1271Green tea consumption is associated with annual changes in hippocampal volumes: a longitudinal study

Background To investigate the association between green tea consumption and the annual rate of change of gray matter (GM), white matter (WM), and hippocampal volumes in middle-aged and older Japanese community-dwellers. Methods A prospective cohort study with two years of follow-up was conducted with 1693 participants (aged 40–89 years). Green tea consumption (mL/day) data were collected with a 3-day dietary record. Volumes of GM, WM, and the hippocampus were estimated by T1-weighted brain magnetic resonance imaging and FreeSurfer software. The GM ratio, WM ratio, and hippocampal ratio (HR) were calculated as the percentages of total intracranial volume, respectively. Results The mean (SD) annual rate of change of hippocampal volume [(HR at baseline - HR at follow-up)/HR at baseline/follow-up years×100%] was 0.499 (1.128) (%). In the multivariable-adjusted general linear model, green tea consumption at baseline was negatively associated only with the annual rate of change of hippocampal volume (%) [β (95% CI) for each 1 mL/day increase in green tea consumption = -20.2E-5 (-35.0E-5 to -5.3E-5); P-value = 0.008]. No associations were observed for the annual rate of change of GM or WM volumes. The results remained significant when the analysis was limited to those with stable green tea consumption and were especially evident among individuals aged 65 years and older and among women. Conclusions In this study, each additional 100 mL/day of green tea intake was related to a reduction of approximately 5% in annual hippocampal atrophy. This association was especially evident among older individuals and among women. Key messages A negative association between green tea consumption and brain atrophy has been suggested; however, this relationship has not yet been investigated in humans. In this study, for the first time, a negative association between green tea consumption and annual hippocampal atrophy was observed in community-dwelling middle-aged and older individuals.