Abstract

Introduction: Dementia in Asia, presents distinct neuropathological features, with White Matter Hyperintensities (WMH) emerging as critical indicators of small vessel disease. WMH, classified into Deep White Matter Hyperintensities (DWMH), Periventricular Hyperintensities (PVH), and Fazekas-Total, exhibit high prevalence in Asian populations. Although WMH are prevalent and recognized indicators of small vessel disease, the domain-specific cognitive impact of WMH location remains unclear. We hypothesized that examining both cognitively normal (CN) and mild cognitive impairment (MCI) groups separately would clarify whether WMH topography exerts distinct effects at different stages of cognitive decline.

Methods: 430 participants were recruited, and a cross-sectional analysis performed. Eight domains of cognition were assessed: global cognition, learning/memory, language, executive function, attention, visuo-spatial, working memory, and processing speed. Correlation and stepwise regression (with False Discovery Rate correction) were performed. 217 participants were classified as Mild Cognitive Impairment (MCI) and 213 participants as Cognitively Normal (CN) as per National Institute on Aging-Alzheimer's Association criteria.

Results: In CN participants, higher Fazekas-Total was associated with impaired attention (p = 0.015, β = 0.422) while higher DWMH was associated with poorer learning and memory (p = 0.460, β = 0.003). In MCI, higher Fazekas-Total was associated with poorer learning and memory (p = 0.195, β = 0.0313).

Discussion: This study demonstrates that WMH burden—particularly DWMH—exerts differential domain-specific effects in CN versus MCI populations, underscoring the importance of WMH topography in early cognitive changes. Our results suggest that evaluating DWMH separately from overall WMH may refine clinical assessments and mechanistic understanding of vascular contributions to cognitive impairment. Future research should target upstream pathophysiological processes underlying region-specific WMH to improve early detection and intervention strategies.