Back to overview

Detection of epileptic activity in presumably normal EEG

Type of publication Peer-reviewed
Publikationsform Original article (peer-reviewed)
Author Baldini Sara, Pittau Francesca, Birot Gwenael, Rochas Vincent, Tomescu Miralena I, Vulliémoz Serge, Seeck Margitta,
Project Exploring brain communication pathways by combining diffusion based quantitative structural connectivity and EEG source imaging : application to physiological and epileptic networks
Show all

Original article (peer-reviewed)

Journal Brain Communications
Volume (Issue) 2(2)
Page(s) 1
Title of proceedings Brain Communications
DOI 10.1093/braincomms/fcaa104

Open Access

Type of Open Access Publisher (Gold Open Access)


AbstractMonitoring epileptic activity in the absence of interictal discharges is a major need given the well-established lack of reliability of patients’ reports of their seizures. Up to now, there are no other tools than reviewing the seizure diary; however, seizures may not be remembered or dismissed voluntarily. In the present study, we set out to determine if EEG voltage maps of epileptogenic activity in individual patients can help to identify disease activity, even if their scalp EEG appears normal. Twenty-five patients with pharmacoresistant focal epilepsy were included. For each patient, 6 min of EEG with spikes (yes-spike) and without visually detectable epileptogenic discharges (no-spike) were selected from long-term monitoring recordings (EEG 31–37 channels). For each patient, we identified typical discharges, calculated their average and the corresponding scalp voltage map (‘spike-map’). We then fitted the spike-map for each patient on their (i) EEG epochs with visible spikes, (ii) epochs without any visible spike and (iii) EEGs of 48 controls. The global explained variance was used to estimate the presence of the spike-maps. The individual spike-map occurred more often in the spike-free EEGs of patients compared to EEGs of healthy controls (P = 0.001). Not surprisingly, this difference was higher if the EEGs contained spikes (P < 0.001). In patients, spike-maps were more frequent per second (P < 0.001) but with a shorter mean duration (P < 0.001) than in controls, for both no-spike and yes-spike EEGs. The amount of spike-maps was unrelated to clinical variables, like epilepsy severity, drug load or vigilance state. Voltage maps of spike activity are present very frequently in the scalp EEG of patients, even in presumably normal EEG. We conclude that spike-maps are a robust and potentially powerful marker to monitor subtle epileptogenic activity.