Altered cortical-subcortical structural connectivity underlying symptom dimensions in schizophrenia.

Abstract

Schizophrenia is characterized by heterogeneous clinical symptoms, which may arise from disruptions in brain network connectivity. This study aimed to characterize alterations in white matter networks, and to investigate their associations with symptom dimensions in schizophrenia. We conducted probabilistic tractography in 61 schizophrenia and 63 matched healthy controls. Whole-brain structural connectomes were constructed using the Desikan-Killiany atlas, which comprises 82 regions. Network-Based Statistics identified significantly altered subnetworks, followed by graph-theoretical analyses to assess their topological properties. Symptom dimensions were derived from the Positive and Negative Syndrome Scale, and regression analyses were performed to examine associations between network alterations and symptom severity. Individuals with schizophrenia showed weakened connectivity in a subnetwork comprising 13 connections across cortical and subcortical regions (threshold edge t > 3.50, component p < 0.001, familywise error corrected). This subnetwork was primarily characterized as frontal-temporal-insular-cingulate connectivity, with limited hippocampal and thalamic contributions. Within this subnetwork, Connectivity intensity(p < 0.001) and global efficiency(p < 0.001) were significantly lower in schizophrenia relative to healthy controls. Exploratory regression analyses (adjusting for potential confounders) revealed that distinct connections were associated with different affective and cognitive symptom dimensions; global efficiency was associated with negative symptom dimensions (all p < 0.05, uncorrected). The analyses identified reduced structural connectivity of cortical networks with limited hippocampal and thalamic involvement among individuals with schizophrenia relative to healthy controls, and provided preliminary evidence linking this subnetwork to clinical heterogeneity in schizophrenia.