The effect of temporary water deprivation for 48 h on the pharmacokinetics and pharmacodynamics of azosemide was examined after intravenous (i.v., 10 mg/kg) and oral (20 mg/kg) administration of azosemide to the control and the water-deprived rats. After i.v. administration of azosemide, the area under the plasma concentration-time curve from time 0 to time infinity and the unbound fraction of azosemide to plasma proteins increased by 36 and 40%, respectively, and total body, renal (CLR), and nonrenal clearances, apparent volume of distribution at steady state, and total amount of azosemide excreted in urine (AeAZ0) decreased by 30, 44, 20, 25, and 33%, respectively, in the water-deprived rats (p < 0.05 vs controls). After oral administration of azosemide, the values of AeAZ0 (591 vs 318 micrograms) and CLR (2.22 vs 0.875 mL/min/kg) decreased significantly in the water-deprived rats. The 12-h urine output per g kidney was also reduced significantly in the water-deprived rats after both i.v. (70.3 vs 24.6 mL) and oral (70.7 and 20.7 mL) administration of azosemide. This could be due to the significantly reduced amount of azosemide excreted in 12 h urine, significantly higher plasma osmolarity, and increased blood vasopressin concentration in the water-deprived rats. The 12-h urinary excretion of sodium, potassium, and chloride per g kidney was also reduced significantly in the water-deprived rats after both i.v. and oral administration. The diuretic efficiency decreased significantly in the water-deprived rats after both i.v. and oral administration. The 12-h urine output per g kidney of i.v. and oral administration of azosemide in the control rats were similar (70.3 +/- 17.3 vs 70.7 +/- 13.8 mL), although the AeAZ0 after oral administration was significantly smaller than that after i.v. administration (473 +/- 98.5 vs 285 +/- 76.3 micrograms), and similar results were also obtained from the water-deprived rats. This could be rationalized by the concept of a single maximally efficient excretion rate of the drug in the pharmacodynamic model (sigmoid Emax), as shown for furosemide.