Both low and high body iron stores relate to metabolic syndrome in postmenopausal women: Findings from the VIKING Health Study-Shetland (VIKING I)

Abstract

Background There are conflicting results among studies on the association between serum ferritin (SF) and metabolic syndrome (MetS), and by groups of sex/menopausal status. To date, there are no studies on British populations. The SF-MetS association might be U/J-shaped. We evaluated whether SF was independently associated with MetS (harmonized definition) in people from Shetland, Scotland.

Methods We analysed cross-sectional data from the Viking Health Study-Shetland (589 premenopausal women [PreMW], 625 postmenopausal women [PostW] and 832 men). Logistic regressions using two approaches, one with the lowest sex and menopausal status-specific ferritin quartile (Q) as the reference and other using the middle two quartiles combined (2–3) as the reference, were conducted to estimate the SF-MetS association. The shape of the association was verified via cubic spline analyses. The associations were adjusted for age, inflammatory and hepatic injury markers, alcohol intake, smoking and BMI.

Results Prevalence of MetS was 18.3%. Among PostMW both low and high SF were associated with MetS (fully adjusted odds ratios [95% confidence interval] compared to the middle two quartiles combined were: 1.99 [1.17–3.38] p =.011 for Q1 and 2.10 [1.27–3.49] p =.004 for Q4) This U-shaped pattern was confirmed in the cubic spline analysis in PostMW with a ferritin range of 15–200 ug/L. In men, a positive association between ferritin quartiles with Q1 as the reference, did not remain significant after adjustment for BMI.

Conclusion Extreme quartiles of iron status were positively associated with MetS in PostMW, while no SF-MetS associations were found in men or PreMW. The ferritin-MetS association pattern differs between populations and U/J-shaped associations may exist