| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Special Features |
Epilepsy Center, Department of Neurosciences (L.B., R.M., S.S.), Department of Endocrinology (R.V.), and Department of Obstetrics and Gynecology (C.D.C., C.N.), Federico II University, 80131 Naples, Italy
Address all correspondence and requests for reprints to: Leonilda Bilo, M.D., Ph.D., Traversa Michele Pietravalle 54, 80131 Naples, Italy. E-mail: lbilo{at}libero.it
Abstract
An increased frequency of reproductive endocrine disorders has been reported in women with epilepsy. A possible role of the seizure disorder or, alternatively, of the use of antiepileptic drugs (AEDs) has been suggested as the pathogenic mechanism. The aim of the present study was to assess the frequency of reproductive endocrine disorders in a series of women with epilepsy, examining the possible relationships of these disturbances with different epilepsy syndromes and AED treatment.
Fifty epileptic women, all of reproductive age and none pubertal, pregnant, or lactating, were submitted to clinical endocrinological evaluation, hormonal assessment, and ovarian ultrasonography. Subjects with abnormal findings in this preliminary study underwent additional evaluations. Reproductive endocrine disorders were diagnosed in 16 (32%), consisting of polycystic ovary syndrome in 13, hypothalamic amenorrhea in 2, and luteal phase deficiency in 1. There was no significant association of these disturbances with epilepsy type or AED treatment. Patients with reproductive endocrine disorders often showed delayed ovulation with shortened luteal phase.
The results of this study suggest that the prevalence of disordered ovulation, in particular polycystic ovary syndrome, is increased in epilepsy, independent of antiepileptic medications or type of seizure disorder.
EPILEPSY IS A major public health problem in all countries. Recent epidemiological studies in industrialized countries estimate its prevalence at 5.0–9.3/1000 (1). This disease has a strong personal, familial, and social impact, and in addition to seizures, patients and their families may have to face many other problems. For instance, impaired fertility is an additional problem of women with epilepsy, as suggested by epidemiological studies (2, 3, 4, 5). An increased frequency of reproductive endocrine disorders, in particular of polycystic ovary syndrome (PCOS), has been described by Herzog et al. in women with temporal lobe epilepsy (6, 7) and by ourselves in women with idiopathic generalized epilepsy (8). Successively, Isojarvi et al. (9, 10) reported a high frequency of polycystic ovaries and/or hyperandrogenism in a large series of women with epilepsy.
The pathogenic mechanisms underlying the association between epilepsy and reproductive endocrine disorders are not yet clear. A pathogenic role of epilepsy itself has been proposed by some researchers (7, 8, 11), whereas others (9, 10) suggest that the use of antiepileptic drugs (AEDs), in particular sodium valproate (VPA), may mediate this association.
The aim of the present study was 1) to evaluate the reproductive endocrinological status in a group of women with epilepsy, assessing the frequency of reproductive endocrine disorders and their distribution in the different epileptic syndromes, and 2) to examine the possible relationships between reproductive endocrine disorders and AEDs.
Subjects and Methods
Selection of subjects
Fifty consecutive women with epilepsy referred to our Epilepsy Center entered the study, provided that they fulfilled the following criteria: of reproductive age, at least 6 yr postmenarche, not pregnant or lactating, not affected by systemic or central nervous system illnesses (apart from epilepsy) that could interfere with hypothalamic-pituitary-ovarian function, and not receiving any hormonal or psychotropic medications apart from AEDs. The mean age of the patient group was 23.4 ± 6.2 yr (range, 16–42). The epileptic disorders were classified in accordance with the International Classification (12). Twenty-five (50%) patients were affected by idiopathic or cryptogenic generalized epilepsy (GE), and 25 patients (50%) were affected by symptomatic or cryptogenic localization-related (focal) epilepsy (LRE). Sixteen patients (32%) were untreated, whereas 34 (68%) were given AEDs. In particular, 19 patients were receiving monotherapy [3 with VPA, 9 with phenobarbital (PB), and 7 with carbamazepine (CBZ)], and 15 were receiving polytherapy (6 with VPA and PB, 4 with VPA and CBZ, and 5 with PB and CBZ). In the 3 months preceding our study and throughout the whole study period, all AED treatments remained constant.
The control group consisted of 18 healthy female volunteers, all of reproductive age, at least 6 yr postmenarche, not pregnant or lactating, with regular menstrual cycles, and not receiving any hormonal or psychotropic medication. Their mean age was 27.4 ± 5.6 yr (range, 18–39 yr).
Informed consent was obtained in all cases, and the study was conducted in accordance with the guidelines laid down in the Declaration of Helsinki. The protocol was approved by our department’s committee on the study of human beings.
Methods
Basal study. In all patients a detailed menstrual history was obtained. Regularity of menstrual cycles was evaluated on the duration of the three cycles preceding the hormonal study and the cycle in which the hormonal study was performed (basal cycle). Regular cycles were defined as those lasting at least 21 days or at most 35 days, with no more than a 4-day variation in length from cycle to cycle in individual subjects (13).
A clinical endocrinological assessment was performed in all patients. Hirsutism was evaluated according to the criteria of Ferriman and Gallwey (14), with normal values below 7. Acne was considered present if currently active and involving the face, neck, or upper trunk or if there was evidence of facial, neck, or upper trunk scarring (15). The body mass index was evaluated according to the criteria of Bray et al. (16), with normal values in women ranging from 19–24.
The hormonal evaluation was performed in all patients in the early follicular phase of the basal cycle (days 5–7), evaluating serum levels of FSH, LH, PRL, 17ß-estradiol (E2), progesterone (P), testosterone (T), androstenedione (A), dehydroepiandrosterone sulfate (DHEA-S), T3, T4, and TSH. P evaluations were repeated on days 21 and 25 of the basal cycle; if the cycle lasted less than 25 days, P evaluation was performed only on day 21. P values of 3 ng/mL or greater were considered indicative of ovulation (17). Amenorrheic patients were submitted to hormonal evaluation on random days, considering the day of the first sample as day 5 and performing the following ones on day 21 and day 25.
All patients underwent a transabdominal echography of the ovaries, performed within day 5 of the basal cycle. The ovarian morphology was defined as polycystic if there were 10 or more cysts, 2–8 mm in diameter, with a thickened stroma (18).
Follow-up study. A successive follow-up study was performed only in 23 patients who had shown menstrual irregularities and/or androgen elevation and/or abnormal echographic ovarian findings during the basal study. In these patients monitoring of daily basal body temperature (BBT) and frequent hormonal sampling for a whole cycle (additional cycle) were carried out. In particular, in 10 patients FSH, LH, E2, and P serum levels were evaluated daily, every other day, or every 2 days depending on length of the previous cycle. In the remaining 13 patients, only P levels were evaluated from samples obtained every 2 or 3 days, according to cycle length, from day 15 until menstrual flow. The day of ovulation was presumed on the basis of BBT charts and of variations of hormonal levels. In particular, ovulation day was defined as the day on which the rise of BBT was observed and accompanied, on the same day or within the following day, by doubling of P levels above the baseline. In patients in whom LH was also monitored, the presence of the LH peak on the day of BBT rise or on the day before was also considered as an indicator of ovulation. Ovulation was considered delayed when it occurred after day 19 (19). Luteal phase was considered shortened when lasting less than 12 days (20).
In hyperandrogenic patients serum levels of cortisol and
17
-hydroxyprogesterone (17-OHP) were also obtained. When 17-OHP
was raised, and/or when hyperandrogenism was particularly severe, ACTH
stimulation tests for cortisol, 11-deoxycortisol, 17-OHP,
corticosterone, T, A, and DHEA-S was performed.
PCOS was diagnosed in presence of echographic picture of polycystic ovaries, irregular menstrual pattern, clinical and/or biochemical hyperandrogenism, and exclusion of other diseases leading to this picture (21, 22). Hypothalamic amenorrhea (HA) was diagnosed in presence of long-standing oligoamenorrhea, normal echographic ovarian pattern, normal gonadotropin, PRL, E2, androgens, cortisol, and thyroid profile, and normal neuroradiological study (23). Luteal phase deficiency (LPD) (20) was diagnosed in presence of a long-standing history of short menstrual cycles with a shortened luteal phase, with normal androgenic and echographic pattern.
Control group. Control subjects were submitted to the same hormonal and echographic evaluation as the patients. Only two controls underwent a follow-up study (daily BBT monitoring and prolonged hormonal sampling of P levels from day 15 until menstrual flow), having shown a pattern of polycystic ovaries at the echographic study.
Hormonal stimulation tests
The ACTH stimulation test was performed between 0800–1000 h in the fasting state, with samples obtained three times in the basal condition and 5 and 6 h after a rapid iv injection of 200 µg ACTH-(1–17 alsactide, Synchrodyn, Hoechst, Frankfurt, Germany) as a long test (360 min) as proposed by New et al. (24).
Hormone assays
FSH, LH, PRL, and TSH were assayed using immunoradiometric assay (CIS Diagnostics, Tronzano Vercellese, Italy), whereas E2, P, T, A, DHEA-S, cortisol, 17-OHP, 11-deoxycortisol, corticosterone, T3, and T4 were evaluated using RIA methods [CIS Diagnostics (Tronzano Vercellese, Italy), RADIM (Pomezia, Italy), and ICN Biomedicals, Inc. (Costa Mesa, CA)]. All hormones were assayed in duplicate, and the samples from an individual woman were analyzed within the same assay to avoid significant interassay interferences. The intra- and interassay coefficients of variation for all hormones were less than 11.5%. All samples were centrifuged and stored at -80 C until assayed. The normal ranges in our laboratory, specific for the general population of our area, were used to evaluate the hormonal data.
Statistical evaluation
Fisher’s exact test was used to evaluate the intergroup distribution of non parametric findings. Parametric data were analyzed by ANOVA with the Bonferroni correction (two-tailed). P < 0.05 was considered significant. Data are presented as the mean ± SE.
Results
General data
Thirty-four patients (68%) had a normal reproductive endocrinological status. In particular, 25 showed regular menstrual cycles, normal ovarian echographic findings, and a normal hormonal profile in the basal cycle, with a luteal P rise consistent with an ovulatory cycle. In 2 patients, P was not increased on either day 21 or 25 of 25- and 27-day cycles despite a history of regular menstrual bleeding. As the remaining clinical, hormonal, and echographic findings were normal, these women were considered to have normal reproductive function. Finally, 7 patients with findings of irregular cycles or elevated androgens or polycystic ovaries in the basal cycle showed ovulatory cycles with adequate luteal phase in the follow-up cycle and were considered endocrinologically normal.
In the remaining 16 patients (32%), all of whom underwent the follow-up study because of irregular menstrual cycles and/or elevated androgen levels and/or abnormal echographic findings in the basal study, the additional investigations allowed diagnosis of a specific reproductive endocrinological disorder. In particular, PCOS was diagnosed in 13 patients (26%). All presented menstrual irregularities and/or disordered ovulation, clinical and/or biochemical hyperandrogenism, and echographic picture of polycystic ovaries. None of these patients had galactorrhea or objective signs of virilization or Cushing’s syndrome. Cortisol levels and thyroid profile were normal in all; 17-OHP was elevated in 2. The ACTH stimulation test ruled out late-onset congenital adrenal hyperplasia in these 2 patients and in the other 3 who presented with severe hyperandrogenism without 17-OHP elevation. Only 2 patients were overweight, and 1 was obese; the remaining 10 were of normal weight. Basal LH levels were elevated in 2. PRL levels were slightly elevated in 1.
HA was diagnosed in 2 patients. They had long-standing severe menstrual irregularities (oligoamenorrhea) and disordered ovulation. Both had normal echographic ovarian pattern and normal gonadotropin, PRL, estradiol, androgens, cortisol, and thyroid profile. None had acne or hirsutism or objective signs of virilization. One patient was obese, and the other was overweight. Cerebral magnetic resonance imaging was normal in both.
LPD was diagnosed in one patient. This subject had a long-standing history of short cycles, lasting from 19–24 days. Her BBT charts, available for several cycles, showed a shortened luteal phase, lasting from 5–8 days. Our basal and follow-up studies confirmed these findings (basal cycle: duration, 21 days, ovulatory; follow-up cycle: duration, 22 days, ovulatory, with luteal phase of 5 days). She had a normal echographic ovarian pattern, normal hormonal profile, and normal body mass index. She had no acne or hirsutism or objective signs of virilization.
All controls showed normal hormonal values in the basal cycle. Two of them showed polycystic ovaries at ultrasonography and consequently underwent a follow-up study, which showed normal ovulatory cycles with adequate luteal phase in both subjects.
Hormone levels in the early follicular phase (basal cycle) in the 18
normal controls were compared with those in the 34 endocrinologically
normal epileptic patients (subdivided into treated and untreated; Table 1
). The 16 epileptic patients affected
by reproductive endocrine disorders were necessarily excluded from this
statistical analysis. E2 and DHEA-S
levels were significantly lower in treated patients compared with
controls (P < 0.05).
|
In the group of PCOS patients (Table 2), the basal study showed anovulatory
cycles in 10 patients of 13 (76.9%). It must be underlined that during
the basal study the last P evaluation was performed on day 25,
independently of cycle duration. In the successive follow-up study,
however, anovulation was documented in only 5 of 13 subjects (38.5%).
In fact, when P levels were frequently sampled until menstrual flow
occurred, it was possible to document ovulation even when it occurred
after day 25. A pattern of delayed ovulation was observed in 8 of these
patients, and a shortened luteal phase was found in 6. The hormonal
profile emerging from the prolonged sampling in one patient (no. 8) is
shown in Fig. 1.
|
|
).
|
Distribution of reproductive endocrine disorders according to epileptic syndromes and AED treatment
Distribution of reproductive endocrine disorders and, in
particular, of PCOS was unrelated to the type of epilepsy (Table 4) or to AED treatment (Table 5; by Fisher’s exact test). As VPA
treatment has been frequently related to menstrual irregularities,
polycystic ovaries, and hyperandrogenism (9, 10), we
examined the association of VPA therapy with these individual findings
and with the clinical picture of PCOS (Table 6). No significant differences in the
distribution of these parameters was observed (by Fisher’s exact
test).
|
|
|
In the present study we confirm a high frequency (26%) of PCOS in women with epilepsy. The actual prevalence of PCOS in the general population has been considered a matter of debate (25), possibly also because of controversy regarding its diagnostic criteria. All of our patients fulfill the three minimal criteria recently proposed for diagnosis of PCOS (21, 22), consisting of irregular menstrual pattern, clinical or biochemical hyperandrogenism, and exclusion of other diseases leading to this condition. Using these criteria for definition of patients with PCOS, recent epidemiological studies have estimated the prevalence of PCOS in the general population as 4.0% (26) and 6.7% (27), which is significantly lower than the 26% observed in our series (P < 0.001).
For HA and LPD, as we describe only three patients with these findings, our data do not allow any specific epidemiological conclusion.
In a previous report we described a high frequency of reproductive endocrine disorders in a series of women affected by GE (8). In the present study reproductive endocrine disorders were found with similar frequency in women affected by GE and in those with LRE, suggesting that epilepsy type is not a relevant factor in determining the occurrence of these conditions. This is in disagreement with other reports (28) describing a significantly higher frequency of ovulatory dysfunctions in women with LRE compared with women with GE, but agrees with researchers (29) who described a higher than average number of anovulatory cycles in epileptic women, without significant differences between GE and LRE.
The assessment of ovulation dynamics in our series of epileptic patients with reproductive endocrine disorders shows most often a pattern of long cycles with delayed ovulation, giving rise to a clinical picture of oligoovulation. This is apparently in contrast with previous reports describing a high frequency of anovulatory cycles in women with epilepsy (28, 29). However, in these studies the ovulatory function was evaluated by means of BBT monitoring (28) and/or by means of a single P evaluation in midluteal phase (28, 29). These techniques are not optimal for the detection of ovulation, as BBT charts are often unclear, and P evaluation carried out only in the supposed midluteal phase may actually miss a delayed ovulation.
Finally, our study does not confirm the association between PCOS and the use of any specific AED, in particular VPA. However, we must underline that our series of patients is considerably smaller than the series in which the association of VPA with polycystic ovaries was described, in particular as far as patients in VPA monotherapy are considered (9, 10). Even with these limitations in mind, however, some considerations may be proposed. One of the most striking differences between our series and the patients reported by Isojarvi et al. (10) is in body weight. In fact, while in that series the occurrence of polycystic ovaries in VPA-treated patients was consistently associated with obesity, considered by the researchers as a key factor in the development of polycystic changes, our epileptic patients with PCOS were generally of normal weight even when they used VPA. It is possible that the VPA-induced weight gain described by Isojarvi et al. (10) might be influenced by an individual susceptibility, possibly related to ethnic, dietary, or environmental factors. The possibility that ethnicity may influence the neuroendocrine-metabolic dysregulation of PCOS patients has been suggested (30) in relation to peculiar patterns of insulin resistance described in Scandinavian populations.
However, the development of PCOS in epileptic subjects does not seem necessarily related to an increase in body weight, as it was observed in our series even in the absence of this factor. It must be underlined that obesity is described in only 50% of PCOS patients (22, 31) and that, for this reason, it is not considered a prerequisite in the development of this condition (30). On the contrary, lean PCOS patients are considered to be affected from the "authentic syndrome," as in the absence of the confounding influence of obesity, the neuroendocrine-metabolic dysregulation underlying PCOS may be fully observed (30). In this light, the neuroendocrine assessment of our series of epileptic subjects with PCOS is particularly interesting, as 77% of them were of normal weight. In our view the overrepresentation of PCOS in women with epilepsy might actually be caused by the epileptic disorder itself, which may act as a deranging influence on the hypothalamic control of reproductive function, whereas the VPA-induced weight gain may be regarded as a possible additional factor contributing to the clinical expression of PCOS.
The epilepsy-related derangement of the reproductive function is probably caused by a dysfunction of the activity of the GnRH pulse generator. This is suggested by experimental data reporting a substantial rearrangement of GnRH efferents in the brain of female rats after induction of epileptic seizures (32) and by evidence in humans suggesting that interictal and ictal paroxysmal discharges might spread within the hypothalamus, leading to derangement of its function (33, 34, 35). Our previous reports of accelerated LH pulse frequency in normally menstruating, untreated epileptic women (36, 37) support the possibility of a disturbing effect of epileptic discharges on GnRH pulsatility, independent from the use of AEDs and also detectable in epileptic patients with normal reproductive endocrine status.
It is possible that in the long run and under the effect of different additional factors, this subclinical dysfunction may give rise to a clinical reproductive endocrine disorder (11, 37) and in particular to PCOS. PCOS is a heterogeneous disorder with complex pathogenesis, whose origin is attributed to a peripheral, ovarian problem or to a central, hypothalamic defect (38, 39). Acceleration of the GnRH pulse generator, which has been suggested as an intrinsic defect of PCOS (30), could in epileptic subjects be related to spreading of paroxysmal discharges within the hypothalamus.
Apart from PCOS, other clinical entities might also be related to a seizure-related hypothalamic derangement. In our series we reported two patients with HA, which has a definite hypothalamic origin (40, 41), and one patient with LPD, the possible hypothalamic origin of which has been suggested (40, 41). It is consequently possible that epileptic disorders may play a pathogenic role in the development of these conditions.
In patients with epilepsy, the progression toward one or another
reproductive endocrine disease probably depends upon several factors,
related to the characteristics of the seizure disorders, the endocrine
susceptibility of the patient, and the use of different AEDs (Fig. 2). A possible influence of AED treatment
on the hormonal status is also suggested by the finding of reduced
E2 and DHEA-S levels in our series
of endocrinologically normal epileptic women receiving AED treatment
(Table 1
). Reduction of sex steroid levels during AED therapy has been
reported by several researchers (see Ref. 42 for review)
and has been attributed to the enzyme-inducing properties of many AEDs.
It is questionable whether the slight hormonal alterations observed in
our series of endocrinologically normal epileptic women might in the
long run play a role in the development of clinical reproductive
endocrine diseases, in particular as far as the reduction of
DHEA-S is concerned. In fact, in our series we report a
reduction of DHEA-S levels, whereas in classical PCOS
androgen levels are usually elevated. The reduction of androgen levels
induced by AED treatment might, on the contrary, have a potential
beneficial role in epileptic women with PCOS, as suggested by previous
reports (25, 43).
|
Received October 20, 1999.
Revised June 12, 2000.
Accepted February 20, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  R. D. Sheth and B. E. Gidal Topical Review: Optimizing Epilepsy Management in Teenagers J Child Neurol, April 1, 2006; 21(4): 273 - 279. [Abstract] [PDF]
|
|
|
|
 |
|
 J. R. Wood, V. L. Nelson-Degrave, E. Jansen, J. M. McAllister, S. Mosselman, and J. F. Strauss III Valproate-induced alterations in human theca cell gene expression: clues to the association between valproate use and metabolic side effects Physiol Genomics, February 10, 2005; 20(3): 233 - 243. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Salmi, H. C. Zisser, and L. Jovanovic Screening for and Treatment of Polycystic Ovary Syndrome in Teenagers Experimental Biology and Medicine, May 1, 2004; 229(5): 369 - 377. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. L. Nelson-DeGrave, J. K. Wickenheisser, J. E. Cockrell, J. R. Wood, R. S. Legro, J. F. Strauss III, and J. M. McAllister Valproate Potentiates Androgen Biosynthesis in Human Ovarian Theca Cells Endocrinology, February 1, 2004; 145(2): 799 - 808. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ferin, M. Morrell, E. Xiao, L. Kochan, F. Qian, T. Wright, and M. Sauer Endocrine and Metabolic Responses to Long-Term Monotherapy with the Antiepileptic Drug Valproate in the Normally Cycling Rhesus Monkey J. Clin. Endocrinol. Metab., June 1, 2003; 88(6): 2908 - 2915. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Joffe, A. E. Taylor, and J. E. Hall EDITORIAL: Polycystic Ovarian Syndrome--Relationship to Epilepsy and Antiepileptic Drug Therapy J. Clin. Endocrinol. Metab., July 1, 2001; 86(7): 2946 - 2949. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
