Abstract
ABSTRACT The association between plasma fatty acid profiles and dementia risk remains debatable, with additional complexity introduced by genetic susceptibility and serum biomarkers. Multivariable-adjusted Cox models were conducted to assess hazard ratios and 95% confidence intervals of dementia. The influences of APOE gene and dementia genetic risk score were investigated by nutrition-genetic interaction analyses. Mediation analyses were performed to assess the role of serum biomarkers. Among 117,884 UK Biobank participants, 1785 cases of dementia emerged after an average follow-up duration of 11.7 years. The highest quartiles of plasma concentrations of monounsaturated fatty acids (MUFAs) and polyunsaturated fatty acids (PUFAs) were associated with 29% and 28% reduced risk of dementia, respectively. Plasma linolenic acid (LA) and non-docosahexaenoic acid (non-DHA) n-3 PUFAs were also found to be inversely associated with dementia risk, while plasma saturated fatty acids, non-LA n-6 PUFAs, and DHA were not. Among White ethnicity individuals, the protective role of MUFAs was more pronounced in high APOE risk individuals, whereas the influence of PUFAs was marginally stronger in low APOE risk individuals. Non-LA n-6 PUFAs were found to be associated with elevated risk of dementia among White ethnicity individuals with low dementia genetic risk score. Serum triglyceride, glucose, and HbA1c partially mediated these associations. Our research underscores the importance of improving plasma fatty acid profiles based on individual genetic background to effectively prevent dementia. </p>