The gut microbiome is involved in nutrient metabolism and produces metabolites that, via the gut–brain axis, signal to the brain and influence cognition. Human studies have so far had limited success in identifying early metabolic alterations linked to cognitive aging, likely due to limitations in metabolite coverage or follow-ups. Older persons from the Three-City population-based cohort who had not been diagnosed with dementia at the time of blood sampling were included, and repeated measures of cognition over 12 subsequent years were collected. Using a targeted metabolomics platform, we identified 72 circulating gut-derived metabolites in a case–control study on cognitive decline, nested within the cohort (discovery n = 418; validation n = 420). Higher serum levels of propionic acid, a short-chain fatty acid, were associated with increased odds of cognitive decline (OR for 1 SD = 1.40 (95% CI 1.11, 1.75) for discovery and 1.26 (1.02, 1.55) for validation). Additional analyses suggested mediation by hypercholesterolemia and diabetes. Propionic acid strongly correlated with blood glucose (r = 0.79) and with intakes of meat and cheese (r > 0.15), but not fiber (r = 0.04), suggesting a minor role of prebiotic foods per se, but a possible link to processed foods, in which propionic acid is a common preservative. The adverse impact of propionic acid on metabolism and cognition deserves further investigation.

Background: Low blood status in several nutritional compounds, including long-chain omega-3 fatty acids (LC n-3 PUFA), carotenoids, and vitamin D, have been associated with a higher risk to develop dementia. Nutritional deficiencies may potentiate each other regarding dementia risk; yet the association of multiple nutritional deficiencies with dementia has been little explored. Objective: To develop an index of micronutritional biological status (MNBS) for the screening of multi-micronutritional deficiencies associated with the risk of dementia in a prospective population-based cohort of older persons. Methods: We included participants from the Bordeaux Three-City study, who were free of dementia at baseline, had blood measurements of LC n-3 PUFA, carotenoids, and 25(OH)D, and who were followed for up to 18 years for dementia. We used penalized splines in Cox models to model dose-response relationships of each nutritional component with the risk of dementia and construct a risk index. Results: 629 participants with an average age of 73.1 years were included in the study. Each increase of 1 SD of the MNBS index was associated with a 46%higher risk of dementia (HR = 1.46, 95%CI 1.23; 1.73). Participants with highest index ([mean+1SD; max]) had a 4-fold increased risk of dementia compared with participants with a low index ([min; mean-1SD]) (HR = 4.17, 95%CI 2.30; 7.57). Conclusion: This index of assessment of micronutritional biological status is a practical tool that may help identify populations with inadequate nutritional status, screen eligible individuals for nutritional prevention in primary care, or for supplementation in preventive trials of dementia.

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY