Authors :
Presenting Author: Hiroki Nariai, MD, PhD, MS – UCLA
Saarang Panchavati, BS – Radiological Sciences – UCLA; Atsuro Daida, MD, PhD – Pediatrics – UCLA; Benjamin Edmonds, MD – Pediatrics – UCLA; Makoto Miyakoshi, PhD – Psychiatry and Behavioral Neuroscience – Cincinnati Children's Hospital; Shingo Oana, MD, PhD – Pediatrics – UCLA; Samuel Ahn, MD – Pediatrics – UCLA; Corey Arnold, PhD – Radiological Sciences – UCLA; Noriko Salamon, MD, PhD – Radiology – UCLA; Raman Sankar, MD, PhD – Pediatrics – UCLA; Aria Fallah, MD, MS – Neurosurgery – UCLA; William Speier, PhD – Radiological Sciences and Bioengineering – UCLA
Rationale:
Neuromodulation with thalamic stimulation has shown promising results in the treatment of medication-resistant epilepsy, primarily due to the pivotal role the thalamus plays as a central hub within the cortical network. Animal and human studies have provided evidence supporting the critical involvement of the thalamus in the generation, maintenance, and termination of seizures. The clinical efficacy of deep brain stimulation targeting the anterior nucleus (AN) of the thalamus has been well-studied in the treatment of limbic epilepsy. Additionally, the effects of centromedian (CM) stimulation have been reported in Lennox-Gastaut syndrome. However, a limited number of studies to date have addressed the dynamic ictal changes of corticothalamic connectivity between thalamic nuclei (AN and CM) and various seizure foci.Methods:
To investigate the complex dynamics of the corticothalamic network during seizures, we examined both undirected and directional connectivity—incorporating coherence and spectral Granger causality analysis (GCA)—between the cortex and thalamic nuclei (AN and CM) across varied seizure foci. We carried out our analysis on 36 seizures from 10 patients aged 2-25, all of whom had medication-resistant focal epilepsy and were undergoing stereotactic EEG (sEEG) evaluation with thalamic coverage.
Results:
Based on the coherence group analysis, we demonstrated a significant increase in undirected connectivity during seizures in all frequency bands (delta, theta, alpha, beta, and gamma), with the delta band notably being the most prominent (Figure 1). In the GCA directional connectivity analysis, the inflow of information to the thalamus from the onsets increased across all frequency bands. In contrast, the outflows to each brain region from the thalamus were largely confined to ipsilateral regions and slower frequency bands (delta, theta, and alpha). During frontal lobe seizures, undirected connectivity increased throughout the brain regions, including the contralateral hemisphere. Conversely, in limbic seizures, undirected connectivity increased mainly in the ipsilateral brain regions (Figure 2). In frontal lobe seizures, the percentage change in delta coherence, as well as in delta GCA in both directions, was larger at CM than at AN. Conversely, in limbic seizures, the percentage change in delta coherence, as well as in delta GCA in both directions, was larger at AN than at CM.
Conclusions:
We have established that frontal lobe seizures predominantly involve CM over AN, while limbic seizures mainly involve AN rather than CM. It appears that the delta frequency plays a pivotal role in modulating the corticothalamic network during seizures. Our results underscore the significance of comprehending the unique role of each thalamic nucleus (AN and CM) during seizures, considering the anatomically distinct seizure onsets. This knowledge could guide personalized neuromodulation treatment strategies.
Funding:
The National Institute of Health, K23NS128318