Plasma alkylresorcinols reflect gluten intake and distinguish between gluten-rich and gluten-poor diets in a population at risk of metabolic syndrome

Research output: Contribution to journalJournal articleResearchpeer-review

Mads Vendelbo Lind, Mia Linda Madsen, Jüri J Rumessen, Henrik Vestergaard, Rikke Juul Gøbel, Torben Hansen, Lotte Lauritzen, Oluf Borbye Pedersen, Mette Bredal Kristensen, Alastair B Ross

BACKGROUND: Many patients with celiac disease experience difficulties in adherence to a gluten-free diet. Methods for testing compliance to a gluten-free diet are costly and cumbersome. Thus, a simple biomarker of gluten intake is needed in a clinical setting and will be useful for epidemiologic studies investigating wider effects of gluten intake.

OBJECTIVE: The aim was to evaluate plasma total alkylresorcinol concentrations as a measure of gluten intake.

METHODS: In this randomized, controlled, crossover intervention study in 52 Danish adults with features of the metabolic syndrome, we compared 8 wk of a gluten-rich and gluten-poor diet separated by a washout period of ≥6 wk. We measured fasting plasma concentrations of alkylresorcinols to determine if they reflected differences in gluten intake as a secondary outcome of the original study. In addition, we investigated in 118 Danish adults the cross-sectional association between self-reported gluten intake and plasma alkylresorcinols in the same and a similar study at baseline. We used mixed-model ANCOVA for examining treatment effects, a classification tree to determine compliance to the gluten-poor diet, and linear regression models for examining baseline correlation between plasma alkylresorcinol concentrations and gluten intake.

RESULTS: Plasma total alkylresorcinols decreased more during the gluten-poor period (geometric mean: -124.8 nmol/L; 95% CI: -156.5, -93.0 nmol/L) than in the gluten-rich period (geometric mean: -31.8 nmol/L; 95% CI: -63.1, -0.4 nmol/L) (P < 0.001). On the basis of the plasma alkylresorcinol profile, we built a classification tree to objectively determine compliance and found an overall participant misclassification error of 3.9%. In the cross-sectional study we found a 5.6% (95% CI: 2.4%, 8.9%) increase in plasma total alkylresorcinols per 1-g increase in reported gluten intake (P < 0.001).

CONCLUSION: We propose the use of plasma alkylresorcinols to monitor compliance to a gluten-free diet as well as to help investigations into the possible effects of gluten in the wider population. This trial was registered at as NCT017119913 and NCT01731366.

Original languageEnglish
JournalJournal of Nutrition
Issue number10
Pages (from-to)1991-1998
Number of pages8
Publication statusPublished - 2016

    Research areas

  • The Faculty of Science - Celiac disease, Biomarkers, Gluten sensitivity, Gluten-related disorders, Coeliac disease, Gluten intolerance, Gluten-free diet

ID: 165940191