Hypercreatinuria is a well-known feature of liver and testicular toxicity and we have recently proposed that hepatotoxin-induced hypercreatinuria would arise as a consequence of increased cysteine synthesis associated with the provision of protective substances (glutathione and/or taurine). Here a direct relationship between hepatotoxin-induced hypercreatinaemia and hypercreatinuria is shown and the possible relationships of hepatotoxin-induced hypercreatinaemia and hypercreatinuria to hepatic damage and to weakened nutritional status are examined. Male Sprague-Dawley rats were dosed with a variety of model hepatotoxins at two dose levels per toxin. Blood plasma samples taken at 24 h post-dosing and urine samples collected from 24-31 h post-dosing were analysed by (1)H NMR spectroscopy. Both hypercreatinaemia and hypercreatinuria were found in rats dosed with allyl formate (75 mg/kg), chlorpromazine (30 and 60 mg/kg), alpha-naphthylisothiocyanate (ANIT, 100 mg/kg) and thioacetamide (200 mg/kg), whilst significant hypercreatinuria, but not hypercreatinaemia, was found after dosing with thioacetamide (50 mg/kg). Neither hypercreatinaemia nor hypercreatinuria were found after dosing with allyl formate (25 mg/kg), ethionine (300 and 1000 mg/kg) or ANIT (30 mg/kg). Reduced feeding is known to cause hypercreatinuria in rats and, of the four hepatotoxins that induced hypercreatinaemia and hypercreatinuria at the given time-points, two, chlorpromazine and ANIT, also affected nutritional status with ketosis being clearly identifiable from the plasma (1)H NMR spectra. Thus, the creatine changes induced by ANIT and chlorpromazine are potentially attributable, in whole or in part, to reduced feeding rather than to liver effects alone and, consequently, the results were examined with and without inclusion of the ANIT and chlorpromazine data. With all of the data included, there were eight out of ten points of correspondence between the incidence of hypercreatinaemia and/or hypercreatinuria and the incidence of increases in plasma alanine aminotransferase (ALT) activity. At the same time there were nine out of ten points of correspondence between the incidence of hypercreatinaemia and/or hypercreatinuria and the incidence of increases in plasma aspartate aminotransferase (AST) activity. However, with the ANIT and chlorpromazine data excluded there was complete (six out of six points) correspondence between the incidence of hypercreatinaemia and/or hypercreatinuria and the incidence of increases in plasma AST and ALT in the remaining data. Likewise, with all of the data included, there was some apparent correlation (correlation coefficient, r=0.80) between the group mean levels of plasma AST and plasma creatine when expressed relative to the mean values for controls sampled at the same time-point. However, with the ANIT and chlorpromazine data excluded, that correlation coefficient was increased to 0.95. The findings of these studies suggest that the ANIT- and chlorpromazine-induced creatine changes may have been caused by reduced feeding rather than by liver toxicity. The allyl formate and thioacetamide data indicate that hepatocellular necrosis is accompanied by increases in plasma and urinary creatine, and suggest the possibility of a quantitative relationship between the increases in plasma AST and the increases in plasma creatine that are associated with hepatocellular necrosis. The ethionine and ANIT data suggest that fatty liver (steatosis) and cholestatic damage may not be associated with hypercreatinaemia and hypercreatinuria.