White matter hyperintensity burden predicts domain-specific cognitive decline across the Alzheimer's disease continuum

Abstract

White matter hyperintensity (WMH), indicative of cerebral small vessel disease, has emerged as a potential biomarker for cognitive decline in Alzheimer's disease (AD). However, their predictive role across specific cognitive domains within the AD spectrum remains unclear. This study investigates the relationship between WMH volume and cognitive performance in memory, executive function, and language across the AD continuum. A cross-sectional analysis was conducted using data from the Alzheimer's Disease Neuroimaging Initiative (ADNI), comprising 557 participants categorized into cognitively normal (CN; n = 158), mild cognitive impairment (MCI; n = 334), and Alzheimer's dementia (AD; n = 65) groups. Cognitive function was assessed using composite scores for memory (ADNI-MEM), executive function (ADNI-EF), and language (ADNI-LAN). WMH volume was quantified through validated Bayesian segmentation of MRI data. Associations between cognitive scores and WMH volume, adjusted for age, gender, APOE epsilon 4 status, and vascular risk factors, were evaluated via multiple linear regression analyses. WMH volume showed numerically progressive increases from CN to MCI and AD groups; however, between-group differences did not reach statistical significance. Within the MCI group, significant negative associations emerged between WMH volume and memory (beta=-0.13, adjusted p = 0.045) and language scores (beta=-0.12, adjusted p = 0.045). Conversely, these relationships were absent in both the CN and AD groups. WMH volume relates specifically to declines in memory and language abilities, particularly in individuals with MCI. These results support using WMH measurements as early markers to identify cognitive decline in AD, potentially helping to guide earlier diagnosis and treatment decisions.