ABSTRACT
Background
Observational investigations have reported correlations between brain imaging-derived phenotypes (IDPs) and dementia, as well as dysfunctions in brain resting-state functional networks in dementia patients. However, the causal nature of these relationships remains largely unknown.
Methods
Herein we applied bidirectional two-sample Mendelian randomisation analysis to infer the causal relationships between 587 IDPs (N = 33 224) and 191 brain resting-state functional networks (n = 34 691) with dementia and its sub-types (AD, PDD, FTD and DLB; n = 3024–216 771) using genetic variants-single nucleotide polymorphism (SNPs) as instrumental variables.
Results
The forward MR identified 14 IDP phenotypes that are causally related to the risk of dementia, including frontotemporal dementia (FTD) and Lewy body dementia (DLB). For example, a decrease in the thickness of the right rostral middle frontal cortex was strongly associated with an increased risk of dementia. The reverse MR analysis revealed significant associations between 153 IDP phenotypes and the risk of FTD and DLB and between 73 rs-fMRI phenotypes and the risk of dementia and AD. For instance, a higher risk of DLB was associated with a decrease in FA in the right posterior thalamic radiation. Additionally, the risk of Alzheimer’s disease dementia is causally associated with reduced connectivity in the default mode and salience networks.
Conclusions
We identified 14 IDPs causally associated with dementia or its subtypes. We also identified potential causal effects of FTD and DLB on 153 IDPs and dementia and AD on 73 rs-fMRI phenotypes. Our findings provide insights into the aetiology of dementia and highlight structural brain changes and functional network impairments throughout the disease process. Furthermore, these results contribute to the identification of potential imaging-based predictors and therapeutic targets for dementia.