Physical activity and metabolic syndrome in relation to incident dementia in UK biobank

The purpose of this study was to investigate the relationship among physical activity (PA), metabolic syndrome (MetS) and dementia based on UK Biobank.

A total of 220,372 dementia-free participants over the age of 50 had available data for outcomes and exposures components were included. Four PA groups [High, Medium, Low and No Moderate-to-vigorous PA (MVPA)], two MetS groups (MetS and No MetS) and eight joint groups were categorized to explore the effects of PA level and MetS status on dementia outcomes. Multivariate Cox proportional hazards regression model was used to explore the associations between PA, MetS and new-onset dementia.

During a median follow-up of 13 years, 4778 all-cause dementia [ACD, including 2997 Alzheimer's disease (AD) and 960 vascular dementia (VD) events] were recorded. In multivariate models adjusting for genetic and biological factors, lifestyle factors and socioeconomic factors, participants with MetS had a 25% increase in the risk of dementia. In terms of PA, the incidence of dementia in individuals without MVPA increased by 46%. Cox proportional hazards regression and Kaplan-Meier analysis all showed that participants with MetS and No MVPA had remarkable highest risk in eight joint groups (for ACD, 1.43 (1.01-2.02); for AD, 1.61 (0.73-1.84); for VD, 3.14 (1.60-6.14), compared with the reference group of No MetS-High PA).

There is a synergistic correlation of MetS and PA level with dementia (ACD, AD and VD), which is significantly exacerbated in participants with MetS and No MVPA. Even low levels of PA may also effectively decrease the harmful associations between MetS and dementia.