Thalamocortical seizure onset patterns in drug resistant focal epilepsy

Abstract

Drug-resistant epilepsy affects tens of millions of people worldwide and is associated with considerable morbidity and mortality. Thalamic deep brain stimulation and cortical responsive neurostimulation are proven treatments for focal epilepsy. Both have been used to target a range of thalamic nuclei, yet the roles of these thalamic nuclei in focal seizure generation remain incompletely understood. Thirteen patients with drug-resistant focal epilepsy undergoing intracranial EEG were consented to undergo investigation of thalamocortical networks. Sampled regions included cortical, mesial temporal, and thalamic brain regions. Visual and spectral analyses were performed to identify seizure onset patterns and correlate thalamic and cortical seizure activity. Thalamic ictal discharges were observed in all patients, including synchronous thalamocortical seizure onset discharges with distinct onset patterns. These onset patterns ranged from hypersynchronous spiking, low-voltage fast activity, ictal baseline shifts, to broadband suppression. Multiple thalamic nuclei were involved in ictal organization and propagation, with the specific nuclei depending on the cortical seizure network. The thalamus plays a crucial role in focal onset seizure generation and propagation, with distinct seizure onset patterns and nuclei involved. These findings support exploring a broader range of thalamic nuclei in epilepsy neurostimulation and have implications for seizure detection settings in intracranial sensing devices.

Keywords: EEG; epilepsy; neurostimulation; seizure; thalamus.

Figure 1.. Simultaneous cortical and thalamic ictal onset recordings, corticothalamic onset delays, and ictal amplitude comparison.

(A) 3D reconstructions of sEEG and chronic ambulatory device electrode implantations surrounding an example of the raw EEG recording from an exemplar cortical seizure onset zone (hippocampus), as well as time-frequency analysis of the cortical seizure onset zone discharge. (B) Top - 3D reconstruction of a thalamic electrode implantation targeting anterior thalamic nuclei bilaterally, parcellated and colored individually according to the Morel atlas. Blue - anteroventral (AV) nucleus; red – mammillothalamic tract. Bottom - example bipolar recording from the same anterior nucleus of the thalamus (AV) with time-frequency analysis of the thalamic ictal discharge. (C & D) Individual corticothalamic ictal onset delays are plotted for each seizure for each subject, sEEG1–10 and RC+S 1–3. All delays are shown in C, while panel D shows only the 0–1 second range. (E) Root mean square (RMS) amplitude of ictal discharge in cortical seizure onset zone (SOZ; in red) and thalamus (blue) are averaged and plotted for each participant, and averaged again across the two groups (sEEG and RC+S^™).

Figure 2.. Corticothalamic propagation and seizure onset patterns.

All panels: top row = raw EEG, bottom row is time-frequency plot (spectrogram); left column = cortical seizure onset zone, right column = thalamus. Dashed lines = seizure onset, solid lines =seizure offset; red = cortical discharge, green = thalamic discharge. (A): seizure onset in the mesial temporal region is essentially synchronous with the thalamus (anterior nucleus), with pre-ictal spiking followed by low-frequency suppression and low voltage fast activity seen in both mesial temporal and thalamic regions. (B): seizure onset in the posterior insula with synchronous thalamic discharge, consisting of hypersynchrony (large spike) followed by low frequency suppression and low-voltage fast activity. The same progression – pre-ictal spiking, ictal baseline shift, and LVFA is seen in the ventral thalamus (ventral lateral nucleus) albeit with slightly different spectro-temporal characteristics to the cortex. (C): seizure onset in the dorsolateral frontal lobe (middle frontal gyrus) is synchronous with thalamic onset (mediodorsal nucleus), the latter manifested as a large ictal baseline shift/slow wave with marked suppression of other frequencies, and eventually hypersynchrony in the form of repetitive spiking (no high pass filter in C, to allow for viewing of slow wave). (D): seizure onset occurs in the lateral temporal region, while thalamic activity (ventral posterior lateral nucleus) is delayed by over 30 seconds; representing a propagation pattern.

Figure 3.. Thalamic ictal recruitment patterns.

(A-D) all panels, left to right: time-frequency plot (left); raw EEG (middle); schematic of sEEG electrode with most distal contact at bottom of image, anatomical location of electrode, and zoomed in version of raw EEG (5 seconds before and after cortical seizure onset; (right). For each panel A-D: rows represent different bipolar recording channels in sampled thalamus. Dashed lines = seizure onset, solid lines = seizure offset; red = cortical discharge, green = thalamic discharge. Different thalamic recruitment patterns are shown: (A) no definite thalamic involvement; (B) long period of focal-appearing activity in a thalamic propagation pattern before more diffuse recruitment; (C) a diffuse suppression of background frequencies, with a more focal onset with the thalamus in the form of a large spike and low voltage fast activity restricted to a single thalamic channel, with evolution of recruitment of different thalamic regions over time; and (D) more ‘diffuse’ thalamic recruitment at seizure onset, with focality limited to a large ictal baseline shift/slow wave (no high pass filter in D, to allow for viewing of slow wave).

Figure 4.. Corticothalamic seizure circuits.

Circuits are grouped according to seizure onset zone and earliest thalamic recruitment. Clockwise from top left: frontal to intra-laminar, medial, & ventral thalamus; insula to intralaminar and ventral group; central/peri-rolandic to intralaminar/ventral groups; parieto-occipital to posterior (pulvinar) group; lateral temporal to medial/lateral groups; mesial temporal to anterior/medial/lateral groups. For each circuit, electrode positions (sEEG or RC+S^™) are shown in 3D reconstructed models, while circuit schematics depicting functional connectivity based on ictal EEG data are shown by arrows.