Trait-related cortical-subcortical dissociation in bipolar disorder: Analysis of network degree centrality

Background: Bipolar disorder is a systemic brain disorder. Accumulated evidence suggested that cortical-subcortical imbalance could be a trait-related pathogenic factor of bipolar disorder. Degree centrality, a robust index of focal connectivity in which the number of direct connections from one node to all nodes is counted, has not previously been studied in bipolar disorder as a whole. Methods: Resting state functional magnetic resonance imaging was performed on 52 patients with DSM-IV bipolar I disorder and 70 healthy controls recruited between September 2009 and July 2014. Degree centrality was calculated within cerebral gray matter for each subject and compared between patients with bipolar disorder and healthy controls. Hub distributions of both groups were explored. Effects of medication exposure and mood state on degree centrality, as well as cortical-subcortical degree centrality correlations, were explored. Results: Compared to healthy controls, patients with bipolar disorder exhibited significant decrease in degree centrality in cortical regions, including the middle temporal pole, inferior temporal gyrus, and ventral prefrontal cortex, but showed significant increase in degree centrality mainly in subcortical regions, including caudate, thalamus, parahippocampal gyrus, hippocampi, anterior cingulate, insula, and amygdala, and a small portion of cortical regions, such as superior and middle frontal gyrus (P < .05, corrected). Spatial distributions of the 2 groups were very similar. No significant effects of medication exposure or mood state on degree centrality were found. Patients with bipolar disorder also showed significant decrease in cortical-subcortical degree centrality correlation (P = .003). Conclusions: These findings further contribute to the mounting evidence of cortical-subcortical dissociation in bipolar disorder pathophysiology. In addition, this study supports the continued development and implementation of graph-based techniques to enhance our understanding of the underlying neural mechanisms in mental disorders such as bipolar disorder, which are increasingly viewed as systemic brain disorders rather than disorders arising from disruption within a single structure or a limited number of structures. Due to the heterogeneity of our sample, as well as the small sample size of each medication and mood state subgroups, further investigation is needed to support our findings.