The mechanism responsible for reduced drug clearance in patients with cirrhosis was investigated using d-propranolol as an example of a drug that is highly extracted by the liver in the normal subject. The pharmacokinetics of d-propranolol, 0.4 mg · kg⁻¹, was studied after intravenous injection in 26 subjects: the total clearance (mean ± SD) of the drug was not different in 6 subjects with normal liver (21 ± 7 ml · min⁻¹ · kg⁻¹) and in 6 patients with alcoholic fibrosis of the liver, without cirrhosis (18 ± 8 ml · min⁻¹ · kg⁻¹); the total clearance of the drug was significantly lower in 14 patients with alcoholic cirrhosis (8 ± 5 ml · min⁻¹ · kg⁻¹). Concomitantly with the pharmacokinetic study, a hepatic vein was catheterized, and indocyanin green was infused in the 6 patients with fibrosis of the liver and in 6 of the 14 patients with cirrhosis; information provided by these techniques allowed measurement of the hepatic clearance of d-propranolol and estimation of its contributing factors: the hepatic clearance of d-propranolol was 18 ± 6 ml · min⁻¹ · kg⁻¹ in patients with fibrosis and only 6 ± 3 ml · min⁻¹ · kg⁻¹ in patients with cirrhosis; the liver blood flow was 25 ± 9 ml · min⁻¹ · kg⁻¹ in the former and 14 ± 6 ml · min⁻¹ · kg⁻¹ in the latter patients; the hepatic extraction ratio of d-propranolol was 0.73 ± 0.09 in patients with fibrosis of the liver and 0.41 ±0.17 in patients with cirrhosis; the intrinsic hepatic clearance of d-propranolol, a mathematical concept expressing the maximal ability of the liver to remove the drug from the blood, was 72 ± 30 ml · min⁻¹ · kg⁻¹ in the former and only 12 ± 7 ml · min⁻¹ · kg⁻¹ in the latter patients. It is concluded that, in patients with cirrhosis, (1) the total clearance of d-propranolol is reduced, (2) the reduced total clearance is a consequence of a reduced hepatic clearance, and (3) the reduced hepatic clearance results mainly from an impaired ability of the liver to remove the drug from the blood and secondarily from a reduced liver blood flow.