The Role of Stress Hormones in Multiday Cycles of Seizure Risk
Abstract number :
3.183
Submission category :
2. Translational Research / 2A. Human Studies
Year :
2024
Submission ID :
767
Source :
www.aesnet.org
Presentation date :
12/9/2024 12:00:00 AM
Published date :
Authors :
Rachel Stirling, PhD – University of Melbourne
Jodie Naim-Feil, PhD – University of Melbourne
Ewan Nurse, PhD – Seer Medical
Dean Freestone, PhD – Seer Medical
Mark Cook, MD – The University of Melbourne
Wendyl D'Souza, MBChB, MPH, FRACP, PhD – St Vincent's Hospital Melbourne
David Grayden, PhD – The University of Melbourne
Presenting Author: Philippa Karoly, PhD – Biomedical Engineering, University of Melbourne
Rationale: Multiday cycles of seizure risk are now a well-established phenomenon in many people with epilepsy (PWE) [1]. However, very little is currently known about the mechanistic drivers of multiday seizure cycles, and few methods exist to assess this. In this work, we illustrate a novel methodology to assess the potential biomolecular drivers of seizure risk, and demonstrate its use in investigating stress hormones (salivary cortisol and DHEAS) as potential seizure risk biomarkers. The influence of stress on epilepsy is multifaceted and complex [2] but it is unknown if this influence extends to multiday seizure cycles.
Methods: Participants with epilepsy wore Fitbit smartwatches & reported seizures electronically. After 6 months, participants (blinded) provided saliva samples at high & low seizure risk times (cycle peak & trough, respectively) based on a personalized forecast [3]. At each sampling period, participants provided six saliva samples over three days (8am/8pm on each day). Samples were analyzed for cortisol & DHEAS. A stress questionnaire (Perceived Stress Scale, PSS) was completed at each sampling period. A summary of the protocol is shown in Figure 1.
Linear mixed models were fitted to predict cortisol, DHEAS and cortisol/DHEAS ratio levels with time of day, retrospective/prospective risk period, pre-sample seizure occurrence, post-sample seizure occurrence and PSS score as fixed effects. The model included age and gender as random effects nested under each participant.
Results: In this ongoing study, 15 participants (10 females) were recruited and 300 saliva samples were collected. On a group level, the standardised cortisol model (Figure 2A) revealed significant associations with time of day (2.95, 95% CI [1.92, 3.99], p < 0.001) (in line with known circadian fluctuations in cortisol), and pre-sample seizures (11.35, 95% CI [5.95, 16.75], p< 0.001). Across the cohort, PWE had high cortisol levels relative to the healthy normal range, and this result was particularly pronounced in the evening samples. Additionally, the cortisol/DHEAS ratio model (Figure 2B) revealed significant associations with time of day (-1.65, 95% CI [-1.85, -1.45], p < .001) and retrospective seizure risk (0.28, 95% CI [0.05, 0.51], p=0.019).
Translational Research