Authors :
Presenting Author: Philippa Karoly, PhD – The University of Melbourne
Rachel Stirling, PhD – The University of Melbourne; Jodie Naim-Feil, PhD – The University of Melbourne; Ewan Nurse, PhD – Seer Medical; Honor Coleman, PhD – The University of Melbourne; Mark Cook, MBBS – The University of Melbourne; Benjamin Brinkmann, PhD – Mayo Clinic; Mark Richardson, BMBCh, PhD, FRCP – Kings College London; Dean Freestone, PhD – Seer Medical
Rationale:
Multiday seizure cycles are an important phenomenon in epilepsy, whereby intrinsic rhythms modulate the risk of epileptic activity over long timescales of weeks to months in an individual-specific way. Numerous studies have now shown that these underlying rhythms can be measured, both invasively (via chronic brain implants) and non-invasively (via wearable signals and seizure diaries). Methods for tracking seizure risk are highly promising for improving the clinical management of epilepsy and quality of life of people living with seizures. This study presents a mobile app for tracking seizure risk cycles, and prospectively validates its utility to improve both individual quality of life and clinical outcomes.
Methods:
Two applications of the seizure risk app were assessed: patient health engagement and pro-ictal scheduling. Pro-ictal scheduling uses multiday cycles of to optimize the timing of epilepsy monitoring for diagnostic or pre-surgical assessment. We validated pro-ictal scheduling in a large cohort (n = 5038) who underwent ambulatory video-EEG between 2019-2022. Clinical outcomes (neurologist assessment of clinical seizures and EEG) were compared in high-risk (pro-ictal) periods.
To assess patient health engagement, self-reported, quantitative measures of mood and stress were taken from a cohort of 111 users of a seizure risk app using validated survey tools in epilepsy. In-depth, semi-structured interviews were conducted with nine users, which focused on how the user managed seizures and uncertainty in epilepsy. An inductive thematic analysis was conducted on user interviews.
Results:
Pro-ictal scheduling improved clinical outcomes for video-EEG (Fig 1). Monitoring in high-risk periods was 25% more likely to detect interictal events (CI[1.137:1.370], p< 0.001) and 63% more likely to capture a confirmed electrographic seizure (CI[1.265:2.101], p< 0.001). In children, pro-ictal scheduling increased the capture of seizures by 93% (CI[1.297:2.885], p=0.002).
Patient-reported outcomes were also positive. Approximately 35% of users said they had used the forecast for helping mental state and/or scheduling (work-related, social or travel activities), and at least 45% of users said the forecast helped them to feel more in control, confident or settled. Users who selected that they had used the forecast for helping mental state had significantly higher BREASI (anxiety) and NDDI-E (depression) scores (Fig 2). Thematic analysis revealed common themes in living with unpredictability, frustrations with current management approach, desire to learn more and hope. In particular, 78% of interviewees enjoy using technology for seizure management, and 89% referred to a desire for data insights to further understand their condition.
Conclusions: This study provides prospective, real-world validation of a mobile app to track cycles of seizure risk. This non-invasive, freely available technology was found to improve clinical outcomes in diagnostic epilepsy monitoring, and may be beneficial for people with epilepsy.
Funding: Australian Government National Health and Medical Research Council
American Epilepsy Foundation 'My Seizure Gauge' grant