Women with epilepsy have a reduction in fertility (7–9) of as much as two thirds of that expected (10), as well as reproductive endocrine disorders, menstrual cycle disturbances, and ovulatory dysfunction (11).
The cause of lower fertility rates is multifactorial. A study in Finland found that persons with epilepsy were less likely to marry and to have offspring (12). In part, this reflects a choice. Much of that choice comes from wrong information suggesting that women with epilepsy are not fit parents, the risk of transmission of epilepsy is very high, or the risk of birth defects in children born to mothers with epilepsy is higher than it really is. A recent survey of healthcare professionals likely to encounter women with epilepsy finds that there is a marked lack of knowledge regarding pregnancy and fetal risks associated with maternal epilepsy and that many physicians would not support the decision of a woman with epilepsy to become pregnant (1).
One basis for infertility is physiological. Reproductive health disturbances in women with epilepsy include menstrual cycle abnormalities, anovulatory menstrual cycles, reproductive endocrine disorders, and sexual dysfunction (13–18). About one third of menstrual cycles in women with epilepsy are anovulatory compared with a rate of ∼10% in women without epilepsy (13,19). Women with primary generalized epilepsy were more likely to have anovulatory cycles than women with localization-related epilepsy (11). The antiepileptic medication VPA, unlike CBZ, GBP, LTG, phenobarbital (PB), or PHT, was significantly associated with anovulatory cycles (11,19). Women with primary generalized epilepsy receiving VPA were at highest risk. In fact, 55% of menstrual cycles were anovulatory in this group of women with epilepsy (11).
Ovulatory failure associated with epilepsy and some antiepileptic medications may be a result of endocrine and end-organ disturbances. Hypothalamic-pituitary axis dysfunction is suggested by observations that pituitary release of luteinizing hormone (LH) in women with epilepsy is altered spontaneously and in response to gonadotropin-releasing hormone (GnRH) (14,15,20) (Fig. 1). Women receiving CYP450 enzyme–inducing AEDs have significant reductions in serum concentrations of estradiol, testosterone, and dihydroepiandrostenedione, as well as elevations in SHBG (21–27). Women taking VPA (which does not induce liver cytochrome enzymes) have higher gonadal and adrenal androgen levels (17). Enhanced steroid metabolism and binding reduces the concentration of biologically active steroid. In contrast, adrenal and gonadal androgens are significantly elevated in women receiving the CYP450 enzyme inhibitor VPA. However, women with epilepsy taking GBP or LTG—two AEDs that do not alter CYP450 enzymes—have sex steroid hormone levels that are no different from those of nonepileptic controls not taking medications (21).
Figure 1. Reproductive endocrine axis disturbances in persons with epilepsy receiving antiepileptic drugs. Hypothalamic-pituitary axis illustrating amygdala interconnections to the hypothalamus. GnRH, gonadotropin-releasing hormone; FSH, follicle-stimulating hormone; LH, luteinizing hormone; PRL, prolactin; +, excitatory feedback; −, inhibitory feedback.
Download figure to PowerPoint
Polycystic ovary syndrome
The polycystic ovary syndrome (PCOS) is a gynecologic disorder affecting ∼7% of reproductive-age women. The phenotype includes signs of excess androgen sensitivity such as hirsutism, truncal obesity, and acne. Women with this syndrome have frequent anovulatory cycles and may have elevated androgen levels, elevated cholesterol with abnormal lipid profiles, an abnormal ratio of pituitary LH to follicle-stimulating hormone (FSH), elevated insulin, and glucose intolerance. Some women also have multiple ovarian cysts—a finding in 15–20% of reproductive-age women. The requirement for a diagnosis of PCOS is phenotypic or serologic evidence of androgen excess, as well as anovulatory cycles. Polycystic ovaries, while often present in women with this syndrome, are not required for diagnosis. In fact, asymptomatic polycystic ovaries may be relatively common in normal women of reproductive age, occurring in 21–23% of women (28–30). The health consequences of PCOS include infertility, accelerated atherosclerosis, diabetes, and endometrial carcinoma, underscoring the importance of detection and treatment.
Women with epilepsy appear to be at risk for developing features of this syndrome, although there is no study in a cohort of women with epilepsy that is adequately designed to permit an accurate diagnosis of this syndrome. Polycystic-appearing ovaries and hyperandrogenism are reported to arise in as many as 40% of women with epilepsy receiving VPA (13,17,19) and may be more likely to occur in women who receive VPA at puberty (31). In a random sample of 20 women with epilepsy of temporal lobe origin, five had PCOS, characterized clinically by oligomenorrhea, hirsutism, and androgen and LH elevation, or by ovarian cysts visualized directly or by ultrasonography (32). Another evaluation of 50 women with partial seizures arising from the temporal lobe found that 28 had menstrual cycle disturbance and 19 had reproductive endocrine disorders and polycystic ovaries (20).
In a prospective assessment of 94 reproductive-age women with epilepsy, polycystic ovaries were detected by transvaginal ultrasound in 26% of women with localization-related epilepsy, 41% of women with primary generalized epilepsy, and 16% of nonepileptic controls (11). Women receiving VPA within the preceding 3 years were more likely to have polycystic-appearing ovaries (38%) than women receiving other AEDs. This condition in women receiving VPA may be reversible when medication is changed to LTG (33).
The relative effect of epilepsy versus antiepileptic therapy can be considered by assessing reproductive health in persons receiving AEDs for conditions other than epilepsy. Two studies have assessed menstrual cycle regularity and ovarian morphology in women with bipolar disease. One study of women with bipolar disease treated with either VPA or lithium found no difference in length of the menstrual cycle or appearance of polycystic ovaries, although both groups had a high prevalence of abnormal menstrual cycle length (34). Another study assessed women with bipolar disorder who were treated with VPA or other agents, and reported abnormal menstrual length in 47% of those receiving VPA as compared with 13% of those not receiving VPA and 0% of healthy controls. Polycystic ovaries and elevated androgens were found in 41% of women with bipolar disease given VPA and in none of the other women with bipolar disease or the controls (35).
Additional evidence that these reproductive health disturbances are a consequence of epilepsy, as well as its treatment, comes from a study in female primates (36). Nonepileptic, regularly cycling, healthy primates were treated with VPA for 1 year, and achieved serum concentrations of VPA similar to those of adults with epilepsy. Over prospective 1-year assessment, the primates did not develop abnormalities in menstrual cycle length, ovarian morphology, or response to GnRH stimulation.
Data such as these suggest that epilepsy and some AEDs individually affect fertility and that these effects may be additive. This implies that the most sophisticated therapy for epilepsy will consider disease-treatment effects on reproductive health.
Men and women with epilepsy appear to have a higher incidence of sexual dysfunction than in other chronic neurologic illnesses; the dysfunction manifests primarily as diminished sexual desire and potency. Sexual dysfunction affects 30–66% of men with epilepsy (37–40) and 14–50% of women (38,41,42). More than one third of women with epilepsy report dyspareunia, vaginismus, and lack of vaginal lubrication, with normal sexual desire and experience (43).
Physiological sexual arousal has been quantitatively evaluated in several studies. Fenwick et al. found impaired nocturnal penile tumescence in a group of men with epilepsy and low testosterone (44). Men with localization-related epilepsy arising from the temporal lobe achieved inadequate penile tumescence and rigidity during REM sleep as determined by an in-home monitoring device (45). Both men and women with localization-related epilepsy arising from the temporal lobe had significantly lower increases in genital blood flow in response to an erotic audiovisual stimulus than did control subjects, even given normal subjective sexual arousal (46).
The cause of sexual dysfunction is probably multifactorial (see reference 47 for review). Social development is impaired in some individuals with epilepsy. Poor self-esteem as a result of having seizures may lead an individual to feel sexually unattractive. Sexual arousal may be negatively reinforced, particularly when sexual activity precipitates seizures or when sexual sensations or behaviors become identified as part of the seizure or postictal period. Realistic acceptance of psychosomatic aspects of a chronic illness is positively correlated with sexual function, whereas poor disease acceptance is often associated with sexual dysfunction (48).
Epileptic discharges in limbic structures may also contribute to sexual dysfunction. Sexual dysfunction usually arises after the onset of seizures (38,49,50) and may be more common in patients with partial, rather than generalized, seizures (40,49,51). Some patients treated for partial epilepsy with temporal lobectomy report postoperative improvement in libido and sexual potency, with the greatest improvement seen in those patients with the best seizure control (37,38,52).
Sexuality in people with epilepsy may be adversely affected by alterations in the pituitary gonadotropins and prolactin, and in the sex steroid hormones (20,53,54). Reductions in LH and elevated prolactin are associated with sexual dysfunction (55). Estrogen and progesterone must be present in adequate amounts to support sexual behavior in females (55). Low total and/or free testosterone has been correlated with sexual dysfunction in persons with epilepsy (25,54).
AEDs may contribute to sexual dysfunction by direct cortical effects or secondarily through alterations in the hormones supporting sexual behavior. Occasional or chronic impotence is most likely to arise in men using barbiturate AEDs, such as PB (56). Women with epilepsy receiving the CYP450 enzyme–inducing medications PHT, PB, or CBZ were significantly more likely to have sexual dysfunction than women receiving VPA or LTG (M.J.M, unpublished data, 2003).
When a patient presents with the complaint of sexual dysfunction, the clinician must consider the patient's somatic, psychological, and social well-being, as well as the dynamics of the couple and family (48). The frequency with which patients volunteer sexual complaints may depend to a great extent on the attitude of the physician (48). Given a complaint of sexual dysfunction, the patient should be questioned about precipitating factors, such as acute or chronic life stresses, recent medications, illnesses, surgery, or symptoms of depression. A recommended evaluation strategy includes a thorough physical and neurologic examination; thyroid function tests; testosterone, estrogen, prolactin, and LH levels; complete blood cell count; and fasting blood glucose measurement. Urologic or gynecologic consultation should be obtained.
Effects of AEDs on bone health
Some AEDs may alter bone mineral metabolism and compromise bone health, especially in women who have smaller bone mass. Women using PHT, PB, and perhaps CBZ and VPA are at higher risk for bone disorders such as osteopenia, osteomalacia, and fractures (57–59). A prospective study evaluating the risk of hip fractures in women >65 years found that women taking AEDs were two times more likely to have a hip fracture (60).
Bone biochemical abnormalities described in people with epilepsy include hypocalcemia, hypophosphatemia, elevated serum alkaline phosphatase, elevated parathyroid hormone (PTH), and reduced levels of vitamin D and its active metabolites (60–62). The most severe bone and biochemical abnormalities are found in patients receiving AED polytherapy (60,61) and in patients who have taken AEDs for a longer time (59).
AEDs may alter bone mineral metabolism by decreasing calcium and by increasing bone turnover. AEDs that interfere with intestinal calcium absorption could directly affect bone cell function, possibly through inhibition of cellular responses to PTH (57,62), but this cannot be the only mechanism. Reproductive-age women taking CBZ, PHT, and VPA had significantly reduced calcium levels compared with women receiving LTG, but only PHT was associated with increased bone turnover (63). Given available data, women with epilepsy should engage in good bone health practices, including adequate daily intake of calcium (1,200 mg/day) and vitamin D, gravity-resisting exercise, and bone density scans if they have taken PHT, CBZ, or VPA for ≥5 years. Bone density scans should be repeated at 3- to 5-year intervals in premenopausal women.
AEDs and lipid metabolism
Changes in lipid metabolism and body weight are associated with some AEDs and may cause long-term adverse health effects. CBZ, PB, and PHT increase high-density lipoproteins, CBZ has cholesterol-lowering effects, and PB and PHT may exert a similar cholesterol-lowering effect (64–67). Counteracting these favorable lipid trends, elevations in low-density lipoproteins are reported with CBZ and PB, and VPA increases LDLs as well as HDLs, leading to an unfavorable lipid profile. VPA-associated obesity and increases in insulin may account for VPA-associated dyslipidemia (33). Until the nature and mechanisms of AED-associated alterations in lipid metabolism are better understood, clinicians should monitor cholesterol and lipid profiles in persons receiving AEDs.