Liver injury has long been associated with occupational exposure to a wide variety of chemicals. The controversial data existing in relation to hepatotoxicity of organic solvents might be explained as a consequence of the different exposures or it may well be that the tests used for evaluating liver function might not be sensitive enough to detect any mild changes at an early stage. To study liver function during exposure to solvent mixtures, we determined serum bile acid (SBA) concentrations as compared with conventional liver function tests in a selected group of workers (n = 30) occupationally exposed to a mixture of organic solvents (mostly toluene, xylene, acetone, n-butylacetate, n-butanol, ethylacetate) and in a reference group (n = 20). The mean levels of liver enzyme activities and bilirubin concentrations in the two groups were similar, whereas mean SBA levels increased in the exposed group (8.0 +/- 6.0 mumol/l vs 2.8 +/- 1.4 mumol/l) and the difference as compared with the controls was significant (P less than 0.01). In 73% of the exposed workers, SBA levels were higher than 5.6 mumol/l (the cut-off value) as compared with 5% of the controls. These results demonstrate the higher sensitivity in detecting liver dysfunction achieved with the SBA test as compared with conventional hepatic function tests. As increased SBA concentrations are considered to reflect an impairment of anion transport across the liver, higher SBA levels in the group of workers exposed to organic solvents might be explained as a slight and early sign of liver dysfunction. Therefore, SBA determination in biological monitoring of workers exposed to potentially hepatotoxic chemicals might be proposed.