If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Correspondence to: Harvard Neuroendocrine Unit, Beth Israel Deaconess Medical Center, 422 Worcester Street, Suite 303, Wellesley, MA 02481, USA. Tel.: +1 781 431 0277; fax: +1 781 431 0274.
Seizure occurrence and numbers vary by the day and phase of the menstrual cycle.
•
Seizures vary by the ovulatory versus anovulatory status of menstrual cycles.
•
3 patterns of catamenial epilepsy: perimenstrual, peri-ovulatory, luteal phase.
•
Reproductive steroids have neuroactive properties that affect neuronal excitability and seizures.
•
Cyclic progesterone supplement may benefit perimenstrually exacerbated seizures.
Abstract
Purpose
To extend our knowledge and practical application of the concept of catamenial epilepsy.
Methods
The review focuses on the impact of the NIH Progesterone Trial on our understanding of the pathophysiology and treatment of catamenial epilepsy.
Results
Catamenial epilepsy refers to the cyclic exacerbation of seizures in relation to the menstrual cycle. An interaction between seizures and the menstrual cycle is suggested by variations in seizure frequency according to the day, phase and ovulatory status of the menstrual cycle. There are three commonly recognized patterns: perimenstrual (C1: Day −3 to +3), peri-ovulatory (C2: Day 10 to 3) and entire luteal phase in anovulatory cycles (C3: Day 10 to 3). Pathophysiological determinants include 1) the neuroactive properties of reproductive steroids, 2) the variation of neuroactive steroid levels across the menstrual cycle and 3) the differential susceptibility of epileptic substrates to neuroactive steroid effects. Perimenstrual seizure exacerbation may result from the premenstrual withdrawal of progesterone which is accompanied by withdrawal of allopregnanolone, a potent positive allosteric modulator of the GABAA receptor, and changes in the subunit composition of the GABAA receptor to the α4 subtype which is insensitive to benzodiazepine and GABA. Bioidentical progesterone supplement is no better than placebo in the treatment of women with focal onset epilepsy overall but shows superior efficacy in women whose seizures show robust perimenstrual exacerbation.
Conclusion
There is sound evidence for the existence of catamenial epilepsy and class 3 evidence for adjunctive progesterone treatment of the perimenstrually exacerbated subtype.
]. Although generally supportive evidence has been mounting since 1881 when William Gowers published his observation that the majority of the women in his clinic reported worsening of seizures perimenstrually [
], there is still some controversy regarding the existence and prevalence of catamenial epilepsy. Differences in opinion stem largely from differences in the definitions of catamenial epilepsy that investigators have used for designation. The concept of catamenial epilepsy is based principally on three pathophysiological determinants: (1) the neuroactive properties of reproductive steroids, (2) the variation of neuroactive steroid levels across the menstrual cycle and (3) the susceptibility of the epileptic substrate to neuroactive steroid effects [
]. If reproductive steroids have a role in seizure occurrence, it follows that reproductive steroids may also have a role in treatment. Yet, despite the 30% prevalence of individuals with intractable seizures [
], hormonal treatment trials have been lacking. The purpose of this review is to discuss how the findings of the NIH Progesterone Trial, the first large-scale, randomized, placebo controlled, multicenter hormonal treatment trial, may contribute to our understanding and practical application of the concept of catamenial epilepsy [
Comprehensive management of epilepsy may need to consider how seizures vary over time. Chronobiology, that investigates how physiological and pathological processes vary over time, has identified circadian rhythms that approximate the 24-h day length (e.g. wake-sleep cycles), ultradian rhythms that have periodicities that are much shorter than 24 h (e.g. the pulsatile secretion of gonadotropin releasing hormones, sleep cycles), and infradian rhythms that have periodicities that are much longer than 24 h (e.g. the menstrual cycle) [
]. Some infradian rhythms synchronize with or are entrained by external astronomical factors such as lunar (e.g. menstrual cycle) and solar (e.g. seasonal affective disorders) cycles [
]. Further, the time of predilection may vary with the location of the epileptic focus, e.g. nocturnal predilection of frontal lobe epilepsy versus morning and afternoon occurrence of temporal lobe seizures [
]. There is also evidence that seizure occurrence may show a circalunar periodicity. This may occur not only in 35% of women in relation to their menstrual cycles but also in 29% of men who show predilection for seizure occurrence at 3–6 week intervals [
]. The women, 13–45 years of age, had intractable focal onset seizures. They recorded the number of seizures that they had each day and the day of each menstrual onset during the 3-month baseline phase. Statistical analysis of the data using a cosinor method that fitted the daily seizure numbers to a sine wave using the method of least squares in order to identify any periodicity of seizure frequency and the “strength” of the periodicity as determined by its level of statistical significance, found that seizures in both ovulatory and anovulatory cycles showed circalunar periodicities of 28 days but anovulatory cycles also showed periodicities of 14 and 9 days [
]. These findings support the notion that the lunar cycle influences seizure occurrence. The lunar periodicity of menstrual cycles and seizures represent either a parallel but independent effects of the lunar cycle or parallel effects of the lunar cycle with interaction between seizures and the menstrual cycle. Differences in periodicities between ovulatory and anovulatory cycles would suggest the latter.
2.1.1 Day-by-day comparisons of seizures across the menstrual cycle
The NIH Progesterone Trial carried out day-by-day comparisons of all menstrual cycle days for seizure numbers and seizure occurrence regardless of numbers [
]. Results for average daily seizure frequency are presented in Fig. 1. The results of day-by-day comparisons are presented in color coded nomograms (Fig. 2, Fig. 3). The highest average daily seizure frequency and the day with the greatest likelihood of seizure occurrence was on Day 1 of the cycle, i.e. the day of menstrual onset, and the lowest on Day −8, i.e. during the midluteal phase. Seizure numbers were two times greater on Day 1 (average daily seizure frequency: 0.68) than on Day −8 (0.30). Likewise, seizure occurrence regardless of numbers were two times more likely on Day 1 (0.33) than on Day −8 (0.17). These findings are consistent with a number of previous studies that have shown greater seizure frequency perimenstrually than at other times during the menstrual cycle and the fewest seizures on a mid luteal day [
Fig. 1The distribution of seizure numbers varies across the days of the menstrual cycle (nested repeated measures ANOVA for the number of seizures on each day, p < 0.0001).
Fig. 2Directional changes and levels of statistical significance of comparisons of each day in the columns with each day in the rows are presented as a color coded nomogram with legend. The nomogram shows significantly more seizures occurring on Day 1 and fewer seizures on Day −8 of the cycle as compared to most other days of the cycle.
Fig. 3Directional changes and levels of statistical significance of comparisons of each day in the columns with each day in the rows are presented as a color coded nomogram with legend. The nomogram shows that the likelihood of seizure occurrence is significantly greater on Day 1 and less on Day −8 of the cycle as compared to most other days of the cycle.
2.2 Phase by phase comparisons of seizures across the menstrual cycle
The menstrual cycle consists of follicular and luteal phases which are divided by the principal events of the menstrual cycle, ovulation and menstruation [
]. These events are heralded by major changes in hormonal levels: a pre-ovulatory surge in estradiol levels and a premenstrual drop in progesterone levels [
]. On the basis of the neuroactive properties of estradiol and progesterone and the changes in their serum levels across the menstrual cycle, especially the proconvulsant effects of the estradiol surge and the anticonvulsant effects of progesterone withdrawal, one can further divide the cycle into 4 phases to determine whether, in fact, there is an increase in seizures during the peri-ovulatory and perimenstrual phases in comparison to the intervening mid-follicular and mid-luteal phases that could be used as baseline comparators [
]. Specifically, seizures would be more likely to occur perimenstrually, defined in our 1997 investigation arbitrarily as Days −3 to 3 and identified as catamenial type 1 or C1 pattern and during or right after the pre-ovulatory estradiol surge between Days 10 to −13, identified as catamenial type 2 or C2 pattern, than during the remainder of the cycle, i.e. mid follicular phase (Days 4–9) and mid luteal phase (Days −12 to −4) [
]. The investigators proposed that anovulatory cycles, by virtue of having higher estradiol to progesterone level ratios during the luteal phase, would show a different pattern of seizure exacerbation in comparison to ovulatory cycles with more seizures from the time of the preovulatory estradiol surge though the perimenstrual phase as compared to the mid follicular phase (Days 4–9). The 1997 study enrolled 184 women with intractable seizures of focal onset and compared seizure frequencies across the various phases of the menstrual cycle during a single cycle. It showed statistically significant support for the greater occurrence of seizures according to the C1 and C2 patterns during ovulatory cycles and the C3 pattern during anovulatory cycles (Fig. 4). The NIH Progesterone Trial investigated the rationale for the separate consideration of ovulatory and anovulatory cycles. Anovulatory cycles differ from ovulatory cycle by their lack of progesterone secretion during the luteal phase which generally yields high proconvulsant estradiol to anticonvulsant progesterone serum level ratios throughout the luteal phase which, unlike ovulatory cycles, would then be expected to result in greater seizure frequency throughout the luteal phase. The NIH Progesterone Trial investigated the sensitivity and specificity of the 3 patterns in distinguishing ovulatory from anovulatory cycles in the first 100 enrolled subjects [
]. Retention during baseline phase was 80%. Complete data were available for 157 cycles at the time of analysis. A midluteal progesterone level ≥5 ng/ml was used to identify ovulatory cycles. The study found that the association between C1 or 2 patterns and ovulatory cycles showed an 82.4% sensitivity and 95.7% specificity whereas the association between the C3 pattern and anovulatory cycles showed 76.2% sensitivity and 40.0% specificity. Ovulatory cycles which showed the C3 pattern had significantly higher mid luteal serum estradiol (pg/ml)/progesterone (ng/ml) ratios than those that showed C1 and/or 2 patterns: 12.5 ± 5.6 versus 8.0 ± 3.7. The lack of specificity of the C3 pattern for anovulatory cycles was not predicted and suggests that ovulatory, as well as anovulatory cycles, can show the C3 pattern if they have high mid luteal estradiol/progesterone ratios. Although the C3 pattern is not considered by the Herzog et al. 1997 criteria for catamenial designation in ovulatory cycles, the great majority (83.3%) of these ovulatory cycles, nevertheless, were still categorized as being catamenial on the basis of demonstrating the C1 or C2 patterns. Of note, C1 and C2 patterns coexist with each other in the same cycle in only 33.3% of ovulatory cycles that show one or other pattern.
Fig. 4Phase by phase comparisons of average daily seizure frequency across the menstrual cycle shows significantly more seizures during the perimenstrual (p < 0.001) and peri-ovulatory (p < 0.001) phases than during the mid-follicular and mid-luteal phases during ovulatory cycles, lending statistical support for the existence of the C1 and C2 patterns of catamenial seizure exacerbation in ovulatory cycles and significantly more seizures during the peri-ovulatory, mid-luteal and perimenstrual phases combined (p < 0.001) than during the mid-follicular phase in anovulatory cycles, lending support for the existence of the C3 pattern in anovulatory cycles.
]. Specifically, investigations have differed in the definition of the phase of the cycle with seizure exacerbation under consideration, as well as the cutoff levels for seizure increase required for designation. The Herzog et al. 1997 investigation demonstrated a relationship between the proportion of women who showed one of the patterns of catamenial exacerbation and the level of seizure exacerbation expressed as multiples of the average daily seizure frequencies during the phase of exacerbation relative to baseline comparator phases, i.e. the mid-follicular and mid-luteal phases for ovulatory cycles and the mid follicular phase for anovulatory cycles [
]. The relationship took the form of reverse S-shaped curves for each of the three patterns of catamenial exacerbation. This relationship suggested that there were large proportions of women with epilepsy who manifested some level of each of these patterns of catamenial seizure exacerbation and small proportions who had a high level of each of these patterns of catamenial seizure exacerbation. The points of inflection of these reverse S-shaped curves (C1: 1.69, C2–1.83 and C3–1.62) were determined and are considered to be the mathematically optimal cutoffs for exacerbation levels that distinguish women whose seizures show substantial versus little hormonal sensitivity.
Using the cutoffs provided by the points of inflection of the 3 reverse S-shaped curves for the designation of catamenial epilepsy, the 1997 investigation found that 78 of the 184 or 42.3% of the women demonstrated at least one of the 3 patterns of catamenial epilepsy (ovulatory cycles: C1–35.7%, C2–28.5%; and anovulatory cycles: C3–41.4%) [
]. The reverse S-shaped curves also help to reconcile some apparent discrepancies in past reports regarding the prevalence of catamenial epilepsy. Specifically the findings of a high prevalence reported in investigations by Tauboll et al. – 78% [
], all of whom defined catamenial epilepsy in terms of just more seizures perimenstrually as compared to other phases of the cycle, corresponded closely with our finding of 71.4% that fell at the high end of the curve for perimenstrual exacerbation, whereas the finding of Duncan et al. (12.5%) [
] who required a 6-fold increase in average daily seizure frequency for designation (i.e. that ¾ of seizures occur between Days −5 and 5 which is approximately one-third of an average cycle) was consistent with the low end plateau found in the curve at approximately 10% for threefold or greater seizure increase.
Using these cutoffs for the designation of catamenial epilepsy and the requirement that one or more be present in at least 2 of the 3 baseline cycles, the NIH Progesterone Trial found that 44.2% of the 294 randomized subjects who completed the 3-month baseline phase could be categorized as having catamenial epilepsy [
]. The prevalence of catamenial epilepsy by pattern was as follows: C1 pattern seizure exacerbation occurred in 39.8%, C2 pattern in 33.9% and C3 pattern, in 47.1%. These values compare quite closely to the findings of the 1997 study which, using the same cutoffs, showed the C1 pattern of seizure exacerbation in 35.7%, the C2 pattern in 28.5% and the C3 pattern in 41.4% of anovulatory cycles [
]. Enrollment bias for a hormonal treatment trial may account for the somewhat higher values in the NIH Progesterone Trial.
3. Pathophysiology of catamenial epilepsy
There are three critical pathophysiological determinants of catamenial epilepsy: (1) neuroactive properties of reproductive steroids, (2) variation of neuroactive steroid levels across the menstrual cycle and (3) susceptibility of the epileptic substrate to neuroactive steroid effects [
3.1 Neuroactive properties of reproductive steroids
The most consistent and compelling preclinical evidence with regard to the neuroactive properties of reproductive steroids that may be pertinent to catamenial epilepsy is that progesterone has a reduced metabolite, allopregnanolone, that is a potent positive allosteric modulator of GABAA neurotransmission [
] and that the withdrawal of progesterone or allopregnanolone, such as may occur premenstrually, changes GABAAR subunit composition to the α4δ subtype which is insensitive to benzodiazepine and GABA but remains sensitive to neurosteroid [
]. This would provide a mechanism for perimenstrual seizure exacerbation and a rationale for treatment trials with cyclic progesterone supplement that include gradual, rather than abrupt, progesterone withdrawal premenstrually.
The neuroactive properties of estradiol may also be pertinent. The potential importance of estradiol in the regulation of temporolimbic function is highlighted by the presence of the estradiol synthesizing enzymes, cytochromes P45017a and P450 aromatases, that are localized in neurons in the hippocampus and the measurement of hippocampal estradiol levels that surpass serum levels [
]. The pertinent actions of estradiol, however, are not as clear as for progesterone. Historical investigations have demonstrated the complexity of estrogen effects and the impact of the estrogen concentration, mode and site of administration, blood–brain barrier and epileptic substrate [
]. Assuming a proconvulsant effect of estradiol, mid cycle seizure exacerbation correlates temporally with and could result from the pre-ovulatory estradiol surge [
]. Entire luteal phase seizure exacerbation in anovulatory cycles and ovulatory cycles with high serum estradiol to progesterone level ratios could result from unopposed estradiol effect. In favor of a proconvulsant effect, estradiol potentiates a number of animal kindling paradigms [
Comparative effects of estradiol benzoate, the antiestrogen clomiphene citrate, and the progestin medroxyprogesterone acetate on kainic acid-induced seizures in male and female rats.
]. By way of underpinnings for this neuroexcitatory effect, there are a number of proposed candidate mechanisms. Estradiol may potentiate neuronal excitability by regulating neuronal plasticity, specifically by increasing the density of NMDA receptor-containing synapses on the apical dendrites of hippocampal CA1 pyramidal neurons via an N-methyl-d-aspartate receptor-dependent mechanism [
]. Estradiol, however, may also potentiate kainate, and quisqualate mediated neurotransmission, thereby implicating non-NMDA receptors in the short-term action of estrogen. Specifically, estradiol can produce a rapid and reversible increase in the amplitude of the Schaffer collateral-activated EPSP which can be blocked by non-NMDA, but not NMDA, antagonists [
]. A non-NMDA mechanism of action is supported by a more recent preclinical model which suggests that E2 binds ERβR to increase glutamatergic AMPAR mediated EPSPs [
]. The complex role of estrogen, however, is illustrated by evidence in some models that estradiol can raise seizure thresholds in the hippocampal region and provide neuroprotection against seizure induced injury [
3.2 Differences in seizure frequency between ovulatory and anovulatory menstrual cycles
The NIH Progesterone Trial findings provide additional support for the potential importance of neuroactive steroids. One of the tertiary outcomes of the trial was to determine whether seizure frequency differs between anovulatory and ovulatory cycles [
]. 92 of the women had at least one ovulatory and one anovulatory cycle during the 3 baseline months of the trial. Serving as their own controls, paired t-tests showed a statistically significantly greater secondary generalized tonic–clonic seizure average daily seizure frequency by 29.5% during their anovulatory cycles as compared to their ovulatory cycles. Complex partial and simple partial seizure frequencies did not differ significantly. For secondary generalized tonic–clonic seizures, the proportional differences in anovulatory/ovulatory average daily seizure frequency ratios correlated significantly with differences in mid-luteal anovulatory/ovulatory estradiol/progesterone serum level ratios. These findings support a possible role for reproductive steroid levels and ratios in secondary generalized tonic–clonic seizure occurrence. The finding of greater seizure frequency in anovulatory than ovulatory cycles is consistent with our past report as well as the reports of others [
3.3 Epilepsy substrates and the effects of epilepsy laterality and focality on the prevalence of catamenial epilepsy
Epileptic brain substrates differ from non-epileptic brain at the molecular biological (e.g. GABAARs) as well as gross anatomical (e.g. mesial temporal sclerosis, heterotopias) levels. At the molecular biological level, for example, GABAARs in the molecular layer of the dentate gyrus in a rat model of epilepsy differ from controls in that they have a preponderance of α4 γ2 subunits which are insensitive to both BDZ and neurosteroids since benzodaiepines lack activity at receptors that have the α4 subunit and neurosteroids preferentially bind the δ subunit [
]. Brain substrates may also show differential propensity for catameniality. As one of its tertiary outcomes, the NIH Progesterone Trial investigated whether the laterality and focality of the epileptic foci were factors in catameniality in the first 100 subjects who completed the baseline phase [
]. Principal investigators at each site determined the laterality and focality of the epileptic foci on the basis of all available pertinent clinical data and neurological tests including EEG, EEG monitoring, MR cranial imaging, interictal and ictal SPECT and PET. Of the 100 subjects, 71 were identified as having temporal lobe (left = 25, right = 29, bilateral = 17), 10 extratemporal, 14 mixed, and 5 indeterminate seizure foci. Extratemporal and mixed seizures occurred randomly across the cycle. Cosinor analysis determined the existence of periodicities of seizures during the three month baseline phase. Temporal lobe seizures (left = 875, right = 706) did not show a random distribution. Temporal lobe seizures occurred cyclically at intervals of 27 days with peak occurrence near the onset of menses. Laterality was a factor. Left-sided temporal lobe seizures occurred maximally and cyclically at onset of menses, whereas right sided temporal lobe seizures occurred randomly. This investigation determined that the circalunar rhythms of seizures in women vary with the neuroanatomic substrate of the seizure focus. Temporal lobe seizures, especially left-sided foci, are more susceptible to circalunar rhythmic occurrence than those arising from extratemporal or mixed foci.
4. Treatment
The NIH Progesterone Trial was a randomized, placebo-controlled, double-blind, clinical trial of progesterone versus placebo therapy in the treatment of intractable seizures in women with and without catamenial epilepsy [
]. The principal outcomes were the proportion of ≥50% responders and the change in seizure frequency between the 3-month baseline and 3-month treatment phases. Sample size of 640 was determined as the enrollment requirement to show a significant difference (p ≤ .05) between treatments for ≥50% responders with 80% power for 35% progesterone versus 15% placebo responders in the catamenial stratum. The large sample size was required since only about one-third of the women were expected to show a catamenial pattern of seizure exacerbation. Catamenial designation was based on the demonstration of catameniality in 2 of 3 baseline cycles using the Herzog et al. [
] established points of inflection cutoffs for designation of C1–C3 patterns. The trial enrolled only 462 women and randomized the 294 subjects who completed the baseline phase. Randomization was carried out seprarately for the catamenial and non-catamenial strata, 2:1 to progesterone or matching placebo treatment. Treatment regimen consisted of baseline optimal antiepileptic drug treatment plus adjunctive progesterone 200 mg lozenges or matching placebo. A whole lozenge was taken three times daily on days 14–25, ½ lozenge three times daily on Days 26–27, ¼ lozenge taken three times daily on Day 28 and then no lozenges until the next Day 14.
The findings of the NIH Progesterone Trial showed that cyclic progesterone is comparable to placebo in the treatment of intractable seizures in women with partial epilepsy [
]. A pre-specified secondary analysis identified a subset of women with perimenstrual seizure exacerbation who were responsive to progesterone treatment. This post hoc predictor analysis using binary logistic regression analysis (dependent variable being ≥50% progesterone responder: yes or no) found that the level of perimenstrual catameniality (C1 level) is a predictor of the efficacy of progesterone treatment. There was a significant interaction between C1 level and treatment. With increasing C1 levels, responder rates increased progressively from 21.3 to 57.1% for progesterone versus only 19.6% to 20.0% with placebo (Fig. 5A) . Changes in average daily seizure frequency progressed from −25.5 to −71.0% for progesterone versus only −25.0 to −26.3% for placebo (Fig. 5B). There was also significant interaction between C1 level and progesterone treatment for the most severe seizure type, secondary generalized tonic–clonic seizures and complex partial seizures but not simple partial seizures. The separation between responder rates for all seizures combined for progesterone (27.3%) versus placebo (14.3%) treatments was not significant at C1 level ≥1.69, the C1 cutoff level selected for designation to the catamenial stratum. The separation did achieve statistical significance at C1 level ≥2 (28.6% versus 12.9%) and at C1 level ≥3, the separation (37.8% versus 11.1%) was both significant (p = .0372) and achieved the anticipated clinically important separation goal of the trial, i.e. ≥35% responder rate for progesterone versus ≤15% responder rate for placebo.
Fig. 5(A) Percent ≥50% Responders in Relation to C1 Level: Progesterone versus Placebo.
This is a plot of ≥50% responders versus the level of perimenstrual seizure exacerbation (C1 level). C1 levels were determined during baseline and are expressed as multiples of the combined mid-follicular and mid-luteal seizure frequencies. Each level includes all women who had seizure exacerbation greater than or equal to that specific level of catameniality. With increasing C1 levels, the rate of ≥50% responders increased from 21.3% to 57.1% with progesterone treatment as compared to an increase of only 19.6% to 20.0% with placebo treatment. The anticipated primary outcome that 35% of catamenial progesterone treated versus 15% of placebo treated women would show a ≥50% reduction in seizure frequency is realized at C1 level ≥3 where 37.8% of progesterone treated as compared to 11.1% of placebo treated women were ≥50% responders (p = .0372). In comparison to the responder rate of the combined placebo group, the progesterone responder rates are significantly greater at each C1 level ≥3. (B) Percent Change in Average Daily Seizure Frequency in Relation to C1 Level: Progesterone versus Placebo. With increasing C1 levels from 1 to 10, the percent reduction in ADSF (mean ± SEM) progressed from 25.5% to 71.0% for progesterone as compared to 25.0–26.2% for placebo. Separation between the treatments reached significance at C1 levels ≥4. In comparison to change in ADSF in the combined placebo group, the changes in ADSF in progesterone treated subjects are significant at each C1 level ≥3. Prog, progesterone; Plac, placebo; C1, perimenstrual seizure exacerbation.
], 38.1% of the subjects had C1 level ≥1.69, 34.4% had C1 level ≥2 and 21.4% had C1 level ≥3 levels of perimenstrual exacerbation (Table 1). Of note, 12.2% had C1 level ≥6 which is almost identical to the 12.4% found in the Duncan study (Table 1) [
]. The findings suggest that 21.4% of women with intractable seizures, i.e. the percent that had C1 level ≥3 baseline, might be candidates for cyclic progesterone supplement.
Table 1Percent of women with various levels of perimenstrual seizure exacerbation.
C1 level
# WWE
% WWE
≥0
294
100.00
≥1
196
66.67
≥1.69
112
38.10
≥2
101
34.35
≥3
63
21.43
≥4
51
17.35
≥5
44
14.97
≥6
36
12.24
≥7
31
10.54
≥8
24
8.16
≥9
22
7.48
≥10
19
6.46
C1 level, level of perimenstrual seizure exacerbation; WWE, women with epilepsy.
Another tertiary outcome of the Trial was to determine whether allopregnanolone (AP) may mediate seizure reduction in progesterone treated women with epilepsy [
]. AP levels were significantly greater in treated than in baseline cycles for women treated with progesterone but not placebo, regardless of catamenial designation. There was a significant inverse correlation between changes in seizure frequency and changes in AP levels for the subset of subjects who showed a significantly greater responder rate in the post hoc analysis of the trial, i.e. subjects who had a threefold or greater increase in average daily seizure frequency perimenstrually as compared to the mid follicular and mid luteal phases (C1 ≥3): r = −0.442, p = .013 and specifically for C1 ≥3 progesterone treated subjects (r = −0.452, p = .035), but not other groups (C1 ≥3 placebo: r = −0.318, C1 <3 progesterone: r = 0.099, C1 <3 placebo: r = 0.131; p = NS). The findings support AP as a mediator of seizure reduction in progesterone treated women who have a substantial level of perimenstrually exacerbated seizures.
Failure of the trial to prove the principal hypothesis may relate to the design that attempted to treat 3 patterns of catamenial epilepsy which likely differ in pathophysiology with a single treatment regimen [
]. Specifically, cyclic progesterone supplement may have greater efficacy where progesterone withdrawal (C1 pattern), rather than estrogen surge (C2) or high luteal phase estradiol/progesterone serum level ratios (C3 pattern), are causally implicated. The design also assumed that the mathematically determined cutoff for catamenial designation would match the cutoff for a significant progesterone response. The absence of a significant difference between progesterone and placebo responders at the C1 cutoff level of ≥1.69 and finding of a significant difference at a clinically important level at C1 level ≥3 may suggest that there is a difference between the catamenial level that mathematically best distinguishes hormonally sensitive seizures and the level that distinguishes progesterone responders at a statistically significant and clinically important level. Actual enrollment of a larger sample size might have achieved a significant difference, i.e. 234 progesterone and 117 placebo-treated subjects might show the demonstrated C1 ≥1.69 progesterone responder rate of 27.3% versus placebo rate of 14.2% with p ≤ 0.05 and power of .80. Even with these larger numbers, however, the responder rate would still not achieve what we considered to be a clinically important response level of ≥35%.
Progesterone is a naturally occurring hormone with a long history of clinical use for the treatment of reproductive disorders, e.g. inadequate luteal phase cycles, infertility and postmenopausal hormone replacement [
]. This novel use of cyclic progesterone supplement as a treatment for women with epilepsy showed a favorable short term safety profile that did not differ significantly from placebo in the frequency and types of adverse events [
]. The profile is also favorable in comparison to AEDs. There were 22 serious adverse events (SAEs) among the 462 women, 13 during baseline and 9 during treatment. Of the 9 during the treatment phase, 6 occurred on progesterone and 3 on placebo, commensurate with the 2:1 randomization. The most common SAE (12/22) was hospitalization for seizures: baseline phase – 7, treatment phase – 5; progesterone – 2, placebo – 3. Two subjects died, one in baseline and one during treatment. The baseline death was by suicide, committed after the initial visit. The treatment phase death was attributed to SUDEP and considered unlikely to be related to progesterone treatment. The 8 other SAEs had singular occurrence. Five (nausea, fever and vomiting, cellulitis, insomnia, AED overdose) occurred during baseline and 3 (stomach flu, thyroid carcinoma, blurred vision) on treatment, all on progesterone but considered unlikely to be related to progesterone. There were no significant differences in proportions of occurrences of any adverse event between the two treatment groups within either stratum. Fatigue was the only adverse symptom that was reported by >10% of subjects during the treatment phase (range 11.1–16.5%) and the proportions were very similar in the progesterone and placebo groups.
The practical applications are that the detection of a ≥3-fold level of perimenstrual seizure exacerbation by charting of seizures and menses might suggest a favorable response to treatment with adjunctive cyclic progesterone supplement. The findings of the secondary analysis still require formal confirmation in an adequately powered investigation of the derived hypotheses. Periovulatory seizure exacerbation may not respond to progesterone or may require the start of treatment earlier in the cycle than Day 14 which was the start day in the NIH Progesterone Trial. Since an earlier start to progesterone supplement may disrupt the menstrual cycle and lead to intermenstrual bleeding, there may be a role for the use of synthetic GABAergic steroids, analogs of allopregnanolone such as ganaxolone that are devoid of reproductive hormonal properties and, therefore, can be used throughout the cycle in women as well as in men [
]. Alternatively, perhaps a more effective approach to suppression of the pre-ovulatory estrogen surge may include parenteral treatment with depomedroxyprogesterone [
]. Finally, the role of progesterone treatment in primarily generalized epilepsy, especially Juvenile Myoclonic Epilepsy, that is more common in women and tends to develop in adrenarchal and pubertal years, remains to be investigated.
Conflict of interest statement
The NIH Progesterone Trial was supported by a grant from the National Institute of Neurological Disorders and Stroke (NIH R01 NS39466).
Comparative effects of estradiol benzoate, the antiestrogen clomiphene citrate, and the progestin medroxyprogesterone acetate on kainic acid-induced seizures in male and female rats.
00052-7&id=gr1.jpg){kind=link}
00052-7&id=gr2.jpg){kind=link}
00052-7&id=gr3.jpg){kind=link}
00052-7&id=gr4.jpg){kind=link}
00052-7&id=gr5.jpg){kind=link}