Abstracts

A higher incidence of symptoms suggestive of polycystic ovary syndrome (PCOS) has been reported both for women with epilepsy (WWE) who are receiving antiepileptic drugs (AEDs) and untreated WWE. We are conducting an international, randomized trial (LAM30007) to evaluate the development of PCOS symptoms in WWE initiating valproate (VPA) or lamotrigine (LTG) therapy. This is a preliminary analysis of the underlying incidence of polycystic ovaries (PCO), hyperandrogenism (HA), and metabolic disorders in these women.
Eligibility criteria included WWE 13-40 years; regular menstrual cycles; no concurrent hormonal medications; no prior LTG or VPA; and either newly diagnosed ([lt]2 weeks prior AED) or inadequately controlled epilepsy (only 1 prior chronic AED[ge]3 months). Baseline evaluations included an ultrasound (ovarian morphology), a hormone sample (day 1-6 of menstrual cycle), a fasted metabolic sample, and a 2-hour glucose loading test. HA was defined as a total or free testosterone or DHEAS above the upper limit of the reference range. An independent sonographer determined PCO status.
We report only preliminary baseline data for women with available ultrasound data (n=316); all baseline data are not yet available. The mean age, weight, and BMI were 22.3 years, 55.0kg, and 21.6 kg/m², respectively. At enrollment, 69% of women had newly diagnosed epilepsy and 16% had inadequately controlled epilepsy (15% data unavailable). More women had a history of generalized seizures (59%) than partial seizures (41%). PCO was noted in 130 (41%) women and HA was noted in 17 (5%) women; five (2%) women had both. There was no apparent difference between the percentage of newly diagnosed versus inadequately controlled women with PCO or HA. Women with generalized seizures had a slightly higher rate of PCO than those with partial seizures but similar rates of HA. More women with HA had a BMI[gt]25kg/m² than women without HA (35% vs. 15%); however, there were no apparent differences in median fasting insulin or glucose levels or impaired glucose tolerance (IGT) rates. Women with PCO had a lower mean FSH level (6.2IU/L vs. 7.1IU/L without PCO; p=0.008) and a higher mean LH/FSH ratio (0.93 vs. 0.74 without PCO; p=0.013) than women without PCO. Women with and without PCO did not have different mean androgen, LH, fasting insulin or glucose levels, or rates of IGT.
The incidence of PCO for WWE with regular menstrual cycles is higher than that reported for the general population. With the exception of lower FSH levels and a higher LH/FSH ratio, the hormone and metabolic parameters of women with PCO did not differ significantly from women without PCO. The effect of LTG vs. VPA therapy on these parameters will be evaluated upon study completion.
[Supported by: GlaxoSmithKline]