The effect of furosemide on hepatic bile formation was studied in isolated perfused rat liver to determine if 1) the observed cholestatic effect at lower dose of furosemide in vivo is a primary effect or a secondary effect due to decreased hepatic blood flow caused by the furosemide-induced volume contraction and if 2) the observed choleretic effect at higher doses can be explained by the osmotic effect of furosemide and its metabolites in bile. A single dose of furosemide (initial perfusate concentration 0.01, 0.1 or 1 mM) produced choleresis, whereas 0.001 mM furosemide did not affect bile flow significantly. Because furosemide failed to produce cholestasis at tested doses, the observed cholestasis in vivo at similar blood concentrations must be a secondary effect. Furosemide choleresis was associated with biliary secretion of furosemide and its metabolites. However, the choleretic effect expressed as microliters per micromole of drug secreted declined with increasing dose and biliary secretion. Furosemide choleresis was also associated with an increase in the net biliary secretion of Na+ and Cl-. The effect of Na+ and Cl- replacement on furosemide choleresis was studied to determine if the choleresis was a result of direct effect of furosemide on hepatic electrolyte transport. Replacement of perfusate Na+ completely by Li+ or partially by choline+ resulted in a 30 to 50% reduction in choleretic effect and furosemide-induced biliary Cl- secretion. A similar decline in choleretic effect and net furosemide-induced biliary Na+ secretion was also observed when perfusate Cl- was replaced by nitrate, acetate or isethionate.(ABSTRACT TRUNCATED AT 250 WORDS)