Dietary carotenoids are associated with cardiovascular disease risk biomarkers mediated by serum carotenoid concentrations

J Nutr. 2014 Jul;144(7):1067-74. doi: 10.3945/jn.113.184317. Epub 2014 Apr 17.


Hyperlipidemia and elevated circulating C-reactive protein (CRP) and total homocysteine (tHcy) concentrations are cardiovascular disease (CVD) risk factors. Previous studies indicated that higher serum carotenoid concentrations were inversely associated with some of these biomarkers. However, whether dietary carotenoid intake is inversely associated with these CVD risk biomarkers is not well known. We assessed the associations between individual dietary carotenoid intake and CVD risk biomarkers and tested whether the serum carotenoid concentrations explain (mediate) or influence the strength of (moderate) the associations, if any association exists. Dietary data collected from 2 24-h dietary recalls and serum measurements in adult men (n = 1312) and women (n = 1544) from the NHANES 2003-2006 were used. Regression models designed for survey analysis were used to examine the associations between individual dietary carotenoids and log-transformed blood cholesterol, CRP, and tHcy. The corresponding individual serum carotenoid concentration was considered as mediator (and moderator if applicable). After adjustment for covariates, significant inverse associations with LDL cholesterol were observed for dietary β-carotene (P < 0.05) and lutein + zeaxanthin (P < 0.001), and with tHcy for dietary β-carotene (P < 0.05), lycopene (P < 0.05), and total carotenoids (P < 0.05). Dietary lutein + zeaxanthin intake was also positively associated with HDL cholesterol concentrations (P < 0.01). Most of these associations were null after additional adjustment for corresponding serum carotenoid concentrations, indicating the complete mediation effects of serum carotenoids. Serum β-carotene significantly moderated the associations between dietary β-carotene and CRP (P-interaction < 0.05), and quartile 4 of dietary β-carotene was associated with lower CRP concentrations only among participants with serum β-carotene > 0.43 μmol/L. In this population-based cross-sectional study, serum carotenoids were mediators of dietary carotenoids and CVD risk biomarker associations. Serum β-carotene was also a moderator of the dietary β-carotene and CRP association. These findings may help in the design of future intervention studies on dietary carotenoids in the prevention of CVD.

MeSH terms

  • Adult
  • Biomarkers / blood
  • C-Reactive Protein / analysis
  • Cardiovascular Diseases / blood
  • Cardiovascular Diseases / epidemiology
  • Cardiovascular Diseases / etiology
  • Cardiovascular Diseases / prevention & control*
  • Carotenoids / blood
  • Carotenoids / therapeutic use*
  • Cholesterol, HDL / blood
  • Cholesterol, LDL / blood
  • Cross-Sectional Studies
  • Diet* / adverse effects
  • Female
  • Homocysteine / blood
  • Humans
  • Hypercholesterolemia / physiopathology
  • Hyperhomocysteinemia / physiopathology
  • Lutein / blood
  • Lutein / therapeutic use
  • Lycopene
  • Male
  • Nutrition Surveys
  • Risk Factors
  • United States / epidemiology
  • Xanthophylls / blood
  • Xanthophylls / therapeutic use
  • Zeaxanthins
  • beta Carotene / blood
  • beta Carotene / therapeutic use


  • Biomarkers
  • Cholesterol, HDL
  • Cholesterol, LDL
  • Xanthophylls
  • Zeaxanthins
  • beta Carotene
  • Homocysteine
  • Carotenoids
  • C-Reactive Protein
  • Lycopene
  • Lutein