Participants
This research was approved by the First Affiliated Hospital of Sun Yat-Sen University ethics committee. Written informed consents were gained from each participant. We applied the following inclusion criteria: (1) a first unilateral supratentorial subcortical infarct (<7 days) (Fig. 1); (2) no intra-cranial or extra-cranial artery occlusion on magnetic resonance angiography confirmed by either ultrasound or magnetic resonance angiography; (3) baseline scores of < 34 on the lower limb (LL) component of the Fugl–Meyer (FM-LL); (4) age 18-75 years; and (5) patients receiving routine rehabilitation therapies, as reported earlier [15-18].121314 Exclusion criteria were: (1) medical implants contraindicated for cerebral magnetic resonance imaging (MRI); (2) additional neurological diseases other than stroke; (3) traumatic brain injury or psychiatric disease or alcohol abuse; (4) any revascularization therapy; (5) presence of aphasia or apraxia, or use of any medications likely to affect motor examinations during the follow-up period. We also recruited age- and gender-matched healthy controls.
Experimental design
Each patient underwent three MRI acquisitions: within the first week (week 1; < 7 days), the fourth week (week 4; 28 ± 4 days), and the 12th week (week 12; 84 ± 4 days) [15-18]. Neurological and FM-LL assessments were performed before MRI examination at each time point. Controls, however, were examined only once.
Behavioral assessments
The National Institutes of Health Stroke Scale was used to assess initial neurological deficits. FM-LL scale was administered to evaluate LL motor deficits at 1-week, 4-weeks, and 12-weeks after stroke.
Image acquisition
MRI was performed using a 3 T MRI system (Tim Trio; Siemens, Erlangen, Germany). High resolution (1 × 1 × 1 mm3) 3D T1-weighted images were acquired using MPRAGE with repetition time (TR) = 2,530 ms, echo time (TE) = 3.45 ms, inversion time = 1,100 ms, flip angle = 7°, field of view = 256 × 256 mm2, 192 slices. DTI data were gained using a single-shot echo-planar imaging sequence with TR = 7,000 ms, TE = 91 ms, flip angle = 90°, matrix = 128 × 128, voxel size = 2 × 2 × 3 mm3, axial slices = 50, field of view = 256 × 256 mm2, non-collinear directions = 64, b = 1,000 s/mm2. Resting-state functional images were collected using an echo-planar imaging sequence with TR = 2,000 ms, TE = 30 ms, flip angle = 90°, field of view = 220 × 220 mm2, voxel size = 3.44 × 3.44 × 3 mm3, averages = 1, axial slices = 33.
DTI and resting-state functional data preprocessing
DTI data preprocessing were performed using the Pipeline for Analyzing Brain Diffusion Images toolkit (PANDA, http://www.nitrc.org/projects/panda/) [19] and FSL (http://fsl.fmrib.ox.ac.uk/fsl) [20]. The detailed preprocessing procedures can be obtained from our previous research [15, 17, 18]. The functional MRI images were preprocessed using Data Processing & Analysis for Brain Imaging (DPABI; http://rfmri.org/DPABI), which is based on Resting-state Data Analysis Toolkit (REST; http://www.restfmri.net) and Statistical Parametric Mapping (SPM12; http://www.fil.ion.ucl.ac.uk/spm). Briefly, preprocessing procedures included the first ten volumes’ removal, slice timing and motion correction, the transformation from individual space to Montreal Neurological Institute space, spatial smoothing with a Gaussian kernel of 4 × 4 × 4 mm3, and band-pass filtering (0.01-0.08 Hz). Participant data were excluded if they met the head motion criteria, which included head motion > 2 mm translation or a 2° rotation in any direction. For the ALFF calculation, the time courses were first converted to the frequency domain using fast Fourier transform. The square root of the power spectrum was then computed and averaged across 0.01-0.08 Hz at each voxel. This averaged square root was considered as the ALFF [12-14]. To reduce the global effects of variability across subjects, the ALFF of each voxel was divided by the global mean ALFF value obtained for each participant.
Regions of interest (ROIs) analysis
For ROI analysis, we selected the bilateral ICP and the anterior lobe of the cerebellum (lobules IV-VI). The bilateral ICP was extracted from the Johns Hopkins University white matter tractography atlas (JHU-ICBM-DTI-81-WMPM-90p) [21] and the bilateral anterior lobe of the cerebellum from the Automated Anatomical Labeling atlas. Averaged FA and MD values for each ICP were obtained for each subject using the corresponding normalized diffusion metric maps. The mean ALFF values of each anterior lobe of the cerebellum were acquired for each participant using the individual ALFF maps.
Statistical analysis
Categorical variables were compared using the Pearson χ2 or Fisher exact tests (when the expected number was ≤ 5). Age was compared between the patient group and the control group using the independent t test after normality testing by the Shapiro-Wilk test. In the patient group, FM-LL scores obtained at multiple time points were compared using the paired-samples Wilcoxon signed-rank test. Analyses of variance with repeated measures were employed to compare absolute values within ROIs between time points, and a post hoc analysis was conducted based on the Bonferroni correction for multiple testing. A Spearman’s partial correlation analysis was used to determine correlations between changes in FM-LL scores and changes in ICP diffusion parameter values when adjusting for age and gender as covariates, which was defined as the difference between FM-LL scores or diffusion parameter values at baseline and 12 weeks. All tests were 2-sided and P < 0.05 was considered significant. The FA and MD values of the ICP and the ALFF values of the anterior lobe of the cerebellum at each time point were compared with those of controls by independent t-test with a corrected P value threshold (corrected Pthreshold = 0.05/3). All analyses were conducted using SPSS 16.0 for Windows (SPSS, Chicago, IL, USA).