Subjects
Thirty of 7-12 years old children with PMNE were recruited from among outpatients of Shanghai Children’s Medical Center. These patients were diagnosed as PMNE by a senior doctor according to the international Classification of diseases and The Diagnostic and Statistical Manual of Mental Disorders. The inclusion criteria were as follows: 1) The patient wetted the bed during the night more than or equal twice a week for three months but did not during the day. 2) Their symptoms were not caused by any related disease or drug, 3) Their symptoms continued for more than six months. Twenty-eight age- and gender-matched healthy controls were recruited if they had not wet their bed after age 5. All subjects were right-handed and all neurological and psychiatric diseases were excluded based on clinical examination and a structured interview; more details, see Table 2 and additional file 1.
Table 2. General clinical information for the PMNE group and healthy group.
Measures
|
PMNE group
(n = 30)
|
Healthy group
(n = 28)
|
Age, years
|
9.34 ± 1.37
|
9.62 ± 1.45
|
Gender, male/female
|
15/15
|
14/14
|
Years of education
|
3.31 ± 1.23
|
3.22 ± 1.42
|
Bed-wetting frequency, per week
|
2.16 ± 0.05
|
-
|
Number of patients never waking up for voluntary voiding
|
15 (n=30)
|
-
|
PMNE = primary monosymptomatic nocturnal enuresis.
MRI image acquisition
The MRI data were acquired using a 3.0 Tesla scanner (Siemens Trio Tim, Erlangen, Germany) that utilized a 12 channels head coil. The children were required to drink approximately 300 ml of water, and the functional MRI scan began when they felt like urinating. During the scan, the subjects were instructed to hold urine naturally, remain at rest with close their eyes, not fall asleep, and minimize movement. Head motion was confined to less than 2 mm or 2°.
We used a high-resolution T1-weighted 3-dimensional magnetization-prepared rapid acquisition gradient-echo (MPRAGE) pulse sequence to obtained the whole brain anatomical images. The parameters were as follows: echo time (TE) =3.42 ms, repetition time (TR) = 1900 ms , inversion time = 900 ms, flip angle=9° , field of view (FOV) = 240 × 240 mm2, matrix size = 256 × 256, slice thickness = 1.0 mm, and 192 sagittal slices. The resting-state functional data were acquired using a T2*-weighed gradient-echo echo-planar-imaging (EPI) sequence, the parameters of the sequence were as follows: FOV = 220 × 220 mm2, matrix size = 64 × 64, TR = 2000 ms, TE = 30 ms, slice thickness = 3 mm, 32 axial slices to cover the whole brain, 210 volumes.
fMRI image analysis
fMRI data preprocessing
Functional images analysed with statistical parametric mapping software (SPM12; http://www.fil.ion.ucl.ac.uk/spm/), the Data Processing Assistant for Resting-State fMRI (DPARSF; http://rfmri.org/DPARSF), and Matlab (MathWorks, Natick, MA) software on a personal computer. The first ten volumes of functional data were discarded for each participant to avoid MRI system instability and allow the participant adaptation to the machine noise. The remaining 200 volumes were then corrected for the acquisition time delay using slice-timing, realignment was used to correct for head motion. After these corrections, the series of functional images for each participant were segmented and normalized, which included individual structural and functional images that were coregistered and segmented into gray matter, white matter and cerebrospinal fluid, after that the functional images were normalized to Montreal Neurological Institute stereotaxic space (3-mm isotropic voxels). In this study, the tissue probability template was created according to children’s age and gender by the Template-O-Matic toolbox (TOM, http://141.35.69.218/wordpress/software/tom/). After that, the images were bandpass filtered (0.01-0.08 Hz) to reduce the influences of low-frequency drift and high-frequency respiratory and cardiac noise. The spatial smoothing was performed on the functional images using a 6-mm full-width at half-maximum (FWHM) isotropic Gaussian kernel.
Each participant’s fMRI data were denoised through linear regression of confounding effects including movement estimates (Friston 24-parameter, including current and past position parameters) and blood oxygen level-dependent signals in cerebrospinal fluid and white matter by the CompCor technique, which removed MRI signals derived from the respiratory and cardiac cycles. We did not regress out the global mean timecourse in consideration of its relation to global brain activity, which would be problematic for calculating DC. DC values were based on the correlation between one voxel and the rest of the whole brain, as described later in this article. Framewise displacement (FD) was calculated for each participant and included in the group analysis.
DC analysis
After preprocessing, the normalized and filtered images were used to calculate binarized DC measures using DPARSF. First, Pearson correlations between the time series of all gray voxel pairs were calculated, and a whole-brain FC matrix was constructed for each subject. Then, we obtained an undirected binarized matrix. If the Pearson's coefficient between the two voxels was < 0.25, the elements of the binarized matrix were set to 0; otherwise, they were set to 1. Then, the voxels that had low temporal correlation due to signal noise were excluded. From the perspective of graph theory, the DC value of each voxel reflects the number of edges connected to the given voxel (vertex), which indicates its central role in transferring information across brain regions. Standardized binarized DC maps were obtained by dividing by the global average of DC values.
Seed-based FC analysis
The brain regions in which the patient group showed significantly different DC from the healthy group were defined as ROIs. In brief, Pearson correlation coefficients of the functional time series were calculated between the ROIs and the whole brain voxel-by-voxel using DPARSF. Individual correlation coefficients were then transformed into z-scores to improve normality and acquire the zFC maps.
Statistical analysis
Individual DC and zFC images entered into the group-level analyses using SPM. Two sample t-tests were conducted to analyse the between-group differences. Each subject’s age, gender and FD were as covariates. All the statistical results reported reached voxel-level uncorrected p < 0.001 and cluster-level p < 0.05 with family-wise error correction.