Abstracts

Rationale: Previous studies suggest a relationship between the clinical semiology of generalized tonic-clonic seizures (GTCS), presence and duration of post-ictal generalized EEG suppression (PGES), and sudden unexpected death in epilepsy (SUDEP). Moreover, increased SUDEP risk has been associated with a rise in the electrodermal activity (EDA) in the setting of GTCS. This study examined the association of EDA signals and clinical semiology of GTCS in epilepsy patients. Methods: We asked patients to wear a portable wristband sensor that records EDA signals on wrists or ankles during video EEG long-term monitoring. Findings were correlated with seizures detected by video EEG monitoring. Patients who experienced GTCS during the recording interval were included.  Seizure videos were reviewed by two board-certified epileptologists and clinical and electrographic characteristics were collected. Based on ictal semiology, seizures were categorized into three groups: type 1 (with symmetric arm extension during tonic phase); type 2 (with asymmetric arm extension during tonic phase); type 3 (not classifiable in type 1 or 2) (Alexandre et al., Neurology, 2015; 85:1598-1603). EDA signals were analyzed in MATLAB software and the area under the curve (AUC) of the EDA signal around each seizure onset (from seizure onset up to 60 minutes after seizure onset) was calculated. Data analysis was performed by linear mixed effects regression model using SAS v9.4 allowing for nested events within subjects. Results: We prospectively enrolled 204 patients admitted to the epilepsy monitoring unit at Boston Children’s Hospital between February 2015 and September 2017. Thirty-four GTC seizures were recorded from 18 patients. Patients’ age ranged from 9 to 27 years (average=14.4). Epilepsy duration, MRI findings, and epilepsy type are depicted in Table 1. After semiologic classification, 10 seizures were categorized in group 1, 10 seizures in group 2, and 14 seizures in group 3. Average AUC measures were 1122736 µS x s (microsiemens x seconds), 93897 µS x s, and 102213 µS x s in group 1, 2, and 3 respectively. Comparison of AUC measures between 3 groups showed a trend in pairwise comparisons of group 1 and group 3 (P value= 0.05). Other pairwise comparisons between group 1 and 2 (P value= 0.13), and group 2 and 3 (P value= 0.74) were not significant.  Conclusions: In epilepsy patients, EDA signals correlated with clinical semiology of GTCS: greater EDA responses were observed in seizures with symmetric arm extension during the tonic phase of GTCS. Larger numbers and inclusion of additional covariates and confounders are needed to validate these findings. Funding: Epilepsy Research Fund