Abstract
AbstractBackgroundThere is increasing evidence from pathology and neuroimaging suggesting that the structural abnormalities which characterize leukoaraiosis are present within regions of normal-appearing white matter. In this study, we aimed to verify whether signal intensity on FLAIR imaging in normal-appearing white matter is related to the leukoaraiosis burden.MethodsWe performed a cross-sectional study of adult patients admitted with a diagnosis of acute ischemic stroke or transient ischemic attack. Leukoaraiosis was segmented using a semi-automated method involving manual outlining and signal thresholding. White matter regions were segmented based on the probabilistic tissue maps from the International Consortium for Brain Mapping 152 atlas. Also, white matter was further segmented based on voxel-distance from leukoaraiosis borders. Normalized mean FLAIR signal intensity on normal-appearing white matter (NAWMM) was used as a dependent variable in univariate and multivariate statistical analysis, and leukoaraiosis volume quartiles (LKAV) and clinical data as independent variables.ResultsOne-hundred consecutive patients were selected for analysis (53% female, mean age 68 years). NAWMMwas higher in the vicinity of leukoaraiosis and progressively lower at increasing distances from leukoaraiosis. NAWMMwas independently associated with leukoaraiosis volume. In voxels in the vicinity of leukoaraiosis borders, there was a linear association between LKAVand NAWMM(B=0.03;P<0.01). In voxels non-adjacent to leukoaraiosis borders, there was a non-linear association, with higher values of NAWMMamong patients in the second and third quartiles, and lower in the first and fourth quartiles (B=-0.21 for the second order term;P<0.01). In multivariate analysis, leukoaraiosis was the only variable independently related to NAWMM.ConclusionsOur results show that normal-appearing white matter exhibits subtle signal intensity changes that are related to leukoaraiosis burden. The neuroimaging signature of these subtle changes suggests that they might not be a milder form of leukoaraiosis, but rather rather reflect widespread etiopathogenic processes underlying small-vessel disease and thus, leukoaraiosis development.
Publisher
Cold Spring Harbor Laboratory