Teratogenicity is a considerable problem for women with epilepsy. Under most circumstances it would be detrimental to their health to have the antiepileptic medication withdrawn during pregnancy. At the same time, there is risk to the fetus exposed to certain anticonvulsants. The problem is compounded if the epileptic woman has multiple seizure types or seizures refractory to monotherapy. Under these circumstances, polytherapy is required to control seizures and the rate of teratogenicity is two to threefold higher with multiple drugs. In rodent systems, growth and other physical anomalies apparent in the embryo or at birth, resulting from phenytoin (PHT) exposure during gestation, have been linked to reactive metabolite exposure. However, no studies have sought to identify the cause of learning and behavioral deficits which become apparent after birth. Carbamazepine (CBZ) apparently forms at least one reactive arene oxide metabolite, but no studies have been conducted to link any form of carbamazepine-induced teratogenicity to reactive metabolite exposure, although the likelihood of this mechanism has been recognized. The proposed research will test the hypothesis that teratogenicity in the form of growth problems at birth and (more importantly) developmental delays after birth, first in PHT therapy, and second in CBZ therapy, is due to reactive metabolite exposure. We propose to conduct in vivo inhibition, pharmacokinetic, and teratogenicity studies using a pregnancy monkey model. These studies will be accomplished by utilizing a nonteratogenic second drug stiripentol (STP), which in polytherapy with PHT (or CBZ) inhibits cytochrome P-450 and thus decreases reactive metabolite exposure while maintaining the plasma concentration of the parent compound.