Compensation or Pre-stroke Advantage? An Investigation of the Relationships Between Structural and Diffusion MRI Measures in the Right Hemisphere and Language and Cognitive Abilities in Left-hemisphere Stroke Survivors
Previous work indicates that structural changes in the right hemisphere following left hemisphere stroke may be related to language abilities. However, the mechanisms behind this relationship remain unclear, particularly regarding the relative contributions of gray and white matter. The present study examined how structural and diffusion measures in the right hemisphere differ between chronic left hemisphere stroke survivors and matched control subjects, and the relationships between language and cognition measures and these right hemisphere measures. T1-weighted MRI, diffusion tensor images (DTI), and a battery of cognitive tests were obtained from 27 chronic left hemisphere stroke survivors and 44 neurologically intact matched control participants. Cortical and volumetric measures of gray and white matter in regions of interest were obtained from the T1 images and compared between groups, and correlated with behavioral measures. Tract-Based Spatial Statistics and tractography methods from the DTI were examined in a similar manner. The T1 MRI-based analyses revealed that the stroke survivors did not differ from the control group in any of the gray or white matter volume measurements. The cortical thickness and mean curvature analyses identified right lateral frontal and insular ROIs exhibiting thinner and greater curvature (an indication of atrophy) in the left hemisphere stroke survivors compared to controls. The DTI-based results showed that the stroke survivors had lower fractional anisotropy and fewer reconstructed fibers in the right language ventral-stream tracts. Regarding correlations between the right hemisphere measures and behavioral performance, there were no significant results within the DTI data, and only one significant result in the gray matter analyses: faster processing speed was correlated with greater cortical thickness in the right frontal cortex in chronic left hemisphere stroke survivors. Overall, the present study provides support for the idea that the right hemisphere exhibits post-stroke changes, particularly in right dorsal stream gray matter and the ventral stream’s white matter, and that these differences are not captured by T1-imaging alone; in fact, the DTI tract-specific analyses were perhaps the most revealing. Future studies are needed, perhaps incorporating functional neuroimaging, to elucidate how these right hemisphere differences in left hemisphere stroke survivors is related to language recovery.