Data-Driven Identification of Multidien Seizure Cycles in Women with and without Clinical Diagnoses of Catamenial Epilepsy
Abstract number :
2.288
Submission category :
4. Clinical Epilepsy / 4E. Women's Issues
Year :
2025
Submission ID :
1019
Source :
www.aesnet.org
Presentation date :
12/7/2025 12:00:00 AM
Published date :
Authors :
Presenting Author: Elaine Sinclair, DO, PhD – VCU Health
Emma Osterhaus, BS – University of Pittsburgh School of Medicine
Maxime Baud, MD, PhD – University Hospital, University of Bern
Cynthia Harden, MD – Xenon Pharmaceuticals Inc.
Jacqueline French, MD – Department of Neurology and Comprehensive Epilepsy Center, New York University Grossman School of Medicine, NYU Langone Health
Page Pennell, MD – University of Pittsburgh
Wesley Kerr, MD, PhD – University of Pittsburgh Neurology
Rationale: Catamenial epilepsy (CE) exists in 33-45% of women with epilepsy (WWE) and is defined as a twofold or greater increase in average daily seizure frequency during a phase of menstruation (C1: peri-menstrual, C2: peri-ovulatory, or mixed C1/C2). Clinical, hypothesis-driven criterion for CE aims to identify a relationship between seizure frequency and menstrual phase, but the relationship between this hypothesis-driven criterion and other data-driven approaches is unclear. In men and women with focal epilepsy, the data-driven approach of Morelet wavelet transformations applied to seizure (and interictal) count data reveals inherent seizure cycles, so this data-driven approach may also be used to reveal further characteristics of both CE and other causes of multidien seizure cycles in women. Thus we compared this data-driven approach to the clinical, hypothesis-driven criterion for CE, in a secondary analysis of seizure frequency and associated menstrual data obtained from women previously enrolled into the Women With Epilepsy: Pregnancy Outcomes and Deliveries (WEPOD) study.
Methods: The WEPOD study enrolled 89 women with epilepsy who were off contraception and attempting to conceive. Enrolled women tracked daily seizure frequency and menstrual data before and during pregnancy. At screening, women also commented on their perceived relationship between seizure frequency and menstruation. In a substudy of the WEPOD cohort, 23/89 women who recorded more than one menstrual cycle with more than one seizure in the pre-pregnancy phase were also analyzed for clinical CE or non-CE diagnoses. Here, Morelet wavelet transformations were then applied to pre-pregnancy seizure frequency data from 13/23 of these women, excluding those with less than 3 pre-pregnancy seizures, to identify weekly (7-14d), monthly (21-35d), or > monthly ( >35d) seizure cycles. Concordance between clinical and perceived CE diagnoses as well as wavelet-derived seizure cycles was assessed.
Results: Of the 13 women analyzed here, 46% (6/13) met clinical criteria for CE (C1: 4/6; C2: 1/6; mixed C1/C2: 1/6). Of the women with a C1 pattern, 50% had at least a monthly seizure cycle identified via wavelet analyses (Figure 1), whereas the other 50% had weekly but not monthly cycles identified. A weekly but not monthly cycle was identified in the one woman with a mixed C1/C2 pattern, but no significant seizure cycles were identified in the one woman with a C2 pattern. Of the remaining seven women without clinical CE diagnoses, wavelet analyses revealed monthly cycles in 29% (2/7), non-monthly cycles in 43% (3/7), and no significant cycles in 57% (4/7; Figure 2). Both clinical criteria and wavelet-identified seizure cycles had poor correlation with perceived CE diagnoses.
Conclusions: Morelet wavelet transformations identify multidien seizure cycles from daily seizure frequency diaries to supplement clinical CE criterion. In the future, this quantitative method may be used to improve the understanding of catamenial and non-catamenial multidien seizure cycles in WWE.
Funding: No funding supported this research.
Clinical Epilepsy