3.96 hours (tablets). Stimulation of PBZ treated whole blood with lipopolysaccharide and calcium ionophore resulted in an inhibition of TXB2 , PGE2 , LTB4 and 15-HETE production for a prolonged period of time post drug administration. The results of this study suggest that PBZ has a prolonged anti-inflammatory following IV or oral administration of 2 g to horses.

Role of laboratory in the management of phenylbutazone poisoning.

We report a rare case of intentional overdose of phenylbutazone in a 15-yr-old female. The patient exhibited symptoms of phenylbutazone toxicity and the presence of the drug was confirmed by gas chromatography mass-spectrometry (GC-MS) analysis of the initial urine sample. The patient underwent plasmapheresis to remove the drug from the circulation. Semiquantitation of sequential serum samples by GC-MS revealed elimination of phenylbutazone by day 5 of admission at which time the plasmapheresis was discontinued. Elevated blood urea nitrogen (BUN) and creatinine returned to normal. Analysis of biomarkers for liver necrosis and regeneration in sequential serum samples revealed the restoration of normal liver function by day 5. This case further confirms our previous observations that biomarkers for liver necrosis and regeneration can predict the outcome of patients with liver damage due to toxins.

Pharmacokinetic/pharmacodynamic approach to assess irrelevant plasma or urine drug concentrations in postcompetition samples for drug control in the horse.

The current performance of analytical techniques used for drug control in horses lead the Regulatory Authorities to decide whether trace levels of drugs legitimately used for therapeutic medication should or should not be reported. Here, we propose a well-ordered and nonexperimental pharmacokinetic/pharmacodynamic approach for the determination of irrelevant drug plasma (IPC) and urine concentrations (IUC). The published plasma clearance is used to transform an effective (marketed) dose into an effective concentration (EPC). EPC is transformed into an IPC by applying a safety factor (SF). This method is based on several assumptions (eg, drug effects reversibly driven by plasma concentration, linearity of drug disposition). The suitability of the computed IPC and IUC can be checked by calculating the residual amount of drug at IPC and computing a minimal drug withdrawal time. It is concluded that controlling the drug effect (using drug or any analyte concentration as a marker) rather than the drug exposure will be more demanding and also makes urine a less than ideal matrix.

Study of the solid-phase extraction of diclofenac sodium, indomethacin and phenylbutazone for their analysis in human urine by liquid chromatography.

A selective semi-automated solid-phase extraction (SPE) of the non-steroidal anti-inflammatory drugs diclofenac sodium, indomethacin and phenylbutazone from urine prior to high-performance liquid chromatography was investigated. The drugs were recovered from urine buffered at pH 5.0 using C18 Bond-Elut cartridges as solid sorbent material and mixtures of methanol-aqueous buffer or acetonitrile-aqueous buffer as washing and elution solvents. The extracts were chromatographed on a reversed-phase ODS column using 10 mM acetate buffer (pH 4.0)-acetonitrile (58:42, v/v) as the mobile phase, and the effluent from the column was monitored at 210 nm with ultraviolet detection. Absolute recoveries of the anti-inflammatory drugs within the range 0.02-1.0 microg/ml were about 85% for diclofenac and indomethacin, and 50% for phenylbutazone without any interference from endogenous compounds of the urine. The within-day and between-day repeatabilities were in all cases less than 5% and 10%, respectively. Limits of detection were 0.007 microg/ml for diclofenac sodium and indomethacin and 0.035 microg/ml for phenylbutazone, whereas limits of quantitation were 0.02 microg/ml for diclofenac and indomethacin and 0.1 microg/ml for phenylbutazone.

Pharmacokinetics of intravenous and intragastric cimetidine in horses. I. Effects of intravenous cimetidine on pharmacokinetics of intravenous phenylbutazone.

Cimetidine was administered intravenously and by the intragastric route to six mares at a dose of 4.0 mg/kg of body weight (bw). Specific and sensitive high performance liquid chromatographic methods for the determination of cimetidine in horse plasma and urine and cimetidine sulfoxide in urine are described. Plasma cimetidine concentration vs. time data were analysed by non-linear least squares regression analysis to determine pharmacokinetic parameter estimates. The median (range) plasma clearance (Cl) was 8.20 (4.96-10.2) mL/min.kg of body weight, that of the steady-state volume of distribution (Vdss) was 0.771 (0.521-1.15) L/kg bw, and that of the terminal elimination half-life (t1/2 beta) was 92.4 (70.6-125) minutes. The median (range) renal clearance of cimetidine was 4.08 (2.19-6.23) mL/min.kg bw or 55.4 (36.3-81.8)% of the corresponding plasma clearance. Cimetidine sulfoxide was excreted in urine and its urinary excretion through 8 h accounted for 12.0 (9.8-16.6)% of the plasma clearance of cimetidine. The median (range) extent of intragastric bioavailability was 14.4 (6.82-21.8)% and the maximum plasma concentration after intragastric administration was 0.31 (0.24-0.50) microgram/mL. Intravenous cimetidine had no effect on the disposition of intravenous phenylbutazone or its metabolites except that the maximum plasma concentration of gamma-hydroxyphenylbutazone was less after cimetidine treatment.

The effect of inflammation on the disposition of phenylbutazone in thoroughbred horses.

The effect of inflammation on the disposition of phenylbutazone (PBZ) was investigated in Thoroughbred horses. An initial study (n = 5) in which PBZ (8.8 mg/kg) was injected intravenously twice, 5 weeks apart, suggested that the administration of PBZ would not affect the plasma kinetics of a subsequent dose. Two other groups of horses were given PBZ at either 8.8 mg/kg (n = 5) or 4.4 mg/kg (n = 4). Soft tissue inflammation was then induced by the injection of Freud's adjuvant and the administration of PBZ was repeated at a dose level equivalent to, but five weeks later than, the initial dose. Inflammation did not appear to affect the plasma kinetics or the urinary excretion of PBZ and its metabolites, oxyphenbutazone (OPBZ) or hydroxyphenylbutazone (OHPBZ) when PBZ was administered at 8.8 mg/kg. However, small but significant increases (P < 0.05) in total body clearance (CLB; 29.2 +/- 3.9 vs. 43.8 +/- 8.1 mL/ h.kg) and the volume of distribution, calculated by area (Vd(area); 0.18+/- 0.05 vs. 0.25 +/- 0.03 L/kg) or at steady-state (Vd(SS); 0.17 +/- 0.04 vs. 0.25 +/- 0.03 L/ kg), were obtained in horses after adjuvant injection, compared to controls, when PBZ was administered at 4.4 mg/kg which corresponded to relatively higher tissues concentrations and lower plasma concentrations (calculated) at the time of maximum peripheral PBZ concentration. Soft tissue inflammation also induced a significantly (P < 0.05) higher amount of OPBZ in the urine 18 h after PBZ administration but the total urinary excretion of analytes over 48 h was unchanged. These results have possible implications regarding the administration of PBZ to the horse close to race-day.

Determination of phenylbutazone and oxyphenbutazone in plasma and urine samples of horses by high-performance liquid chromatography and gas chromatography-mass spectrometry.

A method is described for the qualitative and quantitative determination of phenylbutazone and oxyphenbutazone in horse urine and plasma samples viewing antidoping control. A horse was administered intravenously with 3 g of phenylbutazone. For the qualitative determination, a screening by HPLC was performed after acidic extraction of the urine samples and the confirmation process was realized by GC-MS. Using the proposed method it was possible to detect phenylbutazone and oxyphenbutazone in urine for up to 48 and 120 h, respectively. For the quantitation of these drugs the plasma was deproteinized with acetonitrile and 20 microliters were injected directly into the HPLC system equipped with a UV detector and LiChrospher RP-18 column. The mobile phase used was 0.01 M acetic acid in methanol (45:55, v/v). The limit of detection was 0.5 microgram/ml for phenylbutazone and oxyphenbutazone and the limit of quantitation was 1.0 microgram/ml for both drugs. Using the proposed method it was possible to quantify phenylbutazone up to 30 h and oxyphenbutazone up to 39 h after administration.

Disposition and urinary excretion of phenylbutazone in normal and febrile greyhounds.

Five days after the induction of acute systemic inflammation in greyhounds by intramuscular and subcutaneous injections of Freund's adjuvant, the hepatic concentrations of cytochromes P-450 and b5, the activities of the hepatic microsomal enzymes aniline p-hydroxylase and aminopyrine n-demethylase and the disposition and urinary excretion of phenylbutazone were determined. The mean plasma concentrations of phenylbutazone after intravenous administration were described by the bi-exponential equations: Cp = 144.2e-34.6t + 171.5e-0.104t for five normal greyhounds and Cp = 113.6e-16.13t + 163.1e-0.108t for five febrile greyhounds. The elimination half-lives, total body clearances and apparent volumes of distribution were 6.7 hours, 18.4 ml kg-1 hour-1 and 0.18 litre kg-1, for the normal greyhounds, and 6.4 hours, 19.5 ml kg-1 hour-1 and 0.18 litre kg-1, for the febrile greyhounds. There were no significant differences between the pharmacokinetic parameters describing the distribution and elimination of phenylbutazone, or between the quantities of phenylbutazone, oxyphenbutazone and hydroxyphenylbutazone excreted in the urine. In the febrile greyhounds, there were significant decreases in the hepatic microsomal concentrations of cytochromes P-450 and b5 and in the activities of aniline p-hydroxylase and aminopyrine n-demethylase.

Phenylbutazone in racing greyhounds: plasma and urinary residues 24 and 48 hours after a single intravenous administration.

The concentrations of phenylbutazone (PBZ), oxyphenbutazone (OPBZ) and gammahydroxyphenylbutazone (OHPBZ) in plasma and urine from 50 Greyhounds 24 and 48 h after the intravenous administration of a single dose of PBZ (30 mg/kg) were measured. The 24 h plasma concentrations of OPBZ and OHPBZ, the 48 h plasma concentration of OHPBZ and the 24 h urinary concentration of PBZ were normally distributed, while log transformations were required before the 24 h plasma concentration of PBZ and the 24 and 48 h urinary concentrations of OPBZ and OHPBZ became normally distributed. The 95%, 99%, 99.9% and 99.99% upper predicted confidence intervals for both 24 h and 48 h plasma and urinary concentrations demonstrated wide potential variation in the concentration of the analytes should PBZ be administered to Greyhounds. The 24 h plasma and urinary concentrations of PBZ were weakly correlated, but no similar relationship existed for OPBZ or OHPBZ. The urinary concentrations of each analyte were not affected by the trainer or sex of the Greyhound or the urinary pH. We conclude that it would be impossible to predict the timing of the PBZ administration or the plasma concentration of PBZ from the measurement of the concentration of PBZ in a single sample of urine.

Simultaneous analysis of flunixin, naproxen, ethacrynic acid, indomethacin, phenylbutazone, mefenamic acid and thiosalicylic acid in plasma and urine by high-performance liquid chromatography and gas chromatography-mass spectrometry.

Simple and reproducible high-performance liquid chromatographic (HPLC) and gas chromatographic-mass spectrometric (GC-MS) methods have been developed for the simultaneous analysis of several acidic drugs in horse plasma and urine. Although the capillary GC-MS column provided better separation of the drugs than the reversed-phase C8 (3 microns, 75 mm) HPLC column, the total analysis time with HPLC was shorter than the total analysis time with GC-MS. The HPLC system equipped with a diode-array detector provided simultaneous screening (limit of detection 100-500 ng/ml) and confirmation (limit 1.0 micrograms/ml) of the drugs. The HPLC system equipped with fixed-wavelength ultraviolet and fluorescence detectors provided a relatively sensitive screening [limit of detection 50-150 ng/ml for ultraviolet and 10 ng/ml for fluorescence (naproxen only) detectors] of the drugs. However, the positive samples had to be confirmed by using either the diode-array detector or the GC-MS system. The GC-MS system provided simultaneous screening and confirmation of the drugs at very low concentrations (20-50 ng/ml).

Disposition of human drug preparations in the horse. I. Rectally administered indomethacin.

A high-performance liquid chromatographic method to measure urinary indomethacin levels is described. In 0.5 ml urine, 1 micrograms/ml of indomethacin could be detected. Alkaline hydrolysis of urine resulted in the decomposition of indomethacin. When two suppositories of Indocid corresponding to 200 mg indomethacin were administered rectally to four horses the drug was rapidly absorbed and remained detectable in urine from 1 to 12 h. The excretion rate peaked after 2-3 h while the maximal concentration ranged from 18.5 to 80.6 micrograms/ml. Only 8 to 16% of the indomethacin dose was eliminated in urine after 12 h. A fraction of the dose was excreted as the glucuronide conjugate.

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.

We have introduced large scale non-isotopic immunoassay testing into pre- and post-race drug testing in racehorses. The technologies utilized are Particle Concentration Fluorescence Immuno Assay (PCFIA) and the one-step Enzyme Linked Immuno Sorbent Assay (ELISA). These technologies are rapid, inexpensive, and highly effective. On introduction into post-race testing in the Western United States, these ELISA tests exposed several previously undetected patterns of drug abuse. The drugs detected were buprenorphine, oxymorphone, mazindol, sufentanil and cocaine. This led to the suspension of a large number of trainers and exposed the high false negative rate of thin layer chromatography (TLC) based testing. More recently, we have introduced both PCFIA and ELISA assays into pre- and post-race testing in Illinois. Within days, our pre-race PCFIA tests detected signs of acepromazine abuse. Directed searches of post-race urines from these horses showed evidence for acepromazine metabolites in the urine of these horses. Examination of frozen samples from associated horses yielded about 70 ELISA "positives" for acepromazine. To date, about 25 of these ELISA "positives" have been confirmed by mass spectrometry. We have also raised antibodies to phenylbutazone and furosemide to enable rapid and inexpensive quantitation of these permitted medications. Furosemide is a particular problem since its use requires a pre-race detention barn. For furosemide, we have developed a regulatory schedule based on our immunoassay test that allows elimination of the detention barn. For phenylbutazone, we have developed a similar immunoassay that allows rapid and inexpensive quantitation of this drug in blood. To enable racing authorities to test jockeys and other racetrack personnel, we have adapted PCFIA technology to human drug testing, and a full range of very sensitive tests for human drugs of abuse is available. These immunoassays are sufficiently sensitive to control abuse of the most potent drugs available to horsemen. In principle, an immunoassay can be raised to any drug within about six months, and made available worldwide at competitive rates. It appears clear that these non-isotopic immunoassays provide racing with the only technological basis that is sufficiently sensitive to detect the most potent abused drugs pre- and post-race, and has the flexibility to be readily adaptable to different drugs. Because of the high false negative rate generated by TLC, credible pre- and post-race testing programs cannot be based on TLC alone.(ABSTRACT TRUNCATED AT 400 WORDS)

Evaluation of the potential for interference by dimethyl sulfoxide (DMSO) in drug detection in racing animals.

Dimethyl sulfoxide (DMSO) had been postulated to be a 'masking agent' when used concurrently with therapeutic or prohibited drugs in racing animals. Eight drugs (flunixin, furosemide, caffeine, apomorphine, phenylbutazone, lidocaine, cocaine, and acepromazine maleate) were administered to six horses singly and with concurrent intravenous DMSO. Urine samples were analyzed for the presence of the drugs and/or their metabolites by thin layer chromatography. Direct comparison of thin layer chromatograms of extracts of positive urine samples with and without DMSO verified that DMSO did not interfere with the detection of these drugs.

Metabolism, excretion, pharmacokinetics and tissue residues of phenylbutazone in the horse.

The pharmacokinetics, metabolism, excretion and tissue residues of phenylbutazone (PBZ) in the horse were studied following both intravenous and oral administration of the drug at a dose rate of 4.4 mg/kg. A 72-hour blood sampling schedule failed to demonstrate a third exponential phase; the plasma disposition following intravenous injection being described by a two compartment open model, with the following elimination phase parameters: beta = 0.13h-1, t1/2 beta = 5.46h, Vdarea = 0.141 1/kg and C1B = 17.9 ml/kg/h. The hydroxylated metabolites oxyphenbutazone (OPBZ) and gamma-hydroxyphenylbutazone (OHPBZ) were present in detectable concentrations in plasma for 72 and 24 h, respectively. After 36 h OPBZ concentrations exceeded plasma PBZ concentrations. In urine the principal metabolites were OPBZ and OHPBZ but smaller concentrations of another compound, probably gamma-hydroxyoxyphenbutazone (OHOPBZ), were also detected. The percentages of the administered dose recovered from urine were 30.7, 39.0 and 40.3 after 24, 48 and 72 h from the time of injection. Recovery of PBZ and its metabolites from urine was significantly reduced in the first 24 h after oral dosing when the horses had free access to hay, probably as a result of markedly delayed absorption, but this did not occur in animals deprived of food for a few hours before and after dosing. Determination of approximate values of urine/plasma (U/P) concentration ratios for PBZ and its metabolites relative to endogenous creatinine U/P concentration ratio suggested that PBZ was filtered in small amounts only because of the high degree of plasma protein binding and then excreted by diffusion trapping in the alkaline urine. Much higher U/P ratios were obtained for the hydroxylated derivatives, and one at least (OHPBZ) was secreted into urine.

Efficacy of testing for illegal medication in horses.

The efficacy of testing for illegal drugs in race horses was surveyed by evaluating 27 questionnaires received from 28 racing jurisdictions polled. Large variations in the number of samples tested and drugs detected were reported. Some jurisdictions reported only illegal medications, whereas others also reported permitted medications. To facilitate comparison, stimulants, depressants, local anesthetics, narcotic analgesics, and tranquilizers were classified as hard drugs. Other drugs, which are legal in some jurisdictions, were classified as soft. To evaluate the efficacy of testing, positive test results were compared for hard drugs only. Positive test results varied from zero in some jurisdictions for some years to 14.8/1,000 samples tested for one small jurisdiction in one year. The mean rates over the years 1975 to 1983 varied from 0.2 to 6.5/1,000, with a modal positive test result of about 1/1,000. Beside the fact that prerace blood testing is less effective than is postrace urine testing, no cause for these variations in the positive test results could be identified. The positive test results also were compared for jurisdictions with differing medication rules for phenylbutazone (PBZ). Jurisdictions that did not allow PBZ had a mean positive test result for hard drugs of about 1.3 +/- 0.9/1,000 samples tested. Jurisdictions that allowed more liberal use of PBZ had a mean positive test result for hard drugs of about 1.3 +/- 1.0/1,000 samples tested. Seemingly, the presence of PBZ in equine forensic samples did not reduce the ability of forensic laboratories to detect the use of hard or illegal drugs.

Effect of diuresis on urinary excretion and creatinine clearance in the horse.

Endogenous creatinine clearance and renal excretion of phenylbutazone, osmotically active material, and compounds contributing to the urinary refractive index were studied in 12 Thoroughbred mares after no treatment, after water administration, or after furosemide administration. Urine was quantitatively collected, using urinary bladder catheters. On average, urine flow of the mares was 9 microliters/min/kg without treatment and increased to about 50 microliters/min/kg after water administration and to about 70 microliters/min/kg after furosemide administration. Water administration increased creatinine clearance values and excretion of phenylbutazone. Furosemide administration increased urinary excretion of osmotically active compounds and compounds contributing to urinary refractive index and decreased excretion of phenylbutazone.

Effects of phenylbutazone and oxyphenbutazone on basic drug detection in high performance thin layer chromatographic systems.

Interference or 'masking' in thin layer chromatography occurs when the presence of one drug on a thin layer plate physically obscures or interferes with the detection of another drug. We investigated the ability of phenylbutazone and oxyphenbutazone to mask or interfere with the detection by high performance thin layer chromatography (HPTLC) of basic drugs used illegally in horse racing. Of fifty-five basic drugs called 'positive' since 1981 by laboratories affiliated with the Association of Official Racing Chemists (AORC), forty did not comigrate with phenylbutazone or oxyphenbutazone and could not, therefore, be masked. When 75 micrograms/ml of oxyphenbutazone was spiked into urine samples, subjected to an extraction procedure for basic drugs, and then run in our routine HPTLC systems, no 'spots' due to oxyphenbutazone appeared. 'Masking' by oxyphenbutazone, therefore, did not and could not occur in our test systems. When phenylbutazone at a concentration of 30 micrograms/ml was spiked into urine samples and run in the routine HPTLC system, phenylbutazone spots were visible under ultraviolet light and after certain specific oversprays were used to visualize basic drugs. These spots, however, did not interfere with routine thin layer testing for basic drugs. It was concluded that phenylbutazone and oxyphenbutazone had no significant ability to interfere with detection of the parent forms of these basic drugs under the conditions described in these experiments.

Phenylbutazone and its metabolites in plasma and urine of thoroughbred horses: population distributions and effects of urinary pH.

A survey of plasma and urinary concentrations of phenylbutazone and its metabolites in thoroughbred horses racing in Kentucky was carried out. Post-race blood samples from more than 200 horses running at Latonia Racetrack and Keeneland in the Spring of 1983 were analysed. The modal plasma concentration of phenylbutazone was between 1 and 2 micrograms/ml, the mean concentration was 3.5 micrograms/ml and the range was up to 15 micrograms/ml. Oxyphenbutazone had a modal plasma concentration between 1 and 2 micrograms/ml, a mean concentration of 2.07 micrograms/ml and a range of up to 13 micrograms/ml. gamma OH-phenylbutazone had a modal plasma concentration of less than 1 microgram/ml, a mean level of 1.39 micrograms/ml and a range of up to 7.32 micrograms/ml. All plasma concentration frequency distributions were well fitted by log normal distributions. Urinary concentrations of phenylbutazone yielded modal concentrations of less than 1 microgram/ml, a mean urinary concentration of 2.9 micrograms/ml, with a range of up to 30.5 micrograms/ml. This population fitted a log-normal distribution. For oxyphenbutazone the modal concentration was less than 3 micrograms/ml, the mean concentration was 15.26 micrograms/ml, with a range to 81.5 micrograms/ml. The frequency distribution of these samples was apparently bimodal. For gamma OH-phenylbutazone, the modal concentration was less than 4 micrograms/ml, the mean concentration 21.23 micrograms/ml, with a range of up to 122 micrograms/ml. The population frequency distribution for gamma OH-phenylbutazone was indeterminate. Analysis of the pH of these post-race urine samples showed a bimodal frequency distribution. The pH values observed ranged from 4.9 to 8.7, with peaks at about pH 5.25 and 7.25. This bimodal pattern of urinary pH values is consistent with observations made in England and Japan. Urinary pH influenced the concentrations of phenylbutazone, oxyphenbutazone and gamma OH-phenylbutazone found in the urine samples. The concentration of these metabolites found in alkaline urines were from 32 to 225 times greater than those found in acidic urines. Plasma concentrations of phenylbutazone and its metabolites, however, were unaffected by urinary pH. In interlaboratory experiments, horses running at Hollywood Park were dosed with phenylbutazone at about 2 g/1000 lbs 24 and 48 h before racing, and a mean dose of 0.6 g/1000 lbs at 72 h prior to racing. Post-race plasma samples from these horses showed phenylbutazone concentrations ranging from 0.44 to 9.97 micrograms/ml, with a mean concentration of 4.09 micrograms/ml.(ABSTRACT TRUNCATED AT 400 WORDS)

Plasma and serum concentrations of phenylbutazone and oxyphenbutazone in racing Thoroughbreds 24 hours after treatment with various dosage regimens.

The plasma and serum concentrations of phenylbutazone (PBZ) and oxyphenbutazone were measured in 158 Thoroughbred horses after various doses of PBZ wer given. All horses were competing or training at racetracks in various parts of the country. All horses used in the study had not been given PBZ 24 hours before they were placed on a specific dosage schedule. Samples were collected 24 hours after the last PBZ administration. Four grams of PBZ were given daily by stomach tube, paste, or tablet for 3 days. On day 4, 24 hours before sample collection, an IV dose of 2 g of PBZ was given, regardless of the dose and method of administration. The 24-hour PBZ plasma concentrations were 3.51, 6.13, and 6.40 micrograms/ml, respectively. After 2 g of PBZ was administered IV daily for 4 days, the plasma PBZ concentration was 4.16 g/ml; after a single 2-g IV administration, the serum concentration was 0.87 g/ml. Concentrations of oxyphenbutazone were 3.35 (stomach tube), 4.29 (paste), 3.60 (tablet), 3.65 (4-day IV), and 1.11 g/ml (single IV). A significant relationship was not found between the serum and the urinary concentrations at this 24-hour measurement. Split samples sent to various laboratories confirmed the stability of high-performance liquid chromatography as a method of analysis.