Reliability of fasting plasma alkylresorcinol concentrations measured 4 months apart

Alkylresorcinols (ARs) have been suggested as specific dietary biomarkers of whole-grain intake. Evaluation of long-term reliability in repeatedly collected samples in population-based studies is needed to assess whether plasma AR reflect long-term exposure and may be used to study exposure-disease relationships in large-scale epidemiologic studies. The objective of this study was to analyze the reliability (reproducibility) of AR concentrations in fasting plasma measured 4 months apart.

AR concentrations (C17:0, C19:0, C21:0, C23:0 and C25:0 and total ARs) were measured in fasting plasma samples from 100 participants from the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam Study with an interval of 4 months between blood sampling.

Fasting plasma AR concentrations were not significantly different between the first and second measurement over the 4-month period. Geometric means of total AR concentrations were 48.8 (95% confidence interval (CI)=41.5-57.3) nmol/l at the first and 48.7 (95% CI=41.4-57.4) nmol/l at the second measurement (P=0.99). The intraclass correlation coefficients for total AR concentrations were 0.42 (95% CI=0.25-0.57) for all, 0.55 (95% CI=0.32-0.72) for women, and 0.17 (95% CI=-0.11-0.42) for men. Comparable results were obtained for the individual homologues.

Overall, we observed moderate reliability in concentration of total AR and individual homologues over time, although reliability was weaker in men than in women. These findings should be taken into account when using AR as blood biomarkers for whole-grain intake in large-scale epidemiologic studies.