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

Abstract

Neurostimulation is a mainstream treatment option for major depression. Neuromodulation techniques apply repetitive magnetic or electrical stimulation to some neural target but significantly differ in their invasiveness, spatial selectivity, mechanism of action, and efficacy. Despite these differences, recent analyses of transcranial magnetic stimulation (TMS) and deep brain stimulation (DBS)-treated individuals converged on a common neural network that might have a causal role in treatment response. We set out to investigate if the neuronal underpinnings of electroconvulsive therapy (ECT) are similarly associated with this causal depression network (CDN). Our aim here is to provide a comprehensive analysis in three cohorts of patients segregated by electrode placement (N = 246 with right unilateral, 79 with bitemporal, and 61 with mixed) who underwent ECT. We conducted a data-driven, unsupervised multivariate neuroimaging analysis Principal Component Analysis (PCA) of the cortical and subcortical volume changes and electric field (EF) distribution to explore changes within the CDN associated with antidepressant outcomes. Despite the different treatment modalities (ECT vs TMS and DBS) and methodological approaches (structural vs functional networks), we found a highly similar pattern of change within the CDN in the three cohorts of patients (spatial similarity across 85 regions: r = 0.65, 0.58, 0.40, df = 83). Most importantly, the expression of this pattern correlated with clinical outcomes (t = -2.35, p = 0.019). This evidence further supports that treatment interventions converge on a CDN in depression. Optimizing modulation of this network could serve to improve the outcome of neurostimulation in depression.

Fig. 1. The relationship between Electric Field (EF) and volume change (ΔVOL).

On the left side of the figure, brain images illustrate the spatial distribution of EF (1st, 3rd, and 5th rows, V/m) and ΔVOL (2nd, 4th, and 6th rows, Cohen’s d). The first two rows depict results from RUL placement (n = 246), the next two rows show results from BT placements (n = 79), and the last two rows display results from MIX placements (n = 61). These regions were segmented into 85 cortical and subcortical areas (refer to methods). Throughout the manuscript, we designated the results from RUL as blue, BT as green, and MIX as orange for color coding consistency. The right panel illustrates the corresponding values across the 85 regions, showcasing the correlation between EF and ΔVOL in RUL, BT, and MIX, respectively.

Fig. 2. Unsupervised multivariate analysis was conducted using Principal Component Analysis (PCA) on the ΔVOL values.

A The outcomes (loadings of PCA) are displayed in three rows: the top row represents the findings for subjects with RUL placement, the middle row shows results for patients with BT placement, and the bottom row illustrates subjects with MIX placement. The first component (PC1_ΔVOL; left side) shows the main effect. The spatial distribution of the second PC of the ΔVOL (PC2_ΔVOL) was very similar to the Causal Depression Network (CDN) recently reported by Siddiqi et al. (2021) regardless of the electrode placement (B) The CDN’s spatial distribution is depicted for visual comparison at the same coronal, sagittal and axial sections, respectively. C The scatterplot to illustrate the close similarity between CDN and PC2_ΔVOLacross all types of electrode placements. D The more similar the pattern was to the Causal Depression Network the better the clinical outcome was.

Fig. 3. Unsupervised multivariate analysis was conducted using Principal Component Analysis (PCA) on the EF values.

The outcomes (loadings of PCA) are displayed in three rows: the top row represents the findings for subjects with RUL placement, the middle row shows results for patients with BT placement, and the bottom row illustrates subjects with MIX placement. The first component (PC1_EF; left side) shows the main effect. The second PC of the EF (PC2_EF) reflects the spatial distribution stemming from the different electrode placements.