Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Right hemisphere functional connectivity increases in large left-hemisphere strokes but does not relate to aphasia severity
Poster Session D, Saturday, October 26, 10:30 am - 12:00 pm, Great Hall 3 and 4
Natalya Vladyko1, Andrew T. DeMarco1, Peter E. Turkeltaub1,2; 1Center for Brain Plasticity and Recovery, Georgetown University Medical Center, Washington DC, 2MedStar National Rehabilitation Hospital, Washington DC
Left hemisphere (LH) strokes can lead to widespread disruption of the LH language network (LN), which relates to aphasia severity. Recruitment of homotopic right hemisphere (RH) regions may partially support aphasia recovery, at least in some individuals. Most of the evidence for RH involvement in recovery comes from task-based fMRI, but there is emerging evidence for changes in RH functional connectivity (FC) after LH stroke as well. The nature of RH functional network reorganization and whether it contributes to aphasia outcomes remains unclear. Previously, we observed that RH FC is greater in LH stroke survivors than controls, with larger increases in individuals with large lesions than those with small lesions. In this study, we further examine RH FC following LH stroke, focusing on LN homotopes. Additionally, we examine how these changes in RH FC relate to behavioral outcomes, specifically to aphasia severity as measured by the Western Aphasia Battery (WAB). This study included 79 chronic LH stroke survivors (35F, mean age=61, mean time post-stroke=53 months, median lesion size=80cc) and 72 matched controls (36F, mean age=61). Participants underwent a 14.5-minute movie-watching fMRI scan. Following standard preprocessing, we computed FC between all 246 regions defined by the Brainnetome atlas. The LH stroke survivors were median-split into small lesion (n=37, median size=25cc) and large lesion (n=38, median size=148cc) groups. We compared the edgewise RH FC of each stroke group to that of controls. We then defined a LH LN as Brainnetome parcels that activated significantly during a semantic decision fMRI task in the control group. RH homotopic parcels defined the RH LN. Parcels outside the LH and RH LNs were classified as the non-language network (NLN). We calculated the proportion of significant FC changes of edges as compared to controls, both within and between the LN and NLN in both hemispheres. Additionally, we correlated the mean FC within and between the LN and NLN and the WAB scores, controlling for lesion size. In the small lesion group, 14% of LH LN edges and 5% of LH NLN edges had significantly lower FC than controls. Only 1% of RH edges had greater connectivity than controls. In the large lesion group, 28% of LH LN edges were decreased. The RH mainly showed increases in FC. The highest proportion of these increases was within the RH NLN (29% vs. 17% in the RH LN) and between the RH LN and NLN (28%). RH FC within and between networks did not correlate with WAB scores for either group. For the small lesion group, within LH LN (r=0.4, p<0.05), within LH NLN (r=0.37, p<0.05), and interhemispheric homotopic LN FC (r=0.43, p<0.05) correlated with WAB scores. There are widespread increases in RH FC following a large LH stroke. These changes are not specific to the RH LN or NLN and occur both within and between these defined networks. Additionally, these changes, at least as measured here, do not relate to aphasia severity, thus it remains unclear whether RH network reorganization plays a meaningful role in aphasia recovery.
Topic Areas: Disorders: Acquired,