Associations between gait velocity parameters and brain amyloid‐β levels in cognitive normal older adults: a cross‐sectional analysis from the AGUEDA trial

Background Amyloid beta (Aβ) is a protein associated with dementia‐related pathologies such Alzheimer’s Disease (AD). The accumulation of Aβ in the brain leads to neurodegeneration, cognitive decline, and motor impairment. Those factors might appear years before the onset of apparent symptoms. There is evidence for an association between gait performance and Aβ levels. Previous studies showed that gait variability has been associated with amyloid in older adults with mild dementia. Moreover, reduced gait velocity, has been associated with greater brain Aβ accumulation in adults with spontaneous memory complaints. However, the nature of this relationship needs further investigation. Thus, the aim of this study was to examine the associations between gait velocity and gait velocity variability, and brain amyloid burden in cognitively normal older adults. Our hypothesis was that gait parameters might differently predict amyloid‐β levels. Method Eighty‐six cognitively normal older adults participated in this study. The Optogait system (Microgate,Italy) was used to assess gait parameters. Participants walked back and forth across a 5‐m long track made up of two bars set 1‐meter apart at their maximum walking pace for 5 minutes. The mean and coefficient of variation (CV) of gait velocity (m/s) were calculated as gait parameters. All participants underwent T1‐weighted MRI scans and [18F] Florbetaben amyloid‐PET (positron emission tomography) scans. PET images were quantified using the Centiloid (CL) method with a 12 CL cutoff for amyloid positivity. The associations between gait parameters and amyloid burden were assessed by individual linear regressions unadjusted and adjusted by age, sex, and education as covariates. Result Table 1 shows descriptive characteristics of the participants (22% Aβ+; 19 participants). Results showed a positive trend for an association between gait variability and amyloid burden (unadjusted: β = 0.21, p = 0.05; adjusted: β = 0.20, p = 0.08). There was no association between gait velocity and amyloid burden (unadjusted: β = ‐0.08, p = 0.42; adjusted: β = ‐0.08, p = 0.54). Conclusion Greater gait velocity variability, but not gait velocity, was marginally associated with higher amyloid burden. This result suggests that gait velocity variability might be an important marker of amyloid burden in cognitively normal older adults.