Network integration and segregation changes in schizophrenia: impact of electroconvulsive therapy.

BACKGROUND

Studies have confirmed brain network topology disruption in schizophrenia (SZ). Electroconvulsive therapy (ECT) rapidly improves acute psychiatric symptoms, yet the exact mechanism by which it impacts brain network topology in SZ patients remains unclear. This study aims to explore topological changes in SZ patients' whole-brain functional networks during ECT, ultimately elucidating implicated neurological mechanisms.

METHODS

This study collected resting-state functional magnetic resonance imaging (rs-fMRI) data from 53 patients with schizophrenia before and after ECT, as well as data from 46 age-, gender-, and education-matched healthy control participants (HC). Using the Brainnetome Atlas, brain functional networks were constructed for each participant. Graph theory methods were applied to measure global and nodal topological properties. Clinical symptoms of patients were assessed using the Positive And Negative Syndrome Scale (PANSS). Independent sample t-tests were employed to compare topological properties between patients and healthy controls, while paired t-tests were used to assess before and after ECT differences within the patient group. Finally, partial correlation analyses were conducted to examine the relationship between changes in topological properties and changes in PANSS scores among patients before and after ECT.

RESULTS

Before ECT, compared to the HC group, the patient group demonstrated reduced local efficiency (Eloc) and clustering coefficient (Cp). In the right superior temporal gyrus, degree centrality (Dc) and nodal global efficiency (Ne) were lower, whereas in the left cingulate gyrus, Ne and Dc were higher. Following ECT, Eloc and Cp normalized in the patient group. Additionally, nodal local efficiency (NLe) and nodal clustering coefficient (NCp) increased in the bilateral superior frontal gyrus. Conversely, in the left inferior parietal lobule, Ne and Dc decreased, and nodal shortest path length (NLp) increased. Both NLe and NCp were lower in the bilateral lateral occipital cortex, both before and after ECT. However, no significant correlation was observed between changes in PANSS scores and alterations in global and nodal topological properties before and after ECT treatment.

CONCLUSIONS

Our study suggests that ECT may improve psychiatric symptoms by modulating the integration and dissociation functions within damaged brain networks in SZ patients. Specifically, the balance between the integration and dissociation functions of the default mode network (DMN), central executive network (CEN), and auditory networks (AN) may play a crucial role in the improvement of psychiatric symptoms.