The estimated absolute cardiovascular disease (CVD) risk level is known to be a useful surrogate marker for future cognitive impairment; however, evidence regarding its predictive validity in terms of cognitive subtypes is limited. We aimed to examine subtype-dependent differences in the associations between absolute CVD risk and the incidence of cognitive impairment in a community-dwelling older Japanese cohort.
This study comprised 1,641 cognitively intact older Japanese participants without CVDs at baseline. We estimated absolute CVD risk using WHO region-specific risk estimation charts and included age, sex, diabetes mellitus, smoking, systolic blood pressure, and total cholesterol at baseline, and the CVD risk level was stratified into the three following risk categories: low (<10%), moderate (10 to <20%), and high (≥20%). Objective cognitive screening was performed using a multicomponent neurocognitive test at baseline and follow-up, and the incidence of cognitive impairment over 48 ± 2 months was determined. The incidence of cognitive impairment in low-, moderate-, and high-CVD risk participants was 1.2, 3.0, and 5.4%, respectively, for amnestic subtypes and 5.8, 10.1, and 14.0%, respectively, for non-amnestic subtypes. After adjusting for potential confounding factors, the absolute CVD risk level was significantly associated with non-amnestic impairment but not with amnestic impairment.
The absolute CVD risk estimated using region-specific risk estimation charts in old age is useful to predict incidence of cognitive impairment. Strategies to screen populations at risk of cognitive impairment and to prevent progression to dementia should be cognitive subtype-specific.