Abstract
Women with Epilepsy and Infertility Have Different Reproductive Hormone Profile than Others.
Thomas SV, Sarma PS, Nirmala C, Mathai A, Thomas SE, Thomas AC. Ann Indian Acad Neurol. 2013;4:544–548.
PURPOSE: One-third of women with epilepsy (WWE) may experience infertility (failure to conceive after 12 months of regular unprotected intercourse). We aimed to compare the hormone profile of WWE and infertility (WWE-I) with that of WWE who had conceived earlier (WWE-F). MATERIALS AND METHODS: In the Kerala Registry of Epilepsy and Pregnancy, we compared the clinical and hormone profile of 50 WWE-I and 40 age-matched WWE-F. Subjects were examined and blood samples were drawn in follicular phase (1–14 days) for 21 WWE-I and 18 WWE-F, in luteal phase (15–30 days) for 23 WWE-I and 15 WWE-F and beyond 30 days for 6 WWE-I and WWE-F who had irregular cycles. RESULTS: The two groups were comparable regarding physical, epilepsy syndrome, duration of epilepsy, body mass index, and serum cholesterol levels. Menstrual periods were irregular for 6 WWE-I and 5 WWE-F. The WWE-I group (compared to the WWE-F group) had significantly (P < 0.01) higher levels of dehydroepiandrostenedione (2.0 ± 1.7 ug/mL vs. 1.0 ± 0.7 ug/mL) and luteinizing hormone-LH (26.4 ± 37.3 mIU/mL vs 9.9 ± 14.5 mIU/mL) and lower levels of progesterone (5.2 ± 9.2 ng/mL vs. 10.4 ± 13.4 ng/mL). There was no significant difference in the levels of FT3, FT4, thyroid stimulating hormone, prolactin, follicle-stimulating hormone (FSH), progesterone, testosterone, or androstenedione levels. The WWE-I had 8.5 times higher risk (95% confidence interval 1.2–59.9) of abnormal LH/FSH ratio. WWE who were on antiepileptic drugs (AEDs) (compared to WWE who were not on AEDs) had higher risk of elevated LH/FSH ratio. CONCLUSION: The hormone profile of WWE-I is significantly different from that of WWE-F. These variations need to be interpreted with caution as a causal relationship to epilepsy or use of antiepileptic drugs need to be established through further studies.
Commentary
The recent report by Thomas et al. showing different reproductive hormone profiles in infertile women with epilepsy and fertile women with epilepsy adds to the growing body of evidence supporting clinical endocrine disruption in epilepsy. The investigators evaluated 50 women with epilepsy who had not conceived within 1 year of trying, as well as 40 women with epilepsy who readily became pregnant. One set of reproductive hormone levels per patient were obtained at varying points during menstrual cycling, either in the luteal or follicular phases, during days 13–14 after the onset of menstrual bleeding, presumably during ovulation, or after 28 days since the last menstrual cycle if menstruation did not occur.
The numbers of samples in each portion of the cycle were small; therefore, analysis by cycle phase was limited. However, the findings across all cycle phases were elevated luteinizing hormone (LH) levels, an elevated LH to follicular stimulating hormone (FSH) ratio and elevated DHEA levels in infertile women with epilepsy compared to fertile women with epilepsy.
How important is it to examine these parameters according to phase of the menstrual cycle? It is an important consideration, given that there are significant cyclic changes for most female reproductive hormones. Testosterone levels are generally considered stable across the menstrual cycle for women. However, mean testosterone levels in normal cycling women increase from a mean of 15 ng/dL in the midfollicular phase to 23 ng/dL at ovulation and remain around 20 ng/dL in the midluteal phase (1). LH and FSH both peak at days 13–14, with LH always at slightly higher levels than FSH throughout the cycle. Progesterone levels are very low in the follicular phase and then rise rapidly after ovulation, remaining elevated throughout the luteal phase with an abrupt fall at menstrual onset.
What is the clinical meaning of the hormonal differences found in this study? The overall increased LH level in infertile women with epilepsy compared to the fertile group was driven by the increase in the follicular phase, and the increase in LH/FSH ratio was present throughout the cycle except for the ovulatory days 13 and 14. These findings are consistent with clinical picture on the spectrum of polycystic ovary syndrome, which is a syndrome of ovarian dysfunction associated with elevated androgens, anovulation-, and oligo-ovulation infertility.
The elevated DHEA is also consistent with PCOS; however, it is often normal in this syndrome. More usual for PCOS is elevated testosterone, which was not found in this study. DHEA is produced in the adrenal glands and the ovaries, and it is a precursor to testosterone; conversion takes place in the ovaries and in peripheral target cells. As ovarian follicular reserve decreases with age or from other causes, the DHEA production from the ovary also decreases (2). Therefore, two findings mitigate against primary ovarian dysfunction as a cause of infertility in this cohort, which are the normal testosterone levels and the elevated DHEA levels. Notably, other investigators have reported an elevated LF/FSH ratio (in the luteal phase) as a predictor of anovulation in women with epilepsy (3), but also found decreased DHEA in women taking the powerful CYP inducing antiepileptic drugs phenytoin (3) and carbamazepine (3, 4).
Do the findings in this study suggest a mechanism of infertility in women with epilepsy? The categories of causes of infertility for couples include female pelvic structural disease and uterotubal factors, male infertility including semen quality and motility disorders, and ovulatory dysfunction, which specifically accounts for approximately 20% of infertility in couples (5). Adequate progesterone levels of greater than 3 ng/mL in the midluteal phase at days 20–22 indicates that ovulation has occurred. In the current study, progesterone levels in the luteal phase were not provided. However, the mean progesterone level in the infertile group was 5.17 (SD = 9.19), but was much lower than that of the fertile group which was 10.42 (SD = 13.36). One cannot conclude much from these results; the possibility of increased rates of anovulation as a contributor to infertility cannot be ruled out.
Therefore, these results suggest that anovulation—not due to ovarian dysfunction but perhaps due to central reproductive hormonal irregularities—is the mechanism of infertility in women with epilepsy. This is somewhat satisfying with regard to connecting this gonadal dysfunction to brain influences, since epilepsy is a disease of the brain and not the gonads. Certainly, the findings in the exploration of reproductive dysfunction in epilepsy are complex and at times do not reconcile across studies. The homeostasis of the reproductive endocrine feedback system contributes to the subtlety of the findings. Should one evaluate these questions in detail within a patient's cycle or with broad strokes, grouping patients and ignoring cycle days to identify overall trends, as in this study? Notably, no effect of seizure frequency or of antiepileptic drug use—including dose or polytherapy, or valproate use specifically—was found to be associated with variant hormonal profiles This is contrary the author's own study showing antiepileptic drug polytherapy and phenobarbital to be specifically associated with infertility (6), but the sample size in the current study is small.
Some issues could be resolved via a prospective study of women with epilepsy compared to control women, who are both actively trying to get pregnant. Ovulatory status in these women could be compared by obtaining midluteal phase progesterone levels and by comparing frequency and timing of sexual activity. By monitoring seizure occurrence, the effect of seizures on ovulation and fertility could be determined. A measure of ovarian reserve, mullerian inhibitory substance, could also shed light on whether ovarian dysfunction is present. Antiepileptic drug influences could be analyzed given a larger sample size; the possibility that the lesion localizes to a pill bottle is not yet excluded.