A 9-wk-old infant with familial giant cell hepatitis and severe intrahepatic cholestasis had low plasma concentrations of chenodeoxycholic acid and cholic acid and elevated plasma concentrations of 5 beta-cholestane-3 alpha,7 alpha,12 alpha,25-tetrol, 5 beta-cholestane-3 alpha,7 alpha,12 alpha,24 xi-tetrol, and 5 beta-cholest-24-ene-3 alpha,7 alpha,12 alpha-triol. Analysis of the urine by fast atom bombardment mass spectrometry and by gas chromatography-mass spectrometry after treatment with Helix pomatia glucuronidase/sulfatase showed that the major cholanoids in urine were the glucuronides of 5 beta-cholestane-3 alpha,7 alpha,12 alpha,24S,25-pentol, 5 beta-cholestane-3 alpha,7 alpha,12 alpha,25-tetrol, and 5 beta-cholestane-3 alpha,7 alpha,12 alpha,24 xi-tetrol. These results are consistent with an inborn error of the 25-hydroxylase pathway for bile acid synthesis, specifically one of the enzymes responsible for conversion of 5 beta-cholestane-3 alpha,7 alpha,12 alpha,24S,25-pentol to cholic acid and acetone. Treatment with chenodeoxycholic acid was tried on two occasions. On the first it appeared to precipitate a rise in bilirubin, on the second the liver function tests improved and the improvement was maintained when treatment was modified to a combination of chenodeoxycholic acid and cholic acid and finally, cholic acid alone. Despite the normalization of liver function tests, a liver biopsy at 1.25 y showed an active cirrhosis. Nonetheless, the child is thriving at the age of 3.5 y, whereas an affected sibling died at 13 mo.