Is valproate monotherapy a practical possibility in chronically uncontrolled epilepsy?

Abstract

In a prospective open study of 20 male epileptic residents of a mental handicap institution, polytherapy was gradually reduced to valproate monotherapy in 18 subjects. In terms of seizure frequency this was significantly disadvantageous but when carbamazepine was added or substituted, seizure control improved significantly. Drugs with documented adverse effects on cognitive function such as phenobarbitone and phenytoin were phased out. In the 18 subjects who achieved valproate monotherapy, no association between serum levels and seizure control could be demonstrated. Adverse effects of valproate were pancreatitis and thrombocytopenia; in one subject thrombocytopenia appeared to be associated with levels in the toxic range but in six other subjects 'toxic' levels of valproate did not give rise to any clinically detectable toxic signs. There was no instance of tremor or weight gain. It was concluded that, in the population studied (institutionalized patients with chronically uncontrolled seizures) valproate monotherapy was inappropriate but carbamazepine with or without valproate was a better option. Phasing out phenytoin and phenobarbitone was successful. Valproate serum levels did not contribute significantly to the conduct of the study; no general relationship between valproate serum levels and either seizure control or toxicity could be demonstrated.