The aim of the current study was to compare the pharmacokinetics of phenobarbital (PB) in extensive metabolizers (EMs) and poor metabolizers (PMs) of S-mephenytoin. Ten healthy volunteers (5 EMs and 5 PMs) were given 30 mg PB daily for 14 days. PB and p-hydroxyphenobarbital (p-OHPB) in serum and urine were measured by high-performance liquid chromatography (HPLC). Urinary excretion (12.5% versus 7.7%) and formation clearance (29.8 versus 21.1 mL/h) of p-OHPB, one of the main metabolites of PB, were significantly lower (p < .05) in PMs than in EMs. However, area under the serum concentration-time curve (153.3 in the EMs versus 122.9 microg x h/mL in the PMs), total (210.8 versus 254.9 mL/h) and renal clearance (53.1 versus 66.1 mL/h) of PB were identical between the two groups. To compare the inducibility of CYP2C19, mephenytoin was also given prior to and on the last day of PB treatment. The urinary level of 4'-hydroxymephenytoin was analyzed by a validated gas chromatograpy/mass spectrometry (GC/MS) method. The mephenytoin hydroxylation index did not change in either EMs (1.42 versus 1.42) or PMs (341.4 versus 403.5), showing that CYP2C19 was not induced by treatment with PB. These results indicated that the p-hydroxylation pathway of PB co-segregates with the CYP2C19 metabolic polymorphism. However, the overall disposition kinetics of PB were not different between EMs and PMs, and therefore polymorphic CYP2C19 seems have no major clinical implications.