Abstracts

In a previous study of the effect of 3 different anticonvulsants on ovarian function in a group of women with primary generalised epilepsy who had only ever taken one anticonvulsant (sodium valproate, lamotrigine or carbamazepine), we showed that women taking sodium valproate but not an oral contraceptive were significantly more likely to suffer from the polycystic ovary syndrome than women taking the other two anticonvulsants (who were no different from a group of women of comparable age who did not have epilepsy). We used the American hormonal definition of the polycystic ovary syndrome as it is more accurate. (Betts T [italic]et al[/italic] Seizure 2003;12:323-329).
We have now investigated whether women taking sodium valproate and suffering with the polycystic ovary syndrome lose evidence of this disorder if their anticonvulsant therapy is changed. All 16 women taking valproate, who had the polycystic ovary syndrome, were switched to a different anticonvulsant (11 to lamotrigine and 5 to levetiracetam). The initial anticonvulsant drug switch was from sodium valproate to lamotrigine but some later additionally received levetiracetam (and were withdrawn from lamotrigine subsequently) to preserve seizure freedom. All but one patient (on lamotrigine) lost the hormonal evidence of the polycystic ovary syndrome during the switch. This patient achieved pregnancy after one treatment with clomiphene. Our findings show that evidence of polycystic ovary syndrome is likely to be lost in women taking sodium valproate if they are switched to either lamotrigine or levetiracetam. It is possible therefore that the increased number of women with the polycystic ovary syndrome on sodium valproate is a property of the drug itself (possibly its effect on insulin metabolism) and not a permanent effect if the drug is withdrawn.