The objectives of this study were to determine the effect of phenylbutazone premedication on the pharmacokinetics and urinary excretion of frusemide in horses; and on frusemide-induced changes in urinary electrolyte excretion. Six Standardbred mares were used in a 3-way crossover design. The pharmacokinetics and renal effects of frusemide (1 mg/kg bwt i.v.) were studied with and without phenylbutazone premedication (8.8 mg/kg bwt per os 24 h before, followed by 4.4 mg/kg bwt i.v. 30 min before frusemide administration). A control (saline) treatment was also studied. Administration of frusemide without phenylbutazone led to diuresis, natriuresis, kaliuresis and chloruresis, and altered the ratio of sodium:chloride excretion from 0.4 to 1.0 in the first hour of diuresis. When frusemide and phenylbutazone were administered, sodium and chloride excretion in the first hour were significantly (P<0.05) reduced by 40 and 32%, respectively, when compared to frusemide administrationwithout phenylbutazone. The fractional clearance of sodium and chloride was also significantly reduced. Potassium excretion, potassium fractional clearance and the ratio of sodium to chloride excretion were not affected by administration of phenylbutazone. During peak diuresis, phenylbutazone did not affect the efficiency of frusemide with respect to electrolyte excretion. The plasma disposition of frusemide was not affected by phenylbutazone. However, the renal excretion of frusemide decreased by approximately 25%. We conclude that the decreased urinary excretion of frusemide by phenylbutazone led to an attenuation of frusemide-induced increases in urinary excretion of sodium and chloride. Since the efficiency of frusemide was not affected by phenylbutazone, we conclude that phenylbutazone attenuates the renal excretion of frusemide without inhibiting the intrarenal activity of frusemide in horses.