OBJECTIVE: The neural network mechanisms of consciousness are not fully understood. This study aimed to construct resting-state whole-brain functional networks using functional near-infrared spectroscopy (fNIRS) to explore the functional network mechanisms underlying disorders of consciousness (DOC) and to identify potential neural network biomarkers for diagnosis and prognosis. METHODS: Resting-state fNIRS data were collected from healthy controls (HC) and patients in vegetative state (VS) or minimally conscious state (MCS). Whole-brain weighted functional networks were constructed for each participant, and group differences in functional connectivity (FC) and global/local topological properties were analyzed, along with their correlation with Glasgow Outcome Scale-Extended (GOS-E) scores in DOC patients. RESULTS: DOC patients (both VS and MCS) had significantly lower average FC strength than HC, with the VS group showing significantly lower FC than the MCS group-differences most prominent within the sensorimotor network. Globally, both DOC groups exhibited reduced clustering coefficient, global efficiency (Eglob), and local efficiency (Eloc), as well as increased characteristic path length (Lp) compared to HC. Furthermore, the VS group had significantly lower Eglob and Eloc and higher Lp than the MCS group. Local topological differences were also mainly found in sensorimotor regions. Several network metrics, including average FC, Lp, Eglob, Eloc, and nodal efficiency, were significantly correlated with GOS-E scores in DOC patients. CONCLUSION: Global connectivity disruption centered around the sensorimotor network, along with reduced functional integration and segregation capabilities, may underlie DOC. Resting-state functional network characteristics could serve as neurobiological markers to differentiate VS from MCS and predict patient prognosis.
Wang et al. (Fri,) studied this question.