| Epilepsy  

Predicting seizures based on cyclical activity data using chronic electroencephalographic recordings (cEEG)

The major problem for epilepsy patients is unpredictability of seizures attacks, which carry a risk of accidents and has negative impact on normal life activities. There is a great need for reliable tools and devices allowing to predict the onset of seizure clusters.

Chronic electroencephalographic recordings (cEEG) by implantable devices generate extensive data revealing that epileptic brain activity shows robust cycles, operating over hours (circadian) and days (multidien), but so far has not been appreciated for practical value (Leguia et al., 2021). In a recent study it was hypothesised that these cycles can be leveraged to estimate future seizure probability, and authors tested the feasibility of forecasting seizures days in advance (Proix et al., 2021). Retrospective analysis of cEEG data recorded with an implanted Responsive Neurostimulation System (RNS) in adults (age ≥18 years) with drug-resistant focal epilepsy at 35 centres across the USA between Jan 19, 2004, and May 18, 2018 was done. Existing cEEG data and seizure logs were screened for eligibility: patients were required to have at least 6 months of continuous hourly IEA data from cEEG without large gaps and 20 or more electrographic or self-reported seizures, but with 50% or less days with seizures, as the usefulness of forecasting in individuals with frequent seizures is likely to be low. The primary outcome was the percentage of patients with forecasts showing improvement over chance (IoC).

Authors included 18 and 157 patients in the development and validation cohorts, after screening 72 and 256 patients, respectively. Models incorporating information about multidien IEA cycles alone generated daily seizure forecasts for the next calendar day with IoC in 15 (83%) patients in the development cohort and 103 (66%) patients in the validation cohort. The forecasting horizon could be extended up to 3 days while maintaining IoC in two (11%) of 18 patients and 61 (39%) of 157 patients. Forecasts with a shorter horizon of 1 h, possible only for electrographic seizures in the development cohort, showed IoC in all 18 (100%) patients.

This is a groundbreaking study showing that cyclic seizure patterns could allow seizures to be predicted over much longer time frames than previously shown (days vs minutes or hours), and that the complex cycles can be simply extracted from accurate capture of the timing of events. The study confirms that the capture of event frequency and event times alone can lead to clinically useful seizure forecasting. Further prospective atudies are needed to fully validate this new approach for rutine clinical practice.



1: Leguia MG, Andrzejak RG, Rummel C, Fan JM, Mirro EA, Tcheng TK, Rao VR, Baud MO. Seizure Cycles in Focal Epilepsy. JAMA Neurol. 2021 Feb 8:e205370.  doi: 10.1001/jamaneurol.2020.5370. https://pubmed.ncbi.nlm.nih.gov/33555292

2: Proix T, Truccolo W, Leguia MG, Tcheng TK, King-Stephens D, Rao VR, Baud MO. Forecasting seizure risk in adults with focal epilepsy: a development and validation study. Lancet Neurol. 2021 Feb;20(2):127-135. doi: 10.1016/S1474-4422(20)30396-3. https://pubmed.ncbi.nlm.nih.gov/33341149/