This study was designed to test the hypothesis that increasing the infusion rate of bile salts could overcome drug-induced cholestasis. Cholestasis was induced by administration of 17.5 mumol/L estradiol-17 beta-D-glucuronide during the infusion of taurocholate, tauroursodeoxycholate or dehydrocholate at 20 nmol/min/gm liver. After 30 min, a bolus of 10 mumol of the bile salts was added to the perfusate, and the infusion rate of each bile salt was increased. Taurocholate at a rate of 62 or 125 nmol/min/gm liver, caused a prompt dose-dependent increase of the depressed bile flow and bile salt excretion. A higher rate of taurocholate infusion (180 nmol/min/gm liver) was less effective than either the 62 or 125 rate in increasing bile flow. Infusion of tauroursodeoxycholate at 250 or 390 nmol/min/gm liver also led to a dose-dependent recovery. Further increase of tauroursodeoxycholate infusion rate of 580 nmol/min/gm liver did not provide any additional recovery in bile flow. Dehydrocholate, at rates of 62 or 125 nmol/min/gm liver, gave only a slight enhancement of bile flow. Both taurocholate and tauroursodeoxycholate caused a marked removal of the estradiol-17 beta-D-glucuronide, which had accumulated in the liver. At lower taurocholate infusion rates, the estradiol-17 beta-D-glucuronide was excreted mainly in the bile. At the highest rate, however, biliary excretion of estradiol-17 beta-D-glucuronide declined significantly, and a marked back-efflux of the estrogen into the perfusate was noted. In contrast, tauroursodeoxycholate led to enhanced biliary estradiol-17 beta-D-glucuronide excretion at all increased tauroursodeoxycholate infusion rates and to only a small increase in back-efflux of estradiol-17 beta-D-glucuronide at the two highest tauroursodeoxycholate infusion rates.(ABSTRACT TRUNCATED AT 250 WORDS)