Altered Functional Connectivity Of Cingulate Subregions In Schizophrenia

Objective:Schizophrenia patients have shown altered resting-state functional connectivity(rs FC) of the cingulate cortex; however, it is unknown if rs FCs of the cingulate subregions are differentially affected in this disorder. We aimed to clarify the issue by comparing rs FCs of each cingulate subregion between healthy controls and schizophrenia patients.Subjects and Methods:A total of 94 schizophrenia patients(52 males and 42 females)and 102 healthy controls(45 males and 57 females) underwent resting-state functional MRI data and high-resolution structural imaging by a GE Discovery MR750 3.0 Tesla. Diagnoses for patients were confirmed using the Structured Clinical Interview for DSM-IV. Two sets of resting-state f MRI data were acquired, which are gradient-echo single-shot EPI sequence and gradient-echo SENSE-SPIRAL(spiral in) sequence.1. The high-resolution structural imaging and the rs-fMRI data were all carried out using statistical parametric mapping(SPM8, http://www.fil.ion.ucl.ac.uk/spm) based on Matlab.2. The cingulate cortex was divided into 9 subregions, including the subgenual anterior cingulate cortex(s ACC), areas 24 and 32 of the pregenual ACC(p ACC), areas 24 and 32 of the anterior mid-cingulate cortex(a MCC), posterior MCC(p MCC), dorsal(d PCC) and ventral(v PCC) posterior cingulate cortex(PCC), and retrosplenial cortex(RSC). Projecting ROI of the s ACC onto the mean normalized template of the f MRI data derived from the two sequences to compare signal loss and distortion of the s ACC between the two acquisition sequences.3. Choosing the cingulate subregions as ROIs,and conducting ROI-based rs FC analyses(with global signal regression and without global signal regression).4. Voxel-based morphometry(VBM) analysis of structural MRI data was performed to investigate group differences of grey matter volume, using general linear model to compare intergroup differences in these rs FCs without and withGMV correction.5. Extracted these significant intergroup differences of rs FCs and calculated Spearman’s correlation coefficients between these rs FCs and clinical parameters(i.e., PANSS, illness duration, and antipsychotic dosage)(p < 0.05, uncorrected).Results1. The s ACC images derived from the SENSE-SPIRAL sequence exhibited less signal loss and distortion. Therefore, we selected f MRI data acquired by the SENSE-SPIRAL sequence for rs FC analyses.2. Results with and without global signal regression(GSR) were reported. After global signal regression, schizophrenia patients showed significantly reduced rs FCs in most subregions, increased rs FC was only observed in a few brain regions. The main decreased rs FCs are the prefrontal lobe, insular, striatum and thalamus, connected with the affective network, salience network, sensorimotor network and default mode network. Schizophrenia patients showed increased rs FCs of the bilateral a MCC(area 24), d PCC, v PCC and RSC without GSR, and the results are connected to default mode network. Consistently increased connectivity can be found between the results of with and without GSR.3. Compared with healthy subjects, schizophrenia patients had significantly reduced GMV. Compared with rs FC results without GMV correction, most of the rs FCs with significant group differences remained significant after GMV correction.Conclusions1. SENSE-SPIRAL sequence exhibits less signal loss and distortion, it was superior to the EPI sequence in terms of imaging quality;2. GSR can greatly affect results of between-group comparisons, which should be considered when analyzing the rs-fMRI data;3. The altered rs FC of the cingulate subregions in schizophrenia patients are complex, and the default mode network had important role in it. The consistently increased rs FCs may challenge the functional disconnection hypothesis of schizophrenia.