Background Structural brain changes associated with Alzheimer’s disease (AD) can occur decades before the onset of symptoms. The Cardiovascular Risk Factors, Aging, and Dementia (CAIDE) score has been suggested to be associated with accelerated brain atrophy in middle-aged subjects but the regional specificity of atrophic areas remains to be elucidated. Methods 3T T1-weighted magnetic resonance imaging scans of 160 cognitively healthy middle-aged participants (mean age = 52) in the PREVENT-Dementia cohort, from baseline and from follow-up after 2 years, were examined. Images were preprocessed using Computational Anatomy Toolbox 12. Voxel-based morphometry was performed in FSL 6.0.1 to identify areas of grey matter (GM) volume differences both cross-sectionally and longitudinally between subjects with high and low baseline CAIDE score (CAIDE score was dichotomized at cohort-median). A GM percentage of change map was created for each subject for evaluation of atrophy over 2 years. Analyses were adjusted for age, gender, education and total intracranial volume. Results Compared to subjects with CAIDE score ≤ 6 (low risk), subjects with CAIDE score > 6 (high risk) showed lower GM volume in the temporal, occipital, and fusiform cortex and lingual gyrus at baseline, and greater percentage of GM loss over 2 years in the supramarginal gyrus, angular gyrus, precuneus, lateral occipital cortex, superior parietal lobule and cingulate gyrus (corrected P < 0.05). Conclusion This study demonstrated accelerated GM atrophy concentrated in several AD signature cortical regions in healthy middle-aged subjects with high CAIDE scores.

No abstract

Cognitive ageing is a complex process which requires multimodal approach. Neuroimaging can provide insights into brain morphology, functional organization and vascular dynamics. However, most neuroimaging studies of ageing have focused on each imaging modality separately, limiting the understanding of interrelations between processes identified by different modalities and the interpretation of neural correlates of cognitive decline in ageing. Here, we used linked independent component analysis as a data-driven multimodal approach to jointly analyze magnetic resonance imaging of grey matter density, cerebrovascular, and functional network topographies. Neuroimaging and behavioural data (n = 215) from the Cambridge Centre for Ageing and Neuroscience study were used, containing healthy subjects aged 18 to 88. In the output components, fusion was found between structural and cerebrovascular topographies in multiple components with cognitive-relevance across the lifespan. A component reflecting global atrophy with regional cerebrovascular changes and a component reflecting right frontoparietal network activity were correlated with fluid intelligence over and above age and gender. No meaningful fusion between functional network topography and structural or cerebrovascular signals was observed. We propose that integrating multiple neuroimaging modalities allows to better characterize brain pattern variability and to differentiate brain changes in healthy ageing.

Maintaining good cognitive function is crucial for well-being across the lifespan. We proposed that the degree of cognitive maintenance is determined by the functional interactions within and between large-scale brain networks. Such connectivity can be represented by the white matter architecture of structural brain networks that shape intrinsic neuronal activity into integrated and distributed functional networks. We explored how the function-structure connectivity convergence, and the divergence of functional connectivity from structural connectivity, contribute to the maintenance of cognitive function across the adult lifespan. Multivariate analyses were used to investigate the relationship between function-structure connectivity convergence and divergence with multivariate cognitive profiles, respectively. Cognitive function was increasingly dependent on function-structure connectivity convergence as age increased. The dependency of cognitive function on connectivity was particularly strong for high-order cortical networks and subcortical networks. The results suggest that brain functional network integrity sustains cognitive functions in old age, as a function of the integrity of the brain's structural connectivity.