Abstract
BackgroundThe concept of metabolic syndrome (MetS) was developed to identify individuals at higher risk of type 2 diabetes and cardiovascular disease, but its relevance for dementia remains unclear. We examined MetS in midlife for association with late-onset dementia, focusing on the thresholds of MetS components that carry risk for dementia.MethodsMetS components (waist circumference, blood pressure, triglycerides, HDL-C, and fasting glucose) were measured on 6,137 white participants < 60 years from the Whitehall II (WII) cohort study. A changepoint method in time-to-event analyses was used to identify optimal thresholds, and those exhibiting better performance for dementia were retained to develop a revised MetS definition. Results were validated on 171,886 participants in the UK Biobank (UKB) study.ResultsOver a median follow-up of 22.6 years in WII and 13.8 years in UKB, 522 and 418 late-onset dementia cases were recorded, respectively. Optimized thresholds for triglycerides and fasting glucose performed better than original MetS thresholds in WII, and were used to develop a revised MetS definition. The MetS scale had a linear association with dementia, and 1-component increment (range 0 to 5) was associated with higher dementia risk using the revised MetS definition (HR, 95% CI: 1.11, 1.03-1.19) but not the original MetS definition (HR, 95% CI: 1.06, 0.98-1.14) in WII. In UKB, the revised MetS definition exhibited better performance for dementia risk than the original definition (p for HR comparison < 0.01).ConclusionsMetS in midlife is potentially an important target for dementia prevention. However, the thresholds for triglycerides and glucose that carry risk need to be tailored specifically for dementia.