BRIEF COMMUNICATION
Distribution of seizures across the menstrual cycle in women with epilepsy *Andrew G. Herzog, *Kristen M. Fowler, †Michael R. Sperling, ‡Joseph M. Massaro, and Progesterone Trial Study Group1 Epilepsia, 56(5):e58–e62, 2015 doi: 10.1111/epi.12969
SUMMARY
Dr. Herzog is professor of neurology at Harvard Medical School and director of the Neuroendocrine Unit at Beth Israel Deaconess Medical Center.
The purpose of this study was to determine whether seizure frequency and cycle days with seizure occurrence vary across the menstrual cycle. The subjects were the first 100 women with intractable focal onset seizures, 13–45 years old, who completed the baseline phase of the National Institutes of Health (NIH) Progesterone Trial. Each subject recorded seizures and menses during a 3-month baseline phase. Data consisted of (1) seizure numbers for each cycle day and (2) cycle days with seizure occurrence. Statistical comparisons of seizure frequency and days with seizures were performed using generalized estimating equation one-way analysis of variance (ANOVA) and logistic regression followed by pairwise multiple comparisons of days based on the least square means. Seizure numbers and cycle days with seizure occurrence varied across the menstrual cycle. There was an approximately twofold difference between the highest (day 1) and lowest (day 8) values for both seizure frequency and days with occurrence. The demonstration of variation in seizure frequency and cycle days with seizure occurrence across the menstrual cycle, as well as identification of specific days that have substantially higher or lower frequencies than other days, supports the existence of catamenial epilepsy. KEY WORDS: Epilepsy, Seizures, Women, Catamenial, Clinical trials randomized controlled.
The term “catamenial epilepsy” refers to cyclic seizure exacerbation in relation to the menstrual cycle.1 Catameniality implies that reproductive hormones may influence seizures.1 The proposed concept and pathophysiology draw support from clinical findings of nonuniform patterns of seizure occurrence in a number of investigations,2–5 as well as basic science and animal experimental findings that show Accepted February 12, 2015; Early View publication March 30, 2015. *Harvard Neuroendocrine Unit, Beth Israel Deaconess Medical Center, Boston, Massachusetts, U.S.A.; †Thomas Jefferson University, Philadelphia, Pennsylvania, U.S.A.; and ‡Harvard Clinical Research Institute, Boston, Massachusetts, U.S.A. 1 Complete list of the Progesterone Trial Study Group members is in Appendix 1. Address correspondence to Andrew G. Herzog, Harvard Neuroendocrine Unit, Beth Israel Deaconess Medical Center, 422 Worcester Street, Suite 303, Wellesley, MA 02481, U.S.A. E-mail: [email protected]. edu Wiley Periodicals, Inc. © 2015 International League Against Epilepsy
that some reproductive steroids have neuroactive properties that can modulate glutamatergic and c-aminobutyric acid (GABA)ergic neurotransmission to impact neuronal excitability and seizure thresholds.6–9 Nevertheless, there has been controversy regarding the existence and prevalence of catamenial epilepsy.1 A review of the literature suggests that differences in the definitions of the phase or phases of the menstrual cycle under consideration for seizure exacerbation and the cutoffs for levels of seizure exacerbation that have been required for the designation of catamenial epilepsy may explain much of the apparent discrepancies in findings.1,4 The purpose of this study was to determine whether (1) average daily seizure frequency (ADSF) and (2) cycle days with seizure occurrence (CDWSO) vary across the menstrual cycle without the potential biases that are introduced by definitions of specific menstrual cycle phases or prerequisite levels of seizure exacerbation for catamenial designation.
e58
e59 Distribution of Seizures in Women
Methods Patients The subjects were the first 100 women with clinical and electroencephalography (EEG) features of localizationrelated epilepsy, 13–45 years old, who completed the baseline phase of the National Institutes of Health (NIH) Progesterone Trial, a large multicenter, randomized, placebocontrolled investigation of adjunctive cyclic progesterone therapy for intractable seizures.10 At entry, all of the women had two or more seizures per month despite trials of at least two first-line antiepileptic drugs (AEDs) and had menstrual cycle intervals between 23 and 35 days. None of the women took hormones or psychotropic medications during the 3 months preceding entry into the protocol. Standard protocol approvals, registrations, and patient consent Written consent was obtained from all participants. This study was part of a larger multicenter clinical trial that was approved by the institutional review boards of the Beth Israel Deaconess Medical Center and each of the participating study centers. Data collection and definitions Each subject completed a three menstrual cycle baseline phase during which she recorded seizures and menstrual onset on a calendar. The data consisted of (1) numbers of seizures on each cycle day and (2) days of the cycle on which seizures occurred. Cycle days were identified as follows: day 1, first day of menstrual flow; day 14, presumed day of ovulation. To normalize the cycle intervals, which ranged from 23–35 days when cycles were longer than 28 days, seizure numbers that occurred on and after day 14 were described as an average day 14 that comprised the total number of seizures on and after day 14 until day 14 divided by the number of days on which they occurred. All discrete seizures, including secondary generalized, complex partial, and simple partial, were counted. The protocol called for the review of seizure calendars at each site by the investigators each month at the monthly patient visit to ascertain to the extent possible that subjects were reporting epileptic events rather than nonepileptic events or drug side effects. Statistical analysis Seizure-menses calendar data were used to compute ADSF for each day of the cycle and the proportion of cycles with seizure occurrence, regardless of numbers of seizures, on each day of the cycle. Statistical comparisons of ADSF across the individual days of the cycle were performed using generalized estimating equation (GEE) one-way analysis of variance (ANOVA) followed by pairwise multiple comparisons between days. There was no adjustment for multiple comparisons. An unstructured correlation structure was
assumed for the within-subject correlations. The results (pairwise p-values) were then plotted as numerical and color-coded tables for ease of visualization of resulting patterns. A similar analysis comparing the proportions of cycles for CDWSO (yes/no) was carried out using GEE logistic regression. p-Values are two-sided and are considered statistically significant at the 0.05 level; SAS Version 9.3 (SAS Institute, Cary, NC, U.S.A.) was used to conduct the analysis.
Results The average age standard deviation (SD) of the subjects was 31.7 8.5 years. The 100 subjects provided complete seizure-menses calendar data for 292 of their 300 baseline phase menstrual cycles during which time they recorded 3,466 seizures. Secondary generalized motor (SGMS), complex partial (CPS), and simple partial (SPS) seizures were included in the analysis. Their numbers were as follows: SGMS, 475 (14%); CPS, 1,225 (35%); and SPS, 1,766 (51%). The median of the per-patient average monthly frequency is 6.7. The quartiles are 3.7–11.4. The minimum is 1.3 and the maximum is 99.3. Sixty-two percent of subjects had a history of secondary generalized convulsive seizures; 31% had one or more during the study. The average age at onset of epilepsy was 13.4 10.5 years of age; mean duration was 18.3 10.4 years. All were treated with AEDs: 65% on monotherapy and 35% on polytherapy. In descending order, the AEDs were carbamazepine (36 subjects), levetiracetam (16), lamotrigine (14), phenytoin (11), topiramate (9), oxcarbazepine (4), and valproate (4). Seizure numbers ADSF for each day of the menstrual cycle is presented in Figure 1A. The distribution of the ADSF varies across the days of the menstrual cycle, p < 0.0001. There was a greater than twofold difference between the day with the highest average daily seizure frequency and the day with the lowest: day 1 was 0.6838 versus day 8 was 0.2990; p < 0.001. Results of the pairwise multiple comparisons of ADSF for all cycle days are presented in Figure 2A. Greater ADSFs in comparison to other days of the cycle occurred between days 4 and +3, with the most significant results occurring for day 1. Lesser seizure frequencies in comparison to the other days of the cycle occurred on day 8 and, less notably, day 9 of the cycle. Seizure days The proportions of cycles for CDWSO likewise varied across the menstrual cycle, p < 0.001 (Fig. 1B). There was an almost twofold difference between the highest and lowest proportions for CDWSO: day 1 was 0.3299 versus day 8 was 0.1724, p < 0.001). Pairwise comparison showed significant differences between individual days of the cycle Epilepsia, 56(5):e58–e62, 2015 doi: 10.1111/epi.12969
e60 A. G. Herzog et al. A
Figure 1. (A) The 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 < 0.001). (B) Cycle days with seizure occurrence vary across the menstrual cycle (p < 0.001). Epilepsia ILAE
B
with respect to incidence rate regardless of seizure numbers (Fig. 2B). Seizures occurred more often on days 2 to +3 and less often on days 8 and less notably, day 7.
Discussion The findings support the existence of catamenial epilepsy by demonstrating variation in ADSF and proportions of cycles for CDWSO across the menstrual cycle without the potentially confounding definitions and cutoffs that may have contributed to the historical controversy. The findings are consistent with a previous chronobiologic analysis of the data, which found that during the 3-month baseline phase of the trial, seizures showed a circalunar periodicity that coincided approximately with the menstrual cycle interval.5 The present findings, which identify day 1 of the cycle as the day with the highest number of seizures and the day with the greatest likelihood of seizure occurrence, are consistent with the majority of studies, which have demonstrated that the Epilepsia, 56(5):e58–e62, 2015 doi: 10.1111/epi.12969
late luteal phase or perimenstrual phase of the cycle has a predilection for increased seizures2–5 and with robust basic and animal experimental evidence that progesterone withdrawal, as might occur clinically with the premenstrual drop in serum progesterone levels, may change the subunit composition of the GABAA receptor to a form that is less sensitive to benzodiazepines and GABA-mediated neurotransmission.6–9 The findings of the lowest seizure frequency and day with seizure occurrence on day 8 are consistent with previous clinical findings, which have demonstrated that seizure numbers vary with the ratios of serum estradiol/progesterone levels across the cycle and specifically with the lowest ratios being found in the mid-luteal phase.2 By way of potential limitations of the study, there is the consideration that enrollment of women for a hormone treatment trial might have a bias that favors recruitment of women with catamenial epilepsy. The proportion of women in this subset of the NIH Progesterone Trial population who met the Herzog et al. cutoffs for designation of catamenial epilepsy
e61 Distribution of Seizures in Women A
Figure 2. (A) Directional 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. (B) Directional 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. Epilepsia ILAE
B
in at least two of three cycles was 38%. This is actually less than the 42.3% demonstrated in the single menstrual cycle in the 1997 study of Herzog et al.4 in which no treatment was offered. The influence of seizure type (SGMS, CPS, or SPS) and ovulatory versus anovulatory status of menstrual cycles on the variation of ADSF will be the focus of further analysis.
Epilepsy & Behavior. Ms. Fowler received support as clinical trials specialist for the study from the NIH NINDS R01 39466. Dr. Sperling was a coinvestigator in this research that was supported by NIH NINDS R01 39466. Dr. Massaro received support for statistical analysis of the study data from the NIH NINDS grant R01 39466. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
References Acknowledgments This study was supported by the National Institute of Neurological Disorders and Stroke (NINDS) NIH R01 NS39466. Statistical analysis: conducted by Dr. Joseph Massaro, PhD, Harvard Clinical Research Institute, Boston, MA.
Conflict of Interest Dr. Herzog was the principal investigator on this research that was supported by NIH NINDS R01 39466. He serves on the editorial board of
1. Herzog AG. Catamenial epilepsy: definition, prevalence, pathophysiology and treatment. Seizure 2008;17:151–159. 2. Laidlaw J. Catamenial epilepsy. Lancet 1956;271:1235–1237. 3. Backstrom T. Epileptic seizures in women related to plasma estrogen and progesterone during the menstrual cycle. Acta Neurol Scand 1976;54:321–347. 4. Herzog AG, Klein P, Ransil BJ. Three patterns of catamenial epilepsy. Epilepsia 1997;38:1082–1088. 5. Quigg M, Fowler KM, Herzog AG, et al. Circalunar and ultralunar periodicities in women with partial seizures. Epilepsia 2008;49:1081– 1085.
Epilepsia, 56(5):e58–e62, 2015 doi: 10.1111/epi.12969
e62 A. G. Herzog et al. 6. Smith SS, Gong QH, Hsu F-C, et al. GABAA receptor alpha4 subunit suppression prevents withdrawal properties of an endogenous steroid. Nature 1998;392:926–929. 7. Maguire JL, Stell BM, Rafizadeh M, et al. Ovarian cycle-linked changes in GABAA receptors mediating tonic inhibition alter seizure susceptibility and anxiety. Nat Neurosci 2005;8:797–804. 8. Joshi S, Rajasekaran K, Kapur J. GABAergic transmission in temporal lobe epilepsy: the role of neurosteroids. Exp Neurol 2013;244:36–42. 9. Herzog AG, Frye CA, the Progesterone Trial Study Group. Allopregnanolone levels and seizure frequency in progesterone treated women with epilepsy. Neurology 2014;83:345–348. 10. Herzog AG, Fowler KM, Smithson SD, et al. Progesterone versus placebo therapy for women with epilepsy: a randomized clinical trial. Neurology 2012;78:1959–1966.
Appendix 1 The Progesterone Trial Study Group Dr. Andrew G. Herzog (Beth Israel Deaconess Medical Center, principal investigator); Kristen M. Fowler (Beth Israel Deaconess Medical Center, clinical trials specialist); Sarah D. Smithson (Beth Israel Deaconess Medical Center, clinical coordinator); Dr. Donald L. Schomer (Beth Israel Deaconess Medical Center, coinvestigator); Dr. Edward B. Bromfield and Dr. Barbara A. Dworetzky (Brigham & Women’s, site investigator); Sonia Replansky and Katherine Gleason
Epilepsia, 56(5):e58–e62, 2015 doi: 10.1111/epi.12969
(Brigham & Women’s, site coordinator); Dr. Alison Pack and Alison Randall (Columbia, site investigator); Dr. Cynthia Harden (Cornell, site investigator); Dr. Blagovast Nikolov (Cornell, site coordinator); Dr. Barbara Jobst and Dr. Gregory L. Holmes (Dartmouth, site investigator); Emily Clough and Tracy Ostler (Dartmouth, site coordinator); Dr. Page B. Pennell and Dr. Kimford Meador (Emory, site investigator); Melanee Newman (Emory, site coordinator); Dr. Gregory L. Krauss (Johns Hopkins, site investigator); Dr. Peter W. Kaplan (Hopkins, site investigator); Faith Mugai and Amanda Cole (Johns Hopkins, site coordinator); Dr. Eva Andermann and Dr. Frederick Andermann (Montreal Neurological Institute, site investigator); Dr. Suha Mercho (Montreal Neurological Institute, site coordinator); Dr. Teresa Tran (Minnesota Comprehensive Epilepsy Program [MINCEP], site investigator); Dr. Sabina Gapany (MINCEP, site coordinator); Dr. Michael R. Sperling and Dr. Joyce D. Liporace (Thomas Jefferson, site investigator); Sevie Shuman and Gwendolyn Manney (Thomas Jefferson, site coordinator); Dr. Laura A. Kalayjian and Dr. Christianne N. Heck (USC, site investigator); Sandra Oviedo and Guadalupe Corral-Leyva (USC, site coordinator); Dr. Marianna Spanaki (Henry Ford, site investigator); Dr. Tricia Ting (University of Maryland, site investigator); Karen Callison (University of Maryland, site coordinator); Dr. Nathan B. Fountain and Dr. Mark Quigg (University of Virginia, Charlottesville, Site Investigator); Dr. Cheryl Frye (University of Albany, director of Neurosteroid Assay Lab); Joseph Massaro (Harvard Clinical Research Institute, biostatistician); Brian Marquis (Freedom Drug, Boston, research pharmacist).
Distribution of seizures across the menstrual cycle in women with epilepsy *Andrew G. Herzog, *Kristen M. Fowler, †Michael R. Sperling, ‡Joseph M. Massaro, and Progesterone Trial Study Group1 Epilepsia, 56(5):e58–e62, 2015 doi: 10.1111/epi.12969
SUMMARY
Dr. Herzog is professor of neurology at Harvard Medical School and director of the Neuroendocrine Unit at Beth Israel Deaconess Medical Center.
The purpose of this study was to determine whether seizure frequency and cycle days with seizure occurrence vary across the menstrual cycle. The subjects were the first 100 women with intractable focal onset seizures, 13–45 years old, who completed the baseline phase of the National Institutes of Health (NIH) Progesterone Trial. Each subject recorded seizures and menses during a 3-month baseline phase. Data consisted of (1) seizure numbers for each cycle day and (2) cycle days with seizure occurrence. Statistical comparisons of seizure frequency and days with seizures were performed using generalized estimating equation one-way analysis of variance (ANOVA) and logistic regression followed by pairwise multiple comparisons of days based on the least square means. Seizure numbers and cycle days with seizure occurrence varied across the menstrual cycle. There was an approximately twofold difference between the highest (day 1) and lowest (day 8) values for both seizure frequency and days with occurrence. The demonstration of variation in seizure frequency and cycle days with seizure occurrence across the menstrual cycle, as well as identification of specific days that have substantially higher or lower frequencies than other days, supports the existence of catamenial epilepsy. KEY WORDS: Epilepsy, Seizures, Women, Catamenial, Clinical trials randomized controlled.
The term “catamenial epilepsy” refers to cyclic seizure exacerbation in relation to the menstrual cycle.1 Catameniality implies that reproductive hormones may influence seizures.1 The proposed concept and pathophysiology draw support from clinical findings of nonuniform patterns of seizure occurrence in a number of investigations,2–5 as well as basic science and animal experimental findings that show Accepted February 12, 2015; Early View publication March 30, 2015. *Harvard Neuroendocrine Unit, Beth Israel Deaconess Medical Center, Boston, Massachusetts, U.S.A.; †Thomas Jefferson University, Philadelphia, Pennsylvania, U.S.A.; and ‡Harvard Clinical Research Institute, Boston, Massachusetts, U.S.A. 1 Complete list of the Progesterone Trial Study Group members is in Appendix 1. Address correspondence to Andrew G. Herzog, Harvard Neuroendocrine Unit, Beth Israel Deaconess Medical Center, 422 Worcester Street, Suite 303, Wellesley, MA 02481, U.S.A. E-mail: [email protected]. edu Wiley Periodicals, Inc. © 2015 International League Against Epilepsy
that some reproductive steroids have neuroactive properties that can modulate glutamatergic and c-aminobutyric acid (GABA)ergic neurotransmission to impact neuronal excitability and seizure thresholds.6–9 Nevertheless, there has been controversy regarding the existence and prevalence of catamenial epilepsy.1 A review of the literature suggests that differences in the definitions of the phase or phases of the menstrual cycle under consideration for seizure exacerbation and the cutoffs for levels of seizure exacerbation that have been required for the designation of catamenial epilepsy may explain much of the apparent discrepancies in findings.1,4 The purpose of this study was to determine whether (1) average daily seizure frequency (ADSF) and (2) cycle days with seizure occurrence (CDWSO) vary across the menstrual cycle without the potential biases that are introduced by definitions of specific menstrual cycle phases or prerequisite levels of seizure exacerbation for catamenial designation.
e58
e59 Distribution of Seizures in Women
Methods Patients The subjects were the first 100 women with clinical and electroencephalography (EEG) features of localizationrelated epilepsy, 13–45 years old, who completed the baseline phase of the National Institutes of Health (NIH) Progesterone Trial, a large multicenter, randomized, placebocontrolled investigation of adjunctive cyclic progesterone therapy for intractable seizures.10 At entry, all of the women had two or more seizures per month despite trials of at least two first-line antiepileptic drugs (AEDs) and had menstrual cycle intervals between 23 and 35 days. None of the women took hormones or psychotropic medications during the 3 months preceding entry into the protocol. Standard protocol approvals, registrations, and patient consent Written consent was obtained from all participants. This study was part of a larger multicenter clinical trial that was approved by the institutional review boards of the Beth Israel Deaconess Medical Center and each of the participating study centers. Data collection and definitions Each subject completed a three menstrual cycle baseline phase during which she recorded seizures and menstrual onset on a calendar. The data consisted of (1) numbers of seizures on each cycle day and (2) days of the cycle on which seizures occurred. Cycle days were identified as follows: day 1, first day of menstrual flow; day 14, presumed day of ovulation. To normalize the cycle intervals, which ranged from 23–35 days when cycles were longer than 28 days, seizure numbers that occurred on and after day 14 were described as an average day 14 that comprised the total number of seizures on and after day 14 until day 14 divided by the number of days on which they occurred. All discrete seizures, including secondary generalized, complex partial, and simple partial, were counted. The protocol called for the review of seizure calendars at each site by the investigators each month at the monthly patient visit to ascertain to the extent possible that subjects were reporting epileptic events rather than nonepileptic events or drug side effects. Statistical analysis Seizure-menses calendar data were used to compute ADSF for each day of the cycle and the proportion of cycles with seizure occurrence, regardless of numbers of seizures, on each day of the cycle. Statistical comparisons of ADSF across the individual days of the cycle were performed using generalized estimating equation (GEE) one-way analysis of variance (ANOVA) followed by pairwise multiple comparisons between days. There was no adjustment for multiple comparisons. An unstructured correlation structure was
assumed for the within-subject correlations. The results (pairwise p-values) were then plotted as numerical and color-coded tables for ease of visualization of resulting patterns. A similar analysis comparing the proportions of cycles for CDWSO (yes/no) was carried out using GEE logistic regression. p-Values are two-sided and are considered statistically significant at the 0.05 level; SAS Version 9.3 (SAS Institute, Cary, NC, U.S.A.) was used to conduct the analysis.
Results The average age standard deviation (SD) of the subjects was 31.7 8.5 years. The 100 subjects provided complete seizure-menses calendar data for 292 of their 300 baseline phase menstrual cycles during which time they recorded 3,466 seizures. Secondary generalized motor (SGMS), complex partial (CPS), and simple partial (SPS) seizures were included in the analysis. Their numbers were as follows: SGMS, 475 (14%); CPS, 1,225 (35%); and SPS, 1,766 (51%). The median of the per-patient average monthly frequency is 6.7. The quartiles are 3.7–11.4. The minimum is 1.3 and the maximum is 99.3. Sixty-two percent of subjects had a history of secondary generalized convulsive seizures; 31% had one or more during the study. The average age at onset of epilepsy was 13.4 10.5 years of age; mean duration was 18.3 10.4 years. All were treated with AEDs: 65% on monotherapy and 35% on polytherapy. In descending order, the AEDs were carbamazepine (36 subjects), levetiracetam (16), lamotrigine (14), phenytoin (11), topiramate (9), oxcarbazepine (4), and valproate (4). Seizure numbers ADSF for each day of the menstrual cycle is presented in Figure 1A. The distribution of the ADSF varies across the days of the menstrual cycle, p < 0.0001. There was a greater than twofold difference between the day with the highest average daily seizure frequency and the day with the lowest: day 1 was 0.6838 versus day 8 was 0.2990; p < 0.001. Results of the pairwise multiple comparisons of ADSF for all cycle days are presented in Figure 2A. Greater ADSFs in comparison to other days of the cycle occurred between days 4 and +3, with the most significant results occurring for day 1. Lesser seizure frequencies in comparison to the other days of the cycle occurred on day 8 and, less notably, day 9 of the cycle. Seizure days The proportions of cycles for CDWSO likewise varied across the menstrual cycle, p < 0.001 (Fig. 1B). There was an almost twofold difference between the highest and lowest proportions for CDWSO: day 1 was 0.3299 versus day 8 was 0.1724, p < 0.001). Pairwise comparison showed significant differences between individual days of the cycle Epilepsia, 56(5):e58–e62, 2015 doi: 10.1111/epi.12969
e60 A. G. Herzog et al. A
Figure 1. (A) The 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 < 0.001). (B) Cycle days with seizure occurrence vary across the menstrual cycle (p < 0.001). Epilepsia ILAE
B
with respect to incidence rate regardless of seizure numbers (Fig. 2B). Seizures occurred more often on days 2 to +3 and less often on days 8 and less notably, day 7.
Discussion The findings support the existence of catamenial epilepsy by demonstrating variation in ADSF and proportions of cycles for CDWSO across the menstrual cycle without the potentially confounding definitions and cutoffs that may have contributed to the historical controversy. The findings are consistent with a previous chronobiologic analysis of the data, which found that during the 3-month baseline phase of the trial, seizures showed a circalunar periodicity that coincided approximately with the menstrual cycle interval.5 The present findings, which identify day 1 of the cycle as the day with the highest number of seizures and the day with the greatest likelihood of seizure occurrence, are consistent with the majority of studies, which have demonstrated that the Epilepsia, 56(5):e58–e62, 2015 doi: 10.1111/epi.12969
late luteal phase or perimenstrual phase of the cycle has a predilection for increased seizures2–5 and with robust basic and animal experimental evidence that progesterone withdrawal, as might occur clinically with the premenstrual drop in serum progesterone levels, may change the subunit composition of the GABAA receptor to a form that is less sensitive to benzodiazepines and GABA-mediated neurotransmission.6–9 The findings of the lowest seizure frequency and day with seizure occurrence on day 8 are consistent with previous clinical findings, which have demonstrated that seizure numbers vary with the ratios of serum estradiol/progesterone levels across the cycle and specifically with the lowest ratios being found in the mid-luteal phase.2 By way of potential limitations of the study, there is the consideration that enrollment of women for a hormone treatment trial might have a bias that favors recruitment of women with catamenial epilepsy. The proportion of women in this subset of the NIH Progesterone Trial population who met the Herzog et al. cutoffs for designation of catamenial epilepsy
e61 Distribution of Seizures in Women A
Figure 2. (A) Directional 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. (B) Directional 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. Epilepsia ILAE
B
in at least two of three cycles was 38%. This is actually less than the 42.3% demonstrated in the single menstrual cycle in the 1997 study of Herzog et al.4 in which no treatment was offered. The influence of seizure type (SGMS, CPS, or SPS) and ovulatory versus anovulatory status of menstrual cycles on the variation of ADSF will be the focus of further analysis.
Epilepsy & Behavior. Ms. Fowler received support as clinical trials specialist for the study from the NIH NINDS R01 39466. Dr. Sperling was a coinvestigator in this research that was supported by NIH NINDS R01 39466. Dr. Massaro received support for statistical analysis of the study data from the NIH NINDS grant R01 39466. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
References Acknowledgments This study was supported by the National Institute of Neurological Disorders and Stroke (NINDS) NIH R01 NS39466. Statistical analysis: conducted by Dr. Joseph Massaro, PhD, Harvard Clinical Research Institute, Boston, MA.
Conflict of Interest Dr. Herzog was the principal investigator on this research that was supported by NIH NINDS R01 39466. He serves on the editorial board of
1. Herzog AG. Catamenial epilepsy: definition, prevalence, pathophysiology and treatment. Seizure 2008;17:151–159. 2. Laidlaw J. Catamenial epilepsy. Lancet 1956;271:1235–1237. 3. Backstrom T. Epileptic seizures in women related to plasma estrogen and progesterone during the menstrual cycle. Acta Neurol Scand 1976;54:321–347. 4. Herzog AG, Klein P, Ransil BJ. Three patterns of catamenial epilepsy. Epilepsia 1997;38:1082–1088. 5. Quigg M, Fowler KM, Herzog AG, et al. Circalunar and ultralunar periodicities in women with partial seizures. Epilepsia 2008;49:1081– 1085.
Epilepsia, 56(5):e58–e62, 2015 doi: 10.1111/epi.12969
e62 A. G. Herzog et al. 6. Smith SS, Gong QH, Hsu F-C, et al. GABAA receptor alpha4 subunit suppression prevents withdrawal properties of an endogenous steroid. Nature 1998;392:926–929. 7. Maguire JL, Stell BM, Rafizadeh M, et al. Ovarian cycle-linked changes in GABAA receptors mediating tonic inhibition alter seizure susceptibility and anxiety. Nat Neurosci 2005;8:797–804. 8. Joshi S, Rajasekaran K, Kapur J. GABAergic transmission in temporal lobe epilepsy: the role of neurosteroids. Exp Neurol 2013;244:36–42. 9. Herzog AG, Frye CA, the Progesterone Trial Study Group. Allopregnanolone levels and seizure frequency in progesterone treated women with epilepsy. Neurology 2014;83:345–348. 10. Herzog AG, Fowler KM, Smithson SD, et al. Progesterone versus placebo therapy for women with epilepsy: a randomized clinical trial. Neurology 2012;78:1959–1966.
Appendix 1 The Progesterone Trial Study Group Dr. Andrew G. Herzog (Beth Israel Deaconess Medical Center, principal investigator); Kristen M. Fowler (Beth Israel Deaconess Medical Center, clinical trials specialist); Sarah D. Smithson (Beth Israel Deaconess Medical Center, clinical coordinator); Dr. Donald L. Schomer (Beth Israel Deaconess Medical Center, coinvestigator); Dr. Edward B. Bromfield and Dr. Barbara A. Dworetzky (Brigham & Women’s, site investigator); Sonia Replansky and Katherine Gleason
Epilepsia, 56(5):e58–e62, 2015 doi: 10.1111/epi.12969
(Brigham & Women’s, site coordinator); Dr. Alison Pack and Alison Randall (Columbia, site investigator); Dr. Cynthia Harden (Cornell, site investigator); Dr. Blagovast Nikolov (Cornell, site coordinator); Dr. Barbara Jobst and Dr. Gregory L. Holmes (Dartmouth, site investigator); Emily Clough and Tracy Ostler (Dartmouth, site coordinator); Dr. Page B. Pennell and Dr. Kimford Meador (Emory, site investigator); Melanee Newman (Emory, site coordinator); Dr. Gregory L. Krauss (Johns Hopkins, site investigator); Dr. Peter W. Kaplan (Hopkins, site investigator); Faith Mugai and Amanda Cole (Johns Hopkins, site coordinator); Dr. Eva Andermann and Dr. Frederick Andermann (Montreal Neurological Institute, site investigator); Dr. Suha Mercho (Montreal Neurological Institute, site coordinator); Dr. Teresa Tran (Minnesota Comprehensive Epilepsy Program [MINCEP], site investigator); Dr. Sabina Gapany (MINCEP, site coordinator); Dr. Michael R. Sperling and Dr. Joyce D. Liporace (Thomas Jefferson, site investigator); Sevie Shuman and Gwendolyn Manney (Thomas Jefferson, site coordinator); Dr. Laura A. Kalayjian and Dr. Christianne N. Heck (USC, site investigator); Sandra Oviedo and Guadalupe Corral-Leyva (USC, site coordinator); Dr. Marianna Spanaki (Henry Ford, site investigator); Dr. Tricia Ting (University of Maryland, site investigator); Karen Callison (University of Maryland, site coordinator); Dr. Nathan B. Fountain and Dr. Mark Quigg (University of Virginia, Charlottesville, Site Investigator); Dr. Cheryl Frye (University of Albany, director of Neurosteroid Assay Lab); Joseph Massaro (Harvard Clinical Research Institute, biostatistician); Brian Marquis (Freedom Drug, Boston, research pharmacist).
