Volumetric changes after convulsive therapies in treatment-resistant depressive patients

88 

Abstract Submission 

Pedro da Silva¹, Helena Bellini¹, Eric Cretaz¹, Adriana Carneiro¹, Leonardo dos Santos¹, José Gallucci-Neto¹, Andre Brunoni¹

¹Service of Interdisciplinary Neuromodulation, Laboratory of Neurosciences, Universidade de São Paulo, São Paulo, Brazil

Pedro da Silva  
Service of Interdisciplinary Neuromodulation, Laboratory of Neurosciences, Universidade de São Paulo
São Paulo, Brazil

Helena Bellini  
Service of Interdisciplinary Neuromodulation, Laboratory of Neurosciences, Universidade de São Paulo
São Paulo, Brazil

Eric Cretaz  
Service of Interdisciplinary Neuromodulation, Laboratory of Neurosciences, Universidade de São Paulo
São Paulo, Brazil

Adriana Carneiro  
Service of Interdisciplinary Neuromodulation, Laboratory of Neurosciences, Universidade de São Paulo
São Paulo, Brazil

Leonardo dos Santos  
Service of Interdisciplinary Neuromodulation, Laboratory of Neurosciences, Universidade de São Paulo
São Paulo, Brazil

José Gallucci-Neto  
Service of Interdisciplinary Neuromodulation, Laboratory of Neurosciences, Universidade de São Paulo
São Paulo, Brazil

Andre Brunoni  
Service of Interdisciplinary Neuromodulation, Laboratory of Neurosciences, Universidade de São Paulo
São Paulo, Brazil

Electroconvulsive Therapy (ECT) is the most effective treatment for depression. However, self-limiting physical side effects and serious cognitive dysfunctions limits the application of ECT. In Magnetic Seizure Therapy (MST), as there is no direct electrical stimulation of structures such as the hippocampus, implicated in ECT-related memory impairment, memory dysfunctions are not expected. However, MST-related studies are limited by small sample sizes and open label designs. Then, it was proposed the "Electro-Magnetic Convulsive Therapies for Depression (EMCODE): a double blind, randomized, non-inferiority study", comparing 100Hz MST with bitemporal ECT. The primary objective involves assessing changes in depressive symptoms and cognitive measures in treatment-resistant depression (TRD) patients undergoing MST as compared to ECT treatment. A secondary objective involves the investigation of structural MRI changes and associations with clinical and cognitive outcomes.

Details of the study protocol can be found in [1]. The co-primary outcome scales were the Hamilton Rating Scale for Depression (HAMD-17) and the Autobiographical Memory Task (AMT). Longitudinal volumetric weighted 3D T1 MRI images were acquired using a 3.0 T device (Achieva, Phillips, Amsterdam, The Netherlands). Processing was performed using the Longitudinal pipeline provided by the Computational Anatomy Toolbox 12 (CAT12), a toolbox of the Statistical Parametric Mapping software (SPM12). Voxel-Based (VBM) and Region-Based Morphometry (RBM) were performed. We used the Neuromorphometrics atlas for RBM analysis. A statistical model was built with the flexible factorial design. The following contrasts were used: post-treatment > baseline in ECT and MST, post-treatment > baseline in ECT & MST and post-treatment > baseline & ECT > MST (interaction). Associations between MRI findings with outcomes were performed using R Version 4.1.2. We used linear mixed models with a three-way interaction between the fixed predictors volume, time, and treatment. HAMD-17 and ATM sub scores were the outcomes. Participants were included as random intercept. To each three-way significant interaction, we performed two-way interaction between brain volume and time separately to each group.

Forty-one patients were included. There was a progressive reduction of HDRS-17 scores for both groups (Fig. 1a). The MST group presented a better profile for the AMT sub scores (Fig. 1b-c). In VBM analysis, no significant structural changes were observed for any contrast. In RBM analysis, the following structures present significant changes in patients who received ECT: left and right hippocampus (t = 8.23, p = .00008; t = 8.23, p = .0004), left and right entorhinal area (t = 4.73, p = .025; t = 4.71, p = .035), left posterior insula (t = 3.15, p = .014), right amygdala (t = 5.01, p = .023), and right temporal pole (t = 4.46, p = .03) (Fig. 2a). No significant structural changes were observed in patients who received MST. The following structures present significant changes when considering patients who received ECT or MST: left and hippocampus (t = 6.36, p = .003; t = 4.88, p = .008), left entorhinal area (t = 3.54, p = .019), and right amygdala (t = 3.79, p = .009) (Fig. 2b). Only the left hippocampus (t = 5.73, p = .006) presented significant change after treatment when comparing both interventions (Fig 2c). For the contrast post-treatment > baseline in ECT & MST in RBM analysis, significant interactions with HAMD-17 as outcome were found for the right hippocampus (t = -7.12, p = 0.04) and amygdala (t = -7.12, p = 0.04). For both interventions, the higher the volume increase, the bigger the antidepressant effect. However, only the MST group showed significant two-way interactions.

Regional-based, and not voxel-based, volumetric changes after MST in TRD patients were associated with depressive symptoms.

Non-Invasive Stimulation Methods Other ¹

Psychiatric (eg. Depression, Anxiety, Schizophrenia) ²

Segmentation and Parcellation

Morphometrics

Psychiatric Disorders