Dynamic Functional Connectivity Predicts Treatment Response to Electroconvulsive Therapy in Major Depressive Disorder

Background: Electroconvulsive Therapy (ECT) is one of the most effective treatments for major depressive disorder (MDD). There is recently increasing attention to evaluating ECT's effect on resting-state functional magnetic resonance imaging (rs-fMRI). This study aims to investigate whether dynamic functional connectivity (dFC) estimated from rs-fMRI predicts the ECT outcome.
Methods:
Resting-state fMRI data were collected from 119 MDD patients (76 females) with an average age of 55.94 ±15.87 years old. This dataset includes 71 responder patients, with a 50% reduction in symptom severity after ECT, and 48 non-responder patients. Twenty-four independent components from default mode and cognitive control network were extracted using group-ICA form pre-ECT rs-fMRI. Then, a sliding window approach was used to estimate the pre-ECT dFC of each subject. Next, k-means clustering was used to put dFC of all patients in three distinct states. We calculated the amount of time each subject spends in each state, called occupancy rate or OCR. Finally, we calculated the partial correlation between pre-ECT OCRs and Hamilton Depression Rating Scale (HDRS) change while controlling for age and gender.
Results:
We found the pre-ECT OCR in a state with higher positive connectivity among CCN components predicts the HDRS changes in the responder patients (R=-0.30, corrected p=0.03), while we did not find any significant link between the pre-ECT OCR and HDRS change in non-response patients.
Conclusion: Our finding suggests that the dFC features, estimated from CCN and DMN, could successfully predict the ECT outcome of MDD patients.