Ethanol elimination in rats following bolus intravenous administration (0.5, 1, 2, 3 g/kg body weight) was investigated with and without pyrazole pretreatment. Elimination time was significantly longer in the pyrazole-pretreated group than in the control. Simultaneous multilines fitting of two-compartment models to the ethanol concentration-time courses proved that the two-compartment model with Michaelis-Menten elimination kinetics was optimum to the curves in both groups. the V(max) value (0.41 +/- 0.05 mg/ml/hr) in the pyrazole-pretreated group was significantly lower than that (0.70 +/- 0.03 mg/ml/hr) in the control. The K(m) value (1.44 +/- 0.12 mg/ml) in the pyrazole-pretreated group was significantly higher than the control (0.07 +/- 0.01 mg/ml), similar to the in vitro value in hepatic alcohol dehydrogenase (ADH). This higher K(m) value in the pyrazole-treated group than the in vitro value in the microsomal ethanol oxidizing system (MEOS) suggests that pyrazole-insensitive pathways may involve pathways other than MEOS. The first-order elimination rate from the two-compartment model with parallel first-order and Michaelis-Menten kinetics was a very low value of 10(-5)min(-1) ,which shows that a pathway with a higher K(m) than blood ethanol concentration does not contribute to in vivo ethanol elimination. The relative contribution of the pyrazole-sensitive pathway calculated from the ratio of total clearance in two groups was 83% at a low ethanol concentration of 0.5 mg/ml and 60% at a high concentration of 5.0 mg/ml. The K(m) value (0.02 mg/ml) from fitting the double Michaelis-Menten model is similar to that in vitro in ADH3. These findings suggest that the ADH3 pathway plays a major role in ethanol elimination.