Authors :
Presenting Author: Stephanie Gollwitzer, MD – University Hospital Erlangen
Hajo Hamer, Professor – University Hospital Erlangen.de; Tamara Welte, MD – University Hospital Erlangen.de; Caroline Reindl, MD – University Hospital Erlangen.de; Stefan Rampp, MD – University Hospital Erlangen.de; Ruediger Hopfengaertner, MD – University Hospital Erlangen.de
Rationale:
Absence seizures as observed in genetic generalized epilepsies (GGE), especially in childhood and juvenile absence epilepsy (CAE, JAE) are non-focal onset seizures presenting as brief episodes of impaired behavioral responsiveness accompanied on EEG by characteristic frontally predominant 2.5–4Hz generalized spike-wave discharges (SWD). Differentiation between absence seizures and subclinical generalized SWD on EEG is essential for therapy management and patient counselling, especially regarding driving privileges. In this study we aimed to identify differences in the spectral composition of clinical and subclinical SWD.
Methods:
Video-EEG of patients with GGE and a history of absences was visually screened. Only SWD of at least three seconds duration unequivocally identified as either clinical or subclinical events based on behavioral testing recorded on video were selected. Testing paradigms comprised repetitive tapping, responding to verbal or acoustic stimuli or reading. As EEG signal over the midline showed lowest artifact contamination and high SWD amplitudes, we chose Fz-Cz electrodes using a bipolar reference for further analysis. Raw EEG of SWD epochs underwent power spectral analysis. Spectral power in delta (0.5-3 Hz), theta (4-7 Hz), alpha (8-12 Hz) and beta (13-35 Hz) bands was calculated as well as total power including the whole frequency range (0.5-35 Hz).
Results:
We analyzed 157 SWD in 42 patients with GGE (mean age 30.1 years, 31 female, 12 male, CAE: n=21; JAE: n=13; GGE with absences not further classifiable: n=8). A total of 95 SWD were classified as clinical, 62 as subclinical. In clinical absence seizures, power spectral analysis revealed a significantly higher delta power proportion of the spectrum (67.03%) than in subclinical SWD (47.56%; ROC: p< 0.0001, AUC 0.75). Theta power percentage was higher in subclinical episodes than in absences (33.09%, vs 21.05%; ROC: p< 0.0001, AUC 0.75). In addition, total power was significantly higher in clinical than sublinical events (mean power 16340.3 µV2 vs 6093.3 µV2, ROC: p< 0.0001, AUC 0.83).