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

 
Poster B30, Wednesday, November 8, 3:00 – 4:15 pm, Harborview and Loch Raven Ballrooms

Brain Network Reorganization for Language after Complete Prenatal Hemispheric Infarction

Salomi S. Asaridou1, Özlem Ece Demir-Lira2, Danny Siu1, Susan Levine2, Steven L. Small1;1Department of Neurology, University of California, Irvine, 2Department of Psychology, The University of Chicago

Recent studies have identified a set of regions that comprise a “core” language network with many critical nodes in the left hemisphere. Disruptions of this network have been described in adults with stroke, with language outcomes that correlate with the degree of left intrahemispheric connectivity. Here, we present a rare case of an individual born without a left hemisphere who nevertheless developed age-appropriate language skills. Using a graph theoretic approach, we investigated the topological properties of brain networks that were congenitally restricted to the right hemisphere. Our patient suffered a prenatal infarction involving the entire left cerebral hemisphere. We collected MRI data on our case at age 14y;2mo as part of a larger study comparing individuals with perinatal stroke with typically developing control cases. Thirty-one typically developing children (age 13y;6mo± 9.9m; 5 left-handed, 18 males) with no reported history of neurological or developmental disorders served as controls. Language performance was tested with standardized language tests and our case fell within the control performance range. Diffusion weighted imaging (60 directions, b-values: 0, 1,000, 2,000s/mm2 ,voxel size=1.5mm3 ) data were preprocessed with FSL, the tensors were fitted, and deterministic fiber tracking was performed with Diffusion Toolkit. The T1-weighted image was segmented and parcellated into 83 areas with Freesurfer using the Lausanne 2008 atlas. Structural connectomes were computed with the Connectome Mapper using fiber length to construct 83-by-83 connectivity matrices, and the Brain Connectivity Toolbox enabled estimation of graph theoretical network measures of the thresholded, weighted, undirected connectivity matrices. Bayesian hypothesis testing was used to estimate the probability that a member of the control population would obtain a different score than the case in the network measures. In our case, the right hemisphere was significantly more densely connected compared to either hemisphere or whole brain in the control sample; it was also significantly less efficiently connected (more connections, less efficient network). This pattern could reflect the result of compensatory network changes, as having a more densely connected unique hemisphere could help accommodate for the absence of the left hemisphere. With respect to individual nodes, the left dorsal IFG (a core language network node) had significantly higher degree centrality than its right homologue in the control participants, contrasting with our patient, who had degree centrality in the right dorsal IFG that did not differ from that of the homologous left dorsal IFG of controls. Whereas in controls, degree centrality in the left STG was significantly higher than the right, our patient's right STG degree centrality was not significantly different than either the left or right STG from the control sample. These nodal findings are in agreement with evidence from left hemispherectomy patients showing left-like patterns of activation in their contralesional hemisphere, with particularly increased activity in the right IFG. To conclude, being born without a left hemisphere leads to dramatic reconfiguration of brain network connectivity, as demonstrated by global density and efficiency, as well as mirror-like network properties in core language nodes, as demonstrated by degree centrality.

Topic Area: Language Disorders

Back to Poster Schedule