There is evidence that the sex-dependent expression of individual forms of the human cytochrome P450s (CYPs) results in gender-related differences in the hepatic metabolism of certain drugs. Previous work has shown that conflicting evidence exists relating to the sex differences in the activity of (S)-mephenytoin 4'-hydroxylase (CYP2C19). Accordingly, we assessed the effect of gender on CYP2C19 activity in a phenotyped and genotyped healthy unrelated Chinese population for further evidence of such a gender-based differentiation. One hundred and sixteen females and 129 males took one tablet of 100 mg racemic mephenytoin (Mesantoin, Sandoz) after emptying their urinary bladders. Amounts of (S)- and (R)-mephenytoin and its metabolite 4'-hydroxymephenytoin (4'-OH-M) excreted in the postdose 0-8 h urine collection were determined by GC and HPLC methods, respectively. The CYP2C19 activity was expressed as the ratio of S/R-mephenytoin (S/R-ratio), the percentage of the dose excreted as 4'-OH-M (D%), and the log10 of the hydroxylation index which was defined as the ratio of micromoles of (S)-mephenytoin dose to micromoles of 4'-OH-M excreted in urine (1g HI). From all the subjects studied, 53 extensive metabolizers (EMs) and 19 poor metabolizers (PMs) phenotyped were randomly selected and the DNA extracted from their blood samples was utilized for genotyping analysis according to the previously developed standard procedures. In this population, the phenotype PMs were identified in 10.9% (14/128) of the males, as compared with 11.2% (13/116) of the females (chi 2 = 0.0045, df = 1; p > 0.05). In all phenotyped subjects, the S/R-ratio of EM males was significantly higher than that of EM females (mean +/- SD; 0.28 +/- 0.17 vs. 0.24 +/- 0.15; p = 0.030), but no sexual differentiation was observed (p > 0.05) in 4'-OH-M excreted among all EMs and PMs, or the S/R-ratio among all PMs. In all genotyped EMs, the frequency of homozygous EMs was 18.4% higher in females (51.7%, 15/29) than in males (33.3%, 8/24) although there was no significant difference (chi 2 = 1.1370, df = 1, p > 0.05), but the S/R-ratio was lower in homozygous females than in homozygous males (0.22 +/- 0.14 vs. 0.33 +/- 0.09; p = 0.046). Thus, we conclude that the higher CYP2C19 activity in females exists among both the phenotyped EMs and the genotyped homozygous EMs compared with that in males, and that the defect frequency of the enzyme activity is equal between the genders. We also conclude that the S/R-ratio is more a sensitive metabolic marker of CYP2C19 enzyme activity than the D% and 1g HI.