Ictal Involvement of the Pulvinar and the Anterior Nucleus of the Thalamus in Patients With Refractory Epilepsy.

Abstract

Background and objectives: Deep brain stimulation (DBS) targeting the anterior nucleus of the thalamus (ANT) has been shown to be effective in treating some patients with medically refractory epilepsy. However, it remains unknown how seizures spread through the ANT relative to other thalamic nuclei. This study aimed to investigate, through simultaneous recordings from both ANT and pulvinar (PLV) nucleus, their roles in seizure propagation. Our goal was to determine whether the ANT is the primary site of seizure propagation in the human thalamus, especially for focal seizure originating in the medial temporal lobe.

Methods: In a retrospective design, we studied EEGs and clinical notes of patients with refractory epilepsy who were implanted with stereo-EEG (sEEG) electrodes across cortical regions, some of which were extended to reach various sites of the thalamus (i.e., multisite thalamic recordings). We selected patients from the Stanford Comprehensive Epilepsy Center with both ANT and PLV electrodes and collected information about the timing and anatomy of seizure activity in the seizure onset zones, usually temporal, and the 2 thalamic sites.

Results: We recruited 17 (5 female, mean age 32 years) adult patients with simultaneous ipsilateral ANT and PLV recordings. In all patients, the procedure was safe without any complications. In 100% of patients, the thalamus was involved during seizures (in 88% both ANT and PLV and in 82% first the PLV). In patients with confirmed hippocampal or amygdalar onset seizures, 62% had initial involvement and 100% had subsequent involvement of the PLV nucleus. Only 31% showed initial propagation to ANT. All focal-to-bilateral tonic-clonic seizures and most of the focal impaired awareness seizures had early involvement of both ANT and PLV, with rapid spread to the contralateral nuclei.

Discussion: sEEG of thalamic nuclei simultaneously provides an opportunity to understand propagation patterns of seizures with respect to each thalamic subdivision at the individual level. The patterns of seizure propagation, as we report here, provide insights about the prominent involvement of the PLV nucleus during seizure propagation. This may motivate future prospective work in larger cohorts of patients to understand how thalamic propagation may predict response to resective/ablative surgery or whether personalization of DBS (for instance, PLV instead of, or together with, ANT) could improve clinical outcomes.