Background: Iron may promote coronary atherosclerotic disease (CAD) by increasing lipid peroxidation. Studies on biochemical or genetic markers of body iron stores as risk factors for CAD have yielded conflicting results.
Methods: We studied 849 individuals with a clear-cut definition of the CAD phenotype, i.e., with (CAD; n = 546) or without (CAD-free; n = 303) angiographically documented disease. We determined serum ferritin, as a biochemical estimate of iron stores, and the C282Y mutation in the HFE gene, i.e., the main cause of hemochromatosis in Caucasians. The relationships of ferritin with serum markers of either inflammation [C-reactive protein (CRP)] or lipid peroxidation (malondialdehyde) were also investigated.
Results: Mean ferritin concentrations were slightly higher in CAD vs CAD-free individuals, but this difference disappeared after adjusting for sex and CRP. Ferritin was significantly correlated with CRP (Spearman's test, rho = 0.129; P <0.001). Heterozygotes for Cys282Tyr were 4.8% among the CAD group and 6.6% among the CAD-free group (P = 0.26). The prevalence of high concentrations of stored iron, defined as ferritin concentrations above the sex-specific upper quintiles of the control distribution, was also similar in the two groups. There was a higher prevalence of "iron depletion" in CAD-free vs CAD females (20% vs 8.8%, respectively), but this difference disappeared after adjustment for age and other cardiovascular risk factors (odds ratio, 0.66; 95% confidence interval, 0.21-2.08). No differences in iron markers were found in CAD patients with or without myocardial infarction.
Conclusions: Our results do not support a role for biochemical or genetic markers of iron stores as predictors of the risk of CAD or its thrombotic complications.