Characterizing small-world networks in the early stage of bipolar I disorder

Objective To explore the small world characters of patients in the early stage of bipolar disorder by using graph theory. Methods Resting-state fMRI data were collected from 22 patients who met with DSM-IV diagnostic criteria for current manic episode and with duration less than 2 years, and 22 age- and gender-matched healthy control subject. Time series of AAL 90 regions were extracted for each subject and correlation matrices were analyzed. Results Compared with health controls, bipolar disorder patients showed significantly decreased global efficiency in sparsity threshold of 18% and longer characterized path length in sparsity threshold of 13%, 16%-31%(P<0.05). BRMS score was positively correlated with characterized path length (r=0.426, P=0.048)and negatively correlated with global efficiency (r=-0.426, P=0.048)in sparsity threshold of 18% in bipolar manic patients. Conclusion The alteration of global efficiency and characterized path length reflected abnomality of connection between cortical and sub-cortical regions. Characterized path length could be a possible neuroimaging marker of bipolar-I mania. Key words: Bipolar disorder; Network; Small world; Functional connectivity