A meta-analysis of polycystic ovary syndrome in women taking valproate for epilepsy.

PURPOSE

To conduct a meta-analysis to quantify whether valproate (VPA) is associated with an increased risk of polycystic ovary syndrome (PCOS) in women with epilepsy.

METHODS

A comprehensive literature search for all published studies of MEDLINE and EMBASE was performed for English language studies published from 1st January 1990 to 4th February 2011. The studies included should be prospective and controlled. We then performed a meta-analysis to identify polycystic ovary syndrome in women with epilepsy treated with VPA.

RESULTS

The meta-analysis included 11 studies that met inclusion/exclusion criteria, involving 556 women with epilepsy treated with VPA, 593 women treated with other antiepileptic drugs (AEDs), 120 women with untreated epilepsy and 329 healthy controls. The calculated incidence of PCOS in women treated with VPA was higher than that in women treated without VPA (10 studies included, P<0.05, OR 3.04, 95% CI 2.09-4.43). According to different definitions or diagnostic criteria of PCOS, the results were different. VPA's treatment was associated with higher incidence of PCOS in the criteria of hyperandrogenism and/or hyperandrogenemia, oligoovulation (4 studies included, P<0.05, OR 3.22, 95% CI 1.79-5.80), and in the criteria of two of the three following fulfilled: appearance of polycystic ovary (PCO) on the ultrasonogram, elevated serum testosterone levels and irregular (oligo-/amenorrhea) menstrual cycles (1 post hoc reanalysis study included, P<0.05, OR 5.01, 95% CI 2.40-10.49. Another prospective and crosssectional included, P<0.05, OR 3.50, 95% CI 1.30-9.40). However, in the criteria of PCO with ovulatory dysfunction (polymenorrhea, amenorrhea, or oligomenorrhea), clinical and/or biochemical evidence of hyperandrogenism according to National Institutes of Child Health and Human Development (NICHHD) and The National Institutes of Health (NIH) Consensus Conference, VPA did not increase the PCOS incidence compared with other AEDs (4 studies included, P>0.05, OR 1.37, 95% CI 0.59-3.19). Data on PCOS's components were pooled. PCO was higher in women treated with VPA than in those without VPA (6 studies included, P<0.05, OR 1.94, 95% CI 1.28-2.95) or in healthy controls (4 studies included, P<0.05, OR 4.99, 95% CI 2.97-8.37). VPA treated women were more subject to hyperandrogenism than non-VPA treated women (5 studies included, P<0.05, OR 2.35, 95% CI 1.57-3.53). There is significant difference of menstrual disorders between VPA treated women and non-VPA treated women (7 studies included, P<0.05, OR 1.64, 95% CI 1.19-2.25).

CONCLUSIONS

Results of this meta-analysis suggest that the raw incidence of PCOS in VPA treated women with epilepsy is approximately 1.95 folds that in other AEDs treated women. PCOS and its components may be associated with VPA treated women with epilepsy. We recommend monitoring PCOS and its components in women with VPA mono-/polytherapy. In light of limitations and heterogenicity, there is surely the need for more prospective studies to identify VPA and PCOS.