Obesity is an important health issue in modern society. The prevalence of obesity has been increasing in these years [1, 2] and morbid obesity is related to cardiovascular disease and overall mortality [3]. Past reviews regarded binge eating as a manifestation of dysfunctional reward system and disinhibition. Some authors considered binge eating as a kind of addiction. Recent study demonstrated the more extensive involvement of brain pathways other than reward system [4]. The mechanism of change is not clear. There is scanty of research regarding correlation between change of activation pattern in brain areas in functional MRI, binge eating and psychiatric illness. To gain insight into the correlation of physiological alteration and psychiatric illness and to develop subsequent detectable biomarker are crucial. The goal of our study was to investigate the morphological changes in gray and white matter between obese patients and healthy subjects using voxel-based morphometry (VBM). Apply these novel markers to monitoring and improving comorbid psychiatric illness will be an essential part of multidiscipline integral care for obese patients.

Twenty morbid obese participants (BMI = 37.9 ± 5.2)aged between 20 and 65 years-old who are preparing to undergo bariatric surgery will be recruited from an obesity clinic in a single regional teaching hospital. The participants meet the criteria of severe binge eating defined by scores > 27 on Binge Eating Scale [5]. The control group as comparison objects for the cross sectional study includes 30 subjects (BMI = 22.6 ± 3.4) who are morbid obese, preparing to undergo bariatric surgery and scores < 17 on Binge Eating Scale. The control group will be recruited from the same clinic. All participants are not taking psychotropic agents during the study period and are not in need of immediate psychiatric intervention, such as suicidal ideation. The participants receive neuropsychological tests, including anxiety and depression test, and MRI before the bariatric surgery. All participants were scanned to obtain 3D gradient echo T1-weighted images (T1WI) using 1.5T MRI (Ingenia, Philips, Netherlands) with an 8-channel head coil. All images were acquired with TR/TE= 7.6/3.5 ms, Flip angle= 8°, voxel resolution = 1 x 1 x 1 mm³.

In data analysis, all structural data were processed using Statistical Parametric Mapping software package (SPM8, http://www.fil.ion.ucl.ac.uk/spm) with VBM toolbox (VBM8, http://dbm.neuro.uni-jena.de/vbm). A segmentation approach was used to perform tissue classification after brain extraction, image registration and normalization. Then gray and white matter images were smoothed for statistical analysis. The gray and white matter images were also calculated to find the correlation between the brain regions and body mass index (BMI), anxiety and depression scores.

In our results, we found the significant increased volume in the amygdala of obese patients compared to healthy subjects, and the significant decreased volume in the premotor cortex of obese patients was also found (p < 0.001) (Fig. 1a, b). We also found the significant increased volume in the inferior frontal gyrus and the thalamus of obese patients compared to healthy subjects (p < 0.001) (Fig. 1c, d). In the correlation analysis, as shown in Fig. 2, the positive correlation was found between amygdala and BMI; the positive correlation was found between thalamus and anxiety score; the positive correlation was found between amygdala and depression (p < 0.001).Previous reports in animal and human showed deficiency in reward system might be related to eating disorder [6]. The reward system is composed by the mesolimbic pathway which projecting from the ventral tegmental area to the nucleus accumbens. The amygdala and the hippocampus were also regarded as part of the system. Besides the reward system, the prefrontal cortex is implicated in initiating behavioral responses based on hedonic appraisals of food cues.Our study used neuroimaging and neuropsychological test for evaluation the difference in obese patients and healthy subjects, and we explored the possible neuromarker in obese patients. Our results suggested the changes of the volume in the brain structures may closely linked to the symptom and behavior of obese patients. The finding in this study confirmed the potential psychobiological biomarkers (neuroimaging and neuropsychological test results) for binge eating among obese patients.