The dysfunctional brain dynamic of Lewy body dementia and its behavioural and clinical correlates : an fMRI and EEG analysis

Abstract

Background: Lewy body dementia (LBD), which comprises dementia with Lewy bodies (DLB) and Parkinson’s disease dementia (PDD), is characterised by transient clinical symptoms such as cognitive fluctuations which may be caused by alterations of intrinsic brain dynamics. The aim of this thesis is to investigate how dysfunctional brain connectivity and dynamics relate to the cognitive LBD phenotype, especially to attentional impairment and cognitive fluctuations. Methods: In order to investigate behavioural aspects of cognitive fluctuations in LBD, reaction time (RT) data from an attention task were analysed to study how attentional impairment in LBD differs from Alzheimer’s disease (AD) and healthy controls. Additionally, brain structural correlates of attentional dysfunction were assessed using voxel-based morphometry. Subsequently, resting-state fMRI data were analysed using static and dynamic functional connectivity and dynamic network analyses. Faster brain dynamics were assessed by EEG microstate analysis. Results: AD and LBD patients exhibited slower and more variable RTs than controls, with greater impairment in LBD than AD. Extremely slow responses occurred with comparable frequency in both dementia groups. There were widespread correlations between RT abnormalities and structural changes in AD patients, but not LBD. Functional connectivity was decreased in DLB patients compared to controls, mainly in motor, temporal, and frontal networks with sparing of the DMN. Differences between AD and DLB were subtle. Considering time-varying connectivity, AD and DLB patients spent more time in sparse connectivity configurations than controls and switched less often into more highly connected states. Compared to controls, variability of global network efficiency was reduced in patients with DLB. Microstate analysis revealed a marked and generalised increase in microstate duration in LBD patients compared to controls, which was not seen in AD and was related to a loss of dynamic connectivity between basal ganglia/thalamic and large-scale cortical networks. Microstate slowing was correlated with fluctuation severity in the DLB group and with RT slowing and variability across all participants. Conclusions: Different aspects of RT performance are differentially affected by AD and LBD, with a ii difference in structural neural correlates. The dynamic connectivity and microstate results indicate a loss of brain dynamics in LBD which might lead to a breakdown of the intricate dynamic properties of the brain, thereby causing loss of flexibility that is crucial for healthy brain function. This might lead to a network configuration which gives rise to the cognitive LBD phenotype characterised by attentional impairment and cognitive fluctuations.