Stiripentol, a new
antiepileptic drug inhibiting
cytochrome P450-enzymes, suggested some efficacy when combined with
carbamazepine in an open trial in refractory
partial epilepsy of childhood. Our objective was to test these results in a placebo-controlled trial. To limit the number of patients included, we used an enrichment and withdrawal design. Among the 67 children entered in a 4-month open add-on
stiripentol study following a 1-month single-blind placebo baseline, the 32 responders were randomized for 2 months either to continue
stiripentol (n = 17) or to withdraw to placebo (n = 15). If
seizures increased by at least 50% after randomization compared with baseline, the patients dropped out (primary end point): there were six patients on
stiripentol and eight patients on placebo (not significant). However, a decrease in seizure frequency compared with baseline (secondary end point) was greater on
stiripentol (-75%) than on placebo (-22%) (P < .025). Twelve patients experienced at least one adverse event on
stiripentol (71%) compared with four patients on placebo (27%); none were reported as severe. The combination of
stiripentol and
carbamazepine proved to reduce seizure frequency in children with refractory
partial epilepsy, although it failed to show a significant impact according to the escape criteria selected as the primary end point in the present study, for ethical reasons.