Vigabatrin is a structural analogue of the inhibitory neurotransmitter gamma-aminobutyric acid (GABA). It is supplied as a racemic mixture, with the S(+) enantiomer possessing pharmacological activity. [R,S]-Vigabatrin plasma concentrations can be estimated using high-performance liquid chromatographic methods. Only gas chromatography-mass spectrometry methods allow quantification of the S(+) and R(-) enantiomers. Vigabatrin was rapidly absorbed reaching peak concentrations within 1 to 2h. Area under plasma concentration-time curves indicated dose-linear pharmacokinetics. There was no effect of food on the absorption of vigabatrin. The absorption characteristics of the enantiomers were similar to those of the [R,S]-vigabatrin. No chiral inversion was detected after administration of the pure S(+) enantiomer. Vigabatrin is not protein bound. The apparent volume of distribution of [R,S]-vigabatrin was approximately 0.8 L/kg. Despite the lack of protein binding, cerebrospinal concentrations of the [R,S]-vigabatrin were only 10% of the plasma concentration 6h after a single oral dose. The half-life of [R,S]-vigabatrin was between 5.3 and 7.4h, the half-life of the enantiomers were 7.5 and 8.1h for the S(+) and the R(-) forms, respectively. The major route of elimination was renal excretion; urinary recovery of the [R,S]-vigabatrin was close to 70%. Pharmacokinetic studies in epileptic children did not show any significant effect of maturation on the disposition of the S(+) enantiomer: the half-life and the renal clearance were similar to adult values. Data suggest a lower bioavailability in children. In adults with epilepsy, the half-life of the [R,S]-vigabatrin ranged from 4.2 and 5.6h, similar to that measured in healthy adults. In elderly nonepileptic volunteers the pharmacokinetics of the enantiomers of vigabatrin showed delayed absorption, a major increase in peak concentration and a prolonged half-life. These changes were attributed to decreased renal clearance of vigabatrin. A nonlinear relationship between renal clearance and creatinine clearance was suggested. Vigabatrin caused a 20% fall in plasma phenytoin concentrations, the mechanism of which has not been elucidated. There were no other interactions with most concurrently administered anticonvulsants. The usual dosage of vigabatrin as add-on treatment in adults is 2 to 4g daily. Higher dosages up to 80 mg/kg daily were required in children. A dosage adjustment was recommended in any patient with decreased renal clearance. Although anticonvulsant effects were clearly related to dosage, monitoring of plasma concentrations of vigabatrin as a guide to dosage is unlikely to be of as much value as with other antiepileptic drugs. The action of the drug long outlasts its presence in plasma.