A Magnetic Resonance Imaging Study On The Gray Matter, White Matter And Intrinsic Networks In Schizophrenic Patients

Objective: Using magnetic resonance imaging (MRI), we examined the volume and density difference of gray matter (GM) and white matter (WM), and functional connectivity difference of intrinsic networks between schizophrenic patients and healthy controls, to investigate the disease-specific biological makers, and to provide new clues and evidences for the understanding of the mechanism and diagnosis of schizophrenia.Methods:(1) The schizophrenic patients (n=85) and healthy controls (n=85) were assigned into four groups: male and female patients, male and female controls. The "unified segmentation" approach of voxel-based morphometry (VBM) was used to preprocess structural images, and four images represent GM volume, GM density, WM volume and WM density, were obtained for each subject. The GM/WM difference between the patients and controls was examined by two-factor analysis of variance (ANOVA), with diagnosis and sex as two factors. The age, total incrainial volume (TIV), or both of them, were introduced as covariants in the statistic models to examine their effects on VBM results.(2) Functional images of schizophrenic patients (n=25), their healthy siblings (n=25), and healthy controls (n=25) were acquired by using resting state functional MRI. The posterior cingulate cortex and adjacent precuneus (PCC/PCu) and right dorsolateral prefrontal cortex (DLPFC) were set as seed regions for functional connectivity analysis. The brain regions that were significantly positively correlated to PCC/PCu and significantly negatively correlated to right DLPFC constitute the TNN, which was at most activation in resting state, while deactivated by attention-demanding or goal-directed tasks in previous functional studies. The brain regions that were significantly negatively correlated to PCC/PCu and significantly positively correlated to right DLPFC constitute the TPN, which was at most activation when an individual is performing attention-demanding or goal-directed tasks. Thus, the intrinsic networks (including the TPN and TNN) were constructed for each group and the overlapped intrinsic networks were obtained. Functional connectivity difference of the overlapped intrinsic networks between the three groups was examined by ANOVA. Post hoc tests (LSD) were used to examine the difference of connectivity strength between every two groups.Results:(1) We found significant main effects of diagnosis factor in examining the GM difference between schizophrenic patients and healthy controls by two-factor ANOVA (p<0.05, FDR correction). Patients were associated with reduction of GM volume/density, which were located in the bilateral superior temporal gyri, inferior frontal gyri and insula, and the left medial temporal regions including the amygdale, hippocampus and hippocampal gyrus. Additionally, GM density reduction was found in the bilateral medial and dorsolateral prefrontal cortices, and the anterior cingulate gyrus in patients. Significant main effects of diagnosis factor was found in examining the WM difference between patients and controls by two-factor ANOVA (p<0.001, uncorrection; negative results were found when set the threthold at p<0.05, FDR correction). Patients were associated with reduced WM volume/density, which were located in the bilateral frontal and temporal white matter, and the brain's princinpal fiber tracts, such as the left uncinatus fasciculus, inferior longitudinal fasciculus and, optic radiation, and right arcuate fasciculus. No significant diagnosis X sex interaction was found in examining the group difference of GM/WM by two-factor ANOVA (p<0.05, FDR correction). The covariants, such as age, TIV, or both of them, can significantly increase the F values of two-factor ANOVA in detecting the GM changes.(2) Significant connectivity difference of the overlapped intrinsic networks was found between schizophrenic patients, healthy siblings and healthy controls (ANOVA, p<0.05). These were the connectivities between the left DLPFC and right inferior frontal gyrus (IFG) (F=3.409, p=0.038), the PCC/PCu and left inferior temporal gyrus (ITG) (F=6.938, p=0.002), and the left ITG and right ITG (F=4.016, p=0.022). Post hoc tests (LSD) revealed that in the TNN, patients and siblings shared higher connectivity between the bilateral ITG, while patients alone showed higher connectivity between the PCC/PCu and left ITG In the TPN, only the patients showed higher connectivity between the left DLPFC and IFG. No significant connectivity difference was found between any two regions from different networks.Conclusion:(1) Schizophrenic patients involve reduction of GM/WM volume and density in the frontal and temporal lobe, limbic regions, and the connecting WM fiber tracts. Structural changes in the frontotemporal limbic circuit may be the key substrates of the neuropathology of schizophrenia. The covariants, such as the age, TIV, or both of them, can increase the sensitivity of VBM to detect the GM changes.(2) Schizophrenic patients and heathly siblings shared higher functional connectivity in the intrinsic networks. The pathophysiology of schizophrenia involves higher connectivity of the TPN and TNN. Higher connectivity of the TNN may be associated with the genetic liability for schizophrenia.