Self‐reported physical activity levels do not predict brain beta amyloid levels in older cognitively normal adults over time

Abstract

AbstractBackgroundAnimal studies have consistently shown that physical activity reduces brain levels of beta‐amyloid (Aβ). However, human studies of physical activity and Alzheimer’s disease (AD)‐related brain biomarkers have produced somewhat inconsistent findings. To date, most human studies examining physical activity and Aβ have utilised cross‐sectional analyses and relatively small sample sizes. As brain Aβ is detectable decades prior to the onset of clinical symptoms, examination using longitudinal brain imaging is vital to provide insight into how ongoing physical activity levels impact the progression of AD. In addition, the causal direction of these relationships requires further investigation; some research suggests physical activity levels may decline prior to dementia diagnosis.MethodHabitual physical activity and Aβ measures were collected over 15 years from 731 cognitively normal older adults (60 years and over) participating in the Australian Imaging, Biomarkers and Lifestyle Study of Ageing. Physical activity was assessed via the self‐reported International Physical Activity Questionnaire and brain Aβ measures obtained via PET imaging. Regression modelling examined the cross‐sectional and longitudinal relationships between physical activity and brain Aβ. Moderation analyses examined the impact of apolipoprotein (APOE) ε4 carriage on the physical activity‐Aβ relationship. To further examine the causal nature of this relationship, the trajectory of physical activity levels for individuals as a predictor of either high or low Aβ brain load was also examined.ResultPhysical activity was not associated with brain Aβ at baseline (β = ‐0.001, p = 0.131) or over time (β = ‐0.101, p = 0.571). APOE ε4 status did not moderate the relationship between physical activity and brain Aβ over time (β = 0.04, p = 0.917). Brain Aβ levels were not associated with physical activity trajectory (β = ‐54.26, p = 0.590).ConclusionOur study did not identify a relationship between habitual physical activity and brain Aβ levels. Future longitudinal studies should examine middle‐aged participants to assess whether physical activity influences Aβ accumulation earlier in life.