Plasma uric acid levels and risk of dementia in a population-based cohort study

Background: Given the opposing properties of uric acid (UA), which are intracellular prooxidant action and extracellular antioxidant action, the association of circulating UA levels with dementia remains controversial. We aimed to ascertain whether both lower and higher plasma UA levels are associated with the risk of incident dementia among middle-aged and older population. Methods: 1685 participants (530 men and 1155 women) aged 40-69 years at baseline (1990-1993) were randomly selected for plasma UA measurement from base cohort participants who responded to the baseline questionnaire and provided blood samples. They were followed for dementia (disabling dementia requiring care; hereinafter dementia) from 2006 to 2016 using certification records for national long-term care insurance in Japan. A Cox proportional hazards model adjusted for various lifestyle factors and past medical history (cardiometabolic disease) was applied for overall participants and sex. Results: Dementia was diagnosed in 240 participants (14.2 % overall: 16.0 % in men and 13.4 % in women). No statistically significant association with plasma UA was found in overall participants. Compared to participants with UA of 5.1-6.0 mg/dL, men with >= 6.1 mg/dL showed fully adjusted hazard ratios of 1.46 (95% confidence interval: 0.78-2.75) for 6.1-7.0 mg/dL and 1.89 (0.97-3.66) for >= 7.1 mg/dL, while women with >= 6.1 mg/dL showed 1.13 (0.54-2.38). Conclusions: No statistically significant association between plasma UA level and risk of dementia was found in overall participants or by sex.