White matter hyperintensity patterns in vascular cognitive impairment

Abstract

The overarching aim of the work described in this thesis was to study white matter hyperintensity (WMH) patterns in order to better understand vascular cognitive impairment (VCI). We focused on memory clinic patients (part I) and patients with ischemic stroke (part II). More specifically, our objectives were 1) to evaluate different aspects of lesion patterns, in particular volume, 3D distributions and tract-based distributions, 2) to assess the relation between WMH patterns and cognition, and 3) to assess the relation between WMH patterns and dementia etiologies. Prior to the research projects described in this thesis, the knowledge on the clinical relevance of 3D WMH patterns in memory clinic patients and patients with ischemic stroke was limited. To enable studies of WMH patterns in VCI in sufficiently powered datasets, we pooled individual patient data from multiple cohorts through the Meta VCI Map Consortium. This way we were able to achieve high lesion coverage, both in memory clinic patients and patients with ischemic stroke, allowing us to perform comprehensive studies of WMH patterns in terms of causes and consequences. In memory clinic patients (part I) WMH distribution patterns were highly variable, and when assessing unusual patterns the total WMH distribution pattern should be considered. We also showed that different dementia etiologies have distinct WMH patterns, in particular vascular risk factors were associated with more anterior and infratentorial WMH, whereas amyloid-1-42 (Aβ42) pathology was associated with posterior WMH. We established WMH location as an important determinant of cognitive impairment. In particular, we identified four strategic white matters tracts and developed a strategic WMH score based on these tracts which was strongly associated with cognitive performance. In patients with ischemic stroke (part II), we showed that total WMH volume was associated with post-stroke cognitive functioning. We also showed that the concept of strategic WMH locations, as observed in memory clinic patients in part I, also applies to patients with ischemic stroke. The knowledge gained through these studies thus provides new leads on WMH patterns in VCI with regard to the following aspects: WMH lesion patterns: • For lesion symptom studies, to obtain sufficient WMH lesion coverage studies should scale up to sample sizes of thousands of patients • WMH distribution patterns are highly variable and when assessing unusual WMH patterns the total distribution pattern should be considered WMH patterns and cognition: • WMH volume is associated with post-stroke cognitive functioning • WMH location is an important determinant of cognitive impairment in memory clinic patients and patients with ischemic stroke WMH patterns and dementia etiologies: • Different dementia etiologies have distinct WMH patterns