Differences in cerebral haemodynamics associated with long-term endurance training in older adults

Regular exercise can benefit cognitive functions in later life, but the mediating mechanisms remain unclear. Changes to brain vascular health are hypothesised to underpin training-induced brain health benefits, yet how long-term exercise training impacts cerebral haemodynamics is poorly understood. Therefore, this pre-registered cross-sectional study compared cerebral haemodynamics and cognitive functions between untrained (self-report <150 minutes/week moderate-intensity exercise) and long-term endurance trained (≥10 years) older adults (n = 28 per group, respective age = 66.0 ± 5.3 and 66.2 ± 5.2 years, respective peak oxygen consumption (V̇O_2peak) = 28.4 ± 4.7 and 42.3 ± 7.2 mL/kg/min). Multiple post-labelling delay arterial spin labelling (ASL) data were used to estimate cerebral blood flow (CBF) and arterial transit time (ATT) in grey matter at rest. Furthermore, in trained participants, global cerebral oxygen extraction fraction (OEF) was estimated using T₂-relaxation-under-spin-tagging (TRUST) data (n = 26). There were no significant between-group differences (trained - untrained) in global CBF (5.6 [-2.6, 13.9] mL/100 g/min, P = 0.197), but global ATT was shorter in trained participants (-0.11 [-0.21, -0.02] seconds, P = 0.031). Within regions fed by anterior cerebral arteries, CBF was greater in trained participants (15.8 [3.2, 29.1] mL/100 g/min), but the effect did not survive multiple comparison adjustment (P _adjusted = 0.064). Cognitive functions did not differ between the groups and were not associated with cerebral haemodynamics. OEF was not associated with V̇O_2peak in trained participants, but was strongly inversely associated with global CBF (β = -0.96 [-1.21, -0.75], P = 0.002). In conclusion, older adults engaging with long-term endurance training have shorter ATT, counteracting normal age-related ATT lengthening. Longer-term longitudinal studies are required to confirm training-induced regional CBF elevations, which would also counteract normal age-related CBF declines. There was no evidence of relationships with cognitive functions, although the sample was predominantly healthy, educated, and aged 60-70 years. Using multiple post-labelling delays is crucial for future ASL research to isolate CBF and ATT from one another, improving estimation accuracy and delineating between different age-related cerebrovascular mechanisms.