Lesion mapping in acute stroke aphasia and its implications for recovery

Abstract

Patients with stroke offer a unique window into understanding human brain function. Mapping stroke lesions poses several challenges due to the complexity of the lesion anatomy and the mechanisms causing local and remote disruption on brain networks. In this prospective longitudinal study, we compare standard and advanced approaches to white matter lesion mapping applied to acute stroke patients with aphasia. Eighteen patients with acute left hemisphere stroke were recruited and scanned within two weeks from symptom onset. Aphasia assessment was performed at baseline and six-month follow-up. Structural and diffusion MRI contrasts indicated an area of maximum overlap in the anterior external/extreme capsule with diffusion images showing a larger overlap extending into posterior perisylvian regions. Anatomical predictors of recovery included damage to ipsilesional tracts (as shown by both structural and diffusion images) and contralesional tracts (as shown by diffusion images only). These findings indicate converging results from structural and diffusion lesion mapping methods but also clear differences between the two approaches in their ability to identify predictors of recovery outside the lesioned regions.

Keywords: Acute stroke; Aphasia recovery; Lesion mapping; Tractography; VLSM; White matter atlas.

Fig. 1

Structural, perfusion, and diffusion images acquired in a single patient with acute stroke and aphasia. The arrows indicate a large ischemic lesion in the left parieto-occipital region and smaller ischemic lesion in the frontal lobe. A. Non-contrast computerised tomography (CT) scan on admission that shows altered signal only in posterior lesion. B. Conventional structural MRI sequences where - similarly to the CT scan - the frontal lesion is barely visible on these contrasts. C. Pulsed continuous arterial spin labelling (pCASL) shows reduced cortical and subcortical blood perfusion in both posterior and anterior regions of the left hemisphere. D. Mean apparent diffusion coefficient (ADC) map and multishell (b-value = 0/500/1500/3000 s/mm2) diffusion weighted imaging (DWI) maps. The boundaries of the lesions vary according to the value of the diffusion weighting. A lower b-value (i.e. less diffusion-weighting) is more sensitive to fast diffusion and therefore to very early ischemic changes, for example due to oedema, but it is also affected by partial volume effects from cerebrospinal fluid. Higher b-value images are more sensitive to slow diffusion and better tissue contrast, for example, between grey and white matter or different white matter tracts. E. Diffusion tensor imaging permits for the generation of maps with additional anatomical information not available from structural MRI images. The fractional anisotropy (FA) map encodes scalar values between zero (indicating diffusion isotropy) and 1 (high degree of diffusion anisotropy). Low FA is typical of voxels containing fluid, cortex, or crossing fibres. The red-green-blue (i.e. RGB) colour-coded map indicates the main orientation of the tensor and therefore the average orientation of the fibres contained in each voxel. By convention red identifies a medial-to-lateral orientation (i.e. commissural pathways), green a longitudinal orientation (i.e. long association pathways), and blue a vertical orientation (i.e. projection pathways). F. Anisotropic power (AP) maps quantify the absolute amount of anisotropic information present in any measured High Angular Resolution Imaging (HARDI) signal profile and is more robust to noise compared to other anisotropy maps. With a higher b-value the anatomical detail encoded in the AP maps increases with better differentiation between lesioned and healthy tissue.

Fig. 2

Percentage lesion overlay maps based on lesion masks delineated on T1-weighted images.

Fig. 3

Percentage lesion overlay map based on lesion masks defined on anisotropic power (AP) images.

Fig. 4

Voxel-based lesion symptom mapping (VLSM) analysis based on A) acute T1-wighted masks and B) AP masks indicating damaged voxels predicting aphasia severity at six months (longitudinal AQ). Results are not corrected for multiple comparisons (P < 0.05 for Z > 1.64). Axial slices are numbered according to MNI z coordinate. The AP-based analysis shows a larger territory of symptom-lesion association extending into posterior regions of the temporal and fronto-parietal lobes.

Fig. 5

Tract-based spatial statistics (TBSS) analysis of acute imaging fractional anisotropy (FA) values and longitudinal language recovery (follow-up AQ). Cold colours indicate FA voxels negatively associated with longitudinal symptom severity; warmer colours indicate FA voxels positively associated with longitudinal symptom severity. Results are not corrected for multiple comparisons. Contrasts are controlled for age, sex, thrombolysis, and lesion size.

Fig. 6

Left (ipsilesional) and right (contralesional) hemisphere white matter tractography reconstruction from a single aphasic stroke patient (male, 28 years, AQ baseline = 45, AQ follow-up = 96.2). The upper panel shows the lesioned hemisphere and the reconstructions of the language pathways within this hemisphere. The lower panel shows the same tracts in the non-lesioned hemisphere (right). The scalar index of fractional anisotropy (FA) is mapped along each pathway. Brighter colours indicate reduced FA. The tract volume in the left hemisphere was reduced for most of the pathways and the anterior segment was completely damaged. In the right hemisphere, all pathways were reconstructed, including the long segment of the arcuate fasciculus. LS, long segment of the arcuate, AS, anterior segment of the arcuate, PS, posterior segment of the arcuate, FAT, frontal aslant tract, IFOF, inferior fronto-occipital fasciculus, UF, uncinated fasciculus.