Vascular Health Is Associated With Functional Connectivity Decline in Higher-Order Networks of Older Adults

Miranka Wirth; Malo Gaubert; Theresa Köbe; Antoine Garnier-Crussard; Catharina Lange; Julie Gonneaud; Robin de Flores; Brigitte Landeau; Vincent de la Sayette; Gaël Chételat

doi:10.3389/fnint.2022.847824

Vascular Health Is Associated With Functional Connectivity Decline in Higher-Order Networks of Older Adults

Front Integr Neurosci. 2022 Apr 26:16:847824. doi: 10.3389/fnint.2022.847824. eCollection 2022.

Authors

Miranka Wirth¹, Malo Gaubert¹, Theresa Köbe¹, Antoine Garnier-Crussard^{2

3

4}, Catharina Lange^{1

5}, Julie Gonneaud⁴, Robin de Flores⁴, Brigitte Landeau⁴, Vincent de la Sayette^{4

6}, Gaël Chételat⁴

Affiliations

¹ German Center for Neurodegenerative Diseases (DZNE), Dresden, Germany.
² Clinical and Research Memory Center of Lyon, Lyon Institute for Aging, Hospices Civils de Lyon, Lyon, France.
³ INSERM 1048, CNRS 5292, Neuroscience Research Centre, Lyon, France.
⁴ UNICAEN, INSERM, U1237, PhIND "Physiopathology and Imaging of Neurological Disorders," Institut Blood and Brain @ Caen-Normandie, Cyceron, Normandy University, Caen, France.
⁵ Department of Nuclear Medicine, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Berlin, Germany.
⁶ Department of Neurology, CHU de Caen, Caen, France.

Abstract

Background: Poor vascular health may impede brain functioning in older adults, thus possibly increasing the risk of cognitive decline and Alzheimer's disease (AD). The emerging link between vascular risk factors (VRF) and longitudinal decline in resting-state functional connectivity (RSFC) within functional brain networks needs replication and further research in independent cohorts.

Method: We examined 95 non-demented older adults using the IMAP+ cohort (Caen, France). VRF were assessed at baseline through systolic and diastolic blood pressure, body-mass-index, and glycated hemoglobin (HbA1c) levels. Brain pathological burden was measured using white matter hyperintensity (WMH) volumes, derived from FLAIR images, and cortical β-Amyloid (Aβ) deposition, derived from florbetapir-PET imaging. RSFC was estimated from functional MRI scans within canonical brain networks at baseline and up to 3 years of follow-up. Linear mixed-effects models evaluated the independent predictive value of VRF on longitudinal changes in network-specific and global RSFC as well as a potential association between these RSFC changes and cognitive decline.

Results: We replicate that RSFC increased over time in global RSFC and in the default-mode, salience/ventral-attention and fronto-parietal networks. In contrast, higher diastolic blood pressure levels were independently associated with a decrease of RSFC over time in the default-mode, salience/ventral-attention, and fronto-parietal networks. Moreover, higher HbA1c levels were independently associated with a reduction of the observed RSFC increase over time in the salience/ventral-attention network. Both of these associations were independent of brain pathology related to Aβ load and WMH volumes. The VRF-related changes in RSFC over time were not significantly associated with longitudinal changes in cognitive performance.

Conclusion: Our longitudinal findings corroborate that VRF promote RSFC alterations over time within higher-order brain networks, irrespective of pathological brain burden. Altered RSFC in large-scale cognitive networks may eventually increase the vulnerability to aging and AD.

Keywords: aging; cognition; longitudinal resting-state functional connectivity; modifiable risk factors; vascular risk.