Understanding seizure networks to improve outcomes in electroconvulsive therapy

About This Grant

PROJECT SUMMARY Electroconvulsive therapy (ECT) is a highly effective treatment for severe treatment-refractory depression and other conditions, in which carefully titrated electrical stimulation elicits brief, generalized seizures to change the brain to improve symptoms. An abundance of longitudinal MRI studies report robust and replicable brain plasticity after ECT, including increased hippocampal gray matter. However, it remains unclear how or why seizures are therapeutic in this context. Epilepsy research has demonstrated that seizure activity progresses through different brain regions and networks. Initial detection of seizure activity often occurs in a specific brain region (e.g., in medial temporal lobe), which can propagate locally, and in some cases spread via highly coordinated thalamo- cortical activity during generalization. Endogenous processes terminate the seizure, involving regions like anterior thalamus, basal ganglia, and cerebellum. A similar process appears to occur in ECT targeting temporal lobes, where electrical current initiates seizure activity in seizure-genic regions of medial temporal lobe (MTL), progressing to generalized seizure activity. Some seizure-network nodes have been implicated in antidepressant response to ECT, including parts of the hippocampus and thalamus. However, many seizure-network nodes are understudied in both ECT and epilepsy research in humans, because they are not included in standard MRI atlases (e.g., piriform cortex, substantia nigra, cerebellar nuclei) or due to limited spatial resolution in other neuroimaging techniques (e.g., coarse spatial resolution in molecular imaging, difficulty resolving deep structures in scalp EEG, limited number and position of pre-surgical recording electrodes in intracranial EEG). Thus, a precise, comprehensive understanding of seizure-network connectivity both in therapeutic seizure in ECT and pathological seizure in epilepsy remains elusive. The proposed studies will leverage pre-existing multi-modal MRI datasets to provide fundamental, mechanistic knowledge of entire seizure-network function before and after therapeutic and pathological seizure. The overall goal is to understand how seizure-network nodes interact in typical states and after therapeutic and pathological seizure, and to use that mechanistic knowledge to improve the administration of ECT, by using pre-treatment brain state to predict susceptibility and response to seizure therapy and by manipulating stimulus dose to influence the putative site of seizure initiation. Beyond improving the administration of ECT, the proposed studies have the potential to inform new neuromodulation strategies for depression, epilepsy, and other disorders, and to further knowledge of brain network function.