Resting-state fMRI mapping of cerebellar functional dysconnections involving multiple large-scale networks in patients with schizophrenia.

BACKGROUND: Recent findings have indicated that patients with schizophrenia have altered cortico-cerebellar connectivity, but the nature of these network dysconnections remains unclear. AIMS: We applied a network-based approach to investigate the functional connectivity (FC) of the cerebellum in patients with schizophrenia. METHODS: Thirty-six patients with schizophrenia and 36 healthy controls underwent resting functional magnetic resonance imaging. We derived the following 6 major networks by applying group independent component analysis: (1) the cingulo-opercular network (CON); (2) the dorsal default-mode networks (dDMN); (3) the ventral default-mode network (vDMN); (4) the left frontoparietal networks (lFPN); (5) the right frontoparietal network (rFPN); and (6) the motor network (MOT). We defined 12 regions of interest (ROIs) by selecting the first 2 peaks of each network in the cerebellum. The FC map of all ROIs was calculated for each participant and compared between groups. RESULTS: The schizophrenic patients showed a decrease in FC between the cerebellar ROIs and the thalamus in all networks except the MOT. The FC decreased between cerebellar ROIs and the frontal cortex in the CON, rFPN, and MOT. However, the FC increased with the precentral gyrus and postcentral gyrus for the CON, lFPN, and dDMN. An increased FC with the occipital fusiform gyrus and the temporal occipital fusiform gyrus was also noted in the dDMN, vDMN, and MOT. CONCLUSIONS: The dysconnection of the cerebellum in the examined patients with schizophrenia was network-specific. The cerebellar-thalamic dysconnections were the most prominent findings and were common to all cognitive-related networks, whereas the cortico-cerebellar connectivity involved both an increase and decrease in FC, and depended more on the nature of the specific network.