The bidirectional relationship between brain structure and physical activity: a cohort study in the UK Biobank

Physical activity has been suggested to project again brain atrophy, but also to be a determinant of physical activity in older adults. However, research is needed to confirm this bidirectional relationship. Therefore, this research aimed to study the bidirectional relationship between physical activity and brain structure in older adults from the UK Biobank. A total of 3,027 (62.45 {+/-} 7.27 years old, 51.3% females) had physical activity and MRI data at baseline and follow-up. Physical activity was assessed through a self-reported questionnaire. T1-weighted MRI and diffusion tensor imaging were used to quantify brain volumes and white matter microstructure, respectively. Cross-lagged panel models were performed to estimate bidirectional associations, and linear mixed-effects models (LMM) to investigate the consistency of findings. Overall, our main findings suggest that higher hippocampal volume (Standardized Beta Coefficient ({beta})=0.075, pFDR<0.001), frontal volume ({beta}=0.043, pFDR=0.037), and global fractional anisotropy ({beta}=0.042, pFDR=0.028) were associated with more do-it-yourself (DIY) activities (e.g., watering the lawn or digging) levels at follow-up. In addition, strenuous sports at baseline were positively associated with hippocampal volume over time ({beta}=0.011, pFDR=0.108). Although this association did not survive/was not significant after adjustment for multiple comparisons, was confirmed by the LMM. In contrast, higher levels of walking for pleasure were negatively associated with white matter volume at follow-up ({beta}= -0.026, pFDR=0.008). In conclusion, there is a bidirectional association between physical activity and brain structure in healthy middle-aged and older adults. However, further research is needed to understand why physical activity subdomains are associated differently with brain structures and features during aging.