|| Checking for direct PDF access through Ovid
Introduction: The DASH dietary pattern emphasizes vegetables, fruits, and low-fat dairy products and is associated with improved cardiometabolic outcomes. No biomarkers exist for assessing adherence to this dietary pattern. The objective of the study was to use metabolomics to identify serum compounds associated with the DASH diet.Methods: We conducted untargeted metabolomic profiling in stored serum specimens collected from participants at the end of an 8-week multi-center, randomized, controlled feeding study (N=218) comparing the DASH diet to a diet typical of intake in the U.S. (control). Multivariable linear regression was used to compare the association of individual log-transformed metabolites between the two diets after adjusting for age, sex, race, education, body mass index, and hypertension. Partial least squares-discriminant analysis was used to identify a composite of compounds that discriminate between the DASH and control diets. The area under the curve was calculated as the cumulative ability to distinguish between diets.Results: Serum levels of 97 known metabolites were significantly different among participants randomized to the DASH diet compared to the control diet at the Bonferroni threshold (p<6.11x10-5; Figure). The majority of these 97 metabolites were lipids (n=64; 66.0%), followed by amino acids (n=15), xenobiotics including food components (n=10), cofactors and vitamins (n=6), carbohydrate (n=1), and nucleotide (n=1). The ten most influential metabolites for discriminating between the DASH and control diets were: N-methylproline, stachydrine, tryptophan betaine, theobromine, 7-methylurate, chiro-inositol, 3-methylxanthine, methyl glucopyranoside, β-cryptoxanthin, and 7-methylxanthine (C statistic=0.986).Conclusion: An untargeted metabolomic platform identified a broad array of serum metabolites that differed between the DASH and control dietary patterns. The composite of top ten metabolites may be used to assess adherence to the DASH dietary pattern.