Rationale:
The thalamus is a central hub both for coordination with and between cortical areas in the brain. Despite its quintessential relationship with the cortex, it has received only passing interest in human epilepsy prior to the definitive trial for deep brain stimulation (DBS). The anterior nucleus of the thalamus (ANT) was chosen as a site for stimulation due to its contribution to the circle of Papez; however, it’s involvement in seizures relative to other nuclei have not been well studied.
In this study we describe thalamic propagation patterns of our first seventeen patients with multisite thalamic stereo-encephalography (sEEG) recordings in the thalamus, and their effects on clinical decision making.
Methods:
The Stanford Institutional Review Board approved the procedures and consents. Patients received depth recordings of the thalamus if they were potential candidates for DBS. Implantation plan considered known thalamic connectivity, and consisted primarily of ANT, mediodorsal (MD), and/or pulvinar (PUL) nuclei. Patients were excluded if there were not ANT and PUL targets included in at least one thalamus. Two senior epilepsy fellows reviewed thalamic and cortical EEG, as well as surgical case conference notes for analysis of decision making after sEEG. Discrepencies were reviewed by a staff epileptologist. Changes of clinical decision making were to include the following: 1) Responsive neurostimulation (RNS) used to target the thalamus and cortex, 2) DBS differed from bilateral ANT stimulation, 3) resective plans were altered, or 4) stimulation tolerability affected decision-making.
Results:
Twenty four consecutively enrolled patients between 7/23/2021 and 1/27/2023 were selected for thalamic sEEG. Seventeen patients had implantations in at least one thalamus of both ANT and PUL. Of 17 patients, 12 (71%) had a change in management based on thalamic implantation.
Nine (52%) constituted a change in the primary plan, and three (18%) in the backup plan.
Prominent thalamic involvement of the PUL occurred in nine (52%) of patients for which pulvinar was a stimulation target. Early involvement of both ANT and PUL, or MD and PUL nuclei occurred in four patients (24%), for which 4-lead DBS systems were implanted.
One patient (6%) had complex propagation patterns with differing thalamic nuclei targeted on the left and right.
One patient with late thalamic spread was selected for resection and was subsequently seizure free, and one patient did not tolerate PUL thalamic stimulation at time of sEEG using standard DBS settings, and was implanted with bilateral hippocampal RNS.
Conclusions:
Propagation of seizures through the thalamus follow multivarious pathways, and are not simply or predominantly through the anterior thalamus as previously assumed, even for temporal lobe epilepsy.
Thalamic ictal patterns changed management in the majority of our patients and yields surprising information on ictal spread, with common and early involvement of the pulvinar nucleus in many of our patients. Further work is needed to understand the effect of stimulating regions of the thalamus with known ictal spread on clinical outcome, compared to standard stimulation of the anterior nucleus of the thalamus alone.
Funding:
None