Betel nut is the seed of the betel palm, which grows in the tropical or subtropical regions, especially in Taiwan, India, China and parts of east Africa. It has been confirmed that chewing betel nuts can be addicted to it and is also carcinogenic to humans [1, 2]. Chewing betel nuts lead to a greatly increased risk of developing a range of serious diseases, including oral and esophagus cancers [3]. In Taiwan, 88% of people who were diagnosed to oral cancer have the habit of chewing betel nuts, but very few researches studied on how chewing betel nuts effects brian structure [4]. Therefore, in this study we tried to find out the structural volume and shape changes in the brain between betel nut chewers (BNC) and healthy controls (HC) with voxel-based morphometry (VBM) and vertex-wise shape analyses. In addition, the relationship between brain structural volume size and implicit attitude or inhibitory control were also discussed.

We enrolled 11 BNCs and 13 HCs who all underwent magnetic resonance imaging (MRI) examination on a 3T imaging system (Skyra, Siemens Medical System, Germany) with a 20 channel head neck coil. T1-weighted images were obtained using the 3D MP-RAGE sequence, and the scan parameters included TR/TE = 2500/22.27 ms; flip angle = 8°; voxel size = 1 x 1 x 1 mm³, and 160 slices.

To investigate focal brain volume differences in both gray matter and white matter between BNCs and HCs, voxel-based morphometry (VBM) analysis of Statistical Parametric Mapping (SPM) was used. Vertex-wise shape analysis of FMRIB Software Library (FSL) was also performed to enable 15 structures to be separated from brain and see the differences in shape between BNCs and HCs of each structure. In addition, two psychometric testing, D score (sensitivity index) and Stop-Signal Reaction Time (SSRT), were tested for 24 participants to find the correlation between brain structural volume size and implicit attitude or inhibitory control, respectively. The higher D score means the more disgust participants feel about betel nuts, and the higher SSRT indicates worse inhibitory control.

In the VBM analysis, reduced gray matter volume of anterior cingulum cortex (ACC) were found in the BNCs compared to HCs (Fig. 1a) (p = 0.01). Decreased white matter volume of superior frontal gyrus (Fig. 1b), and enlarged white matter volume of cingulate gyrus (Fig. 1c) were also observed in the BNCs compared to HCs (p = 0.01). In the vertex-wise shape analysis, altered shape was found in right accumbens (Fig. 2a), left amygdala (Fig. 2b), right caudate (Fig. 2c), and right hippocampus (Fig. 2d) between BNCs and HCs.

In the correlation of D score and gray matter volume, positive correlation was found in left inferior frontal gyrus (pars triangularis) (Fig. 3a) and right anterior cingulum (Fig. 3b), and negative correlation in superior frontal gyrus (Fig. 3c) (p = 0.006). In the correlation of D score and white matter volume, positive correlation was found in cuneus (Fig. 3d), and negative correlation in right supplementary motor area (rSMA) (Fig. 3e) (p = 0.005). In the correlation of SSRT and gray matter volume, positive correlation was found in right middle temporal gyrus (Fig. 3f), and negative correlation in superior frontal gyrus (Fig. 3g) (p = 0.03). In the correlation of SSRT and white matter volume, positive correlation was found in right superior temporal (Fig. 3h), and negative correlation in midbrain (Fig. 3i) (p = 0.04).

The atrophy in gray matter of ACC of BNCs indicated that BNCs might have less error detection ability, attention, and motivation. Previous study also found ACC decreased in BNC [4]. The decreased white matter volume of superior frontal gyrus of BNCs might be related to less self-awareness or less coordination with the action of the sensory system [5]. Larger white matter volume of cingulate gyrus of BNCs might effect their emotion formation and processing, learning, and memory.

The positive correlation with D score indicated those who like chewing betel nuts frequently had increased structural volumes of left pars triangularis inferior frontal, right anterior cingulum, and cuneus, which were related to emotional disorder and addiction. On the contrary, negative correlation with D score indicated those who do not like or not usually chewing betel nuts had decreased structural volumes of superior frontal gyrus and right supplementary motor area, which were related to action of sensory system and coordinating temporal sequences of actions.

The positive correlation with SSRT indicated those who had worse inhibitory control had increased structural volumes of right middle temporal and right superior temporal gyrus. Temporal gyrus linked to the perception of emotions in facial stimuli and the processing information ability of changeable face characteristics. The negative correlation with SSRT indicated who had worse inhibitory control had smaller structural volumes of superior frontal gyrus and midbrain, which were related to motor system.

In our study, the volume and shape changes of brain structures in BNC included motor system, cognitive system and emotion system, which might reflect the underlying mechanism in BNC. The study validated the application of VBM and vertex-wise shape analyses in the habitual betel nut’s addictive brains. Together they provided better structural information of the brain changes and might facilitate understanding of the underlying mechanism in chewers.