Default‐mode connectivity mediates the association between cardiovascular and metabolic risks and cognition: a pilot study. (20th December 2022)

• Record Type: Journal Article
• Authors: Rashid, Barnaly; Glasser, Matthew F.; Nichols, Thomas; Van Essen, David; Juttukonda, Meher R; Schwab, Nadine A; Yacoub, Essa; Lovely, Allison; Terpstra, Melissa; Harms, Michael P; Bookheimer, Susan Y.; Ances, Beau M; Arnold, Steven E.; Salat, David H
• Is Part Of: Alzheimer's & dementia. Volume 18(2022)Supplement 5
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-12-20
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.067131
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Prior studies have linked increased vulnerability to cognitive decline later in life with exposure to cardiovascular and metabolic disease risk factors. Further, these classic risk factors are associated with alteration to brain structure and function. This project investigates whether altered default‐mode network (DMN) functional connectivity (FC) mediates the association between cardiovascular and metabolic (CVM) health and cognition. Method: Structural, Resting‐state fMRI, and fieldmap scans were collected in a large sample of 930 healthy volunteers aged 36 to 90+ years old (mean ± SD: 61.6 ± 16 years; female= 55.6%) from the Human Connectome Project in Aging (HCP‐A) study. Data were preprocessed using the currently recommended HCP‐A pipelines, including precise spatial preprocessing, multi‐modal cross‐subject alignment, and selective temporal denoising. FC values were computed using full correlation with FSLNets amongst the neocortical brain areas of HCP‐A's multi‐modal parcellation and standard FreeSurfer subcortical structures and converted to z‐scores with the Fisher transform. A composite CVM index (CVMI) was computed with variables related to general CVM health, and a cognitive composite index (CCI) encompassed measures of general cognition, memory, processing speed, and executive functioning. Figure 1 presents the mediation model implemented using MATLAB. The mediation effects were assessed in the full sample (FDR corrected, p<0.05), as well as separately in 'younger' participants below 65 years (n = 546) and 'older' participants aged 65+ years (n = 384) with and without sex stratification (uncorrected). Result: Figure 2 highlights an indirect effect between CVMI and CCI that is mediated by DMN FC. In the full sample, the CVMI shows a significant negative direct effect on CCI (b=‐0.54, p<.001). This effect is mediated by DMN FC as seen in Figure 2(a), which is well‐replicated in the sex‐stratified analyses. Similar overlapping effects are observed in the younger and older groups. Figure 3 presents an example of connectivity and age association differences between low and high CCI groups. Conclusion: Our results indicate that cardiovascular and metabolic health impacts cognitive functioning, and that this is, at least in part, mediated by DMN FC.