Biophysical mechanisms of electroconvulsive therapy-induced volume expansion in the medial temporal lobe: A longitudinal in vivo human imaging study

Abstract

Background: Electroconvulsive therapy (ECT) applies electric currents to the brain to induce seizures for therapeutic purposes. ECT increases gray matter (GM) volume, predominantly in the medial temporal lobe (MTL). The contribution of induced seizures to this volume change remains unclear. Methods: T1-weighted structural MRI was acquired from thirty patients with late-life depression (mean age 72.5 ± 7.9 years, 19 female), before and one week after one course of right unilateral ECT. Whole brain voxel-/deformation-/surface-based morphometry analyses were conducted to identify tissue-specific (GM, white matter: WM), and cerebrospinal fluid (CSF) and cerebral morphometry changes following ECT. Whole-brain voxel-wise electric field (EF) strength was estimated to investigate the association of EF distribution and regional brain volume change. The association between percentage volume change in the right MTL and ECT-related parameters (seizure duration, EF, and number of ECT sessions) was investigated using multiple regression. Results: ECT induced widespread GM volume expansion with corresponding contraction in adjacent CSF compartments, and limited WM change. The regional EF was strongly correlated with the distance from the electrodes, but not with regional volume change. The largest volume expansion was identified in the right MTL, and this was correlated with the total seizure duration. Conclusions: Right unilateral ECT induces widespread, bilateral regional volume expansion and contraction, with the largest change in the right MTL. This dynamic volume change cannot be explained by the effect of electrical stimulation alone and is related to the cumulative effect of ECT-induced seizures.

Electroconvulsive therapy, electric field, neuroplasticity, and clinical outcomes

“…First, seizure is an important therapeutic component of ECT [ 56 ], but the impact of seizure on neuroplasticity and clinical outcomes was not assessed with this investigation. Seizure activity may be related to neuroplasticity with and without E-field generation (i.e., Metrazol therapy) [ 57 , 58 ]. The lateralization of ictal power with RUL electrode placement suggests that E-field and ictal power may be interrelated [ 59 ].…”

“…In recent work by Takamiya et al, E-field models were computed in thirty depressed subjects receiving RUL ECT, a subgroup of which incorporated DTI-derived anisotropic conductivity in the models. The E-field correlations with regional brain volumetric changes were similar between both models [ 57 ]. Future work will need to validate E-field modeling accuracy (with or without DTI) prior to clinical use (“E-field-informed ECT”).…”

Multimodal multi-center analysis of electroconvulsive therapy effects: brainwide gray matter increase without functional changes

“…Similar to a recent anatomical parcellation study from GEMRIC [14], the structural increase in patients receiving ECT was widely distributed across the brain. The fact that the highest statistical peak was found in the right hippocampal area could be attributed to the electrode placement and coinciding electric field strength [31-32], presumably due to the majority of the patient sample used in this study receiving RUL stimulation. Again, in line with the previous report from GEMRIC, the large-scale volume increase was not related to clinical response, as indicated by both the factorial ANOVA and multiple regression analysis.…”

Electroconvulsive therapy-induced volumetric brain changes converge on a common causal circuit in depression

“…In agreement with these preclinical ndings, recent large-scale studies in the Global ECT-MRI Research Collaboration (GEMRIC) dataset 15 found robust volume increases 16,17 in a wide range of cortical and subcortical regions, which correlated with the number of ECT sessions. Subsequent EF modeling based on the individual head MRI consistently demonstrated that the ECT-induced EF strongly correlated with volume increase [18][19][20] . These results veri ed that despite the widespread activation of the brain through seizure, the direct electrical stimulation effect of ECT is much more spatially selective and individually diverse than rst assumed.…”