- SI-specific structural connectome alterations: increased network regularisation and topological deficits in the posterior cingulate gyrus, insula and hippocampus.
- Nodal centrality changes in the posterior cingulate gyrus and insula correlate longitudinally with suicidal ideation severity.
- Significant longitudinal interaction in the posterior cingulate gyrus implicates its role in structural network recovery and treatment response in MDD with SI.
BMC Psychiatry. 2026 May 29. doi: 10.1186/s12888-026-08231-6. Online ahead of print.
ABSTRACT
BACKGROUND: Major depressive disorder with suicidal ideation (MDD-SI) is associated with a high risk of mortality. The specific longitudinal alterations in brain structural networks associated with SI remain unclear. We aimed to identify treatment-related topological reorganization in patients with MDD, with and without SI.
METHODS: This longitudinal study leveraged graph theory and structural MRI to analyze brain network changes in 101 MDD patients (66 MDD-SI, 35 MDD-nSI) and 41 healthy controls (HC). Network properties were evaluated before and after 2 weeks of clinical treatment, with global and nodal metrics calculated to track topological reorganization. Subsequently, group × time interaction analyses were conducted to characterize treatment-specific trajectories of brain network evolution.
RESULTS: We identified SI-specific structural connectome alterations, characterized by increased network regularization and topological deficits in the posterior cingulate gyrus (PCG), insula, and hippocampus. Longitudinal analysis revealed that nodal centrality in the PCG and insula was significantly associated with SI severity. Crucially, a significant longitudinal interaction effect was observed in the PCG.
CONCLUSION: Our findings underscore the important role of the PCG in the structural network recovery of MDD-SI. The topological changes of the PCG may provide insight into the neurobiological mechanisms underlying suicide-related pathophysiology and treatment response in MDD-SI.
PMID:42216167 | DOI:10.1186/s12888-026-08231-6
AI Search
Share Evidence Blueprint

Search Google Scholar
Save as PDF

