The aim of the present prospective study was to evaluate changes in plasma GABA concentration in relation to clinical response during vigabatrin treatment of epilepsy. We studied 29 patients with uncontrolled partial-onset seizures during open add-on vigabatrin treatment and measured plasma GABA and vigabatrin concentrations by a sensitive HPLC method. Following short-term treatment 17 out of 28 patients had a seizure reduction of > 50% (responders). After long-term treatment 16 out of 22 patients were responders. There was no difference between responders and nonresponders regarding pretreatment seizure frequency, treatment duration, vigabatrin dose, or plasma vigabatrin concentration. Responders had a significant (p < 0.001) increase in mean plasma GABA both after short-term (from 0.380 to 0.530 nmol/ml; mean increase: 48%) and after long-term (from 0.392 to 0.618 nmol/ml; mean increase: 71%) vigabatrin treatment, whilst nonresponders had no significant changes in GABA levels. However, since plasma GABA increased in a subgroup of nonresponders, mean plasma GABA levels did not differ between responders and nonresponders. Although plasma GABA increased significantly in the responder but not in the nonresponder group during vigabatrin treatment of patients with epilepsy, it does not seem to be a reliable marker of individual clinical response to vigabatrin treatment.