[Efficacy of sodium valproate in partial epilepsy. Crossed study of valproate and carbamazepine].

An open response-conditional cross-over study of valproate versus carbamazepine has been done in previously untreated patients with partial seizures. Thirty-one patients entered the study. Nineteen were followed up to one year. It appeared that valproate was at least as effective as carbamazepine: at one year, 11 patients were seizure-free on valproate and only 8 were seizure-free on carbamazepine. Furthermore no side-effect was noted in valproate therapy, whereas carbamazepine was stopped in 2 patients because of skin rashes. The efficacy of sodium valproate in partial epilepsy remains controversial. It is of course limited when given as co-therapy in severe epilepsies, uncontrolled with other major antiepileptic drugs. However in naive patients, with recent and previously untreated partial epilepsies, a one-drug treatment with valproate appears to be as effective as carbamazepine or phenytoin. It has less unwanted side-effects and should be prescribed as first line treatment.