Cardiorespiratory fitness and cerebral blood flow in cognitively normal older adults and individuals with coronary artery disease: the AGUEDA and Heart-Brain projects

Age and coronary artery disease (CAD) are known risk factors for cognitive decline and dementia, in which cerebral blood flow (CBF) has as a key role. Cardiorespiratory fitness (CRF) has shown consistent links with brain health and dementia, though its association with CBF and whether it differs depending on age or disease status remains limited. The main aim of this study was to examine the association of CRF, assessed through the six-minute walk test (6MWT) and peak oxygen uptake (VO2peak), with CBF in cognitively normal older adults and individuals with CAD. We hypothesized that CRF will be positively associated with global and regional CBF. Seventy-nine cognitively normal older adults from the AGUEDA trial and 84 individuals with CAD from the Heart-Brain trial were included in this cross-sectional analysis. Participants underwent 6MWT, and a 3D arterial spin labelling magnetic resonance imaging scan to assess global and regional CBF. In the Heart-Brain project, participants additionally conducted a cardiopulmonary exercise test from which VO2peak was determined. In the Heart-Brain sample, after adjusting for age, sex, education, mean arterial pressure and APOE4, CRF was positively associated with global CBF (6MWT: {beta}=0.26, P=0.04; VO2peak: {beta}=0.33, P=0.02). At a regional level, CRF was positively associated with CBF in the posterior cingulate cortex (6MWT: {beta}=0.26, P=0.04; VO2peak: {beta}=0.31, P=0.02), the anterior cingulate cortex (6MWT: {beta}=0.29, P=0.02), the precuneus (6MWT: {beta}=0.28, P=0.02; VO2peak: {beta}=0.32, P=0.01) and the hippocampus (VO2peak: {beta}=0.29, P=0.03). No significant associations were observed in the AGUEDA sample (all P>0.05). When adding body mass index (BMI) to the models, the associations were no longer statistically significant in either sample. The association between VO2peak and CBF was significantly mediated (i.e. indirect effect) by BMI (Indirect effect: {beta}=0.250 (95% CI 0.02;0.486), percentage of mediation=72.67%). CRF was positively associated with CBF in individuals with CAD, but not in cognitively normal older adults. Interestingly, the association of CRF with CBF was largely explained and mediated by BMI. Further studies are warranted to clarify the role of CRF and BMI in relation to CBF, the mechanisms involved and the implications for dementia risk prevention in older adults and individuals with CAD.