Assessing the validity of plasma phospholipid fatty acids as biomarkers of dairy fat intake using data from a randomized controlled intervention trial

Abstract

Background: Plasma phospholipid pentadecanoic acid (C15:0), heptadecanoic acid (C17:0), and trans-palmitoleic acid (trans-C16:1n-7) are correlates of dairy fat intake. However, their relative concentrations may be influenced by other endogenous factors, such as liver fat content, and their validity as biomarkers of dairy fat intake has yet to be established.

Objectives: We investigated whether liver fat content modifies relations between concentrations of C15:0, C17:0, and trans-C16:1n-7 (alone and in combination with iso-C17:0) and known dairy fat intake in the context of a randomized controlled intervention study. We further examined the proportion of dairy fat intake explained by these fatty acids on their own and when considering liver fat content.

Methods: We used data from a 12-wk intervention trial in which participants (n = 62) consumed diets limited in dairy (0.3 g/d of dairy fat), rich in low-fat dairy (8.7 g/d of dairy fat), or rich in full-fat dairy (28.5 g/d of dairy fat). We used linear regression models to examine relations between relative fatty acid concentrations and grams per day of dairy fat intake, liver fat percentage, and their interaction.

Results: Only trans-C16:1n-7 in isolation (β: 0.0004 ± 0.0002, P = 0.03) and combined with iso-C17:0 (β: 0.002 ± 0.0005, P < 0.0001) were consistently positively associated with dairy fat intake regardless of liver fat content. Trans-C16:1n-7 combined with iso-C17:0 also explained the greatest proportion of variation (35.4%) in dairy fat intake. C15:0 and C17:0 were not associated with dairy fat intake after adjusting for liver fat and were predicted to be higher in relation to increased dairy fat intake only among individuals with elevated liver fat.

Conclusions: The potential for liver fat to affect relative plasma phospholipid concentrations of C15:0 and C17:0 raises questions about their validity as biomarkers of dairy fat intake. Of the fatty acid measures tested, trans-C16:1n-7 combined with iso-C17:0, especially with adjustment of liver fat, age, and sex, may provide the most robust estimate of dairy fat consumption.

Keywords: biomarker; dairy; full-fat dairy; liver fat; plasma phospholipid fatty acids.

FIGURE 1

Distribution of plasma phospholipid fatty acid correlates of dairy fat intake at the end of the run-in (baseline) and intervention (follow-up) phases by diet intervention group (n = 62). Distribution of (A) relative C15:0 concentration, (B) relative C17:0 concentration, (C) relative trans-C16:1n–7 concentration, and (D) the combined measure of relative trans-C16:1n–7 and iso-C17:0 concentrations at baseline (following a 4-wk run-in diet period consisting of the limited dairy diet) and at follow-up at the end of the 12-wk diet intervention period in the limited dairy group (n = 20), the low-fat dairy group (n = 22), and the full-fat dairy group (n = 20). Circles represent baseline and follow-up fatty acid concentrations of individuals whose respective relative plasma phospholipid fatty acid concentrations decreased or did not change from baseline to follow-up; triangles represent baseline and follow-up fatty acid concentrations of individuals whose respective relative plasma phospholipid fatty acid concentrations increased from baseline to follow-up. The rhombus and the bold line indicate the means at each time point. Boxes indicate the 25th and 75th percentiles, and the middle lines represent the medians.

FIGURE 2

Predicted concentration of plasma phospholipid fatty acids in relation to dairy fat intake among those with 1%, 5%, 10%, and 15% liver fat content (n = 62). Predicted changes in relative concentrations of (A) C15:0, (B) C17:0, (C) trans-C16:1n–7, and (D) trans-C16:1n–7 + iso-C17:0 in the plasma phospholipids at the end of the intervention. Estimates derived from multiple linear regression models with robust standard errors testing for the interaction between total dairy fat intake during the intervention and liver fat content at follow-up and adjusting for age and sex. Gray circles represent the observed concentration of the respective plasma phospholipid fatty acid for each individual in relation to their estimated dairy fat intake during the intervention.