Cognitive impact of lesion location in cerebrovascular disease: Expanding the boundaries of lesion-symptom mapping

Abstract

Cerebrovascular disease is a major cause of cognitive impairment and dementia, either by itself or in combination with Alzheimer’s disease. The contribution of vascular brain injury to cognitive impairment and dementia is collectively referred to as Vascular Cognitive Impairment (VCI). In the context of VCI, lesion location is considered a key determinant of cognitive impairment. Lesion-symptom mapping (LSM) is a powerful approach to studying the relationship between clinical symptoms and lesion location. The overarching aim of this thesis was to gain further insight into the cognitive impact of lesion location in cerebrovascular disease using LSM. We focused on infarct location in patients with acute ischemic stroke and white matter hyperintensity (WMH) location in memory clinic patients. Prior to the projects described in this thesis, a comprehensive map of strategic infarct locations was lacking and the relevance of WMH location was unclear. An important limiting factor in previous studies was incomplete brain lesion coverage due to limited sample sizes. Previous studies have shown that even with data from several hundred subjects, only a limited part (approximately 20%) of the brain could be analyzed. We proposed that large-scale multicenter studies might overcome this challenge. In Part 1 we aimed to enable LSM studies with better brain lesion coverage. For this purpose, we established an international collaborative research network: the Meta-VCI-Map consortium. We successfully implemented multicenter data processing, harmonization and analysis procedures to enable large-scale LSM studies. Upscaling of sample sizes resulted in marked improvement of brain lesion coverage. In Part 2 we aimed to create a more comprehensive map of strategic infarct locations, and further pinpoint the neuroanatomical correlates of specific cognitive and neuropsychiatric functions. Through a large-scale LSM study we created the most comprehensive map for risk of post-stroke cognitive impairment (PSCI) to date, in which 87% of the brain was included. We identified the left frontotemporal lobes, left thalamus and right parietal lobe as infarct locations with the highest risk of PSCI. Furthermore, we developed a visual rating scale for assessment of PSCI risk for direct clinical use. Finally, we identified novel neuroanatomical correlates for verbal memory, semantic and phonemic fluency, and depressive symptoms. In Part 3 we examined the relevance of WMH location in memory clinic patients with cerebral small vessel disease. We found that WMH in specific white matter tracts were associated with cognitive impairment in executive functioning, visuomotor speed and memory. Importantly, regional WMH volumes showed a stronger association with cognitive performance than total WMH burden. Furthermore, WMH location in posterior white matter regions were associated with increased cerebral amyloid burden, suggesting a relationship between WMH and amyloid-related processes. In conclusion, the work in this thesis put lesion location better on the map as key determinant of VCI.