A double-blind, placebo-controlled, crossover study of oral Vigabatrin (2-3 g/d) as add-on therapy for 31 patients with drug-resistant seizures is reported from the Neurological Clinic, Univ of Bologna School of Med, Italy. Both children and adults were included. Those with complex partial seizures and temporal spikes responded whereas patients with mixed seizure types and multi-focal EEG abnormalities were not benefitted. Drowsiness was the most frequent side-effect and concomitant phenytoin serum concentrations fell during Vigabatrin treatment. 
COMMENT. Vigabatrin (y-vinyl GABA) is an inhibitor of y-aminobutyric acid (GABA)-transaminase. Increases in CNS-GABA concentrations in laboratory animals and CSF-GABA in patients treated with Vigabatrin have been associated with anticonvulsant activity. Several laboratory and clinical reports of this drug have appeared in the literature in the past decade, mostly with favorable results in patients with complex partial seizures. [2, 3]
The finding of microvacuoles in the white matter of the CNS of laboratory animals has not been duplicated in autopsy reports on patients who have died from causes independent of Vigabatrin therapy or in CT scans but has prompted the FDA to put a hold on clinical trials in the USA since 1983. At present, 31 adult patients with complex partial seizures are in the on-going collaborative study but no patient may be added (R. Miketta, M.D., Merrell Dow Pharmaceuticals, personal communication). Phase III trials are continuing in other countries and registration of the drug is expected in France in 1988. Side effects in adults treated with Vigabatrin for complex partial seizures have included drowsiness, ataxia, dizziness, headache and skin rash. Levels of SGPT have shown decreases, as might be expected, but no liver or blood disorders have been reported.
Every anticonvulsant drug, both old and new, has its problems, and clinical trials are fraught with potential hazards (e.g. liver fatalities with valproate, leukopenia with carbamazepine, erythe multiforme and lymphoma with phenytoin, and learning disorders with barbiturates). Hopefully, the Vigabatrin-induced CNS vacuoles in animals will prove to be a species specific effect but close monitoring in man is required.