Abstract

The purpose of this study is to examine the association between serum ferritin and transferrin saturation with all-cause, and CVD mortality among 13 858 persons (men: 6532, women: 7326) aged 20 years and older from death certificate data linked to the NHANESIII of a nationally representative sample of the noninstitutionalized USA population. Serum ferritin and transferrin saturation levels were categorised according to sex-specific quintiles. RR and 95% CIs were calculated from Cox proportional hazards regression models adjusted for age, race-ethnicity, poverty index, education, body mass index, smoking, alcohol intake, systolic blood pressure, total cholesterol, and Charlson Comorbidity Index. There were no statistically significant associations between serum ferritin and all-cause, and CVD mortality. There were statistically significant u-shaped associations between transferrin saturation and all-cause mortality in men (first quintile vs third quintile, RR 0.73 (95% CI 0.61 to 0.88), first quintile vs fifth quintile, RR 0.79 (95% CI 0.65 to 0.95) and between transferrin saturation and CVD mortality in women (first quintile vs fourth quintile, RR 0.58 (95% CI 0.48 to 0.84), first quintile vs fifth quintile, RR 0.68 (95% CI 0.48 to 0.98), all tests for trend, p<0.01). In this large cohort, there was consistent evidence of increasing risk of mortality at lower transferrin saturation levels. In fact, lower transferrin saturation levels were associated with an increased risk of all-cause and CVD mortality. The results are compatible with the possibility that there is an inverse association between transferrin saturation levels and risk of mortality.