While remote neuronal plasticity change has been previously described after stroke and may have an impact on clinical outcome^{1, 2}, the nature of changes in brain activation related to good recovery of motor function after stroke is still unclear. We aimed to use multimodal MRI to investigate cortical structural, functional and perfusion changes in recovered patients 6 months after subcortical ischemic stroke.

A total of 29 stroke patients who were well-recovered in global motor functional (Fugl-Meyer test score > 60 and whole extremity Fugl-Meyer test score > 90) (Figure 1) with a unilateral ischemic infarct, involving of the internal capsule and neighboring regions (Figure 2), and 30 age, gender and the years of education matched healthy subjects were investigated to undergo multimodal MRI techniques and behavioral tasks (Figure 3). The imaging data were acquired using GE Discovery MR 750 3.0 Tesla MR scanner.

1. The structural images were acquired by a brain volume (BRAVO) sequence with the following parameters: TR/TE= 8.2/3.2 ms; FOV= 256×256 mm²; matrix= 256×256; slice thickness= 1.0 mm, no gap; 188 slices. The perfusion imaging was performed using a 3D pcASL sequence with the following imaging parameters: TR/TE = 5025/11.1 ms; FOV= 240×240mm²; spiral in readout of eight arms with 512 sample points; reconstruction matrix= 128; slice thickness= 3mm, no gap; 48 axial slices. rs-fMRI data were obtained using a gradient-echo single-shot echo-planar imaging sequence with the following imaging parameters: TR/TE= 2000/41ms; FOV= 220×220mm² ; matrix= 64× 64; flip angle= 90°; slice thickness= 4mm; 0.5mm gap; 32 slices;190 time points.

2. The GMV were calculated using the optimized VBM technique implemented with SPM8. The 3D-T1W structure images were segmented into gray matter (GM), white matter and cerebrospinal fluid with the standard unified segmentation model in SPM8. Subsequently, the non-linear warping of GM images was performed with the exponentiated Lie algebra (DARTEL) technique³. The GM population templates were extracted from the entire image dataset using the diffeomorphic anatomical registration and were resampled to 1.5-mm cubic voxels. Finally, the modulated images were smoothed with an FWHM kernel of 8 mm.

3. The CBF maps were derived from the ASL difference images that were calculated via the subtraction between label images and control images. The individual CBF images were converted into the MNI-standard ASL template. The voxel size of the written normalized images was 2mm×2mm×2mm. We normalized the CBF images via the CBF of each voxel by dividing the mean CBF of the whole brain⁴. The normalized CBF images were spatially smoothed with a Gaussian kernel of 8 × 8 × 8 mm³ FWHM.

4. The rs-fMRI data were preprocessed using SPM8 and DPARSFA software including slice timing, realignment, spatial normalization, resampling to 3×3×3 mm³ voxels, filter and smoothed with an 8×8×8 mm³ full-width at half maximum. Then the rsFC analysis was performed.

5. We performed partial correlation analysis to investigate the association between the clinical behavior scores and the statistics results of GMV, CBF and rsFC.Compared with healthy controls, we found the stroke patients exhibited increased GMV in contralesional supplementary motor area (SMA) (Figure 4A), increased CBFs in contralesional superior frontal gyrus (SFG) (Figure 4B) and supramarginal gyrus (Figure 4C), and increased rsFC in contralesional middle temporal gyrus (MTG) (Figure 4D). Moreover, the increased GMV was negative correlation with the accuracy rate (Figure 5A) and the positive correlation with the reaction time (Figure 5B) of ANT. The increased CBFs of the contralesional SFG (Figure 5C), and supramarginal gyrus (Figure 5D) were all significant negative correlation with TMT-B.This study showed that the brain structural plasticity, regional CBF aberrant and functional reorganization were coexisted in the contralesional hemisphere in subcortical stroke patients. Stroke-induced alterations may present beyond the motor system and manifest as the involvement of cognitive functional systems even if the infarcts located in the motor pathways. The regional aberrant functional, perfusion and structural alteration may underlie neurophysiology of chronic stroke patients involving the cortical-subcortical regions.These findings highlight the importance of remote neuronal plasticity and functional reorganization in stroke recovery. It may have the potential of the imaging biomarkers for stroke recovery.