Abstracts

RATIONALE: Deep brain stimulation (DBS) of the centromedian nucleus (CM) or anterior nucleus (AN) of the thalamus has been proposed as a treatment for medically refractory epilepsy. Hypothetically, thalamic stimulation may prevent generalization of seizures through interruption of thalamic recruitment during seizure propagation. Additional long-lasting beneficial changes in thalamocortical network sensitivity could result from chronic stimulation. To address the extent to which the thalamus is indeed recruited in generalized seizures of presumed cortical onset, we performed combined scalp-thalamic EEG recordings in patients with intractable epilepsy post-insertion of DBS electrodes, prior to internalization of the electrodes and pulse generators.
METHODS: In pilot studies of DBS for epilepsy performed at the Toronto Western Hospital 2 patients have been treated with bilateral CM stimulation and 6 patients with bilateral AN stimulation. After implantation, continuous scalp-thalamic EEG recording was performed for 3-4 days prior to electrode internalization. 3 patients had spontaneous seizures recorded: 1 CM patient with symptomatic generalized epilepsy, 1 AN patient with multifocal or symptomatic generalized epilepsy and 1 AN patient with focal epilepsy and secondarily generalized convulsions.
RESULTS: One generalized tonic-clonic seizure was recorded in the CM patient: scalp EEG showed generalized low amplitude slow wave activity 2 seconds before clinical onset. Simultaneous CM recording showed, 500 msec prior to clinical onset, 1.5 sec after the first scalp EEG changes, rhythmic ictal activity in the left CM and 1 sec later in the right CM, persisting until seizure offset. 6 sudden onset drop attacks progressing to generalized tonic seizures were recorded in the second patient. A subtle generalized attenuation of the scalp background activity just prior to clinical onset was the only EEG finding: no rhythmic ictal activity was recorded from the scalp or AN electrodes. 31 focal motor seizures with secondary generalization were recorded in the third patient. A generalized alteration in scalp background activity predeeded by many seconds the first rhythmic ictal activity recorded from the right AN electrode, ultimately apparent synchronously in the scalp EEG over the right fronto-parietal region prior to bilateral scalp/AN involvement during generalization. Scalp/AN coherence analysis of the right hemispheric ictal activity before generalization showed a 24 msec cortical lead with 98% coherence.
CONCLUSIONS: Scalp-thalamic EEG showed early ictal recruitment of CM in a patient with symptomatic generalized epilepsy and of AN in a patient with focal onset seizures. Coherence analysis in the latter patient demonstrated a consistent cortical lead to the synchronous scalp/thalamic ictal activity. These findings support the use of thalamic DBS for epilepsy. Unexpectedly, no rhythmic ictal activity was recorded from scalp or thalamic electrodes in a patient with drop attacks and tonic generalization, suggesting that these seizures may be generated below the level of the diencephalon.
Objective: At the end of this activity participants should be able to discuss the role of the thalamus in maintenance and propagation of seizures and the rationale for thalamic DBS as a treatment for epilepsy.
[Supported by: Bloorview Epilepsy Research Program (RW)]; (Disclosure: Consulting - Drs. Wennberg and Lozano are consultants for Medtronic, Inc.)