Hypoconnected state characterized in first-episode schizophrenia: topological abnormalities and clinical correlations

Disrupted connectivity among cerebral networks is a key feature of schizophrenia (SZ). Dynamic functional connectivity (dFC) analysis, which captures the time-varying properties of brain networks, has emerged as a pivotal approach for investigating functional abnormalities in SZ. However, the topological properties of patients across different connectivity states remain poorly understood. We recruited 50 patients with acute, first-episode, treatment-naive SZ and 43 healthy controls (HCs). A sliding-window approach was employed to examine dFC states in patients relative to HCs. Functional networks under different states were constructed based on the Dosenbach 160 atlas. Graph theory analyses were performed to calculate topological metrics (global and nodal properties), and correlations between abnormal metrics and clinical scales were analyzed within patient group. Patients stayed more frequently and persistently in a hypoconnected state. In this state, patients demonstrated widespread nodal topological impairments in the default mode network, cingulo-opercular network, and sensorimotor network. Conversely, the hyperconnected state was characterized by a lower clustering coefficient and a shorter shortest path length, and higher network efficiency in patients compared to HCs. Furthermore, aberrant nodal clustering coefficients in hypoconnected state correlated with psychotic symptoms and cognitive deficits. Individuals with SZ exhibit distinct topological abnormalities across different network states. The hypoconnected state, in particular, may serve as a critical window into the pathophysiological mechanisms of the disease.