Premium
Perfusion and free water show opposite trends across tissue layers surrounding white matter hyperintensities in participants with dementia and mild cognitive impairment
Author(s) -
Donnay Corinne A.,
Fan Audrey P.,
DeCarli Charles S.,
Maillard Pauline
Publication year - 2021
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.052256
Subject(s) - hyperintensity , dementia , cardiology , diffusion mri , white matter , voxel , penumbra , medicine , cerebral blood flow , fluid attenuated inversion recovery , nuclear medicine , cognitive decline , cognition , psychology , magnetic resonance imaging , radiology , neuroscience , disease , ischemia
Abstract Background White matter hyperintensities (WMH) are associated with an increased risk of cognitive impairment and AD. While WMHs are thought to reflect vascular disease, the physiological basis of WMHs remain unclear. New imaging biomarkers are needed to understand the cascade of events leading to the development of WMH. The present MRI study aims to investigate changes in both Free Water (FW) and Cerebral Blood Flow (CBF) in WMH and its penumbra elderly cohort with and without cognitive impairment. Method Fluid‐attenuated inversion recovery, diffusion tensor imaging, and pulsed arterial spin labeling (ASL) perfusion images were acquired in participants with dementia (n=17), MCI (n=41), and normal cognition (n=130) with an 8‐channel coil on a 3 Tesla MRI (Siemens Tim Trio). ASL scan parameters included TR/TE: 4000/13ms, inversion time=1800, voxel size: 4x4x10mm 3 . CBF maps, calculated using the FMRIB software library, and FW maps were co‐registered into structural space. WMHs were segmented from fluid‐attenuated inversion recovery and T1‐weighted images using an automated protocol. For each subject, individual WMH lesions were identified and successively dilated by 1mm to create 4 WM layers surrounding the lesion core (Figure 1). For each biomarker (CBF and FW), linear mixed‐effect models were used to compare (1) normal WM to WMHs and their surrounding layers; and (2) WMHs to their surrounding layers (lme4 in R software). All models specified individual lesions per participant as a nested random effect as random effects, and age and cognitive diagnosis as fixed effects. Result CBF in WMHs and Layers 1‐4 were lower than the reference WM (p<0.001, Figure 2A). CBF reduced inlayer 1 (p=0.013) and increased toward normal values in layers 2‐4 (p<0.001) compared the WMH core (Figure 2B). In contrast, the mean FW content was higher in the WMHs and layers 1‐2 compared to the WM (p<0.001). The WMH core had higher FW content than the surrounding layers 1‐4 (p<0.001, Figure 3A, B). Conclusion Decreased CBF and increased FW was observed in tissue surrounding WMHs, suggestive of vascular and microstructural damage in a penumbra region beyond the WMH core. FW and CBF measures provide complementary information to understand the pathological mechanisms of WMH.