Betel nut, also known as areca, is the fourth most commonly used drug worldwide after tobacco, alcohol, and caffeine and also a stimulant and addictive substance. Previously, CM Chen et al. probed into the influence of religious affiliation on heavy betel nut chewing [1], and studied on the relationship between health risk perception and betel nut chewing [2]. Feng Chen et al. analyzed gray matter abnormalities between betel nut chewers and healthy subjects with voxel-based morphometry (VBM) [3]. However, there were few studies mentioned about the functional activity and brain network changes in betel nut chewers using functional magnetic resonance imaging (fMRI). Therefore, our aim was to use resting-state fMRI (rs-fMRI) to investigate the functional differences between betel nut chewers and healthy participants with amplitude of low frequency fluctuations (ALFF) and regional homogeneity (ReHo). The graph theoretical and network-based statistic (NBS) analyses were also used to find the network difference between two groups.

In the study, 24 participants were divided into two groups, 13 healthy subjects and 11 betel nut chewers. All subjects were asked to be relaxation, closing their eyes, and thinking of nothing but could not fall asleep. All participants were scanned using 3T MRI (Skyra, SIEMENS, Germany) imaging system with echo planar image (EPI) sequence to obtain resting-state functional images. The parameters of images were: TR/TE = 2000/30 ms, in-plane resolution (pixel size) = 2.7 x 2.7 mm², thickness =4 mm, number of repetition = 240, and 28 axial slices aligned along AC-PC lines.

In data analysis, Statistical Parametric Mapping (SPM) was used to perform pre-processing for images, including slice-timing, realignment, normalization, and smoothness. ALFF and ReHo were calculated after removing physiological noises. Then voxel-based two-sample t-test analysis was used to compare two groups. The graph theoretical analysis was also used to obtain the topological parameters of the brain network, including clustering coefficient (C), normalized clustering coefficient (γ), local efficiency (E_local), characteristic path length (L), normalized characteristic path length (λ), global Efficiency (E_global), smal-worldness(σ), modularity, assortativity, and transitivity. The NBS analysis was applied to find the difference of cerebrum functional connectivity between two groups of participants.

In ALFF analysis (Fig. 1), we found higher ALFF activation of right putamen, right inferior orbitofrontal gyrus, and left superior orbitofrontal gyrus in the betel nut chewers compared with healthy participants. We also found lower ALFF activation of left insula in the betel nut chewers. In ReHo analysis (Fig. 2), higher regional homogeneity of left supplementary motor area, right medial superior frontal gyrus, and right insula were found in the betel nut chewers compared with healthy participants. In addition, lower regional homogeneity of right middle frontal gyrus in the betel nut chewers was found.

In the graph theoretical analysis, lower degree in high regime of degree distribution was observed in the betel nut chewers compared with healthy participants (Fig. 3a). In the topological parameter, we found lower E_global but higher C, L, E_local in betel nut chewers (Fig. 3b, c), which means the poor ability of global integration but better local segregation of brain in the betel nut chewers compared with healthy participants. In NBS analysis, we compared the edges of the brain networks between two groups. More edges were found in the healthy participants compared with betel nut chewers (Fig. 4). The edges included the connections from the right inferior frontal triangularis gyrus to right hippocampus and left middle temporal pole gyrus, the left precuneus to right paracentral lobule, left middle temporal gyrus, and left middle temporal pole gyrus, the left superior frontal gyrus to left middle temporal gyrus, right inferior frontal triangularis gyrus to left middle temporal pole gyrus. Although all participants maintained small-worldness functional brain network according the σ calculation, the network was more like regular network in the betel nut chewers.

ALFF and ReHo results showed different activation and regional homogeneity of several brain regions in the betel nut chewers. Graph theoretical and NBS analyses revealed different topological organization and poor global integration of the brain network in the betel nut chewers. We suggested that heavy chewers may have abnormal brain function, that may further affect their chewing behaviors.