The contribution of white matter pathology, hypoperfusion, lesion load, and stroke recurrence to language deficits following acute subcortical left hemisphere stroke

• Published in: PloS one 2022-10, Vol.17 (10), p.e0275664-e0275664
• Main Authors: Sharif, Massoud S, Goldberg, Emily B, Walker, Alexandra, Hillis, Argye E, Meier, Erin L
• Format: Article
• Language: English
• Subjects: Aphasia; Basal ganglia; Biology and Life Sciences; Blood vessels; Brain damage; Brain injury; Complications and side effects; Degeneration; Diagnosis; Diffusion rate; Disease; Ganglia; Hemispheric laterality; Impairment; Ischemia; Language; Language disorders; Lesions; Leukoaraiosis; Medical imaging; Medicine and Health Sciences; Neurodegeneration; Neuroimaging; Research and Analysis Methods; Risk factors; Semantics; Social Sciences; Stroke; Stroke (Disease); Substantia alba; Thalamus
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0275664

Aphasia, the loss of language ability following damage to the brain, is among the most disabling and common consequences of stroke. Subcortical stroke, occurring in the basal ganglia, thalamus, and/or deep white matter can result in aphasia, often characterized by word fluency, motor speech output, or sentence generation impairments. The link between greater lesion volume and acute aphasia is well documented, but the independent contributions of lesion location, cortical hypoperfusion, prior stroke, and white matter degeneration (leukoaraiosis) remain unclear, particularly in subcortical aphasia. Thus, we aimed to disentangle the contributions of each factor on language impairments in left hemisphere acute subcortical stroke survivors. Eighty patients with acute ischemic left hemisphere subcortical stroke (less than 10 days post-onset) participated. We manually traced acute lesions on diffusion-weighted scans and prior lesions on T2-weighted scans. Leukoaraiosis was rated on T2-weighted scans using the Fazekas et al. (1987) scale. Fluid-attenuated inversion recovery (FLAIR) scans were evaluated for hyperintense vessels in each vascular territory, providing an indirect measure of hypoperfusion in lieu of perfusion-weighted imaging. We found that language performance was negatively correlated with acute/total lesion volumes and greater damage to substructures of the deep white matter and basal ganglia. We conducted a LASSO regression that included all variables for which we found significant univariate relationships to language performance, plus nuisance regressors. Only total lesion volume was a significant predictor of global language impairment severity. Further examination of three participants with severe language impairments suggests that their deficits result from impairment in domain-general, rather than linguistic, processes. Given the variability in language deficits and imaging markers associated with such deficits, it seems likely that subcortical aphasia is a heterogeneous clinical syndrome with distinct causes across individuals.