Premium
Associations between white matter hyperintensities and cognitive dysfunction in Alzheimer’s, vascular and mixed dementias
Author(s) -
Lee Hyunwoo,
Wiggermann Vanessa,
Rauscher Alexander,
Beg Mirza Faisal,
Popuri Karteek,
Tam Roger,
Lam Kevin,
LiuAmbrose Teresa,
Dao Elizabeth,
Keridy Walid Ahmed Al,
Jacova Claudia,
Sossi Vesna,
Pettersen Jacqueline,
Benavente Oscar R.,
Hsiung GingYuek Robin
Publication year - 2020
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.043984
Subject(s) - hyperintensity , dementia , montreal cognitive assessment , cardiology , psychology , vascular dementia , cognitive decline , white matter , medicine , cognition , atrophy , audiology , magnetic resonance imaging , neuroscience , disease , radiology
Background Increased burden of white‐matter hyperintensities (WMHs) on MRI is common among different types of dementia. WMHs can be caused by various underlying etiologies and are non‐specific markers of tissue abnormality. Therefore, global measures such as WMH burdens may not distinguish between degrees of cognitive impairment. We explored the associations between lobar‐specific WMH volumes and cognitive dysfunction across different dementia subtypes. Method N=57 participants (7 Alzheimer’s disease [AD]/5 subcortical vascular dementia [SVaD]/7 mixed dementia [MixD]/38 vascular cognitive impairment [VCI]; Sex:34M/23F; Age:73±8yrs; pooled from three independent studies) were scanned on 3T MRI. WMHs were segmented on 3D‐FLAIR (MixD/AD/SVaD) or dual‐echo PD/T2‐weighted (VCI) images. T1‐weighted images were processed with Freesurfer 6.0. Left and right frontal, temporal, occipital and parietal lobes plus basal ganglia volumes were constructed using the Freesurfer segmentation outputs. We calculated the proportion of WMH burden within each lobe, i.e. the ratio of lobar WMH volume over the lobar volume. The Montreal Cognitive Assessment (MoCA) was used as a measure of cognitive function. We used a general linear model to determine the association between the WMH/lobar volume ratios and the MoCA scores. Covariates included age, sex, and presence of AD‐related (yes for MixD/AD, no for rest) and vascular‐related conditions (yes for SVaD/VCI/MixD, no for AD). Additionally, we assessed the relationship between the MoCA scores and the ratio of total WMH burden over the whole‐brain parenchymal volume. Result The model yielded a significant association between the MoCA scores and WMH/lobar ratios within the left (p=0.031) and the right (p=0.035) frontal lobes, with higher ratios predicting lower MoCA scores. The effect was not significant within the temporal, occipital, parietal lobes and the basal ganglia. The effect was insignificant at the whole‐brain level (p=0.08). Conclusion We found a significant association between the frontal lobe WMH burden and the MoCA scores, especially sensitive to frontal executive functions. This suggests that region‐specific assessments of WMHs may provide improved imaging‐cognitive correlations over the global WMH burden. Future studies with more consistent imaging protocols and broader cognitive/clinical assessments are warranted to further characterize the relationship between WMHs and cognitive impairment.