Related Experiment Video
Updated: Aug 11, 2026
Robotically Delivered fMRI-Guided Personalized Transcranial Magnetic Stimulation Therapy for Treatment-Resistant Depression
Published on: April 10, 2026
Connectome-based graph metrics predict electroconvulsive therapy response in adults with depression
François Ramon1, David Attali2, Cécile Paolillo3
1Université Paris Cité, Institute of Psychiatry and Neuroscience of Paris (IPNP), INSERM U1266, Ima-Brain team, 75014, Paris, France; Université Paris Cité, Laboratory for the Psychology of Child Development and Education, CNRS UMR 8240, F-75005, Paris, France.
Brain network efficiency at baseline predicts electroconvulsive therapy (ECT) remission in depression. Lower structural efficiency indicates better treatment response, suggesting network characteristics can guide depression treatment.
Area of Science:
- Neuroimaging
- Network Neuroscience
- Psychiatry
Background:
- Major depressive disorder (MDD) is characterized by widespread brain dysconnectivity.
- Electroconvulsive therapy (ECT) is highly effective for depression, but predicting treatment response remains challenging.
- Graph theory metrics offer a way to quantify brain network architecture and identify potential biomarkers.
Purpose of the Study:
- To investigate whether baseline structural and functional brain network properties predict remission in patients undergoing ECT for depression.
- To identify specific graph theory metrics that can serve as biomarkers for ECT treatment response.
Main Methods:
- A prospective longitudinal study involving 41 adult MDD patients and 24 healthy controls.
- Structural and functional MRI scans (anatomical, diffusion, resting-state fMRI) were acquired at baseline, mid-treatment, and post-treatment.
- Brain networks were constructed using diffusion tractography and rs-fMRI, with global and local efficiency analyzed using graph theory.
- Linear mixed-effects models and penalized binomial regression identified predictors of remission (defined as MADRS ≤10).
Main Results:
- Brain network graph theory metrics remained stable throughout ECT, indicating no large-scale network reorganization.
- Patients with depression showed reduced structural local efficiency compared to controls at baseline.
- Individuals who remitted after ECT had lower baseline structural local and global efficiency (in FA and NDI-weighted networks) compared to non-remitters.
- Non-remitters exhibited reduced baseline functional local efficiency relative to controls.
- Baseline NDI-weighted global efficiency was a significant predictor of remission.
Conclusions:
- Baseline structural and functional brain network profiles differentiate between ECT remitters and non-remitters.
- Remission is associated with preserved functional organization but lower baseline structural efficiency.
- Non-remission is characterized by reduced functional integration and relatively preserved structural connectivity.
