You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

Poster E21, Friday, November 10, 10:00 – 11:15 am, Harborview and Loch Raven Ballrooms

Lower axon density in residual temporal white matter is related to semantic paraphasia prevalence

Emilie McKinnon1, Jens Jensen1, Julius Fridriksson2, Chris Rorden2, Joseph Helpern1, Leonardo Bonilha1;1Medical University of South Carolina, 2University of South Carolina

Introduction: It is well-recognized that secondary white matter (WM) loss occurs after ischemic strokes, but the importance of residual WM integrity in shaping post-stroke language deficits (aphasia) is not completely defined. While most lesion-based neuropsychological studies use the post-stroke lesion to define the relationship between structural brain integrity and aphasia severity, diffusion MRI (dMRI) can be used to quantify the degree of WM compromise beyond the necrotic stroke lesion and inform on the impact caused by loss of WM. Biophysical modelling provides biological interpretations of dMRI, permitting the assessment of the microstructural properties that are mostly affected by the stroke, and the ones with the highest clinical significance. In this study, we investigated the relationship between damage to WM networks beyond the stroke lesion and semantically related errors during a confrontation naming in individuals with aphasia. We used diffusional kurtosis imaging (DKI) to model WM axonal water fraction (AWF) (Fieremans, 2011) and assess the integrity of WM axons in contrast with the extracellular space. We hypothesized that axonal loss, identified as AWF decrement, in the residual temporal lobe would be a strong determinant of semantic processing deficits and linearly related to semantic paraphasias. Methods: Twenty-four subjects (age=57±11y; 75%male; MRI time post-stroke=28±32m; WAB-AQ=53±23; semantic paraphasias=15±11%) with chronic (> 6 months) post-stroke aphasia underwent the Philadelphia Naming Test (PNT) and MRI imaging. Structural images (T1 and T2) and DKI (b=0, 1000, 2000 s/mm2, 64 directions) were acquired. Diffusional Kurtosis Estimator (Tabesh, 2011) was used to estimate diffusion and kurtosis tensors, from which AWF was calculated. A probabilistic WM mask was created using the clinical toolbox in SPM12. The WM mask and the JHU WM atlas were warped into native diffusion space using the nonlinear deformation field calculated by the clinical toolbox. The following regions of interest (ROI) were defined for both ipsilateral and contralateral sides: whole-hemisphere WM, temporal lobe WM (defined by the inferior longitudinal fasciculus) and parietal lobe WM (defined by the superior longitudinal fasciculus). All ROIs excluded the WM that was part of the stroke lesion. Correlation coefficients were performed to evaluate the relationship between the number of semantic paraphasias and AWF, controlling for the total lesion size. Results: Average AWF was significantly lower in the left-hemisphere, the left temporal and the left parietal lobes (p<0.001) compared with their right homologues. A decrease in AWF in the left compared to the right temporal (but not in the whole-hemisphere, and not in the parietal lobe) was associated with an increase in semantic paraphasias (r=0.53, p=0.009, controlling for lesion size). Conclusion: Our results suggest that widespread axonal loss occurs in the left hemisphere after a stroke. Importantly, axonal loss within the left temporal lobe is directly associated with semantic processing problems and this relationship is independent from lesion size, indicating that residual WM integrity plays an important role in post-stroke language deficits.

Topic Area: Language Disorders

Back to Poster Schedule