Associations Between Accelerometer- Measured Movement Behaviour and MRI-Derived Adipose Tissue Depots in Type 2 Diabetes

Abstract Background The association between waking-time movement composition and depot-specific adiposity in type 2 diabetes is unclear. We examined associations of accelerometer-derived movement behaviours with MRI-derived abdominal subcutaneous adipose tissue (ASAT), visceral adipose tissue (VAT), and liver proton density fat fraction (PDFF). Methods In this cross-sectional EPSONIP study, adults with type 2 diabetes underwent neck-to-knee MRI and 7-day wrist accelerometry. Sedentary time, light-intensity activity, and moderate-to-vigorous physical activity (MVPA) were analysed compositionally using isometric log-ratio coordinates. Outcomes were standardised; ASAT and VAT were height-normalised. Models adjusted for age and sex. Results Of 257 participants with valid accelerometry, 239 had ASAT/VAT and 234 liver PDFF data. More non-sedentary relative to sedentary time was associated with lower ASAT (β − 0.405 SD; 95% CI − 0.685 to − 0.125) and VAT (β − 0.523 SD; 95% CI − 0.909 to − 0.137), but not liver PDFF. MVPA relative to light activity was not clearly associated with any outcome. Reallocating 10 min/day from sedentary to non-sedentary time corresponded to − 0.019 SD lower ASAT and − 0.025 SD lower VAT. Conclusions More non-sedentary relative to sedentary waking time was associated with lower ASAT and VAT, but not liver PDFF. The cross-sectional design precludes causal inference.