Participants
Eighty-seven patients who had first-episode schizophrenia and had never been medicated were recruited from Nanjing Brain Hospital between January 2017 and December 2018. The following inclusion criteria were considered: patients were in compliance with the DSM-IV diagnostic criteria for schizophrenia; and patients presented no evidence of affective disorders, head trauma, and substance abuse.
Eighty-two HC that matched in terms of age and gender were recruited in the local community, and they had no history of a major psychiatric disorder and no family history of psychotic disorder.
The general inclusion criteria for all of the groups were right handed, aged 16–45 years, and able to understand survey instructions and execute cognitive tests. The general exclusion criteria were a history of the head injury with a loss of consciousness, substance abuse, and use of drugs that might affect the functions of the central nervous system. The study was approved by the Nanjing Brain Hospital Ethics Committee. After the study was completely described, a written informed consent was obtained from all of the participants.
Neuropsychological and clinical assessments
MCCB [24, 25] was used to assess cognitive function. The present study includes nine tasks across seven cognitive domains, including speed of processing (Category Fluency, Trails A, Brief Assessment of Cognition in Schizophrenia Symbol Coding), attention/vigilance(Continuous Performance Test), working memory(Wechsler Memory Scale-III Spatial Span), verbal learning (Hopkins Verbal Learning Test–Revised), visual learning (Brief Visuospatial Memory Test–Revised),reasoning and problem solving (The Mazes test), and social cognition (Mayer–Salovey–Caruso Emotional Intelligence Test). We corrected the original score for age, gender, and education to obtain a T score to evaluate cognitive function, and high scores showed a good performance. We adopted the Wechsler adult scale of intelligence[26] to test the intelligence quotient. The PANSS was used to assess the symptoms of the patients[27], and the process was carried out by two experienced psychiatrists.
Image data acquisition
MRI scanning was performed with a Siemens 3.0-T signal scanner, and a standard head coil padded with foam was used to reduce head motion and scanner noise. All of the participants laid in a supine position and were instructed to remain as still as possible, close their eyes, remain awake, and not think of anything[9, 28].Three-dimensional T1-weighted sagittal images were acquired using a brain volume sequence with the following parameters: repetition time (TR) = 2,300 ms, echo time (TE) =2.96ms, inversion time = 900 ms, flip angle (FA) = 9°, field of view (FOV) = 256 mm × 256 mm, matrix =256 × 256, slice thickness = 1 mm, 192 sagittal slices, and acquisition time = 554 s. The images were acquired using a gradient echo single-shot echo planar imaging sequence with the following parameters: TR/TE = 2,500/30 ms, FOV =224 mm × 224 mm, matrix = 64 × 64, FA = 90°, slice thickness = 3.5 mm, no gap, 37 interleaved transverse slices, 149 volumes, and acquisition time = 379s.
Data preprocessing and processing
Functional images were preprocessed with the Matlab2013b platform and the Data Processing Assistant for rs-fMRI (DPARSF4.4, advanced edition)[29].Data were calculated in an original space warped by diffeomorphic anatomical registration via an exponentiated Lie algebra (DARTEL).The first four volumes of the BOLD data were discarded for each subject to allow the signal to reach equilibrium. The remaining images were corrected for the acquisition time delay between slices. All of the subjects should have no > 2 mm maximum displacement in any plane, 2° of angular motion, and 0.2 mm mean frame-wise displacement[30]. Then, the images were spatially realigned to the first image of each dataset, and movement parameters were assessed for each subject and corrected using the Friston 24 approach [31]. Several nuisance covariates (global brain, white matter, and cerebrospinal fluid signals) were regressed out. The datasets were band-pass filtered to reduce low-frequency drift and high-frequency physiological respiratory and cardiac noise(0.01<f<0.1 Hz). ReHo was calculated on a voxel-by-voxel basis by calculating Kendall’s coefficient of concordance on the basis of regional homogeneity hypothesis, which estimates similarity in the time series of a given voxel to its nearest 26 voxels[9]. Each subject’s value was divided by the mean value of their whole-brain ReHo to eliminate the whole-brain effect at utmost. The standardized ReHo images were spatially smoothened with a Gaussian filter with a full width at half maximum (FWHM) of 4 mm. Finally, ReHo values were used for statistical analysis.
Statistical analysis
SPSS software (version20.0) was used to statistically analyze demographic data, clinical symptoms, IQ, and cognitive data. For continuous variables, a two-sample t-test was used to compare the differences between the two groups, and achi square test was used for categorical variables.
The ReHo maps were compared between the two groups by using the threshold-free cluster enhancement (TFCE) method with family wise-error (FWE) correction for multiple comparisons. Age, gender, education, and probability of gray matter were treated as covariates, and the threshold for significance was p<0.05[32]. In the following correlation analysis, the resultant significant ReHo map was used as inclusion masks.
Patients with schizophrenia were subjected to voxel-wise correlation analyses to explore the correlation between the ReHo map and PANSS (positive symptoms, negative symptoms, general, and all totals) and the correlation between the ReHo map and MCCB (speed of processing, verbal learning, working memory, reasoning/problem solving, visual learning, attention/vigilance, social cognition, and overall composite). In addition, age, gender, education, and illness duration were placed in the model as covariates. The permutation-based nonparametric inference was undertaken with 5,000 permutations, and significance level was thresholded for the correction of multiple comparisons by using a TFCE of 0.05.