The definition of infertility is pregnancy not occurring after 1 year of normal sexual activity for a couple not using contraceptives. Birth rates of live offspring to women and men with epilepsy are decreased compared to the general population¹ and same-sex siblings,⁶⁶ suggesting that persons with epilepsy are at risk for infertility. Although these studies adjust for marriage rates, they cannot determine the exact factors accounting for lower birth rates, such as choosing not to bear children in the setting of a chronic illness. One clear biologic factor that may affect fertility is the higher number of anovulatory menstrual cycles per year for women with epilepsy.⁴⁸

Women with epilepsy also have higher rates of polycystic ovary syndrome (PCOS), which is a frequent cause of infertility, than the general population.⁴ The current definition of polycystic ovary syndrome is the presence of two of the following three factors: (a) polycystic ovaries, (b) oligo-/anovulation, and/or (c) clinical or biochemical evidence of hyperandrogenism.²⁵ Other frequently present features of PCOS include an elevated ratio of luteinizing hormone (LH) to follicle-stimulating hormone (FSH) and insulin resistance with or without obesity.^40,56

The association between PCOS and epilepsy is complex because epilepsy itself may be a cause or precipitant for PCOS. Epilepsy dysregulates hypothalamic activity by propagation of ictal and interictal discharges through the structure, and it therefore disrupts normal LH pulsatility. Several studies have shown altered LH pulsatility in both men and women with epilepsy compared to normal control subjects.^15,27 LH pulsatility is altered in men with epilepsy, although the effects on LH pulsatility differ in the interictal and postictal states.⁵⁵ For men in the interictal state, the pulsatile secretion of LH is slower in onset and has higher peak concentrations compared to healthy controls, whereas postictally the pulsatility becomes more irregular.⁵⁵

The alteration of this exquisitely balanced hypothalamic-pituitary-gonadal axis may be further dysregulated via increased gonadotropin-releasing hormone (GNRH) secretion caused by contiguous ictal and interictal physiologic activity. Increased GNRH secretion preferentially increases the LH-to-FSH ratio.³⁸ Consistent with this postulated effect, elevated LH-to-FSH ratios have been documented in women with epilepsy.⁴⁸ LH stimulates ovarian steroidogenesis, and an elevated LH-to-FSH ratio will produce follicles that do not fully mature but, rather, become numerous and cystic. Immature follicles are deficient in aromatase, which is the enzyme that produces estrogen in the ovary by converting it from its precursor testosterone. In this manner, the PCOS ovarian follicle primarily manufactures androgens.

The question of whether valproate can cause, exacerbate, or imitate PCOS remains unclear, but it is clear that valproate is associated with the three primary features of PCOS. Valproate is associated with increased androgens^{31,32,45,49,70} and cystic ovaries.^31,32 However, the association between valproate and anovulation has not been consistently found,^32,49,50 but, when present, could contribute to difficulty conceiving. The contribution of valproate to PCOS is confounded by an increased occurrence of PCOS in women with epilepsy in general; a similar rate of PCOS in women with epilepsy taking carbamazepine, valproate, or no AEDs has been reported, around 11% for each group.⁴ Valproate itself inhibits aromatase and can block the conversion of testosterone to estrogen and, in this manner, among other mechanisms, contribute to hyperandrogenism.^22,71 Another well-known adverse effect of valproate now clearly emerging as an endocrinopathy is weight gain, which occurs in 40% to 50% of patients.³² The mechanism of valproate-induced obesity may be due to its association with increased insulin and leptin levels and consequent decreased ghrelin and adiponectin levels.²⁰

Because valproate is an effective AED for many persons with epilepsy, the decision to use or continue it despite the risk of endocrine dysfunction is difficult. For women with evidence of reproductive dysfunction, such as anovulation, hyperandrogenism, or PCOS itself, valproate should not be the first choice of AED because it may exacerbate these features. Weight gain with valproate use is also a reason for discontinuation for both
men and women, not only for cosmetic reasons, but also due to the health risks associated with obesity.

Decreased levels of free testosterone have been found in a large group of men with epilepsy (n = 200) compared to normal controls (n = 105).⁵ Some nuances were found, however, within the epilepsy group. Decreased testosterone-to-LH ratios were present in the temporal lobe epilepsy group only, whereas this important ratio was normal for patients with idiopathic generalized epilepsy. This ratio is an indicator of testicular function because LH increases in response to low testosterone levels and stimulates testicular testosterone production. Therefore, an increased testosterone-to-LH ratio suggests testicular dysfunction and an inability to respond to LH stimulation. Carbamazepine-treated men in this study had the most altered free testosterone and testosterone-to-LH ratios, consistent with the postulated effect of carbamazepine in increasing the hepatic production of testosterone-hormone binding globulin and inducing aromatase, which metabolizes the conversion of testosterone to estradiol. Valproate, on the other hand, was associated with normal total testosterone and testosterone-to-LH ratios, although the free testosterone level was still significantly decreased.⁵

In an evaluation of 65 men with epilepsy on AEDs, a higher rate of sperm abnormalities, including sperm morphology, motility, and concentration, was found.³³ Of the AEDs evaluated in this report, oxcarbazepine had the least detrimental effect on sperm quality compared to carbamazepine and valproate. The authors stated, therefore, that AED use could be a cause of infertility for men with epilepsy.³³

In an evaluation of 63 men with localization-related epilepsy, low bioactive testosterone (similar to free testosterone) was found in men not taking AEDs (n = 9) and in men taking hepatic enzyme-inducing AEDs (n = 36).²⁸ Enzyme-inducing AEDs were associated with increased sex-hormone–binding globulin as well. Men taking lamotrigine (n = 18) showed more normalized bioactive testosterone and better testicular functioning based on the bioactive testosterone-to-LH ratio compared to enzyme-inducing AEDs,²⁸ indicating that lamotrigine has less detrimental effect on reproductive and sexual functioning than do enzyme-inducing AEDs.

Lamotrigine has been reported to be associated with improved sexual functioning in 141 women and men with epilepsy, including those who were initiated on monotherapy or switched to monotherapy from another AED.¹⁹ Women with epilepsy reported decreased sexual functioning in association with phenytoin use and with low levels of estradiol and dehydroepiandrosterone sulfate, as well as mild depression.⁴⁷

Enzyme-inducing AEDs therefore appear to increase the risk of sexual and reproductive dysfunction in persons with epilepsy, whereas valproate may have less of this negative effect, although it has the risk of other endocrinopathies. Evidence shows lamotrigine to be more neutral with regard to these variables, but epilepsy itself is associated with reproductive disturbances.

Perimenopause is a life epoch when seizures tend to increase, and therefore alterations in epilepsy treatment in response to this increase may be required.²⁴ At menopause, however, seizures may decrease with the cessation of menses, especially if there had been a catamenial pattern during the reproductive years.²⁴ An earlier age at menopause has been reported in women with epilepsy, particularly in those with frequent seizures, but no clear relationship to AEDs has been found.²³ However, only a few of the newer-generation AEDs were evaluated in this study.²³

Treating women with epilepsy of childbearing age should include systematic, ongoing, and accurate counseling concerning optimal choice of contraception. A computerized database study demonstrated that 16.7% of women with epilepsy are prescribed oral contraceptives (OCs) compared to 25% of nonepileptic women at fertile age in an English population.⁶⁹ OCs and AEDs can interact in two ways: AEDs can alter the effects of OCs, and OCs can influence the metabolism of AEDs.