The associations of plasma carotenoids and <i>α</i> -tocopherol concentrations with fasting glucose in Cameroon

Previous estimations of the associations between fruits and vegetables (F/V) intake with diabetes markers showed mixed results, possibly partly because of the subjective assessment of dietary intake. We aimed to examine the relationship between plasma carotenoids and <i>α</i>-tocopherol (as objective markers of F/V intake) and fasting glucose. This was a population-based cross-sectional study in 592 adults in Cameroon. Self-reported F/V intake was assessed using the WHO STEPS questionnaire, and the biomarkers were analysed by HPLC. The mean age of participants was 38·5 (sd 8·6) years (63·7 % women). The median (IQR) number of times participants self-reported consuming fruits in a typical week was 2(1-5) and vegetables was 4(2-7) times/week. Plasma total carotenoids was positively correlated with self-reported intake of fruits (<i>r</i> = 0·13) and vegetables (<i>r</i> = 0·29), both <i>P</i>-values < 0·01. In unadjusted analysis, the difference in fasting glucose comparing the highest against the lowest tertile of the biomarkers concentrations was -0·28 (95 % CI -0·56, -0·001) mmol/l for total carotenoids and -0·31 (-0·59, -0·03) mmol/l for plasma <i>α</i>-carotene. The associations became stronger after adjusting for sociodemographics, lifesyle factors and cholesterol (-0·36 (-0·73, -0·002) mmol/l for total carotenoids and -0·41 (-0·79, -0·03) mmol/l for <i>α</i>-carotene). There was no evidence of an association between <i>α</i>-tocopherol and fasting glucose. We showed an inverse association of total carotenoids and <i>α</i>-carotene, objective indicators of F/V intake, with fasting glucose; suggesting that higher F/V intake may be beneficial for diabetes prevention in African populations where F/V intake is low.