Carotenoids as biomarkers of fruit and vegetable intake in men and women.

High fruit and vegetable (FAV) intake is associated with a lower prevalence of chronic diseases. Identifying the ideal number of FAV servings needed to reduce chronic disease risk is, however, difficult because of biases inherent to common self-report dietary assessment tools. The aim of our study was to examine the associations between daily FAV intake and plasma carotenoid concentrations in men and women enrolled in a series of fully controlled dietary interventions. We compiled and analysed data from a group of 155 men and 109 women who participated in six fully controlled dietary interventions and compared post-intervention fasting plasma carotenoid (α-carotene, β-carotene, β-cryptoxanthin, lutein, lycopene, zeaxanthin) concentrations with regard to the daily FAV servings consumed by the participants. We found that plasma β-cryptoxanthin, lutein and zeaxanthin concentrations were positively associated with daily FAV servings (P≤0·005). However, daily FAV intake was negatively associated with plasma α-carotene (P<0·0005) and lycopene (P<0·0001) concentrations, whereas no association was noted with plasma β-carotene. When men and women were analysed separately, we found that for any given number of FAV servings consumed women had higher circulating lutein concentrations compared with men (P<0·01). Significant sex×FAV (P<0·0001) and sex×dietary β-cryptoxanthin (P<0·0005) interactions were also noted favouring higher plasma β-cryptoxanthin concentrations in women than in men for a given FAV consumption. Results from these fully controlled dietary feeding studies indicate that plasma β-cryptoxanthin and lutein concentrations can be used as robust biomarkers of FAV consumption. They also suggest the existence of sex differences influencing circulating β-cryptoxanthin and lutein concentrations following FAV consumption.