A Resting-state Functional Magnetic Resonance Imaging Study In Patients With Generalized Anxiety Disorder

PurposeThe study used regional homogeneity (ReHo) and functional connectivity (FC) based on Resting-state functional magnetic resonance imaging (rs-fmri) to reveal altered the regional spontaneous activity adjacent brain regions and functional connectivity non-adjacent brain regions in patients with Generalized anxiety disorder (GAD),and analyzed the correlation between the clinical characteristics and value of these abnomal brain regions,which might provide objective imaging basis for explortion neural mechanisms of GAD.Materials and Methods1.SubjectsThe subjects through the screening finally were comprised of 31 patients with Generalized anxiety disorder (15 males,16 females),mean age=36.87±9.16 years (range 19-56y); and 36 healthy controls (13 males,23 females),mean age=39.53 ± 8.83 years (range 20～58y). Furthermore, we recruited the gender, age, education, all of the matched healthy controls. Each subject was right-handed, native language is Chinese, the han nationality. Patients with GAD inclusion criteria were as follows:(i) the patients conformed to the criteria of the Diagnostic and Statistical Manual of Mental Disorders, version 4 (DSMIV); (ⅱ) the patients had a complaint of uncontrollable anxiety and worry about everyday events and problems for at least six months; (ⅲ) had the anti-anxiety and antipsychotic drugs at least one month washing-out; (ⅳ) Hamilton Anxiety Rating Scale (HAMA) score≥14; (ⅴ) the patient was younger than 60 years old. Patients with GAD exclusion criteria were as follows: (ⅰ) other psychiatric disorders and secondary anxiety disorder; (ⅱ) with drugs or alcohol dependence; (ⅲ) patients had a past history of depression or episodes of depression; (ⅳ) 24-item Hamilton Depression Rating Scale (HAMD) score>20; (ⅴ) with other significant medical conditions, such as diabetes, high blood pressure, etc.; (ⅵ) the patient had an abnormal signal as verified by conventional T1 or T2-fluid-attenuated inversion recovery (FLAIR) MRI; (ⅶ) pregnant and breastfeeding women, (ⅷ) magnetic resonance imaging contraindication such as claustrophobia, cardiac pacemaker placement, etc.In addition, healthy controls inclusion criteria were as follows:(ⅰ) HAMA score< 7,24-item HAMD<8; (ⅱ) the conventional MRI found no organic disease; (ⅲ) no history of inborn or other psychiatric disorder. (ⅳ) no metal foreign body in the body. The study was approved by the Research Ethics Committee of People’s Hospital of Yuxi City (Yuxi, China), all subjects signed informed consent.2.Behavioral ratings and MRI data acquisitionBefore MRI examinations,subjects were rated by HAMA and 24-item HAMD, State-Trait Anxiety Scale (STAI) of emotional quantifiable indicators by consensus of two experienced psychiatrists which adopt the method of conversation and observation.Content includes:(1) the Hamilton anxiety scale was used in the diagnosis of anxiety and the severity of the symptoms of anxiety division, based on a total of 14 items, scoring criteria for five grade, grade 0 to 4, a level 0 for no symptoms, grade 1 is mild, level 2 is moderate, level 3 is severe, level 4 is very severe, including< 7 points without anxiety;7 or higher may have anxiety; 14 points must have anxiety or higher;There was an obvious anxiety acuity 21 points;29 points or more severe anxiety.(2) 24 Hamilton depression rating scale is the most common clinical applications assess depression scale, the results of analysis:according to the division of DavisJM> 35 points, could be severe depression;＞＞20 points, may be mild or moderate degree of depression;Such as< 8 points, no depression symptoms.(3) the Trait Anxiety inventory (State-Trait Anxiety Scale-traits Scale, STAI-T) reflect the anxiety condition of the subjects.MRI data acquisition using a 3-Tesla MRI scanner (Ingenia; phillips) in the Department of MRI, People’s Hospital of Yuxi city, using a 8 channel head coil Accept the signal of MRI. Each subject was supine with comfortable head by foam pads to reduce head motion, wearing headphones to reduce noise. During rs-fMRI, all the subjects keep close their eyes, relax, hold still, not to think of anything systematically or fall asleep. Each subject both conventional MRI sequences including cross-sectional T1WI, T2WI, T2-FLAIR and sagittal T2WI, to detect the clinical occult and organic diseases.The scanning parameters are as follows:cross-sectional T1WI(TR/TE=189/4.6ms),T2-FLAIR(TR/TE=8000/120 ms),cross-sectional and sagittal T2WI(TR/TE= 1984/80ms)、OV=230x230mm2,matrix= 384x217,4mm thick slices, with a 1 mm gap.The rs-fMRI scans were performed using a gradient echo planar imaging (EPI) sequence. Acquisition parameters were as follows:repeat time (TR)= 2000ms, echo time (TE)= 35 ms, flip Angle=90 °, field of view (FOV)= 230 mm x 230 mm, matrix = 64 x 64, total volume= 240,35 transverse plane parallel AC-PC line,3.6 mm thick slices, with a 0.7 mm gap, voxel size:3.6 mm x 3.6 mm x 3.6 mm,240 dynamic scanning, total time 8min 6sec,8400 images. The fMRI scanning lasted 8 min on each subject. After scanning, the patient was interviewed to assess whether he or she complied with the instructions. Subjects who failed to comply with the instructions would be excluded from the study.3.Data processingData processing were based on Matlab 2012b operation interface, using the Data Processing Assistant for Resting-State fMRI (DPARSF) integrated software (http://restfmri.net/forum/DPARSF) and Resting-State fMRI Data Analysis Toolkit (REST) (http://resting-fmri.sourceforge.net) software analysis.Data preprocessing as follows:(i)the data transformation:converts the DICOM format of the original image to NIFTI format.(ii) The first 10 time points were discarded due to transient signal changes before magnetization reached a steady-state and adaptation of the subject to the circumstances, leaving 230 time points for the preprocessing steps of slice timing, (iii) slice timing:the center of the scanning layer as a reference level,to correct the subjects’time differences, (iv) head motion correction:no subjects’data was discarded due to excessive head motion (more than 1.5mm in translation or 1.5 degree in rotation) for minimizing the influence of head motion, (v) nomalization:the fMRI images were normalized to the Montreal Neurological Institute (MNI) template (resampling voxel size= 3×3×3 mm3), (vi) linear trends and temporal filtering:with the removal of linear trends and temporal filtering (band pass,0.01-0.08 Hz) was used to reduce very low frequency drift and physiological high-frequency respiratory and cardiac noise by using REST software.3.1 ReHo analysisBased on preprocessing figures computing each subject whole brain voxels of regional homogeneity namely ReHo value, get standardized ReHo map.Ultimately,the data using an 8 mm full width at half maximum (FWHM) spatial smoothing gaussian kernel to reduce noise and residual differences in gyral anatomy for improving the image signal to noise ratio (SNR).3.2 FC analysisBased on preprocessing normative figures of spatial smoothing (Smooth), with 8 mm full- width at half maximum (FWHM) spatial smoothing gaussian kernel function to increase the signal-to-noise ratio of data, and then to linear drift and time filtering.Combining previous research results to select the area of interest (ROI), anterior cingulate cortex was chosen as the seed point, extract the average of the BOLD signal of time series, the removal of nine additional covariates, including:6 head motions parameters, global mean signals, white matter and cerebrospinal fluid signals to reduce the effect of head motion and physiological, and calculate the seed point and whole brain other voxel the correlation of time series, looking for seeds and some have stronger time correlation of brain regions.Standardization of Fisher Z transformation Z values, the correlation coefficient can be converted to gain strength value of the functional connectivity.Statistical analysisUsing two sample t test to evaluate the differences of the age, education, HAMA, HAMD and STAI-T scale scores between patients with GAD and healthy controls. By chi-square test analysis of gender differences between the two groups. Statistical analysis is to use software (SPSS, version 16.0),data with mean ± standard deviation.P< 0.05 was considered statistically significant difference.1.ReHo Statistical analysisTwo sample t test was used to explore the ReHo differences between GAD patients and healthy controls, correction cluster level threshold setting at p< 0.05 credited with significant difference (corrected for multiple comparisons with AlphaSim). Obvious differences in the brain region, using REST software to extract the difference of brain regions in patients with GAD Reho values. Then, Pearson’s correlation analysis with HAMA, HAMD-24 and STAI-T scale score.2.FC Statistical analysisUsing SPM8 to statistical analysis of functional connectivity map, select ALL the template of the left ACC and right ACC, respectively as a seed point.Two sample t test was used to explore between patients with GAD and normal subjects the differences of the FC intensity with seed point.Correction cluster level threshold setting at p< 0.05 credited with significant difference (corrected for multiple comparisons with AlphaSim). Obvious differences in the brain region, using REST software to extract the difference of brain regions with the FC intensity. Then, Pearson’s correlation analysis with HAMA, HAMD-24 and STAI-T scale score.Results1.Demographic and clinical characteristicsTwo groups showed no significant difference in gender (P= 0.310), age (P= 0.232), education (P= 0.088) and handedness. GAD group of HAMA and HAMD-24, STAI-T scale score significantly higher than the healthy controls group.2.Changes in ReHo and clinical characteristics Correlation analysisCompared to the healthy controls, GAD patients had significantly increased ReHo values in left middle temporal gyrus (MTG), left superior temporal gyrus (STG), right superior occipital gyrus (SOG). Decreased ReHo was observed in right orbital middle frontal gyrus, left anterior cingulate cortex (ACC), right middle frontal gyrus (MFG), bilateral supplementary motor area (SMA) (P> 0.05,corrected for multiple comparison).We found the Reho value in the left MTG was significantly positively correlated with HAMA scores (r= 0.488, P= 0.005; figure 3).3.Changes in FC and clinical characteristics Correlation analysisBrain regions showing abnormal functional connectivity with left ACC in GAD patients compared to the healthy controls.The increased functional connectivity areas in left precentral gyrus.and decreased functional connectivity areas in right orbital inferior frontal gyrus, left medial superior frontal gyrus, right middle frontal gyrus, right dorsolateral superior frontal gyrus,right middle temporal pole gyrus, right middle temporal gyrus, left posterior cingulate cortex (P> 0.05,corrected for multiple comparison).Further Pearson correlation found on the left ACC and the left medial superior frontal gyrus (r= 0.415, P= 0.039), and right right middle temporal pole gyrus (r= 0.494, P= 0.494) function connection strength value and HAMA score was significantly positively related.Brain regions showing abnormal functional connectivity with right ACC in GAD patients compared to the healthy controls.The decreased functional connectivity areas in right orbital inferior frontal gyrus, right middle temporal gyrus, right median cingulate cortex,and increased functional connectivity areas in left parahippocampal gyrus (P> 0.05,corrected for multiple comparison).Conclusion1.Compared to the healthy controls, GAD patients with multiple brain regions local abnormal consistency had significantly increased ReHo values in left middle temporal gyrus (MTG), left superior temporal gyrus (STG), right superior occipital gyrus (SOG). Decreased ReHo was observed in right orbital middle frontal gyrus, left anterior cingulate cortex (ACC), right middle frontal gyrus (MFG), bilateral supplementary motor area.We found the Reho value in the left MTG was significantly positively correlated with HAMA scores.2.Brain regions showing abnormal functional connectivity with left ACC in GAD patients compared to the healthy controls.The decreased functional connectivity areas in right orbital inferior frontal gyrus, left medial superior frontal gyrus, right middle frontal gyrus, right dorsolateral superior frontal gyrus,right middle temporal pole gyrus, right middle temporal gyrus, left posterior cingulate cortex,and increased functional connectivity areas in left precentral gyrus,. Brain regions showing abnormal functional connectivity with right ACC in GAD patients compared to the healthy controls.The decreased functional connectivity areas in right orbital inferior frontal gyrus, right middle temporal gyrus, right median cingulate cortex, and increased functional connectivity areas in left parahippocampal gyrus.Further Pearson correlation found on the left ACC and the left medial superior frontal gyrus, and right right middle temporal pole gyrus function connection strength value and HAMA score was significantly positively related.In brief, our study found abnormal spontaneous activities in multiple brain regions, to further understand the neural activity characteristics of GAD.GAD in patients with multiple brain regions, such as sensorimotor cortex and emotional regions with altered ReHo and FC, these might be related to the pathophysiology in GAD.