Associations between carotenoids and rheumatoid arthritis risk: evidence from two population-based studies

BACKGROUND: Carotenoids are common natural anti-inflammatory antioxidants produced by plants and some microorganisms. However, there are limited reports on the association between carotenoids and risk of rheumatoid arthritis (RA). METHODS: We analyzed data of participants from the National Health and Nutrition Examination Survey (NHANES) and UK Biobank (UKB). In the NHANES, a total of 40,410 and 13,440 participants were included to evaluate the association of dietary carotenoid intake and serum carotenoid levels with RA, respectively. Among 179,171 participants in the UKB, 2,156 individuals developed RA over a median follow-up of 7.74 years. A weighted multivariable logistic regression model and quantile G-Computation (QGC) model were established to investigate single and joint effect of carotenoid levels (alpha-carotene, beta-carotene, beta-cryptoxanthin, lycopene, lutein and zeaxanthin) on RA in NHANES by cross-sectional study design. Furthermore, Cox proportional hazards regression model was used to validate the potential association in UKB cohort, and restricted cubic spline (RCS) regression was applied to estimate the possible nonlinear association. RESULTS: Compared to the lowest quartile (Q1), dietary intake level of alpha-carotene in Q3 were consistently associated with a decreased risk of RA in NHANES (odds ratio [OR]: 0.771, 95% confidence interval [CI]: 0.628 to 0.946, P-value = 0.013) and UKB (hazards ratio [HR]: 0.875, 95% CI: 0.775 to 0.987, P-value = 0.030). At serum level, all kinds of carotenoids in Q4 were significantly inversely associated with RA when compared with Q1. In QGC model, each one-quartile increase in the intake (OR: 0.886, 95% CI: 0.841 to 0.933, P-value < 0.001) or serum levels (OR: 0.795, 95% CI: 0.732 to 0.864, P-value < 0.001) of the six carotenoids showed a negative correlation against of RA in NHANES. RCS analysis indicated that dietary intake of total carotene (P-value = 0.0020 for nonlinear), alpha-carotene (P-value = 0.0175 for nonlinear), and beta-carotene (P-value = 0.0021 for nonlinear) exhibited a nonlinear relationship with the risk of RA in UKB. The protective effects against RA were found within the ranges of log natural transformation for total carotene (5.28 to 8.69 µg/day), and beta-carotene (5.18 to 8.61 µg/day). CONCLUSION: Both cross-sectional and prospective studies indicate that carotenoids, as natural and effective anti-inflammatory antioxidants, were negatively associated with RA. Daily intake of an appropriate dose can significantly reduce the risk of RA.