The association of ovulatory dysfunction with epilepsy and antiepileptic drugs (AEDs) was evaluated in women aged 18 to 40 years not receiving hormones recruited from the Stanford and Columbia Universities Comprehensive Epilepsy Centers and from other sources. Patients were followed for three menstrual cycles, a transvaginal ovarian ultrasound was obtained, and multiple endocrine and metabolic factors, including luteinizing hormone were sampled over 8 hours on days 2 to 5 of one cycle. Anovulatory cycles occurred in 10.5% of cycles in control patients without epilepsy (23), 14.3% of cycles with localization-related epilepsy (59 patients), and 27.1% of cycles with idiopathic (primary) generalized epilepsy (35 patients). At least one anovulatory cycle occurred in 38.1% of women with epilepsy who were taking valproate currently or within 3 years, and in 10.7% of non-valproate medicated patients. Risk factors for ovulatory dysfunction include idiopathic generalized epilepsy, exposure to valproate AED, high free testosterone, and reduced luteinizing hormone pulses. Patients with polycystic-appearing ovaries (41% of those with idiopathic generalized epilepsy cf 16% women without epilepsy) are not at increased risk and may ovulate normally. 
COMMENT. Women with idiopathic generalized epilepsy are at increased risk for ovarian dysfunction, anovulatory cycles, and polycystic-appearing ovaries. Those treated with valproate, a cytochrome P450 enzyme inhibiting AED, are at highest risk, and this AED has an additive adverse effect. Cytochrome P450 inducing AEDs (carbamazepine, phenytoin, phenobarbital), and AEDs with no effect on cP450 (lamotrigine, gabapentin) have no adverse additive effect on ovarian function. Ovarian function should be monitored in women with epilepsy. The anovulatory cycles may be the only sign of reproductive dysfunction.