In a recent study published in the British Journal of Nutrition, Spencer et al. ⁽Reference Spencer, Abd El Mohsen, Minihane and Mathers¹⁾ reviewed the strengths and the limitations of the biomarkers of dietary polyphenol intake, since nutritional biomarkers may be a better measure of dietary exposure than self-reported dietary data. These authors identified the criteria that must be considered in the development of such biomarkers as the following: (i) robust methodology; (ii) sensitivity; (iii) specificity; (iv) bioavailability. Different polyphenols were reviewed as potential biomarkers by the authors; we suggest that resveratrol should also be considered. We analysed resveratrol metabolites as potential biomarkers of wine consumption in two randomised cross-over trials and a cohort study⁽Reference Zamora-Ros, Urpí-Sardà, Lamuela-Raventós, Estruch, Vázquez-Agell, Serrano-Martínez, Jaeger and Andres-Lacueva²⁾. Using a cut-off of 90 nmol/g, we were able to use urinary total resveratrol metabolite concentration to differentiate wine consumers from abstainers with a sensitivity of 72 % (60–84 %) and a specificity of 94 % (87–100 %). In these trials, urinary resveratrol was specific, as wine has been reported as the most important source of dietary resveratrol (98·4 %)⁽Reference Zamora-Ros, Andres-Lacueva and Lamuela-Raventós³⁾, has an adequate half-life and provided a good correlation between these measured values and the dietary data reported (r 0·654; P < 0·001). In addition, there is a robust analytical technique⁽Reference Urpí-Sardà, Jáuregui, Lamuela-Raventós, Jaeger, Miksits, Covas and Andres-Lacueva^4, Reference Urpi-Sarda, Zamora-Ros, Lamuela-Raventos, Cherubini, Jauregui, de la Torre, Covas, Estruch, Jaeger and Andres-Lacueva⁵⁾ using LC-MS-MS to determine urinary resveratrol metabolites and their pharmacokinetic parameters have been recently studied by Boocock et al. ⁽Reference Boocock, Faust and Patel⁶⁾.

Taking these points into consideration, we want to propose urinary resveratrol metabolites as a biomarker of grape product consumption; this would be a new nutritional biomarker which accomplish and fulfil the criteria of Spencer et al. ⁽Reference Spencer, Abd El Mohsen, Minihane and Mathers¹⁾.

We declare no conflict of interest.