Summary
This controlled feeding study among 153 postmenopausal women from the Women's Health Initiative evaluated whether serum and urinary metabolomic profiles could serve as biomarkers of macronutrient intake. Metabolites alone provided modest predictive power for protein and carbohydrate (%E) with CV-R² of ~36–37%, but combining selected metabolite panels with established biomarkers (doubly labelled water and urinary nitrogen) substantially improved prediction, particularly for energy and absolute protein intake (CV-R² 45–56%). The findings suggest metabolomic approaches could complement existing dietary biomarkers for improved assessment of macronutrient exposure in epidemiological studies.
Regional applicability
The study was conducted in the United States using postmenopausal women; findings are relevant to dietary biomarker development globally but the predictive models may require validation in other populations and age groups. The metabolomic platforms employed (LC-MS/MS, NMR, GC-MS) are available internationally, supporting potential adaptation of the approach in United Kingdom research and dietary surveillance programmes.
Key measures
Cross-validated multiple correlation coefficients (CV-R²) for prediction of energy (kcal/d), protein (g/d and %E), carbohydrate (g/d and %E), and fat intake from serum metabolites (LC-MS/MS and lipidomics), urinary metabolites (¹H NMR and GC-MS), and combination with doubly labelled water and urinary nitrogen.
Outcomes reported
The study evaluated serum and urinary metabolomic panels for predicting daily intake of energy, protein, carbohydrate and fat, and their percentage contribution to total energy (%E). Cross-validated prediction models were built using targeted and untargeted metabolomic approaches, with and without established dietary biomarkers.
Topic tags
Dig deeper with Pulse AI.
Ask about this record, its theme or its relevance to UK farming and policy. Pulse AI uses selected catalogue evidence and cites the sources it draws on.