P4‐138: Diffusion tensor imaging and cognition in patients with subcortical vascular cognitive impairment

Abstract

Interest in using MRI parameters as surrogate markers of disease to define outcome measures for trials is increasing. We investigated which MRI parameters could differentiate between subcortical vascular dementia (SVD) and subcortcal vascular mild cognitive impairment (SVMCI) and which MRI parameters best correlated with cognitive function in SVD and SVMCI. Twenty-six patients with SVD and thirteen patients with SVMCI were recruited. They underwent multimodal MRI (brain volume, flair-attenuated inversion recovery lesion load, lacunar infarct number, medial temporal atrophy, mean diffusivity [MD] and fraction anisotrophy [FA] from diffusion tensor imaging) and neuropsychological testing. The histograms of MD and FA in whole brain tissue (WBT), white matter hyperintensities (WMH), and normal appearing brain tissue (NABT) were produced. In comparison with the patients with SVMCI, those with SVD had larger median and average MD and smaller peak height of MD histogram in WBT and NABT, but not in WMH. There were no significant differences of normalized brain volume, percentage WMH volume, lacunar infarct number, and FA histogram parameters between the patients with SVD and those with SVMCI. The composite neuropsychology score was significantly associated with median and average MD in NABT as well as in WBT after controlling age, sex, education, normalized brain volume, percent WMH volume, lacunar infarct number, and medial temporal atrophy. The diffusion tensor imaging can differentiate SVD from SVMCI. In particular, ultrastructural abnormalities of NABT have a strong and independent effect on cognitive functions, whereas DTI metrics within WMH have no direct impact.