Cognitive impairment in patients with cerebral small vessel disease: comparison of MRI total burden and diffusion tensor tract-based spatial statistics

Abstract

Objective To compare the value of MRI total burden and diffusion tensor tract-based spatial statistics (TBSS) for identifying cognitive impairment in patients with cerebral small vascular disease (CSVD). Methods Consecutive patients with CSVD diagnosed in Suzhou Hospital Affiliated to Nanjing Medical University from January 2018 to June 2019 were enrolled. They were divided into cognitive impairment group and cognitive normal group. The MRI overall burden was evaluated, and TBSS analysis was performed to obtain regions with differences in diffusion parameters between the two groups. Multivariate logistic regression analysis was used to determine the independent risk factors for cognitive impairment. The receiver operating characteristic (ROC) curve was used to identify the recognition ability of the MRI overall burden and TBSS analysis for cognitive impairment in patients with CSVD. Results A total of 71 patients with CSVD were enrolled, including 41 (57.75%) in the cognitive impairment group and 30 (42.25%) in the cognitive normal group. There was a statistically significant difference in the proportion of patients with white matter hyperintensities between the two groups, but the MRI overall burden was not statistically significant difference. After adjusting for the confounding factors such as age and gender, multivariate logistic regression analysis showed that white matter hyperintensities was an independent risk factor for cognitive impairment in patients with CSVD (odds ratio 3.459, 95% confidence interval 1.196-10.007; P=0.022). Diffuse parameters of the white matter area, such as the genu of corpus callosum, anterior thalamic radiation, fasciculus longitudinalis superior, and frontal-occipital fasciculus in the cognitive impairment group were significantly different from those in the cognitive normal group (all P<0.05). The fractional anisotropy of left anterior thalamic radiation (r=0.252, P=0.036) and the radial diffusivity of the genu of corpus callosum (r=-0.246, P=0.042) were significantly correlated with cognitive scores. ROC analysis showed that the areas under the curve of the MRI total burden, the fractional anisotropy of left anterior thalamic radiation, and the radial diffusivity of the genu of corpus callosum for identifying cognitive impairment were 0.61, 0.56 and 0.57, respectively. Conclusion White matter hyperintensities is an independent risk factor for cognitive impairment in patients with CSVD. TBSS analysis may be superior to MRI overall burden assessment in the specific recognition of cognitive impairment in patients with CSVD. Some diffuse parameters have a certain correlation with the overall cognitive status, and may be used as early imaging markers for cognitive impairment in patients with CSVD. Key words: Cerebral small vessel diseases; Cognition disorders; Magnetic resonance imaging; Diffusion tensor imaging; White matter; Risk factors