Abstract WP156: Therapy-related Structural Plasticity Of Temporal White Matter Is Related To Naming Recovery In Aphasia
Stroke is the leading cause of adult disability in the USA and aphasia is a common consequence of dominant-hemispheric strokes. It is unclear why some recover with speech therapy, while others persist with debilitating deficits. One theory suggests that therapy-related brain plasticity provides the anatomical substrate for improvements in language. In this longitudinal study, we assessed the integrity of the ipsi- and contralateral inferior longitudinal fasciculus (ILF) using diffusional kurtosis MRI (DKI), and examined its relationship with aphasia-therapy related changes in semantic errors. 8 subjects (age = 52.0±7.2y; 62% male; MRI time post-stroke = 50.25±29.8m) with chronic post-stroke aphasia received Language Action Therapy for a period of 3 weeks. Structural images (T1 & T2) and DKI (30 directions, b= [1000, 2000 s/mm2]) were acquired. We applied an innovative form of tractography using Diffusion Kurtosis Estimator and a WM mask as a seeding region. Lastly, we optimized the automated fiber quantification software to acquire along-tract diffusion measurements resulting in 100 nodal mean kurtosis (MK), mean diffusivity (MD) and fractional anisotropy (FA) measurements along major tracts. Compared to the contralateral side, the ipsilateral ILF shows diffusion characteristics often found in damaged neuronal tissue: high MD, low FA, and low MK (figure1).
The variability is larger on the ipsilateral side, and lowest MK correlated significantly with an increase in semantic errors (r=-0.84, p<0.05). None of the associations with FA and MD reached significance. Additionally, a therapy related reduction in semantic errors was associated with a longitudinal increase in MK (r=-0.89, p<0.05). In conclusion, ILF integrity captured using MK relates to clinical performance with lower MK predicting worse semantic language production, whereas therapy related increases in microstructural complexity (higher MK) were associated with a decrease in semantic errors.
Author Disclosures: E. McKinnon: None. R. Glenn: None. A. Wabnitz: None. J. Jensen: None. J. Helpern: None. L. Bonilha: None. J. Fridriksson: None.
This research has received full or partial funding support from the American Heart Association, Mid-Atlantic Affiliate - District of Columbia, Maryland, North Carolina, South Carolina, Virginia.
- © 2017 by American Heart Association, Inc.