Abstract
BACKGROUND: This study is aimed at investigate the glucose metabolic patterns of cerebral small vessel disease (CSVD) at "point-line-plane" levels. METHODS: We retrospectively collected the (18)F-fluorodeoxyglucose PET/MRI images of 174 CSVD patients and 206 healthy controls. First, the brain FDG-PET of each subject was divided into 7 classical networks, and the mean standard uptake value (SUV(mean)) was calculated. Second, the SUV(mean) of different brain networks was compared at the region of interest and voxel level. Third, the metabolic connectivity of intra- and inter-networks was compared. Fourth, the correlation of SUV(mean) changes in different networks was calculated. Finally, the correlation between the SUV(mean) of the whole network and each brain region within the network was analyzed in different brain networks. RESULTS: Increased SUV(mean) was mainly shown in the right postcentral gyrus in the somatomotor network and the right superior parietal gyrus in the dorsal attention network; decreased SUV(mean) was predominantly observed in the right superior temporal gyrus in the somatomotor network in the CSVD group. Significant correlations of SUV(mean) alterations in varying networks and SUV(mean) of the whole network and each brain region within the network were found. CONCLUSION: Affected by CSVD, the glucose metabolism redistribution is heterogeneous in different brain network, and its changes start from brain region and metabolic connectivity within the sub-brain network.