Background: Women with epilepsy may be at risk of developing PCOS, but it is unclear whether this is due to the disease or sodium valproate (VPA) therapy. We sought to determine androgenic status and cardiometabolic risk in women taking VPA monotherapy and compared these with subjects taking lamotrigine and healthy controls.
Methods: Patients with epilepsy (age 1843) with no known vascular risk were divided into 3 groups: VPA monotherapy (n=14), lamotrigine therapy (Lam; n=13) and healthy controls (Con; n=10), matched for age, height, weight and smoking status. Arterial compliance and endothelial function were measured by the Sphygmocor system.
Results: There was a trend towards increased central adiposity (waist circumference: VPA 92.7±4.2, Lam 83±4.3, Con 87.1±4.7 cm; P=0.1) and fat mass (VPA 40.2±2%, Lam 35.5±2.5%, Con 34.1±3.6%; P=0.12) in the VPA group but no difference in the proportion of subjects with the metabolic syndrome. There were no differences in central blood pressure, augmentation index, pulse wave velocity, endothelial function, lipids or hsCRP among groups, but VPA patients exhibited a tendency towards increased insulin resistance (HOMA-IR: VPA 5.7±2.8, Lam 2.6±0.4, Con 2.6±0.8, P=0.28; QUICKI: VPA 0.54±0.02, Lam 0.59±0.02, Con 0.63±0.05; P=0.07). Testosterone (VPA 2.18±0.19, Lam 1.75±0.16, Con 1.26±0.21 mmol/l; P=0.005) and androstenedione (VPA 24.08±2.32, Lam 10.54±1.6, Con 6.7±0.98 mmol/l; P<0.001) levels were higher in the VPA group.
Conclusions: Our results support the contention that VPA therapy may be associated with a PCOS-like phenotype but this is not accompanied by changes in vascular function, at least at a young age. These findings need confirmation in larger, prospective studies but surveillance for cardiometabolic risk is appropriate in the meantime.