Depth-to-scalp spatiotemporal dynamics for stereo-EEG

Abstract

The data obtained from stereo-elecroencephalography (SEEG) in patients with focal epilepsy are crucial for defining the epileptogenic zone and achieving successful resection, but suboptimal electrode placement impairs SEEG results. We demonstrate an approach for concurrent scalp and depth EEG analysis from one patient with successful intracranial workup and one in whom the seizure onset zone was unsampled by SEEG. Intracranial epileptiform discharges were identified and clustered, their scalp correlates were averaged, and electric source imaging (ESI) was applied to the resulting averaged scalp potential - depth-to-scalp ESI (dsESI). We found temporal differences between intracranial and scalp peaks, as well as variations in averaged scalp spikes morphology and propagation, expressed by their amplitudes and width, and by their jitter across involved electrodes. Put together with the relative degree of focality and location of the averaged scalp spikes' ESI on the cortex, these data could differentiate onset from propagation of interictal activity and identify unexplored nodes in the epileptic network. Our novel analysis highlights the importance of temporal, and not just spatial, spike dynamics within the epileptic network, may be used to validate depth electrode placement and aid in understanding the epileptic network.

Keywords: Epilepsy; Epilepsy surgery; Interictal epileptiform activity; SEEG.

Fig. 1

Patient 1 was MRI-negative, investigated with bilateral temporal SEEG (A) that suggested a left basal temporal SOZ rapidly spreading to the hippocampi, and now over 3 years seizure-free with bilateral RNS. Three clusters of averaged iIEDs (B) are shown. The averaged scalp potentials (µV) are shown with time 0 as the intracranial peak, and a vertical red line at mid-peak (C) as well as their respective scalp topographies at mid-peak (µV) (D). Current density maps in 4 different views obtained via sLORETA (µA-m) at mid-peak are shown for each cluster in (E). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Patient 2 was MRI-negative, investigated with right fronto-perisilvian-insular SEEG (A), where all intracranial ictal activity was preceded by clinical seizure onset, therefore no surgical intervention was advised. Four clusters of averaged iIEDs (B) are shown. The averaged scalp potentials (µV) are shown with time 0 as the intracranial peak, and a vertical red line at mid-peak (C), as well as their respective scalp topographies at mid-peak (µV) (D). Current density maps in 4 different views obtained via sLORETA (µA-m) at mid-peak are shown for each cluster in (E). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)